article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the first evidence for a new kind of long - lived isomeric states was obtained in actinide fractions @xcite produced via secondary reactions in a cern w target which had been irradiated with 24-gev protons @xcite .
isomeric states with t@xmath5 @xmath6 0.6 y and @xmath7 30 d ( 10@xmath8 - 10@xmath9 times longer than the expected half - lives of the corresponding ground states ) were found in neutron - deficient @xmath10am and @xmath10bk nuclei , respectively . about 3x10@xmath9 atoms of @xmath10am and 4x10@xmath8 atoms of @xmath10bk
were produced in the isomeric states , and decayed by the @xmath11 or electron capture processes .
the character of these states was not clear : they are far from closed shells where high spin isomers are usually found , and they have very long lifetimes as compared to the known shape isomers .
in addition @xcite , some new particle groups were found in the decay of various actinide fractions separated from the same w target .
for example , an unambiguous 5.14 mev @xmath4-particle group with t@xmath5 @xmath6 4 y was seen in the bk source @xcite .
this half - life is a factor of 2x10@xmath12
3x10@xmath13 too short @xcite for the normal @xmath4-decay of bk pu nuclei .
it was very difficult to understand such an enhancement .
furthermore , 3.0 and 4.0 mev particle groups were seen from the am source in coincidence with l@xmath14 x - rays in the am region @xcite . since the relationship between the particle energies and their lifetimes deviate by about 23 and 12 orders @xcite from the systematics @xcite of @xmath4-particles , it was assumed @xcite that they were protons of unknown origin .
the clue to the understanding of the above mentioned findings has been obtained recently in several studies @xcite of the @xmath15o + @xmath16au reaction at e@xmath2 = 80 mev , where similar isomeric states and particle decays were found .
an isomeric state which decays by emitting a 5.20 mev @xmath4-particle with t@xmath5 @xmath6 90 m has been found in @xmath17fr .
since this half - life is longer than the known half - life of the ground state of @xmath17fr , it was concluded that a long - lived isomeric state had been formed in this nucleus .
a t@xmath5 of 90 m for 5.20 mev @xmath4-particles in @xmath17fr is enhanced by a factor of 3x10@xmath9 as compared to normal transitions @xcite .
however , this group was observed in coincidence with @xmath3-rays which fit predictions for a super - deformed band @xcite .
therefore the effect of large deformations of the nucleus on the @xmath4-particle decay was calculated and found @xcite to be consistent with the observed enhancement .
it was argued @xcite that since the isomeric state decays to a high spin state , it should also have high spin , and since it decays by enhanced @xmath4-particle emission to state(s ) in the second well of the potential , it should be in the second well itself .
the predicted @xcite excitation energies of the second minima in the evaporation residue nuclei and their daughters produced by the @xmath15o+@xmath16au reaction are above the proton separation energies .
therefore a search for long - lived proton decays has been performed @xcite using the same reaction .
two long - lived proton activities with half - lives of about 6 h and 70 h were found @xcite with proton energies of 1.5 4.8 mev with a sharp line at 2.19 mev .
this energy may correspond to a predicted transition @xcite from the second minimum in @xmath18tl to the ground state of @xmath16hg with e@xmath19 = 2.15 mev .
the superdeformed state in @xmath18tl may be produced in an alpha decay chain of transitions from shape isomeric states to shape isomeric states starting from @xmath17fr .
it should be noted that a long - lived high - spin isomeric state in the second minimum of the potential in @xmath20pu has been predicted by s. g. nilsson et al .
@xcite back in 1969 . the present work reports on the discovery of new long - lived isomeric states , produced by the @xmath0si + @xmath1ta reaction at bombarding energies of 125 and 135 mev , which decay by strongly hindered @xmath4-particle and proton transitions .
the lower energy is about @xmath21 below the coulomb barrier .
a fusion cross section of about 10 mb is predicted for this energy using a coupled - channel deformation code @xcite with deformation parameters @xmath22=0.41 for @xmath0si and @xmath22=0.26 for @xmath1ta @xcite , value for @xmath1ta was taken as the average from the corresponding values of @xmath23hf and @xmath24w . ] and allowing for 2@xmath25 and 3@xmath26 excitations in @xmath0si .
only 2 @xmath27b is predicted when no deformations are included in the calculations . for 135 mev
the corresponding predicted fusion cross sections are 95 mb with deformations and 40 mb without . at bombarding energies of 125 and 135 mev ,
the compound nucleus is formed at excitation energies of 42.1 and 50.8 mev , respectively .
preliminary results of this work have been published before @xcite .
in three irradiations , 2.5 mg/@xmath28 ta targets followed by stacks of c catcher foils were bombarded with a @xmath0si beam obtained from the pelletron accelerator in rehovot .
irradiations i and ii were performed with a 125 mev @xmath0si beam and irradiation iii with 135 mev .
carbon catcher foils of 200 @xmath27g/@xmath28 and 60 @xmath27g/@xmath28 were used in order to catch the evaporation residue nuclei and their daughters . in all experiments
the average beam intensity was about 11 pna and the total dose about 1x10@xmath15 particles .
long period off - line measurements were carried out in the laboratory in jerusalem , using the irradiated catcher foils as sources . in the present paper
we summarize the results of the particle-@xmath3 coincidence measurements.-@xmath4 correlation measurements were performed with 60 @xmath27g/@xmath28 c catcher foils situated in between two 300 @xmath27g/@xmath28 si detectors , but no @xmath4 correlation events with @xmath29t@xmath30 10 s could be significantly established . the preliminarily claimed correlations @xcite can not presently be ruled out to be due to other physical or electronic background effects . at longer correlation times the number of random events was too large . ] a 450 mm@xmath31 , 300 @xmath27 m thick , si surface barrier detector and a 500 mm@xmath31 , 10 mm thick , thin window ge(li ) detector were used for these measurements .
the si detector was calibrated using the 3.18 mev @xmath4-particle group of a @xmath32gd source and an accurate pulse generator .
the ge(li ) detector was calibrated using the x- and @xmath3-rays of a @xmath33co source .
the source was sandwiched between the si and the ge detector .
two 0.01 inch thick be foils separated the source from the ge detector .
the first foil was the window for the vacuum chamber which included the source and the si detector , and the second one the window of the ge detector .
the transparency of the two be windows together was close to 100% for gamma - rays with e@xmath34 @xmath7 10 kev .
the solid angle of the si detector was about @xmath35 of 4@xmath36 sr and of the ge(li ) detector @xmath37 .
the peak - to - total ratio of the ge(li ) detector was 100% up to about 120 kev and decreased gradually to 22% at 250 kev .
its fwhm energy resolution was about 900 ev .
the intrinsic resolution of the si detector was about 25 kev .
the full line widths for @xmath4-particles of 5.0 to 8.6 mev were 0.52 to 0.34 mev , respectively , with a 200 @xmath27g/@xmath28 c foil , and the corresponding values for the 60 @xmath27g/@xmath28 foil were 0.16 to 0.10 mev .
the resolving time of the coincidence system was 1 @xmath27s .
the coincidence events together with the singles particle events were recorded event by event with a time accuracy of 1 ms .
1 and 2 show typical singles particle and @xmath3-ray spectra obtained at 125 mev bombarding energy . figs . 3 and 4 show similar spectra obtained at 135 mev beam energy . from the measured @xmath4-particle and @xmath3-ray energies and lifetimes , production cross sections for various evaporation residue nuclei , with estimated errors of @xmath3825% , were deduced and are summarized in table 1 .
( cross sections for some isomeric states ( see below ) , are also given in table 1 . )
figs . 5 and 6 give respective @xmath4-@xmath3 coincidence plots from measurements i and ii , corresponding to irradiations i and ii .
. 7 and 8 are one dimensional projection plots on the @xmath4-particle axis of figs . 5 and 6 , respectively , for e@xmath3 @xmath7 25 kev .
9 presents a particle-@xmath3 coincidence spectrum from measurement iii , corresponding to irradiation iii.e - e system of ref .
@xcite was performed and gave negative results with an upper limit of about 0.5 nb for half - lives of between 20 hours to 70 days . on the other hand , in the 135 mev experiment ( fig .
9 ) we most probably saw protons in the particle-@xmath3 coincidences .
( see further below ) . ]
the estimated numbers of random coincidences in the 3 10 mev particle energy and 10 250 kev @xmath3-energy are 5.4 x 10@xmath39 , 4.4 x 10@xmath39 , and 0.4 in figs . 5 , 6 , and 9 , respectively . some of the coincidence events in figs . 5 , 6 and 9 , between 5 - 6 mev @xmath4-particles and various @xmath3-rays
, may perhaps be due to a contamination from an emanated @xmath40pb source from @xmath41th , which was used in the same chamber about 4 months earlier @xcite . despite the long time delay to our present long term measurements of 77 to 235 days , and careful cleaning of the chamber
, we might have picked up some residual contamination .
also , indication for k x - rays of rn is seen in this region . while in principle
their origin may be from some reaction products , they also may be from the decay of @xmath42ra , which belongs to the @xmath43u chain .
although it is not likely , but because of this ambiguity , we will not try to make a claim about this region .
a two - dimensional background measurement , taken for 8 days , before the @xmath40pb source was used in the chamber , gave zero events in the whole region of 3.5 10 mev particles and 0 250 kev @xmath3 s .
it is seen in figs . 5 and 6 , and figs . 7 and 8 , that while the region of @xmath4-particles of 6 8 mev is relatively empty , coincidence events are seen between various @xmath3-rays and @xmath4-particles of 8 9 mev . in fig . 5 ,
13 coincidence events between 7.99 - 8.61 mev @xmath4-particles and various @xmath3-rays of e @xmath7 20 kev are observed , where 8 of them fall within a narrow range of 190 kev particle energy , between 8.42 - 8.61 mev .
( the estimated @xmath4-particle full line width for the 200 @xmath27g/@xmath44 c foil is around 340 kev ) . from the intensities of the singles in the @xmath4- and @xmath3-ray spectra ( 339 @xmath4-particles and 2.14x10@xmath45 gamma - rays in 76.8 days ) , and the resolving time of 1 @xmath27s of the coincidence system ,
the total number of random coincidences in the 8 - 9 mev region was estimated to be 2.2x10@xmath46 . in fig .
6 one finds 8 coincidence events between 8.19 9.01 mev , where 5 of them fall within 70 kev , between 8.55 8.62 mev , and one at 9.01 mev .
( the @xmath4-particle full line width for a 60 @xmath27g/@xmath44 c foil is about 100 kev ) .
the corresponding estimated total number of random coincidences between 8 to 9 mev is 1.4x10@xmath46 .
one does not see such concentration of events in the same energy region in fig .
9 which was taken for about 2.7 times longer period than figs . 5 and 6 .
3 events are seen in the 8 - 9 mev region in fig .
9 which sets an upper limit on the background in figs . 5 and 6 to be aroud 1 count as compared to the respective 13 and 8 counts seen experimentally .
it was estimated that in the 8 9 mev range of @xmath4-particles , at most 0.8 events in fig . 5 and 0.3 events in fig .
6 may be due to the contamination mentioned above of coincidences between @xmath47 and @xmath3-rays from @xmath40bi and the 8.78 mev @xmath4-particles from @xmath40po ( t@xmath5 = 0.3 @xmath27s ) .
shows time sequence plots for the @xmath4-@xmath3 coincidences with @xmath4-energies of 89 mev obtained in measurements i and ii and seen in figs . 5 and 6 , respectively .
in measurement ii a growth in the intensity of the 89 mev group was found from the beginning up to 97 days after the end of irradiation . in measurement
i no significant change in decay rate was observed between 77 to 154 days and , by binning the data , t@xmath5 @xmath7 40 days was estimated for the lower limit of the decaying half - life .
an additional measurement , taken about 4.5 y after the one presented in fig . 5 and measurement i of fig .
10 , gave 0 counts in 22.0 d. an upper limit for the half - life of t@xmath5 @xmath30 2.1 y is deduced from this measurement . table 2(a ) and fig .
11 show that almost all the coincidence events in figs .
5 and 6 with 7.8 - 8.61 mev @xmath4-particles ( the encircled ones ) fit very nicely with a j(j+1 ) law assuming e@xmath48 = 4.42xj(j+1 ) kev and @xmath29j = 1 .
this fit is very significant from a statistical point of view @xcite .
all the gamma - rays in fig . 5 ( 7 out of 7 events ) , and almost all of them in fig . 6 ( 6 out of 7 events ) , which are above 90 kev , fit , to within @xmath380.5 kev on the average , with this formula .
the probability for 13 out of 14 events which are distributed evenly ( for instance due to compton effect from higher energy gamma - rays ) , to fall into 13 specified energy positions , within 1 kev is : @xmath49 .
p is equal to 36x1.0/(250 - 90 ) .
36 is the number of possible gamma - transitions between states of both integer and half integer spins with e@xmath48 = 4.42 x j(j+1 ) , in the range of 90 to 250 kev .
( the four low energy events of e@xmath50 @xmath30 60 kev in fig . 5 and the one event in fig .
6 which do not fit with the j(j+1 ) rule may be due to compton events ) .
it is also seen in fig . 6 and in table 2(b ) that the @xmath3-rays of the 3 coincidence events in the region of 6 - 7 mev @xmath4-energy of 6.16 mev - 176.1 kev , 6.94 mev - 207.4 kev and 6.41 mev - 242.3 kev , fit with 20 @xmath51 19 ( 176.4 ) , 47/2 @xmath51 45/2 ( 207.3 ) and 55/2 @xmath51 53/2 ( 242.6 ) j @xmath51 j-1 transitions , assuming e@xmath48 = 4.41xj(j+1 ) kev .
the probability in this case that this fit is accidental due to a chance coincidence of evenly distributed events is 1.2x10@xmath46 , taking into account that p = 18x0.47/(250 - 170 ) .
18 is the number of possible gamma - transitions in the range of 170 to 250 kev , and 0.47 kev is twice the average deviation from the specified energies .
( this probability increases to 1.1x10@xmath39 if , instead of 0.47 kev , the value which was used before of 1.0 kev is assumed in the calculation ) .
it should be mentioned that rotational constants around 4.4 kev correspond to superdeformed bands [ sdb ] in this region of nuclei @xcite . in fig .
9 one sees about 10 scattered events in the 6.5 9 mev region .
some of them may also fit with a j(j+1 ) law .
however , the statistical significance of the fit in this case is not so good and we will not discuss them further .
the appearance in table 2a of both integer and half - integer spins at low @xmath3-ray energies indicates that bands in both even and odd nuclei were formed .
( at high energies the observed transitions are probably sum events .
see the following paragraphs . )
@xmath52rn , produced via pn evaporation , and its daughters , are , from the kinematic point of view , the best candidates for the half - integer transitions .
the 8.42 mev @xmath4-particle in coincidence with a photon of 67.1 kev ( seen in fig . 5 and table 3 below ) , which fits with k@xmath14 x - ray of pt ( known energy 66.832 kev ) , may indicate that the @xmath4-transition is from @xmath53hg to @xmath54pt .
@xmath53hg is the daughter after 3 @xmath4-particle decays from @xmath52rn .
it should be mentioned that the high energy @xmath3-rays ( above about 120 kev , table 2 ) are most probably sum events rather than photo - peak events . if , for instance , the three coincidence events in fig . 6 at e@xmath34 of 247.0
( 2 events ) and 225.3 kev are photo - peak events , then [email protected] compton events below 120 kev ( the position of the compton edge ) should have been seen in coincidence with @xmath4-particles of 8.55 - 8.61 mev , but only one event is seen .
similarly if the @xmath3-rays which are in coincidence with 6 - 7 mev @xmath4-particles , seen in the same figure and mentioned above , are photo - peak events , then 7.6@xmath55 compton events should have been seen in this @xmath4-energy range at e@xmath50 @xmath30 120 kev , while only one event is seen .
on the other hand , for a de - excitation of a superdeformed band with @xmath29j = 1 , a measured energy which fits a transition between high - spins , may be due to many sum combinations of lower energy transitions between lower spins , for which low - energy compton events should not be observed .
for instance , an energy of 225.4 kev which fits a 51/2 @xmath51 49/2 transition ( assuming e@xmath48 = 4.42@xmath56j ( j+1 ) kev , table 2 ) may be due to one of 17 different combinations of 3-fold events , if only the first 14 levels of the band up to maximum spin of 27/2 , and maximum transition energy of 119.3 kev are considered , for which the photo - peak to total ratio is 100% .
( 119.3 ( 27/2 @xmath51 25/2 ) + 92.8 ( 21/2 @xmath51 19/2 ) + 13.3 ( 3/2 @xmath51 1/2 ) = 225.4 , or 84.0 ( 19/2 @xmath51 17/2 ) + 75.1 ( 17/2 @xmath51 15/2 ) + 66.3 ( 15/2 @xmath51 13/2 ) = 225.4 are two examples out of these 17 possible combinations ) . because of the large solid angle of the @xmath3-detector ( 0.16 ) and the large number of possible 3-fold combinations , the probability p@xmath57 to see a 3-fold event is larger than the probability p@xmath58 to see a 1-fold event .
( these probabilities for the above example are : p@xmath58 = @xmath59(1-@xmath59)@xmath60 = 0.042(1 - 0.16)@xmath61 = 6.4 x 10@xmath62 , and p@xmath57 = n@xmath57@xmath59@xmath13(1-@xmath59)@xmath63 = 17x0.16@xmath13(1 - 0.16)@xmath64 = 1.2x10@xmath39 , about 20 times larger .
@xmath59 is the photo - peak efficiency of the detector for the particular energy , @xmath59 is the geometrical efficiency , n is the number of transitions in the band counting from the level to which the @xmath4-particle is decaying , and n@xmath57 is the number of 3-fold combinations ) .
many combinations of 2- or 4- fold events of a half - integer spin band will give energies which fit with those of a band of an integer spin .
for instance , the two events at 247.0 kev ( table 2 ) which fit in energy to a 28 @xmath51 27 transition , and the 186.5 kev event which fits with a 21 @xmath51 20 transition , may be due to such combinations , since the @xmath4-particle energy is the same as that of the half - integer transitions of 225.3 and 171.8 kev , and therefore they presumably belong to the same band .
( in an integer - spin band of @xmath29j = 1 various combinations of 2- , 3- and 4-fold events of low energy transitions have large efficiencies and give energies which fit with those of transitions between higher spins .
this property , both for half - integer and integer spin bands , does not exist for @xmath29j = 2 transitions ) .
as shown above the detection of sum events which are energetically degenerate with the diagram line leads to a substantially increased effective detection efficiency ( dependent , however , on the maximum available spin ) for this energy .
the minimum spin which is consistent with the data is 19/2 .
the largest efficiency of the detector ( which corresponds to the smallest production cross section ) , taking into account 2- and 4-fold sum events of transitions between half - integer spins in order to get the 247.0 kev transition , is obtained with a spin of 35/2 .
the lower limit for the production cross section estimated from irradiation ii and taking into account the detector efficiency for various possible combinations of sum events , as described above , is about 130 nb ( table 1).-particles , are photo - peak events due to known transitions in @xmath53hg and @xmath16tl .
however , in this case [email protected] compton events below 120 kev should have been seen in fig . 6 in coincidence with @xmath4-particles of 8.55 - 8.62
mev , but only one event is seen.]-energies are compton events due to higher energy @xmath3-rays .
but then the probability @xcite to have two events in the same position ( within the energy resolution ) is only a few percent , and this occurred twice , at 247.0 ( fig .
6 ) and 225.3 kev ( figs . 5 and 6 ) . ]
as shown above , the observation in table 2 of gamma rays which fit in energies with transitions between high spin sdb states is consistent with various sums of gamma rays due to transitions between lower spin states with @xmath29j = 1 . since the intensity of the characteristic x - rays seen in figs . 5 and 6 in coincidence with 8 - 9 and 6 - 7 mev @xmath4-particles is very low
, the observed @xmath3-rays should be due to e1 transitions , with quite low conversion factors , rather than m1 ones .
such transitions have been seen before @xcite and were predicted for sd wells in nuclei around z = 86 and n = 116 @xcite . in order to identify some of the coincidence events seen in figs .
5 , 6 , and 9 at lower particle energies below 6 mev , we first looked on groups of events .
the existence of groups of 3 or 2 coincidence events in clean regions indicates that they are probably not background .
secondly , one looks on their accurately measured photon energies . at low energies the observed photon may be either an x - ray or a @xmath3-ray .
the energies of the characteristic x - rays of the elements are known very well @xcite .
the k@xmath14 x - rays of the various elements of the evaporation residue nuclei and their daughters are separated one from another by about 2 kev , where the value of a k@xmath14-line of an element with z protons is about the same as the value of the k@xmath65-line of an element with z+1 protons .
if the measured energy turns out to be very close to a known x - ray energy , then it is reasonable to assume that the observed photon is an x - ray rather then a @xmath3-ray .
( in the cases found by us and discussed below the deviations from the known values are from 0.03 to 0.28 kev , while the width of one channel in the @xmath3-spectrum is about 0.24 kev and the fwhm is about four channels ) .
the x - ray may be emitted if the @xmath4-particle is decaying to an excited state which decays by conversion electrons .
such a process is followed by emitting characteristic x - rays of the daughter nucleus .
the z - value of the daughter is thus determined . in order to determine its a - value we tried to identify the @xmath3-rays which are in coincidence with @xmath4-particles of about the same energy .
a consistency is obtained when the @xmath3-ray fit in energy with the measured value , belongs to an isotope of the determined element , and it is consistent with known or expected values of the conversion factors , and also with the decay scheme . in several cases , as seen below ,
some of the observed events have been identified as due to characteristic x - rays ( the events surrounded with squares in figs . 5 , 6 and 9 ) and as known @xmath3-transitions in various nuclei ( the events surrounded with triangles in figs . 5 , 6 and 9 ) . in fig .
6 two coincidences between about 5.50 mev @xmath4-particles and 61.1 and 59.4 kev photons are seen , and fit with k@xmath14 and k@xmath65 x - rays of re of 61.140 and 59.718 kev . at about the same @xmath4-energy two 144.0 kev events are observed and can be identified with a known transition in @xmath66re .
this identification is supported by the observation in fig . 5 , in the same @xmath4-energy range , of coincidence events with respective 59.7 kev and 140 kev x- and @xmath3-rays .
the first fits with k@xmath65 of re , and the second with a transition in @xmath66re , which precedes the above 144.0 kev transition .
a coincidence between 5.45 mev @xmath4-particles and 59.4 or 59.7 kev photons may in principle be also due to the known decay of @xmath20am of 5.486 mev @xmath4 and 59.54 kev @xmath3 .
however , we did not use and never had in our laboratory a @xmath20am calibration source . under these circumstances
the observation of the 61.1 kev photon , which fits very nicely with k@xmath14 of re , and which is very far off ( 1.56 kev ) from the @xmath3-rays of @xmath20am , and of the two consecutive @xmath3-rays of @xmath66re , suggests that the two events at about 59.6 kev are due to k@xmath65 of re and not due to @xmath20am
. a 59.6 kev x - ray may in principle be also due to k@xmath14 of w of 59.3182 kev . here also the observation of the other x - rays and @xmath3-rays suggest that it is k@xmath65 of re rather than k@xmath14 of w. out of total number of 3 x - rays , one is expecting to see 1.9 k@xmath14 events and 1.1 k@xmath65 events .
the observation of 1 event in the first case and 2 in the second , is well within the statistical error . in fig .
9 , for @xmath4-particles between 5.18 and 5.53 mev , groups of two and three events with photon energies of 66.8 kev and 59.6 kev , respectively , are seen .
they may correspond to the known k@xmath14 x - rays of pt of 66.832 kev and of w of 59.318 kev of w. at the same corresponding @xmath4-energies known @xmath3-rays of @xmath67pt and @xmath68w are observed . here
the three 59.6 kev events are more likely to be due to k@xmath14 of w , rather than k@xmath65 of re . if they were k@xmath65 of re then 5.2 events of k@xmath14 at 61.140 kev are expected .
the observation of zero events when 5 are expected is unlikely . on the other hand ,
if the 3 observed events are k@xmath14 x - rays of w , then 1.7 events of k@xmath65 of w at 57.98 kev are expected .
the observation of zero events when 1.7 are expected is well within the statistical errors .
as mentioned above and seen in table 3 , about the same @xmath4-energy corresponds to two different decays .
one leads to @xmath66re and was obtained in the e@xmath2 = 125 mev experiment , and the other presumably leads to @xmath68w and was obtained in the e@xmath2 = 135 mev experiment . as seen below ( section iva ) , these data are interpreted as due to transitions from the second well of the potential in the parent nuclei to the normal states in the daughters . there is nothing against having about the same @xmath4-particle energy in two different transitions . because of the low statistics we consider the identifications mentioned above as tentative only , where the identification of @xmath66re is better than of the other .
the results are summarized in table 3 together with possible reaction channels and corresponding decay chains which lead to the observed transitions .
the deduced production cross sections ( with an accuracy of about a factor of 2 ) are given in table 1 , columns 3 and 5 .
in general lower limits were deduced since the branching ratios along the decay chains are not known . some coincidence events between 6.17 9.01 mev @xmath4-particles and identified x - rays with no corresponding @xmath3-rays were seen .
they are also given in table 3 .
a very well defined coincidence group of three events is seen in fig . 9 at an average particle energy of 3.88 mev and @xmath3-energy of 185.8 kev .
because of the low energy of the particles and the very narrow total width of 40 kev , one may conclude that the particles are protons .
( the total estimated widths at this energy are 55 kev for protons and 630 kev for @xmath4-particles .
the estimated half - life for 3.88 mev @xmath4-particles decaying to , for instance , @xmath69hg is about 1x10@xmath70 y @xcite ) .
the 185.8 kev @xmath3-rays fit with a known transition in @xmath71rn ( table 3 ) .
the production cross section of this group is given in table 1 .
the e@xmath72 values for the g.s . to g.s . transitions for @xmath73ir @xmath51 @xmath66re ; @xmath74os @xmath51 @xmath68w and @xmath75hg @xmath51 @xmath67pt mentioned in table 3 are 2.74 , 2.662 and 2.927 mev @xcite , respectively .
the corresponding observed @xmath4-energies of 5.43 , 5.53 and 5.18 mev ( table 3 ) are clearly due to decay of isomeric states in the parent nuclei .
the estimated half - lives for normal @xmath4-particles of these energies @xcite are 2.9x10@xmath13 ; 3.5x10@xmath31 and 1.2x10@xmath12 s. the observed lifetimes of several months , which may be due to combined lifetimes along the long decay chains , are retarded by 3 and 4 orders of magnitude in the first and second case , respectively . from the above
mentioned measured @xmath4-energies the lower limits for the excitation energies of the isomeric states in the parent nuclei @xmath73ir , @xmath74os and @xmath75hg ( see table 3 ) are deduced to be 3.1 ; 3.0 and 4.5 mev , respectively .
( lower limits are deduced since the @xmath4-decay may proceed through a higher excitation energy than the one seen experimentally ) . extrapolated and interpolated predicted energies @xcite for the second minima of 4.1 @xcite or 4.2 @xcite ; 3.4 @xcite or 3.6 @xcite ; and 4.2 @xcite or 4.6 @xcite mev , respectively , are also given in table 3 .
the observed energies of the isomeric states and the predicted positions of the second minima seem to be in the same range of excitation energies .
this suggests that the isomeric states in these cases are in the second well of the potential , and that the @xmath4 transitions are from the superdeformed well in the parent nuclei to less deformed or normal states in the daughters .
these are the last transitions after several decays , presumably from isomeric state to isomeric state , all within the second wells along the decay chain .
for instance , @xmath73ir in the isomeric state may be produced after 6 decays , 4@xmath4 and 2@xmath76@xmath77 or ec decays , all from superdeformed to superdeformed states ( see table 3).-particle or even by proton emission , is in principle not impossible .
] since the last step of the chain is a decay to a relatively high spin state , it seems that the originally produced isomeric state in the evaporation residue nucleus has high spin , and that the transitions from mothers to daughters along the decay chain are between high spin states .
we now discuss the three coincidence events seen in fig . 6 and mentioned above ( secion iii.b . and table 2b ) of 6.16 mev
176.1 kev ; 6.94 mev 207.4 kev and 6.41 242.3 kev .
these events appeared 96 , 68 and 78 days respectively , after the end of irradiation , and their lifetimes therefore are of the order of several months .
as shown above , the energies of the first two @xmath3-rays fit very nicely with sdb transitions between half - integer spins , where the most probable candidates are @xmath52rn , produced by the pn evaporation process , and its daughters .
the coincidence event of a 6.17 mev @xmath4-particle with 79.1 kev photon ( fig
. 5 and table 3 ) which fits with po k@xmath14 x - rays of 79.290 kev , suggests that the transitions mentioned above with 6 7 mev @xmath4-particles might be from isomeric state(s ) in @xmath52rn to the sdb states in the second well of the potential in @xmath78po .
( the half - life of the ground state of @xmath52rn is 9.3 m ) .
the measured @xmath4-particle energies are in the predicted range for the transition from the second minimum in @xmath52rn to the second minimum in @xmath78po of 6.2 @xcite or 6.9 @xcite mev . similarly the energies of 7.16 mev ( fig .
6 ; e@xmath2 = 125 mev ) and 7.05 mev ( fig . 9 ; e@xmath2 = 135 mev ) @xmath4-particles in coincidence with l@xmath79(at ) ( see table 3 ) may correspond to the predicted @xmath4-particle transitions between the second minima of @xmath80fr@xmath81 and @xmath82at@xmath81 ( 3n reaction ) of 6.9 @xcite or 7.3 @xcite mev , and of @xmath83fr@xmath81 to @xmath84at@xmath81 ( 4n reaction ) of 7.0 @xcite and 7.2 @xcite mev , respectively . in both cases discussed in the present subsection ( iv.b . ) , while the @xmath4-energies are consistent with this picture , the very long lifetime may indicate that the internal structure of the parent isomeric states and the final rotational sdb states are different .
for the 3.88 proton group ( section iii.d . above )
, if the identification given in table 3 is correct , the excitation energy of the isomeric state in @xmath83fr is @xmath7 6.7 mev ( see table 3 ) .
this is quite high as compared to 3.9 @xcite or 5.0 @xcite mev predicted for the second minimum , and may indicate that the origin of the isomeric state in this case is different .
the most striking result is the @xmath4-particle group around 8.6 mev which was found in coincidence with sdb transitions ( figs . 5 , 6 , 7 , 8 , and table 2a ) .
as mentioned above the appearance of @xmath3-transitions between half - integer spins shows that most of these transitions are in an odd a nucleus , where @xmath52rn and its daughters are most probable candidates from the reaction kinematic point of view .
the observation of a coincidence event between a 8.42 mev @xmath4-particle ( which is , within the experimental spread for @xmath4-particles , consistent with 8.6 mev for 200 @xmath27g/@xmath28 c catcher foil ) and a k@xmath14 x - ray of pt ( fig .
5 and table 3 ) indicates that the transition may be from @xmath53hg ( produced after 3 @xmath4-decays from @xmath52rn ) to @xmath54pt .
however , transitions from other hg isotopes produced by fewer @xmath4-decays , which are followed by several @xmath76@xmath77 or ec decays can not be excluded .
the predicted @xcite half - life for normal 8.6 mev @xmath4-particles is below 1 @xmath27s .
therefore in principle the @xmath3-rays of the sdb transitions may either precede or follow the 8.6 mev @xmath4-particles .-
particles is always much longer than 1@xmath27s , and they could not be detected in coincidence with @xmath3-rays which precede them . ] in the case where the @xmath3-rays come first , the @xmath4-particles are due to either a transition from the second minimum in the parent nucleus to the second minimum in the daughter , or from the second minimum in the parent to normal states in the daughter .
however in the first case the energy of 8.6 mev is much larger than the predicted 3.3 mev in @xmath53hg@xmath81 @xmath51 @xmath54pt@xmath81 transition @xcite , and 6.3 @xcite or 6.9@xcite in @xmath52rn@xmath81 @xmath51 @xmath78po@xmath81 transition . on the other hand in the second case of an @xmath4-transition from the second minimum to normal states ,
the @xmath4-particle lifetime will most probably be retarded , and the 8.6 mev @xmath4-particles will not be in coincidence within 1 @xmath27s with the sdb transitions .
( furthermore , the predicted transition energy from the second minimum to the ground state in @xmath53hg@xmath81 @xmath51 @xmath54pt@xmath85 is about 7.2 mev @xcite and in @xmath52rn@xmath81 @xmath51 @xmath78po@xmath85 is 6.3 @xcite or 6.9 mev @xcite .
these values are also not in accord with the experimental value of 8.6 mev ) .
it is therefore reasonable to assume that the 8.6 mev @xmath4-particles decay first and then are followed by the sdb @xmath3-rays , similar to the situation in all the other cases discussed above .
let us first consider the case where the isomeric state is in the second well of the parent nucleus which decays by the 8.6 mev @xmath4-particles to the sdb in the daughter .
one faces here two problems , namely the very large hindrance factor of the @xmath4-particles , and their high energy .
the hindrance factor of the @xmath4-particles is in the range of 10@xmath15 ( t@xmath5 @xmath6 40 d as compared to the predicted value of 1.4 x 10@xmath86 s assuming @xmath76@xmath87 = 0.7 @xcite ) .
as for the energy , if , for instance , the 8.6 mev @xmath4-particles decay to a sdb state of spin 27/2 in @xmath54pt at e@xmath48 = 865 kev ( e@xmath88 = 4.42 kev ) , and taking into account the predicted excitation energy of the second minimum in @xmath54pt at about 4.1 mev @xcite , and the q@xmath89([email protected] . )
value for @xmath53hg of 2.190 mev @xcite , the isomeric state in @xmath53hg turns out to be at about 11.6 mev above the g.s . assuming an excitation energy of the second minimum in @xmath53hg of about 5.2 mev @xcite , the above energy of 11.6 mev corresponds to an excitation energy of about 6.4 mev above the second minimum , which seems unlikely .
( in the same way , about 5.0 and 3.0 mev excitation energies above the second minimum are respectively deduced for more neutron - rich hg isotopes and for @xmath52rn , which is unlikely as well . )
it seems reasonable to conclude that the large hindrance factor might be due to another barrier transition , as , for instance , from the third well ( the hyper - deformed well ) @xcite to the second one.cm at @xmath90 = 0.63 and e@xmath48 = 8.2 mev , using hartree - fock calculations @xcite . ]
it should be mentioned that an extrapolated value of about 11.5 mev is obtained for the excitation energy of the third minimum in @xmath53hg from the predictions of ref .
this value is in accord with the deduced value mentioned above of about 11.6 mev
.- events of 6 - 7 mev discussed in section iv.b .
above is unlikely to be due to transitions from the third minimum , as its predicted excitation energy of about 18.8 mev ( extrapolated from ref.@xcite ) comes out far from the data of 7.8 or 10.3 mev ( table 3 ) . ]
the large production cross section of the isomeric state which decays by the 8.6 mev @xmath4-particles of about 130 nb , as compared to the production of the isomeric states in the second minima of a few nb ( see table 1 ) , may also indicate that its origin is from the third minimum , which can presumably be produced more easily at bombarding energies below the coulomb barrier .
it is also seen in table 1 that while the highest cross section for the production of the normal evaporation residue nuclei is about a factor of 60 larger at 135 mev as compared to 125 mev , the production of the isomeric states in the second minima is about the same at these two bombarding energies , whereas there is an exceptionally large cross section at 125 mev for the 8.6 mev @xmath4-particle group which is in coincidence with the sdb transitions .
in summary , evidence for long - lived isomeric states with lifetimes of up to several months , with abnormal decay properties , was obtained .
relatively high energy @xmath4-particles of 5.1 - 5.5 mev were tentatively found in the usually non-@xmath4-emitting nuclei , @xmath74os , @xmath73ir and @xmath75hg , and interpreted as due to transitions from the second well of the potential in the parent nuclei to less deformed or normal states in the daughters .
the observed transitions are the last in long chains of up to 6 steps of @xmath4-particle and @xmath11 or ec transitions , presumably within the second well itself .
long - lived 6 7 mev @xmath4-particles , in coincidence with sdb @xmath3-rays , were found with energies consistent with transitions from the second minimum in the parent nucleus to the second minimum in the daughter .
the reason for their long lifetimes though is not clear . a 3.88 mev long - lived proton group was found .
its state of origin is not entirely clear .
by far the most exciting observation is the very high energy and strongly retarded ( t@xmath5 @xmath7 40 d ) 8.6 mev @xmath4-particle group in coincidence with sdb transitions , which is interpreted as due to a long - lived isomeric state in @xmath53hg at e@xmath48 @xmath91 11.6 mev , and is consistent with a transition from the third ( hyperdeformed ) well of the parent nucleus to the second ( superdeformed ) well in the daughter .
the production cross sections of the isomeric states in the second well were about the same at e@xmath2 = 125 ( 10% below the barrier ) and at e@xmath2 = 135 mev , while on the average a factor of 15 60 larger cross section was found for the 8.6 mev group at 125 mev .
the existence of such long - lived isomeric states , with lifetimes much longer than those of their corresponding ground states , and with preferential production cross sections at relatively low bombarding energies , may add new considerations regarding the stability and the production mechanism of heavy and superheavy elements @xcite .
in particular it should be mentioned that the extra - push energies needed for the production of such nuclei in their superdeformed and hyperdeformed wells are much smaller than expected for producing them in their normal states @xcite .
we acknowledge the support of the accelerator crew of the weizmann institute at rehovot , and the technical assistance of s. gorni , o. skala and the electronic team of the racah institute . d.k .
acknowledges the financial support of the dfg .
we are grateful to n. zeldes and j. l. weil for very valuable discussions .
a. marinov , s. eshhar and d. kolb , phys .
b191 * , 36 ( 1987 ) .
a. marinov , s. eshhar and j. l. weil , _ proc .
inter . symp . on superheavy elements , lubbock , texas , 1978
_ , edited by m. h. k. lodhi ( pergamon , new york , 1978 ) , p. 72 .
a. marinov , c. j. batty , a. i. kilvington , g. w. a. newton , v. j. robinson and j. d. hemingway , nature * 229 * , 464 ( 1971 ) .
a. marinov , s. eshhar and d. kolb , fizika * 19 * , supplement 1 , 67 ( 1987 ) .
a. marinov , 6th adriatic int .
, _ frontiers of heavy - ion physics , 1987 _ eds .
n. cindro , w. greiner and r. aplan , ( world scientific 1987 ) , p. 179 .
a. marinov , s. gelberg , and d. kolb , int .
_ exotic nuclear states _ , debrecen , hungary , 1997 , eds
dombrdi , z. gsci and a. krasznahorkay ( akadmiai kiad , budapest , aph n.s . , heavy ion physics * 7 * , 47 ( 1998 ) . v. e. viola jr . and g. t. seaborg , j. inorg . nucl . chem . * 28 * , 741 ( 1966 ) .
a. marinov , s. gelberg and d. kolb , mod .
a11 * , 861 ( 1996 ) .
a. marinov , s. gelberg and d. kolb , mod .
* a11 * , 949 ( 1996 ) . w. satula , s. wiok , w. nazarewicz , r. wyss and a. johnson , nucl . phys . *
a529 * , 289 ( 1991 ) .
s. j. krieger , p. bonche , m. s. weiss , j. meyer , h. flocard and p .- h .
heenen , nucl . phys . * a542 * , 43 ( 1992 ) .
s. g. nilsson , g. ohln , c. gustafson and p. mller , phys .
30b * , 437 ( 1969 ) .
j. fernndez - neillo , c. h. dasso and s. landowne , code ccdef , comp .
. comm . * 54 * , 409 ( 1985 ) .
s. raman , c. h. malarkey , w. t. milner , c. w. nestor , jr . and
p. h. stelson , adndt * 36 * , 1 ( 1987 ) .
a. marinov , s. gelberg and d. kolb , _ physics of unstable nuclear beams , so paulo , brazil , 1996 , _ eds .
c. a. bertulani , l. f. canto and m. s. hussein , ( world scientific , 1996 ) , p. 181 .
a. marinov , s. gelberg and d. kolb , @xmath92 , dubna , russia,1997 , eds .
oganessian r. kalpakchieva world scientific , 1998 , p. 437 .
f. s. stephens et al .
lett . * 65 * , 301 ( 1990 ) .
s. @xmath93berg , nucl . phys .
* a520 * , 35c ( 1990 ) .
j. hller , and s. @xmath93berg , z. phys .
* a336 * , 363 ( 1990 ) . w. nazarewicz , p. olanders , i. ragnarsson , j. dudek and g. a. leander , phys
lett . * 52 * , 1272 ( 1984 ) . k .- h .
schmidt , c .- c .
sahm , k. pielenz and h .-
clerc , z. phys . *
a316 * , 19 ( 1984 ) .
g. audi , o. bersillon , j. blachot and a. h. wapstra , nucl .
* a624 * , 1 ( 1997 ) .
r. b. firestone , v. s. shirley , c. m. baglin , s. y. f. chu and j. ziplin , table of isotopes , john wiley and sons , 1996 .
w. m. howard and p. mller , adndt * 25 * , 219 ( 1980 ) .
w. nazarewicz , phys . lett .
* b 305 * , 195 ( 1993 ) .
s. wiok , w. nazarewicz , j. x. saladin , w. plciennik and a. johnson , phys . lett . * b 322 * , 304 ( 1994 ) .
d. kolb and a. marinov , proc .
florence , vol . 1 , p. 92
a. marinov , s. eshhar , j. l. weil and d. kolb , phys .
. lett . * 52 * , 2209 ( 1984 ) ; * 53 * , 1120(e ) ( 1984 ) .
a. marinov , s. gelberg and d. kolb , in _
6th int . conf . on nuclei
far from stability and 9th int . conf . on atomic masses and fundamental constants _ , bernkastel - kues , germany , inst .
132 , p. 437
k. kumar , superheavy elements , adam hilger , bristol and new york , 1989 . j. p. blocki , h. feldmeier and w. j. swiatecki , nucl .
* a459 * , 145 ( 1986 ) .
lllll reaction & @xmath94@xmath85(@xmath27b ) & & @xmath94@xmath85(@xmath27b ) & + & 125 mev & * 125 mev * & 135 mev & * 135 mev * + + 3n & @xmath95 & & @xmath96@xmath97 & + p2n & 4.5 & & @xmath98@xmath97 & + 4n & & & 274 & + p3n & & & 101 & + 5n & & & @xmath99@xmath100 & + p4n & & & @xmath99@xmath100 & + @xmath42n & & & & + pn & & & & + @xmath101 lower limit is given since the branching ratios along the decay chain ( table 3 ) are not known .
+ @xmath97 it was impossible to distinguish between the 3n and the p2n reactions .
the value given was deduced assuming that the relevant observed activity is due to this reaction channel only .
+ @xmath100 it was impossible to distinguish between the 5n and the p4n reactions .
the value given was deduced assuming that the relevant observed activity is due to this reaction channel only .
+ @xmath102 based on the coincidence group of the 3.88 mev protons with 185.8 kev @xmath3-rays and assuming that the @xmath3-rays are from @xmath71rn ( see text )
. + lllll e@xmath72 & e@xmath50(expt.)@xmath101 & e@xmath50(cal . ) & @xmath103e & transition@xmath97 + ( mev ) & ( kev ) & ( kev ) & ( kev ) + + + 7.81 & 25.9 & 26.5 & 0.6 & 3@xmath512 + 8.52 & 40.1 & 39.8 & + 0.3 & 9/2@xmath517/2 + 8.21 & 52.9 & 53.0 & 0.1 & 6@xmath515 + 8.53@xmath100 & 97.9@xmath100 & 97.2 & + 0.7 & 11@xmath5110 + 8.19 & 118.4 & 119.3 & 0.9 & 27/2@xmath5125/2 + 7.88 & 123.4 & 123.8 & 0.4 & 14@xmath5113 + 8.37 & 141.3 & 141.4 & 0.1 & 16@xmath5115 + 7.99 & 145.6 & 145.9 & 0.3 & 33/2@xmath5131/2 + 8.51 & 171.8 & 172.4 & 0.6 & 39/2@xmath5137/2 + 8.60 & 186.5 & 185.6 & + 0.9 & 21@xmath5120 + 8.07 & 212.9 & 212.2 & + 0.7 & 24@xmath5123 + 8.59@xmath102 & 225.3@xmath102 & 225.4 & 0.1 & 51/2@xmath5149/2 + 8.61@xmath104 & 247.0@xmath104 & 247.5 & 0.5 & 28@xmath5127 + + 6.16 & 176.1 & 176.4 & -0.3 & 20@xmath5119 + 6.94 & 207.4 & 207.3 & + 0.1 & 47/2@xmath5145/2 + 6.41 & 242.3 & 242.6 & -0.3 & 55/2@xmath5153/2 + @xmath101 the peak to total ratio was 100% up to about 120 kev and reduced gradually to 22% at 250 kev .
+ @xmath97 the transitions between the maximum possible spins are given .
at relatively high energies the observed @xmath3-rays are most probably sum events due to various combinations of lower energy transitions in the band which , because of the j(j+1 ) law , fit in energy to calculated transitions between higher energies .
( see text ) .
+ @xmath100 average of 2 events : 8.47 mev @xmath4 in coincidence with 97.4 kev @xmath3 in the first measurement and 8.58 mev @xmath4 in coincidence with 98.3 kev @xmath3 in the second measurement .
+ @xmath102 average of 2 events : 8.61 mev @xmath4 in coincidence with 224.9 kev @xmath3 in the first measurement and 8.57 mev @xmath4 in coincidence with 225.7 kev @xmath3 in the second measurement .
+ @xmath104 average of 2 events in the second measurement of 8.62 mev @xmath4 in coincidence with 246.5 kev @xmath3 and 8.60 mev @xmath4 in coincidence with 247.5 kev @xmath3 .
+ e@xmath50(meas . no . ) & e@xmath105 & isotope & transition & reaction & e@xmath106(@xmath107z ) & e@xmath108 + kev & mev & & & * decay * & mev & mev + * time / d*@xmath101 & & & & * chain * & & s ; k + + 61.1(ii ) & 5.48 & & k@xmath14(re ) & & & + * 31.1 * & & & & & & + 59.6(x2)(i ; ii ) & 5.45 & & k@xmath65(re ) & & & + * 7.4 ; 107.2 * & & & & & & + 144.0(x2)(ii ) & 5.43 & @xmath66re & 330(?)@xmath97@xmath51186 & 3n & @xmath73.1(@xmath73ir ) & 4.1@xmath102 ; 4.2@xmath102 + * 60.2 ; 88.8 * & & & ( 5@xmath77)@xmath51(6)@xmath109 & * 4 , * & & + & & & e1 ; 100% & * 2(ec)@xmath100 * & & + 141.0(i ) & 5.29 & @xmath66re & 471(?)@xmath97@xmath51330(?)@xmath97 & 3n & @xmath73.2(@xmath73ir ) & 4.1@xmath102 ; 4.2@xmath102 + * 130.2 * & & & ( 4@xmath77)@xmath51(5@xmath77 ) & * 4 , * & & + & & & m1+e2 & * 2(ec)@xmath100 * & & + 66.8(x2)(iii ) & 5.18 & & k@xmath14(pt ) & & & + * 13.9 ; 58.6 * & & & & & & + 236.2(iii ) & 5.12 & @xmath67pt & [email protected] & p3n ; 4n & @xmath74.53 ( @xmath75hg ) & 4.2@xmath104 ; 4.6@xmath104 + * 139.5 * & & & ( @xmath110@xmath77)@xmath51(@xmath111@xmath109 ) & & & + & & & ( e1 ) & * ; * @xmath100 & & + 59.6(x3)(iii ) & 5.53 & & k@xmath14(w ) & & & + * t@xmath112 = 90 * & & & & & & + 210.3(iii ) & 5.43 & @xmath68w & [email protected] & @xmath42n & @xmath73.04(@xmath74os ) & 3.4@xmath102 ; 3.6@xmath102 + * 168.4 * & & & @xmath113@xmath109@xmath51@xmath114@xmath109 & & & + & & & e2 ; 100% & @xmath100 & & + 70.9(ii ) & 9.01 & & k@xmath115(hg ) & & & + * 89.7 * & & & & & & + 67.1(i ) & 8.42 & & k@xmath115(pt ) & & & + * 81.7 * & & & & & & + 79.1(i ) & 6.17 & & k@xmath115(po ) & & & + * 133.5 * & & & & & & + 13.9(ii ) & 7.16 & & l@xmath116(at ) & & & + * 64.6 * & & & & & & + 13.9(iii ) & 7.05 & & l@xmath116(at ) & & & + * 127.8 * & & & & & & + 185.8(x3)(iii ) & 3.88(p ) & @xmath71rn & [email protected]@xmath117 & 4n & @xmath76.7(@xmath83fr ) & 3.9@xmath104 ; 5.0@xmath104 + * t@xmath112 = 120 * & & & ( 9@xmath109)@xmath51(8@xmath77 ) & & & + & & & e1 ; 55% & & & + sdb(ii)@xmath118 & 6 - 7 & @xmath78po & sdb & pn & 7.8@xmath119;10.3@xmath120(@xmath52rn ) & 6.8@xmath102 ; 9.9@xmath104 + * t@xmath112 = 53 * & & & table 2(b ) & & + sdb(i ; ii)@xmath121 & 8.6 & @xmath54pt@xmath122 & sdb & pn & @xmath611.6(@xmath53hg)@xmath122 & 11.5(@xmath123)@xmath124 +
* t 40 * & & & table 2(a ) & * 3 * & & + @xmath101 for a group of three events the half - lives as estimated according to the formulas of k .- h . schmidt et al . [ 21 ] are given .
+ @xmath97 the excitation energy in @xmath66re is not certain [ 23 ] .
+ @xmath100 the order of the decay is not known .
+ @xmath102 extrapolated value .
+ @xmath104 interpolated value .
+ @xmath117 the identification given in the table is not certain as explained in the text .
+ @xmath118 see text and table 2(b ) for the @xmath3-eneries of this band . + @xmath119 estimated excitation energy in @xmath52rn assuming e@xmath72=6.5 mev , a predicted e@xmath48 of the second minimum in @xmath78po of 6.6 mev [ 10 ] , and decay to a sdb state with a spin of 27/2 at excitation energy above the second minimum of 863 kev .
+ @xmath120 as comment ( h ) above except that e@xmath48 of the second minimum in @xmath78po was taken at 9.1 mev as predicted in ref .
+ @xmath121 see table 2(a ) for the @xmath3-energies of this band .
+ @xmath122 see text .
+ @xmath124 extrapolated value for the excitation energy of the third minimum from ref . [
25 ] is given . | the reaction @xmath0si + @xmath1ta has been studied at e@xmath2=125 and 135 mev .
coincidences between high energy particles and various x- and @xmath3-rays from abnormally long - lived states were observed .
e.g. 7.8 - 8.6 mev @xmath4-particles with @xmath3-rays of a superdeformed band , 5.1 - 5.5 mev @xmath4-particles with x- and @xmath3-rays of w , re , and pt , and 3.88 mev particles ( interpreted as protons ) with 185.8 kev @xmath3-rays .
the data are interpreted in terms of the production of long - lived ( t@xmath5 of several months ) high spin isomeric states in the second well of the potential in the parent nuclei , which decay to the normal states in the daughters , and in the third well of the potential , which decay to the second well .
pacs numbers : 23.60.+e , 21.10.re , 21.10.tg , 27.80.+w |
in two classic papers @xcite , by solving the master - slave optical system @xmath2 the author predicted that ultrafast pulses with energies lower than that required to self - sustain soliton shape in the anomalous dispersion regime of an optical medium may preserve their shape , provided an intense copropagating pump of a different color ( i.e. wavelength ) and a longer duration typical of soliton is present . since this key input as well as
the subsequent experimental evidences reported in @xcite it has become evident that besides their most common applications in long - distance transmission of high - power signals @xcite , solitons could be prime candidates for controlling light by light @xcite .
more advanced studies have been carried out in recent years based on numerical simulations , and revealed that due to their unique stability properties solitons provide excellent and reliable guiding structures for reconfigurable low - power signals and re - routinable high - power pulses @xcite .
particularly attracting is the possibility to construct soliton networks that can be used as nonlinear waveguide arrays , as established in the landscape of recent experiments @xcite emphasizing the outstanding robustness of the periodic lattices of optical solitons with flexible ( hence controllable ) refractive index modulation depth and period that are induced all - optically . in general , these periodic optical - soliton structures form in continuous nonlinear media where they develop into optically imprinted modulations with some effective refractive index . given the tunable character of the effective refractive index as well as the flexible modulation period , a great number of new opportunities for all - optical manipulation of light can be envisaged since in this case , periodic optical - soliton waveguides can operate in both weak and strong - coupling regimes depending on the depth of refractive index modulation .
transmission techniques exploiting optically - induced waveguiding configurations with solitons are now common in several communication media such as laser @xcite , photon @xcite systems and photorefractive semiconductor @xcite media .
+ the present work aims at extending the phenomenon of signal reconfiguration by strong optical fields to the context of an harmonic cw beam trapped in the guiding structure of a periodic pulse train . as in a previous study @xcite
we assume that due to the competition between the self - phase modulation ( spm ) and cross - phase modulation ( xpm ) @xcite effects , the low - power probe can be reshaped giving rise to self - sustained optical signals trapped within the guiding structure of the temporal pulse multiplex .
but before it is relavant to stress that the groundstate spectrum of the trapped probe in the case of one - soliton pump has been discussed in details @xcite .
thus , it is established both analytically and numerically that even in the steady - state regime of the pump propagation the probe groundstate is not well defined at any arbitrary value of the ratio between the cross - phase and self - phase modulation strengths but only for some specific choices of this ratio .
for one particular value of this ratio involving well defined optical structures , it has been found @xcite that the fundamental mode of the trapped probe groundstate consists of a single - pulse soliton with finite momentum .
very recently , considering the same particular value of the xpm to the spm ratio we pointed out @xcite that the physics of the pump - probe system turns to be quite rich if the pump is a periodic train of pulses , time - multiplexed at the input of an optical medium as for instance an optical fiber prior to propagation .
namely , we found that in response to this temporal pulse multiplexing in the pump the trapped probe spectrum could burst into a band reflecting distinct possible induced fundamental soliton modes .
+ still the competition between the spm and xpm effects is never of the same order and can change from one optical medium to another , implying quite distinct features and properties of the probe groundstate . in this last respect , for the value of the xpm to the spm ratio considered in our previous study we found @xcite three distinct soliton modes , and observed that they form a complete orthogonal set of @xmath0 eigenstates two of which were almost degenerate .
+ in the present study we shall go beyond previous considerations in terms of the order of the competition between the xpm and spm effects @xcite , by considering two new representative but larger values of the ratio of their strengths .
our ultimate goal through such assumptions is to point out an increasingly broadband and highly degenerate spectrum for the induced - soliton states in the probe , so broad and degenerate that the probe field can develop into a soliton quasi - continuum for sufficiently strong xpm effect relative to the spm effect .
the pump - probe system of our current interest is described by the two coupled equations @xcite : @xmath3 where ( [ a1a ] ) is the pump equation assumed to describe propagation in a kerr - type optical fiber in the anomalous dispersion regime @xcite , and ( [ a1b ] ) is the linear equation corresponding to the harmonic probe .
the quantities @xmath4 and @xmath5 are envelopes of the pump and probe respectively , @xmath6 is the group - velocity dispersion of the probe ( here taken arbitrary to account for possible difference with that of the pump ) , @xmath7 is the spm coefficient generic of nonlinearity in the pump source , @xmath8 in the second equation is the ( temporal ) walk - off between the pump and probe while @xmath9 measures the strength of the cross - phase interaction between the pump and probe . + our main point of focus is the probe equation ( [ a1b ] ) which , as it stands , requires an explicit knowledge of shape profile of the fundamental mode composing the pump signal . as we are interested in a pump consisting of a train of time - multiplexed pulses , we focus on a classic implementation where the separation between pulses in the soliton train is varied by controlling their mutual interactions , namely via the dispersion - management technique
( see more detailed discussions e.g. in @xcite ) .
it is known that such a periodic structure of equally separated pulses can well be represented by the sum : @xmath10 where , following parameter definitions in @xcite , the quantity @xmath11 represents the number of channels in the time domain , @xmath12 is the initial temporal position of a given @xmath13 pulse in the input multiplex while @xmath14 and @xmath15 are the associate amplitude and central frequency respectively .
the input time - multiplexed pulse structure ( [ a3 ] ) can be reduced to the following single - valued function , using the exact summation rule for @xmath16 functions @xcite : @xmath17e^{i\left[\omega t + \frac{\kappa^2\omega^2}{2 } z + \frac{q^2 - 2\omega^2}{2 } z\right ] } , \label{a2}\ ] ] which quite remarkably , coincides with the exact periodic - soliton solution of the pump equation ( [ a1a ] ) @xcite .
it is instructive specifying that @xmath18 in the above formula is the jacobi elliptic function of modulus @xmath19 , @xmath20 is the amplitude and @xmath21 is the characteristic pulse frequency .
the jacobi elliptic function @xmath18 is periodic in its arguments @xmath22 and for the solution ( [ a2 ] ) , the temporal period is @xmath23 where @xmath24 is the elliptic integral of the first kind . in optical communication such solution
has been claimed @xcite ) to represents a steady - state structure describing a train of pulses , which equal temporal separation coincides with the period @xmath25 of the @xmath18 function .
it is useful to end the current discussion on properties of the @xmath18 function by remarking that the one - to - one correspondance between ( [ a2 ] ) and the piecewise multi - soliton signal intensity defined in ( [ a2 ] ) , can easily be established by setting @xmath26 with @xmath27 an initial position , and constraining all pulses to have common amplitude @xmath14 ( hence common central frequency width @xmath28 ) . + now turning to the probe equation ( [ a1b ] ) we remark to start that since the nonlinearity here proceeds from the xpm coupling to the pump , it is ready to anticipate arguing that any nonlinear mode emerging in the probe should be induced by the pump signal . therefore , to ensure full account of the steady - state feature of the pump soliton as well as the temporal multiplexing at the fiber entry
, we must also express the envelope of the probe as a steady wave i.e. : @xmath29 } , \label{a4}\ ] ] where @xmath30 is the temporal core of the probe envelope which is actually trapped by the pump , @xmath31 is the wavector for the probe modulation in the pump trap while @xmath32 is a momentum - independent temporal phase shift . inserting ( [ a4 ] ) in ( [ a1b ] ) and taking @xmath33
we obtain : @xmath34u=0 , \hskip 0.3truecm \phi_t= -\frac{\lambda}{d } , \label{a5}\ ] ] @xmath35 now setting @xmath36 equation ( [ a5 ] ) becomes : @xmath37 u=0 . \label{a6}\ ] ] when @xmath38 , equation ( [ a6 ] ) reduces exactly to the associated legendre equation studied in refs .
@xcite in the context of a single - soliton pump interacting with an harmonic probe . in the next section
, we investigate spectral properties of the eigenvalue equation ( [ a6 ] ) for values of @xmath39 larger than those considered so far , in particualr for orders of the competition between the xpm and spm effects higher than the lowest order discussed in @xcite .
when @xmath42 , formula ( [ a5c ] ) yields @xmath43 and the spectral problem for the probe in the pump trap takes the explicit form : @xmath44 u=0 . \label{a6a}\ ] ] equation ( [ a6a ] ) , which is member of the so - called lam eigenvalue problem @xcite , is known to admit both extended - wave and localized - wave solutions . for the problem under consideration localized - wave solutions are best appropriate for they represent probe structures that involve low - momentum exchanges with the pump signal .
moreover , their localized shapes are key asset in our quest for stable long - standing signals with soliton features . in the specific case of equation ( [ a6a ] ) there are exactly five distinct localized modes that are induced by the soliton trap in the probe groundstate . these five modes , which can be induced simultaneously in the probe for a pump of sufficiently high intensity , form a complete set and the associate spectrum consists of the following eigenstates listed from the lowest to the highest modes ( in terms of the magntitudes of their respective modulation wavectors @xmath45 ) @xcite : + @xmath46 soliton mode : @xmath47 @xmath48 soliton mode : @xmath49 , \nonumber \\ k_2(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\,\left[2(2-\kappa^2 ) - \sqrt{1-\kappa^2(1-\kappa^2)}\right ] , \label{a7c } \end{aligned}\ ] ] @xmath50 soliton mode : @xmath51 , \nonumber \\
k_3(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\,\left[2(2-\kappa^2 ) + \sqrt{1-\kappa^2(1-\kappa^2)}\right ] , \label{a7d } \end{aligned}\ ] ] @xmath52 soliton mode : @xmath53 @xmath54 soliton mode : @xmath55 in the above formula @xmath56 and @xmath57 are jacobi elliptic functions while @xmath58 is a wavector shift associate with the temporal walk - off @xmath8 .
according to expressions of the characteristic wavectors of the five soliton modes , this wavector shift is uniform and constant so that the temporal walk - off is not relevant to the generation mechanism of the localized modes , but strictly the competition between the xpm and spm couplings .
+ in figure [ fig : one ] , we sketched profiles of the intensities @xmath59 of the five soliton modes obtained above for arbitrary values of their normalization amplitudes .
the left graphs are intensity profiles of the induced - soliton signals for @xmath60 , while the right graphs represent same quantities for @xmath38 . + + + + according to the left graphs of figure [ fig : one ] , the five localized modes have periodic - wavetrain features similar to the pump structure when @xmath61
. moreover , as in this last structure where the fundamental signal is pulse shaped , the five modes are pulse - core periodic signals though with distinct periodicities . indeed , the first ( i.e. lowest ) and fourth modes are period - one double - pulse soliton trains , the second is a period - two one - pulse soliton train whereas the third and fifth modes are period - one one - pulse soliton trains .
shapes of the fundamental solitons characterizing the five soliton modes are displayed in the right graphs of figure [ fig : one ] , they are consistent both with the general analytical expressions ( [ a7a])-([a7e ] ) when plotted for @xmath62 , and with the following explicit formula obtained from these solutions in the limit @xmath63 : + @xmath64 @xmath65 , \nonumber \\
k_2&= & k_{\lambda } + q^2d/2 , \label{a8c } \end{aligned}\ ] ] @xmath66 @xmath67 @xmath68 to give a complete description of spectral properties of the @xmath69 probe soliton modes , we also plotted the variations of their modulation wavectors with the paramater @xmath19 . in effect , the modulus @xmath19 of jacobi elliptic functions emerged above as governing the stability of wavetrain structures of both the pump and the induced probe solitons , as well as their decay toward characteristic fundamental soliton modes .
since modes are characterized by their modulation wavectors within the pump trap and given that the modulation wavector determines the amount of energy cost to the pump for mode formation and stable , we expect the variations of the different @xmath70 obtained above to provide relevant additional insight onto spectral properties of the five modes .
+ figure ( [ fig : two ] ) displays the wavector spectrum incompassing the five probe modes ( [ a7a])-([a7e ] ) , where wavectors are taken relative to the common walk - off - induced uniform shift @xmath58 of the whole spectrum as reflected by analytical expressions of the quantitites @xmath70 derived above .
spectrum of the five localized soliton eigenstates ( [ a7a])-([a7e ] ) , representing their modulation wavectors as a function of @xmath19 .
the bottom curve corresponds to the lowest soliton eigenstate and the top curve to the highest soliton eigenstate of the trapped probe.,width=288,height=240 ] two most stricking features emerging from the graph , that traduce relevant physical implications of the variation of wavectors with @xmath19 on the mode stability , are the decrease of their modulation wavectors when they decay from the wavetrain structure to their respective single fundamental - soliton states , and their degenerate features in the two limit values of @xmath19 .
recall that when @xmath19 tends to zero , the jacobi elliptic functions reduce to harmonic waves which are intrinsically linear . when @xmath72 the ratio of the xpm to the spm coupling strengths takes the value @xmath73 , corresonding to a stronger xpm effect exerted by the pump on the probe .
the resulting spectral problem for the trapped probe now reads : @xmath74 u=0 . \label{a9a}\ ] ] for this spectral problem , the localized - mode spectrum bursts into seven different eigenstates given by @xcite : + @xmath75 soliton mode : @xmath76\,sn\left(\tau - \tau_0\right ) , \nonumber \\
k_1(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\left[7 - 5\kappa^2 - 2\sqrt{4 - 7\kappa^2 + 4\kappa^4}\right ] , \label{a10a } \end{aligned}\ ] ] @xmath77 soliton mode : @xmath78\,cn\left(\tau - \tau_0\right ) , \nonumber \\ k_2(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\left[7 - 2\kappa^2 - 2\sqrt{4-\kappa^2(1-\kappa^2)}\right ] , \label{a10c } \end{aligned}\ ] ] @xmath79 soliton mode : @xmath80\,dn\left(\tau - \tau_0\right ) , \nonumber \\
k_3(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\left[10 - 5\kappa^2 - 2\sqrt{1-\kappa^2 + 4\kappa^4}\right ] , \label{a10b } \end{aligned}\ ] ] @xmath81 soliton mode : @xmath82 @xmath83 soliton mode : @xmath84\,sn\left(\tau - \tau_0\right ) , \nonumber \\
k_5(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\left[7 - 5\kappa^2 + 2\sqrt{4 - 7\kappa^2 + 4\kappa^4}\right ] , \label{a10d } \end{aligned}\ ] ] @xmath85 soliton mode : @xmath86\,cn\left(\tau - \tau_0\right ) , \nonumber \\ k_6(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\left[7 - 2\kappa^2 + 2\sqrt{4-\kappa^2(1-\kappa^2)}\right ] .
\label{a10 g } \end{aligned}\ ] ] @xmath87 soliton mode : @xmath88\,dn\left(\tau - \tau_0\right ) , \nonumber \\ k_7(\kappa)&= & k_{\lambda } + \frac{q^2d}{2}\left[10 - 5\kappa^2 + 2\sqrt{1-\kappa^2 + 4\kappa^4}\right ] , \label{a10f } \end{aligned}\ ] ] figures [ fig : three ] and [ fig : four ] depict temporal profiles of their intensities , for @xmath89 ( left graphs ) and @xmath38 ( right graphs ) .
+ + + + + like in the previous case the seven localized probe modes are dominated by periodic wavetrain profiles , consisting of sequences of pulses of different periodicities when @xmath90 . from the right graphs we find that contrary to the previous case ,
three distinct fundamental soliton modes show up when the xpm strength is higher namely single - pulse , double - pulse and tripple - pulse periodic wavetrains .
their analytical expressions are derived from ( [ a10a])-([a10f ] ) for @xmath63 and are explicitely given by : + @xmath91\,tanh\left(\tau - \tau_0\right ) , \nonumber
\\ k_1&= & k_{\lambda } , \label{a11a } \end{aligned}\ ] ] @xmath92\,sech\left(\tau - \tau_0\right ) , \nonumber \\
k_2&= & k_{\lambda } + \frac{q^2d}{2 } , \label{a11c } \end{aligned}\ ] ] @xmath93\,sech\left(\tau - \tau_0\right ) , \nonumber \\ k_3&= & k_{\lambda } + \frac{q^2d}{2 } , \label{a11b } \end{aligned}\ ] ] @xmath94 @xmath95 @xmath96 @xmath97 figure [ fig : five ] displays the spectrum of the seven localized soliton modes , more precisely here are plotted their modulation wavectors versus @xmath19 .
spectrum of the seven localized soliton eigenstates ( [ a10a])-([a10f ] ) , representing their modulation wavectors as a function of @xmath19 .
the bottom curve corresponds to the lowest soliton eigenstate and the top curve to the highest soliton eigenstate of the trapped probe.,width=288,height=240 ] like the first case studied in the previous section , the modulation wavectors decrease from a high - momentum excitation regime where the seven probe soliton modes collapse into degenerate linear waves , to a low - momentum excitation regime where they reduce to their characteristic fundamental soliton modes .
also remarkable are mode degeneracies , in this case we notice that the number of degenerate modes is higher and is expected to increase with increasing value of the ratio of the xpm to the spm coupling strengths .
the induced - phase - modulation ( or cross - phase - modulation ) process has been predicted two decades ago @xcite , and experimentally observed @xcite to occur when a low - power signal ( either an harmonic cw signal or a weakly nonlinear signal ) couples to a high - intensity pump signal , the later one providing an optical waveguiding structure and reshaping channel to the first for substantial increase of its lifetime .
however while this process is rather well understood for single - soliton pumps and in the case of relatively weak cross - phase - modulation processes , almost no insight exists about the context of pump signals consisting of periodic networks of temporal or spatial solitons @xcite .
the main objective of this work was to point out the great variety of spectral properties of pure harmonic cw probes trapped by a pump field consisting of a periodic wavetrain of optical pulses , when different orders of magnitudes of the cross - phase - modulation coupling are considered for the same self - phase - modulation coupling strength .
the principal motivation for carrying out such analysis is the broad range of technological applications behind the possibility to combine , both quantitatively and qualitatively , multi - color soliton signals generated from on single pump field . on the other hand ,
the very large diversity of physical properties expected from optical materials that can be fabricated via current technological means , makes it hard to think of physical contexts where cross - phase - modulation processes will always be far weaker than the intrinsic nonlinearity of pump fields . from these last standpoints
the present study provides new interesting insight onto the general physics of pump - probe systems .
in particular the increasingly large number and variety of possible soliton modes induced by the soliton pump for relatively strong cross - phase - modulation proceses , offers new opportunity to create multi - component vector solitons from optical low - power optical fields via light - induced waveguiding phenomena , thus enriching current designs of engineered ( i.e. artificially created ) soliton networks from reconfigurable non - soliton signals by means of all - optical techniques .
to end , we wish to underline the richness of the fundamental modes emerging from the present study with increasing cross - phase - modulation coupling strength , for a fixed strength of the self - phase - modulation effect .
while some of these modes have already been predicted @xcite , in the present and our recent @xcite works their spectral properties , conditions of existence as well as stability have been formally established .
the key insight about these fundamental modes is that they put into play multi - pulse structures that are generated from the same pump , but with different modulation wavectors and hence different instantaneous wavelengths .
work done in part at the abdus salam international centre for theoretical physics ( ictp ) trieste italy .
the author wishes to thank m. marsili and m. kiselev for their warm hospitality .
10 manassah j t 1990 _ ultrafast solitary waves sustained through induced phase modulation by a copropagating pump _ _ opt .
lett . _ 15 670 - 672 .
manassah j t 1991 _ induced waveguiding effects in a two - dimensional nonlinear medium _ _ opt .
_ 16 587 - 589
. de la fuente r , barthelemy a and froehly c 1991 _ spatial - soliton - induced guided waves in a homogeneous nonlinear kerr medium _ _ opt .
_ 16 793 - 795
. de la fuente r and barthelemy a 1992 _ spatial soliton - induced guiding by cross - phase modulation _ _ ieee j. quantum electron .
_ 28 547 - 554 .
tomlinson w j , stolen r h and johnson a m 1985 _ optical wave breaking of pulses in nonlinear optical fibers _ _ opt .
_ 10 457 - 459 .
agrawal g p , baldeck p l and alfano r r 1989 _ optical wave breaking and pulse compression due to cross - phase modulation in optical fibers _ _ opt .
_ 14 137 - 139 .
ostrovskaya e a , kivshar yu s , skryabin d v and firth w j 1999 _ stability of multihump optical solitons _ 83 296 - 299 .
shipulin a , onishchukov g and malomed b a 1997 _ suppression of soliton jitter by a copropagating support structure _ b14 3393 - 3402 .
steiglitz k and rand d 2009 _ photon trapping and transfer with solitons _
a79 0218021 - 4(r ) .
steiglitz k 2010 _ soliton - guided phase shifter and beam splitter _
a81 0338351 - 5 .
eugenieva e d , efremidis n k and christodoulides d n 2001 _ design of switching junctions for two - dimensional discrete soliton networks _
_ 26 1978 - 1980 .
fleischer j w , carmon t , segev m , efremidis n k and christodoulides d n 2003 _ observation of discrete solitons in optically induced real time waveguide arrays _ 90 0239021 - 4 .
fleischer j w , segev m , efremidis n k and christodoulides d n 2003 _ observation of two - dimensional discrete solitons in optically induced nonlinear photonic lattices _ _ nature _ 422 147 - 150 .
neshev d , sukhorukov a , kivshar yu s , ostrovskaya e and krolikowski w 2003 _ discrete solitons in light - induced index gratings _ _ opt .
_ 28 710 - 712 .
martin h , eugenieva e d , chen z and christodoulides d n 2004 _ discrete solitons and soliton - induced dislocations in partially - coherent photonic lattices _ 92 1239021 - 4 .
cartaxo a v t 1999 _ cross - phase modulation in intensity modulation - direct detection wdm systems with multiple optical amplifiers and dispersion compensators _ _ j. lightwave technol . _
17 178 - 190 .
naumov a n and zheltikov a m 2000 _ cross - phase modulation in short light pulses as a probe for gas ionization dynamics : the influence of group - delay walk - off effects _ _ laser physics _ 10 923 - 926 .
dikand a m 2010 _ fundamental modes of a trapped probe photon in optical fibers conveying periodic pulse trains _
a81 0136211 - 5 .
chen z and martin h 2003 _ waveguides and waveguide arrays formed by incoherent light in photorefractive materials _
_ optical materials _ 23 235 - 241 .
islam m n , mollenauer l f , stolen r h , simpson j r and shang h t 1987 _ cross - phase modulation in optical fibers _ _ opt . lett .
_ 12 625 - 627 .
shapiro j h and bondurant r s 2006 _ qubit degradation due to cross - phase - modulation photon - number measurement _
a73 0223011 - 4 .
shapiro j h 2006 _ single - photon kerr nonlinearities do not help quantum computation _ a73 0623051 - 11 .
lan s , delre e , chen z , shih m f and segev m 1999 _ directional coupler using soliton - induced waveguides _
_ 24 475 - 477 .
golovchenko e a , pilipetskii a n and menyuk c r 1996 _ minimum channel spacing in filtered soliton wavelength - division - multiplexing transmission
_ 21 195 - 197 .
wai p k a , menyuk c r and raghavan b 1996 _ wavelength division multiplexing in an unfiltered soliton communication system _ _ j. lightwave technol .
_ 14 1449 - 1454 .
del duce a , killey r i and bayvel p 2004 _ comparison of nonlinear pulse interactions in 160-gb / s quasi - linear and dispersion managed soliton systems _ _ j. lightwave technol .
_ 22 1263 - 1271 .
hansen e r 1975 _ a table of series and products _ ,
prentice - hill inc , englewood cliffs(n .
magnus w , oberhettinger f and tricomi f g 1953 _ handbook of transcendental functions _
( mcgraw hill , new york ) aleshkevich v , kartashov y and vysloukh v 2001 _ self - frequency shift of cnoidal waves in a medium with delayed nonlinear response _
b18 1127 - 1136 .
kartashov y , aleshkevich v , vysloukh v , egorov a a and zelenina a s 2003 _ stability analysis of @xmath98-dimensional cnoidal waves in media with cubic nonlinearity _
e67 0366131 - 11 .
dikand a m 1999 _ bound states in one - dimensional klein - gordon systems admitting periodic - kink soliton excitations _ 60 291 - 293 .
arscott f m 1964 , _ periodic differential equations : an introduction to mathieu , lam and allied functions
_ ( pergamon press ltd ) | it has been established for some time that high - power pump can trap a probe beam of lower intensity that is simultaneously propagating in a kerr - type optical medium , inducing a focusing of the probe with the emergence of modes displaying solitonic properties . to understand the mechanism by which such self - sustained modes are generated , and mainly the changes on probe spectrum induced by the cross - phase - modulation effect for an harmonic probe trapped by a multiplex of temporal pulses , a linear equation ( for the probe ) and a nonlinear schrdinger equation ( for the pump ) both coupled by a cross - phase - modulation term , are considered simultaneously . in general
the set of coupled probe - pump equations is not exactly tractable at any arbitrary value of the ratio of the cross - phase to the self - phase modulation strengths .
however , for certain values of this ratio , the probe modulation wavector develops into @xmath0 _ quantum states _ involving @xmath1 soliton - shaped eigenfunctions which spectral properties can be characterized unambiguously .
solutions of the probe equation give evidence that the competition between the self - phase and cross - phase modulations leads to a broadband spectrum , with the possibility of a quasi - continuum of soliton modes when the cross - phase - modulation coupling is strong enough . |
the path - integral formulation of quantum mechanics is the foundation of many numerical methods that allow one to study with great accuracy the rich physics of interacting quantum systems . at finite temperature ,
a path - integral monte carlo ( pimc ) technique for continuous systems has been developed and applied by ceperley and pollock.@xcite recently , this approach has been renovated in a new class of methods known as _ worm algorithms_.
@xcite the zero - temperature counterparts of the pimc algorithm are the reptation quantum monte carlo ( rqmc ) @xcite and the path - integral ground state ( pigs ) methods , @xcite which have been demonstrated useful to simulate coupled electron - ion systems , @xcite as well as to infer spectral properties from imaginary time dynamics .
@xcite a number of important physical problems particularly in the fields of strongly correlated fermions and cold atoms can be fruitfully modeled by lattice hamiltonians .
a first application of path - integral techniques to ( boson ) lattice models was proposed by krauth _
et al . _ in 1991 .
@xcite few other attempts to apply pimc to lattice models have been made ever since , with a recent application of the rqmc idea to the quantum dimer model hamiltonian .
@xcite in this paper , we propose a new method that generalizes and improves the approach of ref . in several ways .
our method is based on continuous - time random walks and is therefore unaffected by time - step errors .
inspired by the work of syljuasen and sandvik @xcite and rousseau , @xcite we adopt a generalization of the _ bounce algorithm _ of pierleoni and ceperley , @xcite called _ directed updates _
, which helps reducing the correlation time in path sampling .
we also introduce a worm - algorithm based method to calculate _
pure _ expectation values of arbitrary off - diagonal observables , which are generally out of the scope of existing lattice ground - state methods .
the resulting algorithm naturally applies to fermions , using the fixed - node approximation .
a technique to improve systematically upon this approximation is proposed , based on the calculation of a few moments of the hamiltonian .
our methodology is demonstrated by a few case studies on the one - dimensional heisenberg and the two - dimensional fermion hubbard models .
this paper is organized as follow : in sec .
[ sec : formalism ] we present the general formalism of ground - state pimc for lattice models ; in sec .
[ sec : reptation ] our implementation of the rqmc algorithm on a lattice is presented . in particular , we give a detailed account of the above mentioned _ directed update _ technique ( sec .
[ sub : directed - updates ] ) and of the continuous - time propagator ( sec .
[ subsec : contin ] ) ; in sec .
[ subsec : offdiag ] , we introduce an extension of the algorithm to cope with off - diagonal observables , while in sec .
[ subsec : signpr ] a further extension to systems affected by sign problems is presented , including a strategy to improve systematically upon the fixed - node approximation .
[ sec : results ] contains a few case applications , including the simulation of the spectral properties and spin correlations of the one - dimensional heisenberg model and the calculation of the ground - state energies of the fermionic hubbard model with a significantly better accuracy than that achieved by the fixed - node approximation . in sec . [
sec : conc ] we finally draw our conclusions .
let us consider a generic lattice hamiltonian @xmath0 and a complete and orthogonal basis set , whose states are denoted by @xmath1 . given the generic wave function @xmath2 , its amplitude on the configuration @xmath1 will be denoted by @xmath3 , namely @xmath4 .
the exact ground - state wave function @xmath5 can be obtained by the imaginary time evolution of a given variational state @xmath6 : @xmath7 where @xmath8 , provided that the variational state is non - orthogonal to @xmath5 , i.e. , @xmath9 .
then , the ground - state expectation value of a quantum operator @xmath10 can be obtained by @xmath11 a practical computational scheme can be conveniently introduced by considering a path - integral representation of the imaginary time evolution .
to such a purpose , we split the total imaginary time @xmath12 into @xmath13 slices of `` duration '' @xmath14 , in such a way that the value of the evolved wave function on a generic many - body state of the system reads @xmath15 where we have introduced the imaginary time propagators @xmath16 within this approach , it is easy to write expectation values of operators @xmath10 that are diagonal in the chosen basis @xmath1 , i.e. , @xmath17 .
in fact , in this case we have that : @xmath18 where the summation is extended to all possible imaginary time paths @xmath19 , and @xmath20 is given by : @xmath21\psi_{v}(x_{2m}).\label{eq : pibeta}\ ] ] the ground - state energy can be conveniently obtained by means of the _ mixed average _ : @xmath22 where @xmath23 is the so - called local energy .
besides the static ( i.e. , equal - time ) correlation functions , this formalism allows one to calculate also dynamical correlations in imaginary time @xmath24 that can be computed as @xmath25 where @xmath26 and @xmath27 are two coordinates of the path such that @xmath28 .
a probabilistic interpretation of the previous expectation values ( [ eq : obeta ] ) , ( [ eq : energybeta ] ) , and ( [ eq : cabtau ] ) can be immediately recovered whenever @xmath29 for all configurations @xmath30 .
indeed , in this case , @xmath20 can be interpreted as a probability distribution that may be readily sampled by using monte carlo algorithms .
this fact allows ground - state expectation values to be calculated exactly , within statistical errors .
the basic idea of the rqmc algorithm is to sample the distribution probability @xmath20 by using a markov process with simple moves . given the configuration @xmath19 , a new configuration is proposed in two possible ways : either @xmath31 ( which we call `` left move '' ) or @xmath32 , ( which we call `` right move '' ) . in both cases , @xmath33 is a new configuration proposed according to a suitable transition probability @xmath34 , where @xmath35 stays for @xmath36 ( @xmath37 ) when the left ( right ) move is considered .
such `` sliding moves '' are depicted in fig .
[ fig : reptmov ] . ideally , the transition probability should guarantee the minimum possible statistical error on the desired observables and , to such a purpose , it has been proved useful to consider the propagator with importance sampling , i.e. , @xmath38 and take the following transition probability @xmath39 where @xmath40 represents the normalization factor .
the explicit form of @xmath34 will be discussed in more detail in sec .
[ subsec : contin ] .
the proposed configuration @xmath41 ( where @xmath42 or @xmath43 ) is accepted or rejected according to the usual metropolis algorithm , where the acceptance rate is given by : @xmath44 in this way , a sequence of configurations @xmath45 is generated , @xmath46 being the ( discrete ) time index of the markov chain .
pictorial representation of the `` sliding moves '' along the right imaginary time direction . in the new configuration ( bottom ) , a new head for the reptile is generated from the old configuration ( top ) and the tail is discarded . ] in order to reduce the auto - correlation time of the observables it is convenient to make several consecutive sliding moves along the same imaginary time direction .
@xcite to such a purpose , a recent development called `` bounce '' algorithm has been proposed .
@xcite although the bounce algorithm sampling procedure does not fulfill the microscopic detailed balance , the equilibrium probability @xmath20 is correctly sampled .
@xcite the rqmc algorithm with bounce moves can be then summarized in the following steps : 1 .
for the current direction of the move and for the present configuration @xmath45 , propose @xmath33 according to the transition probability @xmath34 , where @xmath47 if @xmath42 and @xmath48 if @xmath49 .
2 . given the form of the acceptance ratio @xmath50 of eq .
( [ eq : acceptance ] ) , accept the proposed configuration according to the probability @xmath51 if @xmath42 , or with probability @xmath52 if @xmath49 .
3 . if the move is accepted , update the path configurations according to @xmath53 and continue along the same direction , otherwise @xmath54 and change direction .
4 . go back to 1 .
at this point we introduce a novel alternative sampling approach , which generalizes the bounce idea while strictly fulfilling the detailed balance condition .
such a scheme , which is largely inspired by the _ loop algorithm _ methods devised for the stochastic series expansion @xcite and for the _ worm algorithm _ , @xcite allows one to choose the time direction in a purely markovian way , i.e. , independently of the previous history . in our algorithm ,
a markov step consists of many simple consecutive `` sliding moves '' , whose number is not fixed _ a - priori _ but is determined by a certain probability ( see below ) . the actual monte carlo step takes place at the end of few consecutive updates along the currently chosen direction . in the examples below , we denote the number of these sliding moves between two monte carlo steps by @xmath55 . at the beginning of each markov
step we choose a direction @xmath56 according to the probability @xmath57 , whose form will be specified later .
assuming that the right direction has been chosen , we propose a new configuration @xmath33 , according to the transition probability @xmath58 and the configuration labels are shifted according to @xmath59 , with @xmath60 . at this point , we continue updates along this direction with probability @xmath61 , or stop with probability @xmath62 $ ] .
if it has been decided to continue the updates , then a new configuration is generated according to @xmath63 and the labels of the configuration are again shifted according to @xmath64 .
the markov step finishes after @xmath55 consecutive updates along the right direction only when @xmath65 , where @xmath66 is a random number uniformly distributed in @xmath67 . at this point a metropolis test should be done , in order to accept or reject the sequence of intermediate @xmath55 sliding moves : @xmath68 where ( see appendix [ append ] ) @xmath69 however , in order to avoid time - consuming restorations of the original configuration , it is preferable to accept all the moves , while keeping track of the residual weight @xmath70 .
this is possible since @xmath50 only depends upon initial and final configurations , so that , given that all the intermediate moves are accepted , the sampled distribution probability is @xmath71 .
the contribution of the current configuration to statistical averages must be then weighted by the factor @xmath72 . to proceed to the next markov step ,
a new direction @xmath56 is chosen according to @xmath73 and @xmath74 and the updates are carried along the extracted new direction .
let us now show the actual expressions for the afore - mentioned probabilities . in appendix [ append ] , it is demonstrated that the detailed balance is satisfied if one chooses the probabilities for the directions as @xmath75 where @xmath76 which is positive and , therefore , guarantees that the above defined quantities are well defined probabilities , i.e. , @xmath77 and @xmath78 , with the additional property that @xmath79 .
regarding the probabilities to continue the updates along the current direction , we have a substantial freedom of choice , provided that the condition @xmath80 , is satisfied ( see appendix [ append ] ) .
in this paper we have adopted @xmath81 where we have defined : @xmath82 and @xmath83 is an arbitrary parameter of the algorithm , which controls the average number of consecutive updates along the same direction . summarizing , the rqmc algorithm with directed updates consists of a sequence of markov steps determined by the following rules : 1 .
choose a time direction @xmath56 according to the probabilities of eqs .
( [ eq : pleft ] ) and ( [ eq : prig ] ) .
2 . propose a new configuration @xmath33 according to the transition probability @xmath34 , where @xmath47 if @xmath42 and @xmath48 if @xmath49 .
3 . shift the configuration indexes according to @xmath84 if @xmath42 or @xmath59 if @xmath49 .
4 . according to the probability @xmath85 or @xmath86 , decide whether keep moving in the same direction or change direction . in the former case ,
go to 2 , otherwise go to 5 .
the markov step ends here and the current configuration carries the weight @xmath87 , where @xmath55 is the number of intermediate moves along the direction chosen . the relationship between the directed update scheme and the bounce algorithm is further elucidated in appendix [ sec : on - the - relationship ] , where general considerations about the efficiency of the algorithms are presented .
one of the most striking differences between the original formulation of the rqmc on the continuum and the present formulation on the lattice is the lack of the discretization error appearing in the trotter decomposition of the propagator .
indeed it is easier to carry the propagation in continuous imaginary time on a lattice , @xcite than on the continuum .
@xcite to such a purpose , let us consider the limit of an infinitesimal imaginary time @xmath88 , for which the transition probability of eq .
( [ eq : timpsa ] ) can be written as @xmath89-\epsilon\left[h_{xy}\frac{\psi_{v}(y)}{\psi_{v}(x)}\right]+o(\epsilon^{2}),\label{eq : timpsast}\end{aligned}\ ] ] where @xmath90 is the previously defined local energy and @xmath91 denotes the matrix element of the hamiltonian .
whenever @xmath92 is _ non positive _ for all @xmath35 and @xmath93 , this equation takes the form of a continuous - time markov process , whose analytical properties are well known . in particular , the probability distribution for the `` waiting time '' @xmath94 in a given state @xmath35 , i.e. , the average time that the system spends in the state @xmath35 before making an off - diagonal transition to another state @xmath95 , is exactly known , namely @xmath96\}$ ] . as a consequence
, the finite - time propagator @xmath97 can be directly sampled , giving rise to a succession of a certain number @xmath98 of consecutive transitions @xmath99 , with corresponding waiting times @xmath100 ( such that @xmath101 ) .
the normalization of the whole process is @xmath102,\label{eq : ctw}\ ] ] where the waiting times are extracted according to the exponential probability @xmath103 .
the transitions between the intermediate configurations are done according to the off - diagonal elements of eq .
( [ eq : timpsast ] ) , i.e. , @xmath104 is chosen with probability proportional to @xmath105 $ ] .
the formalism so - far developed allows one to successfully compute _ pure _ ground - state expectation values of operators that are _ diagonal _ in the local basis @xmath35 , with the expectation values of off - diagonal operators restricted to the so - called _ mixed averages_.
@xcite nonetheless , it is often of great interest to remove such a limitation ( whose result is biased by the quality of the variational wave function ) and a dedicated sampling strategy has to be devised in order to cope with such a need . in the following ,
we show that a relatively easy modification of the sampling scheme can accomplish this task , providing us with a general tool to compute ground - state averages of operators that are non local in the chosen basis @xmath35 .
let us consider an arbitrary off - diagonal observable @xmath106 , which can be in turn considered as the summation of many observables we are interested in , i.e. , @xmath107 .
for example , we can imagine these operators to be the components of the one - body density matrix at a given distance , @xmath108 with the summation extended to all lattice coordinates at a fixed distance @xmath56 .
in the spirit of refs we introduce a _ worm - operator _ defined by @xmath109 where @xmath110 is a positive constant , and consider the extended configuration space spanned by the paths @xmath111 the extended paths are broken in two ( space)-discontinuous pieces by the worm operator , which is placed at an imaginary - time @xmath112 .
therefore , paths contain @xmath113 coordinates , including @xmath114 and @xmath115 that refer to the same imaginary time @xmath116 .
the configuration space spanned by eq .
( [ eq : piworm ] ) is clearly larger than the one spanned by eq .
( [ eq : pibeta ] ) , which is recovered whenever @xmath117 , i.e. , when the worm operator is _
diagonal_. the pure ground - state expectation value of the operator @xmath106 is conveniently written in terms of the extended paths as @xmath118 where @xmath119 whenever condition @xmath120 is satisfied .
the modulus of eq .
( [ eq : piworm ] ) can be in turn interpreted as a probability distribution and stochastically sampled by means of the elementary sliding moves considered before .
indeed , whenever the worm operator is far from the ends of the imaginary - time paths , the sampling scheme remains unchanged .
in this case , a move along direction @xmath56 will generate a new head ( or tail ) for the reptile according to @xmath34 while shifting the worm position of @xmath121 . on the other hand , whenever the worm operator reaches the ends of the reptile , a new worm configuration is proposed on the other side ; in analogy with the previous analysis , new configurations are generated according to a transition probability @xmath122 where @xmath123 is the normalization factor .
due to the particular form of the matrix elements ( [ eq : wormopera ] ) , the transition probability will lead either to diagonal configurations @xmath124 or to off - diagonal configurations @xmath125 , thus generating continuous and discontinuous paths . the relative probability for diagonal and off - diagonal configurations
depends on the value of @xmath110 that can be tuned in order to reach a balanced sampling frequency for the different sectors of the extended paths . in order to exemplify the worm updates ,
let us consider the case in which @xmath49 and a configuration @xmath126\psi_{v}\left(x_{2m+1}\right)$ ] , after a sliding update in the right direction , we will have @xmath127\mathcal{w}_{x_{2m+1}x_{t}}\psi_{v}\left(x_{t}\right)$ ] , where @xmath33 is proposed according to the transition probability @xmath128 ( see fig .
[ fig : reptmovworm ] ) . in analogy with the previous case ,
the acceptance factor for the bounce moves reads @xmath129 . pictorial representation of the `` sliding moves '' along the right imaginary - time direction when the worm operator sits at the tail of the reptile .
in the new configuration ( bottom ) , a new head for the reptile is generated from the old configuration ( top ) , the old tail configuration is discarded and the worm discontinuity is moved to the `` neck '' of the reptile . ] summarizing , the rqmc with worm - updates consists of the following steps : 1 . for the current direction of the move @xmath56 and for the present configuration @xmath45 consider the worm - operator position @xmath116 .
if the worm is not at the ends of the reptile ( i.e. , @xmath130 when @xmath42 and @xmath131 when @xmath49 ) go to step ( a ) , otherwise go to step ( b ) .
1 . propose a new configuration @xmath33 according to the transition probability @xmath34 , where @xmath47 if @xmath42 and @xmath132 if @xmath49 .
the new configuration is accepted with probability @xmath133 if @xmath42 , or with probability @xmath134 if @xmath49 . in the proposed state @xmath41 ,
all the configuration labels are shifted in the @xmath56 direction , determining in turn a shift of the worm operator of a time interval @xmath121 , depending on @xmath56 .
2 . propose a new configuration @xmath33 according to the worm transition probability @xmath135 , where @xmath136 if @xmath42 and @xmath137 if @xmath49 .
accept the new configuration with probability @xmath138 if @xmath42 , or with probability @xmath139 if @xmath49 . in the proposed state @xmath41 ,
all the configuration labels are shifted in the @xmath56 direction , and the worm operator is moved from the head ( tail ) to the tail ( head ) of the reptile , depending on @xmath56 .
3 . if the move is accepted , update the path configurations according to @xmath53 and continue along the same direction , otherwise @xmath54 and change direction .
4 . go back to 1 .
this scheme samples the probability density associated to the modulus of eq .
( [ eq : piworm ] ) , and the expectation values of the individual components @xmath140 can be recast as statistical averages over such a probability distribution , while keeping track of the overall sign of the extended paths .
in particular the best estimate of the ground - state expectation values is obtained when the worm is in the central part of the path , at @xmath141 , leading to @xmath142\right\rangle _ { \text{od}}^{\text{center}}}{n_{d}^{\text{center}}},\label{eq : ostatoffdiag}\end{aligned}\ ] ] where @xmath143 denotes statistical averages over the off - diagonal distribution @xmath144 and @xmath145 is the number of configurations sampled with a diagonal worm operator in the center of the paths . when the probability distribution of eq .
( [ eq : pibeta ] ) is not positive defined , as is generally the case with fermions , the probabilistic interpretation of the imaginary time paths breaks down .
this circumstance , which is known as `` sign problem '' , originates whenever @xmath146 for some element @xmath147 . in this case , it is not possible to have polynomial algorithms that are able to obtain an _ exact _ solution of the problem , which would imply to sample correctly the resulting signs .
therefore , approximated schemes are welcome and often adopted , the most widespread one being the so - called fixed - node ( fn ) approximation . for lattice systems
, this approach relies on the definition of an effective hamiltonian , which depends parametrically on the nodal structure of a _ given _ variational wave function @xmath148 .
@xcite the matrix elements of the fn hamiltonian are defined as @xmath149 where the sign - flip potential is @xmath150 , the sum being extended to all the sign - flip states defined by the condition @xmath151 . with such a choice , the transition matrix @xmath152 of eq .
( [ eq : timpsast ] ) is always positive definite and the summation of eq .
( [ eq : psibeta ] ) is now restricted which results in the fn approximation to a region of the hilbert space in which imaginary time paths are positive definite .
therefore , within the fn approximation , the ground - state wave function @xmath153 of @xmath154 can be stochastically sampled without any sign problem .
moreover , it is easy to show that the fn approximation becomes exact whenever the signs of the exact ground state are known .
most importantly , it has been proven @xcite that the fn ground - state energy @xmath155 gives a rigorous upper - bound to the exact ground - state one and improves the pure variational results . at this point
, we introduce a straightforward , although computationally expensive , way to improve further the fn energy .
our strategy amounts to compute the expectation values of arbitrary powers of the original hamiltonian @xmath0 on the fn ground state @xmath156 , namely @xmath157 the fn ground state can be expanded in the basis set of the eigenstates of @xmath0 as @xmath158 and @xmath159 , with @xmath160 .
since very often the fn wave function has a considerable overlap with only few low - energy states , the knowledge of the first few moments of the hamiltonian are enough to approximately reconstruct both the coefficients @xmath161 and the energies @xmath162 .
to such a purpose , let us consider a typical situation in which only the first @xmath163 moments of the hamiltonian have been numerically calculated and are therefore known .
we can then truncate the expansion for @xmath164 to the order @xmath165 having a closed system of @xmath163 equations @xmath166 for @xmath167 that can be solved for the unknowns @xmath168 and @xmath169 . in the limit of large @xmath98 , the approximated @xmath170 converges to the exact ground - state energy .
moreover , we verified that @xmath171 , as a result of a connection between the solutions of the eq .
( [ eq : truncl ] ) and the lanczos procedure written in terms of the moments of the hamiltonian .
@xcite the hamiltonian moments are off - diagonal operators and can , in principle , measured according to the sampling procedure detailed in sec .
[ subsec : offdiag ] . in the present implementation
we are able to achieve sufficient statistical accuracy only for the first moment of the hamiltonian , i.e. , @xmath172 , while higher moments are too noisy . yet , to our knowledge our algorithm is the only one that allows the calculation of the expectation value of the _ original _ hamiltonian @xmath0 .
this is known @xcite to be a better upper bound than the expectation value of the fn hamiltonian accessible with other zero - temperature algorithms .
although we are not currently in position to measure directly the hamiltonian moments @xmath164 we have a controlled access to the _ mixed averages _ @xmath173 which present optimal statistical uncertainty
moreover , an improved estimate of the ground - state energy based on the knowledge of the first few moments @xmath174 can be obtained solving a system of equation similar to eq .
( [ eq : truncl ] ) that leads to the approximate ground - state energies @xmath175 .
unfortunately , the proof that @xmath176 , for @xmath177 , is far from being trivial , requiring a generalization of the already non - trivial upper bound for @xmath178 described in ref . .
nonetheless , we have numerically verified that , in all the cases treated in this paper ( where @xmath179 is _ a - priori _ known ) , the condition @xmath176 is always verified .
we are then led to conjecture that this may always be the case .
hereafter , we present a simple application of the previous ideas to sign - problem free spin hamiltonians .
let us consider the one - dimensional quantum heisenberg model @xmath180 where @xmath181 is the spin 1/2 operator on the site @xmath182 and @xmath183 is the nearest - neighbor super - exchange coupling .
lowest - energy excitations as a function of the wave - vector @xmath184 for an @xmath185 heisenberg chain .
the energies are extracted from the dynamical structure factor @xmath186 and are compared to exact results by the lanczos method . ] the same as in fig .
[ fig : sq20 ] for @xmath187 .
exact results are given by bethe ansatz .
] the total number of sites is denoted by @xmath188 and periodic - boundary conditions are assumed .
this model can be solved exactly by using the so - called bethe ansatz technique .
@xcite information on the excitation spectrum can be obtained from the dynamical structure factor @xmath189 where @xmath190 is the fourier transform of time - evolved spin projection on the z - axis . by introducing a complete set of eigenstates of the hamiltonian @xmath191 with eigenvalues @xmath192
, we have that @xmath193 where @xmath194 . in the thermodynamic limit
, the spin-1 states form a branch , which is very similar to spin waves in standard ordered systems , although no long - range order is found in one dimension .
imaginary time correlation functions of arbitrary ( diagonal ) operators can be efficiently evaluated via eq .
( [ eq : cabtau ] ) .
this fact allows us to have a direct access to @xmath195
. this imaginary time correlation function can be then analytically continued , by using the maximum - entropy method , @xcite in order to have a reasonable numerical estimate for the dynamical structure factor of eq .
( [ eq : sqomega ] ) . before presenting the results ,
let us mention that we consider the following jastrow state as a variational wave function : @xcite @xmath196|fm\rangle\ ] ] where @xmath197 is the ferromagnetic state along the x direction , for which @xmath198 does not depend upon the spin configuration and the variational parameters @xmath199 are optimized by using the method of ref . . in fig
[ fig : sq20 ] , we show the results for a small @xmath185 system , where exact diagonalizations are possible by using the lanczos method .
we report the energy excitations @xmath200 for the lowest state with @xmath201 and fixed momentum @xmath184 . in this case a perfect agreement between our rqmc results and the exact ones is found .
moreover , also on larger systems a very good accuracy is possible ( see fig . [
fig : sq80 ] ) , demonstrating the performances of our numerical algorithm .
ground - state expectation value of the spin - spin correlation function @xmath202 for the heisenberg model on a @xmath203-site chain . ] in order to exemplify the potentialities of the scheme outlined in [ subsec : offdiag ] , we conclude this part of the results devoted to the heisenberg model showing the ground - state expectation value of the spin - spin correlation at distance @xmath56 @xmath204 the desired observable is used as a worm operator and the value of the correlation function at the various distances is computed by means of the estimator of eq . .
[ fig : corrheis ] , we show the expectation value of @xmath202 for a @xmath203-site one - dimensional lattice . in this case
we are able to achieve very good statistics for the off - diagonal observable , with a relatively negligible computational effort , when compared to the evaluation of the ground - state expectation value of other diagonal observables . as an example of the application of the rqmc to sign - problem affected hamiltonians , we present some results for the fermionic hubbard model on a square lattice , defined by : @xmath205 where @xmath206 indicate nearest - neighbor sites , @xmath207 ( @xmath208 ) creates ( destroys ) an electron on the site @xmath182 with spin @xmath209 , and @xmath210 .
as a variational state we consider @xmath211|fs\rangle\ ] ] where @xmath212 is the non - interacting fermi sea and the jastrow factor involves density - density correlations .
the variational parameters @xmath199 entering in the jastrow factor may be optimized again by minimizing the variational energy with the method of ref . .
in order to avoid open shells in @xmath212 , we consider @xmath213-degrees tilted lattices with @xmath214 sites , such that both the half - filled case and selected holes - doped cases are closed shells .
ground - state energy for the fermionic hubbard model at half filling on a @xmath215-sites tilted - square lattice .
the energy difference @xmath216 is computed with distinct approximations described in the text . ]
let us start by showing the results for @xmath215 electrons on @xmath215 sites , where lanczos diagonalizations are possible .
@xcite in fig .
[ fig:18sites ] , we report our results for the ground - state energy .
the fn approach gives rather accurate results for small values of @xmath217 , i.e. , @xmath218 , where @xmath219 . by increasing the on - site interaction ,
the fn approach becomes worse and worse .
this fact is due to the choice of the variational wave function that does not contain antiferromagnetic order .
remarkably , a considerable improvement may be obtained by considering the _ pure _ expectation value of the hamiltonian , which is systematically lower than the fn energy , as demonstrated in ref . and now accessible within our algorithm .
further improvements to the fn energy can be obtained upon considering few ( up to three ) higher moments of the hamiltonian measured as mixed - averages , see fig .
[ fig:18sites ] . the scheme based upon the hamiltonian moments ( described in sec .
[ subsec : signpr ] ) allows us to reach a great accuracy for the ground state energy , with a residual error almost independent of @xmath217 . indeed , in this way we have @xmath220 up to @xmath221 .
this approach remains very effective also for larger systems , even though the variational wave function loses accuracy by increasing the cluster size ( because the ground state has antiferromagnetic order in the thermodynamic limit , while the variational state is paramagnetic ) . in table
[ tab:50sites ] , we report the ground - state energy for @xmath222 sites for the half - filled case , while in table [ tab:50sitesholes ] we report the ground - state energies for selected cases at finite hole - doping , where numerically exact results ( for moderate values of @xmath223 and moderate lattice sizes ) can be obtained by the auxiliary - field monte carlo method .
@xcite .[tab:50sites ] ground - state energy as a function of the hubbard @xmath223 repulsion on the @xmath222-site lattice at half filling . [ cols="^,^,^,^",options="header " , ]
in this paper we have provided an efficient and general formulation of the reptation quantum monte carlo technique on lattice models . in particular , we showed an alternative sampling approach which generalizes the bounce algorithm , previously introduced to reduce auto - correlation time of the observables .
our scheme allows one to choose the time direction in a purely markovian way .
in addition , the average number of consecutive moves along the time directions may be optimized by a fine tuning of a certain parameter that has been expressly introduced in the transition probabilities .
we reported benchmarks for two different models with pure bosonic and fermionic degrees of freedom , by showing to what extent it is possible to have accurate results both on the ground state and low - energy excitations .
the introduction of a general method to compute ground - state expectation values of arbitrary off - diagonal observables also constitutes an important achievement , which will ease the study of relevant properties such as bose - einstein condensation and superconductivity phenomena in strongly interacting models .
in addition , the possibility to directly measure the _ pure _ ground - state expectation values may open the way to a better optimization of the correlated wave function associated to the ground - state of an effective hamiltonian which is not the fn one . it is a pleasure to acknowledge here precious discussions with s. sorella and a. parola .
we also acknowledge support from cineca and cofin 07 .
in this appendix we give a detailed derivation of the probabilities for the directed updates .
the detailed balance condition guarantees that the given probability distribution @xmath20 is sampled if transitions from an initial state @xmath45 to a final state @xmath224 differing for @xmath55 intermediate updates are accepted according to : @xmath225 @xmath226 being the overall transition probability between the two states .
let us first consider the case when @xmath227 and fix the right direction @xmath49 ( a similar derivation can be obtained for @xmath42 ) . in this case , the transition probability from the initial state to the final state reads @xmath228,\end{aligned}\ ] ] namely , it is the product of the probability of having chosen the right direction , times the transition probability for the new tail of the reptile , times the probability of stopping the updates after one intermediate step .
the inverse transition probability instead reads @xmath229,\end{aligned}\ ] ] which can be obtained reversing the time directions and considering transitions from the head of the reptile instead that from the tail .
therefore , the acceptance factor reads as @xmath230 for two intermediate transitions instead the transition probabilities are @xmath231,\end{aligned}\ ] ] and @xmath232,\end{aligned}\ ] ] leading to the acceptance factor @xmath233 the generalization to generic @xmath55 intermediate steps is straightforward and can be written as@xmath234\right\ } .\label{eq : asexp}\end{gathered}\ ] ] to find a simple solution for the unknown probabilities , we first impose a cancellation for the intermediate acceptance factors , namely @xmath235 this condition is satisfied by eqs .
( [ eq : pkeeplefa ] ) and ( [ eq : pkeepriga ] ) .
then , we notice that the acceptance factor can be written only in terms of the final and the initial states as @xmath236 further , we can impose the two factors @xmath184 to be independent on the direction , i.e. , the condition @xmath237 , which is satisfied if @xmath238 since the two time directions are mutually exclusive , it is also true that @xmath79 , which allows us to solve eq .
( [ eq : psatk ] ) and obtain eqs .
( [ eq : pleft ] ) and ( [ eq : prig ] ) .
the same reasoning can be repeated for the left direction and , due to imposed homogeneity for the probabilities , it can be checked that the detailed balance is satisfied for the left direction too .
in this appendix we comment on the relationship between the directed - update scheme and the bounce algorithm . if @xmath239 is taken in eqs .
( [ eq : pkeeplefa ] ) and ( [ eq : pkeepriga ] ) , then after @xmath55 updates along the direction @xmath56 , at the end of the markov step @xmath240 , i.e. , the next markov step will be taken in the opposite direction , just like the bounce algorithm .
although the two algorithms are similar in this particular limit , there is an important difference which eventually leads to a different computational efficiency . in order to elucidate this point and to show the @xmath241-dependence of the efficiency of the directed updates , we have done a systematic comparison of the two algorithms . in particular
, we have compared the efficiency of the directed updates with the bounce algorithm for a one - dimensional heisenberg model .
the computational efficiency is generally defined as @xmath242 where @xmath243 is the square of the statistical error associated to a given observable after a given computational time @xmath244 . in fig .
[ fig : effic ] , we show the ratio between the directed - update scheme efficiency over the bounce algorithm efficiency , for the measurement of the ground - state energy of a one - dimensional chain .
relative efficiency of the directed update scheme and the bounce algorithm .
the measured quantity is the ground - state energy of the one - dimensional heisenberg model on a chain of size @xmath187 sites . ]
we notice that the two sampling schemes have comparable performances , being both based on a similar approach .
as anticipated , it clearly emerges from fig .
[ fig : effic ] that the two algorithms do not have exactly the same behavior at @xmath239 , the maximum efficiency of the directed updates being reached for lower values of @xmath241 .
this feature is due to the fact that when @xmath241 is very close to its saturation value , then a single markov step can consist of a conspicuous number of individual `` sliding moves '' . even if this situation leads to a fast decorrelation of configurations it also leads to a rarefaction of the possibility to measure the desired observables , which can eventually take place only at the end of the markov step and not during the individual moves .
this leads to a worse efficiency if compared to the bounce algorithm , where measurements can be in principle done after every sliding move . in conclusion ,
the performances of the two algorithms are very close , although some advantages may arise from the use of the directed - updates .
we further notice that the purely markovian approach introduced in this paper could be slightly more efficient in cases where the number of rejected configurations by the bounce algorithm is substantial whereas _ all _ the generated configurations are accepted in the directed update scheme . | we provide an extension to lattice systems of the reptation quantum monte carlo ( rqmc ) algorithm , originally devised for continuous hamiltonians . for systems affected by the sign problem , a method to systematically improve upon the so - called fixed - node approximation
is also proposed .
the generality of the method , which also takes advantage of a canonical worm algorithm scheme to measure off - diagonal observables , makes it applicable to a vast variety of quantum systems and eases the study of their ground - state and excited - states properties . as a case study
, we investigate the quantum dynamics of the one - dimensional heisenberg model and we provide accurate estimates of the ground - state energy of the two - dimensional fermionic hubbard model . |
transport properties of deterministic chaotic hamiltonian dynamics can exhibit a rich variety of phenomena . in many situations where chaos is strong ,
transport through phase space tends to be uniform and can imitate random processes of the brownian type . the motion on average is a normal diffusion .
however , in more generic situations where regular as well as chaotic behaviour coexist the phase space has a fractal structure , and this structure can induce important variations of the transport process . in this paper
our purpose is to investigate transport properties of a deterministic two - dimensional area preserving map whose dynamics can be either mixed or strongly chaotic as a parameter is varied .
it is defined by the time - dependent hamiltonian @xmath0 called the kicked harper model .
it arises in several branches of physics , as for example in the description of charged particles interacting with an electrostatic wave packet in a transverse uniform magnetic field @xcite , in the study of the two - dimensional motion of electrons in a periodic potential plus uniform perpendicular magnetic field @xcite and in connection to the steady - state dynamics of two and three - dimensional ideal incompressible fluids @xcite .
a characteristic feature of harper - like hamiltonians is the particular functional form of the kinetic energy .
the equations of motion induced by eq.([1 ] ) can be integrated over one period @xmath1 of the kicking potential , leading to the map @xmath2 where @xmath3 .
we can consider the motion either in the entire phase - space plane or in the fundamental domain @xmath4 .
the existence of a fundamental domain comes from the fact that the map ( [ 2 ] ) is invariant under translations @xmath5 where @xmath6 are two arbitrary positive or negative integers since such a shift of the initial conditions produces a global shift of the trajectory . in the limit
@xmath7 the model reduces to the continuous time evolution of the harper hamiltonian @xmath8 which , from the dynamical point of view , is an integrable system . for simplicity , we restrict the following discussion to the special case @xmath9 ( @xmath10 )
. then the phase space plots of @xmath11 are characterized by the existence of a separatrix defined by @xmath12 connecting both sides of the fundamental domain . in the full phase plane
, the separatrix forms a regular lattice having a square symmetry ( cf fig.1(a ) ) , while the contour lines @xmath13 are closed curves . for @xmath14 the separatrix
is destroyed and becomes a mesh of finite thickness inside which the dynamics is chaotic .
this chaotic mesh , which has been called the stochastic web @xcite , coexists with regions of phase space where regular motion takes place .
as @xmath15 increases the stochastic web grows , and for sufficiently large values of @xmath15 the web disappears and chaotic dynamics covers most of phase space .
this transition is shown in fig.1 for the motion restricted to the fundamental domain . in order to better visualize the geometry of the stochastic web , in fig.2 we have plotted again fig.1(b ) but now in the full phase plane .
the motion of the particle is not chaotic for any initial condition inside the stochastic web , since there are regular islands on it ( clearly visible in fig.2 ) .
however , if an initial condition corresponding to a chaotic motion on the web is chosen then , because the web extends over the whole phase - space plane , the particle can `` diffuse '' arbitrarily far from its initial position . in the @xmath16 direction , the diffusion will be characterized by the coefficient @xmath17 where the average is taken over a set @xmath18 of phase - space initial conditions .
this coefficient can in general be expressed as an infinite series involving the force - force correlation functions ( see @xcite and eq.([7 ] ) below ) .
the existence of @xmath19 ( i.e. , the convergence of the series ) and therefore the existence of a true deterministic diffusive process is in general difficult to prove rigorously .
this , however , has been done for a different map , the sawtooth map @xcite . moreover ,
several aspects of the diffusion process in deterministic maps have been studied in the past by truncation of the series and/or adding some noise .
these studies were mainly concentrated on the standard map @xcite or on different types of stochastic webs @xcite . in particular ,
computations of @xmath19 for the kicked harper map can be found in @xcite .
our aim here is to compute the diffusion coefficient as a function of @xmath15 up to second order in the correlation functions and to analyze its dependence on @xmath15 in terms of the underlying classical dynamics ( section iii ) .
we obtain , for all @xmath15 , a finite value of @xmath19 which is in very good agreement with the numerical results except near some parameter ranges where accelerator modes appear .
these modes are stable periodic orbits which coherently propagate across the web .
the shortest periodic orbits corresponding to accelerator modes as well as their stability are computed in section ii .
the existence of these orbits produces a divergence of the diffusion coefficient if the average in eq.([3 ] ) is taken over all possible initial conditions .
this is because in the presence of accelerator modes the average can be splitted into two parts : the first is over the stable phase - space regions @xmath20 where accelerator modes exist , while the second is over the chaotic component @xmath21 @xmath22 where the index @xmath23 means that the average has been taken over the whole fundamental domain .
because accelerator modes propagate in a ballistic manner , @xmath24 and the diffusion coefficient associated to this region diverges like @xmath25 .
the presence of accelerator modes can have more unexpected consequences on the transport properties of the system than the one just mentioned .
their influence extends into the chaotic sea and may produce non - local effects .
in fact , the partial barriers ( cantori ) induced by their presence in the chaotic regions of the web may modify the nature of the transport in the stochastic web as well . for most of the time
the particle diffuses , but sometimes it is trapped in the neighbourhood of an accelerator modes and drag along in a ballistic manner .
these rare events can dominate the dynamics and produce anomalous diffusion of the lvy type @xcite @xmath26 as a consequence the diffusion coefficient associated to the chaotic component also has a power - law divergence with exponent @xmath27 .
anomalous diffusion was already observed numerically in different hamiltonian systems , like the standard map @xcite , in stochastic webs @xcite and in two - dimensional smooth potentials @xcite .
there are however few systematic quantitative results . at present
there is no theory making the connexion between the existence of phase space barriers in the chaotic layers on the one hand and anomalous diffusion of the lvy type on the other hand . only in one case it was possible to establish @xcite a direct connexion between some parameter of a map and the temporal exponent @xmath28 determining the anomalous diffusion .
but unfortunately this was done in the simple case of a one - dimensional iterative map . in sect.iii
we compute , as a function of @xmath15 , the exponent @xmath28 .
our numerical results , presented in fig.5 and fig.6 , demonstrate that the kicked harper map provides an additional example of anomalous diffusion of the lvy type in hamiltonian systems .
different aspects of this exponent in connexion to the existence of normal modes are discussed .
the period-@xmath29 fixed points of the map ( [ 2 ] ) are determined by the condition @xmath30 , @xmath31 .
we are interested however in the fixed points of the map on the fundamental domain , and thus we identify phase - space points which differ from an integer value : @xmath32 , @xmath33 leading to @xmath34 where @xmath35 and we have used the notation @xmath36 , @xmath37 .
these two integers are the winding numbers of the period-@xmath29 fixed points since they define the number of turns the orbit does around the fundamental domain in one iteration of the map . in the phase plane , periodic orbits with winding numbers @xmath38 propagate ballistically , with @xmath39 and @xmath40 .
but not all of them correspond to accelerator modes either because they are not stable or because they do not propagate ( @xmath41 ) . for simplicity
, we again restrict the discussion to @xmath42 .
for a fixed value of @xmath15 , the solutions of eqs.([5 ] ) are characterized by all integers such that @xmath43 $ ] , where the square brackets mean integer part @xcite . for each possible set @xmath38 of winding numbers
there are four different solutions in @xmath44 : @xmath45 , @xmath46 , @xmath47 and @xmath48 , where @xmath49 and @xmath50 . for @xmath51
only @xmath52 is a solution and as a consequence there are no period-@xmath29 accelerator modes in this parameter range since these orbits do no propagate .
as @xmath15 increases , at each integer value there is a first - order bifurcation ( fold catastrophe ) where several new orbits appear . taking into account all possible new couples of winding numbers and their signs ( where at least one is equal to @xmath53 $ ] while the other
one can be less than @xmath53 $ ] ) , there are at each integer value of @xmath15 a total number of @xmath54 $ ] new orbits appearing .
this makes , for a given value of @xmath15 , a total number of @xmath55 + 2)^2 $ ] period-@xmath29 fixed points .
these orbits are stable if @xmath56 and unstable otherwise , where @xmath57 is the monodromy matrix . for a given couple @xmath38
the trace is given by @xmath58 with @xmath59 . at any integer value of @xmath15 , @xmath60 for all the periodic orbits emerging at that value of the parameter .
@xmath61 is positive for half of them , and negative for the other half . for the latter half
, it follows from ( [ 6 ] ) that @xmath62 as @xmath15 increases from the bifurcation point . on the contrary , those having @xmath63 are stable as @xmath15 increases from @xmath53 $ ] but as soon as @xmath64 they become unstable too .
this range of stability defines the domain of existence in parameter space of period-@xmath29 accelerator modes .
its width is in fact relatively small ( compare to one ) , and its size decreases with @xmath53 $ ] .
for example in the particular case @xmath65 $ ] , @xmath66 at @xmath67 ^ 2 + 1/\pi^2)^{1/2 } \approx
[ \gamma]+1/(2 \pi^2 [ \gamma])$ ] .
at the value @xmath68 where @xmath66 there is generically a period doubling bifurcation , where a period-@xmath69 stable fixed point of the map appears . if @xmath15 further increases , this orbit will soon become unstable through another bifurcation giving rise to the usual cascade of bifurcations .
we are not going to establish here a complete classification of the period-@xmath69 fixed points , but briefly discuss a special family who plays an important role in the variations of the diffusion coefficient .
these are the fixed points that emerge at half integer values , @xmath70 .
their point of emergence is independent of @xmath15 and given by @xmath71 and @xmath72 with winding numbers @xmath73 and @xmath74 , respectively . as for the period-@xmath29 orbits ,
their position and stability varies with @xmath15 and there is a small range in parameter space close to @xmath75 in which some of them are stable .
accelerator modes of period @xmath69 therefore exist around all half integer values of @xmath15 .
the analysis of the accelerator modes can be pursued for arbitrary high periods , with increasing complexity .
from eqs.([2 ] ) it follows that @xmath76 and then @xmath77 \ .\ ] ] our purpose is to compute the diffusion coefficient ( [ 3 ] ) averaged over all possible initial conditions @xmath78 in the fundamental domain . by definition @xmath79
it is therefore implicit when computing any average value that the coordinates and momenta @xmath80 are expressed , using the equations of motion ( [ 2 ] ) , as a function of @xmath78 . because @xmath81 and introducing the force - force correlation function
@xmath82 it follows that @xmath83 \ .\ ] ] where we have used the invariance of @xmath84 under time translations .
if the @xmath84 fall off rapidly enough with @xmath1 , then @xcite @xmath85 \ .\ ] ] if correlations are neglected , i.e. the @xmath86 are considered as random uncorrelated variables , then it follows from eq.([2 ] ) that the motion in the @xmath16-direction reduces to a random walk with length - step @xmath87 .
accordingly , we obtain from eq.([7 ] ) @xmath88 .
this is expected to be a good approximation for large values of the @xmath15 s where the `` randomization '' is more effective .
from this point of view , the corrections for @xmath89 in eq.([7 ] ) can be considered as a large-@xmath15 expansion .
the following relations are needed in order to compute the correlation functions @xmath90 the one - step correlator is @xmath91\rangle_{r_f } \end{array}\ ] ] from eqs.([8 ] ) and simple trigonometric relations the averages can be computed explicitly .
one finds @xmath92 the computation of the correlator for points differing by two time steps @xmath93\sin\left [ 2 \pi q_0 + 2\pi\gamma_2\sin(2\pi p_0 - 2\pi\gamma_1\sin(2\pi q_0))\right]\rangle_{r_f } \end{array}\ ] ] follows the same lines , with increasing algebraic complexity .
we get @xmath94 j_n^2 ( 2 \pi\gamma_2 ) \ .\ ] ] from the computation of @xmath95 we only keep the term @xmath96 , the other being of higher order in powers of bessel functions . then , from eqs.([9 ] ) and ( [ 10 ] ) we finally have @xmath97 \\ & & -4 \displaystyle \sum_{n=1}^\infty\left [ ( -1)^{n+1 } j_0 ( 2 n\pi\gamma_1 ) + j_2 ( 2 \pi n \gamma_1)\right ] j_n^2 ( 2 \pi\gamma_2 ) \ .
\end{array}\ ] ] in order to compare this result with the actual properties of the transport , we have numerically computed the quantity @xmath98 as a function of @xmath99 . for each value of @xmath15 ,
the diffusion coefficient was calculated from the slope of @xmath100 for @xmath101 averaged over @xmath102 different initial conditions taken at random on the elementary cell .
the values obtained are plotted in fig.3 , together with the analytical result ( [ 11 ] ) .
the approximation ( [ 11 ] ) clearly accounts for the main features of the diffusion process .
indeed , for most values of @xmath15 , the numerical points are in very good agreement with the finite value obtained in ( [ 11 ] ) .
the oscillating behaviour of period @xmath103 can be physically related to the strong modifications induced in the transport properties by the appearance of period-@xmath29 accelerator modes at each integer value of @xmath15 . for large values of @xmath15
the oscillations are more regular with decreasing amplitude @xmath104 . for small values of @xmath15
the phase space structure is more rich and accordingly @xmath19 is less regular . in this regime
the agreement with eq.([11 ] ) is less good , as shown in the inset of fig.3 , and below @xmath105 the theoretical value is definitely wrong ( indeed , from eq.([11 ] ) we get @xmath106 , while the numerical values tend to @xmath107 ) . to compute more accurately the diffusion coefficient in this limit ,
we make use of the results obtained in @xcite . in the limit @xmath108 the diffusion coefficient of the stochastic web takes the form @xmath109 } \ , \;\;\;\;\;\;\ ; \gamma \ll 1\ ] ] where @xmath110 .
because our computations are averaged over all possible initial conditions , we shall multiply eq.([12 ] ) by the relative area @xmath111 of the stochastic web @xcite @xmath112 with the result @xmath113\right\ } } \ , \;\;\;\;\;\;\ ; \gamma \ll 1 \ .\ ] ] this function is plotted at the origin in the inset of fig.3 , and is in very good agreement with the numerical values . on the other hand
, the numerical results do not agree with the theory close to low integer and half integer values of @xmath15 , where some of the numerical points lie systematically higher .
if the number of iterations employed in order to compute the slope is increased , these points are not stable but rather move up , a fact clearly pointing out towards a divergence of the diffusion coefficient .
analytically , the divergence should come out from further correlation terms in the expansion ( [ 7 ] ) @xcite .
as mentioned in the introduction , this divergent behaviour is closely related to the appearance of accelerator modes .
because of their stability , these ballistic modes occupy a ( small but ) finite phase - space volume .
it then follows that @xmath114 and @xmath115 as @xmath25 when @xmath116 since the average was computed over the whole fundamental domain ( cf eq.([3p ] ) ) . for large integer or half integer values of @xmath15
the divergence is not visible numerically because the area and range of existence of the accelerator modes is very small ( cf section ii ) .
one could think that the divergence of the transport coefficient is limited to stable regions of phase space where accelerator modes exist .
this in fact is not true .
when this modes are present , two rather different dynamics coexist . on the one hand
, there is the ballistic motion associated with the small stable regions .
on the other hand there is the chaotic sea ( occupying most of phase - space as soon as @xmath117 ) , where incoherent motion takes place and where we may expect ( as in the absence of accelerator modes ) a regular brownian - type diffusion .
but the frontier between both types of motion is not sharp .
stable islands are surrounded by broken tori ( called cantori ) . these broken tori play
the role of partial barriers for the motion in the chaotic sea .
when the particle is trapped by a cantorus in the neighborhood of the accelerator mode , it is dragged across the phase plane ballistically .
these rare events can however dominate the dynamics and produce an anomalous diffusion @xcite .
if the trapping time probability distribution has the asymptotic behaviour @xmath118 then we can classify the dynamics into three distinct regimes @xcite @xmath119 the presence of accelerator modes may then induce strong modifications into the transport properties of the chaotic component .
this happens when the exponent @xmath28 lies between @xmath29 and @xmath69 .
the anomalous trajectories consists of a regular diffusion process punctuated by intermittent ballistic episodes .
we find that this type of transport occurs in the kicked harper map in certain parameter ranges .
as an example , in fig.4 we have plotted @xmath120 iterations of a trajectory lying in the chaotic component obtained for @xmath121 . the orbit is shown in the entire phase plane as well as in the fundamental domain .
the former picture stresses the ballistic lvy - type intermittent behaviour of the orbit , while the latter makes clear the trapping of the orbit around the accelerator modes .
in fact , the long step propagating coherently over more than @xmath122 fundamental domains is clearly visible in part ( b ) of that figure as a concentration of points around the stable region associated with the period-@xmath29 orbit emerging at @xmath123 at @xmath124 . in order to make more quantitative these considerations ,
we have numerically computed the exponent @xmath28 as a function of @xmath15 for the chaotic component of phase space .
this was done propagating for each value of @xmath15 @xmath102 trajectories iterated @xmath120 times .
the average over the trajectories was taken in order to compute from eq.([4 ] ) the exponent @xmath28 .
the initial conditions of these trajectories were taken at random in a small region of size @xmath125 around the phase - space point @xmath126 .
this choice ensures that the orbits lie in the chaotic component of phase space .
the results obtained are plotted in fig.5 for @xmath127 , where it is clearly visible that values of @xmath28 greater than one are obtained for values of @xmath15 slightly higher than @xmath128 and @xmath29 .
this confirms the existence of anomalous diffusion .
however , it is remarkable that in spite of the fully developed fractal structure of phase space existing for @xmath51 , we observe anomalous diffusion only in a very localized region around @xmath129 .
could be that looking with a smaller step in @xmath15 other peaks will emerge . in fig.6
the two peaks of fig.5 are amplified .
the function @xmath130 in the second peak is a strongly oscillating function , and for @xmath131 the anomalous diffusion disappears .
this value is in good agreement with the value @xmath132 obtained in section ii for the disappearance of the ( last ) period-@xmath29 accelerator mode emerging at @xmath123 .
the normal diffusion process of a deterministic two - dimensional area - preserving map can thus be very well described in terms of a truncated expansion of the diffusion coefficient .
the theoretical prediction works well even in parameter ranges where the fractal phase - space structure of the classical dynamics is strong ( mixed regime ) .
however , when stable periodic orbits propagating coherently through the chaotic sea are present , the diffusion coefficient diverges .
this divergence is trivial in the sense that it is directly related to the existence of a finite and stable ballistic region of phase space .
it is not accounted for by the truncated expansion but should be recovered if additional terms are added @xcite .
the presence of stable coherent orbits may also modify the transport properties of the chaotic sea itself .
we find that in certain parameter regions surrounding the emergence of accelerator modes the diffusion is anomalous , with @xmath133 in eq.([4 ] ) ( and ( [ 14 ] ) ) .
the puzzling fact is that this is not always true .
for example , we were not able to detect any anomalous diffusion for parameter values close to ( and slightly higher than ) @xmath134 , where we know that period-@xmath69 accelerator modes emerge , but we do observed anomalous diffusion at @xmath135 . this strengthens the need to develop a quantitative and predictive theory for the anomalous diffusion in hamiltonian systems , which does not exist today .
g. m. zaslavsky _
et al _ , _ sov .
jetp * 64 * ( 1986 ) 294 - 303 . g. m. zaslavsky , _ chaos _ * 1 * ( 1991 ) 1 - 12 .
i. dana , _ phys .
a _ * 197 * ( 1995 ) 413 .
p. lebuf _
et al _ , _ phys .
_ * 65 * ( 1990 ) 3076 ; _ chaos _ * 2 * ( 1992 ) 125 - 130 .
g. m. zaslavsky , r. z. sagdeev and a. a. chernikov , _ sov .
_ jetp * 67 * ( 1988 ) 270 - 277 .
r. a. pasmanter , _ phys . rev . a _ * 42 * ( 1990 ) 3622 - 3625 .
j. r. cary and j. d. meiss , _ phys .
* 24 * ( 1981 ) 2664 - 2668
. b. v. chirikov , _ phys .
_ * 52 * ( 1979 ) 265
. a. b. rechester and r. b. white , _ phys .
* 44 * ( 1980 ) 1586 - 1589
. h. d. i. abarbanel , _ physica d _ * 4 * ( 1981 ) 89 .
a. j. lichtenberg and b. p. wood , _ phys .
a _ * 39 * ( 1989 ) 2153 - 2159 . v. v. afanasiev _ et al _ , _ phys .
lett . a _ * 144 * ( 1990 ) 229 - 236 .
i. dana and m. amit , _ phys . rev .
* 51 * ( 1995 ) r2731-r2734 .
b. mandelbrot , _ the fractal geometry of nature _ , freeman , san francisco , 1982 ; e. montroll and m. schlesinger in _ studies in statistical mechanics _ ,
vol.11 , j. leibowitz and e. montroll eds . , north - holland , amsterdam , 1984 . c. f. f. karney , _ physica d _ * 8 * ( 1983 ) 360 .
b. v. chirikov and d. shepelyansky , _ physica d _ * 13 * ( 1984 ) 394 .
r. ishizaki _ et al _ , _ prog .
* 84 * ( 1990 ) 179 ; * 85 * ( 1991 ) 1013 .
g. zumofen and j. klafter , _ europhys .
_ * 25 * ( 1994 ) 565 - 570 .
a. a. chernikov _
et al _ , _ phys .
a _ * 144 * ( 1990 ) 127 - 133 . v. v. afanasiev , r. sagdeev and g. m. zaslavsky , _ chaos _ * 1 * ( 1991 ) 143 - 159 .
d. k. chaikovsky and g. m. zaslavsky , _ chaos _ * 1 * ( 1991 ) 463 - 472 . t. geisel and s. thomae , _ phys . rev . lett .
_ * 52 * ( 1984 ) 1936 - 1939 .
the complex solutions of eq.([5 ] ) as well as their role in the quantum mechanical transport of the model was analyzed in p. lebuf and a. mouchet , _ phys .
_ * 73 * ( 1994 ) 1360 - 1363 .
m. f. shlesinger , g. m. zaslavsky and j. klafter , _ nature _ * 363 * ( 1993 ) 31 - 37 .
3 : normalized diffusion coefficient as a function of @xmath15 . analytical curve ( solid line ) ; numerical results ( dots ) .
inset : amplification in the range @xmath140 ; the full curve at the origin represents eq.([13 ] ) . | chaotic deterministic dynamics of a particle can give rise to diffusive brownian motion .
in this paper , we compute analytically the diffusion coefficient for a particular two - dimensional stochastic layer induced by the kicked harper map .
the variations of the transport coefficient as a parameter is varied are analyzed in terms of the underlying classical trajectories with particular emphasis in the appearance and bifurcations of periodic orbits .
when accelerator modes are present , anomalous diffusion of the lvy type can occur .
the exponent characterizing the anomalous diffusion is computed numerically and analyzed as a function of the parameter .
ipno / th 96 - 50 novembre 1996 + |
in compressive imaging @xcite , we aim to estimate an image @xmath0 from @xmath1 noisy linear observations @xmath2 , @xmath3 assuming that the image has a representation @xmath4 in some wavelet basis @xmath5 ( i.e. , @xmath6 ) containing only a few ( @xmath7 ) large coefficients ( i.e. , @xmath8 ) . in ( [ eqn : system ] ) , @xmath9 is a known measurement matrix and @xmath10 is additive white gaussian noise .
though @xmath11 makes the problem ill - posed , it has been shown that @xmath12 can be recovered from @xmath13 when @xmath7 is adequately small and @xmath14 is incoherent with @xmath5 @xcite .
the wavelet coefficients of natural images are known to have an additional structure known as _ persistence across scales _ ( pas ) @xcite , which we now describe . for 2d images ,
the wavelet coefficients are naturally organized into quad - trees , where each coefficient at level @xmath15 acts as a parent for four child coefficients at level @xmath16 .
the pas property says that , if a parent is very small , then all of its children are likely to be very small ; similarly , if a parent is large , then it is likely that some ( but not necessarily all ) of its children will also be large .
several authors have exploited the pas property for compressive imaging @xcite .
the so - called `` model - based '' approach @xcite is a deterministic incarnation of pas that leverages a restricted union - of - subspaces and manifests as a modified cosamp @xcite algorithm .
most approaches are bayesian in nature , exploiting the fact that pas is readily modeled by a _ hidden markov tree _ ( hmt ) @xcite .
the first work in this direction appears to be @xcite , where an iteratively re - weighted @xmath17 algorithm , generating an estimate of @xmath12 , was alternated with a viterbi algorithm , generating an estimate of the hmt states .
more recently , hmt - based compressive imaging has been attacked using modern bayesian tools @xcite .
for example , @xcite used markov - chain monte - carlo ( mcmc ) , which is known to yield correct posteriors after convergence . for practical image sizes , however , convergence takes an impractically long time , and so mcmc must be terminated early , at which point its performance may suffer .
variational bayes ( vb ) can sometimes offer a better performance / complexity tradeoff , motivating the approach in @xcite .
our experiments indicate that , while @xcite indeed offers a good performance / complexity tradeoff , it is possible to do significantly better . in this paper , we propose a novel approach to hmt - based compressive imaging based on loopy belief propagation @xcite . for this , we model the coefficients in @xmath18 as conditionally gaussian with variances that depend on the values of hmt states , and we propagate beliefs ( about both coefficients and states ) on the corresponding factor graph .
a recently proposed `` turbo '' messaging schedule @xcite suggests to iterate between exploitation of hmt structure and exploitation of observation structure from ( [ eqn : system ] ) . for the former we use the standard sum - product algorithm @xcite , and for the latter we use the
recently proposed _ approximate message passing _
( amp ) approach @xcite .
the remarkable properties of amp are 1 ) a rigorous analysis ( as @xmath19 with @xmath20 fixed , under i.i.d gaussian @xmath14 ) @xcite establishing that its solutions are governed by a state - evolution whose fixed points when unique yield the true posterior means , and 2 ) very low implementational complexity ( e.g. , amp requires one forward and one inverse fast - wavelet - transform per iteration , and very few iterations ) .
we consider two types of conditional - gaussian coefficient models : a bernoulli - gaussian ( bg ) model and a two - state gaussian - mixture ( gm ) model .
the bg model assumes that the coefficients are either generated from a large - variance gaussian distribution or are exactly zero ( i.e. , the coefficients are exactly sparse ) , whereas the gm model assumes that the coefficients are generated from either a large - variance or a small - variance gaussian distribution .
both models have been previously applied for imaging , e.g. , the bg model was used in @xcite , whereas the gm model was used in @xcite . although our models for the coefficients @xmath18 and the corresponding hmt states involve statistical parameters like variance and transition probability , we learn those parameters directly from the data .
to do so , we take a hierarchical bayesian approach similar to @xcite where these statistical parameters are treated as random variables with suitable hyperpriors .
experiments on a large image database show that our turbo - amp approach yields state - of - the - art reconstruction performance with substantial reduction in complexity .
the remainder of the paper is organized as follows .
section [ sec : model ] describes the signal model , section [ sec : alg ] describes the proposed algorithm , section [ sec : sims ] gives numerical results and comparisons with other algorithms , and section [ sec : conc ] concludes .
_ notation _ : above and in the sequel , we use lowercase boldface quantities to denote vectors , uppercase boldface quantities to denote matrices , @xmath21 to denote the identity matrix , @xmath22 to denote transpose , and @xmath23 .
we use @xmath24 to denote the probability density function ( pdf ) of random variable @xmath25 given the event @xmath26 , where often the subscript `` @xmath27 '' is omitted when there is no danger of confusion .
we use @xmath28 to denote the @xmath29-dimensional gaussian pdf with argument @xmath12 , mean @xmath30 , and covariance matrix @xmath31 , and we write @xmath32 to indicate that random vector @xmath12 has this pdf .
we use @xmath33 to denote expectation , @xmath34 to denote the probability of event @xmath35 , and @xmath36 to denote the dirac delta .
finally , we use @xmath37 to denote equality up to a multiplicative constant .
throughout , we assume that @xmath5 represents a 2d wavelet transform @xcite , so that the transform coefficients @xmath38^t$ ] can be partitioned into so - called `` wavelet '' coefficients ( at indices @xmath39 ) and `` approximation '' coefficients ( at indices @xmath40 ) . the wavelet coefficients can be further partitioned into several quad - trees , each with @xmath41 levels ( see fig .
[ fig : quadtree ] ) .
we denote the indices of all coefficients at level @xmath42 of these wavelet trees by @xmath43 , where @xmath44 refers to the root . in the interest of brevity , and with a slight abuse of notation
, we refer to the approximation coefficients as level `` @xmath45 '' of the wavelet tree ( i.e. , @xmath46 ) .
as discussed earlier , two coefficient models are considered in this paper : bernoulli - gaussian ( bg ) and two - state gaussian mixture ( gm ) .
for ease of exposition , we focus on the bg model until section [ sec : gm ] , at which point the gm case is detailed . in the bg model , each transform coefficient @xmath47
is modeled using the ( conditionally independent ) prior pdf @xmath48 where @xmath49 is a hidden binary state .
the approximation states @xmath50 are assigned the apriori activity rate @xmath51 , which is discussed further below . meanwhile ,
the root wavelet states @xmath52 are assigned @xmath53 . within each quad - tree , the states have a markov structure . in particular , the activity of a state at level @xmath54 is determined by its parent s activity ( at level @xmath15 ) and the transition probabilities @xmath55 , where @xmath56 denotes the probability that the child s state equals @xmath57 given that his parent s state also equals @xmath57 , and @xmath58 denotes the probability that the child s state equals @xmath59 given given that his parent s state also equals @xmath59 .
the corresponding factor graph is shown in fig .
[ fig : fga1 ] .
[ l][bl][0.9]@xmath60 [ l][bl][0.9]@xmath61 [ l][bl][0.9]@xmath62 [ l][bl][0.9]@xmath63 [ l][bl][0.9]@xmath64 [ l][bl][0.9]@xmath65 [ l][bl][0.9]@xmath66 [ l][bl][0.9]@xmath67 [ l][bl][0.9]@xmath68 [ l][bl][0.9]@xmath69 [ l][bl][0.9]@xmath70 [ l][bl][0.9]@xmath71 [ l][bl][0.9]@xmath72 [ l][bl][0.9]@xmath73 [ l][bl][0.9]@xmath74 [ l][bl][0.9]@xmath75 [ l][bl][0.9]@xmath76 [ l][bl][0.9]@xmath77 [ l][bl][0.9]@xmath78 [ l][bl][0.9]@xmath79 [ l][bl][0.9]@xmath80 [ l][bl][0.9]@xmath81 [ l][bl][0.9]@xmath82 [ l][bl][0.9]@xmath83 [ r][b][0.9]@xmath84 [ r][b][0.9]@xmath85 [ r][b][0.9]@xmath86 [ l][bl][0.9]@xmath87 [ l][bl][0.9]@xmath88 [ l][bl][0.9]@xmath89 [ l][bl][0.9]@xmath90 [ b][b][0.8]observation structure [ b][b][0.8]support structure we take a hierarchical bayesian approach , modeling the statistical parameters @xmath91 as random variables and assigning them appropriate hyperpriors . rather than working directly with variances , we find it more convenient to work with precisions ( i.e. , inverse - variances ) such as @xmath92 .
we then assume that all coefficients at the same level have the same precision , so that @xmath93 for all @xmath94 . to these precisions
, we assign conjugate priors @xcite , which in this case take the form @xmath95 where @xmath96 for @xmath97 , and where @xmath98 are hyperparameters .
( recall that the mean and variance of @xmath99 are given by @xmath100 and @xmath101 , respectively @xcite . ) for the activity rates and transition parameters , we assume @xmath102 where @xmath103 , and where @xmath104 are hyperparameters .
( recall that the mean and variance of @xmath105 are given by @xmath106 and @xmath107 , respectively @xcite . )
our hyperparameter choices are detailed in section [ sec : sims ] .
to infer the wavelet coefficients @xmath18 , we would ideally like to compute the posterior pdf @xmath108 & = & \sum_{{\ensuremath{\boldsymbol{s}}}}\underbrace{p({\ensuremath{\boldsymbol{s}}})}_{\displaystyle \triangleq h({\ensuremath{\boldsymbol{s } } } ) } \prod_{n=1}^n \underbrace{p(\theta_n { \,|\,}s_n)}_{\displaystyle \triangleq f_n(\theta_n , s_n ) } \prod_{m=1}^m \underbrace{p(y_m { \,|\,}{\ensuremath{\boldsymbol{\theta}}})}_{\displaystyle \triangleq g_m({\ensuremath{\boldsymbol{\theta } } } ) } , \quad \label{eqn : post2}\end{aligned}\ ] ] where @xmath37 denotes equality up to a multiplicative constant .
for the bg coefficient model , @xmath109 is specified by ( [ eqn : spike_slab ] ) . due to the white gaussian noise model ( [ eqn : system ] )
, we have @xmath110 , where @xmath111 denotes the @xmath112 row of the matrix @xmath113 . while exact computation of @xmath114 is computationally prohibitive , the marginal posteriors @xmath115 can be efficiently approximated using _ loopy belief propagation _ ( lbp ) @xcite on the factor graph of fig .
[ fig : fga1 ] , which uses round nodes to denote variables and square nodes to denote the factors in ( [ eqn : post2 ] ) . in doing so
, we also obtain the marginal posteriors @xmath116 . for now , we treat statistical parameters @xmath117 , as if they were fixed and known , and we detail the procedure by which they are learned in section [ sec : update ] . in lbp
, messages are exchanged between the nodes of the factor graph until they converge .
messages take the form of pdfs ( or pmfs ) , and the message flowing to / from a variable node can be interpreted as a local belief about that variable . according to the _ sum - product algorithm _
@xcite the message emitted by a variable node along a given edge is ( an appropriate scaling of ) the product of the incoming messages on all other edges .
meanwhile , the message emitted by a function node along a given edge is ( an appropriate scaling of ) the integral ( or sum ) of the product of the node s constraint function and the incoming messages on all other edges , where the integration ( or summation ) is performed over all variables other than the one directly connected to the edge along which the message travels .
when the factor graph has no loops , exact marginal posteriors result from two ( i.e. , forward and backward ) passes of the sum - product algorithm @xcite .
when the factor graph has loops , however , exact inference is known to be np hard @xcite and so lbp is not guaranteed to produce correct posteriors . still
, lbp has shown state - of - the - art performance in many applications , such as inference on markov random fields @xcite , turbo decoding @xcite , ldpc decoding @xcite , multiuser detection @xcite , and compressive sensing @xcite .
with loopy belief propagation , there exists some freedom in how messages are scheduled . in this work
, we adopt the `` turbo '' approach recently proposed in @xcite . for this
, we split the factor graph in fig .
[ fig : fga1 ] along the dashed line and obtain the two decoupled subgraphs in fig .
[ fig : turbo ] .
we then alternate between belief propagation on each of these two subgraphs , treating the likelihoods on @xmath118 generated from belief propagation on one subgraph as priors for subsequent belief propagation on the other subgraph .
we now give a more precise description of this turbo scheme , referring to one full round of alternation as a `` turbo iteration . '' in the sequel , we use @xmath119 to denote the message passed from node @xmath120 to node @xmath121 during the @xmath122 turbo iteration .
[ l][bl][0.8]@xmath123 [ l][bl][0.8]@xmath84 [ l][bl][0.8]@xmath85 [ l][bl][0.8]@xmath124 [ l][bl][0.8]@xmath68 [ l][bl][0.8]@xmath69 [ l][bl][0.8]@xmath70 [ l][bl][0.8]@xmath125 [ l][bl][0.8]@xmath76 [ l][bl][0.8]@xmath77 [ l][bl][0.8]@xmath78 [ l][bl][0.8]@xmath126 [ l][bl][0.8]@xmath60 [ l][bl][0.8]@xmath61 [ l][bl][0.8]@xmath62 [ l][bl][0.8]@xmath127 [ l][bl][0.8]@xmath128 [ l][bl][0.8]@xmath129 [ l][bl][0.8]@xmath130 [ l][bl][0.8]@xmath131 [ l][bl][0.8]@xmath132 [ l][bl][0.8]@xmath133 [ l][bl][0.8]@xmath134 [ l][bl][0.8]@xmath135 [ ] [ b][1.0]@xmath90 [ b][b][0.8]amp the procedure starts at @xmath136 by setting the `` prior '' pmfs @xmath137 in accordance with the apriori activity rates @xmath138 described in section [ sec : model ] .
lbp is then iterated ( to convergence ) on the left subgraph in fig .
[ fig : turbo ] , finally yielding the messages @xmath139 .
we note that the message @xmath140 can be interpreted as the current estimate of the likelihood would be referred to as the `` extrinsic '' information about @xmath141 produced by the left `` decoder '' , since it does not directly involve the corresponding prior @xmath142 .
similarly , the message @xmath143 would be referred to as the extrinsic information about @xmath141 produced by the right decoder . ] on @xmath141 , i.e. , @xmath144 as a function of @xmath141 .
these likelihoods are then treated as priors for belief propagation on the right subgraph , as facilitated by the assignment @xmath145 for each @xmath146 . due to the tree structure of hmt , there are no loops in right subgraph ( i.e. , inside the `` @xmath123 '' super - node in fig .
[ fig : turbo ] ) , and thus it suffices to perform only one forward - backward pass of the sum - product algorithm @xcite .
the resulting leftward messages @xmath147 are subsequently treated as priors for belief propagation on the left subgraph at the next turbo iteration , as facilitated by the assignment @xmath148 .
the process then continues for turbo iterations @xmath149 , until the likelihoods converge or a maximum number of turbo iterations has elapsed .
formally , the turbo schedule is summarized by @xmath150 in the sequel , we refer to inference of @xmath118 using compressive - measurement structure ( i.e. , inference on the left subgraph of fig . [
fig : turbo ] ) as _ soft support - recovery _ ( ssr ) and inference of @xmath118 using hmt structure ( i.e. , inference on the right subgraph of fig .
[ fig : turbo ] ) as _ soft support - decoding _ ( ssd ) .
ssr details are described in the next subsection .
we now discuss our implementation of ssr during a single turbo iteration @xmath151 .
because the operations are invariant to @xmath151 , we suppress the @xmath151-notation .
as described above , ssr performs several iterations of loopy belief propagation per turbo iteration using the fixed priors @xmath152 .
this implies that , over ssr s lbp iterations , the message @xmath153 is fixed at @xmath154 the dashed box in fig .
[ fig : turbo ] shows the region of the factor graph on which messages are updated during ssr s lbp iterations .
this subgraph can be recognized as the one that donoho , maleki , and montanari used to derive their so - called _ approximate message passing _ ( amp ) algorithm @xcite .
while @xcite assumed an i.i.d laplacian prior for @xmath18 , the approach for generic i.i.d priors was outlined in @xcite .
below , we extend the approach of @xcite to independent _ non_-identical priors ( as analyzed in @xcite ) and we detail the bernoulli - gaussian case . in the sequel , we use a superscript-@xmath155 to index ssr s lbp iterations . according to the sum - product algorithm
, the fact that @xmath153 is non - gaussian implies that @xmath156 is also non - gaussian , which complicates the exact calculation of the subsequent messages @xmath157 as defined by the sum - product algorithm .
however , for large @xmath29 , the combined effect of @xmath158 at the @xmath159 nodes can be approximated as gaussian using central - limit theorem ( clt ) arguments , after which it becomes sufficient to parameterize each message @xmath156 by only its mean and variance : @xmath160 combining @xmath161 with @xmath162 , the clt then implies that @xmath163 the updates @xmath164 and @xmath165 can then be calculated from @xmath166 where , using ( [ eqn : prod_gauss ] ) , the product term in ( [ eqn : xntogm_ip1 ] ) is @xmath167 assuming that the values @xmath168
satisfy @xmath169 which occurs , e.g. , when @xmath170 is large and @xmath171 are generated i.i.d with variance @xmath172 , we have @xmath173 , and thus ( [ eqn : xntogm_ip1 ] ) is well approximated by @xmath174 in this case , the mean and variance of @xmath175 become @xmath176\nonumber\end{aligned}\ ] ] where @xmath177 according to the sum - product algorithm , @xmath178 , the posterior on @xmath47 after ssr s @xmath179-lbp iteration , obeys @xmath180 whose mean and variance determine the @xmath179-iteration mmse estimate of @xmath47 and its variance , respectively . noting that the difference between ( [ eqn : p_hat_xn ] ) and ( [ eqn : xntogm_ip1 ] ) is only the inclusion of the @xmath112 product term , these mmse quantities become @xmath181 similarly
, the posterior on @xmath141 after the @xmath179 iteration obeys @xmath182\nonumber\end{aligned}\ ] ] where @xmath183\nonumber\end{aligned}\ ] ] since @xmath184 , it can be seen that the corresponding log - likelihood ratio ( llr ) is @xmath185 clearly , the llr @xmath186 and the likelihood function @xmath187 express the same information , but in different ways .
the procedure described thus far updates @xmath188 variables per lbp iteration , which is impractical since @xmath189 can be very large . in @xcite ,
donoho , maleki , and montanari proposed , for the i.i.d case , further approximations that yield a `` first - order '' approximate message passing ( amp ) algorithm that allows the update of only @xmath190 variables per lbp iteration , essentially by approximating the _ differences _ among the outgoing means / variances of the @xmath191 nodes ( i.e. , @xmath192 and @xmath193 ) as well as the differences among the outgoing means / variances of the @xmath194 nodes ( i.e. , @xmath195 and @xmath196 ) .
these resulting algorithm was then rigorously analyzed by bayati and montanari in @xcite .
we now summarize a straightforward extension of the i.i.d amp algorithm from @xcite to the case of an independent but non - identical bernoulli - gaussian prior ( [ eqn : fntoxn ] ) : @xmath197 where @xmath198 , @xmath199 and @xmath200 are defined as @xmath201}{[1+\tau_n(\xi;c)]^2 } \ !
\quad \label{eqn : fnd}\\ \tau_n(\xi;c ) & = & \beta_n(c ) \exp(-\zeta_n(c ) \xi^2 ) .
\label{eqn : taun}\end{aligned}\ ] ] for the first turbo iteration ( i.e. , @xmath202 ) , we initialize amp using @xmath203 , @xmath204 , and @xmath205 for all @xmath206 . for subsequent turbo iterations ( i.e. , @xmath207 )
, we initialize amp by setting @xmath208 equal to the final values of @xmath209 generated by amp at the _ previous _ turbo iteration .
we terminate the amp iterations as soon as either @xmath210 or a maximum of @xmath211 amp iterations have elapsed .
similarly , we terminate the turbo iterations as soon as either @xmath212 a maximum of @xmath211 turbo iterations have elapsed .
the final value of @xmath213^t$ ] is output at the signal estimate @xmath214 .
we now describe how the precisions @xmath215 are learned .
first , we recall that @xmath216 describes the apriori precision on the active coefficients at the @xmath217 level , i.e. , on @xmath218 , where the corresponding index set @xmath219 is of size @xmath220 .
furthermore , we recall that the prior on @xmath216 was chosen as in ( [ eqn : prior_rhoj ] ) .
thus , _ if _ we had access to the true values @xmath218 , then ( [ eqn : spike_slab ] ) implies that @xmath221 which implies ) . ]
that the posterior on @xmath216 would take the form of @xmath222 where @xmath223 and @xmath224 .
in practice , we do nt have access to the true values @xmath218 nor to the set @xmath225 , and thus we propose to build surrogates from the ssr outputs .
in particular , to update @xmath216 after the @xmath122 turbo iteration , we employ @xmath226 and @xmath227 , where @xmath228 and @xmath229 denote the final llr on @xmath141 and the final mmse estimate of @xmath47 , respectively , at the @xmath122 turbo iteration .
these choices imply the hyperparameters @xmath230 finally , to perform ssr at turbo iteration @xmath231 , we set the variances @xmath232 equal to the inverse of the expected precisions , i.e. , @xmath233 .
the noise variance @xmath234 is learned similarly from the ssr - estimated residual . next , we describe how the transition probabilities @xmath235 are learned . first , we recall that @xmath58 describes the probability that a child at level @xmath54 is active ( i.e. , @xmath236 ) given that his parent ( at level @xmath15 ) is active . furthermore , we recall that the prior on @xmath58 was chosen as in ( [ eqn : prior_pij11 ] ) . thus _ if _ we knew that there were @xmath237 active coefficients at level @xmath15 , of which @xmath238 had active children , then the posterior on @xmath58 would take the form of @xmath239 , where @xmath240 and @xmath241 . in practice ,
we do nt have access to the true values of @xmath237 and @xmath238 , and thus we build surrogates from the ssr outputs . in particular , to update @xmath58 after the @xmath122 turbo iteration , we approximate @xmath236 by the event @xmath242 , and based on this approximation set @xmath243 ( as in ( [ eqn : kj ] ) ) and @xmath244 .
the corresponding hyperparameters are then updated as @xmath245 finally , to perform ssr at turbo iteration @xmath231 , we set the transition probabilities @xmath246 equal to the expected value @xmath247 .
the parameters @xmath248 , @xmath249 , and @xmath250 are learned similarly . until now
, we have focused on the bernoulli - gaussian ( bg ) signal model ( [ eqn : spike_slab ] ) . in this section ,
we describe the modifications needed to handle the gaussian mixture ( gm ) model @xmath251 where @xmath252 denotes the variance of `` large '' coefficients and @xmath253 denotes the variance of `` small '' ones . for either the bg or gm prior
, amp is performed using the steps ( [ eqn : xini])-([eqn : cni ] ) . for the bg case ,
the functions @xmath198 , @xmath199 , @xmath200 , and @xmath254 are given in ( [ eqn : fn])([eqn : taun ] ) , whereas for the gm case , they take the form @xmath255 ^ 2}{[1+\bar{\tau}_n(\xi , c)]^2 } + \ , c\,\xi^{-1}\ ! f_n(\xi ; c ) \nonumber\label{eqn : gn2}\\[-1.5mm]&&\\[-3.0 mm ] f'_n(\xi;c ) & = & \frac{\bar{\tau}_n(\xi;c)\bar{\alpha}{_{n,\text{\sf l}}}(c)(1+\bar{\tau}_n(\xi;c ) - 2 \xi^2 \bar{\zeta}_n(c ) ) } { [ 1+\bar{\tau}_n(\xi;c)]^2 } \nonumber\\ & & + \frac{\bar{\alpha}{_{n,\text{\sf s}}}(c)(1+\bar{\tau}_n(\xi;c ) + 2 \xi^2 \bar{\zeta}_n(c)\bar{\tau}_n(\xi;c ) ) } { [ 1+\bar{\tau}_n(\xi;c)]^2 } \quad \label{eqn : fnd2}\\ \bar{\tau}_n(\xi , c ) & = & \bar{\beta}_n(c ) \exp(-\bar{\zeta}_n(c)\xi^2),\end{aligned}\ ] ] where @xmath256 likewise , for the bg case , the extrinsic llr is given by ( [ eqn : amp_llr ] ) , whereas for the gm case , it becomes @xmath257
the proposed turbo approach to compressive imaging was compared to several other tree - sparse reconstruction algorithms : modelcs @xcite , hmt+irwl1 @xcite , mcmc @xcite , variational bayes ( vb ) @xcite ; and to several simple - sparse reconstruction algorithms : cosamp @xcite , spgl1 @xcite , and bernoulli - gaussian ( bg ) amp .
all numerical experiments were performed on @xmath258 ( i.e. , @xmath259 ) grayscale images using a @xmath260-level 2d haar wavelet decomposition , yielding @xmath261 approximation coefficients and @xmath262 individual markov trees . in all cases , the measurement matrix @xmath14 had i.i.d gaussian entries . unless otherwise specified , @xmath263 noiseless measurements were used .
we used normalized mean squared error ( nmse ) @xmath264 as the performance metric .
we now describe how the hyperparameters were chosen for the proposed turbo schemes .
below , we use @xmath265 to denote the total number of wavelet coefficients at level @xmath15 , and @xmath266 to denote the total number of approximation coefficients . for both turbo - bg and turbo - gm ,
the beta hyperparameters were chosen so that @xmath267 , @xmath268 and @xmath269 with @xmath270 , @xmath271 , @xmath272 , and @xmath273 .
these informative hyperparameters are similar to the `` universal '' recommendations in @xcite and , in fact , identical to the ones suggested in the mcmc work @xcite . for turbo - bg ,
the hyperparameters for the signal precisions @xmath274 were set to @xmath275 and @xmath276\!=\![10 , 1 , 1 , 0.1 , 0.1]$ ] .
this choice is motivated by the fact that wavelet coefficient magnitudes are known to decay exponentially with scale @xmath15 ( e.g. , @xcite ) . meanwhile ,
the hyperparameters for the noise precision @xmath277 were set to @xmath278 .
although the measurements were noiseless , we allow turbo - bg a nonzero noise variance in order to make up for the fact that the wavelet coefficients are not exactly sparse , as assumed by the bg signal model .
( we note that the same was done in the bg - based work @xcite . ) for turbo - gm , the hyperparameters @xmath279 for the signal precisions @xmath280 were set at the values of @xmath281 for the bg case , while the hyperparameters @xmath282 for @xmath283 were set as @xmath284 and @xmath285 .
meanwhile , the noise variance @xmath234 was assumed to be exactly zero , because the gm signal prior is capable of modeling non - sparse wavelet coefficients . for mcmc @xcite
, the hyperparameters were set in accordance with the values described in @xcite ; the values of @xmath286 are same as the ones used for the proposed turbo - bg scheme , while @xmath287 . for vb , the same hyperparameters as mcmc
were used except for @xmath288 and @xmath289 , which were the default values of hyperparameters used in the publicly available code .
we experimented with the values for both mcmc and vb and found that the default values indeed seem to work best .
for example , if one swaps the @xmath281 hyperparameters between vb and mcmc , then the average performance of vb and mcmc degrade by @xmath290db and @xmath291db , respectively , relative to the values reported in table [ tab : summary ] . for both the cosamp and modelcs algorithms ,
the principal tuning parameter is the assumed number of non - zero coefficients . for both modelcs ( which is based on cosamp ) and cosamp itself ,
we used the rice university codes , which include a genie - aided mechanism to compute the number of active coefficients from the original image .
however , since we observed that the algorithms perform somewhat poorly under that tuning mechanism , we instead ran ( for each image ) multiple reconstructions with the number of active coefficients varying from @xmath292 to @xmath293 in steps of @xmath294 , and reported the result with the best nmse . the number of active coefficients chosen in this manner was usually much smaller than that chosen by the genie - aided mechanism . to implement bg - amp
, we used the amp scheme described in section [ sec : mp : ssr ] with the hyperparameter learning scheme described in section [ sec : update ] ; hmt structure was not exploited . for this
, we assumed that the priors on variance @xmath295 and activity @xmath296 were identical over the coefficient index @xmath146 , and assigned gamma and beta hyperpriors of @xmath297 and @xmath298 , respectively .
for hmt+irwl1 , we ran code provided by the authors with default settings . for spgl1 ,
the residual variance was set to @xmath57 , and all parameters were set at their defaults .
[ fig : reccomp ] shows a @xmath258 section of the `` cameraman '' image along with the images recovered by the various algorithms .
qualitatively , we see that cosamp , which leverages only simple sparsity , and modelcs , which models persistence - across - scales ( pas ) through a deterministic tree structure , both perform relatively poorly .
hmt+irwl1 also performs relatively poorly , due to ( we believe ) the ad - hoc manner in which the hmt structure was exploited via iteratively re - weighted @xmath17 .
the bg - amp and spgl1 algorithms , neither of which attempt to exploit pas , perform better .
the hmt - based schemes ( vb , mcmc , turbo - gm , and turbo - gm ) all perform significantly better , with the turbo schemes performing best .
[ b][b][0.8]original [ b][b][0.8]hmt+irwl1 [ b][b][0.8]cosamp [ b][b][0.8]modelcs [ b][b][0.8]bg - amp [ b][b][0.8]spgl1 [ b][b][0.8]variational bayes [ b][b][0.8]mcmc [ b][b][0.8]turbo - bg [ b][b][0.8]turbo - gm + [ c][bc][0.8]image type [ c][c][0.8]average nmse ( db ) [ l][l][0.45 ] cosamp [ l][l][0.45 ] hmt - irwl1 [ l][l][0.45]turbo bg [ l][l][0.45]turbo gm [ c][bc][0.8]image type [ c][c][0.8]average runtime ( sec ) [ l][l][0.45 ] cosamp [ l][l][0.45 ] hmt - irwl1 [ l][l][0.45]turbo bg [ l][l][0.45]turbo gm .nmse and runtime averaged over @xmath299 images . [ cols="^,^,^",options="header " , ] [ c][bc][0.8]number of measurements @xmath300 [ c][c][0.8]average nmse ( db ) [ l][l][0.45 ] cosamp [ l][l][0.45 ] hmt - irwl1 [ l][l][0.45]turbo bg [ l][l][0.45]turbo gm [ c][bc][0.8]number of measurements @xmath300 [ c][c][0.8]average runtime ( sec ) [ l][l][0.45 ] cosamp [ l][l][0.45 ] hmt - irwl1 [ l][l][0.45]turbo bg [ l][l][0.45]turbo gm for a quantitative comparison , we measured average performance over a suite of images in a _ microsoft research object class recognition _ database images extracted from the `` pixel - wise labelled image database v2 '' at _ http://research.microsoft.com / en - us / projects / objectclassrecognition_. what we refer to as an `` image type '' is a `` row '' in this database . ] that contains @xmath301 types of images ( see fig . [
fig : categories ] ) with roughly @xmath302 images of each type . in particular
, we computed the average nmse and average runtime on a 2.5 ghz pc , for each image type .
these results are reported in figures [ fig : nmse ] and [ fig : comptime ] , and the global averages ( over all @xmath299 images ) are reported in table [ tab : summary ] . from the table , we observe that the proposed turbo algorithms outperform all the other tested algorithms in terms of reconstruction nmse , but are beaten only by cosamp in speed . between the two turbo algorithms , we observe that turbo - gm slightly outperforms turbo - bg in terms of reconstruction nmse , while taking the same runtime . in terms of nmse performance ,
the closest competitor to the turbo schemes is mcmc , db ( i.e. , @xmath303db better ) at an average runtime of @xmath304 sec ( i.e. , @xmath305 slower ) . ]
whose nmse is @xmath306db worse than turbo - bg and @xmath307db worse than turbo - gm .
the good nmse performance of mcmc comes at the cost of complexity , though : mcmc is @xmath308 times slower than the turbo schemes .
the second closest nmse - competitor is vb , showing performance @xmath309 db worse than turbo - bg and @xmath310db worse than turbo - gm .
even with this sacrifice in performance , vb is still twice as slow as the turbo schemes . among the algorithms that do not exploit pas
, we see that spgl1 offers the best nmse performance , but is by far the slowest ( e.g. , @xmath301 times slower than cosamp ) . meanwhile , cosamp is the fastest , but shows the worst nmse performance ( e.g. , @xmath311db worse than spgl1 ) .
bg - amp strikes an excellent balance between the two : its nmse is only @xmath312db away from spgl1 , whereas it takes only @xmath313 times as long as cosamp .
however , by combining the amp algorithm with hmt structure via the turbo approach , it is possible to significantly improve nmse while simultaneously decreasing the runtime .
the reason for the complexity decrease is twofold .
first , the hmt structure helps the amp and parameter - learning iterations to converge faster .
second , the hmt steps are computationally negligible relative to the amp steps : when , e.g. , @xmath314 , the amp portion of the turbo iteration takes approximately @xmath315 sec while the hmt portion takes @xmath316 sec .
we also studied nmse and compute time as a function of the number of measurements , @xmath170 . for this study , we examined images of type 1 at @xmath317 . in figure
[ fig : nmse_varym ] , we see that turbo - gm offers the uniformly best nmse performance across @xmath170 .
however , as @xmath170 decreases , there is little difference between the nmses of turbo - gm , turbo - cs , and mcmc .
as @xmath170 increases , though , we see that the nmses of mcmc and vb converge , but that they are significantly outperformed by turbo - gm , turbo - cs , and
somewhat surprisingly spgl1 .
in fact , at @xmath318 , spgl1 outperforms turbo - bg , but not turbo - gm .
however , the excellent performance of spgl1 at these @xmath170 comes at the cost of very high complexity , as evident in figure [ fig : comptime_varym ] .
we proposed a new approach to hmt - based compressive imaging based on loopy belief propagation , leveraging a turbo message passing schedule and the amp algorithm of donoho , maleki , and montanari .
we then tested our algorithm on a suite of @xmath299 natural images and found that it outperformed the state - of - the - art approach ( i.e. , variational bayes ) while halving its runtime .
m. f. duarte , m. b. wakin , and r. g. baraniuk , `` wavelet - domain compressive signal reconstruction using a hidden markov tree model , '' in _ proc .
speech & signal process .
_ , las vegas , nv , apr . 2008 , pp . 51375140 .
b. j. frey and d. j. c. mackay , `` a revolution : belief propagation in graphs with cycles , '' in _ adv . in neural inform .
processing syst .
_ , m. jordan , m. s. kearns , and s. a. solla , eds.1em plus 0.5em minus 0.4em mit press , 1998 .
s. som , l. c. potter , and p. schniter , `` on approximate message passing for reconstruction of non - uniformly sparse signals , '' in _ proc . national aerospace and electronics conf . _ ,
dayton , oh , jul .
2010 .
j. k. romberg , h. choi , and r. g. baraniuk , `` bayesian tree - structured image modeling using wavelet - domain hidden markov models , '' _ ieee trans . image process .
_ , vol .
10 , no . 7 , pp . 10561068 , jul .
s. som , l. c. potter , and p. schniter , `` compressive imaging using approximate message passing and a markov - tree prior , '' in _ proc .
asilomar conf .
signals syst .
_ , pacific grove , ca , nov . | we propose a novel algorithm for compressive imaging that exploits both the sparsity and persistence across scales found in the 2d wavelet transform coefficients of natural images . like other recent works ,
we model wavelet structure using a hidden markov tree ( hmt ) but , unlike other works , ours is based on loopy belief propagation ( lbp ) . for lbp , we adopt a recently proposed `` turbo '' message passing schedule that alternates between exploitation of hmt structure and exploitation of compressive - measurement structure . for the latter , we leverage donoho , maleki , and montanari s recently proposed approximate message passing ( amp ) algorithm .
experiments with a large image database suggest that , relative to existing schemes , our turbo lbp approach yields state - of - the - art reconstruction performance with substantial reduction in complexity . |
[ sec : intro ] the classical area of _ combinatorial designs _ ( see , e.g. , @xcite ) asks about extremal subset systems of a finite set that have specified intersection patterns . in this paper , we are interested in extremal families of curves in a closed , orientable surface with specified intersection patterns . more specifically , benson farb and chris leininger brought the following topological variant to our attention : [ farbq ] let @xmath1 be a closed oriented surface of genus @xmath0 , and let @xmath2 be a collection of pairwise non - isotopic essential simple closed curves @xmath3 such that @xmath4 intersects @xmath5 at most once for any @xmath6 .
how large can @xmath2 be ?
in fact , this question goes back further , to work of juvan , malni , and mohar @xcite , which we learned about after circulating an earlier version @xcite of the present paper .
juvan , et al .
@xmath7-system _ of curves on a genus @xmath0 closed , oriented surface to be a collection of non - isotopic , essential , simple closed curves @xmath8 such that each of the @xmath4 intersects any @xmath5 , @xmath9 at most @xmath7 times .
we define @xmath10 to be the maximum size of any @xmath7-system on a closed surface @xmath1 of genus @xmath0 .
the most important related results from @xcite are : * for all @xmath11 and @xmath12 , @xmath10 is finite .
* @xmath13 .
* @xmath14 asymptotics for @xmath7 and @xmath0 growing are also given in @xcite .
we study @xmath15 for @xmath7 fixed .
we are particularly interested in the farb - leininger case @xmath16 , for which we can prove : [ theo : generalbounds ] for @xmath17 , @xmath18 and for all @xmath19 , @xmath20 both the upper and lower bounds are , to the best of our knowledge , the best known .
juvan , et al .
do nt optimize the bound in ( * ? ? ?
* theorem 3.3 ) for @xmath16 , but examining their arguments , we get an upper bound on the order of @xmath21 .
the upper bound of theorem [ theo : generalbounds ] is based on the @xmath22-homology of curves ( proposition [ prop : mod2classupper ] ) .
thus same upper bound statement and proof applies if we replace `` @xmath23-system '' with @xmath7-system in which all pairs of curves are either disjoint or intersect an odd number of times .
thus the upper bound in theorem [ theo : generalbounds ] has some built - in slack , and it is plausible that , for large enough @xmath7 , this upper bound is nearly tight . reducing the gap will require a proof method that can see the difference between `` one intersection '' and `` an odd number of intersections '' .
[ [ surfaces - with - boundary - components ] ] surfaces with boundary components + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for a surface with boundary components , a @xmath7-system is defined similarly to the closed case , except we add the condition that none of the curves is allowed to to be homotopic to a boundary curve .
define @xmath24 to be the maximum size of a @xmath7-system in a surface of genus @xmath0 with @xmath25 boundary components .
holding @xmath0 fixed ( along with @xmath7 ) , we can determine @xmath26 very precisely in terms of @xmath27 : [ theo : boundaryasymptotics ] fix some @xmath28 .
then , for all @xmath29 , @xmath30 for @xmath31 and @xmath32 @xmath33 relaxing the definition of a @xmath7-system to allowed arcs as well as curves on a surface with boundary seems to make the problem as hard as determining @xmath27 . [ [ low - genus ] ] low genus + + + + + + + + + on the torus , it is straightforward to show that a maximal @xmath23-system has @xmath34 curves , i.e. @xmath35 . in genus two , we present an elegant geometric argument , based on the hyperelliptic involution , which shows that : [ theo : genus2 ] n(1,2 ) = 12 .
moreover , there are exactly two mapping class orbits of maximal @xmath23-systems on a genus-@xmath36 surface .
the proof gives exact structural information about the maximal @xmath23-systems on a genus-@xmath36 surface : they are lifts of @xmath37-vertex triangulations of the sphere drawn in the quotient of the surface by the hyperelliptic involution .
[ [ nearly - all - curves - intersecting ] ] nearly all curves intersecting + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + given a @xmath7-system @xmath38 on a closed surface @xmath1 of genus @xmath0 , we define its _ odd intersection graph _ @xmath39 to be the graph that has one vertex for each of the @xmath4 and an edge between vertices @xmath40 and @xmath41 if @xmath4 intersects @xmath5 an odd number of times . with assumptions on @xmath42 , we can get much tighter upper bounds .
[ theo : alloddupper ] let @xmath38 be a @xmath7-system with odd intersection graph @xmath39 . then for @xmath43 and @xmath44 : * if @xmath39 is the complete graph @xmath45 , then @xmath46 , and the bound is sharp for all @xmath47 .
* if @xmath39 has average degree @xmath48 , then @xmath49 .
the first statement is proved using a @xmath22-homology argument and the second statement follows via an application of turn s theorem @xcite to @xmath42 s complement , an idea we learned from van vu s paper @xcite .
[ [ nearly - all - curves - non - intersecting ] ] nearly all curves non - intersecting + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + given a @xmath7-system @xmath38 we define its _ intersection graph _ to be the graph that has one vertex for each @xmath4 and an edge between vertices @xmath40 and @xmath41 if @xmath4 intersects @xmath5 .
[ theo : lowdegree ] let @xmath38 be a @xmath7-system on a closed surface of genus @xmath0 with intersection graph @xmath42 . for @xmath50 and @xmath44 .
if @xmath42 has average degree @xmath51 , then @xmath52 . in the case where @xmath53 , this is just the well - known fact that a maximal system of disjoint nonisotopic simple closed curves has at most @xmath54 curves
jm , ir , and lt received support for this work from rivin s nsf cdi - i grant dmr 0835586 .
lt s final preparation was supported by the european research council under the european union s seventh framework programme ( fp7/2007 - 2013 ) / erc grant agreement no 247029-sdmodels .
our initial work on this problem took place at ileana streinu s 2010 barbados workshop at the bellairs research institute .
[ sec : gen2 ] in this section , we give a geometric argument to prove theorem [ theo : genus2 ] . the argument generalizes to give a sub - quadratic lower bound in higher genus .
the key tool we need is the _
hyperelliptic involution_. here are the facts we need , which can all be found in @xcite .
[ prop : hyper - exists ] every riemann surface @xmath1 of genus @xmath36 admits a `` hyperelliptic involution '' with @xmath37 fixed points . the fixed points of the hyperelliptic involution are called _ weirstrass points _ , and we denote them @xmath55 . [
prop : hyper - quotient ] let @xmath1 be a closed riemann surface of genus @xmath36 .
the quotient @xmath56 of @xmath1 by the hyperelliptic involution is a sphere with six double points corresponding to the weierstrass points ; i.e. , @xmath56 is an orbifold of signature @xmath57 .
simple closed geodesics in the riemann surface @xmath1 have nice representatives in @xmath56 . [
prop : hyper - geodesics ] let @xmath1 and @xmath56 be as in proposition [ prop : hyper - quotient ]
. then any non - separating simple closed geodesic @xmath58 in @xmath1 goes through exactly two of the wierstrass points and projects to a geodesic segment going between exactly two of the doubled points in @xmath56 . by proposition
[ prop : nosep ] proven below , there is no separating curve in a maximal @xmath23-system , and by proposition [ prop : hyper - geodesics ] , the @xmath23-systems in @xmath1 that have only non - separating curves correspond to isomorphism classes of simple planar graphs on @xmath37 vertices .
since planar triangulations on @xmath37 vertices have @xmath59 edges , it follows that @xmath60 .
there are two graph isomorphism classes of planar triangulations on @xmath37 vertices , namely , the octahedron and doubly stellated tetrahedron .
since these graphs are @xmath34-connected , there are only two triangulations up to homeomorphism of the sphere ( permuting weierstrass points ) .
birman and hilden @xcite have shown that the mapping class group of the genus @xmath36 surface modulo the hyperelliptic involution is isomorphic to the mapping class group of the @xmath37-times punctured sphere . consequently , there are two mapping class group orbits of maximal @xmath23-systems .
@xmath61 the lower bound of theorem [ theo : genus2 ] generalizes to surfaces @xmath1 of @xmath62 : there are @xmath63 weierstrass points corresponding to double points in the quotient by the hyperelliptic involution , implying that any planar graph on @xmath63 vertices corresponds to some @xmath23-system .
thus , we obtain the lower bound for @xmath17 from theorem [ theo : generalbounds ] .
[ sec : structure ] we will need the following result on the structure of maximal @xmath23-systems : [ prop : nosep ] let @xmath1 be a closed surface of genus @xmath43 .
then , any maximal @xmath23-system in @xmath1 contains no separating curves . from now on in this section , we assume that @xmath1 satisfies the hypotheses of proposition [ prop : nosep ] .
we also use the standard notions of _ minimal position _ and _ geometric intersection number _ for curves , which may be found in ( * ? ? ?
* section 1.2.3 ) .
we denote the geometric intersection number of ( the isotopy classes of ) @xmath64 and @xmath65 by @xmath66 .
a @xmath7-system is in minimal position when the curves in it are pairwise in minimal position .
every @xmath7-system has such a representative ( * ? ? ?
* corollary 1.9 ) .
the auxiliary lemma [ lemma : annulipoly ] , which describes the complementary regions of a maximal @xmath7-system is also of independent interest .
let @xmath2 be a @xmath7-system and define the _ complementary regions of @xmath2 _ to be the components of the surface @xmath67 obtained by cutting @xmath1 along the curves in @xmath2 .
[ lemma : annulipoly ] let @xmath2 be a maximal @xmath7-system such that the curves in @xmath2 are in minimal position . then
, the complementary regions are polygons and annuli .
furthermore , each annulus has a boundary component consisting of a single curve in @xmath2 .
intuitively , one would expect that if proposition [ prop : nosep ] were to fail , an inductive argument would establish an @xmath68 upper bound on the size of any @xmath23-system , contradicting the fact that there are @xmath23-systems with @xmath69 curves .
however , we are unaware of a rigorous proof along these lines , so instead we use lemma [ lemma : annulipoly ] . the difficult part of the proof of lemma [ lemma : annulipoly ] will be to rule out pairs of pants as complementary regions .
this next lemma is straightforward and gives the starting point .
[ lemma : nobiggerthanpants ] suppose @xmath2 is a maximal @xmath7-system in minimal position .
then , the complementary regions are all polygons , annuli , or pairs of pants .
furthermore , each annulus has a boundary component consisting of a single curve in @xmath2 , and every boundary component of the pairs of pants are curves in @xmath2 .
if a complementary region has nonzero genus , then it contains a simple closed curve that is not isotopic to a boundary curve which we could add to @xmath2 to obtain a larger @xmath7-system . a similar fact is true if a complementary region is a sphere with at least @xmath70 boundary components .
this proves that the complementary regions are all polygons , annuli , or pairs of pants .
now consider a complementary region that is an annulus .
if neither boundary component is a single curve in @xmath2 , then the core curve of this annulus is , by our assumption of minimal position , not isotopic to any curve in @xmath2 and does not intersect any curves in @xmath2 .
this contradicts the maximality of @xmath2 .
similar reasoning shows that all boundary curves of a pair of pants must be elements of @xmath2 .
continuing from lemma [ lemma : nobiggerthanpants ] , it suffices to rule out pants as a complementary region . comparing the ( classical ) upper bound , theorem [ theo : all - disjoint ] , of the size of a @xmath71-system and the lower bound on the size of a @xmath23-system from theorem [ theo : generalbounds ]
, we see that not all the complementary regions can be pairs of pants .
thus , if @xmath2 has some pair of pants as a complementary region , then connectedness of @xmath1 and lemma [ lemma : nobiggerthanpants ] implies that some curve @xmath58 in @xmath2 is disjoint from the rest of @xmath2 and bounds a pair of pants @xmath72 on one side and an annulus @xmath73 on the other .
the other boundary component of @xmath73 consists of arcs of curves in @xmath2 , and there are at least two such arcs .
let @xmath74 be two curves with consecutive arcs in @xmath75 .
@xmath76 at 50 140 @xmath77 at 58 21 @xmath78 at 110 170 @xmath79 at 130 58 @xmath58 at 37 25 @xmath80 at 76 50 @xmath81 at 150 89 @xmath82 at 48 104 @xmath83 at 140 145 and arcs @xmath84.,scaledwidth=50.0% ] we now show that @xmath58 may be replaced with two new curves to obtain a larger @xmath23-system ; the construction is depicted in figure [ figure : newcurves ] .
let @xmath85 be the other boundary components ( aside from @xmath58 ) in the pair of pants @xmath72 , and let @xmath82 be a simple arc connecting @xmath76 to @xmath80 and lying in @xmath86 .
the regular neighborhood of @xmath87 is a pair of pants where two of the boundary curves are isotopic to @xmath76 and @xmath80 ; let @xmath77 be the other boundary component .
similarly , let @xmath83 be a simple arc which connects @xmath78 to @xmath81 , lies in @xmath86 , and is disjoint from @xmath82 ; let @xmath79 be defined similarly to @xmath77 . to show @xmath88 is a @xmath7-system of larger size , we must show pairwise intersections are less than or equal to @xmath7 and that @xmath77 and @xmath79 are not isotopic to other curves in @xmath2 or each other .
it is possible to do the latter by establishing the following geometric intersection numbers since the numbers are an invariant of isotopy classes .
* for all curves @xmath89 , we have @xmath90 and @xmath91 * @xmath92 * @xmath93 before proving ( 1)-(3 ) , let us see that they establish that @xmath94 is a @xmath7-system
. statements ( 1 ) and ( 2 ) show @xmath94 is a @xmath7-system .
( note that ( 1 ) also says @xmath95 and similarly for @xmath96 . )
statement ( 3 ) implies @xmath77 and @xmath79 are not isotopic to any curves in @xmath2 since they are all disjoint from @xmath58 , and statement ( 1 ) implies @xmath97 which establishes that @xmath77 and @xmath79 are not isotopic .
[ [ proof - of-1 ] ] proof of ( 1 ) + + + + + + + + + + + + we only show that @xmath90 since the proof applies also to @xmath78 and @xmath79 . if @xmath98 , this is immediate since @xmath77 and @xmath76 are disjoint .
now suppose @xmath99 .
homotope @xmath77 to a simple closed curve @xmath100 which is the union of an arc @xmath101 from @xmath76 and an arc @xmath102 which lies in @xmath86 , is disjoint from all curves in @xmath103 , and cuts @xmath72 into two annuli .
( see figure [ figure : nopantsfig2 ] . ) clearly , the arc @xmath101 and curve @xmath104 cross @xmath105 times , so by the bigon criterion ( ( * ? ? ?
* proposition 1.7 ) ) , it suffices to show that it is impossible to bound a disc with an arc from @xmath104 and an arc from @xmath100 .
since @xmath2 is in minimal position , @xmath76 and @xmath104 have no bigons , and so any bigon made from @xmath104 and @xmath100 must use the arc @xmath102 .
however , on each side of @xmath102 , we may connect @xmath102 to either @xmath80 or @xmath106 with an arc disjoint from @xmath104 ( since @xmath104 is disjoint from @xmath72 ) ; this would be impossible if an arc containing @xmath102 made a bigon with an arc from @xmath104 .
this established the first equality ; the second follows by a similar argument
. indicated in gray.,scaledwidth=50.0% ] [ [ proof - of-2 ] ] proof of ( 2 ) + + + + + + + + + + + + let @xmath100 be as above . by further homotoping , we can ensure that @xmath102 is disjoint from @xmath79 and so @xmath107 [ [ proof - of-3 ] ] proof of ( 3 ) + + + + + + + + + + + + again , we use the bigon criterion . by construction , @xmath58 and @xmath77 each cut each other into two arcs , @xmath108 and @xmath109 respectively , where w.l.o.g .
@xmath110 lies in @xmath72 .
the curves @xmath111 and @xmath112 are each homotopic to some boundary component in @xmath72 and hence can not be trivial .
if @xmath113 bounded a disc , then we could homotope @xmath77 to lie in @xmath72 which is impossible since @xmath78 and @xmath77 have nonzero geometric intersection number .
@xmath61 suppose , for a contradiction , that @xmath2 is a maximal @xmath23-system with a separating curve @xmath58 ; w.l.o.g .
, we may assume that @xmath2 is in minimal position .
lemma [ lemma : annulipoly ] then applies , so @xmath58 is incident on two complementary regions @xmath73 and @xmath114 that are both annuli , and each of @xmath73 and @xmath114 has a boundary component consisting of arcs from at least two different curves in @xmath2 .
we will show that by removing @xmath58 , two new curves may be added to get a larger @xmath23-system . because @xmath2 is a @xmath23-system
, we may slightly strengthen the conclusion of lemma [ lemma : annulipoly ] : the boundary components of @xmath73 and @xmath114 that are not @xmath58 span arcs from at least three different curves in @xmath2 .
let @xmath74 be curves with consecutive arcs in @xmath75 , and let @xmath80 be a curve contributing a boundary arc to @xmath114 .
let @xmath115 be the surface obtained by cutting along @xmath58 .
there is an arc @xmath101 starting and ending at @xmath116 such that @xmath101 `` follows '' @xmath76 .
more precisely , there is an arc @xmath101 such that @xmath101 , @xmath76 and a part of @xmath116 cobound an annulus and @xmath101 has the same intersection numbers with the other curves in @xmath103 .
similarly , there is such an arc @xmath117 for @xmath78 , and let @xmath118 be two disjoint such arcs for @xmath80 .
we can homotope the arcs @xmath119 so that : @xmath76 at 50 100 @xmath78 at 110 130 @xmath58 at 102 26 @xmath58 at 331 26 @xmath80 at 284 110 @xmath101 at 48 48 @xmath117 at 140 95 @xmath120 at 285 62 @xmath121 at 280 30 * @xmath101 and @xmath117 intersect exactly once transversely and * the endpoints of @xmath101 match up with @xmath120 and @xmath117 with @xmath121 when gluing @xmath115 back together .
( this is depicted in figure [ figure : arcsfornosep ] . )
let @xmath77 and @xmath79 be the resulting arcs from gluing @xmath101 with @xmath120 and @xmath117 with @xmath121 respectively . since @xmath77 and @xmath79 intersect
exactly once transversely , they are distinct non - trivial simple closed curves . furthermore both @xmath77 and @xmath79 intersect some curve exactly once on either side of @xmath58 .
since no other curve in @xmath2 does this , @xmath77 and @xmath79 are isotopically distinct from all other curves in @xmath2 .
thus @xmath122 is a larger @xmath23-system than @xmath2 . @xmath61
[ sec : upper - hom ] now we turn to the general case of a surface @xmath1 of genus @xmath50 .
the point of this section is to prove the following proposition , from which the upper bound in theorem [ theo : generalbounds ] follows readily .
[ prop : mod2classupper ] let @xmath1 be a closed oriented surface of genus @xmath50
. then any @xmath23-system of curves on @xmath1 has at most @xmath123 curves in any nontrivial @xmath22-homology class .
the main lemma we need is : [ lemma : disj - all - sep ] let @xmath43 and let @xmath1 be a genus @xmath0 surface with @xmath36 boundary components .
let @xmath2 be a @xmath71-system of separating curves in @xmath1 such that each curve separates the boundary components .
then , @xmath124 .
we argue by induction .
suppose @xmath31 . in this case ,
the only separating simple curves which are not boundary parallel cut @xmath1 into a torus with a single boundary component and a three - holed sphere .
such a curve does not separate the boundary components of @xmath1 .
suppose @xmath125 . cutting @xmath1 along a curve @xmath58 in @xmath2 yields two surfaces @xmath126 with genus @xmath127 adding up to @xmath0 and each with @xmath36 boundary components .
we note that @xmath103 deposits a @xmath71-system in each of @xmath115 and @xmath128 with the same properties as @xmath2 . thus @xmath129 .
it suffices to show that one can find at most @xmath123 mutually disjoint curves in the same @xmath22-homology class .
let @xmath1 be a closed genus @xmath50 surface , and @xmath2 be a @xmath71-system of curves all in the same @xmath22-homology class .
we reduce the proposition to lemma [ lemma : disj - all - sep ] .
cut along some @xmath130 to obtain a surface @xmath115 of genus @xmath131 with two boundary components corresponding to @xmath58 ; fill in each boundary component with a disk to get a closed surface @xmath128 .
the curves in @xmath132 in @xmath128 are all null - homologous in @xmath133 .
this means that their homology classes in @xmath134 are all either non - primitive or trivial .
the former implies the the classes are not primitive in @xmath135 , which is disallowed by the hypothesis that all the curves in @xmath2 are simple ( see ( * ? ? ?
* proposition 6.2 ) ) .
thus , any curve in @xmath103 separates @xmath128 . furthermore , any @xmath136 must separate the boundary components of @xmath115 . if it did
not , then it would bound a subsurface in @xmath115 with a single boundary component , and thus also in @xmath1 .
this however would imply that @xmath81 is null - homologous as an element of @xmath137 , a contradiction .
consequently , we are in the situation of lemma [ lemma : disj - all - sep ] and are done .
@xmath61 by proposition [ prop : nosep ] , a maximal @xmath23-system contains no separating curve and thus no null - homologous curve .
the curves must then all lie in the @xmath138 nontrivial @xmath22-homology classes , and , by proposition [ prop : mod2classupper ] , each of these contains at most @xmath123 curves in a @xmath23-system .
[ sec : extreme ] in this section we give much sharper upper bounds when the intersection graph @xmath42 of a @xmath23-system @xmath2 is either very sparse or very dense .
the key cases are when @xmath42 or its complement are complete .
we recall the following classical and widely known fact , which may be found in ( * ? ? ?
* section 8.3.1 ) .
[ theo : all - disjoint ] let @xmath139 and let @xmath1 be a closed surface of genus @xmath0 , and let @xmath2 be a @xmath71-system . then @xmath140 . to bound the size of @xmath23-systems with all pairs of curves intersecting , we continue along the lines of section [ sec : upper - hom ] .
let @xmath141 and @xmath142 be vectors in @xmath143 and let @xmath144 denote the standard symplectic pairing .
[ prop : linalg ] let @xmath145 be non - zero vectors in @xmath143 with the property that , for all @xmath6 , @xmath146 . then @xmath46 .
suppose there is a linear dependence @xmath147 among the @xmath148 , with not all the @xmath149 zero .
suppose further that the dependence is non - trivial and that @xmath150 and @xmath151 .
pairing both sides of with @xmath152 tells us the number of non - zero @xmath149 is odd ; similarly pairing both sides of with @xmath153 tells us the number of non - zero @xmath149 is even .
the resulting contradiction implies that , in fact , for any non - trivial linear dependence among the @xmath148 , the @xmath149 are all one .
thus , the @xmath148 are either independent , implying @xmath154 , or there is a unique linear dependence with full support among them , implying @xmath155 . as a corollary
, we obtain : [ prop : all - intersecting ] let @xmath44 , and let @xmath2 be a @xmath7-system in a closed genus @xmath0 surface with any pair of curves intersecting an odd number of times
. then @xmath156 .
if the minimal geometric intersection number between essential , non - isotopic , simple closed curves is odd , then their algebraic intersection number mod @xmath36 is @xmath23 and , furthermore , their @xmath22-homology classes must be distinct .
apply proposition [ prop : linalg ] .
we prove theorem [ theo : alloddupper ] , since the proof of theorem [ theo : lowdegree ] is nearly identical .
the upper bound when the graph @xmath39 is @xmath45 is proposition [ prop : all - intersecting ] . if @xmath39 has average degree @xmath48 , then its complement has average degree @xmath51 , and by turn s theorem @xcite , must contain an independent set of size @xmath157 .
applying proposition [ prop : all - intersecting ] to the curves in @xmath2 represented by the corresponding clique in @xmath39 , we see that @xmath158 .
@xmath61 the bound of proposition [ prop : linalg ] is tight , as the following example shows .
define @xmath159 , @xmath160 , and @xmath161 . for @xmath50 , inductively define @xmath162 for @xmath163 $ ] and then @xmath164 , @xmath165 , and @xmath166 .
( the semi - colons mean concatenating vectors . )
.,scaledwidth=70.0% ] [ prop : linalg - tight ] let @xmath43 , and let @xmath167 be defined as above . for all @xmath9 , we have @xmath168 . for @xmath169 , this is an easy computation , and the @xmath50 cases follow by induction .
figure [ fig : curves - g2 ] shows curves with @xmath22-homology classes given by the @xmath170 .
the vectors @xmath167 are canonical .
[ prop : linalg - canonical ] let @xmath43 and suppose that @xmath171 have the property that , for @xmath6 , @xmath172 . then there is a symplectic automorphism @xmath73 of @xmath143 such that @xmath173 . in the proof , we need the basic fact : [ lemma : linalg - indep ] let @xmath174 and @xmath175 be linearly independent sets of vectors such that @xmath176 for all @xmath40 and @xmath41 .
then there is a symplectic automorphism @xmath73 of @xmath143 such that @xmath173 .
let @xmath73 be the automorphism from lemma [ lemma : linalg - indep ] applied to the @xmath177 and @xmath178 from the statement of the theorem .
this is allowed , because the proof of proposition [ prop : linalg ] says that for any collections meeting the hypothesis of the @xmath167 and @xmath179 , we have @xmath180 with the first @xmath181 vectors independent .
thus , we see that @xmath182 we can explicitly describe a @xmath23-system with @xmath183 curves all intersecting pairwise .
[ theo : canonical - curves ] for any genus @xmath184 closed surface , there is a @xmath23-system of @xmath183 curves which all pairwise intersect .
a genus @xmath0 surface is homeomorphic to a regular @xmath185-gon with opposite sides identified .
we obtain @xmath181 simple closed curves from simple arcs connecting opposite sides of the @xmath185-gon . after identifying sides ,
all vertices become identified , so we can add a diagonal to the configuration as well .
later , we will use the following slightly stronger statement which is clear from the construction .
[ cor : canonical - curves - same - point ] the @xmath23-system from theorem [ theo : canonical - curves ] can be chosen so that all curves pairwise intersect at the same point on the surface . in both genus @xmath23 and @xmath36 , it can be shown that , up to the mapping class group , there is only one maximal configuration of curves all of which pairwise intersect exactly once . in genus
@xmath36 , the configuration corresponds to a star graph in the quotient under hyperelliptic involution .
it would be interesting to know if this generalizes . in higher genus surfaces
, is there only one mapping class orbit of @xmath23-systems such that each pair of curves intersects ?
we now prove the upper bound of theorem [ theo : boundaryasymptotics ] .
first , we improve proposition [ prop : all - intersecting ] [ lemma : all - intersecting - boundary ] the statement of proposition [ prop : all - intersecting ] holds for a genus @xmath0 surface with any number of boundary components .
suppose @xmath2 is a @xmath7-system in a genus @xmath0 surface with @xmath25 boundary components with any pair of curves intersecting an odd number of times .
glue in discs into all the boundary components to obtain a closed surface @xmath1 .
since each curve intersects any other curve transversely an odd number of times , no pair of curves of @xmath2 can be pairwise isotopic in @xmath1 .
consequently , @xmath2 is a @xmath7-system satisfying the hypothesis of proposition [ prop : all - intersecting ] and hence @xmath186 .
we argue by induction and show that @xmath187 let @xmath2 be a maximal @xmath23-system in @xmath188 , the surface of genus @xmath0 and @xmath25 boundary components .
glue in a disc @xmath51 to one of the boundary components to obtain @xmath189 .
since intersection numbers did not increase , curves in @xmath2 still pairwise intersect at most once , but some curves may have become isotopic as a result of gluing in the disc .
we will show that a @xmath23-system @xmath190 on @xmath189 may be obtained by removing at most @xmath183 curves from @xmath2 .
then , we have @xmath191 if two curves become isotopic after gluing in @xmath51 , then since @xmath2 is a @xmath23-system , they are disjoint and hence bound an annulus which necessarily contains @xmath51 . in particular , any curve can become isotopic to at most @xmath23 other curve , so we must understand how many pairs of isotopic curves can occur . if @xmath192 and @xmath193 are two such pairs , then both pairs bound an annulus containing @xmath51 and the annuli must intersect .
consequently , each curve in an isotopic pair intersects every curve in the other pairs .
if we construct a set @xmath94 by taking one curve from each pair , then @xmath94 is as in lemma [ lemma : all - intersecting - boundary ] and so @xmath194 .
thus @xmath195 is the @xmath23-system on @xmath189 as desired . in the case of the torus
, we have @xmath196 .
indeed , if two curves on a one - holed torus become isotopic after gluing in the disc @xmath51 , then , as before , they bound an annulus containing @xmath51 ; however , since the surface is a torus , they also bound an annulus , not containing @xmath51 , on the other side , and so they must have already been isotopic in the one - holed torus .
@xmath61 in the next section , we will see that obtaining our lower bounds essentially amounts to reversing the argument in the previous paragraph .
[ sec : lower ] in this section we prove the lower bounds of theorem [ theo : generalbounds ] and theorem [ theo : boundaryasymptotics ]
. we will first prove theorem [ theo : boundaryasymptotics ] .
the lower bound of theorem [ theo : generalbounds ] then follows easily by attaching handles .
the base case is corollary [ cor : canonical - curves - same - point ] .
suppose we have constructed @xmath2 for @xmath188 .
let @xmath51 be a small disc which contains @xmath199 and intersects only the curves from @xmath94 . via a homeomorphism
, we can identify @xmath51 with the standard unit disc in @xmath200 and the arcs from @xmath94 as straight - line diagonals all intersecting at the center @xmath201 , none of which is vertical . remove a small disc @xmath202 directly above @xmath199 , and construct @xmath183 new simple closed curves as follows . for each arc @xmath101 in @xmath51 , place a new arc @xmath203 in @xmath51 parallel to @xmath101 but above @xmath202 .
see figure [ figure : lowerbound ] . obtain a simple closed curve @xmath77 by continuing @xmath203 outside of @xmath51 along the curve @xmath76 in @xmath94 which contains @xmath101 .
this can be done over all @xmath204 so that : * for all @xmath205 , the corresponding new curves @xmath206 deposit straight lines in @xmath51 and otherwise outside of @xmath51 , the curves @xmath207 are pairwise disjoint .
hence , pairwise intersections among @xmath207 are at most @xmath23 .
* for all @xmath204 and @xmath208 , the new curve @xmath77 has intersection @xmath209 let @xmath190 be @xmath2 with the newly constructed curves added .
each new curve @xmath77 bounds with its `` parent '' @xmath76 an annulus containing the ( removed ) disc @xmath202 . since @xmath28 , the complement of the annulus has genus at least @xmath23 , and thus is not an annulus ; consequently @xmath76 and @xmath77 are non isotopic .
furthermore , @xmath77 is not isotopic to any other curve in @xmath190 since if it were , then , after replacing @xmath202 , we would see that @xmath76 were isotopic to some other curve in @xmath2 , a contradiction .
note that @xmath190 has exactly @xmath183 more curves and @xmath210 still has the desired properties .
notice that this argument fails in @xmath31 only because if @xmath211 , then the complement of the annulus containing @xmath51 would in fact be another annulus .
however , once @xmath32 , the complement would contain a boundary component , and the argument proceeds mutatis mutandis , but with the smaller lower bound of @xmath212 . @xmath61
the cases @xmath213 were established in section [ sec : gen2 ] so assume @xmath214 . let @xmath215 if @xmath0 is even and @xmath216 if @xmath0 is odd and let @xmath217 . by theorem [ theo : boundaryasymptotics ]
, there is a @xmath23-system @xmath2 of size @xmath218 on a surface of genus @xmath219 and @xmath220 boundary components .
gluing @xmath25 handles to the @xmath220 boundary components does not cause curves to become isotopic , the system consisting of @xmath2 and the @xmath25 curves going around the handles is a @xmath23-system of at least @xmath221 curves on a genus @xmath0 surface . @xmath61
joan s. birman and hugh m. hilden . on the mapping class groups of closed surfaces as covering spaces . in
_ advances in the theory of riemann surfaces ( proc .
conf . , stony brook , n.y . , 1969 ) _ , pages 81115 .
ann . of math .
studies , no .
princeton univ . press , princeton , n.j . ,
1971 .
andrew haas and perry susskind .
the geometry of the hyperelliptic involution in genus two .
_ , 1050 ( 1):0 159165 , 1989 .
issn 0002 - 9939 .
doi : 10.2307/2046751 .
url http://dx.doi.org/10.2307/2046751 .
m. juvan , a. malni , and b. mohar .
systems of curves on surfaces .
_ j. combin .
theory ser .
b _ , 680 ( 1):0 722 , 1996 .
issn 0095 - 8956 .
doi : 10.1006/jctb.1996.0053 .
url http://dx.doi.org/10.1006/jctb.1996.0053 . van h. vu .
extremal set systems with weakly restricted intersections .
_ combinatorica _ , 190 ( 4):0 567587 , 1999 .
issn 0209 - 9683 .
doi : 10.1007/s004939970008 .
url http://dx.doi.org/10.1007/s004939970008 . | we give an exponential upper and a quadratic lower bound on the number of pairwise non - isotopic simple closed curves can be placed on a closed surface of genus @xmath0 such that any two of the curves intersects at most once . although the gap is large , both bounds are the best known for large genus . in genus one and two , we solve the problem exactly .
our methods generalize to variants in which the allowed number of pairwise intersections is odd , even , or bounded , and to surfaces with boundary components .
= 1 |
the study of statistical gravitational lensing effects on large numbers of source galaxies due to intervening matter has been a subject of major and rapid progress both from the methodological and from the observational point of view ( e.g. , see kaiser & squires 1993 ; luppino & kaiser 1997 ; taylor et al .
1998 ; lombardi & bertin 1998a , b , hereafter paper i , ii ) .
this indeed offers a very promising approach to study the overall mass distribution , especially on the scale of clusters of galaxies .
in fact , the most appealing aspect of this line of research is its diagnostic potential , with the possibility to measure mass distributions independently of the traditional tools that have to rely on the use of dynamical models .
the interest in this method has gained further momentum from the most recent developments of telescopes and instrumentation .
very deep images are now bringing in accurate data that allow us to analyze the shape of thousands of galaxies , with a density of nearly @xmath0 already reached ( see hoekstra et al .
the use of information from so many sources makes it possible to measure very small lensing effects , which would have been beyond expectation just a few years ago .
the work in this area is thus mostly addressing two directions . on the methodological side ,
the main hope is to develop mass reconstruction algorithms that are simple , flexible , and reliable , with full control of the errors involved in the procedure that goes from the shear data to the inferred properties of the lens ( see , e.g. , bartelmann et al .
1996 ; seitz & schneider 1997 ; paper i ; paper ii ) .
on the side of data analysis , a variety of important issues have to be properly faced and addressed , in order to secure accurate shape measurements as free as possible from all the undesired or spurious distortions that are associated with the instrument and with the observation conditions ( kaiser , squires , & broadhurst 1995 ) . within this general pespective , the present paper belongs to a third direction of research , i.e. the study of potentially interesting effects and phenomena , such as the lensing associated with gravity waves ( kaiser & jaffe 1997 ) or cosmological applications where gravitational lensing probes the geometry of the universe ( kaiser 1998 ; lombardi & bertin 1999 , paper iii ; van waerbeke , bernardeau , & mellier 1999 ) .
measuring these effects , admittedly at the limits of current observations , is expected to become feasible with the advent of the next generation telescopes .
the theory of double lenses has already been subject to some progress .
for example , it has been proven that multiple lenses produce an odd number of images from a pointlike source ( seitz & schneider 1992 ) and that a simple theorem on the magnification of images holds ( seitz & schneider 1994 ) , while crawford et al .
( 1986 ) have discussed the probability that a quasar is strongly lensed by two clusters . however
, this still remains a challenging line of research that is so far largely unexplored ; in particular , the problem has not been discussed , to our knowledge , in the important context of weak lensing analyses .
attracted by the new developments in deep imaging and especially in the observation of very distant clusters of galaxies , which already suggest that a chance alignment of two clusters is not to be considered an unlikely event ( for example , see the case of cl@xmath1 noted by molinari , buzzoni , & chincarini 1996 , and of a@xmath2 reported by wang & ulmer 1997 ) , we found it natural to give further thought to such lens configurations . in this paper , for the context of weak lensing , we generalize the procedure to extract the astrophysical information contained in the observed shear field from the case of a single lens to that of a double lens .
we start by showing how the trace and the asymmetric part of the jacobian matrix associated with the ray - tracing transformation can be measured ( sect . 2 and sect .
3 ) , once the observed ellipticities of a large number of background galaxies have been properly secured .
we then consider some explicit questions that can be tackled in configurations of astrophysical interest ; in particular , we formulate and address a `` dark cluster problem , '' corresponding to the situation where the signature of double lensing is observed at variance with the apparent absence of a second bright cluster partially aligned along the line of sight ( sect .
we then estimate the size of the effects involved ( sect .
5 ) , so that we can state that , based on the known properties of the distribution of bright clusters , a few configurations are likely to be present in the sky , for which the small effects characteristic of double lensing may already be within detection limits . surprisingly , a discussion of the relevant contributions to the noise of the shear measurement reveals an intrinsic limitation of weak lensing analyses .
in fact , we find that the noise contribution associated with the two - galaxy cosmological correlation function does not decrease with the depth of the observation .
a quantitative evaluation of this subtle effect and of its impact on mass reconstructions deserves a separate investigation and will be considered in a future paper . some numerical tests in sect .
6 demonstrate that the analytical framework developed is sound and bring out the existence of some curious effects ( especially on the criticality condition for a double lens and on the possibility of pinpointing the location of the invisible cluster in the dark cluster problem ) .
then , simulations in the same section show that the redshift distribution of the source galaxies , which in principle might also contribute to break the curl - free character of the shear field , actually produces systematic effects typically two orders of magnitude smaller than the double lensing effects we are focusing on .
in addition , they suggest that reasonable assumptions on the observed galaxies lead to noise levels that do not mar the possibility of detecting the double - lensing effect . even if , at present , much of what we have obtained is barely within the limits of direct experimental confirmation , with the advent of new instrumentation ( such as acs on hst ) and powerful telescopes of the next generation ( such as ngst ) the observational impact of analyses of the type provided here is bound to become significant .
in this section we briefly recall the lensing equations for double lenses , in the context where the lensing is produced by intervening clusters . for the purpose
, we mostly follow schneider , ehlers , & falco ( 1992 ) , with some small differences of notation ( see also paper iii ) .
we then summarize the principles of the statistical analysis that leads to a local measurement of the gravitational shear .
let us consider for the moment a single source at redshift @xmath3 .
let @xmath4 be the angular diameter distance between two aligned objects at redshift @xmath5 and @xmath6 , and @xmath7 ( with @xmath8 and @xmath9 ) the distances between two of the elements of the lens configuration , made of observer ( o ) , two deflector planes ( @xmath10 and @xmath11 ) , and source plane ( s ) .
we call @xmath12 ( or simply @xmath13 ) the apparent position of the source , so that @xmath14 is the corresponding linear position on the first deflector plane .
the light ray is traced back to angular positions @xmath15 and @xmath16 referred to the second deflector and to the source ( see fig .
[ fig:1 ] ) . for simplicity ,
in the following we will use the notation @xmath17 for @xmath18 .
the `` dynamics '' of the two lenses is contained in two deflection functions @xmath19 and @xmath20 , such that @xmath21 thus the _ ray - tracing equation _ can be written as @xmath22 with ( @xmath23 ) @xmath24 and @xmath25 for given projected mass distributions of the two lenses @xmath26 , the distances @xmath27 enter the problem through @xmath28 and through the two critical densities , defined as @xmath29 the definition of @xmath30 for @xmath31 just states that foreground sources are unaffected by the lens .
the functions @xmath32 can be expressed by suitable integrals of the two reduced densities @xmath33 . for cases of interest , @xmath28 is smaller than unity .
note that , much like in the case of a single lens , since distances enter only through the ratios @xmath34 and @xmath35 , for a family of very distant sources the ray - tracing equation can be applied without explicit reference to the distance of each source , and @xmath28 remains finite .
if the sources are located at different redshifts but not all of them are at large distance , we can still retain a form similar to eq .
for the ray - tracing equation provided we introduce the appropriate _ cosmological weight functions _ : @xmath36 the functions @xmath37 give the strength of the lens @xmath38 on a source at redshift @xmath39 relative to a source at redshift @xmath3 .
[ note that in paper iii we used a similar definition of cosmological weight function , but took the `` reference '' redshift @xmath3 at infinity . here
we prefer to avoid the limit @xmath40 . ] with this definition , eq . can be rewritten as @xmath41 an important qualitative aspect of the two - lens ray - tracing equation is the following . in the limit where @xmath42 , the ray - tracing equation adds two contributions , @xmath43 and @xmath44 , each deriving from a potential , so that the related jacobian matrix @xmath45 is symmetric . in turn , when @xmath46 , the jacobian matrix is no longer guaranteed to be symmetric ( in fact , the product of two symmetric matrices is not necessarily symmetric ) .
the limit @xmath47 occurs when the two deflectors are very close to each other . in the _ weak lensing limit _
the jacobian matrix is considered to be close to the identity matrix , @xmath48 , with the parameter @xmath49 measuring the strength of the lens .
equation readily shows that for two lenses of comparable strength one expects a very small asymmetry in the jacobian matrix , with @xmath50 . in the following
we will find that , in spite of the smallness of the effect involved , a weak lensing analysis , carried out to second order , can provide interesting indications as to the possibility of detecting , in realistic cases , the presence of a double lens through its signature of an asymmetry in the jacobian matrix .
simulations will show that such indications are further encouraged when the lenses involved are relatively strong .
consider a large number @xmath51 of extended sources all subject to the _ same _ jacobian matrix @xmath52 for the relevant ray - tracing .
in general , as noted above , @xmath53 .
let @xmath54 be the measured quadrupole moments of the individual galaxies .
these observed quantities are related to the source ( unlensed ) quadrupole moments @xmath55 by the equation ( see schneider , ehlers , & falco 1992 ; see also paper i ) @xmath56 in the limit @xmath57 , the mean value of the source quadrupoles for an isotropic population of source galaxies should be proportional to the identity matrix @xmath58 with @xmath59 a positive constant .
this is the starting point of the @xmath60-method described in paper i and allows us to invert the relation for the mean values @xmath61 into @xmath62 note that the matrix @xmath63 is symmetric .
here we have taken @xmath64 to be non - singular .
the process of mass reconstruction makes use of the departures from the identity matrix of the ray tracing matrix as brought out by the observed quadrupoles . in practice ,
the true jacobian matrix @xmath65 is considered to be unknown .
equation shows that if @xmath65 is a solution , any matrix @xmath66 obtained by multiplication by an orthogonal matrix ( so that @xmath67 ) is also a solution .
thus in the determination of the shear associated with gravitational lensing the jacobian matrix is bound to be identified only up to an orthogonal matrix .
it is easy to show that any solution @xmath52 for a problem for which @xmath65 is a solution can be written as @xmath68 , with @xmath69 a suitable orthogonal matrix .
the same set of data @xmath70 can be analyzed by a standard single lens procedure , leading to the determination of a symmetric jacobian matrix @xmath71 , with the property @xmath72 .
then @xmath71 is to be related to the true jacobian matrix @xmath65 by means of an orthogonal matrix .
in fact , for a given @xmath65 , there are in general four different symmetric matrices available ( corresponding to matrices that differ by the sign of their eigenvalues ) .
if we restrict the attention to transformations with positive determinant ( the @xmath73 ambiguity reflects the well - known @xmath74 invariance ; see schneider & seitz 1995 ) , we can write @xmath75 with @xmath76 here @xmath77 is the totally antisymmetric tensor of rank 2 and @xmath78 is the trace of @xmath65 ( we use the summation convention on repeated indices ) . note that the quantity @xmath79 is a measure of the asymmetry of @xmath65 .
the symmetric matrix can thus be given in explicit form : @xmath80 with @xmath81 in what has been described so far , @xmath52 is considered to be the same for all the source galaxies .
therefore at this stage we have addressed only the problem of a _ local _ analysis , applicable to source galaxies very close to each other , in a small patch of the sky , and at a similar redshift . in conclusion
, we have so far shown that a local measurement can not lead to discovering a double lens , because there is always a way to interpret the data by means of a symmetric ray - tracing jacobian matrix .
let us first consider only sources located at a given redshift @xmath39 . in the weak lensing limit ,
the jacobian matrix for such sources is of the form @xmath82 , and the asymmetry is small , @xmath83 .
thus we can write the related symmetric matrix ( the `` effective '' jacobian matrix ) as @xmath84 with ( see eq . ) @xmath85 therefore the elements of the symmetric matrix ( appearing in the mass reconstruction analysis ) can be expressed in terms of those of the true jacobian matrix , as @xmath86 from inspection of eqs .
and and from the definition of the _ true _ jacobian matrix , we find that the _ true _ source positions are related to @xmath87 and to @xmath88 by @xmath89 thus , if we refer to the vector field @xmath90 commonly used in weak lensing analyses , we find @xmath91 if we now combine the definition of @xmath90 with the above relations for @xmath87 and @xmath88 we get @xmath92 the last two equations replace the well - known relations applicable to single lens analyses , i.e. @xmath93 and @xmath94 .
suppose now that the source galaxies follow a redshift distribution @xmath95 .
at each redshift , we can apply eqs .
( [ eq : b17][eq : b24 ] ) provided that all lensing quantities , such as @xmath65 , @xmath87 , @xmath96 , and @xmath88 , are calculated for each source at the correct redshift . in the weak lensing limit
, the shear @xmath96 is estimated from the ellipticities of galaxies that are observed in a particular region of the sky .
since source galaxies have different redshifts , in reality a mean value of the shear is measured : @xmath97 where @xmath98 is the redshift - dependent shear . in the weak lensing limit
, all quantities depend linearly on the shear , and thus eqs . and can be written as @xmath99 unfortunately , these relations can not be used to infer directly , from a given set of data , the amount of asymmetry that would be a characteristic signature of multiple lensing .
the reason is that for this purpose the data should be able to identify the vector field @xmath90 up to @xmath100 , because we know that @xmath101 . in practice , it is well - known that standard weak lensing analyses lead to the determination of @xmath90 only up to @xmath102 .
in fact , observations lead to the determination of the _ reduced shear _ @xmath103 ; the identification of @xmath104 with @xmath96 is correct only to @xmath102 .
thus the above analysis is not yet ready for practical applications .
for this reason we need to extend the discussion so as to include the _ second order _ terms in the weak lensing expansion . in principle , the discussion can be carried out by retaining the redshift dependence of sources . in practice ,
such approach would lead us far beyond the original purpose of this paper .
for this reason we will suppose , in the following , that all sources are located at redshift @xmath3 ( in sect .
5 we will consider again the spread of sources in redshift for the single lens case ) . for a population of source galaxies located at a single redshift ( even in the case of strong lensing
; but in this section , we recall , we are still within the weak lensing expansion ) an `` observable '' field is ( see kaiser 1995 ) @xmath105 for a strong single lens , such field has the important property that @xmath106 $ ] , with @xmath107 the dimensionless projected density that one aims at reconstructing . curiously , _ to second order _ in the weak lensing expansion ( see appendix a ) , it is possible to show that , for multiple weak lenses of comparable strength , @xmath108 in other words , _ we have derived a set of equations for which the contribution of a small asymmetry present in the true jacobian matrix is retained consistently _ ; the structure is similar to that of the set for @xmath90 field presented in eqs . and .
_ the new set of eqs . and is the basis that allows us to generalize the procedure to extract the available information , contained in the observed shear field , from the case of a single lens to that of a multiple lens and thus to investigate quantitatively the characteristics of the coupling of two or more deflectors located along the same line of sight_. the symmetric limit , which we may call in this context the single lens limit , is easily recognized . the results of this section ( eqs . and
) are not generalized easily to the case of sources distributed according to a @xmath95 . however , the simulations described in sect .
6.2 will basically support a description analogous to that given by eqs . and .
the conclusions of this section , that the data contained in the _ shear map _ can be used to detect an asymmetry in the true jacobian matrix @xmath65 , and of the previous sect . 2.2 , that such a detection is impossible
if based on a local measurement only , do not depend on the number of deflectors involved in a multiple lens . in the following
, we will specialize our conclusions to the study of double lenses , for which the ray - tracing equation can be handled in a straightforward manner .
in this section we formulate some questions that may be interesting from the astrophysical point of view . the expected size of some effects and the possibility of an actual measurement will be addressed in separate sections at the end of the paper
. the prime signature of double lensing would be the detection of a significant asymmetry @xmath88 ( see eq . ) .
because of statistical errors , the measured vector field @xmath109 is bound to be associated with a non - vanishing curl .
therefore , a positive detection of asymmetry can be claimed only if the expected statistical error on @xmath88 is smaller than the true value @xmath110 characterizing the double lens .
a quantitative analysis of this condition will be provided in sect . 5 below .
here we only note that , in general , for two clusters of comparable strength , at different distances , with offset centers with respect to the line of sight , the regions on the sky where the signal - to - noise ratio for @xmath88 should be largest are those , on either side , just off the line connecting the cluster centers ( see fig . [ fig : density2 ] and fig .
[ fig : dens_plot ] ) as we have seen in sect . 2
, the asymmetry is expected to be weak ( @xmath111 for two clusters of comparable strength @xmath49 ) .
the optimal conditions for detecting double lensing are then : * the geometric parameter @xmath28 should be close to unity , i.e. , for a given distribution of sources , the distance between the two clusters should not be much smaller than the distance from the observer to the near cluster .
this condition is reasonably satisfied if , for example , the redshifts of the two clusters are in the following relation , @xmath112 .
* the two clusters should be not too weak .
it is preferable to consider cases where the dimensionless densities , @xmath113 and @xmath114 , are of order unity . *
the cluster centers should be offset , but there should be a region of significant overlap between the two clusters in the sky .
given the above criteria , we now estimate the probability that a double lens with desired properties be observed . for the purpose
, we refer to @xmath115 for the hubble constant , within a friedmann - lematre cosmological model characterized by @xmath116 and @xmath117 .
first , we suppose that a nearby cluster is observed at redshift @xmath118 ( the values of the various quantities used here are similar to the values used in the simulations to be described in sect .
[ sect : simulations ] ) and ask what is the probability of finding a cluster at redshift @xmath119 well aligned with the first cluster ; in considering this question , we further require that the second cluster be sufficiently massive , with mass greater than @xmath120 . for the purpose , we assume that the angular distance between the centers of the two clusters be between @xmath11 and @xmath121 and that the redshift of the second cluster be in the range @xmath122 .
a simple calculation then shows that the center of the second cluster must be inside a comoving volume with size @xmath123 .
therefore , based on the cluster density found by borgani et al .
( 1999 ; see also girardi et al .
1998 ) , the expected probability of finding a `` good '' double lens turns out to be @xmath124 . if we consider less strict requirements , with an angular distance between the two cluster centers in the range @xmath10@xmath125 and a redshift range @xmath126@xmath127 for the second cluster , the estimated probability increases up to @xmath128 .
these estimates should be combined with the number of massive clusters at small redshift ( for the object observed at @xmath129 ) that should be available over the whole sky .
the comoving volume for a shell with redshift between @xmath130 and @xmath131 is @xmath132 ; correspondingly , the total number of clusters with mass above @xmath133 in such a volume is about @xmath134 . as a result , we expect from @xmath10 to @xmath135 cases of double lenses with detectable effects and configurations similar to the one considered in our simulations . from the way we have approached the problem , it is clear that the estimate just obtained provides only a _ lower limit _ with respect to the number of good cases expected for the purposes of the present paper .
suppose that a set of data leads to the determination of @xmath136 and that , based on suitable boundary conditions ( e.g. , @xmath137 , @xmath138 for @xmath139 ) , we integrate eqs . and and get the functions @xmath140 and @xmath141 . for a single lens problem the lensing analysis , at this stage , would be complete , since we would have the dimensionless projected mass distribution @xmath142 associated with the deflector ; knowledge of the critical density @xmath143 would lead to the projected mass distribution @xmath144 .
in the case of a double lens , what would be the astrophysically interesting quantities that could be derived ? for the following discussion , the ray - tracing equation can be re - cast in a more transparent form in the case @xmath145 ( so that @xmath146 ) .
if we introduce the two _ curl - free lensing maps _ ( we recall that @xmath147 ) @xmath148 and the _ effective ray - tracing map _ @xmath149
we see that eq .
can be written as @xmath150 in other words , ray - tracing for a double lens can be seen as the composition of two curl - free functions .
the double lens problem in its _ direct _ formulation involves two ( dimensionless ) ray - tracing functions ( @xmath43 and @xmath44 ) , two critical densities ( @xmath151 and @xmath152 ) , and one additional geometric parameter @xmath28 .
any inverse problem would be under - determined , if we start from the two dimensionless functions @xmath140 and @xmath141 alone . even if the geometry is taken to be known ( for example , when the distances to the two clusters acting as deflectors are known ) , it is not possible from the pair @xmath153 to `` reconstruct the two lenses , '' i.e. to get the ray - tracing functions @xmath154 and @xmath155 separately .
this statement is easy to understand , especially if we refer to the curl - free lensing maps @xmath156 and @xmath157 of sect .
in fact , a simple scaling invariance @xmath158 , @xmath159 shows that , unless we are able to provide suitable boundary conditions on the functions @xmath160 ( e.g. , @xmath161 for large @xmath162 ) , there is no way to disentangle from the shear data alone the contributions of the two lenses .
obviously , this does not mean that the lensing analysis is useless ; it is only a reminder that in this more complex situation an unambiguous mass reconstruction based on weak lensing would require additional input from other probes of mass distributions ( e.g. , x - ray data ) . in order to clarify this point further
, it may be instructive to consider the following simple examples , all characterized by @xmath163 .
suppose that @xmath164 with @xmath165 a real continuous function such that @xmath166 .
then it is natural to choose @xmath156 and @xmath157 to be radial , i.e. @xmath167 where @xmath168 and @xmath169 are real functions .
supposing that @xmath170 for all @xmath171 , we can rewrite eq .
as @xmath172 thus reduced to a one - dimensional problem , it is clear that this equation in general admits an infinite number of solutions .
consider an effective ray - tracing map ( see eq . )
such that @xmath174 . in this case
we can solve eq . by choosing an invertible curl - free map @xmath157 and by writing @xmath175
we now require that @xmath156 be curl - free . from the above equation
, this happens if and only if the product of the jacobian matrix @xmath176 , associated with @xmath177 , and @xmath178 , associated with @xmath179 , is a symmetric matrix , i.e. @xmath180 from linear algebra we know that this is equivalent to saying that @xmath181 and @xmath178 have the same eigenvectors . since @xmath157 is curl - free ,
let us introduce the potential @xmath182 so that @xmath183 and @xmath184 . if @xmath185 is a _ local _ system of coordinates where @xmath178 is diagonal , then eq
. is satisfied if @xmath186 is diagonal in the same system of coordinates , i.e. if @xmath187 .
this simply states that @xmath182 is _ separable _ in @xmath188 and @xmath189 , i.e. that we can write @xmath182 as the sum of an arbitrary function of @xmath188 and of an arbitrary function of @xmath189 .
this again demonstrates the freedom at our disposal in solving eq . .
in addition , one can now better appreciate why , even when natural boundary conditions are specified , in general there is no guarantee that the solution be determined uniquely ; in fact , several solutions are expected when the coordinate system @xmath190 is associated with _
( such as the point @xmath191 for polar coordinates ) .
one curious case that may be imagined is the possibility of combining two lenses with no net effect , so that @xmath192 ( see eq . ) . in principle
, one may argue that from any invertible curl - free map @xmath193 one can take a second lens characterized by @xmath194 .
this idea would obviously generate an infinite number of solutions since the starting function @xmath156 is at our disposal .
however , if we require that the density distributions associated with the two lenses be both _ positive definite _
, i.e. that @xmath195 , it can be shown that there is no way for two lenses to exactly compensate for each other : in other words , the only admissible solution is the trivial @xmath196 .
this conclusion is consistent with the theorem that states that any combination of gravitational lenses is bound to produce a net magnification ( @xmath197 ; see seitz & schneider 1992 ) .
a different problem based on double lenses can be formulated in the following way .
suppose that one observes a lensing cluster and that in the process of producing the mass reconstruction one finds evidence that the vector field @xmath109 is not curl - free , well above the expected errors .
clearly , in such a situation , one possibility is that a separate mass concentration , which we may call a _ dark cluster _ , is responsible for the effect .
we may now consider the case when the mass distribution of the visible cluster is well constrained by diagnostics independent of lensing ( e.g. , by x - ray data ) . under these circumstances , what can we tell about the dark cluster properties from the observed lensing effects ?
in particular , is it possible to derive the location and the mass distribution of the invoked dark cluster ?
if the mass distribution of the luminous cluster is taken to be known , one has either the function @xmath198 or the function @xmath199 , depending on whether the dark cluster is near or far ; in either case , the deflection associated with the dark cluster @xmath200 depends implicitly ( see eqs .
) on the geometric parameter @xmath28 , which is unknown .
one way to obtain the value of @xmath28 is to impose that the lensing map associated with the dark cluster is curl - free , i.e. that either @xmath201 = 0 \ ; , \ ] ] or that @xmath202 = 0 \ ; , \ ] ] depending on whether we guess the dark cluster to be near or far . in sect .
6 below , by means of a simulated case , we will show how @xmath28 can be determined with reasonable accuracy , by minimizing the square of the left side of the above equations .
in other words , under the above circumstances _ we will demonstrate that a weak lensing analysis allows us to pinpoint the location of the invisible cluster and to reconstruct its mass distribution_. in closing this section , we may note that the problem of a _ single _ dark cluster , i.e. the case of a matter concentration detected by lensing effects without a visible counterpart for the lens , is less constrained and , in this respect , less interesting than the case discussed above in the sense that the usual mass reconstruction would only lead to the dimensionless projected density @xmath203 , with no hope to derive the distance to the dark cluster that is invoked and the actual scale of the mass involved .
in this section we calculate the expected order of magnitude for the measurable effects associated with double lensing . for simplicity
, the following section discusses the noise properties of quantities related to the vector @xmath204 .
the results obtained are the leading order estimates for the desired quantities related to @xmath109 . the following subsection deals with a population of source galaxies located at a single redshift .
effects related to a spread in redshift will be estimated in sect .
5.3 . consider a single lens , characterized by true jacobian matrix @xmath65 , so that the shear field @xmath206 has vanishing curl , i.e. @xmath207 . because of the finite number of source galaxies used and of the smoothing introduced in the reconstruction process , we expect that the measured @xmath204 in general differs from the true field @xmath206 . in order to calculate the expectation value and the variance of @xmath205
, we may use a technique similar to that described in paper ii .
we first recall that in the reconstruction process eq . is used together with a ( positive ) weight function @xmath208
. such weight function , defined so that @xmath209 is significantly different from zero only for @xmath210 close to @xmath13 , enters in the local averages of quadrupoles ( or ellipticities ) . in particular , if we resort to the so - called q - method associated with eq .
( see paper i ) , the local average quadrupole @xmath211 is calculated from the relation @xmath212 here the superscripts @xmath213 label the galaxies affected by lensing . in paper ii we have calculated expectation values and variances of several quantities , under the hypothesis that the weight function is invariant upon translation , so that @xmath214 , and that it is _ normalized _ , i.e. @xmath215 here @xmath216 is the density of galaxies ( taken to be constant over the field ) .
in particular , we have shown that the expectation values of the relevant quantities ( in particular , the measured shear , the field @xmath204 , and the reconstructed mass density @xmath203 ) are equal to the true quantities _ smoothed _ by the weight function , e.g. @xmath217 from this expression it can be easily shown that the expected value for the measured curl of @xmath204 vanishes .
in fact , @xmath218 here we have used the symmetry property of convolutions .
thus , for a single lens , we expect to measure ( in the average ) a field @xmath204 with vanishing curl
. note that , for a double lens , a similar calculation would show that @xmath219 let us now turn to covariances . using the same notation as in paper iii
( see also lombardi 2000 ) , we call @xmath220 the covariance of the ellipticity distribution for the population of source galaxies ( @xmath221 ) . there
we have shown that the covariance of the shear @xmath96 is @xmath222 note that the covariance of @xmath96 , which we may write in short as @xmath223 , depends only on @xmath224 .
then the covariance of @xmath204 is related to the covariance of @xmath96 in a simple way : @xmath225 similarly , it can be shown that @xmath226 the last expression is the basis of the following discussion .
in fact , we may argue that a lens can be considered to be double only if the measured value of @xmath205 is significantly larger than the expected error on this quantity . if we refer to eqs .
, as definitions of @xmath87 and @xmath88 , we thus find ( under optimal conditions defined in paper ii ) @xmath227 a more useful expression can be found in the simple case of a gaussian weight function : @xmath228 in this case we have @xmath229}{64 \pi \rho \sigma_w^{10 } } \times { } \nonumber \\ & { } \times \exp \left ( - \frac{\| { \mbox{\boldmath $ \delta$\unboldmath } } \|^2 } { 2 \sigma^2_w } \right ) \ ; , \end{aligned}\ ] ] where we have called @xmath230 .
note that @xmath205 shows anti - correlation for @xmath231 in the range @xmath232 .
similar noise properties are associated with the quantity @xmath233 .
now we are finally able to set a quantitative condition for the detection of double lensing .
the requirement that the error on @xmath88 ( which can be estimated from eqs . and ) be smaller than the expected value of @xmath88 , within factors of order unity , can be written as : @xmath234 note that @xmath235 represents the number of galaxies effectively used in the average of eq . .
if we refer to a case with @xmath236 and to the currently `` realistic '' density @xmath237 , with @xmath238 , we thus find that a significant signal requires the use of a gaussian weight function with smoothing size @xmath239 not smaller that @xmath240 . on the other hand ,
the interesting effects are those associated with the structure of the field @xmath141 , which is associated with a length scale at best comparable with that of the clusters under investigation .
therefore , we may argue that the most promising way to identify the effects of double lensing is to work with the smallest smoothing size @xmath239 compatible with condition . these arguments and the applicability of the condition defined by eq .
are clarified and further supported by the simulations that will be presented in sect .
so far , in this section , we have referred to a sheet of source galaxies at a given distance . in real situations where
the source galaxies are distributed at different redshifts , we expect three separate effects that can have a significant impact on our analysis : * an extra source of noise , related to the distribution in redshift of the individual sources , is added to our data . as a result , the measured shear will have a larger covariance than the one calculated in eq . , and
thus a larger covariance is also expected for @xmath205 . *
an additional effect is induced by the extra - clumping of the sources described by the cosmological two - point correlation function ( see peebles 1980 ) .
* outside the limit of very weak lensing , there is actually no guarantee that @xmath241 _ even in the standard case of a single lens_. this happens because the observable is , in this case , @xmath242 , which is not simply related to @xmath243 and @xmath244 .
the first two items contribute to the estimate of the relevant covariances while the third item adds a bias ( to be discussed in sect .
5.3.2 ) . below we will estimate the covariance of the measured shear @xmath96 when the source galaxies are located at different ( unknown ) redshifts .
for the purpose , we assume that the redshift distribution of galaxies is known . in particular , we write the probability to find one galaxy in an area @xmath245 of the sky with redshift between @xmath39 and @xmath246 as @xmath247 here we model the galaxy angular distribution in terms of a constant galaxy density @xmath216 on the lens plane .
the probability to find two galaxies , one in a patch @xmath248 of the sky with redshift in the range @xmath249 $ ] , and one in a patch @xmath250 with redshift in the range @xmath251 $ ] , can be written as @xmath252 \bigr [ 1 + \xi(r ) \bigr ] \ ; .\ ] ] here @xmath253 is the two - point correlation function of galaxies at mutual ( 3d ) distance @xmath254 . for distances smaller than @xmath255 , this function
can be well approximated as a power law ( see peebles 1993 ) @xmath256 where @xmath257 is the reduced hubble constant .
we will now evaluate @xmath223 in the weak lensing limit . in this approximation ,
the observed ellipticities are related to the unlensed ones by the relation @xmath258 and thus the expected covariance on @xmath49 is given by @xmath259 a plot of @xmath260 as a function of the lens redshift for a `` typical '' redshift distribution @xmath95 ( see eq .
below ) is shown in fig .
[ fig : var_wz ] .
a simple estimator of the shear is given by ( see seitz & schneider 1997 ) @xmath261 since the weight function @xmath262 is assumed to be normalized following eq .
, this is an _ unbiased _ estimator .
in fact ( for the procedure of `` spatial averaging '' see paper ii ) @xmath263 thus we recover one simple result already discussed in paper ii : the measured shear is the smoothing of the true shear with the spatial weight function @xmath262
. turning to the covariance of @xmath96 , we find @xmath264 w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } , { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) \times { } \\ & \bigl [ \langle w \rangle_z \gamma_{0j}\bigl ( { \mbox{\boldmath $ \theta$\unboldmath}}^{(m ) } \bigr ) - \epsilon^{(m)}_j \bigr ] w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } ' , { \mbox{\boldmath $ \theta$\unboldmath}}^{(m ) } \bigr ) \biggr\rangle \ ; .
\end{split}\ ] ] inserting here eq .
, we find a long expression composed of three types of terms : terms of the form @xmath265 , terms of the form @xmath266 , and terms independent of @xmath267 . because of the isotropy hypothesis
, the first contribution vanishes unless @xmath268 and @xmath269 , and the second always vanishes .
thus we finally find @xmath270 \gamma_{0i } \bigl ( { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } , { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) \times { } \\ & \phantom{\displaystyle{\frac{1}{\langle w \rangle_z^2 } \biggl\langle \sum_{n , m } } } \bigl [ w \bigl ( z^{(m ) } \bigr ) - \langle w \rangle_z \bigr ] \gamma_{0j } \bigl ( { \mbox{\boldmath $ \theta$\unboldmath}}^{(m ) } \bigr ) w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } ' , { \mbox{\boldmath $ \theta$\unboldmath}}^{(m ) } \bigr ) \biggr\rangle \ ; .
\end{split}\ ] ] this equation is composed of two terms .
the first is related to the intrinsic spread of source ellipticities , and in fact it is proportional to @xmath271 .
the second term ( second and third lines in eq . )
is instead related to the spread in redshift of galaxies . in order to obtain a more general expression ,
we adopt a technique already used in paper ii and average the expression for @xmath223 over the source positions @xmath272 and redshifts @xmath273 . as a result ,
the first term becomes simply @xmath274 this term has already been studied in detail in paper ii . the second term of eq .
deserves a more detailed discussion . in order to carry out the average over the source redshifts ,
we need to split the sum over @xmath275 and @xmath276 in two sums , one involving the terms for which @xmath268 , and the other involving the terms @xmath277 . for the first terms we have @xmath278 ^ 2 \gamma_{0i } \bigl ( { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) \gamma_{0j } \bigl ( { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } , { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } ' , { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) \biggr\rangle \longmapsto \\
\longmapsto \frac{1}{\langle w \rangle_z^2 } \rho { \mathrm{var}}(w ) \int_\omega \gamma_{0i } ( { \mbox{\boldmath $ \theta$\unboldmath } } '' ) \gamma_{0j } ( { \mbox{\boldmath $ \theta$\unboldmath } } '' ) w({\mbox{\boldmath $ \theta$\unboldmath } } , { \mbox{\boldmath $ \theta$\unboldmath } } '' ) w({\mbox{\boldmath $ \theta$\unboldmath } } ' , { \mbox{\boldmath $ \theta$\unboldmath } } '' ) \ , { \mathrm{d}}^2 \theta '' \ ; .\end{gathered}\ ] ] note that this term is related to single galaxies , and thus does not involve the correlation function @xmath253 . on the other hand , for pairs of galaxies ( i.e. if @xmath277 in the last term of eq . )
we find a rather complicated expression @xmath279 \gamma_{0i } \bigl ( { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } , { \mbox{\boldmath $ \theta$\unboldmath}}^{(n ) } \bigr ) \bigl [ w \bigl ( z^{(m ) } \bigr ) - \langle w \rangle_z \bigr ] \gamma_{0j } \bigl ( { \mbox{\boldmath $ \theta$\unboldmath}}^{(m ) } \bigr ) w \bigl ( { \mbox{\boldmath $ \theta$\unboldmath } } ' , { \mbox{\boldmath $ \theta$\unboldmath}}^{(m ) } \bigr ) \biggr\rangle \longmapsto \\
\longmapsto \frac{1}{\langle w \rangle_z^2 } \rho^2 \int_0^\infty \!\ !
p(z '' ) \ , { \mathrm{d}}z '' \int_\omega \ !
\theta '' \int_0^\infty \!\ ! p(z ' '' ) \ , { \mathrm{d}}z ' '' \int_\omega \ ! { \mathrm{d}}^2 \theta ' '' \bigl [ w(z '' ) - \langle w \rangle_z \bigr ] \gamma_{0i}({\mbox{\boldmath $ \theta$\unboldmath } } '' ) w({\mbox{\boldmath $ \theta$\unboldmath } } , { \mbox{\boldmath $ \theta$\unboldmath } } '' ) \times { } \\ \bigl [ w(z ' '' ) - \langle w \rangle_z \bigr ] \gamma_{0j}({\mbox{\boldmath $ \theta$\unboldmath } } ' '' ) w({\mbox{\boldmath $ \theta$\unboldmath } } ' , { \mbox{\boldmath $ \theta$\unboldmath } } ' '' ) \xi(r ) \ ; .\end{gathered}\ ] ] in this expression @xmath254 is the distance between two galaxies at redshifts @xmath280 and @xmath281 and angular positions @xmath282 and @xmath283 .
the covariance of @xmath96 is the sum of the terms in eqs .
( [ eq:16][eq:18 ] ) , each of which , as discussed above , arises from a different source of noise . in summary ,
* the contribution of eq .
is related to the spread in the intrinsic ellipticities of source galaxies ( parameterized by @xmath220 ) .
it is proportional to @xmath284 ( see the normalization condition ) and has a correlation length proportional to the correlation length of @xmath262 . note that this term is proportional to @xmath285 , so that no correlation between @xmath286 and @xmath287 arises . *
the contribution of eq .
is related to the spread in redshift of the sources ( parameterized by @xmath260 ) .
it is proportional to @xmath284 , has the same correlation length of the previous term , and , in addition , is proportional to @xmath288 .
this term is not diagonal . *
the contribution of eq .
is due to the correlation between galaxies ( described by the two - point correlation function @xmath253 ) .
it is independent of the density of galaxies ( provided @xmath253 does not depend on @xmath216 ; see comments by peebles 1993 after eq .
( 19.39 ) ) , has a correlation length somewhat longer than that of the other terms and is proportional to @xmath288 .
this term is not diagonal .
a complete description of the effect of the first term has already been provided in paper ii .
its effect for the detection of a double lens was analyzed in sect .
[ sec : expect - vari - nabla ] .
the order of magnitude of the ratio between the term of eq . and the one of eq .
can be argued to be @xmath289 typical values for @xmath220 are @xmath290 , and typical values for @xmath260 can be estimated from fig .
[ fig : var_wz ] .
thus , if we refer to a single lens at redshift @xmath291 with a shear of the order of @xmath292 , we find @xmath293 .
note that , since calculations here are performed in the weak lensing approximation , the shear can not be much larger than @xmath294 , and thus the ratio obtained should be taken as an ( approximate ) upper limit . in any case , the effects of this term will be included in the simulations that will be described in sect .
6.2 . as to the third term @xmath295
, it is clearly very complex , both in relation to its mathematical structure and to the physical ingredients involved .
therefore , instead of attempting here an order of magnitude estimate for it , we prefer to postpone a complete discussion of its impact to a future paper that will make use of suitable simulations in the cosmological context .
another effect encountered when source galaxies are `` placed '' at different redshifts is a bias on @xmath205 , which is not bound to vanish any more .
this effect could , in principle , be misinterpreted as the signature of a double lens .
do we have to worry about this contribution ?
note that if this effect turned out to be significant , it would have an important impact on all single lens mass reconstruction methods , because these are all based on the curl - free character of the shear induced by gravitational lensing . in order to be more specific about the origin of this effect , we start from some well - known results ( seitz & schneider 1997 ; see also paper iii ) .
let us consider a sample of galaxies with redshift distribution @xmath95 , behind a single cluster at redshift @xmath296 . as the critical density @xmath297 depends on the source redshift , the lens acts with differential strength on the lensed galaxies ( see sect .
. the cosmological weight function @xmath298 controls the ray - tracing transformation .
in particular , the ( redshift dependent ) reduced shear takes on the form @xmath299 $ ] .
the fact that @xmath298 enters the expression of @xmath300 nonlinearly complicates the lensing analysis .
note that even though the relevant ray - tracing transformation for a single lens is symmetric ( e.g. , see eq .
( 9 ) of paper iii ) , because of the spread in redshift distribution the arguments generally used to identify a curl - free field @xmath109 do not hold any more .
in fact , if we observe a single lens and just ignore the redshift distribution of the sources , we would naively start by identifying the observed average ellipticity with the reduced shear , using the relation @xmath301 .
this mistake would lead in general to a quantity @xmath109 , defined by eq . , with non vanishing curl . for subcritical lenses seitz & schneider ( 1997 )
have shown that an accurate , but approximate , lensing analysis can be carried out _ as if _ all galaxies were at a single redshift , provided the reduced shear @xmath302 in eq .
is evaluated from the average ellipticity following the relation @xmath303 .
correspondingly , the dimensionless density distribution obtained from the lens reconstruction changes by a factor @xmath304 .
if the above corrected relation between average ellipticity and reduced shear entering the definiton of @xmath136 is kept into account , then , in this approximate description , the `` spurious curl '' ( that would be associated with the naive use of @xmath305 ) is eliminated . in order to evaluate the magnitude of the effects ( described in this subsection ) associated with the spread in the distances of the source galaxies , we have carried out simulations that will be described in the next section . here
we summarize the main results . using a reasonable redshift distribution for the sources , and for a wide range of redshifts @xmath296 of the single deflector
, we find an extremely small value of @xmath306 .
for example , for a nearly critical single lens at redshift @xmath307 the value of the relevant curl is about @xmath308 times smaller than the value associated with the double lens effect addressed in this paper .
the nearly critical condition adopted in this example certainly overemphasizes the signal , with respect to the case of weaker lenses , and is unfavorable to the approximation suggested by seitz & schneider .
still , if we apply the simple correction factor @xmath309 described above , we gain an extra factor of @xmath310 .
we can thus state that the effect of the redshift distribution discussed in this subsection , in the example considered , is contained to be about @xmath311 times smaller than the effect of double lensing we are interested in . from this
we conclude that the simple arguments and derivations provided in the main part of the paper are not affected by the spread in the distances of the source galaxies significantly .
the main goal of this section is to check the analytical framework developed in the paper and to provide convincing evidence that effects characteristic of double lensing are within reach of the observations .
for this reason we follow two types of investigation : a semi - analytical approach ( sect .
6.1 ) , where we test the applicability of our asymptotic analysis , and a study ( sect .
6.2 ) that may be seen as a numerical observation , where we simulate the distorsions of a set of source galaxies in a given field under realistic conditions ( see below ) and we perform the statistical lensing analysis on the observed ellipticities . for simplicity ,
the simulations are carried out under the assumption that no clustering is present in the source population ( eq . with @xmath312 ) .
a first test is based on the following steps of a semi - analytical approach : * the population of source galaxies is taken to be at a single redshift ( @xmath313 ) . * a simple model of a double lens
is chosen . for simplicity ,
two centrally - symmetric projected mass distributions are used , within a favorable geometric configuration ( see sect .
* the two clusters are taken to be located at redshift @xmath314 and @xmath315 . since the source population and the double lens we have in mind
have such large distances , the values of the cosmological parameters @xmath316 and @xmath317 need to be specified . in the following we refer to the case @xmath116 , @xmath117
; moreover , we adopt @xmath318km@xmath319s@xmath320mpc@xmath321 .
* the two functions @xmath43 and @xmath44 are calculated .
thus the ray - tracing function @xmath322 is derived ( see eq . ) . *
the jacobian matrix @xmath323 is calculated from the map @xmath322 .
this allows us to calculate the various quantities defined in eqs .
( [ eq : b18][eq : b22 ] ) . in the following ,
these are taken to hold as definitions also for strong lenses . *
the `` effective '' jacobian matrix @xmath324 is obtained from @xmath323 using eq . .
* from @xmath324 , the reduced shear @xmath302 and the field @xmath325 are calculated .
* the divergence and the curl of @xmath109 are compared with the laplacians of @xmath326 and @xmath88 ( eqs . and ) .
note that , by following this procedure , we are considering the `` effective '' jacobian matrix @xmath71 _ without errors_. we do so , because errors on @xmath71 are discussed separately ( see next subsection 6.2 , and paper i and ii ) and because we are focusing here mostly on a test of the adequacy of eqs . and .
let us now describe some of the steps in further detail .
the two clusters , taken to be located at redshift @xmath327 and @xmath315 ( so that the value of the parameter @xmath28 defined in eq .
is @xmath328 in the assumed cosmological model ) , have been modeled using the following profiles ( close to isothermal profiles ) for the projected density ( see schneider , ehlers , & falco 1992 ) : @xmath329 \kappa_\mathrm{c } \over \bigl [ 1 + ( \theta / \theta_\mathrm{c})^2 \bigr]^{3/2 } } \ ; , \ ] ] with @xmath330 .
here @xmath331 is the maximum value of the projected density , and @xmath332 is a length scale ( core radius ) .
the deflection @xmath333 for such profiles takes a particularly simple form : @xmath334 the intrinsic _ physical _ parameters of the two clusters , i.e. the core radius in mpc and central density in kg@xmath319m@xmath335 ( @xmath336kg@xmath319m@xmath337pc@xmath335 ) have been assumed to be equal . for simplicity
, the first cluster is centered at @xmath338 and the second cluster at @xmath339 .
in addition , we have considered two cases : a case of `` weak '' lenses , for which the second order expansion considered in sect . 3
is expected to provide accurate results , and a case of `` strong '' lenses , with `` joint mass density '' @xmath340 too high in relation to the asymptotic expansion . for the strong case
the mass associated within our field of @xmath341 for the far cluster is @xmath342 .
the latter case will show that , at least qualitatively , the analysis developed in this paper is applicable even to relatively strong lenses .
the cluster parameters adopted for the tests presented here are summarized in table 1 . in figs .
[ fig : light2 ] and [ fig : density2 ] we display the combined luminosity and dimensionless density distributions .
.the parameters used for the simulations . [ cols="<,^,^,^,^,^,^ " , ] the second step has been the calculation of the ray - tracing function @xmath322 .
this function has been calculated on a grid of @xmath343 points , representing a square with a side of @xmath344 .
the jacobian matrix has been calculated on the same grid using the approximate derivatives determined from the differences of @xmath345 in neighboring points .
the plot of the semi - trace @xmath87 of @xmath65 ( see eq . ) presents some interesting features .
we first note that for weak lenses @xmath346 .
however , when the lens is strong , a curious effect is noted ( cfr.fig .
[ fig : tra0-density2 ] ) : the maximum corresponding to the mass distribution of the second cluster is significantly smoothed and lowered .
this has to do with the coupling of the two deflectors .
the effect can be broadly described as a _ magnification _ of the mass distribution of the second deflector because of the lensing effect of the first cluster .
in fact , while the value of @xmath347 exceeds unity ( see fig . [
fig : density2 ] ) , the double lens remains sub - critical with @xmath340 below unity .
proceeding along the list of steps indicated above , we have calculated the reduced shear @xmath302 and the field @xmath348 . at the end of the process , the two quantities @xmath349 and
@xmath306 are available on a @xmath343 grid .
note that all differentiations have been performed using a 3-point , lagrangian interpolation .
the results of our tests for the weak case are summarized in figs .
[ fig : divu - lapsigma1][fig : curlu - laptau_1 ] . within the numerical errors ,
we find that eqs . and
are verified .
of course , since in this case the magnitude of the effect is very small ( the maximum value of @xmath88 is approximately @xmath350 ) , we should not expect that double lensing of this type could be actually detected .
in passing we note that the order of magnitude for the various quantities shown in the figures is consistent with this value of @xmath88 .
the tests for the strong case ( see figs . [
fig : divu - lapsigma2][fig : curlu - laptau_2 ] ) are rather surprising .
in fact , fig . [
fig : divu - lapsigma2 ] shows that eq .
holds approximately also in this case . however , eq . turns out to be inadequate , as shown by fig .
[ fig : curlu - laptau2 ] .
this behavior reflects the fact that the lens configuration here has a rather high value of @xmath340 ( its maximum is about @xmath351 ) , and a small value of @xmath88 ( maximum about @xmath352 ) . a heuristic way to correct eq .
is the following : @xmath353 the use of @xmath87 in the denominator of the previous equation is suggested by analogy with eq . , where @xmath354 is used
. moreover , we note that , to second order , eq . is identical to eq . .
the new equation turns out to be a better approximation than eq . , especially for regimes away from the strong and the weak limiting cases considered above ; in such intermediate cases eq
. performs better , typically by a factor of @xmath11 .
finally , we have considered the problem of the dark cluster ( see sect .
4.4 ) . in this case
the simplified simulations in the semi - analytical approach described above have been performed with the aim of demonstrating how the redshift of a dark cluster can be determined when the redshift and the mass distribution of the luminous cluster are known . in order to find the unknown redshift we have minimized @xmath355 , the integral of the square of the left hand side of eq . or of eq . ,
depending on whether the dark cluster is considered to be closer to us than the luminous cluster or not .
simulations have been performed using the parameters of the strong lens configuration , by taking either the first or the second cluster as the dark cluster .
figure [ fig : findz-0.1 ] shows @xmath355 for the case where the dark cluster is in the front , while fig .
[ fig : findz-0.4 ] shows the same quantity for the case where the dark cluster is behind .
note that @xmath355 has a minimum ( near or equal to @xmath356 ) for the true redshift of the dark cluster .
note also that the second minimum of @xmath355 occurs where the condition @xmath357 is met : in this case the quantities @xmath156 , @xmath157 , and @xmath179 all reduce to @xmath13 .
we should stress that if a dark cluster is detected from a double lens signature , as described in this paper , and if the key characteristics of the bright cluster are known from independent diagnostics , then _ the reconstruction method developed here leads to the full determination of both the distance and the mass distribution of the invisible cluster_. in order to ascertain the relevance of our analysis with respect to present and future observations , we have performed a set of simulations described in the following steps : * we generate @xmath51 source galaxies .
each galaxy is described by its location in the sky , its redshift , and its source ellipticity .
locations are chosen following a uniform distribution on the observed locations @xmath13 ( thus neglecting the magnification effect ) .
the redshift of source galaxies has been drawn from a gamma distribution used also by other authors ( e.g. , brainerd et al . 1996 ) @xmath358 \ ; , \ ] ] with @xmath359 ( other values of @xmath360 have also been tested ) . finally ,
ellipticities @xmath361 are taken from a truncated gaussian distribution ( see eq . ( 32 ) of paper iii ) characterized by @xmath362 .
this value is probably unrealistically large ; it is adopted in order to test our framework under relatively unfavorable conditions .
* as a model for the double lens we use the `` strong '' double lens model introduced in the previous subsection , within the same field of view of @xmath341 .
* source ellipticities @xmath361 are transformed into observed ellipticities @xmath49 using the true asymmetric jacobian matrix associated with the lens .
note that the redshift of each galaxy has been taken into account when performing this transformation . *
the statistical lensing reconstruction procedure is applied to the observed ellipticities .
the reduced shear map @xmath302 is derived by averaging observed ellipticities of nearby galaxies , from the relation @xmath305 properly corrected for the redshift distribution .
* from the shear map we calculate the other relevant quantities : the vector field @xmath325 , its divergence , and its curl . * we then look for significant peaks in the curl map , which would reveal the double nature of the lens .
several simulations have been performed with different source galaxy densities .
the results obtained are very encouraging , since a significant signal in the the curl of @xmath109 is already detected with a galaxy density of @xmath363 .
such relatively high density has already been reached by current observations .
for example , new observations with the very large telescope of the cluster ms1008@xmath3641@xmath3651224 come close to it ( lombardi et al .
2000 , athreya et al . 2000
; see also hoekstra et al .
2000 for observations of ms1054 with the hst ) . figure [ fig : dens_plot ] shows a typical result obtained in our simulations .
the figure shows the density plot of the observed @xmath306 and , superimposed , the contour plot of @xmath366 , i.e. the true quantity that we should measure in the limit of infinite galaxy density . for both plots
we have used the same smoothing spatial weight function , a gaussian of width @xmath367 ( see sect .
we thus conclude that , given the conditions adopted in our simulations , the characteristic signature of double lensing should be within reach of current observations . finally , in order to test the reliability of the detection , we have performed analogous simulations using a single cluster . in this case , as explained above , we expect a ( nearly ) vanishing @xmath368 , especially after the correction for the spread of background sources is applied ( see sect .
5.3.2 ) . figure [ fig : density - single ] shows a map of this quantity ( with the correction applied ) in the case of a single lens at redshift @xmath369 ( the parameters of this lens are the same as those of the first lens considered in the double configuration ) .
note that the noise observed is well below the signal shown in figure [ fig : dens_plot ] .
in this paper we have generalized the statistical analysis at the basis of mass reconstructions of weak gravitational lenses from the standard case of single lenses to the case of double lenses . from one point of view
, this study leads to practical tools to deal with `` errors '' that might occur when the standard single lens analyses are applied to the not infrequent cases where two massive clusters happen to be partially aligned along the line of sight . in reality , our study of double lenses , based on the analytical framework mentioned in the first item below then checked and extended by means of simulations , has opened the way to a number of interesting results .
this is a list of the main points made in our paper : 1 .
a consistent analytical framework has been constructed where the contribution of the small asymmetry in the jacobian matrix induced by double lensing is retained to two orders in the weak lensing asymptotic expansion .
2 . given the known properties of the distribution of ( bright ) clusters of galaxies
, we have shown that a few configurations are likely to be present in the sky , for which the small effects characteristic of double lensing may already be within detection limits .
3 . as a separate astrophysical application
, we have demonstrated that , if the characteristic signature of double lensing appears in an observed configuration where a single bright cluster exists , with properties well constrained by independent diagnostics , the location and the mass distribution of the dark cluster , if present and responsible for the effect , can in principle be reconstructed unambiguously .
we have checked that the redshift distribution of the source galaxies should not confuse the signature of double lensing .
an examination of the relevant contributions to the noise of the shear measurements has brought up a limitation of weak lensing analyses related to the clustering of source galaxies . in particular
, weak lensing studies of single clusters are found to be characterized by a lower limit for the expected noise , regardless of the depth of the images and of the density of source galaxies .
a firm quantitative estimate of this surprising effect will be provided an a separate article .
this work was partially supported by murst of italy .
we thank peter schneider for several stimulating discussions that have helped us improve this paper significantly .
in this appendix we derive eqs . and .
we start from the definition of @xmath109 , which , to second order , is @xmath370 we recall that @xmath371 . by inserting eqs .
( [ eq : b19][eq : b22 ] ) here , rather long expressions are found : @xmath372 \ ; , \\
\label{eq : u-2_2 } \tilde u_2 = & \frac{1}{4 } \bigl [ 4 \nabla^2 \theta^\mathrm{s}_2 - \theta^\mathrm{s}_{2,22 } \nabla \cdot { \mbox{\boldmath $ \theta$\unboldmath}}^\mathrm{s } - 2 \theta^\mathrm{s}_{2,11 } \theta^\mathrm{s}_{1,1 } - \nonumber\\ & \theta^\mathrm{s}_{1,12 } \bigl ( \theta^\mathrm{s}_{2,2 } - \theta^\mathrm{s}_{1,1 } \bigr ) - \bigl(\theta^\mathrm{s}_{1,2 } + \theta^\mathrm{s}_{2,1 } \bigr ) \nabla \wedge { \mbox{\boldmath $ \theta$\unboldmath}}^\mathrm{s}_{,2 } \bigr ] \ ; . \end{aligned}\ ] ] from these expressions we can calculate the divergence and the curl of @xmath109 .
the expressions obtained by a straightforward application of the definitions contain also terms of order @xmath373 or higher , which are to be discarded .
let us start from the curl of @xmath109 . from the previous equations , after some manipulations ,
we obtain @xmath374 \ ; . \end{split}\ ] ] the terms in this expression are organized by rows .
recalling that @xmath375 , we recognize that the terms in the second and third lines are of order @xmath373 , and thus should be dropped .
a simple analysis also shows that the two terms in the last line cancel out . finally , as to the first line
, we note that @xmath376 and @xmath377 are of the form @xmath378 .
thus at the end we obtain @xmath379 that is eq . .
let us now turn to the divergence of @xmath109 . from eqs . and
, a rather long calculation gives @xmath380 \ ; .
\end{split}\ ] ] the first line of this expression needs no special explanations , since we recall that @xmath381 .
the sum of the terms in the second and third lines can be shown to be equal ( dropping terms of order @xmath373 or higher ) to @xmath382 . by replacing @xmath383 by @xmath384 and @xmath385 by @xmath386 ( which is allowed to second order ) , the terms in parentheses in the fourth and fifth lines become respectively @xmath387 and @xmath388 .
finally , the last line can be discarded , because it contains two terms of order @xmath373 . in conclusion , we obtain @xmath389 this expression can be replaced by @xmath390 .
in fact , to second order we have @xmath391 we now apply the vector identity @xmath392 ( valid for any vector field @xmath393 ) to the last term of eq .
: @xmath394 where the last equality holds because @xmath395 is of order @xmath396 .
thus finally we can rewrite eq . as @xmath397 comparing this expression with eq .
we obtain the desired relation . 99 athreya , r. , mellier , y. , van waerbeke , l. , fort , b. , pello , r. , dantel - fort m. 2000 , submitted ( astro - ph/9909518 ) bartelmann , m. , narayan , r. , seitz , s. , & schneider , p. 1996 , apj 464 , l115 borgani , s. , rosati , p. , tozzi , p. , & norman , c. 1999 , apj 517 , 40 brainerd , t. , blandford , r. , smail , i. 1996 , apj 466 , 623 crawford , c. s. , fabian , a. c. , rees , m. j. 1986 , nature 323 , 514 girardi , m. , borgani , s. , giuricin , g. , mardirossian , f. , & mezzetti , m. 1998 , apj 506 , 45 hoekstra , h. , franx , m. , kuijken , k. 2000 , apj , 532 , 88 kaiser , n. 1995 , apj 439 , l1 kaiser , n. 1998 , apj 498 , 26 kaiser , n. , jaffe , a. 1997 , apj 484 , 545 kaiser , n. , & squires , g. 1993 , apj 404 , 441 kaiser , n. , squires , g. , & broadhurst , t. 1995 , apj 449 , 460 lombardi , m. 2000 , phd thesis , scuola normale superiore , pisa lombardi , m. , & bertin , g. 1998a , a&a 330 , 791 ( paper i ) lombardi , m. , & bertin , g. 1998b , a&a 335 , 1 ( paper ii ) lombardi , m. , & bertin , g. 1999 , a&a 342 , 337 ( paper iii ) lombardi , m. , rosati , p. , nonino , m. , girardi , m. , borgani , s. , squires , g. 2000 , submitted luppino , g. a. , & kaiser , n. 1997 , apj 475 , 20 molinari , e. , buzzoni , a. , & chincarini , g. 1996 , a&as 119 , 391 peebles , p.j.e .
1980 , the large scale structure of the universe ( princeton university press : princeton ) peebles , p.j.e .
1993 , principles of physical cosmology ( princeton university press : princeton ) schneider , p. , ehlers j. , & falco , e. e. 1992 , gravitational lenses ( springer - verlag : berlin ) schneider , p. , seitz c. 1995 , a&a 294 , 411 seitz , c. , & schneider , p. 1997 , a&a 318 , 687 seitz , s. , & schneider , p. 1992 , a&a 265 , 1 seitz , s. , & schneider , p. 1994
, a&a 287 , 349 taylor , a. n. , dye , s. , broadhurst , t. j. , benitez , n. , & van kampen , e. 1998 apj , 501 , 539 van waerbeke , l. , bernardeau , f. & mellier , y. 1999 , a&a 342 , 15 wang , q. d. , & ulmer , m. p. 1997 , mnras 292 , 920 | the analysis of the shear induced by a single cluster , acting as a gravitational lens , on the images of a large number of background galaxies is all centered around the curl - free character of a well - known vector field that can be derived from the data ; in particular , the mass reconstruction methods , currently producing many interesting astrophysical applications , are all based on this tensorial property of the induced shear .
such basic property breaks down when the source galaxies happen to be observed through two clusters at different redshifts , partially aligned along the line of sight , because an asymmetric part in the jacobian matrix associated with the ray tracing transformation is expected , now that the incoming light rays are bent twice . in this paper
we address the study of double lenses and obtain five main results .
( i ) first we generalize the procedure to extract the available information , contained in the observed shear field , from the case of a single lens to that of a double lens .
( ii ) then we evaluate the possibility of detecting the signature of double lensing given the known properties of the distribution of clusters of galaxies .
in particular , we show that a few configurations are likely to be present in the sky , for which the small effects characteristic of double lensing may already be within detection limits ( i.e. , we show that the signal involved is expected to be larger than what could be produced by statistical noise , which includes effects associated with the distribution of the source ellipticities and with the spread in redshift of the lensed galaxies ) .
( iii ) as a different astrophysical application , we demonstrate how the method can be used to detect the presence of a dark cluster that might happen to be partially aligned with a bright cluster studied in terms of statistical lensing ; if the properties of the bright cluster are well constrained by independent diagnostics , the location and the structure of the dark cluster can be reconstructed .
( iv ) in addition , we show that the redshift distribution of the source galaxies , which in principle might also contribute to break the curl - free character of the shear field , actually produces systematic effects typically two orders of magnitude smaller than the double lensing effects we are focusing on .
( v ) remarkably , a discussion of relevant contributions to the noise of the shear measurement has brought up an intrinsic limitation of weak lensing analyses , since one specific contribution , associated with the presence of a non - vanishing two - galaxy correlation function , turns out not to decrease with the density of source galaxies ( and thus with the depth of the observations ) . in our mathematical framework ,
the contribution of the small asymmetry in the jacobian matrix induced by double lensing is retained consistently up to two orders in the weak lensing asymptotic expansion . the analysis is then checked and exemplified by means of simulations . |
the coupling of electronic and mechanical degrees of freedom is at the core of the physics of nanoelectromechanical systems ( nems)@xcite .
pronounced vibrational effects in electronic transport have been observed in several recent experiments on molecules @xcite and single wall carbon nanotube quantum dots @xcite .
stimulated by the experimental works , several groups @xcite have attempted to theoretically explain some features of the measured stability diagrams ( i.e. , of the differential conductance in a bias voltage - gate voltage colour map ) which appear to be ubiquitous . specifically , one observes the following : ( i ) equidistant lines run parallel to the edges of the coulomb diamonds .
( ii ) the probed diamonds show often negative differential conductance ( ndc ) features appearing between excited vibronic states .
( iii ) this corresponds to a sequence of peaks in the current as a function of the bias voltage .
the height of the peaks measured in ref . was found to be in quantitative agreement with predictions of a simple franck - condon model for a single electronic level coupled to a harmonic mode ( the so called anderson - holstein model ) @xcite .
the ndc features have been explained in ref . in terms of local vibronic excitations . in ref .
ndc features are associated with the anderson - holstein model at a single vibronic resonance , while in ref .
ndcs appear due to the electron - phonon - coupling - induced selective unidirectional cascades of single - electron transitions . in a recent work
@xcite , carbon nanotube - specific ndc features have been attributed to a spatially dependent franck - condon factor and to the presence of quasidegenerate electronic levels , while in ref .
an interference effect between the orbitally degenerate electronic states has been proposed . in refs . , , and , however , to see ndc some sort of asymmetry is required either in the coupling to the source and drain contacts or in the coupling to the two orbitally quasidegenerate states of the system or in both couplings at the same time .
+ in this paper we extend these ideas and propose a generic model in which two degenerate or quasidegenerate molecular levels are coupled to many vibronic modes .
we consider the case of a _ symmetric _ setup , i.e. , invariant under exchange of source with drain if simultaneously also the sign of the bias voltage is reversed .
if the frequency @xmath1 of the @xmath0th mode is an @xmath0th multiple , @xmath2 , of the frequency @xmath3 of the fundamental mode , then several energetic degenerate vibronic configurations arise ( involving two or more modes ) which may contribute to transport at finite bias .
+ as we are going to show in our paper , the additional presence of spin and/or orbital degeneracies opens the possibility of getting slow channels contributing to transport .
as a consequence , ndc phenomena can occur despite a _ fully symmetric _
quantum - dot setup .
a peculiarity of the observed features is , in particular , an asymmetry with respect to the gate voltage in the stability diagrams .
finally , we retain source - drain symmetry but allow for an asymmetry in the coupling to the two orbitally degenerate states .
we show that this asymmetry is sufficient to explain the experimental observations ( i)-(iii ) .
+ in some of the experiments @xcite the slope of the ndc lines is the same for both positive and negative biases .
this characteristic has been associated with the left and right asymmetry in the coupling to the leads @xcite .
we confirm that including the higher harmonics does not change the effect and give an analytical interpretation of the numerical results for low biases .
+ the paper is organized as follows : in sec .
ii the model hamiltonian of a single molecule coupled to several vibronic modes is introduced .
a polaron transformation is employed to obtain the spectrum of the system in the presence of electron - vibron interactions . in turn , as known from the theory of franck - condon blockade in the simplest anderson - holstein model @xcite , the polaron transformation also crucially affects the tunneling hamiltonian describing the coupling to the source and drain leads .
iii the consequences on transport are analyzed .
( i ) the tunneling transition amplitudes involve product of franck - condon factors with coupling constants depending on the mode number .
( ii ) at low bias , such that only the lowest vibronic mode is excited , a description of the dynamics only in terms of rate equations involving occupation probabilities of the many - body states of the quantum dots is appropriate .
( iii ) at higher bias , when several vibron modes are excited , a generalized master equation ( gme ) coupling diagonal ( populations ) and off - diagonal ( coherences ) elements of the quantum dot reduced density matrix should be used ( see e.g. , refs . ) .
iv our main results for the current - voltage characteristics in a fully symmetric setup are shown and analyzed .
in particular , we give an explanation of the ndc features observed at low bias in terms of a different spin and orbital degeneracy of states with different electron numbers contributing to transport . in sec .
v asymmetric setups are discussed as well .
conclusions are drawn in sec .
in the following we consider a generalized anderson - holstein model where the hamiltonian of the central system has the form @xmath4 where @xmath5 describes two quasidegenerate levels . the hamiltonian is modeled as @xmath6 where @xmath7 is the orbital and @xmath8 is the spin degree of freedom .
the operator @xmath9 counts the number of electrons with spin @xmath10 in the orbital @xmath11 .
@xmath12 is the total number operator .
the orbital energy is @xmath13 $ ] with @xmath14 an orbital - mismatch .
the coulomb blockade is taken into account via the charging energy @xmath15 and we assume @xmath16 .
+ the vibron hamiltonian is expressed as @xmath17 where @xmath18 annihilates ( creates ) a vibron in the @xmath19 mode of energy @xmath20 .
we assume that the energy of the @xmath19 mode is given by @xmath21 being an @xmath19 multiple of the energy @xmath22 of the fundamental mode as it is , for example , for longitudinal stretching modes in quantum wires and carbon nanotubes .
finally , the electron - vibron interaction hamiltonian is given by @xmath23 where @xmath24 is the coupling constant for the @xmath19 vibronic mode . in order to solve the hamiltonian of the system
, we decouple the electron - vibron interaction part by applying a unitary polaron transformation @xcite .
specifically , we set @xmath25 , where @xmath26 and @xmath27 is the dimensionless coupling constant associated with mode @xmath0 .
notice that @xmath28 is the coupling constant of the fundamental mode .
we assume @xmath29 in the analysis of the spectrum , which is in the range of values observed e.g. , in experiments on carbon nanotubes @xcite . under the polaron transformation
the operator @xmath30 is transformed as @xmath31 where @xmath32 $ ] . in a similar way
the shifted vibronic operator is @xmath33 the transformed form of the system hamiltonian is thus @xmath34 where @xmath35 is the renormalized orbital energy and @xmath36 is the coulomb repulsion modified by the vibron mediated interaction . + the eigenstates of the system are @xmath37 where @xmath38 and @xmath39 the number of electrons in the branch @xmath40 .
notice that @xmath41 defines the total number of electrons on the dot . for later purposes we indicate the ground state and first excited state with @xmath42 electrons as [ see fig .
[ fig2](b ) ] @xmath43 the first excited state with @xmath42-electron contains one vibronic excitation in the first mode , i.e. , @xmath44 . in a similar way we define the ground states and first excited states with @xmath45 electron . for zero orbital mismatch one has fourfold degeneracy [ see again fig .
[ fig2](b ) ] , i.e. , @xmath46 where @xmath47,@xmath48 .
we notice that both for @xmath49 and @xmath45 the second excited states are vibronically degenerate for the dispersion relation eq .
, then the configurations @xmath50 and @xmath51 have the same energy .
for finite orbital mismatch , @xmath52 , the orbital degeneracy is broken . the case @xmath53 is illustrated in fig . [ fig8](b ) .
the corresponding states are : @xmath54 and @xmath55 , + @xmath56 .
in this section we discuss the transport across the system under the assumption of weak coupling to the leads .
the hamiltonian of the full system is described as @xmath57 where @xmath58 denote the source and the drain contact , respectively .
the tunneling hamiltonian @xmath59 is given by @xmath60 where @xmath61 is the electron operator in the leads .
finally , @xmath62 describes the influence of the externally applied gate voltage @xmath63 .
the gate is capacitively coupled to the molecule and hence contributes via a term @xmath64 . in the case of high degeneracy of the spectrum ,
the appropriate technique to treat the dynamics of the system in the weak coupling regime is the liouville equation method for the time evolution of the density matrix of the total system consisting of the leads and the generic quantum dot . to describe the electronic transport through the molecule
, we solve the liouville equation @xmath65,\end{aligned}\ ] ] for the reduced density matrix @xmath66 in the interaction picture , where the trace over the leads degrees of freedom is taken .
we make the following standard approximations to solve the above equation : ( i ) the leads are considered as reservoirs of noninteracting electrons in thermal equilibrium .
( ii ) we factorize the total density matrix as @xmath67 , where we assume weak coupling to the leads , and treat @xmath59 perturbatively up to second order .
( iii ) being interested in long time properties , we make a markov approximation , where the time evolution of @xmath68 is only local in time .
notice that in the regime of interest , @xmath69 , the markovian approximation becomes exact .
( iv ) since the eigenstates @xmath70 of @xmath71 are known , it is convenient to calculate the time evolution of @xmath72 in this basis retaining coherences between degenerate states with the same number of particles .
hence @xmath72 can be divided into block matrices @xmath73 , where @xmath74 and @xmath75 are the energy and number of particles of the degenerate eigenstates @xmath76 .
we obtain an equation of the bloch - redfield form + @xmath77 where the indices @xmath78 refer to the eigenstates of @xmath79 and @xmath80 runs over all degenerate states with fixed particle number .
notice that if @xmath49 then @xmath81 , while if @xmath82 then @xmath83 . the redfield tensors are given by @xcite @xmath84 and @xmath85 , where the quantities @xmath86 are transition rates from a state with @xmath75 to a state with @xmath87 particles .
more explicitly : @xmath88 with @xmath89 and @xmath90 the electrochemical potential of the lead @xmath91 . likewise @xmath92 with @xmath93
moreover , we introduced @xmath94 where @xmath95 is the fermi function while @xmath96 and @xmath97 are the bare transfer rates with the constant densities of states of the leads @xmath98 . knowing the stationary density matrix @xmath99 , the ( particle ) current through lead @xmath91 is determined by @xmath100 @xmath101 if the relation given by eq .
holds , then spin and orbital degeneracies intrinsic in the electronic structure are supplemented by degeneracies related to the vibronic structure .
several vibronic modes with frequencies @xmath102 multiples of the fundamental frequency @xmath103 give rise , in fact , naturally to several degenerate vibronic configurations .
this is the situation we shall focus on in the rest of the paper .
+ a degenerate spectrum is a necessary condition for the appearance of interference effects in the transport characteristics both in the linear and non linear regime @xcite and these effects can be captured only by considering not only populations ( diagonal elements ) but also coherences ( off - diagonal elements ) of the reduced density matrix .
+ for the system at hand we calculated the current both with and without coherences between degenerate states up to five vibronic modes , obtaining though only quantitative but not qualitative differences .
while spin and orbital degeneracies can be a priori excluded from the transport through a single molecule with nonpolarized leads@xcite , the role played by the vibronic coherences requires a more careful analysis .
+ we have confirmed that it is not possible to construct a linear combination of degenerate states @xmath104 with finite transition amplitude to a state @xmath105 at one lead but decoupled from @xmath105 at the other lead , where @xmath105 and @xmath106 represent the states given by eq . .
this observation , complemented by the general method presented in ref . proves the absence of interference blocking states in our system .
thus , interference , even if present , does not have dramatic consequences on the transport characteristics of the system .
+ all the current maps presented in the next section are hence , apart from fig . [ fig6](b ) , calculated neglecting coherences . as shown explicitly in fig .
[ fig6](b ) , this approximation does not affect qualitatively the results ( at least in the low bias regime ) .
moreover , the negative differential conductance and the associated dynamical symmetry breaking that we present in the next section are not related to the interference and can thus be obtained by considering the dynamics of the populations alone .
in this section , we illustrate our predictions for the transport characteristics and focus on the @xmath107 transitions . in the calculation we also assume for the coupling constant of the @xmath19 mode @xmath108 ( as expected for stretching modes in carbon nanotubes @xcite ) .
the system is _ symmetrically _ coupled to source and drain contacts ( @xmath109 ) and the lowest five vibron modes are included .
results for the differential conductance for different values of the dimensionless coupling constant @xmath110 are illustrated in fig .
[ fig1 ] and fig .
[ fig7 ] , obtained for zero and finite orbital mismatch @xmath111 , respectively
. when the orbital mismatch is zero , i.e. , @xmath113 , then the two orbital energies @xmath114 are the same . the minimum energy to produce
a charge excitation is @xmath115 .
we take the value of this energy as @xmath116 ( comparable to the level spacing energy of a suspended single wall carbon nanotube of @xmath117 length ) .
furthermore , we assume the energy of the lowest vibronic mode to be @xmath118 .
thus the charge excitation energy is much larger than the energy of the lowest vibron mode . indeed all the equidistant lines running parallel to the diamond edges observed in fig .
[ fig1 ] are due to vibron excited states . what striking is the occurrence of negative differential conductance ( ndc ) features at moderate coupling ( @xmath119 and @xmath120 ) which , however , disappear when the coupling is increased ( @xmath121 ) .
moreover , the ndc lines are only running parallel to one of the diamond edges , which indicates an asymmetry with respect to the gate voltage @xmath63 . as we are going to explain , at low bias , these features are a consequence of franck - condon assisted tunneling combined with the spin and/or orbital degeneracy in the system .
specifically , let us focus on the low bias region [ see fig .
[ fig2](a ) ] , where only ground - state @xmath122 ground - state transitions ( region a ) , and ground - state @xmath122 first excited - state transitions ( regions b , c ) are relevant .
the @xmath42 and @xmath123-particle states involved are illustrated in fig .
[ fig2](b ) , together with their degeneracy due to spin and orbital degrees of freedom , and have energies below the dashed line in fig .
[ fig2](b ) .
the states above the dashed line require an energy of at least @xmath124 and have thus also a vibron degeneracy . in the considered energy range
no degenerate vibron configurations are involved .
moreover , coherences between degenerate electronic configurations are not present such that a rate equation description only in terms of populations is appropriate . at low bias and in the stationary limit
eq . yields the equation for the populations : @xmath125 or , equivalently , @xmath126 where @xmath127 and @xmath128 , see eqs .
( [ transitionrate1])-([transitionrate3 ] ) .
notice in particular that @xmath129 and @xmath130 , i.e. , they only differ in the fermi factors .
the transition coefficients are the same and given by @xmath131 .
moreover , we find for the current through lead @xmath91 @xmath132 let us now focus on region a. in this case only the @xmath42-particle ground state @xmath133 [ see eq . ] with energy @xmath134 and the four @xmath123-particle ground states @xmath135
[ see eq . ] with energy @xmath136 contribute to transport .
moreover , inside region a it holds @xmath137 such that , if @xmath138 , @xmath139 , it also follows @xmath140 @xmath141 this situation is illustrated in the table of fig . [ fig3 ] , where a dashed red ( black ) arrow indicates a transition involving the source ( drain ) . condition ( [ groundstatestransitions ] ) with ( [ populations2 ] ) then implies that @xmath142 and hence @xmath143 ; @xmath144 , yielding with eq . for the current in region a , @xmath145 . along similar lines we can calculate the current in regions b and c.
let us start with region b where ( see table of fig .
[ fig3 ] ) the gate voltage @xmath63 is such that the @xmath123-particle ground states @xmath146 have energy @xmath136 smaller than the one , @xmath134 , of the @xmath42-particle ground state @xmath133 . moreover , in this region also the first excited state @xmath147 with energy @xmath148 enters the transport window .
we also assume that the rate @xmath149 between the states @xmath150 and @xmath147 is negligible with respect to @xmath151 and @xmath152 .
corrections due to a finite @xmath149 will be discussed later . inside region
b it holds , besides eq .
, and hence eq . ,
@xmath153 and @xmath154 .
hence it follows that @xmath155 @xmath156 eq
. implies thus that in region b it holds @xmath157 and hence @xmath158 ; @xmath159 ; @xmath160 .
the total current in region b follows from eq .
and reads @xmath161 .
the condition to observe ndc is that @xmath162 , which implies @xmath163 along similar lines ( see table in fig . [ fig3 ] ) , one finds for the transition from region a to c that @xmath164 if @xmath165 let us look in more detail at eq .
and eq . .
the rates @xmath151 and @xmath152 describe transitions between states which only differ in their vibronic part . from eqs .
( [ groundstatestransitions ] ) , ( [ groundtoexcitedstatestransitions ] ) and ( [ transitionmatrixappendix1 ] ) it follows that @xmath166 hence , to observe ndc for the transition from region a to b one needs that @xmath167 . on the other hand for ndc in the transition from a to c we must require @xmath168 . indeed , as shown in fig .
[ fig1 ] , ndc for the transition a @xmath122 b is observed for @xmath119 and @xmath120 , but it vanishes for @xmath121 . on the other hand ndc
is never observed for the transition region a @xmath169 c. let us now turn to region b and to a finite @xmath170 with @xmath171 , @xmath172 . because now @xmath150 can get populated , also transitions from @xmath173 to @xmath135 are activated ( see fig .
[ fig4 ] ) .
+ because of it holds @xmath174 hence , the stationary solution with equal probabilities is spoiled , at finite @xmath149 , due to the inequality of @xmath175 and @xmath176 .
in fact , in the case @xmath177 , the same inequality only implies that @xmath178 .
we also notice that @xmath179 .
moreover , the rates @xmath149 and @xmath151 only differ in their vibronic configuration : it holds [ cf .
@xmath180 ^ 2=\left ( 1-\lambda^2\right)^2.\end{aligned}\ ] ] likewise @xmath181 hence , if @xmath182 it is indeed @xmath183 and an expansion to lowest order in the ratios @xmath184 can be performed . in this case the conditions for ndc acquire a more complicated form . the condition to get ndc in the source threshold lines [ from a to b in fig .
[ fig2](a ) ] is @xmath185 while the condition for ndc in the drain threshold lines [ from a to c in fig .
[ fig2](a ) ] is @xmath186 it means that the presence of chain transition processes redistributes the population among the many - body states in a way that privileges the low energy states ( see fig . [ fig5 ] ) ; this in turn weakens ndc since it privileges the conducting channels that carry more current . in this section
, we discuss our results on vibration - assisted transport with the same parameters as in sec .
[ zerobandmismatch ] but with a finite orbital mismatch , i.e , @xmath187 .
in this case the orbital degeneracy is broken .
the corresponding stability diagrams are shown in fig .
the analysis for the ndc conditions at low bias remains almost the same as before .
slight differences occur because in this case the orbital degeneracy is lost and the populations are redistributed over the many - body states in a different way . in fig .
[ fig8](a ) , the different transition regions of the stability diagram have been labeled while the energy level scheme has been shown in fig .
[ fig8](b ) , where the degeneracy of each state is given in brackets .
we again truncate the process at the dashed line to analyze the lowest - energy excitations .
as before @xmath188 and @xmath189 are the orbital degrees of freedom .
the transition scheme for regions d and f is shown explicitly in fig .
[ fig9 ] .
the current - voltage characteristics are qualitatively the same as far as the mismatch is in the moderate regime @xmath190 , the only difference being the position of the resonance lines , that depends on the specific position of the energy levels . in other terms , despite the size and the position of the regions of the stability diagram ( fig .
[ fig7 ] ) depend on the mismatch @xmath191 , the value of the current in each region is independent of it .
thus a unified treatment of the two cases presented in fig .
[ fig7 ] is allowed , despite their apparent qualitative differences .
in particular , we can observe that the current in region b is larger than the current in region a since a new transport channel is opening with the same geometrical coupling ( @xmath151 ) when passing from a to b. this implies that the first source threshold transition line is always a positive differential conductance ( pdc ) line .
the current in c is equal to the one in a since , due to energy conservation , no new transport channel is opening when passing from a to c. the corresponding resonance line is thus invisible in the stability diagram ( see fig .
[ fig7 ] ) .
the transition that defines the threshold line separating a from c ( @xmath192 at the drain ) involves , in fact , states which are not populated in that bias and gate voltage range .
finally , the comparison between currents in the adjacent b and d regions and between the currents in the c and f regions results in conditions for the appearance of ndc lines which are very similar to the one in absence of mismatch . in particular , the condition for ndc at the transition between regions b and d is identical to the one for the transition between regions a and the b corresponding to zero mismatch given in eq . .
the ndc condition for the transition between region c and f reads instead @xmath193 to be compared with the one for the transition between the regions a and c and zero mismatch given in eq . .
a similar analysis can be repeated for higher - energy transitions which participate in the transport for higher biases .
it is already clear though from the low energy transitions that a moderate breaking of the orbital degeneracy introduced by the finite mismatch @xmath191 does not change qualitatively the transport characteristics of the system . in particular
, it preserves the presence or absence of asymmetric ndc lines as a function of the electron vibron coupling @xmath110 ( compare figs .
[ fig1 ] and [ fig7 ] ) . the franck condon factors are still assumed to be the same for the source and drain tunneling and no overall asymmetry is introduced in the tunneling coupling of the molecule dot to the source and drain ( @xmath194 ) .
the theory can produce , though , alternating pdc and ndc traces as discussed in ref .
if we assume that coupling of the @xmath188 and @xmath189 orbitals to be different . in fig .
[ fig10 ] we have plotted the differential conductance @xmath195 for an asymmetry parameter @xmath196 where `` 1 '' , `` 2 '' represent the orbital degrees of freedom , respectively and @xmath91 means source or drain . for convenience , in the numerical calculations , we use parameters as in figs .
[ fig1 ] and [ fig7 ] . as seen by comparing fig .
[ fig7 ] with fig .
[ fig10 ] , at @xmath197 ndc can now occur . moreover , an alternation of pdc with ndc lines , as seen in the experiments @xcite occurs . repeating the same analysis as in sec .
[ sec : lowbias_finmis ] , we indeed find that : ( i ) the transition from region a to b gives a ndc line for independent of the value of @xmath110 .
( ii ) the condition governing the transition from region b to d is now modified to be ( at @xmath177 ) @xmath198 which , for @xmath199 , increases the range of @xmath110 giving ndc also for @xmath197 [ fig .
[ fig10](c ) ] . @xmath200 and @xmath201 are defined analogously @xmath202 and @xmath203 by considering the @xmath11 dependence of the bare tunneling rates @xmath204 .
( iii ) the transition from region d to g is governed by the condition : @xmath205 which explains the persistence of a pdc line also for smaller values @xmath110 ( compare again fig .
[ fig7 ] with fig .
[ fig10 ] ) .
the corrections introduced by a finite @xmath149 rate do not change qualitatively the analysis and can be found for completeness in the appendix [ app : finite_g11 ] .
in this paper we analyzed the spectrum and the vibration - assisted transport properties of a nanostructure where two degenerate or quasidegenerate levels are coupled to several vibronic modes .
our model can capture features of transport properties of suspended carbon nanotubes and of molecules with a fourfold degenerate electronic level coupled to many vibrational modes .
the transport theory is based on vibron - assisted tunneling , mediated by vibrational modes . in order to study the dynamics of the system
, we apply a density matrix approach which starts from the liouville equation for the total density operator and which enables the treatment of degenerate and quasidegenerate states . despite the fact that we considered a fully symmetric setup ,
the stability diagram for the differential conductance shows striking negative differential conductance ( ndc ) features which hints at peculiar features of our nanoelectromechanical system .
we predict that ndcs appear due to the slow channels in the source transition originating from spin and/or orbital degeneracies and the suppression of franck - condon channels .
with source - drain symmetry being preserved but an asymmetry between orbitally degenerate states being allowed , we could explain the alternating pdc and ndc features observed in refs . and .
we acknowledge the support of dfg under the programs sfb 689 and grk 1570 .
a. y. acknowledges the support of kohat university of science & technology , kohat-26000 , khyber pakhtunkhwa , pakistan under the human resource development program .
we thank georg begemann and leonhard mayrhofer for their help in the numerical calculations .
to determine the transition rates , we calculate the matrix elements @xmath206 where @xmath207 and @xmath208 represent the eigenstates given by eq . .
the function @xmath209 determines the coupling between states with a different vibronic number of excitations with effective coupling @xmath110 and is expressed as @xmath210 where @xmath211 .
in this appendix we give the conditions which determines the sign of the current change in the transition from region b to d and d to g with finite mismatch @xmath191 ( fig .
[ fig8 ] ) .
we take only the first order contribution in the ratios @xmath212 and @xmath213 . the validity of these formulas is thus restricted to @xmath214 .
the condition for the transition b to d reads @xmath215 while for the transition d to g one obtains @xmath216 where @xmath217 measures in both cases the asymmetry between the coupling to the different orbitals .
30 k. l. ekinci and m. l. roukes , nanoelectromechanical systems , rev .
* 76 * , 061101 ( 2005 ) .
j. park , a. n. pasupathy , j. i. goldsmith , c. chang , y. yaish , j. r. petta , m. rinkoski , j. p. sethna , h. d. abruna , p. l. mceuen and d. c. ralph , nature ( london ) * 417 * , 722 ( 2002 ) .
l. h. yu and d. natelson , nano lett . * 4 * , 79 ( 2004 ) .
x. h. qiu , g. v. nazin , and w. ho , phys .
lett . * 92 * , 206102 ( 2004 ) . h. park , j. park , a. k. l. lim , e. h. anderson , a. p alivisatos , and p. l. mceuen , nature ( london ) * 407 * , 57 ( 2000 ) .
r. h. m. smit , y. noat , c. untiedt , n. d. lang , m. c. van hemert , and j. m. van ruitenbeek , nature ( london ) * 419 * , 906 ( 2000 ) . s. sapmaz , p. jarillo - herrero , ya . m. blanter , c. dekker , and h. s. j. van der zant , phys .
lett . * 96 * , 026801 ( 2006 ) .
a. k. httel , b. witkamp , m. leijnse , m. r. wegewijs , and h. s. j. van der zant , phys .
lett . * 102 * , 225501 ( 2009 ) .
r. leturcq , c. stampfer , k. inderbitzin , l. durrer , c. hierold , e. mariani , m.g .
schultz , f. von oppen , k. ensslin , nature physics * 5 * , 327 ( 2009 ) .
k. flensberg , new j. phys .
* 8 * , 5 ( 2006 ) .
w. izumida and m. grifoni , new j. phys .
* 7 * , 244 ( 2005 ) .
f. cavaliere , e. mariani , r. leturcq , c. stampfer and m. sassetti , phys .
b * 81 * , 201303(r ) ( 2010 ) .
s. braig and k. flensberg , phys .
b * 68 * , 205324 ( 2003 ) .
d. boese and h. schoeller , europhys . lett . *
54 * , 668 ( 2001 ) .
j. koch , f. von oppen , and a. v. andreev , phys .
b * 74 * , 205438 ( 2006 ) . j. koch and f. von oppen , phys
lett . * 94 * , 206804 ( 2005 ) .
a. mitra , i. aleiner , and a. j. millis , phys .
b * 69 * , 245302 ( 2004 ) ; phys . rev
. lett . * 94 * , 076404 ( 2005 ) .
m. g. schultz , phys .
b * 82 * , 195322 ( 2010 ) .
x. y. shen , bing dong , x. l. lei , and n. j. m. horing , phys .
b * 76 * , 115308 ( 2007 ) . m. g. schultz , phys .
b * 82 * , 155408 ( 2010 ) .
g. begemann , d. darau , a. donarini , and m. grifoni , phys .
b * 77 * , 201406(r ) ( 2008 ) ; * 78 * , 089901(e ) ( 2008 ) .
a. donarini , g. begemann , and m. grifoni , nano letters * 9 * , 2897 ( 2009 ) .
s. braig and p. w. brouwer , phys .
b * 71 * , 195324 ( 2005 ) .
m. braun , j. knig , and j. martinek , phys .
b * 70 * , 195345 ( 2004 ) .
b. wunsch , m. braun , j. knig , and d. pfannkuche , phys .
b * 72 * , 205319 ( 2005 ) .
u. harbola , m. esposito , and s. mukamel , phys .
b * 74 * , 235309 ( 2006 ) .
l. mayrhofer and m. grifoni , eur .
j. b * 56 * , 107 ( 2007 ) .
s. koller , l. mayrhofer and m. grifoni , new j. phys .
* 9 * , 348 ( 2007 ) .
r. hornberger , s. koller , g. begemann , a. donarini and m. grifoni , phys . rev .
b * 77 * , 245313 ( 2008 ) . m. g. schultz and f. von oppen , phys
b * 80 * , 033302 ( 2009 ) .
g. d. mahan , many - particle physics , 3rd edition ( kluwer , new york , 2000 ) .
k. blum , density matrix theory and applications ( plenum press , new york 1996 ) .
a. donarini , g. begemann , and m. grifoni , phys .
b * 82 * , 125451 ( 2010 ) . | a theoretical model of a single molecule coupled to many vibronic modes is presented . at low energies ,
transport is dominated by electron - vibron processes where transfer of an electron through the dot is accompanied by the excitation or emission of quanta ( vibrons ) .
because the frequency of the @xmath0th mode is taken as an @xmath0th multiple of the frequency of the fundamental mode , several energetically degenerate or quasidegenerate vibronic configurations can contribute to transport .
we investigate the consequences of strong electron - vibron coupling in a fully _ symmetric _ setup .
several striking features are predicted . in particular , a gate - asymmetry and pronounced negative differential conductance features are observed .
we attribute these features to the presence of slow channels originating from the interplay of franck - condon suppression of transport channels with spin and/or orbital degeneracies . |
the field of cold atoms has expanded dramatically over the last 15 years .
it has now reached a stage where experimentalists are capable of designing the form of the confining potential .
going to extreme aspect ratios , conditions of quasi - one- and quasi - two - dimensional behavior have been achieved . in other experiments ,
it has also become possible to design toroidal trapping potentials @xcite , in which persistent currents have been observed @xcite .
recent theoretical studies have examined bose - einstein condensates in one - dimensional annular traps .
for example , quantum - tunneling - related effects in vertically @xcite and concentrically @xcite coupled double - ring traps were investigated .
also , the rotational properties of a mixture of two distinguishable bose gases that are confined in a single ring were addressed @xcite .
one of the basic points of the above studies is the fact that the ability to design traps , control and manipulate the atoms with a very high accuracy , may allow the investigation of novel quantum phenomena , like quantum phase transitions , for example . in the present work we consider a mixture of two distinguishable bose gases @xcite which interact via an effectively - repulsive contact potential , and are confined in a two - dimensional concentric double - ring - like trap , as shown in fig.[fig1 ] . using the mean - field approximation ,
we investigate two main questions : first , we identify the various phases in the ground state of the system , varying the interaction strength between the atoms . in the trapping potential that we consider
, we observe that the two gases separate radially via discontinuous transitions ; in this case , each gas resides in one of the two minima of the trapping potential , preserving the circular symmetry of the trapping potential .
we also observe the expected azimuthal ( and continuous ) phase separation between the two gases in each potential minimum @xcite .
a similar effect has also been studied in the case of a single ring @xcite ; see also @xcite .
the second main question that we examine are the rotational properties of this system , including its response to some rotational frequency of the trap @xmath0 , as well as the stability of the persistent currents for variable couplings , and variable relative populations of the two components .
the expectation value of the angular momentum of the system as a function of @xmath0 shows an interesting structure , reflecting the various phase transitions that take place with increasing @xmath0 . regarding the ( meta)stability of the currents , it is remarkable that for equal populations between the two components the vast majority of the coupling strengths that we have examined yield metastable states , except for a very small range where all the coupling strengths are exactly or nearly equal .
it is worth mentioning that analogous single and concentric ring geometries have been addressed in semiconductor heterostructures , both theoretically and experimentally , see e.g. @xcite , and also @xcite for reviews on the subject . in these systems ,
the applied external magnetic field plays the same role as the trap rotation in the present problem and allows the investigation of , e.g. , electron localization effects and persistent electron currents in field - free regions . in what follows
we first describe in sec.ii our model . in sec.iii
we present the results for the ground state of the system , identifying the states where the species coexist , or separate , either radially or azimuthally .
in sec.iv we examine the rotational properties for a fixed rotational frequency of the trap , and the ( meta)stability of the persistent currents .
we study the stability as a function of the coupling between the atoms , as well as of the ratio of the populations of the two components .
finally , in sec.v we present a summary and our conclusions .
we consider two distinguishable kinds of bosonic atoms , labelled as @xmath1 and @xmath2 , which are trapped in a two - dimensional potential of the form @xmath3 where @xmath4 is the usual radial coordinate in cylindrical coordinates and @xmath5 is the atom mass , assumed to be equal for the two components .
the two ( overlapping ) parabolae in @xmath6 with frequencies @xmath7 and @xmath8 are centered at the positions with @xmath9 and @xmath10 , giving rise to the potential plotted in fig.[fig1 ] @xcite . of eq.([trpot ] ) , where @xmath11 , @xmath12 , and @xmath13.,width=264 ] in our calculations we consider @xmath11 and @xmath14 , where @xmath15^{1/2}$ ] is the oscillator length corresponding to @xmath16 , and finally @xmath13 . in the outer ring
the potential is more tight , @xmath17 , in order to compensate for the fact that @xmath18 , i.e. , to make the product of the width " of each annulus times the radius of each annulus to be comparable to each other .
( this is typically also the case in the studies on electrons in quantum rings , in semiconductor heterostructures that we mentioned above . ) to simplify the discussion , we also assume that there is a very tight trapping potential along the @xmath19 axis ( omitted in the potential above ) , which completely freezes out the degrees of freedom of the gases along this direction . with this assumption ,
our problem becomes effectively two - dimensional , with the tight dimension entering only implicitly through the parameters @xmath20 in the hamiltonian of eq.([ham ] ) . with the usual assumption of a contact interatomic potential ,
the hamiltonian becomes @xmath21 here @xmath22 , with @xmath23 being the state of lowest energy of the potential along the @xmath19 axis and @xmath24 being the @xmath25wave scattering lengths for zero - energy elastic atom - atom collisions .
the coupled gross - pitaevskii - like equations for the order parameters of the two components @xmath26 and @xmath27 , resulting from the above hamiltonian , are @xmath28 in the above equations @xmath29 , and @xmath30 .
also , @xmath31 , @xmath32 , @xmath33 , and @xmath34 are the chemical potentials . in what follows
, we consider repulsive interactions only , @xmath35 , @xmath36 , and also assume that @xmath37 .
the method that we adopt to solve eqs.([gpe ] ) is a fourth - order split - step fourier method within an imaginary - time propagation approach @xcite .
we start with a reasonable initial state for the two components and propagate it in imaginary time , making sure that we proceed a sufficiently large number of time steps , which guarantee that we have reached a steady state .
we start with the ground state of the system .
there are three energy scales in the problem , namely the single - particle energy that is set by the trap , the intra - atomic interaction energy , and the inter - atomic interaction energy . for weak interactions ,
the energy is dominated by the single - particle term and the ground state is determined by the minimization of this term . on the other hand , for strong interactions , for a given ratio of the two populations @xmath38 , the actual symmetry of the ground state results from the competition between the intra - and inter - species coupling strengths ; the former favors the maximum possible spread of the two gases within the system , whereas the latter favors the minimization of their spatial overlap . the most pronounced difference of this problem as compared to the case where there is only one potential minimum
i.e. , a single annulus is the existence of a phase where each component resides in only one of the two potential minima , thus separating radially . in addition , we have also observed the expected azimuthal phase separation when both species occupy the same potential minimum @xcite . in fig.[fig2 ]
we illustrate the phase diagram showing the symmetry of the ground - state density distribution of the two gases as @xmath39 and @xmath40 are varied , for @xmath41 . since @xmath42 and @xmath41 , the results are symmetric when the two components are interchanged . the axes in the phase diagram of fig.[fig2 ]
may also be considered to represent the values of the scattering lengths @xmath43 and @xmath44 ( scaled appropriately ) .
we have found three different phases : ( i ) coexistence of the two species ( squares , red color ) , ( ii ) azimuthal separation ( triangles , blue color ) , and ( iii ) radial separation ( circles , green color ) , see fig.[fig3 ] .
the difference between solid and empty symbols in fig.[fig2 ] refers to the stability of the persistent currents and is explained in the following section .
-versus-@xmath40 plane , which shows the symmetry of the ground - state density distribution of the two components .
squares , triangles and circles correspond to phase coexistence , azimuthal phase separation , and radial phase separation , respectively .
the solid ( empty ) symbols denote points where the currents are metastable ( unstable).,width=302 ] as seen in fig.[fig2 ] , the phase boundary between phase coexistence and separation of the two components is , to a very good approximation , a straight line given by ( setting @xmath45 ) @xmath46 as we have found by fitting numerically our data . certain limiting cases in the phase diagram of fig.2
may be analyzed and understood easily . in the case @xmath47 and @xmath48 , the two gases do not interact with each other . in order to minimize their energy ,
they distribute homogeneously along the rings and coexist . in the other limiting case @xmath49 , for @xmath39 smaller than @xmath50
, the two components also coexist ; however , for larger values than 2 , they separate azimuthally .
although in this phase the kinetic energy increases , the interaction energy is lowered due to the repulsion between the two species , and the azimuthal symmetry - breaking persists with increasing @xmath39 .
it is also instructive to understand the internal structure of the phase diagram . as one moves vertically , i.e. , for a fixed value of @xmath39 ( being sufficiently large , such that the components separate azimuthally ) , for small enough @xmath40 the two components occupy mainly the inner ring since the repulsion between the particles can not compensate for the stronger confinement of the outer ring .
as @xmath40 increases , the outer ring becomes progressively more occupied .
when the inter - component repulsion @xmath39 becomes large enough , the gases minimize their energy by separating radially .
however , if @xmath40 increases further , the inner ring becomes too small to host one of the species entirely .
azimuthal phase separation takes place again , now with both gases being largely spread within the whole system .
eventually , for even larger values of @xmath40 , the dominant term in the energy is the intra - atomic interaction , and thus the two components coexist . as anticipated before
, we illustrate this effect in fig.[fig3 ] , where we plot the densities of the cases with @xmath51 , and ( a ) @xmath52 , ( b ) @xmath53 , ( c ) @xmath54 , and ( d ) @xmath55 , corresponding to azimuthal phase separation , radial phase separation , azimuthal phase separation , and phase coexistence , respectively .
it is also of interest to investigate the nature of the phase transitions occurring in the system .
as one crosses the boundary from coexistence to azimuthal phase separation , the two components decrease continuously their overlap , developing sharper profiles as the repulsion increases .
this transition is thus continuous ( second order ) , as it is also the case in purely one - dimensional single rings @xcite . in the corresponding energy surface ,
the minimum ( which determines the ground state ) moves continuously as one crosses the phase boundary . on the contrary ,
the transitions involving radial separation are discontinuous ( first order ) , indicating that two local minima in the energy surface compete and that the system jumps abruptly from the one state to the other .
of the two components , for @xmath51 and ( a ) @xmath52 , ( b ) @xmath53 , ( c ) @xmath54 , and ( d ) @xmath55 , corresponding to azimuthal phase separation , radial phase separation , azimuthal phase separation , and phase coexistence , respectively .
the peaks in the densities correspond to the minima of the inner and the outer parabolae of the confining potential.,width=340 ]
the phase transitions that we described above also have a clear influence on the rotational properties of the system .
we start by examining the response of the system to some finite rotational frequency @xmath56 of the trap . in the following
we determine the total angular momentum per particle @xmath57 as a function of @xmath0 , @xmath58 .
following the usual procedure , we minimize the energy of the system in the rotating frame , i.e. , we minimize @xmath59 , where @xmath60 is the total energy .
the result of this calculation is shown in fig.[fig4 ] , where we have set @xmath61 , @xmath62 , and @xmath63 .
the typical values of @xmath0 are rather small because we have scaled it with @xmath7 . however , at least in the case of a purely one - dimensional ring potential , the scale for the typical @xmath0 is on the order of the frequency @xmath64 of the corresponding kinetic energy @xmath65 @xcite , where @xmath66 is the radius of the ring . in atomic units , @xmath67 , @xmath68 , while @xmath69 ( for @xmath66 , say , equal to 5 ) .
this introduces a factor @xmath70 .
it is also instructive to comment on the behavior of the gas as @xmath0 increases and see the connection of the function @xmath71 with the density distribution of the two components . for @xmath72
the two species are separated radially and the total angular momentum is zero . for @xmath73 ,
the density of the two components breaks its azimuthal symmetry discontinuously , and the angular momentum jumps abruptly to a finite value . beyond this value of @xmath74
the angular momentum increases linearly with @xmath0 and is carried by both components , which undergo solid - body rotation .
this is an expected result due to the azimuthal symmetry - breaking of the density , as we have confirmed by studying the phases of the two order parameters .
when @xmath75 , a new plateau appears , which corresponds to an angular momentum per particle equal to two , and the two species separate radially .
as one can see from the plot , there is a sequence between plateaus at integer values of @xmath76 and straight lines with a positive slope , which are separated by abrupt jumps .
this sequence persists up to @xmath77 . beyond this value of @xmath0
there is no longer radial separation of the two components due to the large centrifugal force , which forces both gases to occupy the two potential minima and therefore to separate azimuthally .
similar discontinuous transitions in the function @xmath71 occur in a single - component weakly - interacting bose - einstein condensate that rotates in a harmonic trap @xcite , and are associated with discontinuous transitions between phases of different symmetries of the single - particle density distribution of the gas . in the present problem ,
the corresponding different symmetries are the ones where the components separate radially , or azimuthally .
furthermore , the jumps in the @xmath71 plot are consistent with the fact that for the parameters considered , the system supports persistent currents . in other words , had this function been continuous , then metastability would have not been possible ( for a discussion of this effect we refer to ref.@xcite ) . from the above observations we see
that the general picture that emerges by considering a finite rotation of the trap resembles the one where the couplings are varied , with a series of phase transitions between radial and azimuthal separation . , as a function of the angular frequency of rotation of the trap @xmath0 , which results from the minimization of the energy in the rotating frame.,width=340 ] another interesting question is the possible existence of ( meta)stable currents . in physical terms
, we investigate the energetic ( meta)stability of current - carrying states . in the case
that there is an energy barrier that separates a current - carrying state from the ground state , in the presence of some dissipative mechanism ( such as a thermal cloud , for example ) such a state does not decay , and therefore the system supports persistent currents .
we examine two separate aspects of this problem .
firstly , we consider the points of the phase diagram shown in fig.[fig2 ] , for a fixed population of the two gases .
secondly , we fix the interaction strengths and vary the ratio @xmath38 . in both cases
we examine the imaginary - time evolution of initial states with some finite , nonzero expectation value of the angular momentum , in the absence of any external rotation of the trap . for a given set of parameters , the existence of a converged final state with a nonzero expectation value of @xmath76
implies that the associated current is metastable .
we have examined all the points that are shown in the phase diagram of fig.[fig2 ] .
those corresponding to a final state with a nonzero expectation value of the angular momentum are represented in fig.2 with `` solid '' symbols , while the ones that decay to the non - rotating ground state are represented with `` open '' symbols .
clearly , the vast majority of states correspond to metastable currents , except those close to the diagonal @xmath78 , as well as those with sufficiently small @xmath40 ( for all values of @xmath39 ) .
the obtained results show that the angular momentum of the metastable states is always an integer multiple of the particle population , which implies that the associated densities are necessarily circularly symmetric .
this is consistent with the statement that circular symmetry is a necessary though not sufficient
condition for the ( meta)stability of the currents , as otherwise the circulation may escape from the gas ( since in this case there is no barrier separating the rotating state from the non - rotating one @xcite ) .
let us now study the effect of a variable relative population between the two gases on the ( meta)stability of the currents .
we thus fix the interaction strengths , as well as the population of the one component @xmath79 , and study this question starting from the case with @xmath80 , all the way up to the limit @xmath81 , with an initial state that has some angular momentum in component @xmath2 . since @xmath82 is fixed , the above procedure corresponds physically to keeping the scattering lengths fixed . as mentioned above
, the ( meta)stability of the persistent currents depends on the competition between azimuthal phase separation and circular symmetry of the density distribution of the two components .
thus , when @xmath80 , the @xmath2 component spreads within the whole system for any value of @xmath40 and the currents can be metastable , provided that the coupling @xmath40 is sufficiently large . if the population of the component @xmath1 becomes nonzero but is small enough , it acts only as a weak perturbation , and azimuthal symmetry is still preserved .
however , beyond a critical ratio @xmath38 , the two species separate azimuthally and the currents are no longer metastable .
this critical value depends on the actual intra- and inter - component interaction strengths .
a further increase of @xmath83 with respect to @xmath82 drives the system to a phase of radial separation , and metastability is recovered . finally , in the limit where @xmath83 becomes too large
, this component can not fit into only one of the two rings .
this leads again to azimuthal phase separation of the gases , which is preserved even in the limit @xmath81 . as a result
, the stability of the currents is lost again .
we show in fig.[fig5 ] the densities for the case @xmath54 , and @xmath51 and for various values of the ratio @xmath38 , in order to illustrate the mentioned sequence of phase transitions .
in particular , the results correspond to @xmath84 , @xmath85 , @xmath86 , and @xmath87 , with the first and the third cases corresponding to the metastable states .
according to our simulations , changing the values of the interaction strengths modifies the sizes of the `` windows '' in @xmath83 that separate the different phases , but yields qualitatively similar results . and @xmath88 with @xmath54 , and @xmath89 of the component @xmath1 ( dark blue ) and @xmath2 ( light yellow ) , for ( a ) @xmath84 , ( b ) @xmath90 , ( c ) @xmath86 , and ( d ) @xmath87 . in the cases ( a ) and ( c ) the states support metastable currents , while in ( b ) and ( d ) they do not .
the peaks in the densities correspond to the minima of the inner and the outer parabolae of the confining potential.,width=302 ]
as shown in the present study , a coupled system of two distinguishable bose gases that interact with an effectively - repulsive contact potential and are loaded in a concentric double annular trap reveals a series of phase transitions and the existence of metastable currents . for weak interactions , when the chemical potential is much smaller than the barrier that separates the two potential minima , the gases are confined in the inner ring , with the width of their transverse profile being smaller than the radius of the ring .
thus , their motion is ( at least ) close to being quasi - one - dimensional . on the other hand , as the couplings increase ,
the barrier plays a decreasingly important role , and their transverse width becomes comparable to the radius of the ring(s ) .
such a trapping potential interpolates up to some extent between one - dimensional and two - dimensional motion , depending on the strength of the coupling between the atoms . the interplay between this effect and the strength of the inter- and intra - atomic couplings gives rise to interesting phase transitions in the ground state of the system , including three different geometries : phase coexistence , radial phase separation , and azimuthal phase separation .
an interesting feature of the system considered is its response to some finite rotational frequency of the trap .
the basic picture resembles very much the one where the couplings are varied , inducing axial and/or radial phase separation of the two components .
the robustness in the ( meta)stability of the currents that we found for the vast majority of the points in the phase diagram of fig.[fig2 ] is another interesting aspect of this study .
metastability is not found for the cases when the couplings are all nearly equal or exactly equal to each other , as well as for the cases with small @xmath40 , independently of @xmath39 . according to ref.@xcite
, a necessary ( but not sufficient ) condition for metastability is that the trapping potential does not increase monotonically from the center of the trap .
the present results suggest that multiple variations in the monotonicity of the trapping potential enhance the stability of the currents .
last but not least , the discontinuous phase transitions we have found both in the ground - state phase diagram when the two gases separate radially as the couplings are varied , as well as in the response of the system when the rotation of the trap is varied , imply that hysteresis should show up as the coupling / rotational frequency increases / decreases .
we thank s. bargi and k. krkkinen for useful discussions .
this work was financed by the swedish research council .
the collaboration is part of the nordforsk nordic network `` coherent quantum gases - from cold atoms to condensed matter '' .
b. szafran and f. m. peeters , phys .
b * 72 * , 155316 ( 2005 ) ; j. m. escartn , f. malet , a. emperador , and m. pi , phys . rev .
b * 79 * , 245317 ( 2009 ) ; t. mano , t. kuroda , s. sanguinetti , t. ochiai , t. tateno , j. kim , t. noda , m. kawabe , k. sakoda , g. kido , and n. koguchi , nanoletters * 5 * , 425 ( 2005 ) ; s. viefers , p. s. deo , s. m. reimann , m. manninen , and m. koskinen , phys . rev .
b * 62 * , 10668 ( 2000 ) ; m. manninen , m. koskinen , s. m. reimann , and b. mottelson , eur . phys . j. d * 16 * , 381 . | a two - component bose - einstein condensate confined in an axially - symmetric potential with two local minima , resembling two concentric annular traps , is investigated .
the system shows a number of phase transitions that result from the competition between phase coexistence , and radial / azimuthal phase separation .
the ground - state phase diagram , as well as the rotational properties , including the ( meta)stability of currents in this system , are analysed . |
lagrangian correspondences were described by weinstein @xcite as generalizations of symplectomorphisms , in an attempt to build a symplectic category with composable morphisms between non - symplectomorphic manifolds . by definition a lagrangian correspondence from @xmath0 to @xmath1
is a lagrangian submanifold in the product , @xmath2 , with respect to the symplectic structure @xmath3 .
the basic examples are graphs of symplectomorphisms .
composition of symplectomorphisms generalizes to _ geometric composition _ of lagrangian correspondences @xmath4 , @xmath5 , defined by @xmath6 in general this will be a singular subset of @xmath7 which is isotropic at smooth points . however , if we assume transversality of the intersection @xmath8 , then the restriction of the projection @xmath9 to @xmath10 is an immersion @xcite , and hence @xmath11 is an immersed lagrangian correspondence
. we will study the class of _ embedded _ geometric compositions , for which in addition @xmath12 is injective , and hence @xmath13 is a smooth lagrangian correspondence .
lagrangian correspondences arise naturally in various contexts .
perutz @xcite proposed a construction of three and four - manifold invariants using floer theory for lagrangian correspondences in symmetric products , which generalize the tori in heegard floer homology @xcite .
seidel proposed a generalized version of his exact triangle in floer cohomology @xcite for fibered versions of symplectic dehn twists , whose vanishing cycle is a spherically fibered lagrangian correspondence .
seidel and smith @xcite proposed a symplectic definition of khovanov homology , using lagrangians constructed as geometric compositions of the fibered vanishing cycles .
finally , moduli spaces of flat bundles on three - dimensional cobordisms define lagrangian correspondences between the moduli spaces of bundles on the boundary surfaces , such that composition of cobordisms corresponds to geometric composition of correspondences .
the associated floer cohomology groups , which we construct in @xcite , may be viewed as symplectic versions of instanton floer homology for three manifolds .
naturally the question arises of how composition of correspondences affects floer cohomology . in this paper
we prove that floer cohomology is isomorphic under embedded geometric composition . for a precise general statement , it is best to use the language of quilted floer cohomology developed in @xcite which defines @xmath14 for a cyclic sequence of lagrangian correspondences @xmath15 between symplectic manifolds @xmath16 .
if the composition @xmath17 is embedded , then we obtain under suitable monotonicity assumptions a canonical isomorphism @xmath18 here the quilted floer cohomology on the left hand side counts @xmath19-tuples of pseudoholomorphic strips @xmath20\to m_j)_{j=0,\ldots , k-1}$ ] , whose boundaries match up via the lagrangian correspondences , @xmath21 . on the right hand side of , no strip in @xmath22 is taken into account , and the strips @xmath23 and @xmath24 match up directly via @xmath25 . rather than going through the general definition in detail
, we will prove in detail the following representative example in the familiar notation of floer cohomology for pairs of lagrangians in the same symplectic manifold .
[ main ] let @xmath26 be symplectic manifolds that are either compact or satisfy the ` bounded geometry ' assumptions as in ( * ? ? ?
* chapter 7 ) .
, as detailed in section [ sec : mon ] .
moreover , we require that transverse floer trajectory spaces be constructed as in section [ sec : qhf ] using almost complex structures @xmath27 such that , with respect to @xmath27-compatible metrics , up to second derivatives of @xmath27 as well as the curvature are uniformly bounded . ] .
let @xmath28 be compact lagrangian submanifolds such that the geometric composition @xmath29 is embedded .
then the canonical bijection @xmath30 induces an isomorphism @xmath31 provided the following assumptions hold : 1 .
[ b ] the pair @xmath32 , @xmath33 of lagrangian submanifolds in @xmath34 is monotone ( or exact ) for floer theory , that is with some @xmath35 ( or @xmath36 ) we have @xmath37 for all maps from the annulus @xmath38\to
m_0\times m_1 ^ -\times m_2 $ ] with lagrangian boundary conditions @xmath39 and @xmath40 .
the maslov index is defined by choosing a trivialization @xmath41\times{\mathbb{c}}^n$ ] , then @xmath42 is the difference of maslov indices of the two loops in the lagrangian grassmannian of @xmath43 .
[ c ] the minimal positive maslov index in is @xmath44 , that is there exists no annulus @xmath45 with @xmath46 .
[ d ] each of the @xmath47 has minimal maslov index @xmath48 .
( here the minimal maslov index of @xmath49 is the positive generator of @xmath50 . )
note that ( a ) implies monotonicity on homotopy groups for the symplectic manifolds , i.e. @xmath51 = \tau c_1(tm_i)$ ] on @xmath52 for @xmath53 , as well as for each lagrangian , i.e. @xmath54 = \tau i_{\rm maslov}$ ] on @xmath55 for @xmath56 given by @xmath57 , @xmath58 , @xmath59 , or @xmath60 .
assumptions and are necessary in their full strength for a subtle bubble exclusion argument , as explained below .
they are met , for example , if all lagrangians are orientable and exact , or if they are orientable , monotone , and the image of either @xmath61 or @xmath62 in @xmath63 is torsion . in @xcite
we discuss some alternative conditions ensuring monotonicity .
note that also is the natural assumption that excludes self - connecting trajectories in the construction of floer homology .
similarly , is needed only to ensure that floer homology is well defined . in @xcite
we generalize theorem [ main ] to an isomorphism in the derived category of matrix factorization , allowing to drop assumption . in this paper , the isomorphism of floer cohomology groups
is completely proven only with @xmath64-coefficients .
the discussion of coherent orientations in the presence of orientations and relative spin structures on the lagrangians can be found in @xcite .
there should also be versions of this result for floer cohomology with gradings , coefficients in flat vector bundles , and novikov rings .
we give a detailed statement and proof for the gradings in @xcite .
( 0,0 ) and for @xmath65,title="fig : " ] # 1#2#3#4#5 @font ( 12252,4524)(136,-4213 ) ( 226,-3880)(0,0)[lb]@xmath66 ( 436,-2600)(0,0)[lb]@xmath67 ( 2296,-2186)(0,0)[lb]@xmath68 ( 2251,-3086)(0,0)[lb]@xmath69 ( 2251,-3986)(0,0)[lb]@xmath0 ( 6211,-350)(0,0)[lb]@xmath70 ( 5931,-1750)(0,0)[lb]@xmath71 ( 8126,-2321)(0,0)[lb]@xmath68 ( 8126,-3446)(0,0)[lb]@xmath0 ( 5511,-3250)(0,0)[lb]@xmath29 ( 11900,-50)(0,0)[lb]@xmath72 ( 11900,-1200)(0,0)[lb]@xmath71 throughout we will use the construction of floer cohomology based on @xcite . the floer differential for @xmath73 counts triples of pseudoholomorphic strips in @xmath74 ( see figure [ layer ] below ) . in the standard definition
, one would take the width of all three strips to be equal , but we show in @xcite that one can in fact allow the widths of the strips to differ .
the main difficulty then is to prove that under the stated assumptions and with the width of the middle strip sufficiently close to zero , the triples of pseudoholomorphic strips in @xmath74 are in one - to - one correspondence with the pairs of pseudoholomorphic strips in @xmath75 that are counted in the floer differential for @xmath76 . as in similar situations in floer theory ,
the proof is an application of the implicit function theorem , on one hand , and compactness results for shrinking the middle strip , on the other . in the limit
various kinds of bubbling can occur : sphere bubbles in @xmath0 , @xmath1 , @xmath68 ; disk bubbles in @xmath57 , @xmath60 , @xmath77 , @xmath78 , @xmath79 ; and a novel type of bubble which we call a _ figure eight bubble_. the latter is a triple of pseudoholomorphic maps
@xmath80 \to m_0 , v_1 : { \mathbb{r}}\times [ -1,1 ] \to m_1 , v_2 : { \mathbb{r}}\times [ 1,\infty ) \to m_2 $ ] such that @xmath81 , @xmath82 . to explain the name , note that under stereographic projection to the sphere , or after transformation @xmath83 of @xmath84 , the lines @xmath85 appear as a figure eight as in figure [ figureeight ] .
these pictures are labeled in the pictorial language of @xcite : the maps @xmath86 form a `` quilt '' on the punctured @xmath87 , whose `` patches '' are the domains of the three maps ( labeled by the target spaces ) , and with seams " on the intersections of these domains ( labeled by the `` seam condition '' @xmath66 or @xmath70 that is satisfied there ) .
we conjecture that the maps @xmath88 can be extended continuously to the closure of their domains in @xmath87 by a point @xmath89 .
however , we can not in general prove this removal of singularities , nor is there a readily available fredholm theory for seams touching tangentially as in figure [ figureeight ] .
thus we are lacking the construction of a moduli space of figure eight bubbles . instead , as in @xcite we exclude bubbling by energy quantization without establishing a geometric description of the bubble .
this method hinges on strict monotonicity with a nonnegative constant @xmath90 as well as the @xmath44-grading assumption .
theorem [ main ] has a wide range of applications : first , it provides a tool for symplectic topology , which has not yet been exhaustively used . in @xcite
we give examples of elementary floer homology calculations arising from the representation of symplectic quotients as lagrangian correspondence .
for example , a simple iteration in @xmath91 confirms the calculation @xmath92 of cho @xcite for the clifford torus in @xmath93 .
we also show that non - displaceability of lagrangians in product symplectic manifolds follows directly if the lagrangian , viewed as correspondence , has an image in one of the factors whose floer homology is nonzero .
this explains e.g. the nondisplaceability of standard lagrangian embeddings @xmath94 ( for example the sphere @xmath95 ) by the fact that their projection to @xmath93 contains the nondisplaceable clifford torus .
an application to non - triviality of symplectic mapping class groups is given in @xcite .
second , our isomorphism is key to proving the topological invariance of various floer cohomology groups arising from decompositions in low - dimensional topology ; for example , the symplectic version of instanton ( knot ) homology constructed in @xcite , and seidel - smith homology and heegard - floer homology , for which it provides alternative constructions @xcite , @xcite . ] .
third , from a more conceptual point of view , theorem [ main ] is used in @xcite to give a solution to the problem in weinstein s construction that composition of lagrangian correspondences is not always defined . using the result here , one may construct a symplectic @xmath44-category , in which all lagrangian correspondences are composable morphisms and floer cohomology groups ( as @xmath44-morphism spaces ) are well defined .
thus one removes the quotes in weinstein s category " by promoting the construction to a @xmath44-category , using floer theory .
+ _ we thank paul seidel and ivan smith for encouragement and helpful discussions . _
in this section we first explain why both floer cohomologies in theorem [ main ] are well defined .
then we give a specific quilted " setup and choice of perturbations for both that reduce the isomorphism of floer cohomologies to a bijection of moduli spaces that is proven in section [ shrink ] .
the significance of the monotonicity and maslov index assumptions in theorem [ main ] is the following energy - index relation and relative grading .
[ prop monotone ] suppose that the pair @xmath96 of lagrangians in @xmath97 is monotone , transverse , and has minimal annulus maslov index @xmath98 .
( that is , @xmath99 is the positive generator of @xmath100 \to m , v(s^1\times\{j\})\subset l_j\}\subset{\mathbb{z}}$ ] . )
then for any @xmath101 there exist constants @xmath102 and @xmath103 such that for all strips @xmath104\to m$ ] with boundary values in @xmath96 and limits @xmath105 we have @xmath106 here @xmath107 is the energy and @xmath108 the linearized cauchy - riemann operator at @xmath109 . given two strips @xmath110\to m$ ] we can glue them together ( reversing the orientation of @xmath111 ) to an annulus @xmath38 \to m$ ] , then @xmath112 and @xmath113 .
so the energy - index relation follows from monotonicity @xmath114 , and the index identity follows from @xmath115 .
the energy - index relation ensures energy bounds for the moduli spaces of fixed index and thus compactness up to bubbling ( ` pointwise blow - up of the gradient ' ) and breaking of trajectories ( ` nontrivial amounts of energy moving off into both ends of the strip').-bounds on the holomorphic maps , before standard gromov compactness can be quoted .
note that the domains of maps under our consideration are such that each interior point has bounded distance from a boundary point , where the maps take values in a compact lagrangian submanifold or in the projection of a compact lagrangian correspondence to one factor .
hence it suffices to establish uniform bounds on the gradient ( i.e. exclude bubbling ) . ] together with the index identity it excludes bubbling in moduli spaces of index less than @xmath99 as follows : any bubbling leads to a new ( possibly broken ) trajectory connecting the same points but with less energy.[multiblock footnote omitted ] by monotonicity , less energy means strictly less index . by the index identity mod @xmath99 that means negative index . by transversality ( previously established for moduli spaces of negative index ) that means an empty set : the new trajectory does nt exist , so the bubbling did nt happen .
we spelled out this argument because we will use it again to exclude figure eight bubbling by only proving energy loss , not actually giving a geometric description of the bubble . working with
@xmath116 there is just one point in the construction of floer cohomology where this argument fails : the @xmath117-dimensional moduli spaces of self - connecting floer trajectories have index @xmath44 , so bubbling could lead to an index @xmath118 solution ( which are always constant due to the @xmath119-action ) .
assumption serves to exclude this scenario by index additivity arguments : any pseudoholomorphic disk bubble with boundary on @xmath120 will reduce the index by at least @xmath121 , the minimal maslov index on @xmath55 .
so @xmath122 ensures that the remaining solution would have negative index ( and the same holds for sphere bubbles whose chern number would be at least @xmath123 ) .
note that this argument , unlike the previous bubbling exclusion by energy loss , requires an identification of the bubbles as spheres and disks .
( e.g. exact lagrangians in a cotangent bundle have @xmath124 ) or use any other valid argument to prove @xmath125 . ] in our case it also requires that we work with a split almost complex structure ( preserving the factors of @xmath34 ) , otherwise pseudoholomorphic disks in the product manifold do nt necessarily have the minimal index of a disk in one of the factors .
we will show in section [ sec : qhf ] that we can achieve transversality with a split almost complex structure , and hence our assumptions indeed ensure that the floer cohomology @xmath126 is well defined .
the next lemma shows that the floer cohomology @xmath127 for the composed lagrangian correspondence is also well defined .
[ neednoviterbo ] in the setting of theorem [ main ] , the assumptions and imply the analogous assumptions for the pair @xmath128 of lagrangians in @xmath129 .
assumption implies that @xmath125 on @xmath130 , and hence the floer cohomology is well defined .
consider any annulus @xmath131\to m_0\times m_2 ^ -$ ] with lagrangian boundary conditions @xmath132 and @xmath133 . by the embedded composition there exists a unique lift @xmath134 such that @xmath135 and @xmath136 .
now we can reverse the parametrization in @xmath137 and extend @xmath138 constant along @xmath139 $ ] to define an annulus @xmath140\to m_0\times m_1 ^ -\times m_2 $ ] as in . here
@xmath141 , hence @xmath142 . to identify the maslov indices ,
pick the same trivializations @xmath143\times v_j$ ] for @xmath144 in both cases , then equality follows from the identity @xmath145 for loops of lagrangians @xmath146 , @xmath147 , and @xmath148 given by @xmath149 . the first equality is simply additivity of the maslov index . to see the second equality we fix lagrangians @xmath150 for @xmath144 ,
then the maslov indices can be expressed as the intersection number with @xmath151 resp .
@xmath152 . with this choice
the intersections are identified , @xmath153 now we need to compare the crossing forms @xmath154 at regular crossings @xmath155 .
fix a lagrangian complement @xmath156 , then @xmath157 , after appropriate transposition of factors , is a lagrangian complement for @xmath158 , due to the assumption of transversality @xmath159 .
so for @xmath160 one finds @xmath161 and @xmath162 such that @xmath163 . for the corresponding vector @xmath164 this automatically gives @xmath165 . with this
we identify the crossing forms @xmath166 this proves equality of the maslov indices in and this finishes the proof of and . in the absence of assumption
we have @xmath167 a multiple of the identity in both floer theories , see @xcite and @xcite .
a derived version of theorem [ main ] implies that the value of @xmath168 is the same for both theories , see remark [ rmk : derived main ] .
assuming for the pair @xmath169 we obtain @xmath170 and thus also @xmath125 on @xmath130 .
the index calculation in analogously holds for strips .
this identifies the index on the two complexes in theorem [ main ] .
recall here from @xcite that the index of the linearized cauchy - riemann operator @xmath108 at a map @xmath104\to m$ ] with lagrangian boundary conditions @xmath171 for @xmath172 and limits @xmath173 at transverse intersection points is given by the maslov - viterbo index , @xmath174 here the maslov index of the pair of paths is defined by choosing a trivialization @xmath175 \times v$ ] ( independent of @xmath176 $ ] for @xmath177 ) so that @xmath178 becomes a path of lagrangian subspaces in the symplectic vector space @xmath179 .
[ index ] let @xmath180 , @xmath4 , @xmath5 , and @xmath181 be lagrangians such that the composition @xmath182 is embedded .
suppose that the intersection @xmath183 ( and hence also @xmath184 ) is transverse and consider a map @xmath185\to m_0\times m_2 $ ] taking boundary values in @xmath186 , and limiting to intersection points as @xmath177 .
let @xmath187\to m_0 \times m_1 \times m_2 $ ] be the corresponding map which takes boundary values in @xmath188 and satisfies @xmath189 .
( here @xmath190 reverses the @xmath139$]-parametrization of @xmath111 . ) then the indices of the linearized operators and the energies are equal , @xmath191 the identity of maslov indices follows as in lemma [ neednoviterbo ] .
alternatively , it could be deduced from a more general result of viterbo ( * ? ? ?
* proposition 3 ) . for
the energies just note that @xmath192 and @xmath193 . as in theorem
[ main ] let @xmath26 be symplectic manifolds and let @xmath28 be lagrangian submanifolds such that the geometric composition @xmath194 is embedded .
the aim of this section is to introduce the quilted " setup and give compatible choices of perturbation data for the two floer cohomologies @xmath126 and @xmath195 .
first , we need to fix hamiltonians such that the perturbed intersection points are finite and nondegenerate .
in fact , the following proposition shows that we can pick a hamiltonian of split type which achieves simultaneous transversality for the intersection points in both floer theories . given a pair of time - dependent hamiltonian functions @xmath196
\times m_0)\times c^\infty([0,1 ] \times m_2)$ ] consider the hamiltonians @xmath197 on @xmath198 and @xmath199 on @xmath200 and denote their time @xmath117 flows by @xmath201 and @xmath202 .
then the perturbed intersection points @xmath203 can be identified with @xmath204 and analogously @xmath205 where @xmath206 is the time @xmath117 flow of the hamiltonian @xmath207 and we use the trivial function @xmath208 on @xmath1 .
note that the hamiltonians are constructed such that the perturbed intersection points for the two floer theories are still canonically identified . indeed , by assumption every point in @xmath209 has a unique lift to @xmath210 .
[ pert split ] there is a dense open subset @xmath211 \times m_0)\times c^\infty([0,1 ] \times m_2)$ ] such that for every @xmath212 and @xmath208 the defining equations for both sets @xmath213 and @xmath214 are transversal . by assumption @xmath215
are embedded submanifolds and so locally they are the zero sets of submersions @xmath216 , @xmath217 , @xmath218 . then the defining equations for @xmath213
are @xmath219 consider the universal moduli @xmath220 space of data @xmath221 satisfying , where now each @xmath222 has class @xmath223 for some @xmath224 .
the linearized equations for @xmath220 are @xmath225 for @xmath226 and @xmath227 \times m_j)$ ] .
the product of the operators on the left - hand sides of are surjective since each of the maps @xmath228 \times m_{j } ) \to t_{\phi^{h_j}(m_{j } ) } m_{j}$ ] , @xmath229 is surjective .
so by the implicit function theorem @xmath220 is a smooth banach manifold , and we consider its projection to @xmath228 \times m_0)\times c^\ell([0,1 ] \times m_2)$ ] . by the sard - smale theorem ,
the set of regular values is dense . on the other hand , the set of regular values is clearly open .
hence the set of smooth functions that are regular values is open and dense .
this is exactly the set of functions @xmath230 such that the perturbed intersection @xmath213 is transversal .
moreover , the perturbed intersection @xmath214 is also transversal , since by assumption @xmath231 is transverse to the diagonal @xmath232 . in the following , instead of working with perturbed intersection points , we will apply the hamiltonian diffeomorphisms to the lagrangians to achieve transversality .
replacing @xmath71 with @xmath233 and @xmath67 with @xmath234 the generators of the two floer chain groups are the transverse intersections @xmath235 the forgetful map @xmath236 is a bijection from @xmath237 to @xmath238 since by assumption @xmath239 is bijective .
so , after a hamiltonian perturbation , we have a natural isomorphism of the floer chain groups @xmath240 and it remains to identify the floer differentials . for that purpose
we now drop the hamiltonian from the notation : by abuse of notation we can assume to start out with unperturbed transverse intersections and a natural bijection @xmath241 to investigate the floer trajectories note that we consider @xmath242 as a pair of lagrangians in @xmath129 and @xmath243 as a pair of lagrangians in @xmath34 . for any symplectic manifold @xmath97 let @xmath244 be the space of almost complex structures on @xmath97 that are compatible with the symplectic structure @xmath245 .
is noncompact , then we assume as in @xcite that the almost complex structure extends to the compact symplectic manifold with boundary and corners , whose interior is @xmath97 .
more generally , it would suffice to work with any noncompact @xmath97 and @xmath27 for which the bubble exclusion arguments hold , as detailed in section [ sec : mon ] and lemma [ bubb ] .
in particular , this requires uniform bounds on the curvature and up to second derivatives of the almost complex structures @xmath27 with respect to @xmath27-compatible metrics .
] we pick time - dependent almost complex structures @xmath246,{\mathcal{j}}(m_0))$ ] and @xmath247,{\mathcal{j}}(m_2))$ ] , then @xmath248 defines a compatible almost complex structure on @xmath129
. now any pseudoholomorphic strip @xmath249\to m_0\times m_2 ^ -$ ] with boundary values on @xmath242 corresponds by unfolding " @xmath250 to a pair of strips @xmath251\to m_i\bigr)_{i=0,2}$ ] satisfying @xmath252 similarly , pick an almost complex structure @xmath253 , then @xmath254 defines a compatible almost complex structure on @xmath34 and any pseudoholomorphic strip with boundary values on @xmath243 corresponds by unfolding " to a triple of strips @xmath255\to m_i\bigr)_{i=0,1,2}$ ] satisfying @xmath256 in both cases , the trajectories have finite energy @xmath257 resp .
@xmath258 iff they converge uniformly to intersection points @xmath259 for any @xmath260 let us denote by @xmath261 the one dimensional ( i.e. index @xmath117 ) component of the moduli space of floer trajectories for @xmath242 .
one can achieve transversality of these moduli spaces ( of any index @xmath262 ) by choosing @xmath263-dependent almost complex structures @xmath264 and @xmath265 that are constant near @xmath266 and @xmath267 .. it implies that the set of regular points , @xmath268 with @xmath269 and @xmath270 , is open and dense .
these points can be used to prove surjectivity of the linearized operator for a universal moduli space of solutions with respect to split almost complex structures @xmath271 .
( the constant solutions are automatically transverse due to the previously ensured transversality of the intersection points . )
note that it suffices to work with almost complex structures that are @xmath263-independent outside of @xmath272 $ ] .
the existence of a comeagre set of regular @xmath271 then follows from the usual sard - smale argument as in @xcite . ]
note that we can not expect a bijection with the moduli spaces of floer trajectories for @xmath243 as in .
however , by the independence theorem in @xcite , the cohomology defined from the above floer differential is isomorphic to the cohomology defined by the quilted floer differential " arising from the moduli spaces @xmath273 for any choice of @xmath274 .
here we consider strips @xmath275 of width @xmath117 as before but middle strips @xmath276\to m_1 $ ] of width @xmath274 , and @xmath277 denotes the same boundary value problem as above except for the seam condition @xmath278 .
moreover , we use almost complex structures @xmath279 that converge to @xmath280 in the @xmath281-topology as @xmath282 .
the specific choice follows from the constructions in the proof are given by rescaling @xmath264 to @xmath283 $ ] and @xmath265 to @xmath284 $ ] , and extending them constantly by @xmath285 and @xmath286 respectively .
the convergence holds since each @xmath287 is smooth and constant near @xmath288 . ] and will also ensure that the moduli spaces @xmath289 are cut out transversely for @xmath274 sufficiently small .
( 0,0 ) # 1#2#3#4#5 @font ( 9549,3135)(589,-2461 ) ( 2476,504)(0,0)[lb]@xmath290 ( 2476,-700)(0,0)[lb]@xmath291 ( 2476,-1111)(0,0)[lb]@xmath292 ( 2476,-2300)(0,0)[lb]@xmath293 ( 8751,-2130)(0,0)[lb]@xmath293 ( 8751,-961)(0,0)[lb]@xmath294 ( 8751,300)(0,0)[lb]@xmath290 ( 7626,-511)(0,0)[lb]@xmath295 ( 7626,-1486)(0,0)[lb]@xmath296 ( 1276,-1600)(0,0)[lb]@xmath296 ( 1276,-900)(0,0)[lb]@xmath297 ( 1276,-200)(0,0)[lb]@xmath295 ( 4750,-850)(0,0)[lb]@xmath298 in order to prove theorem [ main ] it now suffices to show that the isomorphism of chain groups descends to cohomology for an appropriate choice of @xmath274 .
we will prove this by establishing a bijection between the floer trajectories for @xmath299 on strips of width @xmath300 and those for @xmath301 on strips of width @xmath302 for sufficiently small width @xmath274 of the middle strip .
these floer trajectories are pseudoholomorphic quilts associated to the pictures in figure [ shrinkfig ] .
more precisely , we will consider the ( zero dimensional , compact ) moduli spaces of floer trajectories modulo @xmath119-translation and prove the following .
[ traj bij ] under the assumptions , of theorem [ main ] and for all sufficiently small @xmath274 , the moduli spaces @xmath303 are regular and there is a bijection @xmath304 [ rmk : derived main ] in the situation of theorem [ main ] except for assumption , the constructions in this section provide naturally isomorphic chain groups @xmath305 and @xmath306 and well defined differentials @xmath307 resp .
@xmath308 on them , defined from the moduli spaces @xmath309 and @xmath310 . as discussed in section [ sec : mon ] , due to obstructions from disks of minimal maslov index @xmath44 , both differentials square to a multiple of the identity , see @xcite and @xcite .
so we have @xmath311 and @xmath312 for any @xmath274 ( as long as the moduli spaces @xmath310 are regular ) . now theorem [ traj bij ] implies that for sufficiently small @xmath274 and any @xmath313 ( viewed as generator in both chain groups ) we have @xmath314 , and hence @xmath315 .
( if @xmath237 is empty then both theories are trivial . )
if @xmath316 ( e.g. by assumption ) or @xmath317 for some other reason , then this proves that both floer cohomologies are well defined and ( again by theorem [ traj bij ] ) are isomorphic . for any value of @xmath315 this proves that there exists a canonical isomorphism @xmath318 in the derived category of factorizations of @xmath319 .
in this section we prove theorem [ traj bij ] . we start by describing the strategy of proof and introducing the relevant notations .
first we use the assumption that @xmath13 is embedded by @xmath12 . consider a solution @xmath320 , that is a pair @xmath321\to m_0 $ ] ,
@xmath322\to m_2 $ ] of index @xmath117 , with limits @xmath323 , and satisfying @xmath324 we can identify @xmath325 with the map @xmath326\to m_0\times m_2 $ ] given by @xmath327 , which satisfies @xmath328 and @xmath329 here we denoted @xmath330 .
we will also denote @xmath331 and @xmath332 .
finally , we will denote by @xmath333 the obvious transposition of factors .
since @xmath334 is transversal and embedded , there is a unique smooth map @xmath335 such that @xmath336 this provides the lift @xmath337 .
we also denote by @xmath338 the extension @xmath339 \to m_0\times m_2\times m_1\times m_1 $ ] that is constant along @xmath340 $ ] . given @xmath298 these choices are unique , so we can identify @xmath109 with the pair @xmath341 . in the same spirit we find unique points @xmath342 such that @xmath343 . in this notation
we have the limit @xmath344 . given @xmath345 as above and @xmath274 we wish to find a corresponding @xmath346 , that is a triple @xmath347\to m_0 $ ] , @xmath276\to m_1 $ ] , @xmath348\to m_2 $ ] with limits @xmath349 , @xmath350 , and satisfying @xmath351 here @xmath279 are given by linearly rescaling @xmath264 to @xmath283 $ ] and @xmath265 to @xmath284 $ ] , and extending them constantly by @xmath285 and @xmath286 respectively . this choice of almost complex structures is more natural in the following reformulation of the @xmath298-moduli spaces .
let @xmath352 ( or equivalently @xmath353 ) . instead of the triple strip
we consider a quadruple of maps @xmath354 with @xmath355,m_0\times m_2)$ ] , @xmath356,m_0\times m_2)$ ] , @xmath357,m_1)$ ] that have limits @xmath358 , @xmath359 , and satisfy @xmath360 for notational convenience we will also group these quadruples of maps as @xmath361 with @xmath362
. then we can abbreviate @xmath363 with @xmath364 , and reformulate ( [ vvvv ] ) as @xmath365 we denote the moduli space of such solutions @xmath361 by @xmath366 .
it is in one - to - one correspondence to @xmath303 as follows : given @xmath367 we obtain @xmath368 from @xmath369 the two different formulations for double and triple strips each are indicated in figure [ strips ] .
( 12309,6229)(451,-7103 ) ( 10726,-5636)(0,0)[lb]@xmath370 ( 11776,-5636)(0,0)[lb]@xmath371 ( 11551,-4136)(0,0)[lb]@xmath293 ( 11551,-7136)(0,0)[lb]@xmath290 ( 4126,-5636)(0,0)[lb]@xmath372 ( 2851,-5636)(0,0)[lb]@xmath373 ( 1801,-5936)(0,0)[lb]@xmath374 ( 4276,-4136)(0,0)[lb]@xmath293 ( 4276,-7136)(0,0)[lb]@xmath290 ( 1801,-5336)(0,0)[lb]@xmath374 ( 7276,-3836)(0,0)[lb]@xmath375 : ( 9151,-5111)(0,0)[lb]@xmath376 ( 9151,-5711)(0,0)[lb]@xmath377 ( 9701,-6236)(0,0)[lb]@xmath291 ( 9701,-5086)(0,0)[lb]@xmath292 ( 8426,-4436)(0,0)[lb]@xmath378 ( 8426,-5636)(0,0)[lb]@xmath379 ( 8426,-6836)(0,0)[lb]@xmath380 ( 1151,-4436)(0,0)[lb]@xmath378 ( 1076,-5036)(0,0)[lb]@xmath292 ( 1151,-5636)(0,0)[lb]@xmath379 ( 1076,-6236)(0,0)[lb]@xmath291 ( 1151,-6836)(0,0)[lb]@xmath380 ( 4051,-2261)(0,0)[lb]@xmath111 ( 8251,-2261)(0,0)[lb]@xmath381 ( 11326,-2261)(0,0)[lb]@xmath382 ( 451,-2811)(0,0)[lb]@xmath293 ( 7126,-2811)(0,0)[lb]@xmath293 ( 2926,-2811)(0,0)[lb]@xmath294 ( 5326,-2811)(0,0)[lb]@xmath290 ( 10226,-2811)(0,0)[lb]@xmath291 ( 12526,-2811)(0,0)[lb]@xmath290 ( 9526,-2811)(0,0)[lb]@xmath292 ( 1576,-2261)(0,0)[lb]@xmath383 ( 7276,-1436)(0,0)[lb]@xmath384 : ( 9826,-2261)(0,0)[lb]@xmath377 ( 676,-1436)(0,0)[lb]@xmath385 : the bijection @xmath386 to the moduli space @xmath387 can then be established via a bijection @xmath388 this map will be constructed by the implicit function theorem [ solving ] .
we prove injectivity in corollary [ injective ] , and the surjectivity will follow from the compactness theorem [ compact ] . the purpose of this section is to construct the map @xmath389 of theorem [ traj bij ] .
we will do this by constructing the map , with @xmath390 replaced by @xmath298 , from the following implicit function theorem .
[ solving ] there exist constants @xmath391 , @xmath392 , and @xmath393 such that the following holds for every @xmath394 $ ] . for every @xmath345 there exists a unique @xmath395 such that @xmath396 with @xmath397 .
the solution moreover satisfies @xmath398 here @xmath399 is given in terms of @xmath400 and @xmath401 with @xmath402 and @xmath403 .
the precise definitions of the exponential map @xmath404 , the @xmath405-ball @xmath406 , the @xmath407-norm , and the local slice @xmath408 of the @xmath119-shift symmetry will be given in the process of the proof .
to prove the theorem we fix a solution @xmath345 , and in the following will allow all constants to depend on @xmath109 up to translation in @xmath119 .
( since @xmath387 is finite we can then easily find uniform constants @xmath391 and @xmath393 . ) we will then roughly solve @xmath409 for sections @xmath401 , @xmath410 satisfying the boundary conditions @xmath411 the exponential map @xmath412 will then be constructed such that the nonlinear lagrangian boundary conditions are satisfied automatically .
the index of the new solution @xmath413 will coincide with that of the given solution @xmath109 due to lemma [ index ] .
here we identified @xmath413 with a solution @xmath414 , @xmath415 .
then the homotopy between @xmath396 and @xmath341 induces a homotopy @xmath416 .
to set up the implicit function theorem we introduce the space of @xmath417-sections over @xmath341 for @xmath418 , @xmath419,u_{02}^ * t(m_0\times m_2 ) ) , \\
\eta_{02}'\in h^k({\mathbb{r}}\times[0,\delta],\bar u_{02}^ * t(m_0\times m_2 ) ) , \\ \eta_1',\eta_1\in h^k({\mathbb{r}}\times[0,\delta],\bar u_1^ * tm_1 ) \qquad\quad \end{aligned}\right\}\right .. \ ] ] we also write these sections as @xmath420 , where the subscripts indicate the width of the domains of @xmath421 and @xmath422,\bar u^ * t(m_0\times m_2\times m_1\times m_1))$ ] .
the corresponding @xmath417-norm on this space is @xmath423)}^2 + \|\hat\eta\bigr\|_{h^k({\mathbb{r}}\times[0,\delta])}^2 \\ & \qquad = \|\eta_{02}\bigr\|_{h^k({\mathbb{r}}\times[0,1])}^2 + \|\eta_{02}'\bigr\|_{h^k({\mathbb{r}}\times[0,\delta])}^2 + \|\eta_1'\bigr\|_{h^k({\mathbb{r}}\times[0,\delta])}^2 + \|\eta_1\bigr\|_{h^k({\mathbb{r}}\times[0,\delta])}^2 .\end{aligned}\ ] ] we denote the space of @xmath424-sections satisfying the boundary conditions by @xmath425 and equip this space with the norm @xmath426 where we added the @xmath427-norm @xmath428)}^4 + \|\nabla\hat\xi\|_{l^4({\mathbb{r}}\times[0,\delta])}^4 \bigr)^{1/4 } $ ] on the multi - strip .
we denote the @xmath405-ball in @xmath429 by @xmath430 we equip the target space @xmath431 with the norm @xmath432 the reason for adding the @xmath427-norms in domain and target is that we do not have uniform sobolev embeddings on the strips of varying width .
instead , we build the necessary sobolev multiplication properties into the norms .
the definitions of all these norms also involves a choice of metric on each manifold @xmath26 .
different choices yield equivalent norms .
are noncompact , since the images of @xmath372 , @xmath433 are contained in compact sets . ]
( existence of compatible quadratic corrections ) [ qq ] there exists @xmath434 and smooth families of maps ( defined on the @xmath435-balls ) @xmath436 @xmath437,\ ] ] that are a diffeomorphism onto their image and have the following properties : ( quadratic ) : : @xmath438 , @xmath439 , @xmath440 , and @xmath441 for all @xmath442 $ ] .
in particular , there is a constant @xmath443 such that for all @xmath444 and @xmath445 @xmath446 ( linearizing @xmath447 ) : : @xmath448 maps @xmath449 to @xmath450 .
( linearizing @xmath451 ) : : @xmath452 maps @xmath453 to @xmath454 .
( linearizing @xmath294 ) : : @xmath455 maps @xmath456 to @xmath457 .
( compatible ) : : restricting @xmath458 to @xmath459 and composing it with the projection @xmath460 yields a map that is independent of the @xmath461-component .
the resulting family @xmath462 coincides with @xmath463 .
we fix @xmath464 and restrict the exponential map @xmath465 to a geodesic ball around @xmath118 .
the subsequent constructions will depend smoothly on @xmath464 , which we drop from now on . by assumption
the submanifold @xmath466 in the vector space @xmath467 is transverse to the subspace @xmath468 .
their intersection @xmath469 is diffeomorphic to the submanifold @xmath470 by a map @xmath471 with uniquely determined @xmath472 .
so we have a direct sum decomposition @xmath473 as a submanifold we can now write @xmath474 as the graph of a map @xmath475 over a sufficiently small @xmath405-ball , @xmath476 with @xmath477 and @xmath478 .
we moreover pick a complement @xmath479 of @xmath480 , @xmath481 then the transversality @xmath482 implies the splitting @xmath483 we write @xmath484 in this splitting and define a map @xmath485 by @xmath486 ( 7388,5439)(0,-10 ) ( 704.214,2881.071)(582.000,2862.000)(698.811,2821.315 ) ( 582,2862)(6587,2319 ) ( 6464.786,2299.929)(6587.000,2319.000)(6470.189,2359.685 ) ( 3143.483,1259.299)(3162.000,1137.000)(3203.214,1253.625 ) ( 3162,1137)(3447,4137 ) ( 3465.517,4014.701)(3447.000,4137.000)(3405.786,4020.375 ) ( 1872,4677)(387,4812)(12,642 ) ( 6612,12)(6972,4212)(5367,4362 ) ( 1857,4677)(5352,4362 ) ( 4570,2795)(4548,2538 ) ( 2202,5412)(2204,5412)(2207,5410 ) ( 2214,5408)(2224,5405)(2239,5400 ) ( 2260,5394)(2285,5386)(2316,5377 ) ( 2353,5366)(2395,5353)(2441,5340 ) ( 2493,5324)(2548,5308)(2606,5291 ) ( 2667,5274)(2730,5256)(2794,5238 ) ( 2858,5220)(2923,5202)(2986,5185 ) ( 3049,5168)(3111,5152)(3171,5137 ) ( 3230,5123)(3287,5109)(3342,5097 ) ( 3395,5085)(3446,5074)(3495,5065 ) ( 3543,5056)(3590,5048)(3635,5040 ) ( 3678,5034)(3721,5029)(3763,5024 ) ( 3804,5020)(3844,5017)(3884,5015 ) ( 3923,5013)(3963,5012)(4002,5012 ) ( 4041,5012)(4081,5013)(4120,5015 ) ( 4160,5017)(4200,5020)(4241,5024 ) ( 4283,5029)(4326,5034)(4369,5040 ) ( 4414,5048)(4461,5056)(4509,5065 ) ( 4558,5074)(4609,5085)(4662,5097 ) ( 4717,5109)(4774,5123)(4833,5137 ) ( 4893,5152)(4955,5168)(5018,5185 ) ( 5081,5202)(5146,5220)(5210,5238 ) ( 5274,5256)(5337,5274)(5398,5291 ) ( 5456,5308)(5511,5324)(5563,5340 ) ( 5609,5353)(5651,5366)(5688,5377 ) ( 5719,5386)(5744,5394)(5765,5400 ) ( 5780,5405)(5790,5408)(5797,5410 ) ( 5800,5412)(5802,5412 ) ( 5802,5412)(5801,5410)(5799,5407 ) ( 5796,5400)(5791,5389)(5784,5374 ) ( 5774,5353)(5762,5328)(5747,5298 ) ( 5731,5263)(5713,5224)(5692,5181 ) ( 5671,5136)(5649,5088)(5626,5039 ) ( 5603,4989)(5580,4939)(5558,4889 ) ( 5536,4840)(5515,4792)(5494,4746 ) ( 5475,4701)(5457,4658)(5440,4617 ) ( 5424,4577)(5408,4539)(5394,4502 ) ( 5381,4467)(5369,4433)(5357,4399 ) ( 5346,4367)(5336,4335)(5327,4304 ) ( 5318,4273)(5310,4243)(5302,4212 ) ( 5294,4180)(5287,4147)(5280,4114 ) ( 5273,4081)(5267,4048)(5262,4013 ) ( 5256,3977)(5252,3941)(5247,3902 ) ( 5243,3863)(5239,3821)(5235,3778 ) ( 5231,3733)(5228,3686)(5225,3637 ) ( 5222,3587)(5219,3535)(5217,3482 ) ( 5214,3429)(5212,3376)(5210,3325 ) ( 5209,3274)(5207,3227)(5206,3183 ) ( 5205,3144)(5204,3109)(5203,3080 ) ( 5203,3057)(5203,3039)(5202,3026 ) ( 5202,3018)(5202,3014)(5202,3012 ) ( 2202,5412)(2201,5410)(2199,5407 ) ( 2196,5400)(2191,5389)(2184,5374 ) ( 2174,5353)(2162,5328)(2147,5298 ) ( 2131,5263)(2113,5224)(2092,5181 ) ( 2071,5136)(2049,5088)(2026,5039 ) ( 2003,4989)(1980,4939)(1958,4889 ) ( 1936,4840)(1915,4792)(1894,4746 ) ( 1875,4701)(1857,4658)(1840,4617 ) ( 1824,4577)(1808,4539)(1794,4502 ) ( 1781,4467)(1769,4433)(1757,4399 ) ( 1746,4367)(1736,4335)(1727,4304 ) ( 1718,4273)(1710,4243)(1702,4212 ) ( 1694,4180)(1687,4147)(1680,4114 ) ( 1673,4081)(1667,4048)(1662,4013 ) ( 1656,3977)(1652,3941)(1647,3902 ) ( 1643,3863)(1639,3821)(1635,3778 ) ( 1631,3733)(1628,3686)(1625,3637 ) ( 1622,3587)(1619,3535)(1617,3482 ) ( 1614,3429)(1612,3376)(1610,3325 ) ( 1609,3274)(1607,3227)(1606,3183 ) ( 1605,3144)(1604,3109)(1603,3080 ) ( 1603,3057)(1603,3039)(1602,3026 ) ( 1602,3018)(1602,3014)(1602,3012 ) ( 1602,3012)(1604,3012)(1607,3010 ) ( 1614,3008)(1624,3005)(1639,3000 ) ( 1660,2994)(1685,2986)(1716,2977 ) ( 1753,2966)(1795,2953)(1841,2940 ) ( 1893,2924)(1948,2908)(2006,2891 ) ( 2067,2874)(2130,2856)(2194,2838 ) ( 2258,2820)(2323,2802)(2386,2785 ) ( 2449,2768)(2511,2752)(2571,2737 ) ( 2630,2723)(2687,2709)(2742,2697 ) ( 2795,2685)(2846,2674)(2895,2665 ) ( 2943,2656)(2990,2648)(3035,2640 ) ( 3078,2634)(3121,2629)(3163,2624 ) ( 3204,2620)(3244,2617)(3284,2615 ) ( 3323,2613)(3363,2612)(3402,2612 ) ( 3441,2612)(3481,2613)(3520,2615 ) ( 3560,2617)(3600,2620)(3641,2624 ) ( 3683,2629)(3726,2634)(3769,2640 ) ( 3814,2648)(3861,2656)(3909,2665 ) ( 3958,2674)(4009,2685)(4062,2697 ) ( 4117,2709)(4174,2723)(4233,2737 ) ( 4293,2752)(4355,2768)(4418,2785 ) ( 4481,2802)(4546,2820)(4610,2838 ) ( 4674,2856)(4737,2874)(4798,2891 ) ( 4856,2908)(4911,2924)(4963,2940 ) ( 5009,2953)(5051,2966)(5088,2977 ) ( 5119,2986)(5144,2994)(5165,3000 ) ( 5180,3005)(5190,3008)(5197,3010 ) ( 5200,3012)(5202,3012 ) ( 5202,3012)(5202,3010)(5202,3006 ) ( 5202,2998)(5203,2985)(5203,2967 ) ( 5203,2944)(5204,2915)(5205,2880 ) ( 5206,2841)(5207,2797)(5209,2750 ) ( 5210,2699)(5212,2648)(5214,2595 ) ( 5217,2542)(5219,2489)(5222,2437 ) ( 5225,2387)(5228,2338)(5231,2291 ) ( 5235,2246)(5239,2203)(5243,2161 ) ( 5247,2122)(5252,2083)(5256,2047 ) ( 5262,2011)(5267,1976)(5273,1943 ) ( 5280,1910)(5287,1877)(5294,1844 ) ( 5302,1812)(5310,1781)(5318,1751 ) ( 5327,1720)(5336,1689)(5346,1657 ) ( 5357,1625)(5369,1591)(5381,1557 ) ( 5394,1522)(5408,1485)(5424,1447 ) ( 5440,1407)(5457,1366)(5475,1323 ) ( 5494,1278)(5515,1232)(5536,1184 ) ( 5558,1135)(5580,1085)(5603,1035 ) ( 5626,985)(5649,936)(5671,888 ) ( 5692,843)(5713,800)(5731,761 ) ( 5747,726)(5762,696)(5774,671 ) ( 5784,650)(5791,635)(5796,624 ) ( 5799,617)(5801,614)(5802,612 ) ( 1602,3012)(1602,3010)(1602,3006 ) ( 1602,2998)(1603,2985)(1603,2967 ) ( 1603,2944)(1604,2915)(1605,2880 ) ( 1606,2841)(1607,2797)(1609,2750 ) ( 1610,2699)(1612,2648)(1614,2595 ) ( 1617,2542)(1619,2489)(1622,2437 ) ( 1625,2387)(1628,2338)(1631,2291 ) ( 1635,2246)(1639,2203)(1643,2161 ) ( 1647,2122)(1652,2083)(1656,2047 ) ( 1662,2011)(1667,1976)(1673,1943 ) ( 1680,1910)(1687,1877)(1694,1844 ) ( 1702,1812)(1710,1781)(1718,1751 ) ( 1727,1720)(1736,1689)(1746,1657 ) ( 1757,1625)(1769,1591)(1781,1557 ) ( 1794,1522)(1808,1485)(1824,1447 ) ( 1840,1407)(1857,1366)(1875,1323 ) ( 1894,1278)(1915,1232)(1936,1184 ) ( 1958,1135)(1980,1085)(2003,1035 ) ( 2026,985)(2049,936)(2071,888 ) ( 2092,843)(2113,800)(2131,761 ) ( 2147,726)(2162,696)(2174,671 ) ( 2184,650)(2191,635)(2196,624 ) ( 2199,617)(2201,614)(2202,612 ) ( 2202,612)(2204,612)(2207,610 ) ( 2214,608)(2224,605)(2239,600 ) ( 2260,594)(2285,586)(2316,577 ) ( 2353,566)(2395,553)(2441,540 ) ( 2493,524)(2548,508)(2606,491 ) ( 2667,474)(2730,456)(2794,438 ) ( 2858,420)(2923,402)(2986,385 ) ( 3049,368)(3111,352)(3171,337 ) ( 3230,323)(3287,309)(3342,297 ) ( 3395,285)(3446,274)(3495,265 ) ( 3543,256)(3590,248)(3635,240 ) ( 3678,234)(3721,229)(3763,224 ) ( 3804,220)(3844,217)(3884,215 ) ( 3923,213)(3963,212)(4002,212 ) ( 4041,212)(4081,213)(4120,215 ) ( 4160,217)(4200,220)(4241,224 ) ( 4283,229)(4326,234)(4369,240 ) ( 4414,248)(4461,256)(4509,265 ) ( 4558,274)(4609,285)(4662,297 ) ( 4717,309)(4774,323)(4833,337 ) ( 4893,352)(4955,368)(5018,385 ) ( 5081,402)(5146,420)(5210,438 ) ( 5274,456)(5337,474)(5398,491 ) ( 5456,508)(5511,524)(5563,540 ) ( 5609,553)(5651,566)(5688,577 ) ( 5719,586)(5744,594)(5765,600 ) ( 5780,605)(5790,608)(5797,610 ) ( 5800,612)(5802,612 ) ( 4300,2887)(0,0)[lb]@xmath487 ( 4300,4682)(0,0)[lb ] @xmath488 ( 6500,3787)(0,0)[lb]@xmath489 ( 5650,2450)(0,0)[lb]@xmath490 ( 4607,2520)(0,0)[lb]@xmath475 this map linearizes the intersection , @xmath491 , and we have @xmath492 and @xmath493 . in order to linearize the entire lagrangian @xmath488 we remark that @xmath494 .
so we can write @xmath495 as graph of a map @xmath496 with @xmath497 , @xmath498 , and by the previous construction @xmath499 .
( 7388,5439)(0,-10 ) ( 704.214,2881.071)(582.000,2862.000)(698.811,2821.315 ) ( 582,2862)(6587,2319 ) ( 6464.786,2299.929)(6587.000,2319.000)(6470.189,2359.685 ) ( 1872,4677)(387,4812)(12,642 ) ( 6612,12)(6972,4212)(5367,4362 ) ( 1857,4677)(5352,4362 ) ( 1602,4812)(5202,4812)(5202,912 ) ( 1602,912)(1602,4812 ) ( 1614,2773)(1614,4833)(5224,4823)(5204,2443 ) ( 5194,2463)(5194,903)(1634,923)(1604,2783 ) ( 2202,5412)(2204,5412)(2207,5412 ) ( 2214,5412)(2224,5412)(2239,5412 ) ( 2260,5412)(2285,5412)(2316,5412 ) ( 2353,5412)(2395,5412)(2441,5412 ) ( 2493,5412)(2548,5412)(2606,5412 ) ( 2667,5412)(2730,5412)(2794,5412 ) ( 2858,5412)(2923,5412)(2986,5412 ) ( 3049,5412)(3111,5412)(3171,5412 ) ( 3230,5412)(3287,5412)(3342,5412 ) ( 3395,5412)(3446,5412)(3495,5412 ) ( 3543,5412)(3590,5412)(3635,5412 ) ( 3678,5412)(3721,5412)(3763,5412 ) ( 3804,5412)(3844,5412)(3884,5412 ) ( 3923,5412)(3963,5412)(4002,5412 ) ( 4041,5412)(4081,5412)(4120,5412 ) ( 4160,5412)(4200,5412)(4241,5412 ) ( 4283,5412)(4326,5412)(4369,5412 ) ( 4414,5412)(4461,5412)(4509,5412 ) ( 4558,5412)(4609,5412)(4662,5412 ) ( 4717,5412)(4774,5412)(4833,5412 ) ( 4893,5412)(4955,5412)(5018,5412 ) ( 5081,5412)(5146,5412)(5210,5412 ) ( 5274,5412)(5337,5412)(5398,5412 ) ( 5456,5412)(5511,5412)(5563,5412 ) ( 5609,5412)(5651,5412)(5688,5412 ) ( 5719,5412)(5744,5412)(5765,5412 ) ( 5780,5412)(5790,5412)(5797,5412 ) ( 5800,5412)(5802,5412 ) ( 5802,5412)(5801,5410)(5799,5407 ) ( 5796,5400)(5791,5389)(5784,5373 ) ( 5774,5352)(5762,5326)(5747,5295 ) ( 5731,5260)(5713,5220)(5692,5176 ) ( 5671,5129)(5649,5079)(5626,5028 ) ( 5603,4976)(5580,4924)(5558,4872 ) ( 5536,4820)(5515,4770)(5494,4721 ) ( 5475,4673)(5457,4627)(5440,4582 ) ( 5424,4538)(5408,4496)(5394,4455 ) ( 5381,4415)(5369,4376)(5357,4338 ) ( 5346,4300)(5336,4263)(5327,4225 ) ( 5318,4188)(5310,4150)(5302,4112 ) ( 5295,4077)(5289,4042)(5283,4007 ) ( 5277,3971)(5272,3934)(5267,3895 ) ( 5262,3856)(5257,3815)(5253,3773 ) ( 5249,3729)(5245,3684)(5242,3636 ) ( 5238,3586)(5235,3534)(5232,3480 ) ( 5229,3424)(5227,3366)(5224,3305 ) ( 5222,3243)(5219,3179)(5217,3114 ) ( 5215,3049)(5213,2983)(5212,2918 ) ( 5210,2854)(5209,2792)(5207,2733 ) ( 5206,2677)(5205,2627)(5204,2581 ) ( 5204,2540)(5203,2506)(5203,2477 ) ( 5203,2454)(5202,2437)(5202,2425 ) ( 5202,2418)(5202,2414)(5202,2412 ) ( 2202,5412)(2201,5410)(2199,5407 ) ( 2196,5400)(2191,5389)(2184,5373 ) ( 2174,5353)(2162,5327)(2147,5297 ) ( 2131,5261)(2113,5222)(2092,5179 ) ( 2071,5132)(2049,5084)(2026,5034 ) ( 2003,4983)(1980,4931)(1958,4880 ) ( 1936,4830)(1915,4781)(1894,4733 ) ( 1875,4687)(1857,4642)(1840,4599 ) ( 1824,4558)(1808,4518)(1794,4479 ) ( 1781,4441)(1769,4404)(1757,4369 ) ( 1746,4334)(1736,4299)(1727,4265 ) ( 1718,4230)(1710,4196)(1702,4162 ) ( 1695,4129)(1688,4096)(1682,4063 ) ( 1676,4029)(1670,3994)(1665,3959 ) ( 1660,3922)(1656,3885)(1651,3846 ) ( 1647,3805)(1643,3763)(1640,3719 ) ( 1636,3673)(1633,3625)(1630,3575 ) ( 1627,3523)(1624,3469)(1622,3414 ) ( 1619,3357)(1617,3299)(1615,3241 ) ( 1613,3183)(1611,3125)(1610,3069 ) ( 1608,3015)(1607,2964)(1606,2916 ) ( 1605,2873)(1604,2835)(1603,2802 ) ( 1603,2775)(1603,2753)(1602,2737 ) ( 1602,2725)(1602,2718)(1602,2714)(1602,2712 ) ( 5202,2412)(5202,2410)(5202,2405 ) ( 5203,2396)(5203,2382)(5204,2364 ) ( 5205,2340)(5207,2312)(5208,2280 ) ( 5211,2245)(5213,2209)(5216,2171 ) ( 5219,2133)(5223,2096)(5227,2060 ) ( 5231,2024)(5236,1990)(5241,1957 ) ( 5246,1926)(5252,1895)(5259,1864 ) ( 5266,1834)(5274,1804)(5282,1774 ) ( 5292,1743)(5302,1712)(5310,1687 ) ( 5319,1662)(5329,1637)(5339,1610 ) ( 5350,1582)(5361,1553)(5374,1523 ) ( 5387,1491)(5402,1458)(5417,1422 ) ( 5434,1385)(5452,1345)(5471,1304 ) ( 5491,1260)(5512,1214)(5535,1167 ) ( 5558,1118)(5582,1068)(5606,1017 ) ( 5630,966)(5654,916)(5677,868 ) ( 5700,822)(5720,779)(5739,741 ) ( 5755,707)(5769,679)(5780,656 ) ( 5789,639)(5795,626)(5799,618 ) ( 5801,614)(5802,612 ) ( 1602,3012)(1602,3010)(1602,3006 ) ( 1602,2998)(1603,2985)(1603,2967 ) ( 1603,2944)(1604,2915)(1605,2880 ) ( 1606,2841)(1607,2797)(1609,2750 ) ( 1610,2699)(1612,2648)(1614,2595 ) ( 1617,2542)(1619,2489)(1622,2437 ) ( 1625,2387)(1628,2338)(1631,2291 ) ( 1635,2246)(1639,2203)(1643,2161 ) ( 1647,2122)(1652,2083)(1656,2047 ) ( 1662,2011)(1667,1976)(1673,1943 ) ( 1680,1910)(1687,1877)(1694,1844 ) ( 1702,1812)(1710,1781)(1718,1751 ) ( 1727,1720)(1736,1689)(1746,1657 ) ( 1757,1625)(1769,1591)(1781,1557 ) ( 1794,1522)(1808,1485)(1824,1447 ) ( 1840,1407)(1857,1366)(1875,1323 ) ( 1894,1278)(1915,1232)(1936,1184 ) ( 1958,1135)(1980,1085)(2003,1035 ) ( 2026,985)(2049,936)(2071,888 ) ( 2092,843)(2113,800)(2131,761 ) ( 2147,726)(2162,696)(2174,671 ) ( 2184,650)(2191,635)(2196,624 ) ( 2199,617)(2201,614)(2202,612 ) ( 2202,612)(2204,612)(2207,612 ) ( 2214,612)(2224,612)(2239,612 ) ( 2260,612)(2285,612)(2316,612 ) ( 2353,612)(2395,612)(2441,612 ) ( 2493,612)(2548,612)(2606,612 ) ( 2667,612)(2730,612)(2794,612 ) ( 2858,612)(2923,612)(2986,612 ) ( 3049,612)(3111,612)(3171,612 ) ( 3230,612)(3287,612)(3342,612 ) ( 3395,612)(3446,612)(3495,612 ) ( 3543,612)(3590,612)(3635,612 ) ( 3678,612)(3721,612)(3763,612 ) ( 3804,612)(3844,612)(3884,612 ) ( 3923,612)(3963,612)(4002,612 ) ( 4041,612)(4081,612)(4120,612 ) ( 4160,612)(4200,612)(4241,612 ) ( 4283,612)(4326,612)(4369,612 ) ( 4414,612)(4461,612)(4509,612 ) ( 4558,612)(4609,612)(4662,612 ) ( 4717,612)(4774,612)(4833,612 ) ( 4893,612)(4955,612)(5018,612 ) ( 5081,612)(5146,612)(5210,612 ) ( 5274,612)(5337,612)(5398,612 ) ( 5456,612)(5511,612)(5563,612 ) ( 5609,612)(5651,612)(5688,612 ) ( 5719,612)(5744,612)(5765,612 ) ( 5780,612)(5790,612)(5797,612 ) ( 5800,612)(5802,612 ) ( 6500,3787)(0,0)[lb]@xmath489 ( 2000,2722)(0,0)[lb]@xmath500
( 4100,5000)(0,0)[lb]@xmath501 ( 2909,4300)(0,0)[lb]@xmath502 finally we define the entire linearization @xmath503 by @xmath504 for @xmath505 in the splitting ( [ split x ] ) . now @xmath506 is quadratic and linearized @xmath450 by construction .
explicitly , we have @xmath507 the construction moreover ensures that @xmath458 linearizes @xmath508 , that is @xmath509 , since @xmath510 is mapped to @xmath511 . to construct @xmath512
compatible with @xmath458 note that for @xmath513 we have a splitting @xmath514 where @xmath515 only depend on @xmath516 . with this we can see in ( [ q ] ) that indeed @xmath517 is independent of @xmath518 . we then simply define @xmath519 .
moreover , a graph construction as above provides a map @xmath520 that is quadratic and linearizes @xmath457 .
now the two families @xmath463 and @xmath521 can easily be interpolated by the smooth family @xmath522 of quadratic maps . with these quadratic corrections
we can now define the exponential map @xmath404 by @xmath523 for @xmath524 , where @xmath525 note that we have the usual properties of an exponential map , @xmath526 to define @xmath404 on @xmath406 the @xmath392 should be chosen such that @xmath527 and @xmath528 are sufficiently small for the quadratic corrections in lemma [ qq ] to be defined .
are noncompact , in particular the interior of a compact manifold with boundary and corners as in @xcite , then the choice of @xmath392 also ensures that the exponential map at @xmath372 resp .
@xmath433 is well defined .
this is always possible with a uniform @xmath392 since the images of @xmath372 and @xmath433 are contained in compact subsets . ]
lemma [ sobolev ] below ensures that we can pick a uniform @xmath392 for all @xmath274 .
now solutions @xmath395 in a neighborhood of @xmath109 correspond to zeroes of the map @xmath529 given by @xmath530 here @xmath531 denotes the parallel transport @xmath532 along the path @xmath533 . for @xmath534 this parallel transport on @xmath535
can simply use the levi - civita connection .
[ connection ] in the definition of @xmath536 we however use a hermitian connection @xmath537 on the tangent bundle @xmath538 that leaves @xmath539 invariant .
this can be done by the same construction as in ( * ? ? ?
* proposition 3.1.1 ) , which brings the linearized operator into simple form .
next , we introduce projections related to the various lagrangians : @xmath540 @xmath541 are linear operators , given by pointwise orthogonal projection onto the subspaces @xmath542 resp . @xmath543 .
the following lemma contains the estimates resulting from the transversality assumption . 1 . for every
@xmath544 @xmath545 2 . for every @xmath464 and @xmath546 @xmath547 3 . for every @xmath548 @xmath549 and
the same holds with @xmath550 replaced by @xmath551 or @xmath552 for any @xmath553 .
moreover , @xmath554 we fix metrics on each @xmath555 and use the induced split metrics on both @xmath198 and @xmath556 . towards ( a ) note that we evidently have @xmath557 for @xmath558 .
moreover , we can estimate with some constant @xmath479 @xmath559 since the lagrangian @xmath450 intersects @xmath508 transversally .
( in more detail this follows from the linear theory below ; if @xmath560 does not lie in an exponential neighbourhood of @xmath561 , then both sides of the inequality are bounded away from zero , hence the quotient attains a positive minimum on the complement of the exponential neighbourhood .
are noncompact .
indeed , we only need to consider the case of @xmath562 and simply note that @xmath457 as well as the projection @xmath563 are compact subsets .
hence we can bound @xmath564 with a finite constant @xmath565 and obtain @xmath566 . ] ) this proves ( a ) since @xmath567 is bounded by @xmath568 . to approach ( b ) note moreover that @xmath561 injects to @xmath569 .
so at every point of @xmath561 we have a decomposition @xmath570 , where we can change the first factor to @xmath571 . on the other hand ,
the transverse intersection implies @xmath572 so we obtain a splitting @xmath573 this means that the product of the three orthogonal projections onto the factors defines an isomorphism .
the norm of this isomorphism is bounded at each @xmath574 , so for every @xmath575 we have @xmath576 with a uniform constant @xmath479 as claimed in ( b ) .
( here the projection onto @xmath577 is given by @xmath578 .
) moreover , the splitting commutes with @xmath579 via the canonical projection on the left hand side , and on the right hand side the identity on @xmath580 combined with a bounded map @xmath581 .
this implies that @xmath582 with another uniform constant @xmath479 .
this proves ( b ) .
for @xmath548 we can then apply the pointwise estimates to @xmath583 and integrate over @xmath464 to obtain for any @xmath553 including @xmath584 @xmath585 in order to prove the @xmath550- and @xmath551-estimates we also apply the pointwise estimates to @xmath586 , @xmath587 here we will need the inequalities @xmath588 the first inequality ( and similarly the others ) can be seen by writing @xmath589 in a local orthonormal frame given by @xmath590 and @xmath591 . writing @xmath592 we have @xmath593 note here that @xmath594 and @xmath595 are uniformly bounded .
putting things together we obtain the first estimate in ( c ) with an extra @xmath596 or @xmath597 on the right hand side , for which we can use ( [ previous ] ) .
for the last estimate in ( c ) we obtain @xmath598 this finishes the proof of ( c ) .
[ sobolev ] ( uniform sobolev estimate ) there is a constant @xmath599 such that for all @xmath600 $ ] and @xmath601 @xmath602,h^1({\mathbb{r } } ) ) } + \| \hat\xi \|_{{\mathcal{c}}^0([0,\delta],h^1({\mathbb{r } } ) ) } \\ & \leq c_s\bigl ( \|\xi\|_{h^2_{1,\delta } } + \| ( \xi_{02}- \xi'_{02})|_{t=0 } \|_{h^1({\mathbb{r } } ) } + \vert ( \xi_1-\xi_1')|_{t=0 } \vert_{h^1({\mathbb{r } } ) } + \vert \pi_{0211}^\perp\hx|_{t=\delta } \vert_{h^1({\mathbb{r } } ) } \bigr).\end{aligned}\ ] ] in particular , for all @xmath603 including @xmath584 and for @xmath604 satisfying the boundary conditions ( [ ll ] ) , @xmath605 ) } + \| \hat\xi \|_{l^p({\mathbb{r}}\times[0,\delta ] ) } \leq c_s \|\xi\|_{h^2_{1,\delta } } .\ ] ] the @xmath606- and @xmath552-estimates will follow from the continuous embeddings @xmath607 and @xmath608 for @xmath609 .
so it suffices to suppose by contradiction that there are sequences @xmath610 and @xmath611 with @xmath612,h^1({\mathbb{r } } ) ) } + \| \hat\xi^\nu \|_{{\mathcal{c}}^0([0,\delta^\nu],h^1({\mathbb{r}}))}=1 $ ] but @xmath613 . by the standard sobolev embedding @xmath614\times{\mathbb{r } } ) \subset h^1([0,1],x)\hookrightarrow{\mathcal{c}}^0([0,1],x ) \qquad\text{with}\ ; x = h^1({\mathbb{r}})\ ] ] this implies @xmath615,h^1({\mathbb{r } } ) ) } \to 0 $ ] , and so @xmath616 we can moreover integrate for all @xmath617 $ ] to obtain @xmath618)}^2 \to 0 .\ ] ] using lemma [ lltrans ] we then obtain @xmath619 with uniform constants @xmath620 by , , , and a bound on the operator norm of @xmath12 . now combining @xmath621 with ( [ txi1 ] )
proves @xmath622,h^1({\mathbb{r}}))}\to 0 $ ] in contradiction to the assumption and the previously established fact that @xmath615,h^1({\mathbb{r } } ) ) } \to 0 $ ] . the solution @xmath109 of the @xmath118-equation corresponds to @xmath623 , which is an almost zero of @xmath624 .
this and a quadratic estimate for @xmath625 near @xmath118 is the content of the next lemma . for later purposes we also compare @xmath626 with the linearized operator @xmath627 of @xmath628 at @xmath412 . to state the comparison we will need the pointwise linear operator @xmath629 it satisfies @xmath630 , and since @xmath404 maps @xmath429 to the space of maps satisfying the boundary conditions in ( [ vvvv ] ) ,
the linearization @xmath631 maps @xmath429 to the space of sections @xmath632 over @xmath633 , that satisfy the linearized boundary conditions @xmath634 the linearized operator @xmath635 acts on this space of sections and is given by @xmath636 with the connection @xmath537 introduced on page .
in this notation we have @xmath637 .
to estimate @xmath642 we recall that @xmath372 is pseudoholomorphic and @xmath433 is constant in @xmath263 , so @xmath643 here @xmath644 converges exponentially as @xmath177 , and so does @xmath645 , where @xmath646 from has bounded differential .
this shows that the above constant @xmath647 is indeed finite . for the third estimate we differentiate as in @xcite the identity @xmath648 to obtain @xmath649 where the estimate for the right hand side @xmath650 is part of the estimates below
the first component of @xmath624 is independent of @xmath298 , so the quadratic estimates for it simply follow from the continuous differentiability of @xmath624 . for the second component
we follow the argument in ( * ? ? ?
* prop.3.5.3 . ) to obtain a uniform constant for all @xmath600 $ ] .
we need to consider @xmath651 where @xmath652 is the exponential map with quadratic correction defined in .
note that our parallel transport @xmath653 is defined with respect to the path @xmath654 and the hermitian connection @xmath537 on @xmath538 that leaves @xmath539 invariant .
since @xmath655 and @xmath656 , the same path can be used in the definition of @xmath657 instead of the geodesic . now let @xmath658 with @xmath659 .
then by lemma [ sobolev ] @xmath660 with a uniform constant @xmath599 thus a uniform constant @xmath661 that only depends on @xmath405 . in the following , all constants will be uniform in the sense that they only depend on @xmath661 and hence @xmath405 .
next , we consider @xmath662 note that @xmath663 and that @xmath664 since the covariant derivative exactly uses the parallel transport @xmath665 .
moreover , these maps are linear in @xmath666 and @xmath667 , and they depend smoothly on @xmath668 . so given @xmath405 and thus @xmath669 we have linear bounds @xmath670 . with these preparations we calculate from ( [ dfd ] ) , using the notation of ( * ? ? ?
* prop.3.5.3 . ) , @xmath671 we then use the uniform bounds we here need @xmath672 bounded only in a compact neighbourhood of the image of @xmath433 . ] on @xmath673 , @xmath674 , @xmath675 , @xmath676 , @xmath677 , and the estimates @xmath678 with a uniform constant @xmath679 to obtain with a further uniform constant @xmath680 @xmath681 so far these pointwise estimates were standard calculations .
now we have to check that they actually lead to uniform bounds in the @xmath298-dependent norms .
the zeroth order part of the @xmath682-norm over @xmath683 $ ] can be estimated with the help of lemma [ sobolev ] by @xmath684 for the first order part of the @xmath682-norm one differentiates the above identity and uses further bounds on @xmath685 and @xmath686 to find a pointwise bound @xmath687 then we again use lemma [ sobolev ] and @xmath688 to obtain with a final uniform constant @xmath689 @xmath690 theorem [ solving ] now follows from the implicit function theorem ( * ? ? ?
* a.3.4 ) if we can establish surjectivity and a uniform bound on the right inverse for the linearized operator @xmath691 here @xmath692 and @xmath693 are the linearized operators of @xmath694 at @xmath372 ( which is pseudoholomorphic ) and of @xmath695 at @xmath696 ( which satisfies @xmath697 ) respectively .
( see ( * ? ? ?
* prop.3.1.1 . ) for an explicit calculation of the linearized operators , and note that we identify @xmath698,u^*tm)$ ] with sections of @xmath699 by @xmath700 . )
we can identify the cokernel of @xmath701 with @xmath702 . by elliptic regularity any element in this cokernel can be represented by the @xmath703-inner product @xmath704 with a smooth section @xmath705 .
partial integration then shows that @xmath640 satisfies the boundary conditions ( [ ll ] ) and lies in the kernel of the formal adjoint operator , @xmath706 .
note that @xmath707 is given by @xmath708 plus lower order terms .
so @xmath707 has the same analytic properties as @xmath701 , and we will prove the surjectivity of @xmath701 by establishing injectivity for @xmath707 . by our assumptions on the index and regularity of @xmath709
we know that the operator @xmath710 on the space of sections in @xmath711 with boundary conditions at @xmath267 in @xmath712 ( where @xmath713 is the projection at @xmath266 ) is surjective and has a one dimensional kernel @xmath714 .
this is not a subspace of @xmath429 , but we will fix a complement for every @xmath274 in the following sense , @xmath715 here we used the @xmath703-inner product on @xmath716 , u_{02}^*t(m_0\times m_2))$ ] . combining the uniform linear estimates
lemma [ ddelta ] and lemma [ d02 ] we can choose @xmath717 such that for all @xmath718 and @xmath719 @xmath720 ) } \\ & \geq \tfrac 12 c_1 \| \xi \|_{\gamma_{1,\delta } } - c_2^{-1}\sqrt{\delta } \vert \nabla_t\hx \vert_{h^1({\mathbb{r}}\times[0,\delta ] ) } \;\geq\ ; \tfrac 14 c_1 \| \xi \|_{\gamma_{1,\delta } } , \end{aligned}\ ] ] and similarly for all @xmath721 @xmath722 the first estimate shows that @xmath707 is injective and hence @xmath701 is surjective .
the second estimate shows that its right inverse is uniformly bounded .
it remains to check that @xmath701 stays surjective when restricted to @xmath408 .
this follows from the fact that both @xmath692 with boundary conditions in @xmath723 and @xmath724 with boundary conditions ( [ ll ] ) are surjective and have the same index @xmath117 by lemma [ index ] and the identification @xmath725 .
so @xmath701 has a @xmath117-dimensional kernel , which is transversal to @xmath408 by the last estimate , and hence @xmath726 must be surjective .
this finishes the proof of theorem [ solving ] . here
@xmath392 is fixed such that the exponential map @xmath404 is defined on @xmath406 and such that lemma [ quadratic ] holds .
we choose @xmath729 such that theorem [ solving ] applies .
then let @xmath396 be the solution constructed from @xmath730 and consider a shifted @xmath118-solution @xmath731 $ ] .
then @xmath732 satisfies @xmath733 , @xmath734 , and the orthogonality condition to @xmath735 .
hence @xmath736 $ ] , so @xmath728)=[v_u]$ ] is well defined .
the injectivity of @xmath737 follows from the fact that @xmath387 consists of isolated points , so the @xmath606-distance @xmath738,[u'])>\delta_0 $ ] is bounded below by some @xmath739 for all @xmath740\neq[u']$ ] . on the other hand , @xmath741,{\mathcal{t}}_\delta([u])\leq c_0 c_s ( 1+c_q ) \delta^{\frac 14}$ ] by ( [ sqrtd ] ) , ( [ quest ] ) , and lemma [ sobolev ] .
so if we had @xmath728)={\mathcal{t}}_\delta([u'])$ ] then @xmath738,[u'])\leq d_{{\mathcal{c}}^0}([\bu],[\bu ' ] ) \leq 2c_0 c_s ( 1+c_q ) \delta^{\frac 14}$ ] .
this implies @xmath740=[u']$ ] whenever @xmath742 , where we choose @xmath743 . in this section
we establish the uniform linear and nonlinear estimates that are used in sections [ ift ] and [ cpt ] .
we will work in the setup of section [ ift ] and fix a solution @xmath345 .
for convenience we denote the target spaces by @xmath744 and @xmath745 and the symplectic structures by @xmath746 and @xmath747 respectively .
the nonlinear equation for @xmath361 , @xmath748\to m_{02}$ ] , @xmath749\to m_{0211}$ ] is @xmath750 we will need uniform estimates for the nonlinear operator @xmath751 on @xmath752 and the linearized operator @xmath701 . for that purpose
we use the levi - civita connection on @xmath753 and @xmath754 respectively to identify @xmath755 for every @xmath756 . with this
we decompose @xmath757 as @xmath758 we denote the pullback almost complex structure on @xmath407 under @xmath759 by @xmath760 for @xmath761 . with this
we can express @xmath762 in terms of the nonlinear operator on @xmath407 , @xmath763 note that @xmath764 is the usual almost complex structure at @xmath341 , so we can express the linearized operator as @xmath765 1 .
there is a constant @xmath647 such that for all @xmath600 $ ] and @xmath766 @xmath767 2 .
there is a constant @xmath392 and for every @xmath768 there is a constant @xmath647 such that for all @xmath600 $ ] and @xmath769 with @xmath770 , @xmath771 @xmath772 3 .
there is a constant @xmath773 such that for all @xmath600 $ ] and @xmath719 @xmath774 and the same holds with @xmath701 replaced by @xmath707 .
we prove ( a ) in general for @xmath775 and @xmath776 with a lagrangian section @xmath777 over a path @xmath778 .
these expressions vanish if @xmath120 is totally geodesic . to estimate them in general
we pick a smooth family of orthonormal frames @xmath779 , then @xmath780 with @xmath781 . by the orthonormality we have @xmath782 , and using @xmath783 as a trivialization for the definition of sobolev norms on @xmath699 we obtain @xmath784 .
we now use the identities @xmath785 to obtain @xmath786 where the constant @xmath479 only depends on @xmath787 ( that is on @xmath778 ) up to third derivatives . here
we used partial integration @xmath788 to prove ( c )
we can replace @xmath701 by @xmath789 since the difference of the operators is bounded in the different components and norms by @xmath790 ) } + \bigl\| \nabla_{\hx}\hat j(\bu ) j(\bu ) \pd_s \bu \bigr\|_{h^0({\mathbb{r}}\times[0,\delta ] ) } \leq c \bigl\| \xi \bigr\|_{h^0_{1,\delta } } , \nonumber\\ & \bigl\| \nabla_{\xi_{02}}j_{02}(u_{02})\pd_t u_{02 } \bigr\|_{l^4({\mathbb{r}}\times[0,1 ] ) } \leq c \bigl\| \nabla_{\xi_{02}}j_{02}(u_{02})\pd_t u_{02 } \bigr\|_{h^1({\mathbb{r}}\times[0,1 ] ) } \leq c \bigl\| \xi \bigr\|_{h^1_{1,\delta } } , \nonumber\\ & \bigl\| \nabla_{\hx}\hat j(\bu ) j(\bu ) \pd_s \bu \bigr\|_{h^1({\mathbb{r}}\times[0,\delta ] ) } \leq c
\bigl\| \xi \bigr\|_{h^1_{1,\delta } } , \\ & \bigl\| \nabla_{\hx}\hat j(\bu ) j(\bu ) \pd_s \bu \bigr\|_{l^4({\mathbb{r}}\times[0,\delta ] ) } \leq c \| \nabla\hat j\|_\infty \| \pd_s\bu \|_\infty \|\hx\|_{l^4({\mathbb{r}}\times[0,\delta ] ) } \leq c
\bigl\| \xi \bigr\|_{h^2_{1,\delta } } , \nonumber\end{aligned}\ ] ] where @xmath479 denotes any uniform constant
. the extra terms on the right hand side will fit into the proof and will be recalled for the relevant estimates .
the proof for @xmath707 is completely analogous . to prove ( b ) and ( c ) we may moreover fix convenient metrics on @xmath791 and @xmath792 . in order to obtain the boundary terms involving the symplectic forms , we pick the induced metrics @xmath793 resp .
@xmath794 , noting that this introduces a smooth @xmath263-dependence in the metric on @xmath791 .
we will now use the notation @xmath795 to make partial integration calculations for the nonlinear ( @xmath796 ) and linear ( @xmath797 ) operator at the same time . in the nonlinear case
the almost complex structure @xmath798 is not skew - adjoint since the metric is defined by @xmath799 . in order to obtain this property we work with the @xmath800-metric , which uses the pullback metric @xmath801 under @xmath802 on @xmath791 and @xmath803 on @xmath792 respectively . in the linear case
@xmath797 nothing has happened ; in the nonlinear case we can pick @xmath392 and hence @xmath804 sufficiently small such that @xmath805 is @xmath606-close to the identity , and hence the induced @xmath806-norm is uniformly equivalent to the standard @xmath807-norm . with this in mind
we start by calculating for any @xmath808 with @xmath809 ( unless otherwise specified integrals are over two infinite strips of width @xmath298 and @xmath117 ) @xmath810 @xmath811 where we abbreviated @xmath812 these boundary terms occur on the right hand side of ( c ) and they will be estimated by ( a ) to prove ( b ) .
the boundary term at @xmath266 vanishes by the diagonal boundary conditions , and the boundary terms at @xmath813 vanish since @xmath814\}}\to 0 $ ] in the @xmath407-norm .
the error term can be estimated by @xmath815 where the highest order term @xmath816 can be absorbed on the right hand side . from now on @xmath479 will denote any uniform constant ( which is allowed to depend on @xmath661 in the nonlinear case @xmath796 ) . in summary
, the estimates for @xmath817 and @xmath818 are @xmath819 this already proves the first estimate in ( b ) .
we can moreover use the identity @xmath820 to obtain @xmath821 in the linear case ( c ) these estimates combined with ( a ) and ( [ lower order ] ) to prove the first estimate : @xmath822 with a uniform constant @xmath823 . in the nonlinear case ( b )
we obtain similarly @xmath824 with a constant @xmath647 that depends on @xmath825 .
the @xmath427-estimate for the linear and nonlinear operators will arise by rescaling from the following basic estimate . here
@xmath826\to m_{0211}$ ] will be given by @xmath827 for any @xmath600 $ ] .
then for every @xmath828,\hat u^*tm_{0211})$ ] @xmath829 ) } \leq c_0 \bigl ( \|\he|_{t=1 } \|_{l^2({\mathbb{r } } ) } + \|\nabla\he\|_{l^2({\mathbb{r}}\times[0,1 ] ) } \bigr ) .\ ] ] this simply follows from the sobolev embedding @xmath830 ) \hookrightarrow l^4({\mathbb{r}}\times[0,1])$ ] and @xmath831)}^2 \leq \int_0 ^ 1 \biggl\| \he(\cdot,1 ) - \int_t^1 \nabla_t\he(\cdot,\tau ) d\tau \biggr\|_{l^2({\mathbb{r}})}^2 d t \leq 2\| \he|_{t=1 } \|_{l^2({\mathbb{r}})}^2 + 2 \|\nabla_t\he\|_{l^2({\mathbb{r}}\times[0,1])}^2 .\ ] ] when applying this to @xmath832 we encounter the following terms : @xmath833)}^2 & = \biggl ( \int_{{\mathbb{r}}\times[0,1 ] } = \delta^{-1}\| \nabla_s\hx \|_{l^4({\mathbb{r}}\times[0,\delta])}^2 , \\
\| \he|_{t=1 } \|_{l^2({\mathbb{r}})}^2 & = \int_{{\mathbb{r } } } |\nabla_s\hx(\delta s,\delta)|^2 ds = \delta^{-1 } \| \nabla_s\hx|_{t=\delta } \|_{l^2({\mathbb{r}})}^2 ,
\\ \|\nabla\he\|_{l^2({\mathbb{r}}\times[0,1])}^2 & = \int_{{\mathbb{r}}\times[0,1 ] } \delta^2 = \|\nabla\nabla_s\hx\|_{l^2({\mathbb{r}}\times[0,\delta])}^2.\end{aligned}\ ] ] putting this together we find that @xmath834 ) } & \leq c_0 \bigl ( \|\nabla_s\hx|_{t=\delta } \|_{l^2({\mathbb{r } } ) } + \|\nabla\nabla_s\hx\|_{h^2({\mathbb{r}}\times[0,\delta ] ) } \bigr ) \leq c_0 \bigl ( \|\hx|_{t=\delta } \|_{h^1({\mathbb{r } } ) } + \|\xi\|_{h^2_{1,\delta } } \bigr),\end{aligned}\ ] ] where the estimate for @xmath835 is already established .
the @xmath427-estimate for @xmath836 follows from the sobolev embedding @xmath830)\hookrightarrow l^4({\mathbb{r}}\times[0,1])$ ] , and for the last component we have @xmath837 ) } \leq \| \nabla_s\hx + \hat j(\sigma\hat\zeta)\nabla_t\hx \|_{l^4({\mathbb{r}}\times[0,\delta ] ) } + \| \nabla_s\hx \|_{l^4({\mathbb{r}}\times[0,\delta])}.\end{aligned}\ ] ] this finishes the proof of the second estimate , where we allow @xmath838 on the right hand side , and the constant in the nonlinear case depends on @xmath771 . in the linear case
the difference to @xmath839 in ( [ lower order ] ) is bounded by the previous estimate . 1 .
there is a constant @xmath392 and for every @xmath768 there is a constant @xmath840 such that for all @xmath600 $ ] and @xmath769 with @xmath809 , @xmath771 we have @xmath841 ) } + \sqrt{\delta } \vert \nabla_t\hx \vert_{h^1({\mathbb{r}}\times[0,\delta ] ) } + \|\pi_{0211}^\perp\hat\xi|_{t=\delta}\|_{h^1({\mathbb{r } } ) } \\ & \qquad\quad + \|\xi_{02}\|_{l^2({\mathbb{r}}\times[0,1 ] ) } + \|(\xi_1'-\xi_1)|_{t=0}\|_{h^1({\mathbb{r } } ) } + \|(\xi_{02}'-\xi_{02})|_{t=0}\|_{h^1({\mathbb{r } } ) } \bigr).\end{aligned}\ ] ] 2 .
there is a constant @xmath842 such that for all @xmath600 $ ] and @xmath719 @xmath843 ) } + \sqrt{\delta } \vert \nabla_t\hx \vert_{h^1({\mathbb{r}}\times[0,\delta ] ) } , \end{aligned}\ ] ] and for all @xmath721 @xmath844 ) } + \sqrt{\delta } \vert \nabla_t\hx \vert_{h^1({\mathbb{r}}\times[0,\delta ] ) } .\end{aligned}\ ] ] the constant @xmath392 in case ( a ) is chosen such that @xmath845 and thus @xmath846 is defined . to prove ( a ) ( and similar for ( b ) ) we assume by contradiction that we have sequences @xmath610 and @xmath847 such that @xmath848 ( in case ( b ) add @xmath849 here ) , but the right hand sides converges to zero . for technical reasons
we assume in addition @xmath850)}\leq 1 $ ] , which we will also disprove ( i.e. we actually prove a stronger estimate with this term on the left hand side ) .
first we integrate for all @xmath851 $ ] @xmath852 ) } \to 0 .\ ] ] next , lemma [ lltrans ] implies @xmath853 in the two cases of ( b ) we use the boundary conditions for @xmath854 here .
in all three cases the hardest step is now to prove that @xmath855 .
here we exploit the assumption that @xmath856)}$ ] is bounded .
this implies a bound on @xmath857 .
now we find a convergent subsequence @xmath858,u_{02}^*tm_{02})$ ] in the weak @xmath550-topology , and at the same time @xmath859 in the @xmath703-norm on every compact set .
( the sobolev embedding @xmath860 is compact for compact domains @xmath861 $ ] with smooth boundary @xmath862 , see e.g. ( * ? ? ?
* theorem 6.3 ) . ) in case ( a ) the limit has to be @xmath863 since @xmath864 ) } \leq \liminf_{\nu\to\infty } \| \xi^\nu_{02 } \|_{l^2({\mathbb{r}}\times[0,1 ] ) } = 0 $ ] .
this also holds in case ( b ) since the limit satisfies with @xmath865 or @xmath866 @xmath867 ) } & \leq \liminf_{\nu\to\infty } \| d \xi^\nu_{02 } \|_{l^2({\mathbb{r}}\times[0,1 ] ) } = 0 , \\
\qquad \| \pi_{02}^\perp\xi^\infty_{02}|_{t=0 } \|_{l^2({\mathbb{r } } ) } & \leq \liminf_{\nu\to\infty } \| \pi_{02}^\perp\xi^\nu_{02}|_{t=0 } \|_{l^2({\mathbb{r } } ) } = 0 .\end{aligned}\ ] ] since @xmath372 is assumed regular , @xmath868 is injective , and in the second part of case ( b ) we have in addition @xmath869 .
so in all three cases we obtain @xmath870)}\to 0 \qquad \text{for all}\ ; t>0 .\ ] ] the same holds for @xmath871 since we can apply lemma [ lltrans ] on the interval @xmath872 for any @xmath873 $ ] to obtain @xmath874 this together with the fact that @xmath875 as @xmath876 implies that @xmath855 and hence @xmath877 by ( [ l2h1 ] ) . from this
we will move on to prove that @xmath878 ) } \to 0 .\ ] ] for that purpose we denote by @xmath879 any of the three operators @xmath880 in case ( a ) and @xmath881 or @xmath692 in case ( b ) .
then we use the fact that in all three cases the operator @xmath882 is fredholm on the space of sections @xmath705 that satisfy the boundary conditions @xmath883 , see e.g. ( * ? ? ?
* theorem 20.1.2 ) for compact domains .
the corresponding estimates add up to @xmath884 ) } & \leq c \bigl ( \vert d \xi_{02}^\nu \vert_{h^1({\mathbb{r}}\times[0,1 ] ) } + \|\pi_{02}^\perp \xi_{02}^\nu|_{t=0 } \|_{h^1({\mathbb{r } } ) } + \| \xi_{02}^\nu \|_{h^0({\mathbb{r}}\times[0,1 ] ) } \bigr).\end{aligned}\ ] ] in the nonlinear case ( a ) the constant in this estimate depends continuously on @xmath846 in the @xmath551-topology , see e.g. ( * ? ? ?
* appendix b ) . in this case
the above estimate already implies the claim ( [ xito0 ] ) since we assumed @xmath885 . in the linear cases we need to use the injectivity of the operators to remove the last term from the right hand side of .
since @xmath886 is compact only for @xmath887 , we first have to achieve a lower order term on a compact domain : consider the operator @xmath888 , where @xmath889 ( or @xmath890 in the case @xmath891 ) is self - adjoint and invertible on its constant domain @xmath892,t_{x^\pm}m_{02})$ ] with boundary conditions @xmath893 , @xmath894 . then abstract theory ( e.g. ( * ? ? ?
* lemma 3.9 , proposition 3.14 ) ) implies the fredholm property and bijectivity , @xmath895 ) } \leq c \vert d_{x^\pm } \eta \vert_{h^0({\mathbb{r}}\times[0,1 ] ) } .\ ] ] in order to apply this estimate to @xmath896 we first find an extension @xmath897)$ ] of @xmath898 such that @xmath899 .
we moreover fix a cutoff function @xmath900)$ ] with @xmath901 and @xmath902 , where we fix @xmath903 sufficiently large such that @xmath904 for some smooth map @xmath905 .
then we can apply the estimate to @xmath906 , where @xmath907 denotes parallel transport along the path @xmath139\ni\tau\mapsto e_{x^\pm}(\tau\vartheta_{02})$ ] .
we obtain , denoting all uniform constants by @xmath479 , @xmath908 ) } \\ & \leq c \vert \eta \vert_{h^1({\mathbb{r}}\times[0,1 ] ) } + \vert h \zeta \vert_{h^1({\mathbb{r}}\times[0,1 ] ) } \\ & \leq c \bigl ( \vert \bigl ( d_{x^\pm } - d\circ\phi_{x^\pm}(\vartheta_{02 } ) \bigr ) \eta \vert_{h^0({\mathbb{r}}\times[0,1 ] ) } + \vert d ( h\xi^\nu_{02 } ) \vert_{h^0({\mathbb{r}}\times[0,1 ] ) } + \vert h \zeta \vert_{h^1({\mathbb{r}}\times[0,1 ] ) } \bigr ) \\ & \leq c \bigl ( \vert \bigl ( d_{x^\pm } - d\circ\phi_{x^\pm}(\vartheta_{02 } ) \bigr)\bigr|_{\{|s|>t-1\}}\vert \cdot \| h ( \xi^\nu_{02 } - \zeta ) \vert_{h^1({\mathbb{r}}\times[0,1 ] ) } + \vert d\xi^\nu_{02 } \vert_{h^0({\mathbb{r}}\times[0,1 ] ) } \\ & \qquad + \vert \xi^\nu_{02 } \vert_{h^0([-t , t]\times[0,1 ] ) } + \|\pi_{02}^\perp\xi^\nu_{02}|_{t=0}\|_{h^{1/2}({\mathbb{r } } ) } \bigr ) .\end{aligned}\ ] ] here the difference of the operators goes to zero for @xmath876 since @xmath909 with all derivatives .
thus for sufficiently large @xmath910 we can absorb the first term into the left hand side and @xmath911 .
after all this we can finally replace the last term in ( [ basest ] ) by @xmath912\times[0,1])}$ ] .
now in the first case of ( b ) we can deduce ( [ xito0 ] ) from the fact that @xmath913 is surjective by assumption and hence @xmath914 is injective .
so the compact embedding @xmath915)\hookrightarrow h^0([-t , t]\times[0,1])$ ] allows the removal of the lower order term .
similarly , in the second case of ( b ) we can employ the injectivity of the operator on @xmath916 to deduce ( [ xito0 ] ) .
next , ( [ xito0 ] ) and the sobolev trace theorem provide @xmath917 , and again using lemma [ lltrans ] we can deduce that @xmath918 finally , combining this with ( [ txii ] ) in case ( b ) implies @xmath919)}\to 0\ ] ] in contradiction to the assumption
. finally , we establish uniform exponential decay for the solutions of floer s equation ( [ vvvv ] ) on the triple strip .
for that purpose we introduce the following notation for integration over finite strips , @xmath920\sqcup[0,\delta ] } \left|\pd_s v(s , t ) \right|^2 dt : = \int_0 ^ 1 \left|\pd_s v_{02}(s , t ) \right|^2 dt + \int_0^\delta \left|\pd_s \hat v(s , t ) \right|^2 dt , \end{aligned}\ ] ] and similarly for the @xmath606-norm @xmath921 ) } & : = \| \pd_s v_{02 } \|_{l^\infty([s_0,s_1]\times[0,1 ] ) } + \| \pd_s \hat v \|_{l^\infty([s_0,s_1]\times[0,\delta ] ) } , \\
d_{{\mathcal{c}}^0_{1,\delta}([s_0,s_1])}(v , x^\pm ) & : = \sup_{(s , t)\in [ s_0,s_1]\times[0,1 ] } d_{m_{02}}(v_{02}(s , t ) , x^\pm ) , \\ &
\quad + \sup_{(s , t)\in [ s_0,s_1]\times[0,\delta ] } d_{m_{0211}}(\hat v(s , t ) , ( x^\pm , x_1^\pm , x_1^\pm ) ) .\end{aligned}\ ] ] [ expdecay ] there are constants @xmath922 and @xmath479 such that the following holds for every @xmath600 $ ] .
if @xmath923 is a smooth solution of ( [ vvvv ] ) satisfying @xmath924\sqcup[0,\delta ] } \left|\pd_s v(s , t ) \right|^2 dt ds \;<\ ; \hbar , \ ] ] then for every @xmath925 @xmath926\sqcup[0,\delta ] } \left| \pd_s v ( s , t ) \right|^2 dt ds , \ ] ] and the analogous statement holds on @xmath927 $ ] for the convergence to @xmath928 .
* step 1 : * _ for every @xmath929 there is an @xmath930 such that the following holds for all @xmath600 $ ] .
if @xmath923 satisfies ( [ eq : longenergy ] ) with @xmath931 , then _
@xmath932 assume by contradiction that this is wrong .
then there exist @xmath929 and sequences @xmath933 $ ] and @xmath934 such that @xmath935\sqcup[0,\delta^\nu ] } \left|\pd_s v^\nu(s , t ) \right|^2 dt ds = 0 , \ ] ] but the assertion fails .
so after a time - shift we can assume that @xmath936 ) } > \tfrac 12 \kappa .\ ] ] the equation @xmath937 together with implies that @xmath938 in the @xmath703-norm .
if @xmath939 is bounded away from zero , then the standard compactness for a variety of mild ` bounded geometry ' assumptions which can ensure the initial @xmath606-bound on the curves . ] for pseudoholomorphic curves with lagrangian boundary conditions implies that @xmath940 in @xmath281 on every compact set ( for a subsequence ) , in contradiction to the assumption . in the case
@xmath941 the standard compactness theory still implies @xmath942\times(0,\infty)}\to 0 $ ] in @xmath281 on every compact set . for @xmath943 and @xmath371 near the boundary @xmath266 we obtain a @xmath551-bound from lemma [ bubb ] .
so we obtain @xmath606-convergence of a subsequence @xmath944 , @xmath945 to constants @xmath946 , @xmath947 such that @xmath948 .
now we can use the same compactness arguments as in the proof of lemma [ bubb ] ( step 2 , using a cutoff function only in @xmath949 ) to deduce that @xmath950}\to 0 $ ] in the @xmath606-norm .
this again is a contradiction .
* step 2 : * _ there are constants @xmath951 and @xmath647 such that the following holds for all @xmath600 $ ] . if @xmath923 satisfies ( [ eq : longenergy ] ) with @xmath952 , then _
@xmath953\sqcup[0,\delta])}^2 \leq c_1 \int_{[0,1]\sqcup[0,\delta ] } \left|\nabla_t \pd_s v(1,t ) \right|^2 dt .\ ] ] by contradiction we find sequences @xmath933 $ ] and @xmath934 that satisfy ( [ noenergy ] ) , but there is no uniform constant @xmath647 with which the estimate holds .
then as in step 1 we obtain ( for a subsequence ) @xmath551-convergence @xmath954 on @xmath955\times ( [ 0,1]\sqcup[0,\delta^\nu])$ ] to constants @xmath946 , @xmath947 with @xmath956 . by assumption @xmath457 and
@xmath957 intersect transversely in @xmath958 , and hence we have for all @xmath959\to t_{x_{02}}m_{02}$ ] with @xmath960 @xmath961 ) } \leq c \bigl ( \|\nabla_t\xi_{02}\|_{l^2([0,1 ] ) } + \bigl|\pi_{02}^\perp\xi_{02}(0)\bigr| \bigr ) .\ ] ] now consider in addition @xmath962\to
t_{\hat x}m_{0211}$ ] such that @xmath963 and @xmath964 .
we integrate for all @xmath965 $ ] @xmath966 combining this with lemma [ lltrans ] and using the boundary conditions we obtain @xmath967 and thus @xmath961)}^2 \leq c^2 \int_{[0,1]\sqcup[0,\delta ] } \left|\nabla_t \xi \right|^2 dt .\ ] ] we moreover obtain from lemma [ lltrans ] with uniform constants @xmath968 @xmath969\sqcup[0,\delta ] } \bigl|\nabla_t ( \xi_{02},\hx ) \bigr|^2 dt \biggr)^{1/2 } .\end{aligned}\ ] ] together with ( [ diff ] ) this implies @xmath970\sqcup [ 0,\delta])}^2 \leq c_1 \int_{[0,1]\sqcup[0,\delta ] } \left|\nabla_t \xi \right|^2 dt\ ] ] with some uniform constant @xmath647 for all @xmath600 $ ] and all sections @xmath971 over @xmath972 satisfying the boundary conditions . due to the @xmath551-convergence
@xmath973 this estimate continues to hold with a uniform constant for sufficiently large @xmath974 for sections @xmath975,{v^\nu_{02}|_{s=1}}^*tm_{02})$ ] , @xmath976,{\hat v}|_{s=1}^*tm_{0211})$ ] that satisfy the analogous boundary conditions .
( we can write @xmath977 with @xmath978 and use @xmath979 to map @xmath980 to a section over @xmath972 . this preserves the boundary conditions by construction of @xmath981 . ) in particular , we can apply this new estimate to @xmath982 , which provides a uniform estimate and thus finishes the proof by contradiction .
* step 3 : * _ there are uniform constants @xmath983 and @xmath840 such that the following holds for all for all @xmath600 $ ] .
if @xmath923 satisfies ( [ eq : longenergy ] ) with @xmath984 , then for all @xmath985 _ @xmath986\sqcup[0,\delta ] } | \pd_s v ( s_0,t ) |^2 dt \leq c_2 e^{-\delta s_0 } \int_1 ^ 2 \int_{[0,1]\sqcup[0,\delta ] } | \pd_s v ( s , t ) |^2 dt ds .\ ] ] consider the function @xmath987 defined by @xmath988\sqcup[0,\delta ] } |\pd_s v(s , t)|^2 dt .\ ] ] we can use the equation @xmath989 and the bound @xmath990 from step 1 to calculate for @xmath991 @xmath992\sqcup[0,\delta ] } \bigl ( - { \langle}\pd_s v \,,\ ,
j \nabla_t \nabla_s \pd_s v { \rangle}\bigr)\\ & \quad - \int_{[0,1]\sqcup[0,\delta ] } \bigl ( { \langle}\pd_s v \,,\ , j r(\pd_s v,\pd_t v)\pd_s v + 2 ( \nabla_{\pd_s v}j ) \nabla_s\pd_t v + \nabla_s(\nabla_{\pd_s v}j ) \pd_t v { \rangle}\bigr ) \\ & \geq \int_{[0,1]\sqcup[0,\delta ] } \bigl ( 2 |j\nabla_t\pd_s v|^2 + \pd_t\bigl ( { \omega}(\pd_s v,\nabla_s\pd_s v)\bigr ) - c |\pd_s
v|^2 \bigl ( |\pd_s v|^2 + |\nabla_t\pd_s v|\bigr ) \bigr ) \\ & \geq \bigl(2 - c\kappa\bigr ) \int_{[0,1]\sqcup[0,\delta ] } |j\nabla_t\pd_s v(s , t)|^2 dt \;- c'\bigl(\kappa + \kappa^2\bigr ) \|\pd_s v(s,\cdot)\|_{{\mathcal{c}}^0([0,1]\cup[0,\delta])}^2 .\end{aligned}\ ] ] the last step uses @xmath993 and the claim @xmath994\sqcup[0,\delta ] } \pd_t\bigl ( { \omega}(\pd_s v,\nabla_s\pd_s v)\bigr)\biggr| \leq c\bigl(|\pd_s v_{02}(1)|^3 + |\pd_s \hat v(\delta)|^3 \bigr ) .\ ] ] to prove the claim we first use the diagonal boundary conditions to obtain @xmath994\sqcup[0,\delta ] } \pd_t\bigl ( { \omega}(\pd_s v,\nabla_s\pd_s v)\bigr)\biggr| = \bigl| { \omega}_{02}(\pd_s v_{02},\nabla_s\pd_s v_{02})|_{t=1 } + { \omega}_{02}(\pd_s \hat v,\nabla_s\pd_s \hat v)|_{t=\delta } \bigr| .\ ] ] then we use a smooth family of orthonormal frames @xmath995 near @xmath996 ( and similarly for @xmath943 ) , @xmath997 with @xmath781 . by the orthonormality we have @xmath998 , and using the identities @xmath785 one obtains @xmath999 , where the constant @xmath479 only depends on @xmath1000 . since @xmath120 is compact this holds with a uniform constant
. we can now choose @xmath929 sufficiently small and then fix @xmath1001 such that step 1 and step 2 ( applied to time - shifts of @xmath45 ) together with the above calculation yield for all @xmath991 @xmath1002\sqcup[0,\delta ] } |j\nabla_t\pd_s v(s , t)|^2 dt \geq ( ( 1+\delta)c_1)^{-1 } \int_{[0,1]\sqcup[0,\delta ] } |\pd_s v(s , t)|^2 dt \geq \delta^2 f(s)\ ] ] with @xmath1003 .
any such nonnegative convex function satisfies for all @xmath1004 and @xmath1005 @xmath1006 } f(t ) dt + \int_{[2t,2t+1 ] } f(t ) dt \biggr)\ ] ] with a constant @xmath479 that only depends on @xmath489 .
a detailed proof can be found in e.g. ( * ? ? ?
* lemma 3.7 ) ( use the estimate for @xmath1007 , where the function @xmath1008 is shifted by @xmath1009 ) .
if we let @xmath876 then @xmath1010 } f(t ) dt \to 0 $ ] by the finite energy condition @xmath1011 , and this proves the claim .
* step 4 : * _ there are constants @xmath1012 and @xmath1013 such that the following holds for all @xmath600 $ ] . if @xmath923 satisfies ( [ eq : longenergy ] ) with @xmath1014 , then _
@xmath1015 ) } \leq c_3 \left\|\pd_s v \right\|_{l^2_{1,\delta}([\frac 12,\frac 52 ] ) } .\ ] ] by contradiction we find sequences @xmath933 $ ] and @xmath934 that satisfy ( [ noenergy ] ) , but the assertion fails , i.e. we can not find a constant @xmath1013 for which the estimate is satisfied .
then as in step 1 we obtain ( for a subsequence ) @xmath551-convergence @xmath954 on @xmath1016\times ( [ 0,1]\sqcup[0,\delta^\nu])$ ] to constants @xmath946 , @xmath947 with @xmath956 .
so we can find sections @xmath1017 over @xmath1018 such that @xmath1019}=e_x(\xi^\nu)$ ] .
the equation @xmath1020 then becomes @xmath1021 and we have the boundary conditions @xmath1022 and @xmath1023 .
we fix two cutoff functions @xmath1024)$ ] with @xmath1025}\equiv 1 $ ] , @xmath1026 and @xmath1027 and consider the sections @xmath1028 .
note that @xmath1029 with @xmath1030 .
so for sufficiently large @xmath974 we have @xmath1031 ) } & \leq 2 \|h \nabla_s \xi^\nu \|_{{\mathcal{c}}^0_{1,\delta^\nu } } \leq 2c_s \|h \nabla_s \xi^\nu \|_{h^2_{1,\delta^\nu } } , \\ \|\nabla_s \xi^\nu \|_{l^2_{1,\delta^\nu}([\frac 12,\frac 52 ] ) } & \leq 2 \|\pd_s v^\nu \|_{l^2_{1,\delta^\nu}([\frac 12,\frac 52 ] ) } , \end{aligned}\ ] ] where we used lemma [ sobolev ]
. now we apply lemma [ ddelta ] ( b ) to the sections @xmath1032 and @xmath1033 ( for which the boundary terms vanish since @xmath1034 satisfy the boundary conditions ) and @xmath1035 ( which satisfy @xmath1036 and @xmath1037 ) to obtain with uniform constants @xmath620 @xmath1038 } .\end{aligned}\ ] ] now the contradiction follows , @xmath1031 ) } \leq 2 \| h\nabla_s\xi^\nu \|_{h^2_{1,\delta^\nu } } \leq 2c ' \| \nabla_s\xi^\nu \|_{h^0_{1,\delta^\nu}([\frac 12,\frac 52 ] ) } \leq 4c ' \|\pd_s v^\nu \|_{l^2_{1,\delta^\nu}([\frac 12,\frac 52 ] ) } .\end{aligned}\ ] ] * step 5 : * _ we prove the claim , that is for every @xmath1039 @xmath1040\sqcup[0,\delta])}(v(s,\cdot ) , x^+ ) ^2 + \left\|\pd_s v(s,\cdot ) \right\|_{{\mathcal{c}}^0([0,1]\sqcup[0,\delta])}^2 \le c e^{-\delta s } e'(v)\ ] ] with
_ @xmath1041\sqcup[0,\delta ] } | \pd_s v ( s , t ) |^2 dt ds .\ ] ] we choose @xmath1042 , then step 3 and step 4 ( applied to appropriately shifted solutions ) combine as follows for all @xmath1039 @xmath1043)}^2 & \leq c_3 ^ 2 \int_{s-1}^{s+1 } \int_{[0,1]\sqcup[0,\delta ] } | \pd_s v ( s , t ) |^2 dt \\ & \leq c_3 ^ 2 c_2 \int_{s-1}^{s+1 } e^{-\delta s } e'(v ) ds \;\leq\ ; c_3 ^ 2 c_2 \delta^{-1 } e^{\delta } e^{-\delta s } e'(v ) .\end{aligned}\ ] ] this proves the second part of the claim .
the estimate on @xmath1044\sqcup[0,\delta])}(v(s,\cdot ) , x^+ ) $ ] now simply follows by integration : for all @xmath925 and @xmath176 $ ] @xmath1045 and similarly for @xmath943 .
the surjectivity of the map @xmath1046 , as introduced in the previous section , will be a direct consequence of the following compactness result . here
we choose @xmath1047 $ ] with @xmath392 from in theorem [ solving ] .
then @xmath1048 with @xmath1049 implies that @xmath1050={\mathcal{t}}_\delta([u])$ ] by the definition of @xmath737 via theorem [ solving ] . we will denote the time - shift by @xmath1051 .
[ compact ] given @xmath434 there exists @xmath393 such that for every @xmath394 $ ] and @xmath1052 there exist @xmath345 and @xmath1053 such that @xmath1054 with @xmath721 and @xmath1055 .
moreover , the moduli space @xmath1056 is regular for all @xmath394 $ ] in the sense that the linearized operator @xmath1057 is surjective for every @xmath1052 .
we assume by contradiction that there is an @xmath434 , a sequence @xmath941 , and solutions @xmath1058 for which the assertion of the theorem fails .
their energy is fixed , @xmath1059 , by the analogue of proposition [ prop monotone ] for strips of different widths : for any pair of maps @xmath1060 that are not necessarily pseudoholomorphic but satisfy the limits and seam conditions of @xmath1061 we have @xmath1062 here @xmath1063 is independent of @xmath298 since the equations for different @xmath298 apply to the same map , rescaled to different widths , which has the same energy and index .
next , we can exclude bubbling by the following argument based on lemma [ bubb ] below : if @xmath1064 is unbounded near a point @xmath1065 $ ] , then the standard rescaling method gives rise to a nontrivial pseudoholomorphic sphere or disk in @xmath57 , or in @xmath60 , or in both . or @xmath68 are noncompact , this convergence can be ensured by ` bounded geometry ' assumptions , or energy concentration can be proven directly , as outlined in section [ sec : mon ] .
] thus some fixed amount of energy @xmath1066 would have to concentrate near @xmath1067 .
the same energy quantization holds for blowup of @xmath1068 or @xmath1069 by lemma [ bubb ] .
so the energy densities @xmath1070 can only blow up at finitely many points . on the complement the same compactness proof as in the next paragraph provides a @xmath1071 convergent subsequence @xmath1072 , where the limit corresponds to a solution @xmath1073 with finitely many singularities and energy @xmath1074 .
the singularities can be removed by the standard proofs for pseudoholomorphic curves with lagrangian boundary condition ( * ? ? ?
* theorem 4.1.2 ) , so we would obtain a solution @xmath1075 of energy @xmath1076 .
its limits @xmath1077 may not be the same as those of @xmath1078 , in which case we find a sequence of trajectories @xmath1079 connecting @xmath1080 to @xmath1081 , with total energy @xmath1082 .
we claim that monotonicity forces @xmath1083 to have total index @xmath1084 , and hence by regularity of the moduli spaces @xmath1085 consists of a single constant trajectory .
this however would mean that @xmath1078 were self - connecting trajectories of @xmath1086 , i.e. we have annuli with @xmath1087 in contradiction to assumption . to control the index of @xmath1083 we glue the trajectories to a single map @xmath1088\to m_0 ^ -\times m_2 $ ] satisfying all limit and boundary conditions of @xmath1089 except for holomorphicity . its index and energy coincide with the total energy and index of @xmath1083 . with that we obtain @xmath1090 from the monotonicity formula and the index and energy identities in lemma [ index ] applied to @xmath1091 , where @xmath1092 is the @xmath263-independent map given by the lift of @xmath1093 to @xmath1094 .
this proves @xmath1095 as claimed and hence excludes bubbling .
so from now on we assume that @xmath1096 is uniformly bounded .
in addition , the boundary condition in the compact set @xmath450 implies a priori bounds on the map @xmath1097 . together with the uniform gradient bound on @xmath1098 this implies a priori bounds on @xmath1099 , which transfer to @xmath1100 via the boundary condition . finally using the uniform gradient bounds on @xmath1101 we then obtain uniform @xmath551-bounds on both @xmath1101 and @xmath1098 on every ball @xmath1102 of fixed radius ( intersected with the domain of the respective map ) .
so we can fix @xmath603 and find a subsequence and map @xmath1103,m_0 \times m_2)$ ] such that @xmath1072 in the @xmath606-topology and the weak @xmath1104-topology on every compact subset of @xmath1105 $ ] .
we claim that the limit @xmath372 corresponds to a solution @xmath1106 .
indeed , the holomorphicity follows from the weak @xmath1104-convergence , and the boundary condition follows from @xmath1107 . to check the latter ,
recall that lemma [ lltrans ] ( a ) bounds this distance in terms of @xmath1108 , which due to the boundary conditions on @xmath1109 is bounded by @xmath1110 .
we also conclude that @xmath1111 in @xmath1112\times[0,\delta^\nu])$ ] for all @xmath910 , where @xmath1113 is determined uniquely by @xmath1114 .
indeed , @xmath1115 satisfies @xmath1116 as well as @xmath1117 , so @xmath1118 must converge to @xmath1113 on compact sets , and the convergence for @xmath617 $ ] follows from @xmath1119 .
in summary we have @xmath1120 in the @xmath606-topology on every set @xmath1121 for fixed @xmath1122 . in the following
, we will strengthen this convergence using uniform nonlinear estimates and exponential decay , to find sections @xmath1123 such that @xmath1124 and @xmath1125 is surjective in contradiction to the assumption .
let us first note that , by the same monotonicity arguments as above , the limit must be a nonbroken trajectory @xmath1126 of the same index and energy @xmath1127 .
in the next step we strengthen the local convergence . for fixed @xmath910 and sufficiently large @xmath1128
we can write @xmath1129 with a section @xmath1130 ( extended smoothly to @xmath1131 ) . the extension of @xmath1132 can be chosen such that @xmath1133 and @xmath1134 follows from the @xmath606-convergence and @xmath551-boundedness of @xmath1135 .
for the latter note that @xmath1136 , where @xmath1137 is uniformly bounded , and @xmath1138 as @xmath1139 .
this puts us into the position where lemma [ ddelta ] applies with @xmath1035 .
we fix a cutoff function @xmath1140,[0,1])$ ] with @xmath1141}\equiv 1 $ ] , then @xmath1142 now we can use ( [ dbarxi ] ) , @xmath1143 , @xmath1144 , and @xmath697 to obtain @xmath1145 ) } & = \bigl\| h \cdot de_\bu(\hx^\nu)^{-1 } \bigl(\pd_1 e(\bu,\hx^\nu ) \pd_s
\bu \bigr ) \bigr\|_{l^2([-t , t]\times[0,\delta^\nu ] ) } \\ & \leq c \|\pd_s\bu\|_{l^2([-t , t]\times[0,\delta^\nu ] ) } \leq c \sqrt{\delta^\nu}\|\pd_s\bu\|_{l^2([-t , t ] ) } , \end{aligned}\ ] ] and furthermore , using the fact that @xmath1146 commutes with @xmath1147 , @xmath1148 ) } \\ & = \bigl\| h \cdot de_{u_{02}}(\xi_{02}^\nu)^{-1 } \bigl ( \pd_1 e(u_{02},\xi_{02}^\nu ) j(u_{02})\pd_t u_{02 } - j_{02}(u_{02 } ) \pd_1 e(u_{02},\xi_{02}^\nu)\pd_t u_{02 } \bigr ) \bigr\|_{l^2({\mathbb{r}}\times[0,1 ] ) } \\ & \leq c \|\xi_{02}^\nu\|_{l^2([-t , t]\times[0,1 ] ) } .\end{aligned}\ ] ] hence we have @xmath1149 which converges to zero , and thus @xmath1072 in the @xmath550-norm on every compact set .
now we can verify the assumptions of lemma [ expdecay ] ( with the constant @xmath1066 ) and achieve uniform exponential decay : pick @xmath910 such that @xmath1150\times[0,1 ] } |\pd_s u_{02}|^2 \ge e(u ) - \frac 12 \hbar$ ] and pick @xmath1151 such that for all @xmath1152 we have @xmath1153\times[0,1])}^2 - \| \pd_s v_{02}^\nu \|_{l^2([-t , t]\times[0,1])}^2 \leq \frac 12\hbar$ ] and thus @xmath1154 } |\pd_s v^\nu_{02}|^2 + \int_{[0,\delta^\nu ] } |\pd_s \hat v|^2 \biggr ) \;\le\ ; e(v^\nu ) + \tfrac 12 \hbar - e(u ) + \tfrac 12\hbar \;=\ ; \hbar .\ ] ] now the exponential decay lemma [ expdecay ] combined with the local @xmath606-convergence implies that @xmath1155 uniformly for all @xmath1156 .
thus for sufficiently large @xmath974 we can write @xmath1124 with @xmath611 and @xmath1157 .
in fact , the uniform exponential decay implies global convergence , @xmath1158 this puts us into the position where lemma [ ddelta ] and [ d02 ] apply with @xmath1035 , @xmath1159 ) } \bigr).\end{aligned}\ ] ] the terms in the last line converge to zero or can be absorbed into the left hand side for @xmath939 sufficiently small .
we claim that the penultimate line also converges to zero and we thus obtain the convergence @xmath1160 . to check this we recall from ( [ dbarxi ] ) that @xmath1143 implies @xmath1161 recall that @xmath1162 so in zeroth order we have , using the equations @xmath697 and @xmath1163 , @xmath1164 and thus @xmath1165 ) } + \|\xi_{02}^\nu\|_{l^4({\mathbb{r}}\times[0,1 ] ) } + ( \delta^\nu)^{1/2}\|\pd_s\bu\|_{l^2({\mathbb{r } } ) } + ( \delta^\nu)^{1/4 } \|\pd_s\bu\|_{l^4({\mathbb{r } } ) } \bigr ) \to 0 .\end{aligned}\ ] ] for the first derivative we calculate from , denoting all uniform constants by @xmath479 , @xmath1166 and ( in between dropping the subscript from @xmath896 ) @xmath1167 here the estimate for the second summand follows from ( [ pd2 ] ) and the identity @xmath1168 ( and similarly for @xmath1169 ) , where we have @xmath1170 since @xmath1171 and , calculating in local normal coordinates with an extension @xmath1172 of @xmath1173 that is covariantly constant along @xmath1174 , @xmath1175
now the uniform estimate @xmath1176 and the exponential decay of @xmath1177 imply @xmath1178 ) } + ( \delta^\nu)^{1/2}\|\pd_s\bu\|_{h^1({\mathbb{r } } ) } \bigr ) \to 0 .
\end{aligned}\ ] ] this proves @xmath1179 it remains to find a time - shift such that @xmath1180 with some @xmath1181 but still @xmath1182 .
in order to find this shift we write @xmath1183 with @xmath1184 this will satisfy @xmath1185 so it is well defined whenever @xmath1186 , where we fixed @xmath1187 such that @xmath1188 is ensured for sufficiently large @xmath1152 .
the @xmath703-estimate on @xmath1189 can be seen from the pointwise estimate @xmath1190 here @xmath479 is a continuity constant for @xmath1191 .
the higher derivatives of @xmath1192 are estimated similarly .
now consider the function @xmath1193 it satisfies @xmath1194 and ( dropping the @xmath1195-subscript ) we obtain from @xmath1196 the latter is an arbitrarily small error for large @xmath974 and small @xmath1197 .
hence we will find solutions @xmath1198 $ ] of @xmath1199 . with these we have @xmath1200 , where @xmath1201 and @xmath1188 .
so with this small time - shift on @xmath1078 we obtain a contradiction to the assumption that @xmath1202 is not surjective . finally , to prove the transversality we need to check that @xmath1203 is surjective .
( the same then holds for the time shifts @xmath1204 . )
this follows from the quadratic estimate in lemma [ quadratic ] : let @xmath1205 be the right inverse of @xmath1206 , then @xmath1207 where @xmath1208 by ( [ right inverse ] ) and @xmath1209 .
this shows that @xmath1210 and hence also the operator @xmath1211 has a right inverse for all sufficiently large @xmath1152 . here
the parallel transport @xmath1212 is an isomorphism on the target and @xmath1213 identifies @xmath429 with the domain of @xmath1214 . for the latter see the discussion before lemma [ quadratic ] and recall that @xmath630 .
so we have established that @xmath1125 is surjective , and this finishes the proof .
[ bubb ] there exists a universal constant @xmath1066 such that the following holds for any sequence of floer trajectories @xmath1215 with @xmath941 .
if for some @xmath464 @xmath1216 then there exists a sequence @xmath1217 such that @xmath1218 here @xmath1219 is the @xmath405-ball in @xmath1105 $ ] or @xmath1220 $ ] respectively . in the usual analysis of bubbling effects , one would prove this lemma by rescaling around points where the differentials blow up , identifying the limits with pseudoholomorphic spheres or disks , and hence obtaining an energy quantization constant @xmath1221 that is geometrically determined by the minimal nonzero energy of spheres or disks . in the present case however , depending on the relative speed of blow - up and strip - shrinking @xmath941 , the rescaling may lead to sphere bubbles in @xmath0 , @xmath1 , or @xmath68 , disk bubbles in @xmath77 , @xmath78 , or @xmath79 , or the novel _ figure eight bubble _ described in the introduction .
since we do not have a geometric bound on the minimal energy of figure eight bubbles , we use a mean value inequality to obtain @xmath1221 by purely analytic methods .
suppose the lemma is false , that is , for every @xmath1226 there exists a sequence @xmath1227 with @xmath1228 such that ( after time shift to @xmath1229 ) @xmath1230 for some @xmath1231 , but @xmath1232 for every sequence @xmath1217 .
in particular , this will hold for a fixed sequence @xmath1233 that satisfies in addition @xmath1234 , @xmath1235 and @xmath1236 .
we can then find diagonal sequences @xmath1237 with @xmath1238 , and @xmath1239 , @xmath1240 such that @xmath1241 and @xmath1242 next , we use lemma [ hofer ] to refine the choice of the blowup points @xmath1243 . for that purpose
we consider the spaces @xmath1244 $ ] , @xmath1245 $ ] , and @xmath1246 $ ] , with the obvious inclusion @xmath1247 . using the function @xmath1248 on @xmath1249 and @xmath1250 on @xmath1251 one can then vary the point @xmath1252 $ ] by @xmath1253 to find @xmath1254 and @xmath1255 , such that @xmath1256 and @xmath1257 on @xmath1258 . here
( [ no energy ] ) continues to hold on @xmath1259 .
now in a first step we will prove that figure eight bubbles ( arising from rescaling in the case @xmath1260 ) have a minimal energy ( possibly depending on @xmath1003 . )
more precisely , we claim that ( [ no energy ] ) implies @xmath1261 in a second step we will then see that this gives rise to a disk bubble in @xmath79 .
first consider the case @xmath1262 and @xmath1263 .
then for all sufficiently large @xmath19 we can apply the mean value inequality w.r.t .
@xmath287-compatible metrics . ]
( * ? ? ? * lemma 4.3.1 ) to @xmath1264 on the ball @xmath1265 with @xmath1266 , @xmath1267 here we can not have @xmath1268 since @xmath1269 , so we have @xmath1270 and thus @xmath1271 as claimed . in the case
@xmath1272 and @xmath1273 we can apply the mean value inequality ( * ? ? ? * theorem 1.3 , lemma a.1 ) to @xmath1274 with boundary condition @xmath1275 on the partial ball
@xmath1276 \cap b_{{\epsilon}_k}(0)$ ] for @xmath1277 , @xmath1278 as before we can not have @xmath1268 since @xmath1269 , so we have @xmath1279 and thus @xmath1280 as claimed . in the remaining case @xmath1281
we consider the pseudoholomorphic map @xmath1282\to m_0\times m_2 \times m_0\times m_2\times m_1 \times m_1 , \ ] ] which satisfies the lagrangian boundary condition @xmath1283 . by
the above we have @xmath1284 and @xmath1285 .
so for all sufficiently large @xmath19 we can apply the mean value inequality ( * ? ? ? * theorem 1.3 , lemma a.1 ) on the partial ball @xmath1286 for @xmath1277 , @xmath1287 again we can not have @xmath1268 since @xmath1269 , so we have @xmath1279 and thus @xmath1288 as claimed .
we consider the rescaled maps @xmath1289 , where @xmath1290 is defined on half balls of radius @xmath1291 in the half space @xmath1292 by @xmath1293 , and @xmath1294 ) \to m_0\times m_2\times m_1\times m_1 $ ] is defined by @xmath1295 on balls of radius @xmath1296 intersected with the strip of width @xmath1297 .
this rescaling preserves the nontriviality @xmath1298 , but on both domains @xmath1299 is uniformly bounded . hence we can find a subsequence of the @xmath1300 that converges in the @xmath606-topology on the unit half ball @xmath1301 .
the ( scaling invariant ) energy @xmath1302 converges to zero by ( [ no energy ] ) , so the limit has to be constant .
in fact , we have @xmath1303 since the boundary values @xmath1304 converge to @xmath1305 in @xmath1306)$ ] . to see the latter use lemma [ lltrans ] ( a ) to bound the distance to @xmath457 by the distance @xmath1307 , which is zero for @xmath266 replaced by @xmath1308 .
however , the bound on @xmath1309 provides a bound @xmath1310 and thus proves @xmath1311 .
this also proves that @xmath1312 in @xmath1306\times[0,\delta_k r_k])$ ] , where @xmath947 is uniquely determined by @xmath1313 .
the maps @xmath1300 are @xmath1314-holomorphic , so by elliptic regularity the convergence @xmath1315 is in the @xmath281-topology on every compact subset of @xmath1316 .
however , in order to obtain a contradiction to the fact that @xmath1298 with @xmath1317 we need to establish @xmath551-convergence on @xmath1318 up to the boundary .
we begin by noting that due to the @xmath606-convergence we can express @xmath1319 in terms of sections @xmath1320\times[0,\delta_k
r_k],\bar x^*t(m_0\times m_2\times m_1\times m_1))$ ] using the exponential map centered at @xmath1321 .
these sections satisfy the diagonal and lagrangian boundary conditions @xmath1322 and @xmath1323 , the @xmath606-convergence @xmath1324 , and a uniform bound @xmath1325 .
since @xmath1326 and @xmath1327 we obtain from ( [ dbarxi ] ) @xmath1328 now @xmath1329 , so it suffices to prove the @xmath606-convergence of @xmath1330 near @xmath118 . for that purpose
we multiply the sections by cutoff functions @xmath1331 with @xmath1332\to[0,1]$ ] supported in @xmath1318 , @xmath1333 $ ] supported in @xmath1334 $ ] , and both equal to @xmath117 near @xmath118 .
then we obtain sections on the multistrip @xmath1335 that also satisfy the boundary condition @xmath1336 .
these satisfy a uniform bound @xmath1337 due to the bounds on @xmath1338 and @xmath1339 and the compact support of @xmath1340 . from this lemma [ ddelta ] ( b )
provides a uniform bound @xmath1341 indeed , the boundary terms vanish since the constant boundary conditions directly transfer to the derivatives , @xmath1342 and @xmath1343 .
we now fix a pair of cutoff functions @xmath1344 with support in @xmath1345 and still equal to @xmath117 near @xmath118 .
then we apply lemma [ ddelta ] ( b ) to @xmath1346 , again with vanishing boundary terms , to obtain @xmath1347 we can pick the cutoff functions such that @xmath1348 on the half ball @xmath1349 and @xmath1350}\equiv 1 $ ] .
then the compact sobolev embedding @xmath1351 provides @xmath606-convergence of a subsequence @xmath1352 .
we already know that the limit is @xmath118 , so we obtain @xmath1353 and @xmath1354 in @xmath1355 .
it remains to establish @xmath1356\times[0,\delta_k r_k ] ) } \to 0 $ ] and thus @xmath1357\times[0,\delta_k r_k])}\to 0 $ ] in contradiction to @xmath1298 with @xmath1317 . to see this
we follow the argument in lemma [ sobolev ] . using the standard sobolev embedding @xmath1358)\hookrightarrow{\mathcal{c}}^0([-\frac 12,\frac 12])$ ]
we obtain for all @xmath1359 $ ] @xmath1360)}^2 & \leq \| \nabla_s\hat\xi^k|_{t = t_0 } - \nabla_s\hat\xi^k|_{t=\delta_k r_k } \|_{h^1([-\frac 12,\frac 12])}^2 \\ & \leq \delta_k r_k \int_0^{\delta_k r_k } \|\nabla_t\nabla_s\hat\xi^k\|_{h^1([-\frac 12,\frac 12])}^2 \\ & \leq \delta_k r_k \|\nabla_s\hat\xi^k\|_{h^2([-\frac 12,\frac 12]\times[0,\delta_k r_k])}^2 \to 0 .
\nonumber\end{aligned}\ ] ] from the above we moreover have @xmath1361 ) } = \| \nabla_s\xi^k_{02}|_{t=0 } \|_{{\mathcal{c}}^0([-\frac 12,\frac 12 ] ) } \to 0 $ ] . now using lemma [ lltrans ] and the boundary conditions , in particular @xmath1362
, we obtain @xmath1363 ) } \\ & \leq c \bigl(\vert \pi_{02}(\nabla_s\hx^k)|_{t=\delta_k r_k}\vert_{{\mathcal{c}}^0([-\frac 12,\frac 12 ] ) } + \vert \nabla_s(\xi_1^k-\xi_1'^k)|_{t=\delta_k r_k } \vert_{{\mathcal{c}}^0([-\frac 12,\frac 12 ] ) } \bigr ) \\ & \leq c \bigl(\vert \nabla_s\xi'^k_{02}|_{t=0 } \vert_{{\mathcal{c}}^0([-\frac 12,\frac 12 ] ) } + 3 \vert \nabla_s\hat\xi^k|_{t=\delta_k r_k } - \nabla_s \hat\xi^k|_{t=0 } \vert_{{\mathcal{c}}^0([-\frac 12,\frac 12 ] ) } \bigr ) \to 0 .\end{aligned}\ ] ] combining @xmath1364)}\to 0 $ ] with ( [ hatxik ] ) then proves @xmath1365\times[0,\delta_k r_k])}\to 0 $ ] and thus @xmath1366 in contradiction to the assumption .
[ hofer ] let @xmath1367 be a metric space , @xmath1368 topological spaces , @xmath1369 a continuous map , and @xmath1370 a non - negative continuous function .
fix @xmath1371 for some @xmath1372 and @xmath1373 .
suppose that @xmath1374 is complete for each @xmath1375 .
then there exists an @xmath1376 and a positive number @xmath1377 such that @xmath1378 otherwise , the same argument as in the proof of hofer s lemma @xcite shows that there exists a sequence @xmath1379 such that @xmath1380 after passing to a subsequence , we obtain a cauchy sequence @xmath1381 in some @xmath1382 with @xmath1383 , which contradicts completeness and continuity of @xmath1384 . | we prove an isomorphism of floer cohomologies under geometric composition of lagrangian correspondences in exact and monotone settings . |
adaptive network consists of a collection of agents that are interconnected to each other and solve distributed estimation and inference problems in a collaborative manner .
two useful strategies that enable adaptation and learning over such networks in real - time are the incremental strategy @xcite and the diffusion strategy @xcite .
incremental strategies rely on the use of a hamiltonian cycle , i.e. , a cyclic path that covers all nodes in the network , which is generally difficult to enforce since determining a hamiltonian cycle is an np - hard problem .
in addition , cyclic trajectories are not robust to node or link failure . in comparison ,
diffusion strategies are scalable , robust , and able to match well the performance of incremental networks . in adaptive diffusion implementations
, information is processed locally at the nodes and then diffused in real - time across the network .
diffusion strategies were originally proposed in @xcite and further extended and studied in @xcite .
they have been applied to model self - organized and complex behavior encountered in biological networks , such as fish schooling @xcite , bird flight formations @xcite , and bee swarming @xcite .
diffusion strategies have also been applied to online learning of gaussian mixture models @xcite and to general distributed optimization problems @xcite .
there have also been several useful works in the literature on distributed consensus - type strategies , with application to multi - agent formations and distributed processing @xcite .
the main difference between these works and the diffusion approach of @xcite is the latter s emphasis on the role of adaptation and learning over networks . in the original diffusion least - mean - squares ( lms ) strategy @xcite , the weight estimates that are exchanged among the nodes can be subject to quantization errors and additive noise over the communication links . studying the degradation in mean - square performance that results from these particular perturbations
can be pursued , for both incremental and diffusion strategies , by extending the mean - square analysis already presented in @xcite , in the same manner that the tracking analysis of conventional stand - alone adaptive filters was obtained from the counterpart results in the stationary case ( as explained in ( * ? ? ?
* ch . 21 ) ) .
useful results along these lines , which study the effect of link noise during the exchange of the weight estimates , already appear for the traditional diffusion algorithm in the works @xcite and for consensus - based algorithms in @xcite . in this paper ,
our objective is to go beyond these earlier studies by taking into account additional effects , and by considering a more general algorithmic structure .
the reason for this level of generality is because the analytical results will help reveal which noise sources influence the network performance more seriously , in what manner , and at what stage of the adaptation process .
the results will suggest important remedies and mechanisms to adapt the combination weights in real - time .
some of these insights are hard to get if one focuses solely on noise during the exchange of the weight estimates .
the analysis will further show that noise during the exchange of the regression data plays a more critical role than other sources of imperfection : this particular noise alters the learning dynamics and modes of the network , and biases the weight estimates .
noises related to the exchange of other pieces of information do not alter the dynamics of the network but contribute to the deterioration of the network performance .
to arrive at these results , in this paper , we first consider a generalized analysis that applies to a broad class of diffusion adaptation strategies ( see further ahead ; this class includes the original diffusion strategies and as two special cases ) .
the analysis allows us to account for various sources of information noise over the communication links .
we allow for noisy exchanges during _ each _ of the three processing steps of the adaptive diffusion algorithm ( the two combination steps and and the adaptation step ) . in this way , we are able to examine how the three sets of combination coefficients @xmath0 in influence the propagation of the noise signals through the network dynamics .
our results further reveal how the network mean - square - error performance is dependent on these combination weights . following this line of reasoning
, the analysis leads to algorithms and further ahead for choosing the combination coefficients to improve the steady - state network performance .
it should be noted that several combination rules , such as the metropolis rule @xcite and the maximum degree rule @xcite , were proposed previously in the literature especially in the context of consensus - based iterations @xcite .
these schemes , however , usually suffer performance degradation in the presence of noisy information exchange since they ignore the network noise profile @xcite .
when the noise variance differs across the nodes , it becomes necessary to design combination rules that are aware of this variation as outlined further ahead in section vi - b .
moreover , in a mobile network @xcite where nodes are on the move and where neighborhoods evolve over time , it is even more critical to employ adaptive combination strategies that are able to track the variations in the noise profile in order to cope with such dynamic environments .
this issue is taken up in section vi - c .
we use lowercase letters to denote vectors , uppercase letters for matrices , plain letters for deterministic variables , and boldface letters for random variables .
we also use @xmath1 to denote conjugate transposition , @xmath2 for the trace of its matrix argument , @xmath3 for the spectral radius of its matrix argument , @xmath4 for the kronecker product , and @xmath5 for a vector formed by stacking the columns of its matrix argument .
we further use @xmath6 to denote a ( block ) diagonal matrix formed from its arguments , and @xmath7 to denote a column vector formed by stacking its arguments on top of each other .
all vectors in our treatment are column vectors , with the exception of the regression vectors , @xmath8 , and the associated noise signals , @xmath9 , which are taken to be row vectors for convenience of presentation .
we consider a connected network consisting of @xmath10 nodes .
each node @xmath11 collects scalar measurements @xmath12 and @xmath13 regression data vectors @xmath8 over successive time instants @xmath14 . note that we use parenthesis to refer to the time - dependence of scalar variables , as in @xmath12 , and subscripts to refer to the time - dependence of vector variables , as in @xmath8 .
the measurements across all nodes are assumed to be related to an unknown @xmath15 vector @xmath16 via a linear regression model of the form @xcite : @xmath17 where @xmath18 denotes the measurement or model noise with zero mean and variance @xmath19 .
the vector @xmath16 in denotes the parameter of interest , such as the parameters of some underlying physical phenomenon , the taps of a communication channel , or the location of food sources or predators .
such data models are also useful in studies on hybrid combinations of adaptive filters @xcite .
the nodes in the network would like to estimate @xmath16 by solving the following minimization problem : @xmath20 in previous works @xcite , we introduced and studied several distributed strategies of the diffusion type that allow nodes to cooperate with each other in order to solve problems of the form in an adaptive manner .
these diffusion strategies endow networks with adaptation and learning abilities , and enable information to diffuse through the network in real - time .
we review the adaptive diffusion strategies below . in @xcite ,
two classes of diffusion algorithms were proposed .
one class is the so - called combine - then - adapt ( cta ) strategy : @xmath21\\ \end{aligned}\right.\end{aligned}\ ] ] and the second class is the so - called adapt - then - combine ( atc ) strategy : @xmath22\\ { \bs{w}}_{k , i}&=\sum_{l\in{\mc{n}}_k}a_{2,lk}{\bs{\psi}}_{l , i}\\ \end{aligned}\right.\end{aligned}\ ] ] where the @xmath23 are small positive step - size parameters and the @xmath0 are nonnegative entries of the @xmath24 matrices @xmath25 , respectively .
the coefficients @xmath0 are zero whenever node @xmath26 is not connected to node @xmath11 , i.e. , @xmath27 , where @xmath28 denotes the neighborhood of node @xmath11 .
the two strategies and can be integrated into one broad class of diffusion adaptation @xcite : @xmath29\\ \label{eqn : idealdiffusionpostdiff } { \bs{w}}_{k , i}&=\sum_{l\in{\mc{n}}_k}a_{2,lk}{\bs{\psi}}_{l , i}\end{aligned}\ ] ] several diffusion strategies can be obtained as special cases of through proper selection of the coefficients @xmath0 .
for example , to recover the cta strategy , we set @xmath30 , and to recover the atc strategy , we set @xmath31 , where @xmath32 denotes the @xmath24 identity matrix . in the general diffusion strategy , each node @xmath11 evaluates its estimate @xmath33 at time @xmath34 by relying solely on the data collected from its neighbors through steps and and on its local measurements through step .
the matrices @xmath35 , @xmath36 , and @xmath37 are required to be left or right - stochastic , i.e. , @xmath38 where @xmath39 denotes the @xmath40 vector whose entries are all one .
this means that each node performs a convex combination of the estimates received from its neighbors at every iteration @xmath34 . the mean - square performance and convergence properties of the diffusion algorithm
have already been studied in detail in @xcite . for the benefit of the analysis in the subsequent sections , we present below in the recursion describing the evolution of the weight error vectors across the network
to do so , we introduce the error vectors : @xmath41 and substitute the linear model into the adaptation step to find that @xmath42 where the @xmath43 matrix @xmath44 and the @xmath15 vector @xmath45 are defined as : @xmath46 we further collect the various quantities across all nodes in the network into the following block vectors and matrices : @xmath47 then , from , , and , the recursion for the network error vector @xmath48 is given by @xmath49 where @xmath50 each of the steps in involves the sharing of information between node @xmath11 and its neighbors .
for example , in the first step , all neighbors of node @xmath11 send their estimates @xmath51 to node @xmath11 .
this transmission is generally subject to additive noise and possibly quantization errors .
likewise , steps and involve the sharing of other pieces of information with node @xmath11 .
these exchange steps can all be subject to perturbations ( such as additive noise and quantization errors )
. one of the objectives of this work is to analyze the _ aggregate _ effect of these perturbations on general diffusion strategies of the type and to propose choices for the combination weights in order to enhance the mean - square performance of the network in the presence of these disturbances . to node @xmath11 . ]
so let us examine what happens when information is exchanged over links with additive noise .
we model the data received by node @xmath11 from its neighbor @xmath26 as @xmath52 where @xmath53 and @xmath54 are @xmath15 noise signals , @xmath9 is a @xmath13 noise signal , and @xmath55 is a scalar noise signal ( see fig . [
fig : noise ] ) . observe further that in , we are including several sources of information exchange noise . in comparison ,
references @xcite only considered the noise source @xmath53 in and one set of combination coefficients @xmath56 ; the other coefficients were set to @xmath57 for @xmath58 and @xmath59 . in other words , these references only considered and the following traditional cta strategy without exchange of the data @xmath60 compare with ; note that the second step in only uses @xmath61 : @xmath62\\ \end{aligned}\right.\end{aligned}\ ] ] the analysis that follows examines the aggregate effect of all four noise sources appearing in , in addition to the three sets of combination coefficients appearing in . we introduce the following assumption on the statistical properties of the measurement data and noise signals .
[ asm : all ] 1 .
the regression data @xmath8 are temporally white and spatially independent random variables with zero mean and covariance matrix @xmath63 .
2 . the noise signals @xmath64 , @xmath53 , @xmath55 , @xmath9 , and @xmath54 are temporally white and spatially independent random variables with zero mean and ( co)variances @xmath19 , @xmath65 , @xmath66 , @xmath67 , and @xmath68 , respectively .
in addition , @xmath65 , @xmath66 , @xmath67 , and @xmath68 are all zero if @xmath27 or @xmath69 .
3 . the regression data @xmath70 , the model noise signals @xmath71 , and the link noise signals @xmath72 , @xmath73 , @xmath74 , and @xmath75 are mutually - independent random variables for all @xmath76 and @xmath77 .
using the perturbed data , the diffusion algorithm becomes @xmath78\\ \label{eqn : noisydiffusionpostdiffold } { \bs{w}}_{k , i}&\!=\!\sum_{l\in{\mc{n}}_k}a_{2,lk}{\bs{\psi}}_{lk , i}\end{aligned}\ ] ] where we continue to use the symbols @xmath79 to avoid an explosion of notation . from and ,
expressions can be rewritten as @xmath80\\ \label{eqn : noisydiffusionpostdiff } { \bs{w}}_{k , i}&\!=\!\sum_{l\in{\mc{n}}_k}a_{2,lk}{\bs{\psi}}_{l , i}\!+\!{\bs{v}}_{k , i}^{(\psi)}\end{aligned}\ ] ] where we are introducing the symbols @xmath81 and @xmath82 to denote the aggregate @xmath15 zero - mean noise signals defined over the neighborhood of node @xmath11 : @xmath83 with covariance matrices @xmath84 it is worth noting that @xmath85 and @xmath86 depend on the combination coefficients @xmath56 and @xmath87 , respectively .
this property will be taken into account when optimizing over @xmath56 and @xmath87 in a later section .
we further introduce the following scalar zero - mean noise signal : @xmath88 for @xmath89 , whose variance is @xmath90 to unify the notation , we define @xmath91 .
then , from , , and , it is easy to verify that the noisy data @xmath92 are related via @xmath93 for @xmath94 .
continuing with the adaptation step and substituting , we get @xmath95\end{aligned}\ ] ] then , we can derive the following error recursion for node @xmath11 ( compare with ):
@xmath96 where the @xmath43 matrix @xmath97 and the @xmath98 vector @xmath99 are defined as ( compare with and ): @xmath100 we further introduce the block vectors and matrices : @xmath101 and the corresponding covariance matrices for @xmath102 and @xmath103 : @xmath104 then , from , , and , we arrive at the following recursion for the network weight error vector in the presence of noisy information exchange : @xmath105\!-\!{\bs{v}}_i^{(\psi)}\nonumber\\ { } & = { \mc{a}}_2^{\t}\left[(i_{nm}\!-\!{\mc{m}}{\bs{\mc{r}}}_i')({\mc{a}}_1^{\t}{\wt{\bs{w}}}_{i-1}\!-\!{\bs{v}}_{i-1}^{(w ) } ) \!-\!{\mc{m}}{\bs{z}}_i\right]\!-\!{\bs{v}}_i^{(\psi)}\end{aligned}\ ] ] that is , @xmath106 compared to the previous error recursion , the noise terms in consist of three parts : * @xmath107 is contributed by the noise introduced at the information - exchange step _ before _ adaptation .
* @xmath108 is contributed by the noise introduced at the adaptation step .
* @xmath103 is contributed by the noise introduced at the information - exchange step _ after _ adaptation .
given the weight error recursion , we are now ready to study the mean convergence condition for the diffusion strategy in the presence of disturbances during information exchange under assumption [ asm : all ] . taking expectations of both sides of ,
we get @xmath109 where @xmath110 from , , , and , it can be verified that @xmath111 whereas , from and assumption [ asm : all ] , we get @xmath112\nonumber\\ { } & = -\left(\sum_{l\in{\mc{n}}_k}c_{lk}r_{v , lk}^{(u)}\right)w^o\end{aligned}\ ] ] let us define an @xmath113 matrix @xmath114 that collects all covariance matrices @xmath115 , @xmath116 , weighted by the corresponding combination coefficients @xmath117 , such that its @xmath118th @xmath43 submatrix is @xmath119 . note that @xmath114 itself is _ not _ a covariance matrix because @xmath120 for all @xmath11 .
then , from and , we arrive at @xmath121 therefore , using and , expression becomes @xmath122 with a driving term due to the presence of @xmath123 .
this driving term would disappear from if there were no noise during the exchange of the regression data . to guarantee convergence of , the coefficient matrix @xmath124 must be stable , i.e. , @xmath125 . since @xmath126 and @xmath127 are right - stochastic matrices , it can be shown that the matrix @xmath124 is stable whenever @xmath128 itself is stable ( see appendix [ app : meanconvergence ] ) .
this fact leads to an upper bound on the step - sizes @xmath23 to guarantee the convergence of @xmath129 to a steady - state value , namely , we must have @xmath130 for @xmath131 , where @xmath132 denotes the largest eigenvalue of its matrix argument .
note that the neighborhood covariance matrix @xmath133 in is related to the combination weights @xmath117 .
if we further assume that @xmath37 is doubly - stochastic , i.e. , @xmath134 then , by jensen s inequality @xcite , @xmath135 since ( i ) @xmath132 coincides with the induced @xmath136-norm for any positive semi - definite hermitian matrix ; ( ii ) matrix norms are convex functions of their arguments @xcite ; and ( iii ) by , @xmath117 are convex combination coefficients .
thus , we obtain a sufficient condition for the convergence of in lieu of : @xmath137 } } \ ] ] for @xmath131 , where the upper bound for the step - size @xmath138 becomes independent of the combination weights @xmath117 .
this bound can be determined solely from knowledge of the covariances of the regression data and the associated noise signals that are accessible to node @xmath11 .
it is worth noting that for traditional diffusion algorithms where information is perfectly exchanged , condition reduces to @xmath139}\end{aligned}\ ] ] for @xmath131 . comparing with
, we see that the link noise @xmath9 over regression data reduces the dynamic range of the step - sizes for mean stability . now ,
under , and taking the limit of as @xmath140 , we find that the mean error vector will converge to a steady - state value @xmath141 : @xmath142
it is well - known that studying the mean - square convergence of a single adaptive filter is a challenging task , since adaptive filters are nonlinear , time - variant , and stochastic systems .
when a network of adaptive nodes is considered , the complexity of the analysis is compounded because the nodes now influence each other s behavior . in order to make the performance analysis more tractable , we rely on the energy conservation approach @xcite , which was used successfully in @xcite to study the mean - square performance of diffusion strategies under perfect information exchange conditions .
that argument allows us to derive expressions for the mean - square - deviation ( msd ) and the excess - mean - square - error ( emse ) of the network by analyzing how energy ( measured in terms of error variances ) flows through the nodes . from recursion and under assumption
[ asm : all ] , we can obtain the following weighted variance relation for the global error vector @xmath48 : @xmath143\}\\ { } & \quad+\e\|{\mc{a}}_2^{\t}(i_{nm}\!-\!{\mc{m}}{\bs{\mc{r}}}_i'){\bs{v}}_{i-1}^{(w)}\|_{\sigma}^2+\e\|{\bs{v}}_i^{(\psi)}\|_{\sigma}^2 \end{aligned } } \ ] ] where @xmath144 is an arbitrary @xmath113 positive semi - definite hermitian matrix that we are free to choose .
moreover , the notation @xmath145 stands for the quadratic term @xmath146 .
the weighting matrix @xmath147 in can be expressed as @xmath148 where @xmath124 is given by and @xmath149 denotes a term on the order of @xmath150 . evaluating the term @xmath149 requires knowledge of higher - order statistics of the regression data and link noises , which are not available under current assumptions .
however , this term becomes negligible if we introduce a small step - size assumption .
[ asm : smallstepsize ] the step - sizes are sufficiently small , i.e. , @xmath151 , such that terms depending on higher - order powers of the step - sizes can be ignored .
hence , in the sequel we use the approximation : @xmath152 observe that on the right - hand side ( rhs ) of relation , only the first and third terms relate to the error vector @xmath153 . by assumption
[ asm : all ] , the error vector @xmath153 is independent of @xmath154 and @xmath155 .
thus , from , the third term on rhs of can be expressed as @xmath156\cdot\e{\wt{\bs{w}}}_{i-1}\}\nonumber\\ { } & = -2\,{\mathfrak{re}}(z^*{\mc{m}}{\mc{a}}_2\sigma{\mc{a}}_2^{\t}{\mc{a}}_1^{\t}\cdot\e{\wt{\bs{w}}}_{i-1})+o({\mc{m}}^2)\end{aligned}\ ] ] since we already showed in the previous section that @xmath157 converges to a fixed bias @xmath141 , quantity will converge to a fixed value as well when @xmath140 . moreover , under assumption [ asm : all ] , the second , fourth , and fifth terms on rhs of relation are all fixed values .
therefore , the convergence of relation depends on the behavior of the first term @xmath158 .
although the weighting matrix @xmath147 of @xmath153 is different from the weighting matrix @xmath144 of @xmath159 , it turns out that the entries of these two matrices are approximately related by a linear equation shown ahead in .
introduce the vector notation @xcite : @xmath160 then , by using the identity @xmath161 , it can be verified from that @xmath162 where the @xmath163 matrix @xmath164 is given by @xmath165 to guarantee mean - square convergence of the algorithm , the step - sizes should be sufficiently small and selected to ensure that the matrix @xmath164 is stable @xcite , i.e. , @xmath166 , which is equivalent to the earlier condition @xmath125 . although more specific conditions for mean - square stability can be determined without assumption [ asm : smallstepsize ] @xcite , it is sufficient for our purposes here to conclude that the diffusion strategy is stable in the mean and mean - square senses if the step - sizes @xmath23 satisfy or and are sufficiently small .
the conclusion so far is that sufficiently small step - sizes ensure convergence of the diffusion strategy in the mean and mean - square senses , even in the presence of exchange noises over the communication links .
let us now determine expressions for the error variances in steady - state .
we start from the weighted variance relation . in view of
, it shows that the error variance @xmath167 depends on the mean error @xmath129 .
we already determined the value of @xmath168 in .
we continue to use the vector notation and proceed to evaluate all the terms , except the first one , on rhs of in the following . for the _ second _ term
, it can be expressed as @xmath169^*\sigma\end{aligned}\ ] ] where we used the identity @xmath170^*\sigma$ ] for any hermitian matrix @xmath171 , and @xmath172 denotes the autocorrelation matrix of @xmath173 .
it is shown in appendix [ app : rz ] that @xmath172 is given by @xmath174 where @xmath175 is defined in , @xmath123 is in , and @xmath176 are two @xmath113 positive semi - definite block diagonal matrices : @xmath177\end{aligned}\ ] ] from expression and assumption [ asm : smallstepsize ] , the _ third _ term on rhs of is given by @xmath178\sigma\}\nonumber\\ { } & = -\!\left[\vec\left({\mc{a}}_2^{\t}{\mc{a}}_1^{\t}(\e{\wt{\bs{w}}}_{i-1})z^*{\mc{m}}{\mc{a}}_2 + { \mc{a}}_2^{\t}{\mc{m}}z(\e{\wt{\bs{w}}}_{i-1})^*{\mc{a}}_1{\mc{a}}_2\right)\right]^*\sigma\end{aligned}\ ] ] likewise , the _ fourth _ term on rhs of is approximated by @xmath179\right)\right]^*\sigma\nonumber\\ { } & \approx\left[\vec({\mc{a}}_2^{\t}{\mc{r}}_v^{(w)}{\mc{a}}_2)\right]^*\sigma\end{aligned}\ ] ] where we are now ignoring terms on the order of @xmath180 and @xmath150 .
the _ fifth _
term on rhs of is given by @xmath181^*\sigma\end{aligned}\ ] ] let us introduce @xmath182 at steady - state , as @xmath140 , by and , the weighted variance relation becomes @xmath183^*\sigma\end{aligned}\ ] ] where we are using the compact notation @xmath184 to refer to @xmath185 doing so allows us to represent @xmath147 by the more compact relation @xmath186 on rhs of ; we shall be using the weighting matrix @xmath144 and its vector representation @xmath187 interchangeably for ease of notation ( likewise , for @xmath147 and @xmath188 ) .
the steady - state weighted variance relation can be rewritten as @xmath189^*\sigma\end{aligned}\ ] ] where the term @xmath190 is contributed by the model noise @xmath191 while the remaining terms @xmath192 are contributed by the link noises @xmath193 .
recall that we are free to choose @xmath144 and , hence , @xmath187 .
let @xmath194 , where @xmath195 is another arbitrary positive semi - definite hermitian matrix .
then , we arrive at the following theorem .
[ lemma : steadystatevariancerelation ] under assumptions [ asm : all ] and [ asm : smallstepsize ] , for any positive semi - definite hermitian matrix @xmath195 , the steady - state weighted error variance relation of the diffusion strategy is approximately given by @xmath196^*(i_{n^2m^2}-{\mc{f}})^{-1}\vec(\omega ) \end{aligned } } \ ] ] where @xmath197 is given in , @xmath198 in , @xmath199 in , and @xmath164 in .
each subvector of @xmath200 corresponds to the estimation error at a particular node , say , @xmath201 for node @xmath11 .
the network msd is defined as @xcite : @xmath202 since we are free to choose @xmath195 , we select it as @xmath203 . then , expression gives @xmath204^*(i_{n^2m^2}-{\mc{f}})^{-1}\vec(i_{nm } )
\end{aligned } } \ ] ] similarly , if we instead select @xmath205 , where @xmath206 then expression would allow us to evaluate the network emse as : @xmath207^*(i_{n^2m^2}-{\mc{f}})^{-1}\vec({\mc{r}}_u ) \end{aligned } } \ ] ] where , under assumption [ asm : all ] , the network emse is given by @xmath208 we showed in the earlier sections that the link noise over regression data biases the weight estimators . in this section
we examine how the results simplify when there is no sharing of regression data among the nodes .
[ asm : nodatasharing ] nodes do not share regression data within neighborhoods , i.e. , assume @xmath209 . by assumptions [ asm : smallstepsize ] and [ asm : nodatasharing ] , matrices @xmath210 in , , and become @xmath211 where @xmath212 is given in .
then , the network msd and emse expressions and simplify to : @xmath213^ * ( i_{n^2m^2}-{\mc{f}})^{-1}\vec(i_{nm } ) \end{aligned } } \ ] ] and @xmath214^ * ( i_{n^2m^2}-{\mc{f}})^{-1}\vec({\mc{r}}_u ) \end{aligned } } \ ] ] recalling that @xmath198 and @xmath164 are related to the combination matrices @xmath215 , or , equivalently , @xmath216 , results and express the network msd and emse in terms of @xmath216 .
however , it is generally difficult to use these expressions to optimize over @xmath216 to reduce the impact of link noise . instead , by substituting into and using the fact that @xmath164 is stable
, we can arrive at another useful expression for the network msd : @xmath217^ * \sum_{j=0}^{\infty}{\mc{f}}^j\vec(i_{nm})\nonumber\\ { } & = \frac{1}{n}\!\left[\vec({\mc{a}}_2^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}_2\!+\!{\mc{r}}_v)\right]^ * \sum_{j=0}^{\infty}({\mc{b}}^{\t}\otimes{\mc{b}}^*)^j\vec(i_{nm})\nonumber\\ { } & = \frac{1}{n}\!\left[\vec({\mc{a}}_2^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}_2\!+\!{\mc{r}}_v)\right]^ * \sum_{j=0}^{\infty}\vec({\mc{b}}^{*j}{\mc{b}}^j)\end{aligned}\ ] ] that is , @xmath218 } \ ] ] where @xmath124 is given in .
similarly , the network emse can be expressed as @xmath219 } \ ] ] expressions and reveal in an interesting way how the noise sources originating from any particular node end up influencing the overall network performance .
let us denote @xmath220 the error recursion can be rewritten as @xmath221 where @xmath222 then , @xmath223 under assumption [ asm : nodatasharing ] , @xmath224 in and can be simplified as @xmath225 where @xmath226 are given in and . by assumption [ asm : all ] , @xmath224 are temporally independent for different @xmath34 and @xmath227 where @xmath124 is given by . as @xmath140 ,
the first term on rhs of becomes @xmath228\right\}\nonumber\\ { } & \stackrel{(a)}{\approx}\lim_{i\rightarrow\infty}\tr\left[\left(\e{\bm{\phi}}_{0,i}\right ) ( \e{\wt{\bs{w}}}_{-1}{\wt{\bs{w}}}_{-1}^*)\left(\e{\bm{\phi}}_{0,i}\right)^*\right]\nonumber\\ { } & = \lim_{i\rightarrow\infty}\tr\left[{\mc{b}}^{i+1}(\e\,{\wt{\bs{w}}}_{-1}{\wt{\bs{w}}}_{-1}^*){\mc{b}}^{(i+1)*}\right]\nonumber\\ { } & \stackrel{(b)}{=}0\end{aligned}\ ] ] where ( a ) is obtained by approximating the expectation of the product by the product of expectations and ( b ) is due to the stability of @xmath124 .
therefore , the steady - state value of gives @xmath229\nonumber\\ { } & \stackrel{(a)}{\approx}\lim_{i\rightarrow\infty}\sum_{m=0}^{i}\tr\left[{\mc{b}}^{i - m}({\mc{a}}_2^{\t}{\mc{m}}{\mc{s } } { \mc{m}}{\mc{a}}_2+{\mc{r}}_v){\mc{b}}^{(i - m)*}\right]\nonumber\\ { } & \stackrel{(b)}{=}\lim_{i\rightarrow\infty}\sum_{j=0}^{i}\tr\left[{\mc{b}}^{j}({\mc{a}}_2^{\t}{\mc{m}}{\mc{s } } { \mc{m}}{\mc{a}}_2+{\mc{r}}_v){\mc{b}}^{j*}\right]\nonumber\\ { } & = \sum_{j=0}^{\infty}\tr\left[{\mc{b}}^{j}({\mc{a}}_2^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}_2+{\mc{r}}_v){\mc{b}}^{*j}\right]\end{aligned}\ ] ] where , by and , ( a ) is due to @xmath230 and ( b ) is simply a change of variable : @xmath231 . since the @xmath232th term of the summation in or is contributed by the term @xmath233 , which consists of all the noise sources at time @xmath234 , expression shows how various sources of noises are involved and how they contribute to the network msd .
before we optimize the combination matrices @xmath216 , we first specialize the msd expression and the emse expression for the atc and cta algorithms . for the atc algorithm , we set @xmath31 and @xmath235 , and for the cta algorithm , we set @xmath236 and @xmath30 .
let us denote @xmath237 then , we get @xmath238\!\!\\ \label{eqn : noisyatcemse } \!\!{\textrm{emse}}_{\textrm{atc}}&\!\approx\!\frac{1}{n}\!\sum_{j=0}^{\infty}{\mathrm{tr}}\!\left[{\mc{b}}_{\textrm{atc}}^j ( { \mc{a}}^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}\!+\!{\mc{r}}_v^{(\psi ) } ) { \mc{b}}_{\textrm{atc}}^{*j}{\mc{r}}_u\right]\!\!\end{aligned}\ ] ] and @xmath239\\ \label{eqn : noisyctaemse } { \textrm{emse}}_{\textrm{cta}}&\approx\frac{1}{n}\sum_{j=0}^{\infty}{\mathrm{tr}}\left[{\mc{b}}_{\textrm{cta}}^j ( { \mc{m}}{\mc{s}}{\mc{m}}+{\mc{r}}_v^{(w)}){\mc{b}}_{\textrm{cta}}^{*j}{\mc{r}}_u\right]\end{aligned}\ ] ] minimizing the msd expression or the emse expression for the atc algorithm over left - stochastic matrices @xmath240 is generally nontrivial .
we pursue an approximate solution that relies on optimizing an upper bound and performs well in practice .
let us use @xmath241 to denote the nuclear norm ( also known as the trace norm , or the ky fan @xmath242-norm ) of matrix @xmath243 @xcite , which is defined as the sum of the singular values of @xmath243 .
therefore , @xmath244 for any @xmath243 and @xmath245 when @xmath243 is hermitian and positive semi - definite .
let us also denote @xmath246 as the block maximum norm of matrix @xmath243 ( see appendix [ app : meanconvergence ] ) .
then , @xmath247 & = \big\|{\mc{b}}_{\textrm{atc}}^j({\mc{a}}^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a } } + { \mc{r}}_v^{(\psi)}){\mc{b}}_{\textrm{atc}}^{*j}\big\|_*\nonumber\\ { } & \le\|{\mc{b}}_{\textrm{atc}}^j\|_*\cdot\|{\mc{a}}^{\t}{\mc{m}}{\mc{s } } { \mc{m}}{\mc{a}}+{\mc{r}}_v^{(\psi)}\|_*\cdot\|{\mc{b}}_{\textrm{atc}}^{*j}\|_*\nonumber\\ { } & \le c^2\cdot\|{\mc{b}}_{\textrm{atc}}^j\|_{b,\infty}^2\cdot { \mathrm{tr}}({\mc{a}}^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}+{\mc{r}}_v^{(\psi)})\nonumber\\ { } & \le c^2\cdot\|{\mc{b}}_{\textrm{atc}}\|_{b,\infty}^{2j}\cdot { \mathrm{tr}}({\mc{a}}^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}+{\mc{r}}_v^{(\psi)})\nonumber\\ { } & \le c^2\cdot(\|{\mc{a}}\|_{b,\infty}\cdot\|i_{nm}-{\mc{m}}{\mc{r}}_u\|_{b,\infty})^{2j } { \mathrm{tr}}({\mc{a}}^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}+{\mc{r}}_v^{(\psi)})\nonumber\\ { } & = c^2\cdot\rho(i_{nm}-{\mc{m}}{\mc{r}}_u)^{2j}\cdot{\mathrm{tr } } ( { \mc{a}}^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}+{\mc{r}}_v^{(\psi)})\end{aligned}\ ] ] where @xmath248 is some positive scalar such that @xmath249 because @xmath241 and @xmath246 are submultiplicative norms and all such norms are equivalent @xcite . in the last step of we used lemmas [ lemma : rghtstocmat ] and [ lemma : blockdiagonal ] from appendix [ app : meanconvergence ] .
thus , we can upper bound the network msd by @xmath250 ^ 2}\end{aligned}\ ] ] where the combination matrix @xmath251 appears only in the numerator .
the result motivates us to consider instead the problem of minimizing the upper bound , namely , @xmath252 using and , the cost function in can be expressed as @xmath253\end{aligned}\ ] ] problem can therefore be decoupled into @xmath10 separate optimization problems of the form : @xmath254\\ \st & \;\ ; \sum_{l\in{\mc{n}}_k}a_{lk}=1,\;\;a_{lk}\ge0,\;\;a_{lk}=0\;\mbox{if}\;l\notin{\mc{n}}_k \\ \end{aligned } } \ ] ] for @xmath255 . with each node @xmath94 , we associate the following nonnegative _ variance product _ measure : @xmath256 this measure incorporates information about the link noise covariances @xmath257 .
the solution of is then given by @xmath258 we refer to this combination rule as the relative variance combination rule ; it is an extension of the rule devised in @xcite to the case of noisy information exchanges . in particular , the definition of the scalars @xmath259 in is different and now depends on both subscripts @xmath26 and @xmath11 . minimizing the emse expression for the atc algorithm over left - stochastic matrices
@xmath240 can be pursued in a similar manner by noting that @xmath260\le c^2[\rho(i_{nm}\!-\!{\mc{m}}{\mc{r}}_u)]^{2j}\ , { \mathrm{tr}}({\mc{a}}^{\t}{\mc{m}}{\mc{s}}{\mc{m}}{\mc{a}}\!+\!{\mc{r}}_v^{(\psi)})\,{\mathrm{tr}}({\mc{r}}_u)\end{aligned}\ ] ] thus , minimizing the upper bound of the network emse leads to the same solution . using the same argument
, we can also show that the same result minimizes the upper bound of the network msd or emse for the cta algorithm . to apply the relative variance combination rule
, each node @xmath11 needs to know the variance products , @xmath259 , of their neighbors , which in general are not available since they require knowledge of the quantities @xmath261 .
therefore , we now propose an adaptive combination rule by using data that are available to the individual nodes . for the atc algorithm , we first note from and that @xmath262 for @xmath89 .
since the algorithm converges in the mean and mean - square senses under assumption [ asm : smallstepsize ] , all the estimates @xmath263 tend close to @xmath16 as @xmath140 .
this allows us to estimate @xmath264 for node @xmath11 by using instantaneous realizations of @xmath265 , where we replace @xmath51 by @xmath266 .
similarly , for node @xmath11 itself , we can use realizations of @xmath267 to estimate @xmath268 . to unify the notation
, we define @xmath269 .
let @xmath270 denote an estimator for @xmath264 that is computed by node @xmath11 at time @xmath34 .
then , one way to evaluate @xmath270 is through the recursion : @xmath271 for @xmath272 , where @xmath273 is a forgetting factor that is usually close to one . in this way
, we arrive at the adaptive combination rule : @xmath274^{-1}}{\sum_{m\in{\mc{n}}_k}[{\widehat{\bm{\gamma}}}_{mk}^{2}(i)]^{-1 } } , & { \textrm{if $ l\in{\mc{n}}_k$ } } \\ 0 , & { \textrm{otherwise } } \\ \end{cases } } \ ] ]
the diffusion strategy is adaptive in nature .
one of the main benefits of adaptation ( by using constant step - sizes ) is that it endows networks with tracking abilities when the underlying weight vector @xmath16 varies with time . in this section
we analyze how well an adaptive network is able to track variations in @xmath16 . to do so
, we adopt a random - walk model for @xmath16 that is commonly used in the literature to describe the non - stationarity of the weight vector @xcite .
[ asm : randomwalk ] the weight vector @xmath16 changes according to the model : @xmath275 where @xmath276 has a constant mean @xmath16 for all @xmath34 , @xmath277 is an i.i.d .
random sequence with zero mean and covariance matrix @xmath278 ; the sequence @xmath277 is independent of the initial conditions @xmath279 and of all regression data and noise signals across the network for all time instants .
we now define the error vector at node @xmath11 as @xmath280 so that the global error recursion ( [ eqn : noisyerrorrecursion1 ] ) for the network is replaced by @xmath281 where the @xmath282 vector @xmath283 is defined as @xmath284 by assumptions [ asm : all ] and [ asm : randomwalk ] , it can be verified that the condition for mean convergence continues to be @xmath285 , where @xmath124 is defined in .
in addition , it can also be verified that the error recursion converges in the mean sense to the same non - zero bias vector @xmath141 as in . from and under assumption [ asm : smallstepsize ] , we can derive the weighted variance relation : @xmath286\}\nonumber\\ { } & \qquad+{\mathbb{e}}\|{\mc{a}}_2^{\t}(i_{nm}-{\mc{m}}{\bs{\mc{r}}}_i'){\bs{v}}_{i-1}^{(w)}\|_{\sigma}^2\nonumber\\ { } & \qquad+{\mathbb{e}}\|{\mc{a}}_2^{\t}{\mc{m}}{\bs{z}}_i\|_{\sigma}^2+{\mathbb{e}}\|{\bs{v}}_i^{(\psi)}\|_{\sigma}^2\end{aligned}\ ] ] where @xmath164 is given in .
if the step - sizes are sufficiently small , then we can assume that the network continues to be mean - square stable .
the steady - state performance is affected by the non - stationarity of @xmath16 . from assumption
[ asm : smallstepsize ] , at steady - state , expression becomes @xmath287^*(i_{n^2m^2}\!-\!{\mc{f}})^{-1}{\mathrm{vec}}(\omega)\end{aligned}\ ] ] where @xmath197 is given in , @xmath288 in , @xmath199 in , @xmath164 in , and @xmath289 is the covariance matrix of @xmath283 : @xmath290 by , , and , we get @xmath291 then , following the same argument that led to , we find that the network msd is now given by : @xmath292^*(i_{n^2m^2}-{\mc{f}})^{-1}\vec(i_{nm } ) \end{aligned } } \ ] ] similarly , the network emse is given by : @xmath293^*(i_{n^2m^2}-{\mc{f}})^{-1}\vec({\mc{r}}_u ) \end{aligned } } \ ] ] where @xmath212 is defined in .
observe that the main difference relative to and is the addition of the term @xmath289 .
therefore , all the results that were derived in the earlier section , such as and , continue to hold by adding @xmath289 .
in particular , if assumptions [ asm : smallstepsize ] and [ asm : nodatasharing ] are adopted , expressions and can be approximated as @xmath294^*(i_{n^2m^2}-{\mc{f}})^{-1}\vec(i_{nm } ) \end{aligned } } \ ] ] and @xmath295^*(i_{n^2m^2}-{\mc{f}})^{-1}\vec({\mc{r}}_u ) \end{aligned } } \ ] ] where @xmath198 is now given in .
we simulate two scenarios : noisy information exchanges and non - stationary environments . we consider a connected network with @xmath296 nodes .
the network topology is shown in fig .
[ fig : topology ] .
the unknown complex parameter @xmath16 of length @xmath297 is randomly generated ; its value is @xmath298 $ ] .
we adopt uniform step - sizes , @xmath299 , and uniformly white gaussian regression data with covariance matrices @xmath300 , where @xmath301 are shown in fig . [
fig : regressor ] .
the variances of the model noises , @xmath302 , are randomly generated and shown in fig .
[ fig : modelnoise ] .
we also use white gaussian link noise signals such that @xmath303 , @xmath304 , and @xmath305 .
all link noise variances , @xmath306 , are randomly generated and illustrated in fig .
[ fig : linknoise ] from top to bottom .
we assign the link number by the following procedure .
we denote the link from node @xmath26 to node @xmath11 as @xmath307 , where @xmath308 .
then , we collect the links @xmath309 in an ascending order of @xmath26 in the list @xmath310 ( which is a set with _ ordered _ elements ) for each node @xmath11 .
for example , for node @xmath311 in fig .
[ fig : topology ] , it has @xmath312 links ; the ordered links are then collected in @xmath313 .
we concatenate @xmath314 in an ascending order of @xmath11 to get the overall list @xmath315 .
eventually , the @xmath316th link in the network is given by the @xmath316th element in the list @xmath317 .
nodes.,height=172 ] we examine the simplified cta and atc algorithms in and , namely , no sharing of data among nodes ( i.e. , @xmath209 ) , under various combination rules : ( i ) the relative variance rule in , ( ii ) the metropolis rule in @xcite : @xmath318 where @xmath319 denotes the degree of node @xmath11 ( including the node itself ) , ( iii ) the uniform weighting rule : @xmath320 and ( iv ) the adaptive rule in with @xmath321 .
we plot the network msd and emse learning curves for atc algorithms in figs .
[ fig : msd_atc ] and [ fig : emse_atc ] by averaging over 50 experiments . for cta algorithms , we plot their network msd and emse learning curves in figs . [
fig : msd_cta ] and [ fig : emse_cta ] also by averaging over 50 experiments .
moreover , we also plot their theoretical results and in the same figures . from fig .
[ fig : sim2 ] we see that the relative variance rule makes diffusion algorithms achieve the lowest msd and emse levels at steady - state , compared to the metropolis and uniform rules as well as the algorithm from @xcite ( which also requires knowledge of the noise variances ) .
in addition , the adaptive rule attains msd and emse levels that are only slightly larger than those of the relative variance rule , although , as expected , it converges slower due to the additional learning step . the value for each entry of the complex parameter @xmath322 is assumed to be changing over time along a circular trajectory in the complex plane , as shown in fig .
[ fig : track ] .
the dynamic model for @xmath323 is expressed as @xmath324 , where @xmath325 , @xmath326 , and @xmath327 .
the covariance matrices @xmath328 are randomly generated such that @xmath329 when @xmath330 , but their traces are normalized to be one , i.e. , @xmath331 , for all nodes
. the variances for the model noises , @xmath332 , are also randomly generated .
we examine two different scenarios : the low noise - level case where the average noise variance across the network is @xmath333 db and the noise variances are shown in fig .
[ fig : hsnr ] ; and the high noise - level case where the average variance is @xmath334 db and the variances are shown in fig .
[ fig : lsnr ] .
we simulate 3000 iterations and average over 20 experiments in figs .
[ fig : track_hsnr ] and [ fig : track_lsnr ] for each case .
the step - size is 0.01 and uniform across the network . for simplicity
, we adopt the simplified atc algorithm where @xmath209 , and only use the uniform weighting rule . the tracking behavior of the network , denoted as @xmath335 , is obtained by averaging over all the estimates , @xmath336 , across the network . figs .
[ fig : track_hsnr ] and [ fig : track_lsnr ] depict the complex plane ; the horizontal axis is the real axis and the vertical axis is the imaginary axis .
therefore , for every time @xmath34 , each entry of @xmath323 or @xmath337 represents a point in the plane .
when @xmath34 is increasing , @xmath338 moves along the red trajectory ( in @xmath339 ) , @xmath340 along the blue trajectory ( in @xmath341 ) , @xmath342 along the green trajectory ( in @xmath343 ) , and @xmath344 along the magenta trajectory ( in @xmath345 ) . from fig .
[ fig : track ] , it can be seen that diffusion algorithms exhibit the tracking ability in both high and low noise - level environments .
in this work we investigated the performance of diffusion algorithms under several sources of noise during information exchange and under non - stationary environments .
we first showed that , on one hand , the link noise over the regression data biases the estimators and deteriorates the conditions for mean and mean - square convergence . on the other hand
, diffusion strategies can still stabilize the mean and mean - square convergence of the network with noisy information exchange .
we derived analytical expressions for the network msd and emse and used these expressions to motivate the choice of combination weights that help ameliorate the effect of information - exchange noise and improve network performance .
we also extended the results to the non - stationary scenario where the unknown parameter @xmath16 is changing over time .
simulation results illustrate the theoretical findings and how well they match with theory .
following @xcite , we first define the block maximum norm of a vector .
[ def : vecblkmaxnorm ] given a vector @xmath347 consisting of @xmath10 blocks @xmath348 , the block maximum norm is the real function @xmath349 , defined as @xmath350 where @xmath351 denotes the standard @xmath136-norm on @xmath352 . similarly , we define the matrix norm that is induced by the block maximum norm as follows : [ def : blkmaxnorm ] given a block matrix @xmath353 with block size @xmath43 , then @xmath354 denotes the induced block maximum ( matrix ) norm on @xmath355 .
[ lemma : blkunitaryinvariant ] the block maximum matrix norm is block unitary invariant , i.e. , given a block diagonal unitary matrix @xmath356 consisting of @xmath10 unitary blocks @xmath357 , where @xmath358 , for any matrix @xmath353 , then @xmath359 where @xmath360 denotes the block maximum matrix norm on @xmath355 with block size @xmath43 .
[ lemma : rghtstocmat ] let @xmath361 be a right - stochastic matrix . then , for block size @xmath43 , @xmath362 from definition [ def : blkmaxnorm ] , we get @xmath363_{lk}x_k\|_2}{\max_{k}\|x_k\|_2}\nonumber\\ { } & \le\max_{x\in{\mathbb{c}}^{mn}\backslash\{0\ } } \frac{\max_{l}\sum_{k=1}^{n}[a]_{lk}\|x_k\|_2}{\max_{k}\|x_k\|_2}\nonumber\\ { } & \le\max_{x\in{\mathbb{c}}^{mn}\backslash\{0\ } } \frac{\max_{l}(\sum_{k=1}^{n}[a]_{lk})\cdot\max_k\|x_k\|_2}{\max_{k}\|x_k\|_2}\nonumber\\ { } & \le\max_{x\in{\mathbb{c}}^{mn}\backslash\{0\}}\frac{\max_{l}1\cdot\max_{k}\|x_k\|_2}{\max_{k}\|x_k\|_2}\nonumber\\ { } & = 1\end{aligned}\ ] ] where @xmath364 consists of @xmath10 blocks @xmath348 , and @xmath365_{lk}$ ] denotes the @xmath366th entry of @xmath240 . on the other hand , for any induced matrix norm ,
say , the block maximum norm , it is always lower bounded by the spectral radius of the matrix @xcite : @xmath367 combining and completes the proof .
[ lemma : blockdiagonal ] let @xmath368 be a block diagonal hermitian matrix with block size @xmath43 .
then the block maximum norm of the matrix @xmath251 is equal to its spectral radius , i.e. , @xmath369 denote the @xmath11th @xmath43 submatrix on the diagonal of @xmath251 by @xmath370 .
let @xmath371 be the eigen - decomposition of @xmath370 , where @xmath372 is unitary and @xmath373 is diagonal .
define the block unitary matrix @xmath374 and the diagonal matrix @xmath375 .
then , @xmath376 . by lemma [ lemma : blkunitaryinvariant ] ,
the block maximum norm of @xmath251 with block size @xmath43 is @xmath377 where we used the fact that the induced @xmath136-norm is identical to the spectral radius for hermitian matrices @xcite . on the other hand ,
any matrix norm is lower bounded by the spectral radius @xcite , i.e. , @xmath378 combining and completes the proof .
now we show that the matrix @xmath379 is stable if @xmath128 is stable . for any induced matrix norm , say , the block maximum norm with block size @xmath43
, we have @xcite @xmath380 where , from and , @xmath126 and @xmath127
satisfy lemma [ lemma : rghtstocmat ] . by and
, it is straightforward to see that @xmath128 is block diagonal with block size @xmath43 .
then , by lemma [ lemma : blockdiagonal ] , expression can be further expressed as @xmath381 which completes the proof . in appendix
i of @xcite and the matrix @xmath180 in lemma 2 of @xcite are block diagonal , the @xmath382 norm used in these references should simply be replaced by the @xmath360 norm used here and as already done in @xcite . ]
let us denote the @xmath366th submatrix of @xmath383 by @xmath384 . by assumptions [ asm : all ] and expression ,
@xmath385 can be evaluated as @xmath386 where , by expressions and , @xmath387 when @xmath388 , expression reduces to @xmath389 when @xmath390 , expression becomes @xmath391 where @xmath392 denotes the kronecker delta function .
evaluating the last term on rhs of requires knowledge of the excess kurtosis of @xmath393 , which is generally not available . in order to proceed ,
we invoke a separation principle to approximate it as @xmath394 substituting into leads to @xmath395\end{aligned}\ ] ] from and , we get @xmath396\end{aligned}\ ] ] substituting into , we obtain @xmath397\end{aligned}\ ] ] from and , we arrive at expression .
l. li and j. a. chambers , `` distributed adaptive estimation based on the apa algorithm over diffusion netowrks with changing topology , '' in _ proc .
ieee workshop stat . signal process .
( ssp ) _ , cardiff , uk , aug .
2009 , pp . 757760 .
n. takahashi and i. yamada , `` link probability control for probabilistic diffusion least - mean squares over resource - constrained networks , '' in _ proc .
speech , signal process .
( icassp ) _ , dallas , tx , mar .
2010 , pp . 35183521 .
z. towfic , j. chen , and a. h. sayed , `` collaborative learning of mixture models using diffusion adaptation , '' in _ proc .
workshop machine learn . signal process .
( mlsp ) _ , beijing , china , sept .
2011 , pp .
j. chen and a. h. sayed , `` diffusion adaptation strategies for distributed optimization and learning over networks , '' to appear in _ ieee trans .
signal process .
[ also available online at http://arxiv.org/abs/1111.0034 as _ arxiv:1111.0034v2 [ math.oc]_ , oct . 2011 . ]
r. abdolee and b. champagne , `` diffusion lms algorithms for sensor networks over non - ideal inter - sensor wireless channels , '' in _ proc .
sensor systems ( dcoss ) _ , barcelona , spain , june 2011 , pp .
16 .
l. xiao , s. boyd , and s. lall , `` a scheme for robust distributed sensor fusion based on average consensus , '' in _ proc .
acm / ieee int .
sensor networks ( ipsn ) _ , los angeles , ca , apr .
2005 , pp . 6370 .
d. mandic , p. vayanos , c. boukis , b. jelfs , s. l. goh , t. gautama , and t. rutkowski , `` collaborative adaptive learning using hybrid filters , '' in _ proc .
speech , signal process .
( icassp ) _ , honolulu , hi , apr .
2007 , pp . 921924 .
tu and a. h. sayed , `` optimal combination rules for adaptation and learning over netowrks , '' in _ proc .
workshop comput .
advances multi - sensor adapt . process .
( camsap ) _ , san juan , puerto rico , dec .
2011 , pp . 317320 . | adaptive networks rely on in - network and collaborative processing among distributed agents to deliver enhanced performance in estimation and inference tasks .
information is exchanged among the nodes , usually over noisy links .
the combination weights that are used by the nodes to fuse information from their neighbors play a critical role in influencing the adaptation and tracking abilities of the network .
this paper first investigates the mean - square performance of general adaptive diffusion algorithms in the presence of various sources of imperfect information exchanges , quantization errors , and model non - stationarities . among other results
, the analysis reveals that link noise over the regression data modifies the dynamics of the network evolution in a distinct way , and leads to biased estimates in steady - state .
the analysis also reveals how the network mean - square performance is dependent on the combination weights .
we use these observations to show how the combination weights can be optimized and adapted .
simulation results illustrate the theoretical findings and match well with theory .
diffusion adaptation , adaptive networks , imperfect information exchange , tracking behavior , diffusion lms , combination weights , energy conservation . |
the very notion of _ really existing objects _ , i.e. _ physical objects carrying energy - momentum _ , necessarily implies that all such objects must have definite stability properties , as well as propertiies that do not change with time ; otherwise everything would constantly change and we could not talk about objects and structure at all , moreover no memory and no knowledge would be possible . through definite procedures of _ measurement _ we determine , where and when this is possible , quantitative characteristics of the physical objects .
the characteristics obtained differ in : their nature and qualities ; in their significance to understand the structure of the objects they characterize ; in their abilities to characterize interaction among objects ; and in their universality .
natural objects may be classified according to various principles .
the classical point - like objects ( called usually particles ) are allowed to interact continuously with each other just through exchanging ( through some mediator usually called field ) universal conserved quantities : energy , momentum , angular momentum , so that , the set of objects `` before '' interaction is the same as the set of objects `` after '' interaction , no objects have disapeared and no new objects have appeared , only the conserved quantities have been redistributed .
this is in accordance with the assumption of point - likeness , i.e. particles are assumed to have no internal structure , so they are undestroyable . hence
, classical particles may be subject only to elastic interaction . turning to study the set of microobjects , called usually elementary particles : photons , electrons , etc .
, physicists have found out , in contrast to the case of classical particles , that a given set of microobjects may transform into _ another _ set of microobjects under definite conditions , for example , the well known anihilation process : @xmath5 .
these transformations obey also the energy - momentum and angular+spin momentum conservation , but some features may disapear ( e.g. the electric charge ) and new features ( e.g. motion with the highest velosity ) may appear .
hence , microobjects _ allow to be destroyed _
, so they have structure and , consequently , they would not admit the approximation `` point - like objects '' . in view of this
we may conclude that any theory aiming to describe their behaviour must take in view their structure . in particular , assuming that the planck formula @xmath3 is valid for a free microobject , the only reasonable way to understand where the characteristic _ frequency _ comes from is to assume that this microobject has a periodic dynamical structure .
the idea of conservation as a dynamics generating rule was realized and implemented in physics firstly by newton a few centuries ago in the frame of real ( classical ) bodies : he invented the quantity _ momentum _
, @xmath6 , for an isolated body ( treated as a point , or structureless object ) , as its integral characteristic , postulated its time - constancy ( i.e. conservation ) , and wrote down his famous equation @xmath7=@xmath8 . this equation says , that the integral characteristic _ momentum _ of a given point - like object may ( smoothly ) increase _ only _ if some other object loses ( smoothly ) the same quantity of _ momentum_. the concept of _ force _ , @xmath8 , measures the momentum transfered per unit time .
the conservative quantities _ energy _ and _ angular momentum _ ,
consistent with the momentum conservation , were also appropriately incorporated .
the most _ universal _ among these turned out to be _ the energy _ , since , as far as we know , every natural object carries energy .
this property of universality of energy makes it quite distinguished among the other conservative quantities , because its change may serve as a relyable measure of _ any _ kind of interaction and transformation of real objects .
we especially note , that all these conservative quantities are carried by some object(s ) , _ no energy - momentum can exist without corresponding objects - carryers_. in this sense , the usual words `` energy quanta '' are sensless if the corresponding carryers are not pointed out .
so , theoretical physics started with idealizing the natural objects as particles , i.e. objects _ without structure _ , and the real world was theoreticaly viewed as a collection of _ interacting _ , i.e. energy - momentum exchanging , particles .
as far as the behaviour of the real objects as a whole is concerned , and the interactions considered do not lead to destruction of the bodies - particles , this theoretical model of the real world worked well . the 19th century physics , due mainly to faraday and maxwell , created the theoretical concept of _ electromagnetic field _ as the _ interaction carrying object _ , responsible for the observed mutual influence between distant electrically charged ( point - like ) objects .
this concept presents the electromagnetic field as an extended ( in fact , infinite ) continuous object , having dynamical structure , and although it influences the behaviour of the charged particles , it does not destroy them .
the theory of the electromagnetic field was based also on balance relations of new kind of quantities .
actually , the new concepts of _ flux of a vector field through a 2-dimensional surface _ and _ circulation of a vector field along a closed curve _ were coined and used extensively .
the faraday - maxwell equations in their integral form establish , in fact , where the time - changes of the fluxes of the electric and magnetic fields go to , or come from , in both cases of a _ closed _
2-surface , and of a _ not - closed _
2-surface with a boundary , and in this sense they introduce a kind of _ ballance relations_. we note , that these fluxes are new quantities , specific to the continuous character of the physical object under consideration ; the field equations of faraday - maxwell do not express directly energy - momentum balance relations as the above mentioned newton s law @xmath9 does . nevertheless , they are consistent with energy - momentum conservation , as it is well known .
the corresponding local energy - momentum quantities turn out to be quadratic functions of the electric and magnetic vectors .
although very useful for considerations in finite regions with boundary conditions , the pure field maxwell equations have time - dependent solutions in the whole space that could hardly be considered as mathematical models of really existing fields . as a rule ,
if these solutions are time - stable and not static , they occupy the whole 3-space , or its infinite sub - region ( e.g. plane waves ) , and , hence , they carry _ infinite _ energy and momentum ( infinite objects ) . on the other hand , according to cauchy s theorem for the dalembert wave equation [ 1 ] , which is necessarily satisfied by any component of the vacuum field in maxwell theory , every _ finite _ ( and smooth enough ) initial field configuration is strongly time - unstable : the initial condition blows up radially and goes to infinity , and its forefront and backfront propagate with the velocity of light . hence , faraday - maxwell equations _ can not describe finite time - stable _ localized field - objects .
the inconsistences between theory and experiment appeared in a full scale at the end of the last century and it soon became clear that they were not avoidable in the frame of classical physics .
after planck and einstein created the notion of _ elementary field quanta _
, named later by lewis [ 2 ] _ photon _ , physicists faced the above mentioned problem : the light quanta appeared to be real objects of a new kind , namely , they did not admit the point - like approximation like newton s particles did .
in fact , every photon propagates ( transitionally ) as a whole with a _ constant _ velocity and keeps unchanged the energy - momentum it carries , which _ should _ mean that it is a _
free _ object . on the other hand
, it satisfies planck s relation @xmath3 , which means that the very existence of photons is intrinsically connected with a periodical process of frequency @xmath10 , and periodical processes in classical physics are generated by _
external _ force - fields , which means that the object _ should not _ be free .
the efforts to overcome this undesirable situation resulted in the appearence of quantum theory , quantum electrodynamics was built , but the assumption `` the point - like approximation works '' _ was kept as a building stone _ of the new theory , and this braught the theory to some of the well known singularity problems .
modern theory tries to pay more respect to the view that the point - like approximation does not work in principle in the set of microobjects satisfying the above planck s formula .
in other words , the right theoretical notion for these objects should be that of _ extended continuous finite _ objects , or _
finite field objects_. during the last 30 years physicists have been trying seriously to implement in theory the `` extended point of view '' on microobjects mainly through the string / brane theories , but the difficulties these theories meet , generated partly by the great purposes they set before themselves , still do not allow to get satisfactory results .
of course , attempts to create extended point view on elementary particles different from the string / brane theory approach , have been made and are beeng made these days [ 3 ] .
anyway , we have to admit now , that after one century away from the discovery of planck s formula , we still do not have a complete and satisfactory self - consistent theory of single photons .
so , creation of a self - consistent extended point of view and working out a corresponding theory is still a challenge , and this paper aims to consider such an extended point of view on photons as _ screw solitons _ in the frame of the newly developed eed [ 4 ] .
first we summurize the main features / properties of solitons and photons .
the concept of soliton appears in physics as a nonlinear elaboration - physical and mathematical- of the general notion for exitation in a medium .
it includes the following features : 0.4 cm i. physical . 0.4 cm 1 .
the medium is homogeneous , isotropic and has definite properties of elasticity .
the exitation does not destroy the medium . 0.3 cm 3 .
the exitation is finite : 0.3 cm -at every moment it occupies comparetively small volume of the medium , -it carries finite quantities of energy - momentum and angular momentum , and of any other physical quantity too , -it may have translational - rotational ( may time - periodical ) dynamical structure , i.e. besides its straightline propagation as a whole it may have internal rotational degrees of freedom . 0.3 cm 4 . the exitation is time - stable , i.e. at lack of external perturbations its dynamical evolution does not lead to a self - ruin . in particular , the spatial shape of the exitation does not ( significantly ) change during its propagation . 0.3 cm the above 4 features outline the physical notion of a _
solitary wave_. a solitary wave becomes a _ soliton _ if it has in adition the following property of stability : 0.3 cm 5 .
the exitation survives when collides with another exitation of the same nature .
0.3 cm we make some comments on the features 1 - 5 .
feature 1 requires homogenity and some ealstic properties of the medium , which means that it is capable to bear the exitation , and every region of it , subject to the exitation , i.e. draged out of its natural ( equilibrium ) state , is capable to recover entirely after the exitation leaves that region .
feature 2 puts limitations on the exitations considered in view of the medium properties : they should not destroy the medium .
feature 3 is very important , since it requires _ finite _ nature of the exitations , it enables them to represent some initial level self - organized physical objects with dynamical structure , so that these objects `` feel good '' in this medium .
this finite nature assumption admits only such exitations which may be _ created _ and _ destroyed _ ; no point like and/or infinite exitations are admitted .
the exitation interacts permanently with the medium and the time periodicity available may be interpreted as a measure of this interaction .
feature 4 guarantees the very existence of the exitation in this medium , and the shape keeping during propagation allows its recognition and identification when observed from outside .
this feature 4 carries in some sense the first newton s principle from the mechanics of particles to the dynamics of continuous finite objects , it implies conservation of energy - momentum and of the other characteristic quantities of the exitation .
the last feature 5 is frequently not taken in view , especially when one considers single exitations .
but in presence of many exitations in a given region it allows only such kind of interaction between / among the exitations , which does not destroy them , so that the exitations get out of the interaction ( almost ) the same .
this feature is some continuous version of the elastic collisions of particles .
0.4 cm ii .
mathematical 0.4 cm 1 .
the exitation defining functions @xmath11 are components of _ one _ mathematical object ( usually a section of a vector / tensor bundle ) and depend on @xmath12 spatial and @xmath13 time coordinates . 0.3 cm 2 .
the components @xmath11 satisfy some system of nonlinear partial differential equations ( except the case of ( 1 + 1 ) linear wave equation ) , and admit some `` running wave '' dynamics as a whole , together with available internal dynamics . 0.3 cm 3 .
there are ( infinite ) many conservation laws
. 0.3 cm 4 .
the components @xmath11 are localized ( or finite ) functions with respect to the spatial coordinates , and the conservative quantities are finite . 0.3 cm 5 .
the multisoliton solutions , describing elastic interaction ( collision ) , tend to many single soliton solutions at @xmath14 .
0.4 cm comments : 0.3 cm 1 .
feature 1 introduces some notion of _ integrity _ : _ one exitation - one mathematical object _ , although having many algebraically independant but differentially interrelated ( through the equations ) components @xmath11 . 0.3 cm 2 .
usually , the system of pde is of _ evolution _ kind , so that the exitation is modelled as a dynamical system : the initial configuration determines fully the evolution .
the `` running wave '' dynamics as a whole introduces galileo / lorentz invariance and corresponds to the physical feature 4 .
the nonlinearity of the equations is meant to guarantee the spatially localized ( finite ) nature of the solutions
. 0.3 cm 3 .
the infinite many consrvation laws frequently lead to complete integrability of the equations
. the spatially localized nature of @xmath11 represents the finite nature of the exitation
. 0.3 cm 5 .
the many single soliton asymptotics at @xmath15 of a multisoliton solution mathematically represents the elastic character of the interactions admitted , and so it takes care of the stability of the physical objects beeing modelled .
0.4 cm the above physical / mathematical features are not always strictly accounted for in the literature .
for example , the word _
soliton _ is frequently used for a _ solitary wave _ exitation .
another example is the usage of the word soliton just when the energy density , being usually a quadratic function of the corresponding @xmath11 , has the above soliton properties [ 5 ] .
also , one usually meets this soliton terminology for spatially _ localized _ , i.e. going to zero at spatial infinity , but _ not _ spatially _ finite _ @xmath11 , i.e. when the spatial support of @xmath11 is a compact set .
in fact , all soliton solutions of the well known kdv , sg , nls equations are localized and _ not _ finite . it is curious , that the _ linear _ ( 1 + 1 ) wave equation has spatially finite soliton solutions of arbitarary shape . further in this paper we shall present 1-soliton screw solutions of the vacuum eed equations , so we may , and shall , use the more attractive word soliton for solitary wave .
we hope this will not bring any troubles to the reader .
the screw soliton solutions we are going to present are of photon - like character , i.e. the velocity of their translational component of propagation is equal to the velocity of light @xmath16 , and besides of the energy - momentum , they carry also internal angular ( spin , helicity ) momentum accounting for the available rotational component of propagation
. therefore , this seems to be the proper place to recall some of the well known properties of _
photons_. first of all , as it was explained in the introduction , photons should not be considered as point - like objects since they respect the planck s relation @xmath3 and carry internal ( spin ) momentum , so they have to be considered as extended finite objects with periodical rotational - translational dynamical sructure .
we assume that the concept of soliton , as described above , may serve as a good mathematical tool in trying to represent mathematically the real photons , so that , their well known integral properties to appear as determined by their dynamical structure . we give now some of the more important for our purposes properties of photons . 0.3 cm 1 .
photons have zero proper mass and electric charge .
the ( straightline ) translational component of their propagation velocity is constant and equal to the experimentally established velocity of light in vacuum .
photons are time - stable objects .
every interaction with other objects kills them .
the existence of photons is generically connected with some time - periodical process of period @xmath17 and frequency @xmath18 , so that the planck relation @xmath3 , or @xmath4 , where @xmath19 is the planck constant , holds . 0.3 cm 4 .
every single photon carries momentum @xmath6 with @xmath20 and spin momentum equal to the planck constant @xmath19 .
photons are polarized objects . the polarization of every single photon is determined through the relation between the translational and rotational directions of propagation , hence , the 3-dimensionality of the real space allows just two polarizations .
we call the polarization _
right _ , if when looking along its translational component of propagation , i.e. from befind , we find its rotational component of propagation to be _ clock - wise _ , and the polarization is _ left _ when under the same conditions we find _ anti - clock - wise _ rotational component of propagation
. 0.3 cm 6 .
photons do not interact with each other .
i.e. they pass through each other without changes .
0.4 cm these well known properties of photons would hardly need any comments .
however , according to our oppinion , these properties strongly suggest to make use of the soliton concept for working out a mathematical model of their structure and propagation . and
this was one of the main reasons to develop the extension of faraday - maxwell theory to what we call now extended electrodynamics ( eed ) .
we proceed now to recall the basics of eed , in the frame of which the screw soliton model of photons will be worked out .
we are going to consider just the vacuum case of eed in relativistic formulation .
the signiture of the space time pseudometric @xmath21 is @xmath22 , the canonical coordinates will be denoted by @xmath23 , so the components @xmath24 in any canonical coordinate system are @xmath25 , and @xmath26 for @xmath27 .
the corresponding volume 4-form @xmath28 is given by @xmath29 since @xmath30 .
the hodge @xmath31-operator is defined by @xmath32 , where @xmath33 and @xmath34 are @xmath35 and @xmath36 forms respectively . in terms of @xmath37
we have @xmath38 .
we have also the exterior derivative @xmath39 and the coderivative @xmath40 .
the physical interpretation of @xmath41 are : @xmath42 , are the components @xmath43 of the electric vector @xmath44 , and @xmath45 are the components @xmath46 of the magnetic vector @xmath47 , respectively . in terms of @xmath48
the vacuum maxwell equations are given by @xmath49 in eed the above equations ( 1 ) are extended to @xmath50 in components , equations ( 2 ) are respectively : @xmath51 the maxwell energy - momentum tensor @xmath52\ ] ] is assumed as energy - momentum tensor in eed because its divergence @xmath53\ ] ] is obviously zero on the solutions of equations ( 3 ) and no problems of maxwell theory at this point are known .
the physical sense of equations ( 3 ) is , obviously , local energy - momentum redistribution during the time - evolution : the first two equations say that @xmath0 and @xmath1 keep locally their energy - mimentum , and the third equation says ( in correspondence with the first two ) , that the energy - momentum transferred from @xmath0 to @xmath1 is always equal locally to the energy - momentum transferred from @xmath1 to @xmath0 , hence , any of the two expressions @xmath54 and @xmath55 may be considered as a measure of the rotational component of the energy - momentum redistribution between @xmath0 and @xmath1 during propagation ( recall that the spatial part of @xmath56 is @xmath57 and the spatial part of @xmath58 is @xmath59 ) . obviously , all solutions of ( 1 ) are solutions to ( 3 ) , but equations ( 3 ) have more solutions . in particular , those solutions of ( 3 ) which satisfy the relations @xmath60 are called _
nonlinear_. further we are going to consider only the nonlinear solutions of ( 3 ) .
some of the basic results in our previous studies of the nonlinear solutions of equations ( 3 ) could be summarized in the following way : for every nonlinear solution @xmath61 of ( 3 ) there exists a canonical system of coordinates @xmath62 in which the solution is fully represented by two functions @xmath63 , @xmath64 , and @xmath65 , as follows : @xmath66 @xmath67 we call @xmath68 the _ amplitude _ function and @xmath69 the _ phase _ function of the solution
. the condition @xmath70 allows to set @xmath71 , and further we are going to work with @xmath72 , and @xmath72 will be called _
phase_. as we showed [ 4 ] , the two functions @xmath68 and @xmath69 may be introduced in a coordinate free manner , so they have well defined invariant sense .
every nonlinear solution satisfies the following important relations : @xmath73 we recall also the _ scale factor _ @xmath74 , defined by the relation @xmath75 .
a simple calculation shows that it depends only on the derivatives of @xmath72 in these coordinates and is given by @xmath76
note that eed considers the field as having two components : @xmath0 and @xmath1 . as we mentioned earlier ,
the third equation of ( 3 ) describes how much energy - momentum is redistributed locally with time between the two components @xmath0 and @xmath1 of the field : @xmath77 gives the transfer from @xmath0 to @xmath1 , and @xmath78 gives the transfer from @xmath1 to @xmath0 , thus , if there is such an energy - momentum exchange equations ( 3 ) require permanent and equal mutual energy - momentum transfers between @xmath0 and @xmath1 .
since @xmath0 and @xmath1 are always orthogonal to each other and these two mutual transfers depend on the derivatives of the field functions through @xmath56 and @xmath79 ( i.e. through @xmath80 and @xmath81 , which are not equal to zero in general ) , we may interpret this property of the solution as a description of an _ internal rotation - like _ component of the general dynamics of the field .
hence , any of the two expressions @xmath82 or @xmath78 , having the sense of local energy - momentum change , may serve as a natural measure of this rotational component of the energy - momentum redistribution during the propagation . therefore ,
after some appropriate normalization , we may interpret any of the two 3-forms @xmath83 and @xmath84 as local _ spin - momentum _ of the solution .
making use of the above expressions for @xmath0 and @xmath1 we compute @xmath85 : @xmath86 since @xmath87 , this expression says that we shall have nonzero local spin - momentum only if @xmath72 is _ not _ a running wave along @xmath88 . the above idea to consider the 3-form @xmath84 as a measure of the spin momentum of the solution suggests also some additional equation for @xmath72 ,
because the spin momentum is a conserved quantity and its integral over the 3-space should not depend on the time variable @xmath89 .
this requires to have some nontrivial _
closed 3-form _ on @xmath90 , such that when restricted to the 3-space and ( spatially ) integrated to give the integral spin momentum of the solution
. therefore we assume the additional equation @xmath91 in our system of coordinates this equation is reduced to @xmath92 i.e. @xmath93 equation ( 10 ) has the following solutions : 1@xmath94 . running wave solutions @xmath95 , 2@xmath94 .
@xmath96 , 3@xmath94 .
@xmath97 , 4@xmath94 . any linear combination of the above solutions with coefficients which are allowed to depend on @xmath98 .
the functions @xmath99 and @xmath100 are arbitrary in the above expressions .
the running wave solutions @xmath101 , defined by @xmath102 lead to @xmath103 and to @xmath104 , and by this reason they have to be ignored .
the solutions @xmath105 and @xmath106 , defined respectively by 2@xmath107 and 3@xmath107 , give the same scale factor @xmath108 .
since at all spatial points where the field is different from zero we have @xmath109 , we may choose @xmath110 , so we obtain the following _ nonrunning _ wave solutions of ( 10 ) : @xmath111 where @xmath112 accounts for the two different polarizations .
clearly , the physical dimension of @xmath113 is _ length _ ,
@xmath100 is dimensionless and the _ scale _ factor is @xmath114 .
the corresponding electric @xmath115 and magnetic @xmath116 vectors for the case 2@xmath107 in view of ( 11 ) are @xmath117 , % 12%\ ] ] @xmath118 , % 13%\ ] ] a characteristic feature of the solutions , defined by ( 12)-(13 ) , is that the direction of the electric and magnetic vectors at some initial moment @xmath119 , as seen from ( 12)-(13 ) , is entirely determined by @xmath120 ( under @xmath121 , and so , it does not depend on the coordinate @xmath88 .
therefore , at all points of the 3-region @xmath122 , occupied by the solution at @xmath119 , under the additional conditions @xmath123 and @xmath124 , the directions of the representatives of @xmath115 at all spatial points will be the same , independently on the spatial shape of the 3-region @xmath122 . at every subsequent moment
this common direction will be rotationally displaced , but will stay the same for the representatives of @xmath115 at the different spatial points
. the same feature will hold for the representatives of @xmath116 too .
thus , the representatives of the couple @xmath125 will rotate in time , clockwise ( @xmath126 ) or anticlockwise ( @xmath127 ) , coherently at all spatial points , so these solutions show no twist - like , or screw , propagation component even if the region @xmath122 has a screw shape .
hence , these solutions may be soliton - like but not of screw kind .
the electric @xmath115 and magnetic @xmath116 vectors for the case 3@xmath107 in view of ( 11 ) are @xmath128 , % 14%\ ] ] @xmath129 , % 15%\ ] ] we are going to use this particular solution ( 14)-(15 ) to construct a theoretical example of a screw photon - like soliton solution .
we have to give an explicit form of the amplitude function @xmath68 . in accordance with the _ finite _ nature of the solution
@xmath68 must have compact spatial support , and this guarantees that the solution is finite because the phase function @xmath130 , so , the products @xmath131 , @xmath132 are also finite . 25em
first we outline the idea ( fig1 ) .
we mean to choose @xmath133 at @xmath134 in such a way that @xmath135 to be localized inside a screw cylinder of radius @xmath136 and height @xmath137 and we shall denote this region by @xmath122 ; the @xmath88-prolongation of @xmath122 will be an infinite screw cylinder denoted by @xmath138 .
let this screw cylinder have the coordinate axis @xmath88 as an outside axis , i.e. @xmath138 _ windes around _
the @xmath88-axis and _ never _ crosses it .
the initial configuration @xmath122 shall be made propagating along @xmath138 through both : @xmath88-tanslation , and a consistent rotation around itself and around the @xmath88-axis .
this kind of propagation will be achieved only if every spatial point inside @xmath122 will keep moving along its own screw line and will never cross its neighbors screw lines .
we consider the plane @xmath98 at @xmath139 and choose a point @xmath140 in the first quandrant in this plane , so that the points @xmath141 where @xmath142 are inside the circle @xmath143 centered at @xmath144 , and the distance @xmath145 between @xmath146 and the zero - point @xmath147 is much greater than @xmath136 .
now we consider the function @xmath148}.\ ] ] it is concentrated mainly inside @xmath138 and goes to zero outside @xmath138 although it becomes zero just at infinity .
restricted to the plane @xmath139 it is concentrated inside the circle @xmath149 of radius @xmath136 around the point @xmath146 and it becomes zero at infinity ( @xmath150 ) .
we want this concentrated but _ still infinite _ object to become _ finite _ , i.e. to become zero outside @xmath138 . in order to do this
we shall make use of the so called _ localizing _ functions , which are smooth everywhere , are equal to 1 on some compact set @xmath151 and quickly go to zero outside @xmath151 .
( these functions are very important in differential topology for making partition of unity and glueing up various structures [ 6 ] ) .
we shall denote these functions by @xmath152 .
let @xmath153 be a localizing function around the point @xmath146 in the plane @xmath98 , such that it is equal to @xmath13 inside the circle @xmath154 , centered at @xmath146 , of radius @xmath155 , and @xmath155 is considered to be a bit shorter but nearly equal to @xmath136 , and @xmath156 outside the circle @xmath149 .
we modify now the above considered function as follows : @xmath157}.\ ] ] let now @xmath158 be a localizing function with respect to the interval @xmath159 , where @xmath137 , hence , the function @xmath160}\ ] ] is different from zero only inside @xmath122 .
we choose now the scale factor @xmath113 to be a constant equal to @xmath161 , and making use of the phase @xmath106 with @xmath127 we define the function @xmath162 } % 16% { \rm cos}\left(\frac{z}{l_o}+b(x , y)\right),\ ] ] where @xmath163 is a constant with appropriate physical dimension .
this function ( 16 ) represents the first component @xmath164 of the electric vector at @xmath165 .
similarly , for @xmath166 at @xmath165 we write @xmath162 } % 17% { \rm sin}\left(\frac{z}{l_o}+b(x , y)\right),\ ] ] the right hand sides of ( 16 ) and ( 17 ) define the initial state of the solution , and this initial condition occupies the screw cylinder @xmath122 , its internal axis is a screw line away from the @xmath88-axis at a distance of @xmath167 .
hence , the solution in terms of ( @xmath168 ) in this sytem of coordinates will look like ( we assume @xmath169 , and @xmath170 , @xmath171 and we write just @xmath172 for @xmath100 ) .
@xmath173 } { \rm cos}\bigl(\frac{z}{l_o}+b\bigr ) ; \\[0.6ex ] c\frac{\theta(x , y;r_o)\theta(\xi - z;l_o ) } { { \rm cosh}\left[(x - a)^2+(y - a)^2\right ] } { \rm sin}\bigl(\frac{z}{l_o}+b\bigr);\ 0\biggr\ } \end{gathered}\ ] ] @xmath174 } { \rm sin}\bigl(\frac{z}{l_o}+b\bigr ) ; \\[0.6ex ] c\frac{\theta(x , y;r_o)\theta(\xi - z;l_o ) } { { \rm cosh}\left[(x - a)^2+(y - a)^2\right ] } { \rm cos}\bigl(\frac{z}{l_o}+b\bigr);\ 0\biggr\ } \end{gathered}\ ] ] we consider now the vectors @xmath115 and @xmath116 on the screw line passing through the point @xmath175 , where @xmath171 .
we choose the coordinates @xmath176 as usual : @xmath177 grows rightwards , @xmath178 grows upwards , and then @xmath88 is determined by the requirement to have a right orientation .
if we count the phase anticlockwise from the @xmath177-axis , then at @xmath179 , at the point @xmath180 the vector @xmath116 is directed to the point @xmath147 and @xmath115 is orthogonal to @xmath116 in such a way that @xmath181 is directed to @xmath182 .
when @xmath88 grows from @xmath183 to @xmath184 the magnetic vector @xmath116 on this screw line ( which is the central screw line of the screw cylinder ) turns around the axis @xmath88 and stays always directed to it , while the electric vector @xmath115 rotates around the @xmath88-axis and it is always tangent to this rotation ; at the same time the poynting vector keeps its @xmath182-orientation .
this situation corresponds to the clockwise polarizarion ( looking from behind ) .
if @xmath126 then the electric vector @xmath115 stays always directed to the @xmath88-axis and since we have chosen @xmath170 , i.e. the solution propagates along @xmath182 , the magnetic vector @xmath116 rotates around the @xmath88-axis anticlockwise ( looking from behind ) and is always tangent to this rotation .
so , this corresponds to the anticlockwise polarization .
we note once again that any actual assumption @xmath124 would make the representatives of the magnetic vector ( when @xmath127 ) @xmath116 parallel to each other at every point of the plane @xmath185 at @xmath165 .
this may cause some instabilites with time , so in order to have all representatives of @xmath116 ( or @xmath115 in the other polarization ) at every point of that plane to be directed to the @xmath186axis and not to be paralel to each other , we may use @xmath100 for the necessary corrections .
this remark shows the importance of the relation @xmath187 , @xmath100 must be equal to the angle between the two lines passing through the points @xmath188 $ ] and @xmath189 $ ] , where @xmath143 .
now , every point @xmath190 inside the solution region follows its own screw line so , that the distance @xmath191 between this screw line and the @xmath88-axis is kept the same when @xmath88 grows .
since the spatial periodicity along @xmath88 is equal to @xmath184 , ( note that @xmath161 is in fact the maximum value of @xmath113 ) , we obtain an interpretation of the scale factor @xmath192 as the distance between the screw line inside @xmath138 along which @xmath192 and the @xmath88-axis .
this , in particular , means that for @xmath193 , i.e. for a very thin screw cylinder , the scale factor @xmath114 is approximated by @xmath194 ; it also follows that the relation between the longitudinal ant transverse sizes of the solution is approximately equal to @xmath2 .
we turn now to the integral spin - momentum ( helicity ) computation . according to our assumption
its density is given by any of the correspondingly normalized two 3-forms @xmath84 , or @xmath83 . in order to have the appropriate physical dimension we consider now the 3-form @xmath34 defined by @xmath195.\ ] ]
its physical dimension is `` energy - density @xmath196 time '' . since @xmath197 at most , and in view of ( 8)
, we see that @xmath34 is closed : @xmath198 .
the restriction of @xmath34 to @xmath199 ( which will be denoted also by @xmath34 ) is also closed : @xmath200=0,\ ] ] and we may use the stokes theorem .
we shall make use of the solutions @xmath201 of equation ( 10 ) .
we have @xmath202 , @xmath203 , so , in view of our approximating assumption @xmath192 , we integrate @xmath204 over the 3-space and obtain @xmath205 where @xmath206 is the integral energy of the solution , @xmath207 is the intrinsically defined time - period , and @xmath112 accounts for the two polarizations . according to our interpretation
this is the integral intrinsic angular momentum , or spin - momentum , of the solution , for one period @xmath17 .
this intrinsically defined action @xmath208 of the solution is to be identified with the planck s constant @xmath19 , @xmath209 , or @xmath3 , if we are going to interpret the solution as an extended model of a single photon .
0.3 cm * remark*. for the connection of @xmath84 with our earlier definition of the local intrinsic spin - momentum through the nijenhuis torsion tensor of @xmath210 one may look in our paper , cited as the last one in [ 5 ] .
according to our view , based on the conservation properties of the energy - momentum and spin - momentum , photons are real objects and not theoretical imagination .
this view on these microobjects as _ real extended _ objects , obeying the famous planck relation @xmath3 , almost necessarily brings us to favor the soliton concept as the most appropriate and self - consistent working theoretical tool for now , because _ no point - like conception is consistent with the availability of frequency and spin - momentum of photons , as well as with the possibility photons to be destroyed_. so , the dynamical structure of photons is a real thing , clearly manifesting itself through a consistent rotational - translational propagation in space , and their finite nature reveals itself through the finite 3-volumes of definite shape they occupy at every moment of their existence , and through the finite values of the universal conserved quantities they carry .
the dynamical point of view on these objects reflects theoretically in the possibility to make use of , more or less , _ arbitrary _ initial configurations , i.e. to consider them as dynamical systems .
this important moment allows the localizing functions @xmath211 from differential topology to be used for making the spatial dimensions of the solution finite , and not smoothly vanishing just at infinity as is the case of the usual soliton solutions .
our theoretical screw example , presenting an exact solution of the nonlinear vacuum eed equations , was meant to present a , more or less , _ visual image _ of the well known properties of photons , of their translational - rotational dynamical structure , and , especially , of the nature of their spin - momentum . of course
, we do not insist on the function @xmath212 chosen , any other localisable inside the circle @xmath149 function would do the same job in this theoretical example .
so far we have not experimental data concerning the shape of single photons , and we do not exclude various ones , so our choices for the amplitude @xmath68 and for the scale factor @xmath213 are rather admissible approximations than correct mathematical images .
the more important moment was to recognize the dynamical sense of the quantity @xmath214 , and to find that the spin - momentum conservation equation @xmath215 gives solutions for the phase function @xmath216 , which helps very much to visualize the dinamical properties of photons in our aproach .
so , we are able to obtain a general expression for the integral spin - momentum , which , in fact , is the planck s formula @xmath209 .
moreover , the whole solution @xmath61 describes naturally the polarization properties of photons , it clearly differs the clockwise and anticlockwise polarizations through pointing out the different roles of @xmath115 and @xmath116 in the two cases .
an important moment in our approach is that @xmath0 and @xmath1 are considered as two components of the same solution , so the couples @xmath125 and @xmath217 give together the 3-dimensional picture of _ one _ solution .
this , of course , would require the full energy - momentum tensor @xmath218 , to be two times the usual @xmath219 , given by ( 4 ) : @xmath220 .
finally , we d like to mention that , making use of the localizing functions @xmath221 , we may choose an amplitude function @xmath68 of a `` many - lump '' kind , i.e. at every moment @xmath68 to be different from zero inside many non - overlaping 3-regions , probably of the same shape , so we are able to describe a flow of consistently propagating photon - like exitations of the same polarization and of the same phase
. some of such `` many - lump '' solutions ( i.e. flows of many 1-soliton solutions ) may give the macroimpression of , or to look like as , ( parts of ) plane waves . 1.5 cm * references * 0.6 cm 1 . *
s.j.farlow * , partial differential equations for scientists and engineers , john wiley and sons , inc . ,
1982 . \3 . *
j.j.thomson * , philos.mag.ser . 6 , 48 , 737 ( 1924 ) , and 50 , 1181 ( 1925 ) , and nature , vol.137 , 23 ( 1936 ) ; * n.rashevsky * , philos.mag .
ser.7 , 4 , 459 ( 1927 ) ; * w.honig * , found.phys . 4 , 367 ( 1974 ) ; * g.hunter , r.wadlinger * , phys.essays , vol.2,158 ( 1989 ) , * t.waite * , annales de la fondation louis de broglie , 20 , no.4 , ( 1995 ) , * j.p.vigier * , found . of physics , 21 , 125 , ( 1991 ) , * b.lehnert * , photon and old problems in light of new ideas , ed.v.dvoeglazov , nova science publ . , 2000 , pp.3 - 20
* s.donev , m.tashkova * , proc.r.soc.of lond .
, a 443 , 301 , ( 1993 ) ; * s.donev , m.tashkova * , proc.r.soc . of lond .
, a 450 , 281 , ( 1995 ) ; * s.donev , m.tashkova * , annales de la fondation louis de broglie , vol.23 , no.2 , 89 - 97 , lanl e - print : nlin/9709006 ; vol.23 , no.3 - 4 , 103 - 115 ( 1998 ) , lanl e - prints : nlin/9710001 , 9710003 . | this paper aims to present explicit photon - like ( 3 + 1 ) spatially finite soliton solutions of screw type to the vacuum field equations of extended electrodynamics ( eed ) in relativistic formulation .
we begin with emphasizing the need for spatially finite soliton modelling of microobjects .
then we briefly comment the properties of solitons and photons and recall some facts from eed . making use of the localizing functions from differential topology ( used in the partition of unity ) we explicitly construct spatially finite screw solutions .
further a new description of the spin momentum inside eed , based on the notion for energy - momentum exchange between @xmath0 and @xmath1 , is introduced and used to compute the integral spin momentum of a screw soliton .
the consistency between the spatial and time periodicity naturally leads to a particular relation between the longitudinal and transverse sizes of the screw solution , namely , it is equal to @xmath2 .
planck s
formula @xmath3 in the form of @xmath4 arizes as a measure of the integral spin momentum . |
many authors have studied the neutrino charge radius ( ncr ) @xcite . in 1972 ,
bardeen , gastmans and lautrup @xcite showed in the frame of the sm and using the unitary gauge that the ncr is infinite and therefore it is not a physical quantity .
later , in the same year , s.y .
lee @xcite working in the unitary gauge considered the @xmath16 scattering and defined the ncr including besides the usual terms , diagrams in which the photon is replaced by a neutral gauge boson @xmath17 . in this way he obtained a finite , although gauge dependent quantity @xcite .
one of the earliest analyses of the neutrino charge radius , in the context of the general one - loop electromagnetic form factor of a fermion in electroweak theories , was carried out in 1977 by lee and shrock @xcite .
these authors working in the context of the sm and using the linear r@xmath18 gauge showed explicitly that the ncr is not only infinite , but also gauge dependent .
lee and shrock showed in their paper how a full calculation , including not just charge - radius terms , but also box diagrams ( which could not be considered to be corrections to the neutrino electromagnetic vertex ) combined together to yield a gauge - independent total amplitude .
hence , in order to look for a definition of a physical neutrino charge radius one has to consider other diagrams which contribute to the total amplitude of the physical process @xmath19 .
the papers written by s.y .
lee , and by b.w .
lee and r.e .
shrock inspired many works in which finite and gauge independent quantities , based on the ncr , were introduced by considering the @xmath16 scattering @xcite .
we want to end this paragraph pointying out the following .
eventhough that it has been already shown that the neutrino charge radius is an infinite and gauge dependent quantity in the frame of the standard model ( sm ) when just the proper diagrams are taken into account , it is possible to define a physical neutrino charge radius by considering the @xmath16 scattering , which becomes a finite and gauge independent quantity , independent of the lepton @xmath20 used to define it and also which only gets contribution from the proper neutrino electromagnetic vertex @xcite .
discussions on the experimental bounds on the ncr can be found , for example , in refs.@xcite on the other hand , it is well known that any fermion may develop an anapole moment @xmath21 @xcite . in particular , the neutrino , even massless , may have an anapole moment @xmath22 .
measurements of the solar neutrino flux at super - kamiokande established that at least one neutrino is not massless @xcite . besides , from atmospheric and accelerator neutrino oscillations , we know that there is a non - vanishing mass difference @xcite . from solar and reactor neutrino oscillations , we know that there is a different non - vanishing mass difference @xcite .
so , at least two neutrinos are not massless .
the neutrino anapole moment ( nam ) has been discussed in great detail in the literature @xcite . here , we only want to remark that in the frame of the standard model @xcite it is satisfied the relation @xmath23 for a massless dirac neutrino @xcite .
the relation between the charge form factor and the anapole form factor for massless active ( left - handed ) neutrinos is not model - dependent .
in fact , it is a consequence of having an effective vertex for the neutrino with the left - handed chirality projector . in this work ,
we study possible new physics effects on the ncr in a model independent approach by using the effective lagrangian technique @xcite , which is an appropriate scheme to study those processes that are suppressed or forbidden in the sm . motivated by the highly gauge dependent behavior of the charge and the anapole form factors , @xmath4 and @xmath5 , within the context of the sm ,
we will focus on those effects that could be induced by a yang mills sector possessing a richer gauge structure than that of the dimension four theory . to this end
, we will consider an effective electroweak yang
mills sector that includes @xmath2invariants of dimension higher than four . as we will see below ,
there is only one dimension six @xmath2 invariant that induces the @xmath3 vertex and contributes at the one loop level to these form factors , @xmath6 and @xmath7 , respectively .
hence , we can write @xmath10 and @xmath11 , where @xmath24 and @xmath25 are the standard model form factors
. we will show that , as a consequence of the @xmath2 symmetry , the dimension
six @xmath3 vertex gives a contribution which leads to manifest gauge independent expressions for the @xmath6 and @xmath7 form factors .
this result arises , in part , due to the fact that the @xmath26 invariants of dimension higher than four are not affected by the gauge fixing procedure used in the dimension four theory . as a consequence , in the context of effective theories , fermionic form factors would be made of vertices that are not affected by the gauge fixing procedure , which eventually would lead to gauge independent form factors @xcite .
even more , we will show that it is possible to express these form factors as finite functions of @xmath8 by renormalizing them in the sense of effective field theories @xcite .
these form factors get contribution at the one loop level only from the proper neutrino electromagnetic vertex .
besides , for a massless neutrino and as a consequence of having an effective vertex for the neutrino with the left - handed chirality projector the relation @xmath27 is fulfilled .
hence , @xmath28 and therefore we get the relation @xmath29 , as in the standard model .
this paper is organized as follows . in sec.[sm ] , the structure of the electromagnetic and anapole form factors of the neutrino in the context of the sm is briefly discussed , mainly to fix our notation . in sec.[cal ] , the calculation of the electromagnetic and anapole form factors of the neutrino in the context of an effective yang - mills electroweak theory is presented .
the gauge independence of these form factors , as well as the possibility of introducing a renormalization scheme beyond the dyson s sense , will be emphasized . finally , in sec.[con ] the conclusions are presented .
in this section , we will present a schematic discussion on the main features of the charge and anapole form factors of a massless dirac neutrino within the context of the sm @xcite .
we will take advantage to introduce the notation and conventions that will be used through the paper . for a massless left - handed neutrino
the matrix element of the electromagnetic current can be expressed in terms of only one form factor @xmath31 as @xmath32 for a massless neutrino , this expression can easily be rewritten as follows @xmath33 f_a^{sm}(q^2 ) \ } u_\nu ( p)\ ] ] where@xcite @xmath34 and @xmath35 are the charge and anapole form factors of the neutrino , respectively .
@xmath36 satisfies the physical requirement : @xmath37 on the other hand , the ncr and the nam are defined , respectively , by @xmath38 and @xmath39 that is , @xmath40 proper diagrams contributing to the neutrino charge radius in the standard model in the linear @xmath41 gauge.,width=288 ] reducible diagrams contributing to the neutrino charge radius in the standard model in the linear @xmath41 gauge.,width=288 ]
we now turn to calculating the one loop contribution of an anomalous ( nonrenormalizable ) @xmath3 vertex to the charge and anapole form factors of the sm dirac neutrino .
next , we will predict the corresponding charge radius by adopting an appropriate renormalization scheme .
the gauge structure of the @xmath3 vertex ( and also of @xmath42 ) has been the subject of important attention in the literature in diverse contexts .
the one loop radiative corrections to the renormalizable vertex have been calculated in the sm @xcite and some of its extensions @xcite .
its most general structure has been parametrized in a model independent manner using the effective lagrangian approach @xcite and used in countless phenomenological applications @xcite .
one effective electroweak lagrangian can be constructed by adding to the dimension four lagrangian all the @xmath26 invariants of dimension higher than four , which may respect or no discrete transformations , such as @xmath43 , @xmath44 , and @xmath45 or some combinations of them .
the effective lagrangian can be written as @xmath46 where @xmath47 represents the sm lagrangian and @xmath48 are @xmath49 invariants of dimension higher than four .
@xmath50 represents the new physics scale and the @xmath51 are unknown coefficients , which can be calculated once the fundamental theory is known .
the @xmath26 structure of some of the @xmath52 operators depend on the mechanism responsible for the electroweak symmetry breaking .
if this breaking occurs through the higgs mechanism , the @xmath26 symmetry is linearly realized through the higgs doublet . if this is not the case , the electroweak symmetry
is realized nonlinearly through the introduction of the matrix field @xmath53 instead of the higgs doublet . here ,
the @xmath54 fields represent the pseudo goldstone bosons and @xmath55 is the fermi scale . in this case , the effective lagrangian parametrizes new physics that are the responsible for the electroweak symmetry breaking @xcite . in this paper , we will focus on those type of effective interactions that are independent of the mechanism responsible for the electroweak symmetry breaking . as we will see below ,
this class of invariants induce the nonrenormalizable structure of the @xmath3 vertex , which is dictated exclusively by the @xmath2 group . from the known particles , the @xmath56 gauge boson , whose properties
would exhaustively be studied at the next generation of linear colliders , is the one which possesses the richer collection of electromagnetic form factors , as it is charged and has the highest spin within the category of renormalizable theories . to our best knowledge ,
gaemers and gounaris @xcite derived initially 9 form factors for the @xmath3 vertex , but further on a careful analysis carried out by hagiwara
peccei zeppenfeld
hikasa @xcite showed that only 7 of these quantities are independent indeed .
subsequent studies have confirmed these results @xcite .
these form factors define the charge , the magnetic and electric dipole moments , the magnetic and electric quadrupole moments , and the cp even and cp odd anapole moments of this particle .
the @xmath3 vertex is given by the following lagrangian : @xmath57,\end{aligned}\ ] ] where @xmath58 , @xmath59 , etc . while the cp even @xmath60 parameters define the magnetic dipole and electric quadrupole moments , the cp
odd ones @xmath61 determine the electric dipole and magnetic quadrupole moments of the @xmath56 gauge boson @xcite . on the other hand , @xmath62 and @xmath63
represent the cp even and cp odd anapole moments of this particle .
the only terms in the above equation which are explicitly invariant under the electromagnetic @xmath64 gauge group are those associated with the form factors @xmath65 and @xmath66 .
however , the above interactions arise from a more fundamental lagrangian which is invariant under the electroweak @xmath26 gauge group . in the linear realization of the group @xcite , the electroweak effective lagrangian that induces @xmath67
can be written in terms of @xmath68 invariants that comprise interactions of up to dimension eight as follows : @xmath69 where the dimension four yang mills term has been included . here
, @xmath70 is the covariant derivative of the @xmath26 group and @xmath71 ( @xmath72 ) and @xmath73 are the respective field tensors . in addition , @xmath74 is the higgs doublet .
the two dimension eight operators were introduced in order to generate the anapole moments of the @xmath56 boson .
they are made of the @xmath26invariant piece @xmath75 @xcite needed to generate the @xmath56 current .
these operators , which have not been considered in the literature , can be eliminated using the equations of motion , as must be . as for the dimension
six invariants , their properties have widely been studied in the literature in diverse contexts @xcite .
as already mentioned in sec.[sm ] , a massless dirac neutrino possess only two electromagnetic form factors , which are however not independent .
these form factors are the charge , @xmath0 , and the anapole , @xmath1 , which satisfy the relation @xmath76 .
the @xmath3 vertex can contribute at the one loop level to these form factors through the cp conserving structures .
however , in this work we will focus on the contribution given by the cp even @xmath77 component of this vertex , which is governed exclusively by the @xmath2 gauge group . a comprehensive analysis for the off
shell vertex @xmath78 , with @xmath79 an arbitrary fermion and @xmath80 , will be presented elsewhere @xcite .
there are several good reasons to consider the contribution given by the @xmath2invariant @xmath81 to the neutrino form factors . in first place
is its pure yang
mills nature , which leads to the most general structure for the gauge @xmath3 vertex .
this invariant constitutes therefore a good theoretical instrument to studying the gauge structure of electromagnetic form factors of elementary particles .
another important reason to use this operator is , as we will see below , that it leads to a ncr and nam that are gauge independent in manifest way .
also , due its nonrenormalizable nature , @xmath82 is necessarily generated at one loop or higher orders in any renormalizable theory , which means that it maybe sensitive to new physics effects . using the notation given in fig.[v ]
, the vertex function , @xmath83 , for the @xmath84 interaction can be written as : @xmath85 which , as it is evident , satisfies the following simple ward identities : @xmath86 we now turn to calculating the contribution of the above vertex to the nam .
since the @xmath82 operator is not affected by the gauge fixing procedure of the dimension four theory , there are no contributions from ghost fields .
also , there are not contributions from goldstone bosons , as @xmath82 does not depend on the mechanism responsible for the electroweak symmetry breaking . accordingly , @xmath82 only can contribute through the proper diagram of fig.[i]@xmath87 and the self
energy diagrams given in figs.[r]@xmath88 and [ r]@xmath89 .
the latter one being induced by @xmath82 through the quartic @xmath90 vertex . from all the involved vertices ,
only the sm @xmath3 one , which contributes through the diagram in fig.[r]@xmath88 , depends on the gauge fixing procedure .
however , we have verified that due to the above ward identities , this diagram gives a vanishing contribution .
also , we have encountered that there is no contribution from diagram in fig .
[ r]@xmath89 .
so , the contribution to the nam arises only through the proper diagram given in fig.[i]@xmath87 , whose vertices are all independent of the gauge fixing procedure .
this means that the only gauge dependence of the nam could arise through the @xmath56 propagator , which in the @xmath41 gauge is given by @xmath91.\ ] ] however , it is clear that due to the ward identities given above , the longitudinal components of the @xmath56 propagators do not contribute to the proper diagram given in fig.[i]@xmath87 . as a consequence ,
the amplitude for this diagram is manifestly gauge independent .
this amplitude is given by @xmath92[(k - p)^2-m^2_w][(k - p')^2-m^2_w]},\ ] ] where @xmath93 is the mass of the charge lepton to which is associated the neutrino in consideration . using the passarino
veltman @xcite covariant decomposition , a direct calculation leads to the following expressions for the charge and anapole form factors : @xmath94,\end{aligned}\ ] ] where @xmath95 and @xmath96 are passarino veltman two point scalar functions .
in addition , the dimensionless variable @xmath97 was introduced .
the above expression was obtained after expressing a scalar @xmath98 three point function as a combination of @xmath99 functions @xcite . from the above expressions ,
it is clear that the @xmath100 condition is fulfilled .
it is also evident that , although gauge independent , the anapole form factor is divergent . in the next subsection
, the possibility of renormalizing this quantity within the context of effective field theories will be explored .
the anapole form factor can be expressed in terms of elementary functions as follows : @xmath101,\ ] ] where @xmath102 is the dimensional regularization scale and the ultraviolet divergence is contained in : @xmath103 with @xmath104 the euler s constant .
in addition , @xmath105 , & \textrm{if $ x_q>4 $ } \end{array } \right.\ ] ] where @xmath106 . following refs.@xcite , the divergent term in this amplitude can be absorbed by renormalizing the coefficients of @xmath107 since it already contains all the invariants allowed by the sm symmetry .
the invariant needed to absorb the divergence is @xmath108 where @xmath109 is the usual lepton doublet and @xmath110 is the gauge tensor associated with the @xmath111 group . in this way , the divergence can be absorbed by the @xmath112 parameter of the effective lagrangian . here , we want to point out the following .
the divergency of eq .
( 18 ) needs a counterterm in the effective neutrino - photon vertex to be renormalized .
the invariant of eq .
( 21 ) absorbs this divergence .
however , after renormalization , the contribution of the counterterm is not zero , but a finite value . as a consequence ,
the final result is , in fact , the sum of two finite contributions : one explicit depending on the effective @xmath3 coupling and a second one coming from the counterterm proportional to @xmath112 .
the result quoted in this paper originates from the first contribution . using the @xmath113 renormalization scheme with @xmath114
, the renormalized anapole form factor can be written as follows : @xmath115 where @xmath116 the anapole moment is the on shell quantity @xmath117 . in this limit
, the loop function @xmath118 takes the way @xmath119-\frac{x^2_l}{1-x_l}\log(x_l),\\ & \approx&2\log \big(\frac{\lambda}{m_w}\big)+1,\end{aligned}\ ] ] which leads to the following expression for the additional contribution in the frame of an effective theory to the neutrino charge radius @xmath120,\\ & = & 0.95\times 10^{-34}\ cm^2 \
\lambda_\gamma \big[2\log\big(\frac{\lambda}{m_w}\big)+1\big].\end{aligned}\ ] ] we now turn to discussing our results . in order to make predictions ,
we need to assume some value for the @xmath77 parameter and the new physics scale @xmath50 . we will use the most recent experimental limit on @xmath77 obtained by the d0 collaboration @xcite , namely , @xmath121 for @xmath122 tev .
more recently , this collaboration limited the trilinear @xmath42 vertex to @xmath123 at the @xmath124 c.l . and for the same value of @xmath50 @xcite .
we will make predictions assuming that @xmath125 and @xmath126 tev .
using these values , we obtain the following additional contribution to the neutrino charge radius : @xmath127 it is interesting to compare our result with the theoretical one obtained , within the frame of the sm and by using the pinch technique @xcite , in ref.@xcite . in this reference
, it was derived that @xmath128 for @xmath129 respectively .
hence we can conclude that our result is about one order of magnitude lower than the results derived in ref.@xcite . on the other hand ,
our result is of the same order of magnitude than the new physics contribution derived in the context of the minimal supersymmetric standard model @xcite . finally , it is worthwhile to compare our result for the additional contribution in the context of an effective theory to the neutrino charge radius with some experimental bounds existing in the literature .
the upper bound @xmath130 was derived from primordial nucleosynthesis @xcite . in ref.@xcite , the bound @xmath131 for the charge radius of a right handed neutrino
was derived using astrophysical information obtained from observational data on the sn 1987a .
the bound @xmath132 was derived in ref.@xcite from neutral current neutrino reaction . in ref.@xcite
the upper limit @xmath133 was derived from kamiokande ii and chlorine solar experiments . from the above considerations
, we can see that in general terms our result for @xmath134 is about one order of magnitude lower that the best bounds derived so far .
the trilinear @xmath3 vertex.,width=288 ]
our aim in this paper has been to present the calculation of the additional contribution , in the frame of an effective yang - mills theory , to the charge and the anapole form factors , @xmath6 and @xmath7 , for a massless dirac neutrino .
these form factors get contribution at the one loop level only from the proper neutrino electromagnetic vertex .
we showed that this vertex function is independent of the gauge fixing parameter and ultraviolet finite , as it can be renormalized within the framework of the effective theory .
we also showed that the relation @xmath135 is fulfilled and hence we get the relation @xmath136 .
therefore , @xmath137 as in the standard model .
this well - known relation between the charge form factor and the anapole form factor for massless active ( left - handed ) neutrinos does not depend on an specific model .
this relation is a consequence of having an effective vertex for the neutrino with the left - handed chirality projector .
an interesting point , long discussed in the literature , is the gauge independence and non - divergence of the neutrino charge radius . in the sm , this issue is now clarified with a definition of this quantity with all the required properties , which is furthermore gauge independent and finite . in the model discussed by the present paper , one should again treat these points with care , particularly in the context of effective theories .
the question of gauge independence is solved favorably because only the transverse part of the w - propagator appears .
however , the divergency of eq . ( 18 ) needs a counterterm in the effective neutrino - photon vertex to be renormalized . as it was stated in subsection iii.c , the invariant of eq .
( 21 ) absorbs this divergence .
however , after renormalization , the contribution of the counterterm is not zero , but a finite value . as a consequence ,
the final result is the sum of two finite contributions : one explicit depending on the effective @xmath3 coupling and a second one coming from the counterterm proportional to @xmath112 .
the result quoted in this work originates from the first contribution .
finally , we obtain @xmath138 $ ] . using the recent d0 collaboration constraint on the @xmath77 parameter
, we estimate the value @xmath139 , which is of the order of magnitude that may be expected in theories beyond the sm , as supersymmetry , and about one order of magnitude lower than the current bounds . this work was supported in part by the _
consejo nacional de ciencia y tecnologa _ ( conacyt ) and _ sistema nacional de investigadores ( sni ) de mxico_. a.r .
would like to thank prof .
shrock for valuable comments .
d. y. bardin and o. a. mogilevsky , lett .
nuovo cim .
* 9 * , 549 ( 1974 ) .
g. j. aubrecht and k. tanaka , phys .
d * 10 * , 3522 ( 1974 ) .
n. m. monyonko and j. h. reid , prog .
phys . * 73 * , 734 ( 1985 ) .
p. vogel and j. engel , phys .
d * 39 * , 3378 ( 1989 ) .
k. l. ng , nuovo cim .
a * 107 * , 431 ( 1994 ) .
p. minkowski , p. schupp and j. trampetic , eur .
j. c * 37 * , 123 ( 2004 ) [ arxiv : hep - th/0302175 ] . m. dvornikov and a. studenikin , phys .
d * 69 * , 073001 ( 2004 ) [ arxiv : hep - ph/0305206 ] .
m. s. dvornikov and a. i. studenikin , j. exp .
phys . * 99 * , 254 ( 2004 ) [ arxiv : hep - ph/0411085 ] .
s. y. lee , phys .
d * 6 * , 1701 ( 1972 ) . m. j. musolf and b. r. holstein , phys . rev .
d * 43 * , 2956 ( 1991 ) .
j. bernabeu , l. g. cabral - rosetti , j. papavassiliou and j. vidal , phys .
d * 62 * , 113012 ( 2000 ) [ arxiv : hep - ph/0008114 ]
. j. bernabeu , j. papavassiliou and j. vidal , nucl .
b * 680 * , 450 ( 2004 ) [ arxiv : hep - ph/0210055 ] .
j. bernabeu , j. papavassiliou and j. vidal , phys .
lett . * 89 * , 101802 ( 2002 ) [ erratum - ibid . *
89 * , 229902 ( 2002 ) ] [ arxiv : hep - ph/0206015 ] .
e. nardi , aip conf .
proc . * 670 * , 118 ( 2003 ) [ arxiv : hep - ph/0212266 ] .
j. papavassiliou , j. bernabeu and j. vidal , nucl .
suppl . * 114 * , 197 ( 2003 ) [ arxiv : hep - ph/0210312 ] .
m. hirsch , e. nardi and d. restrepo , phys .
d * 67 * , 033005 ( 2003 ) [ arxiv : hep - ph/0210137 ] .
k. fujikawa and r. shrock , phys .
d * 69 * , 013007 ( 2004 ) [ arxiv : hep - ph/0309329 ] .
j. papavassiliou , j. bernabeu , d. binosi and j. vidal , eur .
j. c * 33 * , s865 ( 2004 ) [ arxiv : hep - ph/0310028 ] .
j. bernabeu , j. papavassiliou and d. binosi , nucl .
b * 716 * , 352 ( 2005 ) [ arxiv : hep - ph/0405288 ] .
j. papavassiliou , j. bernabeu and m. passera , pos * hep2005 * , 192 ( 2006 ) [ arxiv : hep - ph/0512029 ] .
j. a. grifols and e. masso , mod .
lett . a * 2 * , 205 ( 1987 ) .
j. a. grifols and e. masso , phys .
d * 40 * , 3819 ( 1989 ) .
r. c. allen _ et al .
_ , phys . rev .
d * 43 * , 1 ( 1991 ) .
a. m. mourao , j. pulido and j. p. ralston , phys .
b * 285 * , 364 ( 1992 ) [ erratum - ibid .
b * 288 * , 421 ( 1992 ) ] .
p. salati , astropart .
phys . * 2 * , 269 ( 1994 ) .
j. barranco , o. g. miranda and t. i. rashba , arxiv:0707.4319 [ hep - ph ] .
b. zeldovich , soviet phys .
jetp , * 33 * , 1531 ( 1957 ) ; ya .
b. zeldovich and a. m. perelomov , soviet phys .
jetp , * 12 * , 777 ( 1961 ) ; r. e. marshak , riazuddin and c. p. ryan : _ theory of weak interactions in particle physics _ : wiley - interscience .
y. fukuda _ et al .
_ [ super - kamiokande collaboration ] , phys . rev . lett . * 81 * , 1158 ( 1998 ) [ erratum - ibid . * 81 * , 4279 ( 1998 ) ] [ arxiv : hep - ex/9805021 ] .
d. g. michael _ et al . _
[ minos collaboration ] , phys .
lett . * 97 * , 191801 ( 2006 ) [ arxiv : hep - ex/0607088 ] .
s. abe _ et al . _
[ kamland collaboration ] , phys .
lett . * 100 * , 221803 ( 2008 ) [ arxiv:0801.4589 [ hep - ex ] ] . s. m. apenko and yu e. lozovik , j. phys .
b * 15 * , 257 ( 1982 ) .
a. barroso , f. boudjema , j. cole and n. dombey , z. phys .
c * 28 * , 149 ( 1985 ) .
a. rosado , rev .
fis . * 47 * , 132 ( 2002 ) .
m. abak and c. aydin , europhys .
* 4 * , 881 ( 1987 ) . a.
gngora - t and robin g. stuart , z. phys .
c * 55 * , 101 ( 1992 ) .
a. rosado , phys .
d * 61 * , 013001 ( 2000 ) .
l. g. cabral - rosetti , m. moreno and a. rosado , arxiv : hep - ph/0206083 . s. weinberg , phys . rev
* 19 * ( 1967 ) 1264 ; a. salam : proc .
8th nobel symposium , s. 367 ( n. svartholm , ed . , stockholm 1968 ) ; s.l .
glashow , j. illiopoulos and l. maiani , phys .
d * 2 * ( 1970 ) 1285 g. couture , j. n. ng , j. l. hewett , and t. g. rizzo , phys .
rev . * 36 * , 859 ( 1987 ) ; c. l. bilachak , r. gastmans , and a. van proeyen , nucl phys .
b * 273 * , 46 ( 1986 ) ; g. couture , j. n. ng , j. l. hewett , and t. g. rizzo , phys .
d * 38 * , 860 ( 1988 ) ; a. b. lahanas and v. c. spanos , phys .
b * 334 * , 378 ( 1994 ) ; t. m. aliyev , _ ibid .
_ * 155 * , 364 ( 1985 ) ; a. arhrib , j. l. kneur , and g. moultaka , _ ibid . _
* 376 * , 127 ( 1996 ) ; n. k. sharma , p. saxena , s. singh , a. k. nagawat , and r. s. sahu , phys .
d * 56 * , 4152 ( 1997 ) ; t. g. rizzo and m. a. samuel , phys .
d * 35 * , 403 ( 1987 ) ; a. j. davies , g. c. joshi , and r. r. volkas , _ ibid . _ * 42 * , 3226 ( 1990 ) ; f. larios , j. a. leyva , and r. martnez , phys . rev .
d * 53 * , 6686 ( 1996 ) ; g. tavares velasco and j. j. toscano , phys .
d * 65 * , 013005 ( 2001 ) ; _ ibid .
_ * 69 * , 017701 ( 2004 ) ; j. l. garca
luna , g. tavares velasco , and j. j. toscano , phys .
d * 69 * , 093005 ( 2004 ) ; j. hern ' andez s ' anchez , c. g. honorato , f. procopio , g. tavares velasco , and j. j. toscano , phys .
d * 75 * , 073017 ( 2007 ) .
h. novales s ' anchez and j. j. toscano , _ effective lagrangian approach to fermion electric dipole moments induced by a cp - violating @xmath3 vertex _ , to appear in phys . rev .
d , arxiv:0712.2008 [ hep - ph ] .
see for instance , a. de r ' ujula , m. b. gavela , o. pene , and f. j. vegas , nucl .
phys . * 357 * , 311 ( 1991 ) ; a. de r ' ujula , m. b. gavela , p. hern ' andez , and e. masso , nucl .
b * 384 * , 3 ( 1992 ) ; c. arzt , m. b. einhorn , and j. wudka , phys .
d * 49 * , 1370 ( 1994 ) ; t. inami , c. s. lim , b. takeuchi , and m. tanabashi , phys .
b * 381 * , 458 ( 1996 ) ; j. ellison and j. wudka , ann .
nucl . part .
sci . * 48 * , 33 ( 1998 ) .
see also , j. ellison and j. wudka , ann .
nucl . part .
sci . * 48 * , 33 ( 1998 ) . a comprehensive discussion on anapole moment of majorana particles is given in : f. boudjema and c. hamzaoui , phys .
d * 43 * , 3748 ( 1991 ) .
see also , f. boudjema c. hamzaoui , v. rahal , and h. c. ren , phys .
lett . * 62 * , 852 ( 1989 ) .
this @xmath26invariant structure was used to construct the operators that generate the trilinear neutral gauge boson couplings @xmath140 , @xmath141 , and @xmath142 : f. larios , m. a. p ' erez , g. tavares velasco , and j. j. toscano , phys .
d * 63 * , 113014 ( 2001 ) . | in this work , we look for possible new physics effects on the electromagnetic charge and anapole form factors , @xmath0 and @xmath1 , for a massless dirac neutrino , when these quantities are calculated in the context of an effective electroweak yang - mills theory , which induces the most general @xmath2invariant lorentz tensor structure of nonrenormalizable type for the @xmath3 vertex . it is found that in this context , besides the standard model contribution , the additional contribution to @xmath4 and @xmath5 ( @xmath6 and @xmath7 , respectively ) are gauge independent and finite functions of @xmath8 after adopting a renormalization scheme .
these form factors , @xmath6 and @xmath7 , get contribution at the one loop level only from the proper neutrino electromagnetic vertex .
besides , the relation @xmath9 ( @xmath10 , @xmath11 ) is still fulfilled and hence the relation @xmath12 ( @xmath13 , @xmath14)is gotten , just as in the sm . using the experimental constraint on the anomalous @xmath3 vertex ,
a value for the additional contribution to the charge radius of @xmath15 is obtained , which is one order of magnitude lower than the sm value .
= = # 1tempboxa= tempdima=0.5tempdima 0.5tempboxa -tempdima tempboxa # 1tempboxa= tempdima=0.5tempdima 0.5tempboxa -tempdima tempboxa # 1 |
the main impediment towards the use of single photons in schemes for deterministic quantum logic and teleportation @xcite as very robust and versatile carriers of quantum information @xcite is the weakness of optical nonlinearities in conventional media . a major trend aimed at the enhancement of optical nonlinearities exploits one - dimensional ( 1d- ) periodic distributed bragg reflectors and 2d- or 3d - periodic photonic crystals ( pcs ) , where light can be slowed down or trapped via multiple reflections in the vicinity of photonic bandgaps ( pbgs ) @xcite .
giantly enhanced nonlinearity has been predicted when dopants with transition frequencies within the pbg are implanted in the structure , so that light near such frequencies resonantly interacts with the dopants and is concurrently affected by the pbg dispersion @xcite .
the resulting soliton - like transmission of very weak pulses within the pbg while filtering out spurious noise is highly advantageous for classical communication @xcite .
however , this mechanism is incompatible with the goals of quantum logic and communications , particularly with photon - photon entanglement , because of the quantum noise associated with resonant field - atom interactions .
another pathway to enhance the nonlinearities is based on electromagnetically induced transparency ( eit ) in atomic media , which comes about when classical driving fields induce coherence between atomic levels and transform the weak fields into atom - dressed dark - state polaritons @xcite .
the ultrahigh sensitivity of the eit polaritonic dispersion to a small field - induced stark shift of its atomic level can result in an appreciable nonlinear phase shift , impressed by one ultraweak field upon another @xcite .
notwithstanding this promising sensitivity , eit - polariton entanglement by a large conditional phase shift of one photon in the presence of another ( also known as cross - phase modulation ) faces serious challenges in spatially uniform media .
one drawback of these schemes has been the mismatch between the group velocities of the probe pulse moving as a slow eit polariton and the nearly - free propagating signal pulse , which severely limits their effective interaction length and the maximal conditional phase - shift @xcite . to enable long interaction times and thus large conditional phase shifts in a medium of finite length ( up to a few centimeters ) , the group velocities of _ both _ interacting pulses should be small @xcite .
this , however , imposes a limitation on the photonic component of the signal polariton whose magnitude determines the phase shift .
copropagating pulses pose yet another difficulty : since the phase - shift of each pulse is proportional to the local intensity of the other pulse , different parts of the interacting pulses acquire different phase shifts , which results in their spectral broadening . here
we put forward a mechanism that may produce strong photon - photon interactions along with suppressed quantum noise and give rise to their entanglement with high fidelity , by combining the advantages of their dispersion in pbg structures and of the strongly enhanced nonlinear optical coupling achievable via eit in an appropriately doped medium .
the main idea is that a single - photon signal pulse is adiabatically converted into a standing - wave polaritonic excitation inside the periodic structure .
this trapped polariton , having an appreciable photonic component , can impress a large , spatially - uniform phase shift upon a slowly propagating probe polariton
. this task can further be facilitated by employing 2d- or 3d - periodic structures with defects where the two pulses interact via tightly confined modes @xcite .
at frequency outside the pbg to the trapped field @xmath0 at frequency inside the pbg .
atoms b provide eit for the probe field @xmath1 and its cross - coupling with the signal field @xmath0 .
( b ) 2d - periodic photonic crystal ( pc ) having the density of modes @xmath2 , is doped with atomic species a and b. ( c ) interaction and the resulting phase - shift of the probe and signal pulses.,width=529 ] the proposed scheme is based upon a periodic structure containing uniformly distributed dopants
atoms a and b ( see fig .
[ as - pr ] ) .
atoms a , having double-@xmath3 level configuration and interacting with classical driving fields on the transitions @xmath4 with the rabi frequencies @xmath5 , respectively , facilitate the trapping of the signal pulse inside the periodic structure , by converting its frequency from outside to inside of the pbg . on the other hand , atoms b , having n level configuration and interacting with the @xmath6 driving field on the transition @xmath7 , serve to simultaneously slowdown the probe pulse and cross - couple it with the signal .
the following procedure is foreseen to this end .
initially , all atoms a are in the ground state @xmath8 , the driving fields @xmath9 and @xmath10 , where @xmath11 is the temporal width of the input signal pulse @xmath12 .
the carrier frequency of @xmath12 is outside the pbg , close to the @xmath13 transition frequency , so that the usual eit for the input signal due to the @xmath3 configuration @xmath14 is realized . upon entering the medium ,
the signal pulse is slowed down and spatially compressed , by a factor of @xmath15 , to the length @xmath16 , where @xmath17 is its group velocity inside the medium . once the signal pulse has fully accommodated in the medium of length @xmath18 , which requires that @xmath19 , the driving field @xmath20 is adiabatically switched off . as a result ,
the signal is stopped , its photonic component being converted into the stationary atomic ( raman ) coherence @xmath21 @xcite .
next , the driving field @xmath22 is adiabatically switched on to a value @xmath23 , converting the atomic coherence into the signal field @xmath0 , whose frequency is inside of the pbg and amplitude is larger than that of the input signal @xmath12 by a factor of @xmath24 @xcite . due to the bragg scattering of the forward and backward propagating components of the signal pulse with the wave vectors @xmath25 , it remains localized ( trapped ) within the medium @xcite .
both components @xmath0 of the signal field dispersively interact with atoms b via transition @xmath26 with the detuning @xmath27 .
this off - resonant interaction causes an ac stark shift of level @xmath28 , thereby strongly affecting the eit dispersion for the probe field @xmath1 , which interacts with atoms b on the transition @xmath29 @xcite . for a large enough product of the signal - field intensity @xmath30 and interaction time @xmath31 ( @xmath32 being the probe group velocity )
, both pulses accumulate a _ uniform _
conditional phase - shift which can reach @xmath33 ( see fig .
[ as - pr](c ) ) .
finally , reversing the sequence that resulted in trapping of the signal pulse , its frequency is converted back to the original frequency and the @xmath34 pulse leaves the medium .
let us now consider the scheme more quantitatively .
to describe the quantum properties of the medium , we use collective slowly varying atomic operators @xmath35 , averaged over a small but macroscopic volume containing many dopants of species @xmath36 around position @xmath37 @xcite : @xmath38 , where @xmath39 is the total number of the corresponding dopants .
the quantum radiation is described by the traveling - wave ( multimode ) electric field operators @xmath40 and @xmath41 , where @xmath42 are the annihilation operators for the field mode with the wave - vector @xmath43 , @xmath44 being the carrier wave vector of the corresponding field .
these single - mode operators possess the standard bosonic commutation relations @xmath45={\delta}_{ij } { \delta}_{qq^{\prime}}$ ] , which yield @xmath46 = l { \delta}_{ij } { \delta}(z - z^{\prime})$ ] . using the standard technique @xcite
, we perturbatively solve the heisenberg equations for the atomic coherences @xmath47 under eit conditions @xcite and substitute the solution into the propagation equations for the slowly varying field operators @xmath48 .
these operators are related to the polariton operators @xmath49 and @xmath50 , which represent the coupled excitation of the corresponding field and atomic raman coherence @xmath51 .
the mixing angles @xmath52 are defined via @xmath53 , where @xmath54 are the atom - field coupling constants , @xmath55 being the corresponding atomic dipole matrix element and @xmath56 the cross - sectional area of the quantum fields .
note that the amplitude of the photonic component of each polariton is proportional to the @xmath57 of the corresponding mixing angle @xmath58 . under the bragg resonance condition @xmath59 , where @xmath60 is the structure period ,
the equations of motion for polaritons are obtained as [ pols ] @xmath61 here @xmath62 and @xmath63 are the group velocities , @xmath64 and @xmath65 are the intensities ( excitation numbers ) of the signal and probe polaritons , respectively ; @xmath66 are the absorption rates , @xmath67 being the raman coherence decay rate of the corresponding atoms , @xmath68 are the @xmath69-correlated noise operators associated with the relaxation ; @xmath70\cos^2\theta_a \sin^2\theta_b g_b^{\prime 2 } /{\delta}_b$ ] is the cross - coupling rate between the polaritons , @xmath71 being the atom - field coupling on the transition @xmath72 and @xmath73 the decay rate of @xmath74 ( for @xmath75 , the cross - absorption vanishes @xcite and then @xmath76 , being purely real , represents the cross - phase modulation rate )
. finally , @xmath77 is the bragg reflection rate , @xmath78 being the pbg bandwidth . equations ( [ pols ] ) constitute the starting point of our analysis .
their general solution , for arbitrary initial / boundary conditions of the traveling - wave quantized polaritons @xmath79 is not known . when absorption is negligible ( see below ) , for a given time- and space - dependence of the signal - polariton intensity @xmath80 , the solution for the probe is @xmath81,\ ] ] where @xmath82 is the retarded time .
an analytic solution for the two counter - propagating components of the signal polariton can be obtained in the long probe limit , i.e. , when the spatial dependence of the probe - polariton intensity is negligible on the scale of @xmath83 : @xmath84 .
this requires that @xmath85 , where @xmath86 is the duration of the probe pulse ( its spectral width being @xmath87 ) .
then eq .
( [ cpol ] ) is soluble by the fourier transform @xmath88 , with the result [ spolms ] @xmath89 , \\ { \hat{\psi}}_{-}(q , t ) & = & i { \hat{\psi}}_{+}(-q,0 ) \ : e^{i { \hat{\phi}}_s(t ) } \frac{\beta}{\chi } \sin(\chi t),\end{aligned}\ ] ] where @xmath90 and @xmath91 , with @xmath92 .
thus all the spatial modes @xmath93 of the signal polariton acquire the same @xmath94-independent phase shift @xmath95 .
it follows from eqs .
( [ spolms ] ) that a signal polariton composed of modes with @xmath96 will be strongly trapped inside the medium , its wavepacket periodically cycling between the forward and backward components while interacting with the probe polariton , yielding [ polsol ] @xmath97 where @xmath98 , with @xmath99 , is the probe phase - shift .
let us dwell upon the approximations involved in the derivation of eqs .
( [ polsol ] ) . during the conversion of the signal pulse into a standing - wave polaritonic excitation inside the periodic structure , the nonadiabatic corrections resulting in its dissipation are negligible
provided the medium is optically thick , @xmath100 , where @xmath101 is the resonant absorption cross - section for the transition @xmath13 and @xmath102 is the density of atoms @xmath103 @xcite .
after the signal pulse has been trapped in the pbg , due to nonzero values of @xmath94 , its gets spatially distorted ( spreads ) at a rate @xmath104 ( @xmath105 ) .
we can estimate the bandwidth of the signal pulse from its spatial extent as @xmath106 , thus obtaining the upper limit for the distortion rate @xmath107 . on the other hand ,
the bandwidth of the probe is limited by the length of the medium via @xmath108^{-1/2 } = |{\omega}_d^{(b)}|^2 k_p [ { \gamma}_a^{(b ) } \sqrt{3 \pi/2 \rho_b l}]^{-1}$ ] , where @xmath109 is the density of atoms @xmath110 @xcite . finally , the interaction time @xmath111 is limited by @xmath112 , and so the fidelity of the cross - phase modulation is given by @xmath113 $ ] .
consider first the classical limit of eqs .
( [ polsol ] ) , where the operators @xmath79 and @xmath114 are replaced by c - numbers .
then for two single - photon pulses @xmath115 , the conditional phase shift accumulated by the probe and signal fields during the interaction is given by @xmath116 note that the phase shift is proportional to the intensity of the photonic component of the signal polariton , as attested by the presence of the @xmath117 term in the nominator of eq .
( [ phsh_cls ] ) . for realistic experimental parameters , relevant to a doped periodic structure discussed below
, one can obtain @xmath118 ( see fig .
[ as - pr](c ) ) with the fidelity @xmath119 .
we now turn to the fully quantum treatment of the system . to compare the classical and quantum pictures , we consider first the evolution of input wavepackets composed of the multimode coherent states @xmath120 .
states @xmath121 and @xmath122 are the eigenstates of the input fields operators @xmath123 and @xmath124 with eigenvalues @xmath125 and @xmath126 , respectively .
the expectation values for the fields are then obtained as [ expval ] @xmath127 \ , , \\ { \langle { \hat{\cal e}}_{\pm } ( z , t ) \rangle } & = & \alpha_{+ } ( z )
\exp \left [ \frac{e^{i \phi}-1}{l } c
\int_0^t |\alpha_p(\tau^{\prime})|^2 \ , \upd t^{\prime } \right ] \times \left [ \begin{array}{c } \cos(\beta t ) \\
i \sin(\beta t ) \end{array } \right ] \ ,
.\end{aligned}\ ] ] these equations are notably different from those obtained for single - mode @xcite and multimode copropagating fields @xcite because all parts of the probe pulse interact with the whole signal pulse ( and vice versa ) , as is manifest in the space ( time ) integration .
similarly to the cases discussed in @xcite , eqs .
( [ expval ] ) reproduce the classical result only in the limit @xmath128 , yielding @xmath129 and @xmath130 .
a @xmath33 phase shift is then obtained for @xmath131 .
this restriction on the classical correspondence of coherent states comes about since , for large enough cross - phase modulation rates @xmath76 , these states exhibit periodic collapses and revivals as @xmath132 changes from 0 to @xmath133 , which limits their usefulness for quantum information applications .
consider finally the input state @xmath134 , consisting of two single - excitation polariton wavepackets @xmath135 , where the fourier amplitudes , normalized as @xmath136 , define the spatial envelopes @xmath137 of the probe and forward signal pulses that initially ( at @xmath138 ) are localized around @xmath139 and @xmath140 , respectively ( fig .
[ as - pr](c ) ) . during the evolution
, the state of the system evolves according to @xmath141 , where @xmath142 $ ] is the effective interaction hamiltonian whose first and second terms commute .
the implicit time - dependence of the effective hamiltonian , due to the propagation of the probe - polariton pulse with the group velocity @xmath32 , is contained in the operator @xmath143 , with @xmath144 .
after the interaction , at time @xmath145 , the output state of the system is @xmath146 , \label{sph}\ ] ] where @xmath147 .
thus , while the signal pulse periodically cycles between the forward and backward modes , the combined state of the system acquires an overall conditional phase shift @xmath148 .
when @xmath149 , transformation ( [ sph ] ) corresponds to the truth table of the _ universal _ controlled - phase ( cphase ) logic gate between the two photons representing qubits , which can be used to realize arbitrary unitary transformation @xcite .
an @xmath150 2d - periodic lattice of dielectric rods or semiconductor stacks ( fig .
[ as - pr](b ) ) @xcite , with controlled structural defects and mirror confinement in @xmath151 , appears to be the most suitable structure for realizing polaritonic entanglement of two single - photons , since both the signal and the probe pulses may be confined in the @xmath152 directions in the vicinity of a `` defect '' row forming a pc waveguide , thereby avoiding diffraction losses and focusing the fields to a radius of @xmath153 m @xcite . using the double-@xmath3 dopants ( atoms a )
, the signal pulse can be trapped in the pc , with the localization distance @xmath83 extending over many periods , and interact with the probe pulse via atoms b. expressing the atom - field coupling constants @xmath154 through the decay rate @xmath155 of the corresponding excited state as @xmath156 , and assuming that @xmath157 and @xmath158 ( @xmath159 ) , we have from eq .
( [ phsh_cls ] ) @xmath160 . among the possible dopants
, iii - v or ii - vi @xmath161-doped semiconductor quantum dots , having large dipole moments and level structure conducive to eit @xcite , could be the best choice for our scheme , provided high densities can be achieved . in
such single - electron doped qds , the spin degeneracy of the ground and the lowest and higher excited ( charged exciton or trion ) states can be lifted with a magnetic field @xcite , realizing the level scheme of fig .
[ as - pr](a ) , where atoms a and b can spectroscopically be selected ( via optical pumping or spectral hole burning ) from the inhomogeneous ensemble of qds . thus states @xmath162 and @xmath163 are represented by the zeeman - split spin - up and spin - down states of the conduction - band electron in the qd .
the excited states @xmath164 and @xmath165 of atoms a can be the first and the second ( or higher ) exciton states , while states @xmath166 and @xmath74 of atoms b are the zeeman sublevels of the lowest excitonic state .
assuming the parameters @xmath167 cm , @xmath168@xmath169 , @xmath170rad / s @xcite , @xmath171@xmath172 , @xmath173rad / s , @xmath174rad / s , @xmath175rad / s , @xmath176rad / s , @xmath177s@xmath178 , and @xmath179s@xmath178 @xcite , we obtain @xmath118 with the fidelity @xmath180% , the main limiting factor being the decay of raman coherence @xmath181 .
other contenders for observing the proposed effects include periodic structures fabricated from rare - earth doped crystals , such as pr : ysp , in which high - fidelity eit has experimentally been demonstrated @xcite , or cryogenically cooled diamond with high density of nitrogen - vacancy defect centers @xcite .
to summarize , we have proposed a new class of multimode quantum - field interactions involving quantized eit - polaritons in pbg structures .
we have shown that such interactions allow efficient cross - phase modulation between a propagating probe pulse and a trapped signal pulse , whose localization is achieved by an adiabatic four - wave mixing process that pulls its frequency into the pbg .
this localization allows multiply repeated interaction of the signal with the entire probe pulse . as a result , the combined two - photon state of the system can acquire a conditional @xmath33 phase - shift , which corresponds to the universal cphase logic gate .
the phase shift is spatially - uniform and the process may have high fidelity .
the experimental realization of the predicted effects requires the fabrication of periodic structures with large densities of optically active dopants @xcite , which may also find useful applications in laser technology , optical communication or quantum computation .
we note that a similar regime of giant cross - phase modulation with suppressed spectral broadening is also realizable in cold atomic vapors using optically - induced pbgs @xcite , on which we intend to report elsewhere
. the proposed scheme may pave the way to quantum information applications such as deterministic all - optical quantum computation , dense coding and teleportation @xcite . | giantly enhanced cross - phase modulation with suppressed spectral broadening is predicted between optically - induced dark - state polaritons whose propagation is strongly affected by photonic bandgaps of spatially periodic media with multilevel dopants .
this mechanism is shown to be capable of fully entangling two single - photon pulses with high fidelity . |
it has long been appreciated that high energy @xmath2-rays from sources at cosmological distances will be absorbed via electron - positron pair production @xcite on the diffuse background of long wavelength photons produced over the history of the universe .
now that several extragalactic tev sources have been discovered , it is beginning to become possible to use this process to probe the extragalactic background light ( ebl ) .
this technique will become increasingly powerful as many more sources will presumably be discovered at greater distances with the new generation of @xmath2-ray telescopes ( glast , celeste , stacee , magic , hess , veritas , and milagro ) .
broadly speaking , there are three approaches to studying the ebl/@xmath2-absorption connection , represented by the three speakers on this topic here at the veritas workshop : * limits on the ebl , and on models for its production , from @xmath2-absorption data @xcite ; * semi - empirical estimates of the ebl @xcite ; and * prediction of the ebl and @xmath2-absorption from physical theories of galaxy formation and evolution in a cosmological framework .
the advantage of the last of these approaches , which we will follow in this talk , is that it permits one to deduce from @xmath2-ray absorption data a great deal about galaxy formation and evolution , including the effects of the stellar initial mass distribution and of dust extinction and reradiation .
it is also arguably the best way to estimate the extent of @xmath2-ray absorption at various energies , as shown by the correct prediction from a simplified model of this sort ( * ? ? ?
* hereafter mp96 ) that there would be rather little absorption of tev @xmath2-rays from the nearest extragalactic sources , mrk 421 and mrk 501 , at redshifts of only @xmath3 . the calculations reported here ( and in more detail in @xcite and @xcite )
are based on state - of - the - art semi - analytic models ( sams ) of galaxy formation , which we summarize briefly below .
but it will be useful to start by summarizing our earlier calculations ( mp96 ) .
although there is a long history of spectral synthesis models leading to predictions for the ebl ( e.g. , @xcite ) , such models typically attempted to account only for the star formation history of the galaxies existing today i.e. , they are pure luminosity evolution models . moreover in spectral synthesis , the star formation history of each galaxy is _ not _ determined from its cosmological history , taking into account the fact that gas is not available to form stars until it has cooled within a dense collapsed structure .
but the evidence is certainly increasing that there was a great deal of galaxy formation and merging in the past , plausibly in agreement with the predictions of hierarchical models of galaxy formation of the cdm type .
the motivation of the approach used by mp96 was to obtain theoretical predictions for the ebl and the resulting absorption of @xmath2-rays in the context of hierarchical theories of structure formation , specifically within the cdm family of cosmological models .
it is relatively straightforward to calculate the evolution of structure in the dark matter component within the cdm paradigm . in mp96 , the number density of dark matter halos as a function of mass and redshift and its dependence on cosmology
was modeled using press - schechter theory , which agrees fairly well with the predictions of n - body simulations . however , obtaining the corresponding radiation field as a function of time and wavelength involves complicated astrophysics with many unknown parameters .
this problem was addressed empirically , using the simple assumption that each dark matter halo hosts one galaxy , with the galaxy luminosity assumed to be a monotonic function of the halo mass . to model the spectrum of each galaxy , the star formation rate ( sfr )
was assumed to be @xmath4 , with ellipticals ( assumed to be 28% of the galaxies ) having @xmath5 gyr , and spirals ( the remaining 72% ) having @xmath6 gyr .
the actual sfr for each galaxy was determined so that a desired local luminosity function ( llf ) was reproduced at redshift @xmath7 . as there is considerable variation in the observed b - band luminosity function derived from different redshift surveys , mp96 considered three representative choices .
stellar emission was modeled in a simplified way , with each stellar population of a given mass and age treated as a black body with the appropriate temperature .
three different power - law initial mass functions ( imfs ) @xmath8 describing the differential distribution of the initial stellar masses were considered : the standard salpeter imf with @xmath9 , and also steeper imfs with @xmath10 and -2.0 .
the evolution of the gas content and metallicity within each galaxy was treated in the instantaneous recycling approximation .
dust absorption was treated in a standard way @xcite , assuming that the mass of dust increases with the galaxy metallicity and gas fraction .
the extinction curve was similar to a galactic extinction curve , but was scaled according to the metallicity ( galaxies with lower metallicities have steeper extinction curves in the uv , as indicated by observations of the lmc and smc ) .
energy was conserved , so that any energy absorbed by dust was reradiated .
the dust emission spectrum was modeled @xcite with three components : pah molecules ( @xmath11 to 30 @xmath12 m ) , warm dust from active star forming regions ( 30 to 70 @xmath12 m ) , and cold `` cirrus '' dust ( 70 to 1000 @xmath12 m ) .
two cosmological models were considered , a standard ( cluster - normalized , @xmath13 ) @xmath14 cold dark matter ( scdm ) model , and a cobe - normalized cold + hot dark matter ( chdm ) model with the then - favored hot dark matter fraction @xmath15 .
since both were @xmath14 models , the hubble parameter was chosen to be @xmath16 ( @xmath17 km s@xmath18 mpc@xmath18 ) in order to obtain a universe with an age of 13 gyr .
galaxy formation occurs fairly early in the scdm model ( because of the large amount of power on small scales ) , and considerably more recently in the chdm model .
the main conclusion from this study was that cosmology is the dominant factor influencing the ebl in the range 1 - 10 @xmath12 m , which is the range most relevant for absorption of @xmath0 tev @xmath2-rays from nearby extragalactic sources . in this wavelength range , the most extreme differences between the three different llfs and three different imfs considered were less than the difference between the scdm and chdm cosmological models , representing early and late galaxy formation respectively .
it is not difficult to understand why this happens : since the optical luminosity at @xmath7 was fixed for each assumed llf , the main factor determining the ebl in the near infrared was the star formation history . because galaxies were assumed to trace halos in a simple way , the star formation history was almost entirely determined by the cosmology .
as mp96 explained , the scdm model predicted a larger ebl flux because ( 1 ) the stars have put out more light since they have been shining longer than in chdm , ( 2 ) there was more redshifting of their light from the optical to the near infrared , and ( 3 ) the scdm galaxies are older at a given redshift and hence are composed of more evolved stars , producing more flux in the red and near - infrared .
as expected , when the optical depth for @xmath2-rays due to @xmath19 production was calculated , more absorption was predicted for scdm than for chdm .
but for sources as near as mrk 421 and 501 , the predicted absorption only steepens the spectrum a little in the 300 gev - 10 tev range for which results have thus far been published , with curvature noticeable mainly above about 3 tev .
these predictions appear to be consistent with the observations @xcite , unlike those from earlier @xcite and later @xcite calculations based on a semi - empirical approach .
the predictions of our new , more complete treatment are qualitatively consistent with this earlier simplified approach .
our new approach is based on semi - analytic models ( sams ) of galaxy formation , which allow one to model the astrophysical processes involved in galaxy formation in a simplified but physical way within the framework of the hierarchical structure formation paradigm .
the semi - analytic models used here are described in detail in @xcite , @xcite , and @xcite .
these models are in reasonably good agreement with a broad range of local galaxy observations , including the relation between luminosity and circular velocity ( the tully - fisher relation ) , the b - band luminosity function , cold gas contents , metallicities , and colors @xcite .
our basic approach is similar in spirit to the models originally presented by @xcite and @xcite , and subsequently developed by these groups in numerous other papers ( reviewed in @xcite and @xcite ) .
significant improvements included in @xcite are that we assumed a lower stellar mass - to - light ratio ( in better agreement with observed values ) , included the effects of dust extinction , and developed an improved `` disk - halo '' model for supernovae feedback . with these new ingredients , we were able to overcome some of the difficulties of previous models , which did not simultaneously reproduce the tully - fisher relation and b - band luminosity function , and produced bright galaxies that were too blue . instead of assuming a one - to - one relationship between galaxies and dark matter halos , as in mp96
, we now determine the galaxy population residing in halos of a given mass by constructing the `` merging history '' of each halo using an extension of the press - schechter technique . using the method described in @xcite , we create monte - carlo realizations of the masses of progenitor halos and the redshifts
at which they merge to form a larger halo .
these `` merger trees '' ( each branch in the tree represents a halo merging event ) reflect the collapse and merging of dark matter halos within a specific cosmology and have been shown to agree fairly well with merger trees extracted from n - body simulations @xcite .
each halo at the top level of the hierarchy is assumed to be filled with hot gas , which cools radiatively and collapses to form a gaseous disk .
the cooling rate is calculated from the density , metallicity , and temperature of the gas .
cold gas is turned into stars using a simple recipe , depending on the mass of cold gas present and the dynamical time of the disk .
supernovae inject energy into the cold gas and may expell it from the disk and/or halo if this energy is larger than the escape velocity of the system .
chemical evolution is traced assuming a constant yield of metals per unit mass of new stars formed .
the spectral energy distribution ( sed ) of each galaxy is then obtained by assuming an imf and using stellar population models ( e.g. @xcite ; in the present work we use the updated gissel98 models with solar metallicity ) . when halos merge , the galaxies contained in each progenitor halo retain their seperate identities until they either fall to the center of the halo due to dynamical friction and merge with the central galaxy , or until they experience a binding merger with another satellite galaxy orbiting within the same halo .
all galaxies are assumed to start out as disks , and major ( nearly equal mass ) mergers result in the formation of a spheriod . new gas accretion and star formation may later form a new disk , resulting in a variety of bulge - to - disk ratios at late times .
this may be used to divide galaxies into rough morphological types , and seems to reproduce observational trends such as the morphology - density relation and color - morphology trend @xcite . the recipes for star formation , feedback , and chemical evolution contain free parameters , which we set by requiring an average fiducial `` milky way '' galaxy to have an i - band magnitude , cold gas mass , and metallicity as dictated by observations of nearby galaxies .
the star formation and feedback processes are some of the most uncertain elements of these models , and indeed of any attempt to model galaxy formation .
we have investigated several different combinations of recipes for star formation and supernova feedback ( sf / fb ) , discussed in detail in @xcite , @xcite and @xcite .
the star formation history for several different scenarios in shown in figure [ fig : sfrz ] . note that the three models shown in figure [ fig : sfrz ] are for the same scdm cosmology ; the only difference is in the mechanism used to convert cold gas into stars .
this illustrates that , unlike in the previous approach of mp96 , the star formation history of the universe is quite sensitive to the assumed astrophysics and not only the cosmology . here
we will discuss results for a single choice of sf / fb recipe , which corresponds to the fiducial `` santa cruz '' model discussed in @xcite , and is similar to the models of kauffmann et al .
( e.g. @xcite ) .
we shall elaborate on the effects of changing the sf / fb recipes on the ebl and gamma ray absorption in @xcite and @xcite . in @xcite
, we included dust extinction using the same approach as mp96 .
as shown in @xcite , this led to better results for the b - band luminosity function , and improved galaxy colors . using a similar approach to modelling dust extinction , @xcite confirmed these results and demonstrated that in addition the inclusion of dust greatly improves the agreement of the galaxy - galaxy correlation function with observations .
the inclusion of dust extinction and the re - radiation of absorbed light at longer wavelengths is of course a crucial ingredient in modeling the ebl .
the current dust model , discussed in @xcite , is an improved version of the one used in mp96 , and is very similar to the approach used by guiderdoni et al .
@xcite . as in mp96 ,
all of the absorbed starlight is re - radiated by the dust , assuming a three - component blackbody emission spectrum . however , in mp96 the shape of the dust emission spectrum was chosen to match that of the galaxy , which is inconsistent with the global data for iras galaxies . in the current models ,
the temperatures and relative contributions of the three components are determined by requiring the colors at 12 , 25 , and 60 @xmath12 m to match the observed colors of iras galaxies @xcite .
details will be given in @xcite .
the stellar initial mass function determines the wavelength distribution of starlight produced by a stellar population of a given age , and as such it is an important ingredient in the calculation of the ebl . in mp96
we considered only the salpeter imf and steeper power - law imfs , but here we will discuss results both for the salpeter imf and the scalo imf ( see figure [ fig : imfs ] ) .
these are two of the most commonly used imfs , although recent studies ( e.g. , @xcite ) indicate that a better representation of the observed imf may be a salpeter - like slope at @xmath20 , with a flattening at @xmath21 .
the most important difference between the salpeter and scalo imfs for our present purposes is that if both are normalized to the same total mass of stars , the fraction of high - mass stars is higher with the salpeter imf .
since only high - mass stars emit significant amounts of ultraviolet light , this results in much more ultraviolet in the spectrum of a typical galaxy with salpeter imf as compared with scalo imf , as shown in figure [ fig : galspectra ] , both without and with inclusion of the effects of dust .
note that the much greater amount of absorbed ultraviolet light with the salpeter imf results in much more reradiated infrared light at long wavelengths . also note that changing the imf has a relatively small effect on the predictions in the 1 - 10 @xmath12 m range .
[ cols="<,<,<,<,<,<,<,<,<,^,^",options="header " , ] [ tab : cosmo ] previous sam calculations of properties of local galaxies ( e.g. @xcite ) assumed a scalo imf .
however , in @xcite and @xcite , it was shown that a more `` top - heavy '' imf ( more high - mass stars ) such as the salpeter imf is favored by observations of uv - bright galaxies at very high redshift ( lyman - break galaxies ) . in @xcite
, we investigate the effects of using different imfs ( scalo or salpeter ) on the observable properties of galaxies at @xmath7 .
in particular , we calculate the luminosity function at 2000 , b , r , and k , the corresponding 2000 - b , b - v , b - i , and b - k colors predicted by our models , and the tully - fisher relation in various bands .
we find that because the mass - to - light ratio in the longer wavebands ( i to k ) is significantly higher for the salpeter imf , the luminosity of a galaxy with a given velocity dispersion is smaller .
this makes it very difficult to obtain a bright enough tully - fisher zero - point in cosmologies with @xmath14 , in which the baryon fraction is low ( we assume a fixed value of @xmath22 , as suggested by observations @xcite ; the baryon fraction @xmath23 is therefore higher in low-@xmath24 cosmologies ) .
if the mass - to - light ratios predicted by the current generation of stellar population models are accurate and the salpeter imf is really representative of typical galaxies , this may suggest that the baryon fraction in bright galaxies must be @xmath25 , similar to the value in groups and clusters .
in @xcite , we calculate the predicted ebl for several cosmologies and also investigate the effects of the assumed imf and star formation recipe , and the variations among stellar population models compiled by different groups .
here we can only present a subset of preliminary results .
the cosmological models considered are summarized in table 1 .
the scdm model is shown for comparison with other work , but is ruled out by many independent observational considerations .
the three remaining models represent currently favored variants of the cdm family of models .
the shape and normalization of the luminosity function that we obtain from the sams depends on the cosmological model , as shown in figure [ fig : lfk ] .
the k - band luminosity function shown in the figure is relatively insensitive to dust and the star formation history . to put the models on an equal footing , following a similar logic to the b - band local luminosity function normalization of mp96 , we renormalize each model to give the same integrated luminosity in the k - band .
the common normalization is obtained by integrating the observed local luminosity function from ref .
the resulting correction factors range from 0.42 for scdm to 1.7 for lcdm .
note that this renormalization also sidesteps the known inaccuracy of the press - schechter model used to estimate the number density of dark matter halos .
the factor of 0.42 for scdm is consistent with the rule - of - thumb factor of 0.5 determined from comparison with n - body simulations @xcite .
figure [ fig : lcdmsalscalo ] shows the ebl for the model for the scalo and salpeter imfs .
as expected from our previous discussion , the predicted ebl is much higher for the salpeter imf at both short and long wavelengths , but the predictions are very similar in the 1 - 10 @xmath12 m band that is most relevant for @xmath0 tev @xmath2-ray attenuation from relatively nearby sources , in agreement with mp96 .
note that both ebl curves are consistent with the lower limits from source counts at ultraviolet , optical , and near - infrared wavelengths ( filled symbols ) , but neither curve is high enough to agree with the new dirbe ebl detections at 140 and 240 @xmath12 m @xcite .
the remaining results that we present are all for the salpeter imf .
figure [ fig : ebl_all ] shows the predicted ebl for the four cosmological models discussed above .
note that the three models in which galaxy and star formation are relatively early predict rather similar ebl , while chdm predicts generally lower ebl .
figure [ fig : evol ] shows the reason for this in more detail . only about 10% of the ebl in the 1 - 10 @xmath12 m band comes from @xmath26 for the chdm model , compared to 20 - 40% for the model , in which galaxies form considerably earlier .
( an alternative way of looking at the evolution of the ebl is given in figure 8 of @xcite . )
keep in mind , however , that even in a model with an `` early formation '' cosmology like but which assumes less efficient star formation at high redshift ( for example , the model with the star formation history indicated by the light solid line in figure [ fig : sfrz ] ) , the ebl will show a steeper evolution than that shown in figure [ fig : evol ] for lcdm .
so we expect some degeneracy between cosmology and astrophysics .
however , to the extent that the background cosmology may soon be determined by other methods , measurements of the ebl will provide useful constraints on the star formation history of the universe .
all of the models fall short of the dirbe detection at 140 @xmath12 m by at least a factor of @xmath27 .
this may be due to a number of effects that we have not yet included in our modeling .
much of the far - infrared and sub - mm light is probably produced by heavily extinguished ultra - luminous starburst galaxies . the phenomenological work of guiderdoni et al .
@xcite suggests that the contribution from this population may increase with redshift .
our independent work suggests a physical reason for this : the galaxy interactions that probably trigger these starburst events are more frequent at higher redshift because of the higher density of the universe , and galaxies are more gas rich so the starburst events may be more dramatic @xcite .
observations of nearby starburst galaxies ( cf .
@xcite ) suggest that the galactic / smc model that we have used here does not provide a good description of dust extinction in actively star forming galaxies .
additional contributions to the far - ir flux that we have not included may come from agn and from energy that is injected into the gas by supernovae and later radiated at long wavelength .
we will include the contribution of starburst galaxies and address these other effects in improved models that we are developing in collaboration with bruno guiderdoni and julien devriendt .
however , we do not expect that this improved treatment will have much effect on the absorption of @xmath2-rays with energies @xmath28 tev , which we now discuss briefly .
here we have space to present only the results for one case , lcdm with salpeter imf .
figure [ fig : tau ] shows the optical depth of the universe @xmath29 as a function of @xmath2-ray energy for varying source redshifts , @xmath30 .
the corresponding attenuation factors , @xmath31 , are shown in figure [ fig : attenuation ] .
note that for sources as near as mrk 421 and 501 ( @xmath3 ) , attenuation is predicted to be rather small between 1 - 10 tev , with little curvature in the spectrum .
gamma - ray energies @xmath32 tev for local sources , or sources at @xmath33 for lower energies , will likely be needed in order to see significant attenuation . in @xcite we will present results for several cosmological models , and discuss dependence on imf and star formation prescriptions .
bullock et al .
@xcite discusses the difference in predicted @xmath2-attenuation between salpeter and scalo imf for this same lcdm model .
the early universe is much more transparent to 10 - 100 gev @xmath2-rays with the scalo imf , since the fraction of high - mass stars is lower , and the ultraviolet flux density is correspondingly reduced ( cf .
* semi - analytic models ( sams ) of galaxy formation provide a convenient and powerful theoretical framework to determine how input assumptions e.g. , cosmology , star formation history , and imf affect the predicted extragalactic background light ( ebl ) and the resulting @xmath0tev @xmath2-ray attenuation . *
the 1 - 10 @xmath12 m ebl and the resulting attenuation of few - tev @xmath2-rays reflect mainly the history of star formation in the universe , with less attenuation for models such as chdm in which galaxies form relatively late . * the ebl at @xmath34 @xmath12 m and @xmath35 @xmath12 m is significantly affected by the imf and the modeling of the absorption and reradiation by dust . *
gamma - ray energies @xmath32 tev and/or sources at @xmath33 will probably be needed to provide clear evidence of attenuation due to @xmath36 . * therefore , both space- and ground - based @xmath2-ray telescopes will be required to probe the spectra of agns at various redshifts , in order to determine both * * the unabsorbed spectra , which will help determine how these @xmath2-rays are produced , and * * the intergalactic absorption , which as we have shown is affected by cosmology , star formation history , imf , and dust .
jrp and jsb were supported by nsf and nasa grants at ucsc , and rss was university fellowship from the hebrew university .
jrp thanks avishai dekel for hospitality at hebrew university .
donn macminn contributed significantly to an early stage of this work , especially the dust modeling .
his life was tragically cut short by a hit - and - run driver as he was bicycling near chicago on august 30 , 1997 . | we present here the extragalactic background light ( ebl ) predicted by semi - analytic models of galaxy formation , and show how measurements of the absorption of gamma rays of @xmath0 tev energies via pair production on the ebl can probe cosmology and the formation of galaxies . semi - analytic models permit a physical treatment of the key processes of galaxy formation including gravitational collapse and merging of dark matter halos , gas cooling and dissipation , star formation , supernova feedback and metal production and have been shown to reproduce key observations at low and high redshift . using this approach
, we investigate the consequences of variations in input assumptions such as the stellar initial mass function and the underlying cosmology .
we conclude that observational studies of the absorption of @xmath1 tev gamma rays will help to constrain the star formation history of the universe , and the nature and extent of the extinction of starlight due to dust and reradiation of the absorbed energy at infrared wavelengths .
cosmology : observations , diffuse radiation infrared : galaxies galaxies : evolution gamma rays : theory |
lyman alpha ( ly@xmath2 ) galaxies have been studied in detail for nearly a decade now , and we are only just beginning to learn about the properties of individual objects .
most initial narrow - band studies ( e.g. , hu et al . 1998 ; rhoads et al . 2000 ) observed blank fields with two filters : a narrow - band filter ( to capture the ly@xmath2 emission line ) , and a broad - band filter encompassing this narrow - band ( used to measure a narrow - band excess ) .
learning the detailed properties of individual ly@xmath2 galaxies would be difficult with these data
. however , these early studies were able to measure the equivalent widths ( ews ) of their objects , calculated from the ratio of narrow - band flux to broad - band flux ( malhotra & rhoads 2002 ) .
some of these initial results were surprising , as in many cases the ly@xmath2 ew distribution was higher than one might have expected from stellar population models , as many objects exhibited rest - frame ly@xmath2 ews over 200 ( e.g. , kudritzki et al .
( 2000 ) ; malhotra & rhoads ( 2002 ) ; dawson et al . ( 2004 ; 2007 ) ; shimasaku et al . ( 2006 ) ; finkelstein et al . ( 2007 ) ) .
a normal stellar population ( defined as having a salpeter ( 1955 ) initial mass function ( imf ) , solar metallicity , and a constant star formation rate ) has a maximum ly@xmath2 ew of 260 at 1 myr , falling to a constant value of 95 at 100 myr .
thus , while objects with ews @xmath0 200 are possible with a normal stellar population , they would all have to be extremely young .
looking at a random sample of ly@xmath2 galaxies , it would be extraordinarily rare if they all had ages of only a few myr , thus something is likely happening to skew the ews higher than they should be .
there are many possible reasons why this could be occurring , and the easiest area to examine is the assumption of a normal stellar population . for example , if a stellar population had sub - solar metallicity , it would form stars with hotter stellar photospheres .
these stars would then pour out more ionizing photons , which , after interacting with the local interstellar medium ( ism ) , would in turn create more ly@xmath2 photons .
another possibility exists , whereby if a galaxy had a top - heavy initial mass function , it would create more massive stars than a salpeter imf , again resulting in more ionizing photons , and thus more ly@xmath2 photons .
both of these scenarios are possible in primitive , dust - free galaxies .
strong ly@xmath2 lines also come from the active galactic nuclei ( agn ) components of some galaxies , but previous work has shown that most ly@xmath2 galaxies detected via narrow - band selection are not agn ( malhotra et al .
2003 ; wang et al .
2004 ) .
ly@xmath2 photons are resonantly scattered by any neutral hydrogen present , thus in a homogeneous ism , they will have to travel much further to escape the galaxy , dramatically increasing the chance of absorption by dust ( meier & terlevich 1981 ) .
however , if the dust is in an inhomogeneous distribution , it is possible that a galaxy could appear to have a strong ly@xmath2 ew at later ages . in this scenario
, ly@xmath2 can escape unaffected by dust if it is somehow screened from interacting with the dust grains .
this can happen if the dust resides in cold , neutral clumps embedded in an ionized inter - clump medium .
this scenario was first proposed by neufeld ( 1991 ) and examined in detail more recently by hansen & oh ( 2006 ) . in this ism geometry
, the ly@xmath2 photons get scattered by neutral hydrogen at the surface of these clumps , and thus never see the dust which is embedded deeper inside .
however , the continuum photons suffer no such restriction , so they pass un - obstructed into the interior of the clump , with many continuum photons being absorbed ( extincted ) and scattered ( resulting in reddening of the continuum colour ) .
the ly@xmath2 photons proceed to bounce around from clump to clump , eventually escaping the galaxy .
the net result of this is an increase in the ly@xmath2 continuum flux ratio over that intrinsic to the stellar population .
we would thus detect this galaxy with a higher ew , irrespective of its stellar population characteristics .
we believe that this scenario may be responsible for some of the high - ew objects seen in narrow - band studies , and in our previous work ( finkelstein et al .
2009 ; hereafter f09 ) , we found ten objects out of a sample of 14 laes where this scenario appears to be in effect ( one of which was also studied in finkelstein et al .
two of these objects are evolved galaxies with large ly@xmath2 ews due to this dust enhancement scenario ( the other eight are still young , but have their ly@xmath2 ews enhanced to be larger than their already large value due to dust ) .
the first of these two evolved galaxies was best fit by a 450 myr , 3.8 x 10@xmath3 m@xmath4 stellar population with a rest - frame ew of @xmath5 100 .
this relatively high ew at old age was due to dust , as the best - fit model had 0.3 magnitudes of dust extinction at 1200 , but this dust was in a geometry such that it did not attenuate ly@xmath2 measurably .
this resulted in a @xmath5 35% increase in the ly@xmath2 ew over a similar stellar population with no dust .
the second of these objects had a most likely age ( see f09 for derivation of most likely age ) of 500 myr , a mass of 5 @xmath6 10@xmath7 m@xmath4 , and 3.5 mag of dust extinction in a very clumpy geometry , greatly enhancing the ly@xmath2 ew . while these objects are very interesting , and could go a long way towards explaining the apparent discrepancy in ew distributions , the rest of the objects in the sample had their own surprises .
they were all young and lower - mass , with ages from 3 15 myr , and masses from 1.6 @xmath6 10@xmath8 6.1 @xmath6 10@xmath3 m@xmath4 .
these ages and masses are indicative of the traditional belief of ly@xmath2 galaxies , where they are very young , and possibly primordial in nature .
however , all of these twelve laes were best fit by stellar populations with considerably non - primordial amounts of dust , with a@xmath9 = 0.5 4.50 , with many galaxies showing indications of a clumpy ism . by fitting their observations to stellar population synthesis models ,
many other recent studies have derived dust extinction in their ly@xmath2 galaxies as well ( e.g. , chary et al .
2005 ; pirzkal et al . 2007 ; lai et al .
2007 ; pentericci et al . 2008 ) . however , these model parameters can be degenerate , as at rest - frame uv and optical wavelengths the reddening of the continuum colour could be due to both dust _ and _ old stars . while observing these objects in the rest - frame optical with the _
spitzer space telescope _ helps reduce this degeneracy , data at longer wavelengths could provide much tighter constraints .
when uv and optical photons are absorbed by dust , they are re - emitted in the far - infrared ( fir ) , resulting in a second observed peak in the spectral energy distribution ( sed ) . observing these galaxies in the sub - millimeter and millimeter ( mm ) regimes would detect this dust emission directly , which could resolve this red - colour degeneracy .
the atacama large millimeter array ( alma ) will be a uniquely sensitive tool to detect these high - redshift galaxies if they have dust , given its excellent observing conditions and large baseline .
thus , obtaining alma observations ( even non - detections ) , will provide a much tighter constraint on the amount of dust in these objects than rest - frame uv - optical observations alone .
we use alma as a measure of future observatory capabilities of detecting dust emission from laes .
if we can confirm that many ly@xmath2 galaxies do in fact have dust , then it will help show that they are not the primitive galaxies that they were once thought to be .
on the other hand , if we do not find evidence for dust , it may indicate that a top - heavy imf or population iii stars are responsible for the strong ly@xmath2 ews .
where applicable , we assume benchmark model cosmology , where @xmath10 = 0.3 , @xmath11 = 0.7 and h@xmath12 = 0.7 ( c.f .
spergel et al .
all magnitudes are listed in ab magnitudes ( oke & gunn 1983 ) . in section 2
we discuss our sample of ly@xmath2 galaxies which we use to make predictions . in section 3
we present our results , and in section 4 we make conclusions about the implications of these results .
we examine samples of laes from two recently published studies : finkelstein et al . 2009 and pirzkal et al . 2007 .
the sample from f09 consists of 14 ly@xmath2 galaxies in the great observatories origins deep survey ( goods ) _ chandra _ deep field south ( cdf s ) narrow - band selected to be at z @xmath13 4.5 .
we will use the identifiers from f09 , which are based on their detection filters : cha-1 to 4 ( detected in nb656 , or @xmath14 ) , ch8 - 1 and 2 ( detected in nb665 , or @xmath14 + 80 ) , and cs2 - 1 to 8 ( detected in nb673 , or [ sii ] ) . see f09 for further details on the data reduction and selection process .
seven of the 14 objects have published photometric redshifts putting them near z @xmath5 4 4.5 ( music , grazian et al .
2006 ; fireworks , wuyts et al . 2008 ) .
as mentioned in the introduction , 12 of these objects were best - fit by young ( 3 15 myr ) , low - mass ( 1.6 60 @xmath6 10@xmath8 m@xmath4 ) stellar populations with a good amount of dust ( a@xmath9 = 0.5 4.5 ) , with the dust in some cases enhancing the ly@xmath2 ew .
two objects ( cha-3 and cs2 - 4 ) were determined to contain older ( 450 500 myr ) , more evolved populations , with dust extinction attenuating the continuum only due to a clumpy ism , thus enhancing the ly@xmath2 ew . the sample from pirzkal et al .
( 2007 ; hereafter p07 ) consists of nine ly@xmath2 galaxies selected in the _ hubble _ ultra - deep field ( hudf ; beckwith et al .
2006 ) , identified by their emission lines with spectra from the grapes ( grism acs program for extragalactic science ; pirzkal et al . 2004
; malhotra et al . 2005 ; xu et al . 2007 ;
rhoads et al . 2008 ) survey .
these objects were selected based on a detectable ly@xmath2 emission line in the grapes grism data , and broadband acs photometry consistent with a lyman break galaxy ( p07 ) . in this paper , we will use the object ids from p07 : 631 , 712 , 4442 , 5183 , 5225 , 6139 , 9040 , 9340 and 9487 .
these objects lie in the redshift range 4.00 @xmath15 z @xmath15 5.76 , with ly@xmath2 line fluxes ranging from 17 60 @xmath6 10@xmath16 erg s@xmath17 @xmath18 and @xmath19 magnitudes from 25.66 28.36 . using an exponentially decaying star formation history ( sfh )
, p07 found best - fit ages of 1 20 myr and stellar masses from 7 @xmath6 10@xmath20 1.8 @xmath6 10@xmath3 m@xmath4 .
three of the objects were best fit by dust , with 0.05 0.6 magnitudes of extinction in the v - band ( four objects were best - fit with dust using a single - burst sfh , and five objects were best - fit with dust using a two - burst sfh ) . for comparison with our laes , we also use the catalog of spectroscopically confirmed hudf lyman break galaxies ( lbgs ) from hathi , malhotra & rhoads ( 2008 ; hereafter hmr08 ; malhotra et al .
2005 ; rhoads et al .
there are 49 lbgs in the catalog , but for objects at redshifts greater than 4.5 , we require j and h band measurements to derive their fir fluxes , thus we were restricted to using only 21 of the 49 lbgs . while we know that photons absorbed by dust are re - emitted in the far - infrared , the stellar population models used in both f08ab and p07 ( bruzual & charlot 2003 ) , while computing fluxes as red as 160 @xmath21 m rest - frame , do not compute dust emission . in order to predict the fir dust emission from the objects in our sample , we exploit a relation between the uv spectral slope ( @xmath22 ; defined as f@xmath23 @xmath24 @xmath25 by calzetti et al .
1994 ) and the infrared excess ( defined as the ratio of fir flux to uv flux ) detailed in meurer et al .
( 1997 ; hereafter m97 ) .
this relation is shown in figure 1 of m97 , with the infrared excess increasing exponentially with the spectral slope .
this relation is due to dust reddening of the spectral slope ( @xmath22 becoming less negative ) .
larger quantities of dust will increase the amount of uv extinction , increasing the amount of flux re - emitted in the fir .
since we have obtained estimates of dust in our previous studies , we could use our derived extinction values to estimate the fir flux . however , not all objects are able to be detected in the observed ir , especially at higher redshift .
these ir data are necessary to break the aforementioned degeneracy of red colors due to dust or old stars .
we aim to predict fir fluxes independent of stellar population modeling results , with the goal that our method can be applied to objects which only have detections in the optical or nir . to use the m97 relation , we first require the uv spectral slope for all objects in our sample , which usually requires spectra .
however , in m97 , the authors detail a method of obtaining @xmath22 from the broad - band v and r magnitudes .
unfortunately , this relation is only valid for z @xmath5 3 , but hmr08 compute similar relations for z @xmath5 4 and z @xmath5 5 6 .
for objects in our sample with z @xmath15 4.5 , we computed @xmath22 via : @xmath26 and for objects with z @xmath27 4.5 , we computed @xmath22 via : @xmath28 thirteen of the 23 objects in our sample were calculated to have @xmath22 @xmath0 -2.5 , indicating that the dust in these galaxies is contributing significantly to their fir fluxes .
table 1 details the relevant parameters for each object in our sample , including the uv spectral slopes calculated with the above method .
once @xmath22 is calculated , we use the relation from m97 to compute the fir flux from the uv flux ( defined in m97 as the flux at 2320 ) .
first , we used the continuum slope calculated in equation 1 or 2 , and then from the definition of continuum slope ( see above ) , we calculated the uv flux via : @xmath29^{\beta+1}}{9200^{\beta}}\right)\ ] ] where f@xmath30 is the z flux density in units of f@xmath23 and 9200 is the central wavelength of the z filter . to find f@xmath31/f@xmath32 , we fit an exponential function to the dashed line in figure 1 of m97 .
the equation of this line was : @xmath33 this equation accurately matched m97 s fig .
1 over the range of -2.5 @xmath34 @xmath22 @xmath34 4 .
while some objects in our sample did have @xmath22 @xmath34 -2.5 , these objects will have a very small f@xmath31/f@xmath32 ratio , and thus the emission from their small amount of dust will very likely not be detectable . in order to obtain the fir flux densities at a few representative wavelengths ,
we used equation 10 from m97 , which details how the fir flux is composed of the flux densities at 60 @xmath21 m and 100 @xmath21 m ( in jy ) : @xmath35 to independently calculate f@xmath36 and f@xmath37 , we require second equation relating these two parameters .
we used a template sed from bendo et al .
( 2006 ) , who computed the best fit modified blackbody curve to ngc 4631 , an edge - on sd galaxy ( a reasonable local analog to laes ) .
they found that a simple blackbody modified by a @xmath38 term , with a dust temperature of 28 k , fit their observed data points from @xmath5 10 1000 @xmath21 m .
@xmath39 adopting an sed of this form and t = 28 k , we calculated a ratio of f@xmath37/f@xmath36 of 3.64 . using this in combination with eqn . 5
, we found : @xmath40
while we know the rest - frame 60 @xmath21 m flux for the objects in our sample , we also would like to know how their fir flux changes with wavelength . to do this , we scaled the bendo et al .
modified blackbody function up to the value of the 60 @xmath21 m flux for each object ( redshifted to the redshift of the object ) , so that we could see how their fluxes behaved at all wavelengths in the range of interest .
we can then compare the flux densities of these objects at any fir wavelength to the sensitivities of sub - mm instruments to see if an object will be detected .
[ cols="^,^,^,^,^,^,^ " , ] with the alma sensitivity calculator , we computed alma sensitivity curves using the default set of parameters for an integration time of four hours in bands 8 ( 385 500 ghz ) and 9 ( 602 720 ghz ) .
continuum measurements are possible with alma in 8 ghz instantaneous bandpasses , corresponding to a bandpass width of 14 @xmath21 m at 730 @xmath21 m ( wilson 2007 ) .
we thus chose three windows in which to compare our objects to the alma sensitivity curves : 450 , 610 and 730 @xmath21 m .
the average sensitivity in an 8 ghz width across these three bandpasses are plotted as thick black lines in figure 1 . from the relative positions of these sensitivities to the sed curves from our objects in this figure , we can see which objects will have their dust emission detected with alma . out of the fifteen objects in the cdf - s sample , fig .
1 shows that six of them appear to contain enough dust to be detected in the observed sub - mm .
these seven objects , cha-4 , ch8 - 1 , cs2 - 2 , cs2 - 3 , cs2 - 4 and cs2 - 6 , contain dust in a range a@xmath9 = 1.0 4.5 mag .
one of these , cs2 - 3 , which was best - fit by a@xmath9 = 4.5 mag , has the brightest f@xmath37 data point ( 28333 @xmath21jy ) in the whole sample , further proof that our derivation of fir fluxes is consistent with our model fitting .
cha-1 , cha-3 , ch8 - 2 , cs2 - 5 and cs2 - 8 have @xmath22 @xmath34 -2.5 , thus we did not calculate their fir fluxes .
objects cha-2 , cs2 - 1 and cs2 - 7 do have measurable 60 @xmath21 m fluxes ( 0.2 4.7 @xmath21jy ) , but these are likely much too faint to be detectable .
it is of note that cs2 - 4 is only detectable in the 730 @xmath21 m band .
three of the grapes laes which we analyzed appear to be bright enough in the fir to be seen with alma , although only in the 730 @xmath21 m band .
these objects are 5183 , 6139 and 9040 . in pirzkal
( 2007 ) , object 5183 has a maximum dust extinction of a@xmath41 = 0.6 magnitudes ( equivalent to a@xmath9 @xmath5 2.4 ) , and object 6139 is best fit by 0.4 magnitudes of dust extinction .
this is consistent with our results , which imply that the uv light absorbed by this dust is being re - radiated in the fir .
while object 9040 was not best - fit by any dust , the authors do acknowledge that its best - fit model was still a poor fit , although they are confident that it is a lae at its measured redshift .
an important point is that the cdf - s laes have a higher detection fraction than the grapes laes .
this is likely due to differences in the sample selection method .
the cdf - s sample was selected via ground - based narrowband imaging ( with space - based broadband photometry ) .
however , the grapes sample was selected via space - based grism spectroscopy over a much smaller area ( only the hudf vs. the whole cdf - s ) , thus probing much deeper into the faint end of the luminosity function . as the detectability of an object scales not only with its color , but with its continuum brightness as well , fainter objects will be more difficult to detect .
figure 1 also plots the derived fir seds of the sample of 21 lbgs .
dust emission from 11 of the lbgs are detectable at 450 , 610 @xmath21 m and 730 @xmath21 m , with two additional lbgs being detectable at only 730 @xmath21 m . in a rough comparison between lbgs and laes ,
39% of the 23 laes are detectable with alma , compared with 62% for lbgs .
while these numbers may be consistent given the relatively small sample sizes , it could be indicative of a larger difference between the types of objects .
traditionally , lbgs are thought to harbor more evolved and more massive stellar populations than those in laes .
however , from this result , we now see that lbgs may also have a higher dust content than laes .
this is confirmed by the average @xmath22 of the two samples , which indicates that lbgs have a redder uv slope with @xmath42 = -1.62 , compared with @xmath43 = -2.03 .
a similar effect was seen in vanzella et al .
( 2006 ) , who found @xmath42 = -1.7 and @xmath43 = -2.1 , as well as in pentericci et al .
( 2007 ) , who found that lbgs were dustier than laes .
this is interesting , given that in f09 we found that our sample of laes had much higher dust extinction rates than we would have thought .
thus , while laes appear to have some amount of processed ism , the ism of lbgs appears to be much more enriched , resulting in higher dust emission in the fir . we have chosen a value of f@xmath37/f@xmath36 = 3.64 from bendo et al .
( 2006 ) , which corresponds to a dust temperature of 28 k. however , there is a large uncertainty in the fir color .
larger values of f@xmath37/f@xmath36 imply a cooler ism ( thus a cooler dust temperature ) , while smaller values imply a warm ism , more consistent with compact star formation ( dale et al .
dale et al . ( 2001 ) and dale and helou ( 2002 ) compute model fir seds assuming a large range in fir color , from 0.8 @xmath34 f@xmath37/f@xmath36 @xmath34 3.3 . since laes are expected to be undergoing significant bouts of star formation , and they are quite compact ( e.g. , pirzkal et al . 2007 ) , it is interesting to see how our derived fir seds would change if we assume a warmer value of f@xmath37/f@xmath36 .
the observations presented in dale & helou ( 2002 ) show that the number of objects drops off considerably below f@xmath37/f@xmath36 = 1 , so we assumed f@xmath37 = f@xmath36 , and redid our calculations . figure 2 shows these results for cs2 - 2 , which was one of the fainter laes detected in all alma bands .
since warmer fir seds will result in fainter fluxes in the alma observable range , this plot can give us an idea of the limits on our predicted fluxes .
we find that although the shape of the sed has changed as expected , this particular object would still be detected .
however , since the peak of the dust emission has moved toward shorter wavelengths , objects which were only detected in the 730 @xmath21 m band may be less likely to be detected if they have a warmer sed . as another illustration of our uncertainties ,
we have plotted the 1 @xmath44 errors on our original predicted 60 and 100 @xmath21 m fluxes , obtained from the photometric i and z errors .
we also plot error bars derived from the uncertainty in the m97 relation , as described in meurer et al .
( 1999 ; 0.04 dex uncertainty in log(f@xmath31/f@xmath32 ) ) . these errors are much smaller than the photometric errors .
however , it is worth deeper investigation into the extent which the photometric errors can alter our results . using equations 1 6
, we can calculate the dust emission from any object given a @xmath19 magnitude and a colour ( @xmath45 for redshifts less than 4.5 , and @xmath46 for redshifts greater than 4.5 ) .
figures 3 and 4 show these results at 730 @xmath21 m , with contours showing the flux levels for any colour - magnitude combination , and one contour specifically showing the alma 4 hour sensitivity at 730 @xmath21 m .
figure 3 plots all objects in our whole lae+lbg sample with z @xmath15 4.53 .
the mean redshift of this subsample is 4.33 , so the flux curves were derived for this redshift .
figure 4 plots all objects with z @xmath27 4.53 , with the flux curves derived for the mean redshift of this subsample , which was 5.26 .
the different coloured circles represent the three different samples ( cdf - s laes , hudf laes , and hudf lbgs ) , with filled circles representing objects we found that would be detectable with alma via figure 1 .
all filled circles lie above the alma 4 hour sensitivity contour ( black line ) showing that figures 1 and 3 & 4 are mutually consistent , although the three detectable hudf laes do lie less than 1@xmath44 above the alma 4 hour sensitivity contour . using these figures
, one could then take the colour , magnitude and rough redshift of any object , and determine if the dust emission could be detected by seeing if the points lie above the alma sensitivity contour .
figures 3 & 4 allow us to examine the robustness of our results , as we have visualized the 1@xmath44 errors on our objects fluxes .
we can thus see how close our objects are to the alma sensitivity limit as a function of their photometric errors .
while 39% of the laes will have their dust emission detectable , only 17% ( 4/23 ) of them lie more than 1@xmath44 from the alma sensitivity contour .
likewise , for lbgs , only 33% lie more than 1@xmath44 from the alma 4 hour sensitivity contour . however , some objects which lie below the alma contour are less than 1@xmath44 away .
thus , we can define a maximum percentage of objects with detectable dust emission , including not only those above the line , but those that are below but within 1@xmath44 .
we find this to be 61% for laes and 81% for lbgs .
this allows us to put an error bar on our lae detection fraction , of 39 @xmath1 22 % ( likewise 62 @xmath47 % for lbgs ) .
while the fir
@xmath22 relation was derived at low redshift , both seibert , heckman & meurer ( 2002 ) and reddy & steidel ( 2004 ) find that it also applies at high - redshift .
in addition , teplitz et al .
( 2006 ) found that when they compared fuv to spitzer observations , the results were in general agreement with the fir
@xmath22 relation .
chapman et al .
( 2005 ) have investigated this relation using a sample of sub - millimeter selected galaxies at z @xmath5 2 .
the authors conclude that the bolometric luminosity of sub - millimeter galaxies are underestimated by 2 orders of magnitude when derived from their uv properties .
this indicates that our derived fir fluxes may be lower limits , meaning that if anything we could be underestimating the detection fraction
. however , reddy et al . ( 2006 ) found that the m97 relation may not apply to all star - forming galaxies at z @xmath5 2 .
they find that galaxies with derived ages @xmath34 100 myr exhibit a smaller fir excess than that predicted by the m97 relation , indicating that these young galaxies have redder uv colors for a given amount of dust extinction that older galaxies .
this is of concern as most of our laes have very young derived stellar population ages . in addition ,
baker et al . ( 2006 ) find that the m97 relation also over - predicts the fir flux for the gravitationally lensed lbg cb58 . however , there are differences between these objects and laes , as they are at a lower redshift , and they are continuum selected .
we thus can not know if the same age dependence will apply to laes until we obtain a statistical sample of rest - frame fir detections of laes .
in addition , the reddy et al .
figure which illustrates the age discrepancy does not have error estimates on their young object points , thus it is hard to discern how significant this effect is .
baker et al .
do show error bars , and these errors imply that their observations are only off the m97 relation by @xmath5 1 @xmath44 . if young objects do have systematically lower f@xmath31/f@xmath32 ratios , it would be worrying , as it could imply that we are over - predicting the fir fluxes of our laes by as much as an order of magnitude .
it is thus prudent to apply our results to any relevant observations to see if our fir flux predictions are accurate .
although very few galaxies with ly@xmath2 emission have been detected in their rest - frame fir , capak et al .
( 2008 ) have recently published a study of a spectroscopically confirmed millimeter starburst galaxy , which does exhibit ly@xmath2 emission .
they publish the observed sed of this galaxy from x - rays to radio wavelengths . using the published optical magnitudes , we can compute @xmath22 , and predict the fir flux . doing this , we calculate a predicted 554 ( rest 100 ) @xmath21 m flux of 2637@xmath48 @xmath21jy . however , the closest point in the observed sed is at 1.1 mm , at 4800 @xmath1 1500 @xmath21jy .
the flux of our derived sed at 1.1 mm is 2054@xmath49 @xmath21jy , thus they are consistent within 1 @xmath44 .
these observations give evidence that our predictions do match observations .
we show the sed from capak et al . along with our predictions for this object in figure 5 .
additionally , we also show the v , i , z and 3.6 @xmath21 m fluxes for ch8 - 1 , which was one of our redder laes .
these points have been scaled to match the i flux of the capak et al .
we can see that the i - z color is somewhat redder than the capak et al .
object ( 0.3 versus 0.15 ) , which indicates that ch8 - 1 is dustier , explaining its higher expected fir flux .
however , as it is intrinsically fainter ( the scaling factor was @xmath5 17 ) , it would be much harder to detect with current instrumentation . although we have assumed a reasonable sed shape , it is possible that the fir sed could be much different .
houck et al .
( 2004 ) discuss the low metallicity blue compact dwarf galaxy sbs 0335 - 052 , which has a fir spectrum much different than typical starburst galaxies , peaking at @xmath5 28 @xmath21 m , indicating much warmer dust , possibly due to the lower metallicity .
high - redshift laes may be expected to have a very low metallicity , thus it is worth mentioning that if any of our objects had this type of sed , we could be vastly overestimating their fir flux .
we have thus investigated the dependence of our results on these derived fir fluxes . to do this
, we reduced our fir flux to be 50% of its derived value , and computed the alma detection fraction of both types of objects .
we calculated a detection fraction of 26% for laes and 62% for lbgs ( 57% and 76% for laes and lbgs respectively within 1@xmath44 of the alma contour ) .
thus , while we believe that the m97 relation is applicable to our objects , we find that our conclusions are robust to a factor of two variation in the fir flux .
directly observing dust emission from ly@xmath2 galaxies will provide undeniable proof that these earliest of galaxies have undergone some enrichment of their ism . while these observations are currently difficult , we have been able to predict whether a sample of 23 laes would be detected with a future sub - mm observatory such as alma based on their optical and nir fluxes . exploiting a relation from meurer et al .
( 1997 ) , which relates the far - infrared excess to the uv continuum slope , we have derived the fir fluxes from these laes , which will be primarily due to dust emission . by comparing these fluxes to the sensitivity of alma at 450 , 610 and 730 @xmath21 m ,
we have determined that 39% ( 9/23 ) of our sample of laes should be detectable based on their dust emission .
we perform the same analysis on a sample of 21 lyman break galaxies , finding that 62% ( 13/21 ) of the lbgs are detectable with alma , @xmath5 60% higher than the detection fraction of laes .
this may be a function of our small sample size , or it may be the result of a true dichotomy in the dust contents of laes and lbgs , providing further proof that lbgs are a more evolved class of high - z galaxy . in order to generalize our results , we computed the fir fluxes for a wide range of continuum magnitudes and colours , comparing these fluxes to the alma 4 hour sensitivity limit .
thus , knowing the continuum magnitude and colour of any object , one can then predict whether its dust emission will be detectable .
while stellar population modeling gives some insight on the amount of dust in high - redshift galaxies , future sub - mm instruments will allow us to directly observe the dust emission from these objects , providing a more complete census of dust . from our analysis of laes and lbgs in the hudf
, we calculate a number density of alma detectable laes and lbgs of 0.26 arcmin@xmath50 , and 1.13 arcmin@xmath50 respectively .
thus , even in a small survey area , one could expect to detect a number of high - redshift objects with alma in a reasonable amount of time . in our previous analysis of laes ,
the addition of the clumpiness parameter allows strong ly@xmath2 flux to an arbitrarily old and dusty galaxy , thus these alma observations will help to constrain the amount of dust , and thus the level of clumpiness .
these results will have considerable impact on several areas of galaxy evolution , specifically on the star formation rate ( e.g. , giavalisco et al .
1996 ) , which , being derived from uv observations , is critically dependent on accurate measurements of the dust content . combining alma with uv observations
, we will have a complete map of star formation in high - redshift galaxies ( i.e. gao 2008 ) .
studying the sub - millimeter emission of these objects will also allow calculation of the total dust mass , which is something one can not calculate from uv - optical data alone .
along with learning about the total amount of dust in these objects , alma will give us the opportunity to investigate the distribution of dust due to its excellent resolution . at 730 ghz
, alma can achieve resolutions down to 001 pix@xmath17 . combining alma observations with high resolution _
hst _ data will allow us to investigate dust on local scales at high - redshift for the first time .
strong sub - millimeter detections combined with strong ly@xmath2 emission could indicate that dust enhancement of the ly@xmath2 ew is widespread .
we thank the anonymous referee for their comments which greatly improved the fidelity of this paper . support for this work
was provided in part by nasa through grant numbers hst - ar-11249 , hst - go-10240 and hst - go-10530 from the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, under nasa contract nas5 - 26555 .
this work was also supported by the arizona state university ( asu ) department of physics and the asu school of earth and space exploration . baker , a. j. et al .
2006 , a&a , 372 , l37 beckwith , s. v. w. et al .
2006 , aj , 132 , 1729 bendo , g. j. et al 2006 , apj , 652 , 283 bruzual , g. & charlot , s. 2003 , mnras , 344 , 1000 calzetti , d. , kinney , a. l. & storchi - bergmann , t. 1994 , apj , 429 , 582 chapman , s. c. , blain , a. w. , smail , i. & ivison , r. j. 2005 , apj , 622 , 772 chary , r .- r . , stern , d. & eisenhardt , p. 2005
, apj , 635 , l5 capak , p. et al .
2008 , apj , 681 , l53 dale , d. a. et al .
2001 , apj , 549 , 215 dale , d. a. & helou , g. 2002 , apj , 576 , 159 dawson , s. , rhoads , j. e. , malhotra , s. , stern , d. , wang , j. x. , dey , a. , spinrad , h. & jannuzi , b. t. 2007 , apj , 671 , 1227 dawson , s. , rhoads , j. e. , malhotra , s. , stern , d. , dey , a. , spinrad , h. , jannuzi , b. t. , wang , j. x. & landes , e. 2004 , apj , 617 , 707 finkelstein , s. l. , rhoads , j. e. , malhotra , s. , pirzkal , n. & wang , j. x. 2007 , apj , 660 , 1023 finkelstein , s. l. , rhoads , j. e. , malhotra , s. , grogin , n. a. & wang , j. x. 2008a , apj , 678 , 655 finkelstein , s. l. , rhoads , j. e. , malhotra , s. & grogin , n. a. 2009 , apj accepted , astroph/0806.3269 gao , y. 2008 , nature , 452 , 417 giavalisco , m. , koratkar , a. & calzetti , d. 1996 , apj , 466 , 831 grazian , a. et al .
2006 , a&a , 449 , 951 hansen , m. & oh , s.
mnras , 367 , 979 hathi , n. p. , malhotra , s. & rhoads , j. e. 2008 , apj , 673 , 686 houck , j. r. et al .
2004 , apjs , 154 , 211 hu , e. m. , cowie , l. l. & mcmahon , r. g. 1998 , apj , 502 , l99 kudritzki , r .-
2000 , apj , 536 , 19 lai , k. et al .
2007 , apj , 655 , 704 malhotra , s. & rhoads , j. e. 2002 , apj , 565 , l71 malhotra , s. wang , j. , rhoads , j. e. , heckman , t. m. & norman , c. a. 2003 , apj , 585 , 25 malhotra , s. et al .
2005 , apj , 626 , 666 meurer , g. r. , heckman , t. m. , lehnert , m. d. , leitherer , c. & lowenthal , j. 1997 , aj , 112 , 54 m97 meurer , g. r. , heckman , t. m. & calzetti , d. 1999 , apj , 521 , 64 meier , d. l. & terlevich , r. 1981 , apj , 246 , l109 neufeld , d. a. 1991 , apj , 370 , l85 oke , j. b. & gunn , j. e. 1983 , apj , 266 , 713 pentericci , l. et al .
2007 , a&a , 471 , 433 pentericci , l. et al .
2008 , accepted to a&a pirzkal , n. et al .
2004 , apjs , 154 , 501 pirzkal , n. et al .
2007 , apj , 667 , 49 p07 reddy , n. a. & steidel , c. c. 2004 , apj , 603 , l13 reddy , n. a. et al .
2006 , apj , 644 , 792 rhoads , j. e. , malhotra , s. , dey , a. , stern , d. , spinrad , h. & jannuzi , b. t. 2000 , apj , 545 , l85 rhoads , j. e. et al .
2008 , submitted to apj , astroph/0805.1056 salpeter , e. e. 1955 , apj , 121 , 161 shimasaku , k. et al .
2006 , pasj,58 , 313 seibert , m. , heckman , t. m. & meurer , g. r. 2002 , aj , 124 , 46 spergel , d. n. et al .
2007 , apjs , 170 , 377 teplitz , h. i. et al .
2006 , aj , 132 , 853 vanzella , e. et al .
2006 , astroph/0612182 wang , j. x. , rhoads , j. e. , malhotra , s. , dawson , s. , stern , d. , dey , a. , heckman , t. m. , norman , c. a. & spinrad , h. 2004 , apj , 608 , l21 wilson , t. l. 2007 , astrophys space sci , springer , 10.1007 wuyts , s. et al .
2008 , astroph/0804.0615 xu , c. et al .
2007 , aj , 134 , 169 | recent results have shown that a substantial fraction of high - redshift lyman alpha galaxies contain considerable amounts of dust .
this implies that lyman alpha galaxies are not primordial , as has been thought in the past .
however , this dust has not been directly detected in emission ; rather it has been inferred based on extinction estimates from rest - frame ultraviolet ( uv ) and optical observations .
this can be tricky , as both dust and old stars redden galactic spectra at the wavelengths used to infer dust .
measuring dust emission directly from these galaxies is thus a more accurate way to estimate the total dust mass , giving us real physical information on the stellar populations and interstellar medium ( ism ) enrichment .
new generation instruments such as the atacama large millimeter array ( alma ) and sub - millimeter array ( sma ) , should be able to detect dust emission from some of these galaxies in the sub - mm . using measurements of the uv spectral slopes , we derive far - infrared flux predictions for of a sample of 23 z @xmath0 4 lyman alpha galaxies .
we find that in only a few hours , we can detect dust emission from 39 @xmath1 22% of our lyman alpha galaxies . comparing these results to those found from a sample of 21 lyman break galaxies ( lbgs )
, we find that lbgs are on average 60% more likely to be detected than ly@xmath2 galaxies , implying that they are more dusty , and thus indicating an evolutionary difference between these objects .
these observations will provide better constraints on dust in these galaxies than those derived from their uv and optical fluxes alone .
undeniable proof of dust in these galaxies could explain the larger than expected ly@xmath2 equivalent widths seen in many ly@xmath2 galaxies today .
[ firstpage ] galaxies : ism galaxies : high - redshift galaxies : evolution |
tunneling represents the wave mechanical analogy to the propagation of evanescent modes @xcite .
evanescent modes are observed , e.g. in the case of total reflection , in undersized waveguides , and in periodic dielectric heterostructures @xcite .
compared with the wave solutions an evanescent mode is characterized by a purely @xmath0 wave number , so that i.e. the wave equation yields for the electric field @xmath1 @xmath2 where @xmath3 is the angular frequency , t the time , x the distance , k the wave number , and @xmath4 the @xmath5 wave number of the evanescent mode .
thus evanescent modes are characterized by an exponential attenuation and a lack of phase shift .
the latter means that the mode has not spent time in the evanescent region , which in turn results in an infinite velocity in the phase time approximation neglecting the phase shift at the boundary @xcite .
two examples of electromagnetic structures in which evanescent modes exist are shown in fig.[examples ] @xcite .
the dispersion relations of the respective transmission coefficients are displayed in the same figure .
quite often a signal is said to be defined by switching on or off light .
it is assumed that the front of the light beam informs my neighbour of my arrival home with the speed of light .
such a signal is sketched in fig.[zeit ] .
the inevitable inertia of the light source causes an inclination of the signal s front and tail . due to the detector s sensitivity ,
level @xmath6 , the information about the neighbours arrival ( switching on ) and departure ( switching off ) becomes dependent on intensity .
in this example the departure time is detected earlier with the attenuated weak signal .
( a ) ( b ) ( c ) , @xmath7 and the tails are @xmath8 , @xmath9 .
they are measured at different places and times in consequence of the detector s responsitivity level @xmath10 . ]
this special signal does not transmit reliable information on arrival and departure time .
in addition to the dependence on intensity , a light front may be generated by any spontaneous emission process or accidentally by another neighbour . obviously , a detector needs more than a signal s front to response properly .
the detector needs information about carrier frequency and modulation of the signal in order to obtain reliable information about the cause . in general
, a signal is detected some time after the arrival of the light s front .
due to the dynamics of detecting systems there are several signal oscillations needed in order to produce an effect @xcite .
an @xmath11 is detected with the energy velocity . in vacuum or in a medium with normal dispersion
the signal velocity equals both , the energy and the group velocities .
hz and the amplitude modulation is limited to a frequency band width of about @xmath12hz.,title="fig : " ] + for example , a classical signal can be transmitted by the morse alphabet , in which each letter corresponds to a certain number of dots and dashes . in general signals
are either frequency ( fm ) or amplitude modulated ( am ) and they have in common that the signal does not depend on its magnitude . a modern signal transmission , where the halfwidth corresponds to the number of digits , is displayed in fig.[signal ] .
this am signal has an infra - red carrier with @xmath13 wave length and is glass fiber guided from transmitter to receiver . as mentioned above ,
the signals are independent of magnitude as the halfwidth does not depend on the signal s magnitude .
the front or very beginning of a signal is only well defined in the theoretical case of an infinite frequency spectrum .
however , physical generators only produce signals of finite spectra .
this is due to their inherent inertia and due to a signal s finite energy content .
these properties result in a real front which is defined by the measurable beginning of the signal .
for example the signals of fig.[signal ] have a detectable frequency band width of @xmath14 , where @xmath15 is the carrier frequency .
frequency band limitation in consequence of a finite signal energy reveals one of the fundamental deficiencies of classical physics . a classical detector can detect a deliberately small amount of energy , whereas every physical detector needs at least one quantum of the energy @xmath16 in order to respond
superluminal signal velocities have been measured by enders and nimtz @xcite .
the experiments were carried out with am microwaves in undersized waveguides and in periodic dielectric heterostructures .
the measured propagation time of a pulse is shown in fig.[impuls ] .
the microwave pulse has travelled either through air or it has crossed an evanescent barrier @xcite .
the linewidth of the pulse represents the signal .
the experimental result is , that the tunneled signal has passed the airborne signal at a superluminal velocity of 4.7 @xmath17 c. the measurements of the traversal time are carried out under vacuum - like conditions at the exit of the evanecscent region , the reason for this will be discussed later .
measured microwave signals are shown in fig.[impuls ] .
the halfwidth ( information ) of the tunneled signal has traversed the evanescent region at a velocity of 4.7 @xmath17 c. as explained above , signals have a limited frequency spectrum since their energy content w is always finite and detectable frequency components with @xmath18 can not exist . in this experiment
all frequency components of the signal are evanescent and move at a velocity faster than c. the beginning of the evanescent signal overtakes that of the airborne signal as seen in fig.[impuls ] .
the superluminal velocity of evanescent modes has some interesting features differing fundamentally from luminal or subluminal propagation of waves with @xmath19 wave numbers .
this will be discussed in the following subsections .
the existence of a superluminal signal velocity ensures the posssibility of an interchange of chronological order .
let us assume an inertial system @xmath20 moves away from system @xmath21 with a velocity @xmath22 .
the special relativity ( sr ) gives the following relationship for the travelling time @xmath23 and for the distance @xmath24 of a signal in the system @xmath21 which is watched in @xmath20 @xmath25 @xmath26 is the condition for the change of chronological order , i.e. @xmath27 , between the systems @xmath21 and @xmath20 .
for example , at a signal velocity @xmath28 10 c the chronological order changes at @xmath29 0.1 c. this result does not violate the sr . the common constraint @xmath30@xmath31
is posed by the sr on electromagnetic wave propagation in a dispersive medium and not on the propagation of evanescent modes .
the schrdinger equation yields a negative kinetic energy in the tunneling case , where the potential u is larger than the particle s total energy w : @xmath32 the same happens to evanescent modes . within the mathematical analogy
their kinetic electromagnetic energy is negative too .
the helmholtz equation for the electric field @xmath33 in a waveguide is given by the relationship @xmath34 where @xmath35 is the cut - off wave number of the evanescent regime .
the quantity ( @xmath36 - @xmath37 ) plays a role analogous to the energy eigenvalue and is negative in the case of evanescent modes .
the dielectric function @xmath38 of evanescent modes is negative and thus the refractive index is imaginary . for the basic mode a rectangular waveguide has the following dispersion of its dielectric function , where @xmath35 = @xmath39 holds and @xmath40 is the waveguide width ,
@xmath41 @xmath42 is the free space wavelength of the electromagnetic wave . in the case of tunneling it
is argued that a particle can only be measured in the barrier with a photon having an energy @xmath43 ( u - w ) @xcite .
this means that the total energy of the system is positive .
according to eq.(5 ) the evanescent mode s electric energy density @xmath44 is given by the relationship @xmath45 where @xmath46 is the electric permeability .
the analogy between the schrdinger equation and the helmholtz equation holds again and it is not possible to measure an evansecent mode .
achim enders and i tried hard to measure evanescent modes with probes put into the evanescent region but failed @xcite . obviously evanescent modes are not directly measurable in analogy to a particle in a tunnel .
we might also say this problem is due to impedance mismatch between the evanescent mode and a probe .
the impedance z of the basic mode in a rectangular waveguide is given by the relationship @xmath47 where @xmath48 is the free space impedance .
in the evanesecent regime @xmath49 the impedance is imaginary .
evanescent modes do not experience a phase shift inside the evanescent region @xcite .
they cross this region without consuming time .
the predicted @xcite and the measured @xcite time delay happens at the boundary between the wave and the evanescent mode regime .
for opaque barriers ( i.e. @xmath50,where @xmath51 is the @xmath52 wave number and @xmath53 the length of the evanescent barrier ) the phase shift becomes constant with @xmath54 which corresponds to one oscillation time of the mode .
in fact , the measured barrier traversal time was roughly equal to the reciprocal frequency in the microwave as well as in the optical experiments , i.e. either in the 100 ps or in the 2 fs time range independent of the barrier length @xcite .
the latter behaviour is called _ hartman effect _ :
the tunneling time is independent of barrier length and has indeed been measured with microwave pulses thirty years after its prediction @xcite .
evanescent modes show some amazing properties with which we are not familiar . for instance
the evansecent region is not causal since evanescent modes do not spend time there .
this is an experimental result due to the fact that the traversal time is independent of barrier length .
another strange experience in classical physics is that evanescent fields can not be measured .
this is due to their @xmath55 energy or to the impedance mismatch .
amazingly enough this is in analogy with the wave mechanical tunneling .
the energy of signals is always finite resulting in a limited frequency spectrum both according to planck s energy quantum @xmath56 .
this is a fundamental deficiency of classical physics , which assumes the measurability of any small amount of energy .
a physical signal never has an ideal front .
the latter needs infinite high frequency components with a correspondingly high energy .
another consequence of the frequency band limitation of signals is , if they have only evanescent mode components , they are not lorentz - invariant .
the signal may travel faster than light .
front , group , signal , and energy velocities all have the same value in vacuum .
bearing in mind the narrow frequency band of signals , the former statement holds also for the velocities of evanescent modes . in first order approximation
the dispersion relation of a stop band is constant and a significant pulse reshaping does not take place .
this result demonstrate that signals and effects may transmitted with superluminal velocities provided that they are carried by evanescent modes .
stimulating discussions with v. grunow , d. kreimer , p. mittelstaedt , r. pelster , and h. toyatt are gratefully acknowledged .
99 arnold sommerfeld , vorlesungen ber theoretische physik , band iv , optik dieterchsche verlagsbuchhandlung ( 1950 ) g.nimtz and w.heitmann , prog.quantum electronics * 21*,81(1997 ) th.hartman , j.appl.phys .
@xmath57,3427(1962 ) a.enders and g.nimtz , j.phys .
i ( france ) @xmath58,1693(1992 ) + a.enders and g.nimtz , j.phys .
i ( france ) @xmath59,1089(1993 ) a.enders and g.nimtz , phys.rev .
e * 48*,632(1994 ) | it recently has been demonstrated that signals conveyed by evanescent modes can travel faster than light . in this report
some special features of signals are introduced and investigated , for instance the fundamental property that signals are frequency band limited .
evanescent modes are characterized by extraordinary properties : their energy is _ negative _ , they are not directly measurable , and the evanescent region is not causal since the modes traverse this region instantaneously .
the study demonstrates the necessity of quantum mechanical principles in order to interprete the superluminal signal velocity of classical evanescent modes . |
quantum fisher information ( qfi)@xcite which is one of the most important ingredients in parameter estimation fields has been extensively investigated in both the classical and quantum worlds .
qfi characterizing the sensitivity of a state with respect to perturbation of the parameters is employed as a measure to quantify the precision of parameter estimation .
the parameter estimation is the purpose of the information in a finite number of observation value to obtain highest precision of the parameter and design schemes to attain that precision which is bounded by the quantum cramr - rao inequality,@xcite i.e. , @xmath7 , where @xmath8 is the number of experiments and @xmath9 denotes the qfi of the parameter @xmath4 .
this inequality indicates that the inverse of the qfi provides the lower bound of the error of the estimation . on the other hand
, qfi is formally generalized from the classical one , and there are several variants of quantum versions of fisher information@xcite , among which the one based on the symmetric logarithmic derivative operator has been used most widely .
for example , assume that a parameter @xmath4 is encoded in quantum states @xmath10 , and then the qfi of @xmath4 is defined as@xcite @xmath11,\ ] ] where @xmath12 is the so - called symmetric logarithmic derivative , which is determined by @xmath13 with @xmath14 .
making use of the spectrum decomposition , one can diagonalize the matrix as @xmath15 . and
then the qfi can be rewritten as @xcite @xmath16 where @xmath17 is the qfi for pure state @xmath18 with the form @xmath19 $ ] . given any single qubit states @xmath20 , one can rewrite the form as @xmath21 in the bloch sphere representation , where @xmath22 is the real bloch vector and @xmath23 denotes the pauli matrices . according to ref.@xcite ,
a simple and explicit expression of qfi could be obtained in terms of the bloch sphere representation @xmath24 @xmath25 from the perspective of parameter estimation , the larger qfi represents the higher estimation precision in general .
hence , how to improve the qfi becomes a key problem to be solved .
making use of quantum resources such as coherence , entanglement and so on , one can find that qfi of quantum systems can provide much more sensitivity than the classical ones even to beat the shot noise limit in principle , in particular , under some circumstances , one can even achieve the heisenberg limit @xcite .
however , in the realistic physical parameter estimation process , the quantum system unavoidably interacts with the surrounding environments , so that the qfis dynamics decreases monotonically with time resulting in the decrease of parameter estimation precision . in recent years , how to improve the estimation precision has become a significant issue in both experimentally and theoretically .
different protocols and strategies have been proposed and realized to improve the precisions of various parameters .
for example , tan et al . reported the improvement precision of parameter estimation in atom interferometer under dephasing noise by using dynamical decoupling pulses @xcite .
zhang et al . proposed a scheme to enhance the precision of quantum estimation by increasing entanglement of the input state @xcite .
inspired by the classical driving , abdel - khalek studied the qfi for a two - level atom system driven by a phase noise laser under non - markovian dynamics @xcite .
later , parameter estimation precision of a non - markovian dissipative two - state system by classical driving is also studied by li @xcite .
taking non - markovian effect into consideration , a qubit system in non - markovian environment becomes a good candidate for implementation of quantum optics schemes and information with high precision @xcite . applying quantum feedback , zheng et al . analyzed the parameter precision of optimal quantum estimation of a dissipative qubit @xcite . besides
, the enhancement of the precision of phase estimation in quantum metrology is investigated by employing weak measurement and quantum measurement reversal @xcite .
so far , most previous literature studying the precision of parameter estimation have been mainly focused on the hermitian or quasi - hermitian systems which govern the interest system unitary evolution @xcite . in particular , others have been focused on estimating an overall multiplicative factor of a hamiltonian @xcite , for example , estimating @xmath26 in a hamiltonian @xmath27 @xcite . in a recent work
@xcite , liu et al . studied a general hamiltonian parameter estimation problem where the qfi for an exponential form initial state involved in a hermitian operator .
they found that the qfi is totally determined by this hermitian hamiltonian and the initial state .
however , to the best of our knowledge , few detailed investigations are concern on estimating a non - hamiltonian parameter .
motivated by these works , we here take advantage of the presence of @xmath0 operator instead of hermitian one . in particular , we would like to answer questions such as the following : for a @xmath0 hamiltonian , is it much more efficient than hermitian one to increase the precision of parameter estimation ?
how does the @xmath0 process achieve better precision than hermitian one ? to anticipate these questions , this paper extends parameter estimation to a @xmath0 case , we propose two schemes to enhance the parameters estimation by applying property of the non - hermitian hamiltonian process .
the proposed schemes are related to apply the @xmath1 or the @xmath2 non - hermitian operators process .
the former case corresponding to before the interest system suffers from decoherence channel , a non - hermitian hamiltonian operator is operated , and the later case is opposite . in contrast with the results obtained in ref.@xcite , we derive the exact expressions of the qfis with respect to the amplitude parameter @xmath3 and the phase parameter @xmath4 in terms of the bloch sphere representation , and in detail investigate the influence of non - hermitian operator on the qfis . from our results , it turns out that ( i)one can improve the precision of parameter estimation by applying the non - hermitian operator process .
( ii)the qfis not only depend on the decoherence rate @xmath28 , but also depend on the non - hermitian parameter ( @xmath29 , @xmath30 ) .
( iii)the @xmath2 non - hermitian operators process is superior to the @xmath1 one to enhance the precision of parameter estimation . finally , a generalized non - hermitian operator process effect on the parameters estimation is also considered .
estimation of parameter in non - hermitian operators processing .
after the phase - gate operation , ( a ) a non - hermitian operator is performed after the interest system interacts with a environment ; ( b ) a non - hermitian operator is applied before the interest system interacts with a environment.,width=283 ] as we are all known , in conventional quantum mechanics , any physical obervables must be represented by self - adjoint or hermitian operators in a hilbert space , e.g. @xmath31 .
the hamiltonian @xmath32 of a closed system in conventional quantum mechanics is hermitian , which requires not only the eigenvalues are real , but also the time - evolution of the system is unitary . however , bender et al .
@xcitehave recently found that the parity - time - symmetric hamiltonian , which is non - hermitian , can also have real energy spectra under some conditions .
given a hermitian hamiltonian @xmath32 of the interest system , the evolution of the final state from time zero to @xmath33 is given by a map @xmath34 , where @xmath35 is a time evolution operator .
according to conventional quantum theory , the evolution of the state is also generated by the schr@xmath36dinger equation @xmath37throughout this paper , we set @xmath38 .
the corresponding density matrix of the entire system can be written as @xmath39 , which satisfies @xmath40 and @xmath41 . in what follows we consider a different scenario by replacing the @xmath42 hamiltonian with a @xmath0 one .
for simplified , we assume this non - hermitian hamiltonian is @xcite @xmath43where @xmath44 and @xmath45 are real numbers . based on bender et al @xcite ,
it is shown that the non - hermiticity of @xmath32 can have real eigenvalues if it possesses parity - time symmetry , namely @xmath46 $ ] . here
the operator @xmath47 is unitary and it is given by transform @xmath48 , @xmath49 , and the operator @xmath50 is also unitary , one can make a transform @xmath51 , @xmath49 , and @xmath52 .
hence , it is not difficult to conclude the eigenvalues of @xmath32 real satisfy the condition @xmath53 and its eigenvalues are @xmath54 . according to non - hermitian quantum theory , to obtain the evolution of the system , one should introduce a time - evolving operator @xmath55 , @xmath56defining @xmath57 which is related to @xmath44 when fixed @xmath58 .
different from the conventional quantum theory , the evolution of the system with a non - hermitian hamiltonian is not trace - preserving in general @xmath59therefore , make sure the probability of normalization , we renormalize the system density operator @xmath60}. \label{renorm22}\]]in the following , we utilize this non - hermitian hamiltonian process given by eq .
( [ unitary ] ) to enhance and preserve qfi suffering from decoherence . before studying the precision of parameter estimation in non - hermitian operator process
, we firstly review the framework of parameter estimation .
usually the precision of parameter estimation is composed of three stages : the first stage is the preparation of initial state , the second stage is the sampling stage involved in the parameter code , and the third stage is the measurement corresponding to the parameter estimate . in following , we will adopt this framework to investigate the effect of non - hermitian operators process on the parameter estimation of the amplitude parameter @xmath3 and the phase parameter @xmath4 in terms of qfis for a two - level qubit subjected to a decoherence noisy environment .
we start with the scheme @xmath61 as shown in figure 1a . for simplicity , we consider the same scenario as ref .
@xcite , where an open system consisted of a two - level qubit subjected to an amplitude damping decoherence noisy environment after the phase gate @xmath62 is performed .
we assume that the input state is a two - level superposition state @xmath63 .
after the phase gate operation @xmath64 operated on the input state @xmath65 , the output state is given by @xmath66 which contains two parameters , one is the amplitude parameter @xmath3 and the other is the phase parameter @xmath4 .
next , let the state @xmath20 subject to the noisy environment , e.g .
an amplitude damping decoherence environment , which is described as the quantum operation @xmath67 . due to noisy decoherence
, the state @xmath68 inevitably evolves into a mixed , namely @xmath69 , where @xmath70 , and @xmath71 . in the next step of the scheme
we perform the non - hermiticity operation eq .
( [ unitary ] ) on the open system after the implementation of the amplitude damping decoherence process .
then the finial evolution state @xmath72 of the system is obtained @xmath73as discussed above , the evolution of a system with a non - hermitian hamiltonian is not trace - preserving and has to be re - normalized @xmath74 . in the standard qubit basis @xmath75
, the non - zero elements of @xmath76 are described in terms of the bloch sphere representation given by @xmath77 @xmath78 @xmath79 where @xmath80 and @xmath81 .
finally , the scheme ends with one implementing a measurement on this finial state . substituting eqs.(10)-(12 ) into eq .
( [ eq0 ] ) , it is not difficult to calculate the qfis of @xmath76 with respect to @xmath3 and @xmath4 .
the analytical expressions of the qfis can exactly be obtained @xmath82 ^ 2\nonumber\\ & + & 4\cos^4\alpha[\delta\cos2\tau\cos\alpha\cos\theta\sin\phi+\delta\sin2\tau(\delta^2\sin\alpha\sin\phi+\eta^2\cos\theta\sin\alpha\sin\phi-\delta\sin\theta)]^2\nonumber\\ & + & 4\eta^2\delta^2\cos^2\alpha\sin^2\frac{\theta}{2}[(2-\cos2\tau+\cos2(\tau-\alpha))\cos\frac{\theta}{2}-4\delta\sin^2\tau\sin\alpha\sin\phi\sin\frac{\theta}{2}]^2\nonumber\\ & + & [ \delta\sin\phi\sin2\alpha(\cos^2\tau+\eta^2\cos^2\tau\cos\theta+\eta^2+\eta^2\sin^2\tau)-2\delta\sin\phi\sin2\tau\cos^2\alpha\cos\theta\nonumber\\ & + & \delta^2\cos\alpha(1 - 2\cos2\tau-\cos2\alpha)\sin\theta+\delta\sin2\alpha\sin\phi(\delta^2\cos^2\tau-\eta^2\cos\theta(\sin^2\tau+1)-2)]^2\ } , \label{master1}\end{aligned}\ ] ] @xmath83 ^ 2\ } , \label{master2}\end{aligned}\ ] ] where @xmath84 ^ 4 $ ] . the calculations show the qfis are deeply related to the input initial state parameters ( @xmath3 , @xmath4 ) , the non - hermitian parameters ( @xmath29 , @xmath30 ) and the decoherence rate @xmath28 . before investigating the non - hermitian - assisted enhancement precision of parameter estimation
, one should maximize the qfis for all input states .
it is easy to determine from eqs .
( [ master1 ] ) and ( [ master2 ] ) that the optimal input state is @xmath85 which is defined by the analytic calculation of @xmath86 and @xmath87 .
fig.2 shows the qfi with respect to parameter @xmath3 as a function of @xmath30 ( @xmath88 ) for different phase parameter @xmath4 .
it can be seen qfi exhibits periodically oscillation .
qfi possesses the maximum and minimum values depending on @xmath4 . with the phase parameter @xmath4 increasing , the transitions of qfi from concave into convex function . for @xmath89
, qfi has the minimum values , while @xmath90 , qfi has the maximum values .
hence , one might deduce that the optimal input state is @xmath85 and @xmath90 .
in fact , this deduction is true due to @xmath91 and @xmath92 .
however , it is worth noting that the initial value of qfi is independent of @xmath4 , but dependent of the decoherence rate @xmath28 .
one can easily determine from eqs .
( [ master1 ] ) and ( [ master2 ] ) which reduces to @xmath93 at initial time @xmath94 .
the initial value of qfi varies from 0 to 1 . moreover , for @xmath95 or @xmath96(@xmath97 ) , eqs .
( [ master1 ] ) and ( [ master2 ] ) also reduce to @xmath93 .
these results are consistent with ref @xcite , where the analytical expression of the qfi dynamics under amplitude damping decoherence channel without the @xmath0 process . note that the qfi with respect to parameter @xmath3 is the same as the one with respect to parameter @xmath4 when the optimal input state parameter is @xmath85 . for simplicity , in the remainder of the section we only consider the qfi with respect to parameter @xmath3 .
qfi of the amplitude parameter @xmath3 as a function of @xmath30 ( @xmath98 ) for different phase parameters @xmath4 .
here we take @xmath85 , @xmath99 and @xmath100.,width=207 ] in following , we restrict our investigation to the optimal input states ( @xmath85 and @xmath90 ) , and then eq . ( [ master1 ] ) reduces to @xmath101 ^ 2 } { [ 2 - 2\sin\alpha(2\delta\sin^2\tau+\cos2\tau\sin\alpha)+\eta^2\sin2\tau\sin2\alpha]^4}.\end{aligned}\ ] ] to quantify the efficiency of the @xmath0 process on enhancing qfi , and concentrate only on how these process achieve better precision than hermitian one , we introduce the difference of the qfi between with and without the implementation of @xmath0 operator .
@xmath102 where @xmath103 corresponds to the pure decoherence process .
according to ref .
@xcite , the analytical expression of the qfi under amplitude damping decoherence channel is @xmath104 . based on eq .
( [ eq4 ] ) , @xmath105 implies enhancement precision of parameter estimation , while @xmath106 corresponds to decrease precision of parameter estimation .
fig.3 clearly shows that qfis with respect to the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath29 and @xmath28 for @xmath107 . from this graph
, one can see clearly that the @xmath2 non - hermitian operator process can greatly enhance the qfi . if fixed parameter @xmath30 , one can choose appropriate non - hermitian operator parameters @xmath29 to make @xmath105 , which reflects to enhance the parameters estimation . to illustrate parameters @xmath29 effect more clearly , we plot qfi as a function of @xmath30 for different parameters @xmath29 in fig.4 .
several distinct features can be found from this graph : first , the evolution of qfi displays the same periodical oscillation independent of @xmath29 .
second , when @xmath95 , qfi is invariable and @xmath108 for the fixed value @xmath28 .
third , when @xmath109 , there exists @xmath110 , while @xmath111 is related to @xmath112 . in the above , only one of non - hermitian parameters case is discussed . in fig.5
we show qfis as a function of @xmath30 and @xmath28 for @xmath99 . from this graph
, it can be clearly seen that the @xmath2 non - hermitian operators process can greatly enhance the qfis . by fixed parameter @xmath29
, one can choose appropriate parameters @xmath30@xmath113 to make @xmath105 .
besides , the optimal parameters @xmath29 and @xmath30 could be obtained by calculating the following conditions : @xmath114 and @xmath115 correspond to @xmath116 and @xmath117 , respectively . by numerical calculation ,
we plot qfi with respect to the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath30 and @xmath29 for @xmath100 .
it can be seen that the maximum @xmath118 corresponds to parameters @xmath119 and @xmath120 as shown in fig.6 .
qfis of the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath29 and @xmath28 with @xmath107 .
the red lower surface represents @xmath103 and the upper surface represents @xmath121 .
here we take @xmath85 and @xmath90.,width=207 ] qfi of the amplitude parameter @xmath3 as a function of @xmath30 for different parameter @xmath29 . here
we take @xmath85 , @xmath90 and @xmath100.,width=207 ] qfis of the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath30 and @xmath28 with @xmath99 .
the red lower surface represents @xmath103 and the upper surface represents @xmath121 .
here we take @xmath85 and @xmath90.,width=207 ] qfis of the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath30 and @xmath29 with @xmath100 . here
we take @xmath85 and @xmath90.,width=207 ] based on scheme @xmath61 above , even though the precision of parameter estimation in @xmath5 operator process can be increased if the parameters @xmath29 and @xmath30 are chosen appropriately , there are still some limitations .
following , we show an improved scheme that can completely circumvent the decoherence .
the whole procedure is depicted in fig.1 @xmath123 , we first perform the @xmath32 operator on the qubit system before it goes through the amplitude damping channel .
the output state is given by @xmath124}$ ] , and the corresponding three bloch vector components are @xmath125 @xmath126}{2[\cos2\tau+\cos\theta\sin2\tau\tan\alpha+2\sec\alpha\sin^2\tau(\sec\alpha-\sin\theta\sin\phi\tan\alpha)]},\end{aligned}\ ] ] @xmath127}{\cos2\tau+\cos\theta\sin2\tau\tan\alpha+2\sec\alpha\sin^2\tau(\sec\alpha-\sin\theta\sin\phi\tan\alpha)}.\end{aligned}\ ] ] similarly to the first scheme , we restrict our investigation to the optimal input states ( @xmath85 and @xmath90 ) . substituting eqs.(17)-(19 ) into eq .
( [ eq0 ] ) , then the analytical expressions of the qfi of parameter @xmath3 ( @xmath4 ) can be obtained @xmath128 from the above equation , it can be seen the qfi only depends on the non - hermitian parameters ( @xmath29 , @xmath30 ) and the decoherence rate @xmath28 . in particular ,
when @xmath95 or @xmath96 , eq .
( [ eq5 ] ) reduces to @xmath129 . in fig.7
we plot qfis with respect to the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath29 and @xmath28 for @xmath107 .
one can see clearly that the @xmath1 non - hermitian operator process could completely eliminate the effect of decoherence to improve the parameters estimation .
the larger the non - hermitian parameter @xmath29 is , the higher the estimation precision . on the other hand , we also plot qfis with respect to the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath30 and @xmath28 for @xmath99 . from this graph , it is clearly shown that qfis in the presence of the @xmath1 non - hermitian operator process could be greatly improved .
in contrast to the results obtained in ref.@xcite , our schemes i and ii all can enhance and preserve the qfis .
the underlying physical mechanism can be understood as follows : according to eq.(7 ) , the evolution of a system with a non - hermitian hamiltonian is not trace preserving .
this means the non - hermitian operation is not a genuine local operation in the conventional quantum world , so that the non - hermitian natural has less than unity probability .
therefor , the enhancement of qfis in our schemes can be attributed to the probabilistic nature of the non - hermitian . from this viewpoint ,
the success probability of improving qfis in scheme i is @xmath130,\end{aligned}\ ] ] which not only depends on the non - hermitian parameter ( @xmath29 , @xmath30 ) , but also depends on the decoherence rate @xmath28 . while the success probability of improving qfis in scheme ii is @xmath131 which is only dependent of the non - hermitian parameters ( @xmath29 , @xmath30 ) but independent of the decoherence rate @xmath28 .
however , by compared with scheme i , scheme ii is much better to enhance the parameters estimation .
this indicates the input initial state performed the non - hermitian operator is much more robust against the amplitude damping decoherence than the case where the input initial state in the absence of the non - hermitian operator performing goes through the decoherent channel .
estimation of parameter in non - hermitian operators process .
qfi of the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath29 and @xmath28 with @xmath107 .
the red lower surface represents qfi @xmath103 and the upper surface represents qfi @xmath132 . here
we take @xmath85 and @xmath90.,width=207 ] estimation of parameter in non - hermitian operators process .
qfi of the amplitude parameter @xmath3 ( the phase parameter @xmath4 ) as a function of @xmath30 and @xmath28 with @xmath99 .
the red lower surface represents qfi @xmath103 and the upper surface represents qfi @xmath132 . here
we take @xmath85 and @xmath90.,width=207 ] in the above discussion , we have assumed a specific non - hermitian hamiltonian with parity - time symmetry . here
, we consider a generalized non - hermitian hamiltonian @xmath133where @xmath134 are all real numbers .
one can easily conclude this hamiltonian violates parity - time symmetry and corresponds to the non - hermitian operator process which is given @xmath135 where @xmath136 . according to non - hermitian quantum theory ,
the evolution of a system with a non - hermitian hamiltonian is not trace - preserving .
when we deal with the non - hermitian process instead of eq.(6 ) , the renormalization procedure is also introduced .
therefore , if we repeat the processes of scheme i and ii , the same phenomenon is also observed for a generalized non - hermitian hamiltonian .
namely , the precision of parameter estimation by applying the non - hermitian operator process can be probabilistically improved if chosen appropriate non - hermitian parameters @xmath134 .
in this work , we propose two schemes to enhance the parameters estimation by applying property of the non - hermitian hamiltonian procession . the first one , where the actions of the @xmath1 non - hermitian operator process before the system under decoherence , the second one dealing with the @xmath2 non - hermitian operators process after the system under decoherence .
we calculate the exact expressions of the qfis with respect to the amplitude parameter @xmath3 and the phase parameter @xmath4 , and in detail investigate the influence of @xmath0 operator parameters on the qfis . from our results , it turns out that : ( i)the qfis not only depend on the input initial state ( @xmath3 , @xmath4 ) , but also depend on the non - hermitian parameter ( @xmath29 , @xmath30 ) .
( ii)one can improve the precision of parameter estimation by applying the @xmath1 non - hermitian operator process . finally , a generalized non - hermitian operator process effect on the parameters estimation is also considered .
our results provide an active way to suppress decoherence and enhance the parameter - estimation precision , which is rather significant in quantum precision measurement and quantum metrology .
this research is supported by the funds of the national natural science foundation of china under ( grant no .
11374096 ) , the natural science foundation of hunan province ( grant no .
2016jj2045 ) , the start - up funds for talent introduction and scientific research of changsha university 2015 ( sf1504 ) and scientific research project of hunan province department of education ( 16c0134 and 16c0469 ) 51ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty c. w. helstrom , _ quantum detection and estimation theory _
( academic press , new york , 1976 ) .
s. l. braunstein , and c. m. caves , _ phys .
lett . _ * 72 * , 3439 - 3443 ( 1994 ) .
a. s. holevo,_probabilistic and statistical aspects of quantum theory _ ( northholland , amsterdam , 1982 ) .
e. p. wigner , and m. m. yanase , _ proc .
sci . _ * 49 * , 910 ( 1963 ) r. a. fisher , _ proc.cambridge philos .
* 22 * , 700 ( 1925 ) .
s. l. luo , _ phys .
lett . _ * 91 * , 180403 ( 2003 ) j. liu , x. jing , and x. wang , _ phys .
a _ * 88 * , 042316 ( 2013 ) x. xiao , y. yao , w. j. zhong , y. l. li , and y. m. xie , _ phys .
rev . a _ * 93 * , 012307 ( 2016 ) .
s. boixo , a. datta , s. t. flammia , a. shaji , e. bagan , and c. m. caves , _ phys .
rev . a _ * 77 * , 012317 ( 2008 ) .
m. napolitano , and m. w. mitchell , _ new j. phys .
_ * 12 * , 093016 ( 2010 ) . m. napolitano , m. koschorreck , b. dubost , n. behbood , r. j. sewell , and m. w. mitchell , _ nature ( london ) _ * 471 * , 486 ( 2011 )
. q. s. tan , y. x. huang , l. m. kuang , and x. g. wang , _ phys .
a _ * 89 * , 063604 ( 2014 ) . q. s. tan , y. x. huang , x. l. yin , l. m. kuang , and x. g. wang , _ phys .
a _ * 87 * , 032102 ( 2013 ) . y. m. zhang , x. w. li , w. yang , and g. r. jin , _ phys .
a _ * 88 * , 043832 ( 2013 ) . y. l. li , x. xiao , and y. yao , _ phys .
rev . a _ * 91 * , 052105 ( 2015 ) .
s. abdel - khalek , _ annals of phys._. * 351*,0952 - 959 ( 2014 ) .
k. berrada , _ phys .
a _ * 88 * , 035806 ( 2013 ) . k. berrada , _ j. opt
. soc . am . b _ * 32 * , 571 ( 2015 ) .
w. zhong , z. sun , j. ma , x. g. wang , and franco nori , _ phys .
a _ * 87 * , 022337 ( 2013 ) .
x. x. jing , j. liu , h. n. xiong , and x. g. wang , _ phys .
a _ * 92 * , 012312 ( 2015 ) . s. pang , and todd a. brun , _ phys .
a _ * 90 * , 022117 ( 2014 ) . j. liu , x. x. jing , and x. g. wang , _ sci .
rep . _ * 5 * , 8565 ( 2015 ) .
m. d. lang , and c. m. caves , _ phys .
lett . _ * 111 * , 173601 ( 2013 ) .
m. jarzyna , and r. demkowicz - dobrzanski , _ phys .
rev . a _ * 85 * , 011801 ( 2012 ) .
j. joo , k. park , h. jeong , w. j. munro , k. nemoto , and t. p. spiller , _ phys . rev .
a _ * 86 * , 043828 ( 2012 ) . y. m. zhang , x. w. li , w. yang , and g. r. jin , _ phys .
a _ * 88 * , 043832 ( 2013 ) . | recently , zhong et al . [ phys . rev . a 87 , 022337(2013 ) ] investigated the dynamics of quantum fisher information ( qfi ) in the presence of decoherence .
we here reform their results and propose two schemes to enhance and preserve the qfis for a qubit system subjected to a decoherence noisy environment by applying @xmath0 operator process .
the proposed schemes are related to the @xmath1 ( or the @xmath2 ) non - hermitian operator process corresponding to before ( or after ) the interest system suffers from decoherence , respectively . resorting to the bloch sphere representation , we derive the exact analytical expressions of the qfis with respect to the amplitude parameter @xmath3 and the phase parameter @xmath4 , and in detail investigate the influence of @xmath0 operator parameters on the qfis . compared with pure decoherence process ( without non - hermitian operator process ) , we find that the @xmath5 operator process can potentially enhance and preserve the qfis by choosing appropriate @xmath0 operator parameters , while with the help of the @xmath6 operator process one could completely eliminate the effect of decoherence to improve the parameters estimation . finally , a generalized non - hermitian operator parameters effect on the parameters estimation is also considered . |
the linking number @xmath1 of disjoint , oriented curves @xmath2 is among the most basic invariants in knot theory , and as such it admits many different descriptions .
we begin with three .
let @xmath3 be euclidean coordinates on @xmath0 and consider the global angular form @xcite @xmath4^{3/2}}\,\in\,\omega^2\big(\br^3-\{0\}\big)\,.\ ] ] with the given normalization , @xmath5 is the pullback of an , unit - volume form @xmath6 on @xmath7 under the retraction @xmath8^{1/2}}\ , .
\end{aligned}\ ] ] without loss , we take the embedded curves @xmath9 to be parametrized by smooth maps @xmath10 in terms of which we write the difference @xmath11 the gauss formula for the linking number of @xmath12 and @xmath13 is then given by an integral over the torus @xmath14 , @xmath15 essential here , since @xmath12 and @xmath13 are disjoint space curves , the singularity of @xmath16 at the origin is avoided , and the gauss integrand is everywhere smooth and bounded on @xmath17 . alternatively , because @xmath6 represents a generator for the cohomology @xmath18 , the gauss linking integral computes the topological degree @xmath19 where @xmath20 is the composition @xmath21 in its second description as the degree of @xmath20 , the linking number is clearly an integer and invariant under smooth isotopies of the curves @xmath22 .
the overall sign of the linking number depends upon the orientations for @xmath17 and @xmath23 . throughout , if @xmath24 are angular coordinates on @xmath17 , we orient the torus by @xmath25 .
we similarly give @xmath26 the orientation induced from the standard orientation @xmath27 , for which the cohomology generator @xmath28 is positive .
@xmath29{over.eps}\qquad & \qquad & \includegraphics[scale=0.50]{under.eps}\\ & \operatorname{\mathsf w}=+1 \qquad & \qquad & \operatorname{\mathsf w}=-1 \end{matrix}\ ] ] famously the linking number admits a third , diagrammatic description , convenient for computations . again without loss , we consider a generic plane projection for @xmath12 and @xmath13 in which only simple , double - point crossings are present .
the index set @xmath30 of all crossings in the planar diagram for the link divides into subsets @xmath31 , depending upon whether the upper and lower strands at each crossing belong respectively to @xmath12 or @xmath13 .
to each crossing @xmath32 we attach a local writhe @xmath33 according to the handedness , as in figure [ writhe ] . in terms of these data ,
the linking number is computed by any of the following sums , @xmath34 the signed sum of crossings by @xmath13 over @xmath12 can be interpreted as a signed count of preimages @xmath35 for @xmath36 at the north pole , so the first equality in follows directly from the topological description of @xmath1 as the degree of @xmath20 .
one can check that our orientation conventions for @xmath17 and @xmath23 are consistent with the assignment of signs for @xmath37 in figure [ writhe ] .
the second equality in follows similarly with @xmath36 at the south pole , and the third equality is the symmetric combination of the preceding two .
the present article is concerned not with linking but with self - linking , for which one would like to make sense of @xmath1 as a knot invariant in the degenerate case @xmath38 .
this problem does not have a unique solution , and several notions of self - linking already exist .
our purpose is to propose another , motivated by gauge theory @xcite and with an eye towards higher - order @xcite self - linking invariants .
the most naive attempt to define a self - linking number for an embedded , oriented curve @xmath39 is based upon the gauss integral in . again parametrizing @xmath40 as the image of a smooth map
@xmath41 we follow our nose to set @xmath42 unlike , the self - linking integrand is now singular along the diagonal @xmath43 , due to the divergence of @xmath16 at the origin in @xmath0 . to deal with the singularity , we excise a tubular neighborhood @xmath44 of width @xmath45 about the diagonal , depicted as the shaded region in figure [ deleps ] , and we integrate only over the resulting cylinder @xmath46 .
we finally take the limit @xmath47 to remove any dependence on the auxiliary parameter .
@xmath48{deleps.eps } } \,&\simeq\,\,\,\raisebox{-42pt}{\includegraphics[scale=0.30]{cyl.eps } } \end{matrix}\ ] ] there are two essential statements to make about the naive self - linking integral in , both of which go back to the classic works by clugreanu @xcite and pohl @xcite . first ,
the singularity along the diagonal @xmath49 is integrable , so the limit defining @xmath50 does exist .
second , @xmath51 is _ not _ a deformation - invariant of @xmath40 , but rather varies to first - order under a change @xmath52 in the embedding , @xmath53 to be a regular , unit - speed parametrization , @xmath54 , for which @xmath55 is the arc - length measure on @xmath40 .
also , @xmath56 for @xmath57 is the fully anti - symmetric tensor , normalized so that @xmath58 . as a corollary of
, the value of the naive self - linking integral depends non - trivially on the geometry of @xmath59 , and it can not generically be an integer , @xmath60 .
where did the naive attempt to define a self - linking invariant go wrong ?
the answer to this question has both a topological and an analytical component . from a topological perspective
, the primary difficulty is that the domain of integration @xmath61 has a boundary .
briefly , when the embedding map @xmath62 varies in , the integrand changes by a cohomologically - trivial two - form on the cylinder @xmath63 . by stokes theorem , the variation of @xmath51 then reduces to an integral over the boundary circles in figure [ deleps ] . with some calculation
, the boundary integrand can be explicitly evaluated in the limit of small @xmath64 to yield the expression in .
see the ph.d .
thesis of bar - natan @xcite for this calculation , also reviewed in appendix a of @xcite .
the analytical aspect of the self - linking anomaly is not as obvious , but no less important .
irrespective of the particular integrand in , the appearance of a boundary contribution to @xmath65 from the variation of @xmath62 on @xmath66 is inevitable .
what is not inevitable , but what does happen in the detailed computation leading to , is that the boundary contribution to @xmath65 remains non - vanishing even in the limit @xmath67 .
this phenomenon depends very much on the analytic behavior of the angular form @xmath16 near the origin , and need not have occurred had the divergence of @xmath16 been less rapid
. this observation will be central to our work and is exemplified by the fundamental lemma in section [ fundamental ] .
because of the anomaly , all notions of self - linking involve a modification to the naive gauss integral in , as well as a refinement of the equivalence relation by smooth isotopy for the curve @xmath68 .
for instance , in the original approach of @xcite , one adds to @xmath51 a counterterm @xmath69 whose variation with respect to @xmath62 is exactly opposite the variation in , @xmath70 the required counterterm is nothing more than the total torsion of @xmath40 measured with respect to arc - length , @xmath71 where the local frenet - serret @xcite torsion @xmath72 is given in terms of @xmath62 by @xmath73 for sake of brevity , we do not review the proof of here .
a direct calculation of @xmath74 can be found in appendix b of @xcite . from the relation in ,
the sum @xmath75 does not change under any small deformation of @xmath40 , so it provides a reasonable notion of self - linking .
however , @xmath76 is also not invariant under arbitrary smooth isotopies of @xmath40 . due to the denominator in
, the torsion @xmath72 is only defined at points of @xmath40 where @xmath77 , or equivalently , where the frenet - serret curvature is non - zero . to make sense of @xmath69
, we assume that @xmath40 has everywhere non - vanishing curvature . the latter condition is open and so holds for the generic space curve , but of course it does not hold for all space curves . as a result ,
@xmath76 is only invariant under those ` non - degenerate ' isotopies which preserve the condition of non - vanishing curvature along @xmath40 . . ]
like the linking number , the self - linking number in has a simple diagrammatic description . to derive it ,
consider the smooth isotopy @xmath78 where @xmath79 is a positive real parameter . applied to any curve @xmath39
, this isotopy flattens @xmath40 to the @xmath80-plane as @xmath81 grows to infinity .
rotating @xmath40 if necessary , we assume that both @xmath40 and its projection to the @xmath80-plane have everywhere non - vanishing curvature
. see figure [ trefoil ] for a diagram of the trefoil knot which satisfies this condition .
then @xmath82 is non - degenerate for all values of @xmath81 , and @xmath76 can be evaluated in the limit @xmath83 . for any plane curve , @xmath84 identically , so @xmath76 reduces to @xmath51 in this limit . otherwise , when @xmath81 is large and @xmath40 is nearly planar , the naive self - linking integral in reduces to a sum over the local writhe at each self - crossing of @xmath40 .
we omit a proof of the latter claim , which is hopefully plausible on the basis of the similar description for @xmath1 . in section
[ local ] , we will establish a closely related result as part of our main theorem .
altogether , @xmath76 is given by the writhe @xmath85 of the planar diagram for @xmath40 , @xmath86 in particular , @xmath87 is an integer , which is not manifest from the analytic description .
also , because the assignment of the writhe @xmath88 in figure [ writhe ] is invariant under a reversal of orientation for both strands , @xmath76 does not depend upon the orientation of @xmath40 .
finally , the various versions of the unknot in figure [ wunknot ] illustrate that @xmath76 can not be a full isotopy invariant of @xmath40 .
@xmath29{unknot1.eps } \qquad & \qquad \includegraphics[scale=0.50]{unknot2.eps } \qquad & \qquad \includegraphics[scale=0.50]{unknot3.eps}\\[1 ex ] & \operatorname{\mathsf w}(c ) = 0 \qquad & \qquad \operatorname{\mathsf w}(c ) = 1 \qquad & \qquad \operatorname{\mathsf w}(c ) = 2 \end{matrix}\ ] ] the frenet - serret self - linking number @xmath76 is really a special case of the more general , and perhaps more familiar , notion of framed self - linking . by definition ,
a framing of @xmath39 is a trivialization of the normal bundle @xmath89 , up to homotopy .
such a trivialization can be specified by a nowhere - vanishing normal vector field @xmath90 along the knot . in turn
, the normal vector field determines a new curve @xmath91 obtained by displacing @xmath40 a small amount in the direction of @xmath92 , as shown in figure [ framed ] . .
] given the pair @xmath93 , the framed self - linking number is defined as the ordinary linking number of the two disjoint curves @xmath40 and @xmath91 , @xmath94 on the upside , @xmath95 is manifestly an integral isotopy invariant of the pair @xmath93 . on the downside
, @xmath95 carries no information about the knot @xmath40 itself , since the invariant takes all possible values as the winding number of @xmath92 about @xmath40 shifts .
had one a canonical choice of framing , the framed self - linking number @xmath95 could be converted into an honest invariant of @xmath40 .
no such choice exists for all smooth curves simultaneously , but a variety of choices can be made if one restricts to special classes of curves .
we have already encountered an example in the discussion of the frenet - serret self - linking @xmath76 , for which we require @xmath68 to have everywhere non - vanishing curvature . not coincidentally
, such curves also carry a canonical frenet - serret framing , with unit normal @xmath96 .
indeed , a natural guess is that @xmath97 for the frenet - serret normal .
this guess is correct , as can be seen most easily by considering the behavior of the frenet - serret frame in the planar limit @xmath83 from . in this limit
, the frenet - serret framing reduces to the blackboard framing in which the unit normal vector @xmath92 lies everywhere in the plane of the knot diagram . after displacing @xmath40 by @xmath92 ,
one finds a planar ribbon graph , shown in the neighborhood of a positive crossing on the left in figure [ fsframe ] .
for such a graph , each self - crossing of @xmath40 corresponds to a crossing of @xmath40 by @xmath91 with identical chirality , so automatically @xmath98 .
the relation between the writhe and the frenet - serret framing can also be understood in three - dimensional terms , as indicated to the right in figure [ fsframe ] .
@xmath99{ribbon.eps } } \quad&\simeq\quad & \raisebox{-40pt}{\includegraphics[scale=0.50]{twist.eps}}\\[1.5 ex ] & \operatorname{\mathsf w}\,=\ , + 1 \qquad&\qquad & \operatorname{\mathsf{slk}}_{\rm f } \,=\ , + 1 \end{matrix}\ ] ] this article serves as the companion to a longer work @xcite in which we develop a new , effectively supersymmetric formulation of chern - simons perturbation theory . both the frenet - serret and
the framed self - linking invariants were rediscovered physically @xcite in the setting of bosonic chern - simons theory , and our purpose is to present what one finds with the addition of @xmath100 supersymmetry , after all the baggage from gauge theory is removed .
details about the gauge theory are discussed in @xcite .
supersymmetry has two consequences . first and foremost , @xmath39 must be legendrian with respect to the standard contact structure on @xmath0 , and the supersymmetric self - linking number will be a legendrian isotopy invariant .
for very enjoyable introductions to contact topology and the study of legendrian knots , see @xcite . at the moment , we recall only the minimum necessary to state and prove our main theorem . throughout ,
we represent the standard contact structure on @xmath0 with the contact form @xmath101 for which the top - form @xmath102 is positive with respect to the standard orientation on @xmath0 .
this choice of contact form respects many symmetries , which are important here and in @xcite .
manifestly , @xmath103 is preserved under translations generated by the reeb field @xmath104 as well as rotations in the @xmath80-plane .
though @xmath103 is not preserved by translations in the @xmath80-plane , @xmath103 is preserved by the left - action of the heisenberg lie group @xmath105 on itself , where the heisenberg multiplication @xmath106 is given by @xmath107 the origin remains the identity in @xmath108 , and the heisenberg inverse is @xmath109 .
finally , @xmath103 is homogeneous of degree two under the parabolic scaling @xmath110 which commutes with heisenberg multiplication . as a result , the parabolic scaling fixes each contact plane @xmath111 in the kernel of @xmath103 .
a picture of the family of contact planes @xmath112 appears in figure [ standardh ] .
the contact planes are approximately horizontal near @xmath113 , but they twist vertically as one moves outward from the origin in the @xmath80-plane . .
( courtesy of the mathematica routine ` csplotter ' by matias dahl . ) ] we also require a few facts about legendrian knots . by definition , @xmath59 is legendrian when the tangent line @xmath114 at any point @xmath115 lies in the contact plane @xmath116 at that point .
equivalently , the pullback of @xmath103 to @xmath40 vanishes , @xmath117 or in terms of a parametrization @xmath118 , @xmath119 any smooth knot admits a legendrian representative , so the theory of legendrian knots is extremely rich .
moreover , equivalence by legendrian isotopy , ie .
continuous isotopy through a family of legendrian knots , strictly refines the usual topological equivalence .
a given topological knot may admit infinitely - many inequivalent legendrian representatives . unlike topological knots , legendrian knots have canonical plane projections .
for this reason , legendrian knots behave in many ways like plane curves .
our interest lies in the so - called lagrangian projection to the @xmath80-plane , @xmath120 for which the image @xmath121 of a legendrian knot @xmath39 is a smoothly immersed curve .
the smoothness of @xmath121 is already a non - trivial feature of the legendrian condition , since this condition implies that @xmath122 at any point on @xmath40 where @xmath123 .
hence if @xmath62 is a regular parametrization of @xmath40 , then @xmath124 is a regular parametrization of @xmath121 .
trivially , @xmath121 is a lagrangian submanifold of @xmath125 with respect to the symplectic form @xmath126 , whence the name . and @xmath127 .
] in figure [ lgtref ] we display the lagrangian projection of a legendrian trefoil knot . to guide the eye
, we indicate over- and under - crossings in the figure .
unlike for topological knot diagrams , the crossing information for a legendrian knot is redundant , since the spatial configuration of @xmath39 can be completely recovered from the plane curve @xmath121 by integrating the contact relation in , @xmath128.\ ] ] here @xmath129 is the height of @xmath40 at the basepoint corresponding to @xmath130 . due
the symmetry of @xmath103 under translations in @xmath131 , this constant is both arbitrary and irrelevant . not every immersed curve can be the plane projection of a legendrian knot .
for instance , if we take the parameter @xmath132 in to have periodicity @xmath133 , then @xmath134.\ ] ] equivalently by stokes theorem , the plane region @xmath135 enclosed by @xmath121 must have zero symplectic area , @xmath136 where each component of @xmath135 is oriented consistently with @xmath137 .
also , if @xmath138 are distinct parameter values for which @xmath139 and @xmath140 , corresponding to the location of a crossing in @xmath121 , then @xmath141.\ ] ] the necessary conditions in and are sufficient for the immersed plane curve to lift to an embedded legendrian knot .
these conditions depend upon the signed areas of the regions enclosed by @xmath121 , so the lagrangian projection can not be manipulated in a wholly topological fashion la reidemeister .
legendrian knots always carry a canonical framing by the reeb vector field @xmath142 . concretely from , a legendrian curve @xmath39 can not have a vertical tangent , where @xmath123 but @xmath143 . with the choice @xmath144 , the framed self - linking number @xmath95
can then be converted into a legendrian invariant @xmath145 a kind of self - linking number for @xmath40 .
the thurston - bennequin invariant @xmath146 can be easily computed from the lagrangian projection of @xmath40 . after a rigid rotation ,
the vertical framing by the reeb field @xmath147 becomes equivalent to the planar , blackboard framing of @xmath121 .
but again by figure [ fsframe ] , the self - linking number in the blackboard framing is exactly the writhe of the knot diagram .
thus , @xmath148 since the writhe is fixed under orientation - reversal , so too is @xmath149 the thurston - bennequin invariant is one of a pair of classical legendrian invariants . to state the main theorem , we also need the other . because @xmath40 is determined by its lagrangian projection @xmath121 , any isotopy invariant of immersed plane curves yields a legendrian invariant of @xmath40 .
according to the whitney - graustein theorem @xcite , the unique such invariant of an immersion @xmath150 is the rotation number @xmath151 defined as the topological degree of the derivative @xmath152 .
equivalently , @xmath153 is the total ( signed ) curvature of the immersed plane curve , @xmath154 where we use the shorthand ` @xmath155 ' for the scalar cross - product , @xmath156 as will be essential later , the formula for @xmath153 in presents the rotation number as a local invariant , in the sense of being the integral of a locally - defined geometric quantity along the curve . with our conventions ,
@xmath157 when @xmath158 is a circle traversed in the counterclockwise direction . for the legendrian knot @xmath39
, we set @xmath159 see definition @xmath160 in @xcite for an intrinsically three - dimensional characterization of @xmath161 . in terms of the diagram for @xmath121 , the rotation number can be computed as a signed count of upwards vertical tangencies , as in figure [ rotat ] . for the legendrian trefoil in figure [ lgtref ] ,
two upwards vertical tangencies occur , but they do so with opposite signs , so @xmath162 .
note that the rotation number depends upon the orientation of the curve , and under a reversal of orientation , the rotation number changes sign .
so in contrast to the behavior of the thurston - bennequin invariant , @xmath163 we have not specified an orientation for the trefoil knot in figure [ lgtref ] , but because @xmath164 , the orientation does not matter .
@xmath29{rotp.eps}\qquad\quad & \quad\qquad & \includegraphics[scale=0.50]{rotn.eps}\\ & \operatorname{\mathsf{rot}}=+1 \qquad\quad & \quad\qquad & \operatorname{\mathsf{rot}}=-1 \end{matrix}\ ] ] as its second consequence , supersymmetry modifies the angular form @xmath165 which enters the elementary gauss linking integral .
because the legendrian condition on @xmath39 does not respect the euclidean action by @xmath166 , we forgo the seemingly - natural requirement that @xmath16 itself be @xmath166-invariant . in recompense ,
the supersymmetric version of @xmath16 will enjoy superior analytic behavior near the origin in @xmath0 . as heuristic motivation for the following
, one should imagine that we alter the angular form @xmath16 in by concentrating the support for the generator of @xmath18 at the north pole on the sphere .
this trick is well - known to aficionados of chern - simons perturbation theory , but here we must take care to preserve the underlying symmetries of the contact structure on @xmath0 . more precisely , we introduce the gaussian two - form @xmath167 on the @xmath80-plane , @xmath168 here @xmath81 is a positive real parameter which sets the width of the gaussian , and in the limit @xmath83 , the gaussian becomes a delta - function concentrated at the origin in @xmath125 . clearly @xmath169 is invariant under rotations , and @xmath169 is normalized so that for all @xmath81 , @xmath170 the various factors of two in are standard and could be absorbed into @xmath81 if desired .
though the support of @xmath169 is not compact , @xmath169 decays very rapidly at infinity .
at least morally , @xmath169 should be regarded as a generator for the compactly - supported cohomology @xmath171 of the plane , on the same footing as the unit - area form on the sphere . because the normalization condition in does not depend on @xmath81 , neither does the cohomology class of @xmath169 .
explicitly , a small computation shows @xmath172 \e{\!-\lambda\,(x^2 + y^2)/2}\,dx\^dy \,=\ , d\alpha_\lambda\,,\ ] ] where @xmath173 the transgression form @xmath174 will reappear in the proof of our fundamental lemma . in the meantime , note that @xmath174 is also @xmath175-invariant , as required by the relation to @xmath169 in .
we next introduce the planar analogue for the retraction onto @xmath23 in . to preserve the parabolic scaling in , we consider the map @xmath176 defined on the upper half - space @xmath177 by @xmath178 trivially , the image of @xmath179 is preserved under the scaling for which @xmath180 and @xmath181 have weight one and @xmath131 has weight two . using the planar retraction in , we pull the gaussian form @xmath182 back to a new two - form @xmath183,\qquad z>0\ , . \end{aligned}\ ] ] by construction , @xmath184 is invariant under the parabolic scaling and the action of @xmath175 , but not @xmath166 .
of course , @xmath184 strongly resembles the heat kernel for the laplacian in two dimensions . as with the heat kernel , so long as @xmath185 , the expression in vanishes smoothly as @xmath186 from above . to define @xmath184 on the entire punctured space @xmath187 , we simply extend by zero , @xmath188 with this choice
, @xmath189 is automatically closed away from @xmath190 and generates the cohomology @xmath191 .
for instance , over the unit sphere @xmath192 , @xmath193 we have yet to incorporate the heisenberg symmetry of the contact structure . in the elementary gauss linking integral , the abelian structure of @xmath0 as a vector space enters implicitly through the definition of the difference map @xmath194 in .
to preserve instead the non - abelian symmetry by left - translation in @xmath195 , we consider a heisenberg difference map @xmath196 , given by @xmath197 since @xmath198 is the heisenberg multiplication , @xmath199 in the usual shorthand .
this combination of @xmath200 and @xmath201 is invariant under simultaneous left - multiplication , @xmath202 and we have selected the relative signs of @xmath200 and @xmath201 in to agree with the convention for the abelian difference in .
[ hlink ] suppose @xmath203 are disjoint oriented curves , not necessarily legendrian , with respective parametrizations @xmath204 .
then @xmath205 this proposition follows from the fact that the heat form @xmath184 is equivalent in cohomology to the global angular form @xmath16 , @xmath206\,=\,[\psi ] \,\in\ , h^2\big(\br^3-\{0\};\bz\big)\,.\ ] ] also , the heisenberg difference @xmath207 in is homotopic to the abelian difference @xmath194 . to see this , set @xmath208\,.\ ] ] then @xmath209 smoothly interpolates from the abelian to the heisenberg difference as the planck constant @xmath210 ranges over the interval from @xmath211 to @xmath212.@xmath213 though the angular form @xmath16 and the heat form @xmath184 agree in cohomology , they behave very differently near the origin in @xmath0 .
this analytic distinction matters crucially for approaches to self - linking .
let @xmath39 be an oriented legendrian curve , with regular parametrization @xmath118 .
by analogy to the naive gauss self - linking integral in , we consider a new heisenberg self - linking integral @xmath214\,,\qquad\quad \lambda > 0\,.\ ] ] the remainder of the article is devoted to the proof of the following main theorem .
[ mt ] the limit defining @xmath215 exists .
the value of @xmath215 is independent of @xmath81 and depends only upon the legendrian isotopy class of @xmath40 . in terms of the thurston - bennequin invariant @xmath146 and the rotation number @xmath161 , @xmath216 informally ,
the main theorem states that the framing anomaly for knots in bosonic chern - simons theory is absent in supersymmetric chern - simons theory .
the corresponding statement for seifert - fibered three - manifolds was observed previously in @xcite .
the main theorem is also consistent with results of fuchs and tabachnikov @xcite identifying @xmath146 and @xmath161 as the only order @xmath217 finite - type invariants of legendrian knots .
the strategy of proof for theorem [ mt ] is straightforward .
we first demonstrate that @xmath215 is independent of the parameter @xmath81 which sets the width of the gaussian in @xmath184 . for any @xmath218 ,
the derivative of @xmath215 with respect to @xmath81 is given in terms of a boundary integral of the transgression form @xmath174 in .
the content of our fundamental lemma in section [ fundamental ] is to demonstrate that the potentially anomalous boundary contribution from @xmath174 vanishes in the limit @xmath67 , due to the rapid decay of the heat kernel away from the diagonal . in section [ mainthm ]
we directly evaluate the self - linking integral in the limit @xmath83 . in this regime ,
the self - linking integrand is non - negligible only in the neighborhood of points on @xmath17 which map under the product @xmath219 to pairs of points @xmath220 that become coincident under the lagrangian projection to the @xmath80-plane , ie . @xmath221 .
such points on @xmath17 correspond either to the preimage of crossings in @xmath121 , or more trivially , to points on the diagonal @xmath222 .
the local contribution from the crossings leads to the appearance of the thurston - bennequin invariant @xmath146 , while the local contribution from the diagonal @xmath49 leads to the appearance of the rotation number @xmath161 for @xmath223 . at the special value @xmath224
, the heisenberg symmetry of the integrand is restored , and the anomalous contribution from @xmath49 vanishes .
the legendrian condition is used crucially throughout the analysis .
invariance under legendrian isotopy follows _ a postiori _ from the formula in .
, achieved for @xmath65 in , involves some formidable differential algebra and did not appear practical to these authors . ]
let us emphasize one striking feature of theorem [ mt ] , which is perhaps best appreciated after one works through the localization computation in section [ local ] .
namely , since the coefficients of @xmath146 and @xmath161 in are integers , so is the value of @xmath215 ! the coefficient of @xmath146 is directly related to our normalization condition on the heat form @xmath184 , required to recover the usual linking number in proposition [ hlink ] .
thus the coefficient of @xmath146 is fixed by fiat to unity .
in contrast , the coefficient of @xmath161 is determined by a delicate calculation near the diagonal @xmath222 , so its integrality for all @xmath210 is a non - trivial feature of the legendrian self - linking integral . from the physical perspective
, integrality of @xmath215 is necessary for gauge invariance as well as consistency with standard lore about non - renormalization and the infrared behavior of supersymmetric yang - mills - chern - simons theory . see
@xmath225 of @xcite for a discussion of this statement .
however , on the principle that no integer appears by chance , a simple topological explanation for the integrality of @xmath226 would be nice to have .
for instance , the difference of classical legendrian invariants @xmath227 in occurs naturally in contact topology as the transverse self - linking invariant @xmath228 of the canonical positive transverse push - off @xmath229 of @xmath40 ( proposition @xmath230 in @xcite ) . the transverse self - linking invariant @xmath228 can be interpreted in terms of a relative euler class on a seifert surface for @xmath229 , so its appearance in theorem [ mt ] is surely no accident . for the convenience of the reader , we summarize the notation and conventions used in the rest of the paper
. * @xmath0 has euclidean coordinates @xmath3 and is oriented by @xmath231 . *
the map @xmath232 is the projection onto the @xmath80-plane .
* @xmath233 is the standard radially - symmetric contact form , positive with respect to the orientation on @xmath0 .
* @xmath68 is an oriented legendrian knot , with regular parametrization @xmath118 .
* @xmath234 is an angular coordinate on @xmath235 , compatible under @xmath62 with the given orientation on @xmath40 .
we abbreviate @xmath236 , and so on . *
the torus @xmath237 has angular coordinates @xmath24 and is oriented by @xmath238 .
the diagonal @xmath43 is the subset where @xmath239 . * for @xmath240 , a tubular neighborhood @xmath44 of the diagonal @xmath222 is parametrized by @xmath241 and @xmath242 for @xmath243 .
the cylinder @xmath63 has boundary circles @xmath244 on which @xmath245 , respectively .
* @xmath246 is an immersed plane curve which is the lagrangian projection of @xmath40 , ie .
* we use the abbreviation ` @xmath155 ' for the scalar cross - product on the plane , as in @xmath248 * the contact form @xmath103 is left - invariant with respect to the heisenberg multiplication @xmath249\\ & \mu\big(x_1,x_2\big ) \,=\ , \left(x_1 + x_2,\ , y_1 + y_2,\ , z_1 + z_2 - x_1 y_2 + x_2 y_1\right ) .
\end{aligned}\ ] ] more generally , for other values of @xmath210 set @xmath250 * the left - invariant heisenberg difference @xmath251 for @xmath252}$ ] is given by @xmath253 * the gaussian fundamental class @xmath254 of the @xmath80-plane is given by @xmath255 * the transgression form @xmath256 satisfies @xmath257 , where @xmath258 * the planar retraction @xmath259 is defined on the upper half - space @xmath177 by @xmath260 * the heat form @xmath261 is the pullback @xmath262 explicitly , @xmath263,\qquad z>0\,.\ ] ]
we first demonstrate that the value of the legendrian self - linking integral @xmath215 does not depend upon the parameter @xmath264 which sets the width of the gaussian in the heat form @xmath184 .
[ fundlm ] the limit which defines the legendrian self - linking integral @xmath215 exists , @xmath265\,,\qquad\quad \lambda>0\,,\ ] ] and the value of @xmath266 is independent of the parameter @xmath81 . precisely at the special value @xmath212 ,
the self - linking integrand is heisenberg - invariant .
for this reason , the analysis to prove lemma [ fundlm ] will differ slightly depending on whether @xmath267 or @xmath224 .
throughout , we make two extra topological assumptions about the legendrian knot @xmath40 .
both assumptions hold generically and help to simplify the proofs . 1 .
the lagrangian projection @xmath121 is an immersed plane curve @xmath158 with only double - point singularities , as in figure [ lgtref ] .
2 . the height function @xmath268 on @xmath39 is morse , with isolated non - degenerate critical points .
that is , @xmath269 vanishes only at isolated points @xmath115 , at which @xmath270 .
because @xmath40 is legendrian , the morse condition on @xmath268 is equivalent by to the condition that the function @xmath271 vanish only at isolated points @xmath272 , at which @xmath273 .
the first assumption is standard .
otherwise , the morse condition is used only in the case @xmath223 and could possibly be relaxed , at the cost of further effort . before embarking on the proof of our fundamental lemma ,
let us mention an easy corollary which is handy for the gauge theory analysis in
@xmath274 of @xcite .
let @xmath275 be a positive scaling parameter .
as in the remarks following proposition [ hlink ] , we consider a version of the heisenberg difference @xmath276 with rescaled planck constant @xmath277 for fixed @xmath252}$ ] , @xmath278 for each value of @xmath279 , we associate the contact form @xmath280 the relative power of @xmath279 between the two terms in ensures that the contact form is left - invariant under the heisenberg multiplication with @xmath281 in .
the overall power of @xmath279 ensures that the contact condition @xmath282 is satisfied trivially for all values @xmath275 .
finally , suppose that @xmath68 is a legendrian knot with respect to the standard contact form @xmath283 . just as we consider the family of isotopic contact forms in , we would like to consider a family of isotopic knots @xmath284 , each of which is legendrian with respect to @xmath285 for the given value of @xmath279 .
such a family of knots is determined if we simply require the lagrangian projection of @xmath286 to coincide with that of @xmath40 , @xmath287 either by integrating the contact condition as in or just by scaling , the embedding map @xmath288 for @xmath286 is then related to the original embedding @xmath62 for @xmath40 via @xmath289 in particular , the abelian limit @xmath290 of the heisenberg structure corresponds to a limit in which @xmath286 flattens to a curve in the @xmath80-plane , and the lagrangian projection @xmath121 is realized geometrically . given the family of curves @xmath286 , we consider the _ three_-parameter self - linking integral @xmath291\,,\qquad\quad t,\lambda>0\,.\ ] ] precisely for @xmath212 , the self - linking integrand is invariant under the heisenberg symmetry with multiplication map @xmath292 .
the @xmath279-dependence of the integrand in is very simple . in terms of the finite differences @xmath293 the formula for the heat form @xmath184 in implies @xmath294 , \end{aligned}\ ] ]
provided @xmath295 . otherwise , the pullback of @xmath184 vanishes .
evidently , @xmath279 just multiplies @xmath296 in , and all dependence on @xmath279 can be absorbed by rescaling the gaussian parameter @xmath81 . for all @xmath297 , @xmath298 by the scaling identity ,
the behavior of the self - linking integrand in limit @xmath83 with fixed @xmath279 is the same as the behavior in the limit @xmath290 with fixed @xmath81 . because @xmath286 flattens to a plane curve in the latter limit
, @xmath81 plays a similar role to the parameter of the same name in .
the value of @xmath299 is independent of both @xmath279 and @xmath81 .
this corollary follows immediately from lemma [ fundlm ] and the scaling identity in .@xmath213 the non - trivial aspect of our work concerns the local analysis of the self - linking integrand in the vicinity of the diagonal @xmath300 . in practice
, this analysis amounts to the taylor expansion of expressions such as occur in . rather than scatter such expansions willy - nilly throughout the paper ,
we collect here the basic ingredients to be used again and again .
let @xmath24 be angular coordinates on @xmath17 . to parametrize the tubular neighborhood @xmath301 of the diagonal , we set @xmath302 here @xmath303 is an angular coordinate along the diagonal , and @xmath44 is the subset where @xmath243 .
our local expansions near the diagonal will then be taylor expansions in @xmath304 , appropriate for the regime @xmath305
. we must be careful about orientations . in terms of the coordinates @xmath306 , the orientation form on @xmath17
is given by @xmath307 .
topologically , the configuration space @xmath63 is a cylinder with oriented boundary circles @xmath308 as shown in figure [ cylinder ] , the boundary orientation of @xmath309 is positive with respect to the direction of increasing @xmath303 and the orientation of @xmath310 is negative , so @xmath311 . ]
we now expand the various terms appearing in the self - linking integrand @xmath312 , given by the expression in for @xmath313 .
we start with @xmath314 and similarly for @xmath315 .
hence @xmath316 where @xmath317 is the lagrangian immersion @xmath318 as before . for the related two - form ,
we can write @xmath319 after collecting terms in the product , @xmath320 recall the definition @xmath321 the legendrian condition on @xmath40 is absolutely critical for our results , because it implies that the local behavior of @xmath322 near the diagonal is controlled by the geometry of the lagrangian immersion @xmath158 .
moreover , for @xmath322 the heisenberg group rears its head !
the expansion of the abelian difference @xmath323 is fixed by the legendrian condition @xmath324 thus @xmath325 in passing to the second line of , we repeatedly differentiate the legendrian condition on @xmath269 in .
the attentive reader may wonder why we have expanded @xmath326 all the way to cubic order in @xmath304 .
the question answers itself once we expand the remaining quadratic terms in the heisenberg difference @xmath322 , @xmath327 so long as @xmath328 , the expansion of @xmath322 begins at linear order in @xmath304 , @xmath329 as one naively expects .
but precisely at @xmath212 , cancellations occur in the sum of and , and the leading term in the expansion of @xmath322 near the diagonal begins at cubic order , @xmath330 the cancellation in is forced by the heisenberg symmetry at @xmath212 . clearly , the quantity @xmath331 in is not invariant under translations @xmath332 for constant @xmath333 .
upon projection to the @xmath80-plane , such translations are generated by the heisenberg action , so @xmath331 is forbidden to appear at @xmath212 . combining the expansions in , , and , we see that the argument of the heat kernel is given in the neighborhood @xmath44 by @xmath334$}\\[1 ex ] & -\frac{3\ , ||\dot{\gamma}||^2}{\left(\dot{\gamma}\times\ddot{\gamma}\right)\eta } \,+\ , \co(1)\ , . &
\hbox{\small$[\hbar=1]$ } \end{aligned}\ ] ] for generic @xmath328 , the argument of the heat kernel in vanishes linearly near the diagonal .
however , at the symmetric point @xmath212 , the argument instead diverges as @xmath335 . in section [ mainthm ] , this difference will ultimately lead to the discontinuity in the value of @xmath215 at @xmath224 .
the pullback of the heat form @xmath184 vanishes identically unless @xmath295 . on the neighborhood @xmath44 , positivity of @xmath322
becomes equivalent via and to the local sign condition @xmath336$}\,,\\ & \eta\left(\dot{\gamma}\times\ddot{\gamma}\right ) < 0\ , .
& \hbox{\small$[\hbar=1]$}\ , . \end{aligned}\ ] ] again , the nature of the positivity condition depends upon whether or not @xmath212 . for generic values of @xmath210 ,
the sign of @xmath304 is determined by the sign of @xmath337 and hence the sign of the derivative @xmath269 . for @xmath212 ,
the sign of @xmath304 is instead fixed by the sign of @xmath338 , proportional to the plane curvature of @xmath121 .
@xmath339{etap.eps}\quad\qquad & \quad\qquad & \includegraphics[scale=0.60]{etan.eps}\\[1 ex ] & \eta>0\,,\,\dot{\gamma}\times\ddot{\gamma}<0 \quad\qquad & \quad\qquad & \eta<0\,,\,\dot{\gamma}\times\ddot{\gamma}>0 \end{matrix}\ ] ] since the local positivity condition depends upon the sign of @xmath304 , the self - linking integrand always vanishes in the neighborhood of one or the other of the boundary circles @xmath244 in figure [ cylinder ] , and the geometry of @xmath40 at any given point determines on which boundary circle the integrand vanishes .
see figure [ positv ] for a geometric illustration of the local positivity condition @xmath295 in the symmetric case @xmath212 . for clarity , we exaggerate the small separation between the points @xmath340 and @xmath341 in the figure . in both cases ,
the positivity condition is sensitive to the orientation of @xmath40 , as a reversal of orientation flips the signs of @xmath337 and @xmath338 in .
let us complete the expansion for small @xmath304 of the self - linking integrand .
the second bracketed term in involves the angular one - form @xmath342 from the expansions of @xmath343 in and , @xmath344$}\\ & -\ha\left(\dot{\gamma}\times\ddot{\gamma}\right)^2 \eta^4\ , d\phi\^d\eta\ , , & \hbox{\small$[\hbar=1]$ } \end{aligned}\ ] ] to leading order , so @xmath345$}\\ & 3\,\eta\left(\dot{\gamma}\times\ddot{\gamma}\right ) d\phi\^d\eta\,+\ , \co\big(\varepsilon^2\big)\ , d\phi\^d\eta\ , .
& \hbox{\small$[\hbar=1]$ } \end{aligned}\ ] ] after a bit of algebra , one finds that the pullback of @xmath184 behaves near the diagonal @xmath300 for @xmath328 as @xmath346}\,d\phi\^d\eta\,+\ , \cdots\ , , \end{aligned}\ ] ] assuming the sign condition @xmath347 in holds .
otherwise , the pullback is equal to zero .
the sign condition ensures that the exponential in is always decaying , and it leads to a non - analytic dependence on the sign of @xmath304 .
for @xmath348 , and @xmath349 for @xmath350 . ]
the ellipses in indicate subleading terms which vanish as @xmath351 .
by contrast , at the symmetric value @xmath212 , @xmath352}\,d\phi\^d\eta \,+\,\cdots\,,\\ & \,=\ , \frac{3\,\lambda}{2\pi } \exp{\!\left[-\frac{3\,\lambda\,||\dot{\gamma}||^2}{\left|\dot{\gamma}\times\ddot{\gamma}\right| } \,|\nu|\right]}\,d\phi\^d\nu \,+\,\cdots\ , , \end{aligned}\ ] ] under the sign condition @xmath353 in . in passing to the second line
, we make the substitution @xmath354 for clarity . in this case
, the pullback of the heat form @xmath184 vanishes exponentially as @xmath355 , or equivalently @xmath356 .
the proof of our fundamental lemma [ fundlm ] is now an exercise in calculus . as a brief formality
, we first establish that the singularity in the pullback of the heat form @xmath184 is integrable , so that the defining limit @xmath67 in does exist . by assumption , @xmath357 is everywhere non - vanishing , and the functions @xmath358 and @xmath359 are bounded from above on @xmath40 .
integrability in the region of small @xmath243 follows immediately from the local expressions in and , both of which remain finite as @xmath335 . otherwise , we must check that the value of @xmath215 does not depend upon the parameter @xmath79 .
this argument will be equally straightforward but the result is significant ; the analogue for the naive gauss self - linking integral @xmath51 is simply false .
our strategy will be to show that the derivative of @xmath215 with respect to @xmath81 vanishes for all values of @xmath360 .
the details differ somewhat depending upon whether @xmath361 or @xmath212 , but the main idea is the same in both cases .
we compute @xmath362_+ } \left(x \times x\right)^*\!\widehat\gamma^*_\hbar\,\varrho_+^*\!\left(\frac{\partial\omega_\lambda}{\partial\lambda}\right)\ , .
\end{aligned}\ ] ] here @xmath363_+$ ] indicates the closed subset of the cylinder where @xmath364 is positive , @xmath365_+ \,=\ , \big\ { ( \theta_1,\theta_2 ) \,\big|\ , \widehat\delta z(\theta_1 , \theta_2 ) \ge 0\big\}\,,\ ] ] on which the pullback of the heat form @xmath184 is non - vanishing .
of course , the positive subset in depends upon the legendrian embedding @xmath62 .
we omit this dependence from the notation as @xmath62 is fixed throughout . by the calculation in ,
@xmath366 for the transgression one - form @xmath367 we apply the commutativity of the de rham operator with pullback , followed by stokes theorem , to reduce the bulk integral in the second line of to a boundary integral , @xmath368_+}\mskip-5mu \left(x \times x\right)^*\!\widehat\gamma^*_\hbar\,\varrho_+^*\!\left(\frac{\partial\omega_\lambda}{\partial\lambda}\right ) \,=\ , \int_{\partial[t^2 - \delta(\varepsilon)]_+}\mskip-5mu \left(x \times x\right)^*\!\widehat\gamma^*_\hbar\,\varrho_+^*\alpha_\lambda\,.\ ] ] explicitly , the boundary integrand in is given , where non - zero , by @xmath369 this expression vanishes smoothly whenever @xmath370 from above with @xmath371 .
the boundary of the positive subset @xmath372_+}$ ] in includes those curves where @xmath373 as well as the intersection of @xmath374_+}$ ] with the boundary circles @xmath244 themselves . recall that points on @xmath244 satisfy @xmath245 , respectively . by the preceding ,
only the boundary integral over the intersection @xmath375_+}$ ] is relevant , because the boundary integrand in vanishes on the locus where @xmath373 . altogether , in terms of the boundary integral on the right in , @xmath376_+}\mskip-30mu\left(x \times x\right)^*\!\widehat\gamma_\hbar^*\,\varrho_+^*\alpha_\lambda \,-\ , \int_{s^1_-\cap[t^2 - \delta(\varepsilon)]_+}\mskip-30mu\left(x \times x\right)^*\!\widehat\gamma_\hbar^*\,\varrho_+^*\alpha_\lambda\right].\ ] ] the minus sign for the boundary integral over @xmath310 accounts for the relative orientation in figure [ cylinder ] . despite the minus sign
, no possibility exists for a trivial cancellation between the two boundary integrals in for any fixed @xmath240 . according to the local positivity condition in for respectively @xmath328 or @xmath212 , @xmath377_+\,,\\ & \left(1-\hbar\right)\left(\gamma\times\dot{\gamma}\right ) \hbox { or } \dot{\gamma}\times\ddot{\gamma } \ge 0 \quad\hbox{on}\quad s^1_-\cap[t^2 - \delta(\varepsilon)]_+\,.\\ \end{cases}\ ] ]
the domains of integration over the two boundary circles @xmath244 in are therefore disjoint away from the degeneracy locus where @xmath378 or @xmath379 , so no cancellation can occur . generically , the degeneracy locus consists of a finite set of isolated inflection points on the curve .
let us examine the behavior of the boundary integrand via the expansion near the diagonal from section [ locald ] .
* symmetric case @xmath212 * we initially consider the heisenberg - symmetric case @xmath212 . similar to the bulk integrand in
, the boundary integrand behaves to leading - order at @xmath380 as @xmath381}\,d\phi \,+\ , \cdots\,,\ ] ] where the omitted terms vanish more rapidly as @xmath67 . by a conspiracy of signs ,
the difference on the right of can be rewritten as the single integral @xmath382=\ , 0\,.\ ] ] to deduce the vanishing of the limit @xmath67 , we note that the ratio @xmath383 is everywhere bounded from below on @xmath235 by a positive constant @xmath384 , so the integrand in is dominated by the exponentially - small constant @xmath385 since @xmath360 has been arbitrary throughout , @xmath215 is independent of @xmath81 for @xmath212.@xmath213 * generic case @xmath328 * the analysis for generic @xmath328 is slightly more delicate . here
@xmath386}\,d\phi \,+\ , \cdots\,,\ ] ] so the derivative becomes @xmath387,\ ] ] with @xmath388_+}\mskip-20mu d\phi\,\,\frac{\dot{\gamma}\times\ddot{\gamma}}{|\gamma\times\dot{\gamma}|}\ , \exp{\!\left(-\frac{\lambda\,||\dot{\gamma}||^2\,\varepsilon}{2\,|1-\hbar|\,|\gamma\times\dot{\gamma}|}\right)}\,.\ ] ] the functions @xmath389 differ only in the domain of integration over @xmath235 , and our goal will be to show individually @xmath390 were the function @xmath337 to be everywhere non - zero on @xmath235 , the conclusion in would be immediate , as we would know the integral in to be bounded in magnitude even for @xmath391 .
the explicit prefactor of @xmath64 then ensures the vanishing of @xmath389 in the limit @xmath67 .
however , @xmath392 always vanishes for at least two points ( the highest and the lowest ) on the knot @xmath39 , and we must worry about what happens to the integral in near a zero of @xmath331 , when @xmath64 is very small .
let us make an elementary simplification .
since @xmath359 is bounded from above and @xmath393 is bounded from below on @xmath235 , @xmath394_+}\mskip-20mu d\phi\,\,\frac { a\,\varepsilon}{|\gamma\times\dot{\gamma}|}\ , \exp{\!\left(-\frac{b\,\varepsilon}{|\gamma\times\dot{\gamma}|}\right)}\,,\qquad a , b\,>\ , 0\,,\ ] ] for some positive constants @xmath395 and @xmath396 , into which we also absorb the dependence on @xmath81 and @xmath210 and the various other numerical factors in . to deduce the limit for @xmath389
, we show that @xmath397 vanishes in the same limit . by assumption ,
the height function @xmath398 is morse , with isolated non - degenerate critical points .
equivalently , the function @xmath399 vanishes non - degenerately at an isolated set of points on @xmath235 . by the criteria in , these points are precisely the endpoints of the intervals which compose each integration domain @xmath400_+}$ ] .
locally near such an endpoint @xmath401 , @xmath402 when we examine @xmath403 in the limit @xmath67 , only the contribution to the integral from a ( one - sided ) neighborhood of @xmath404 can be non - zero , so we simplify further by replacing @xmath403 by the model @xmath405 here @xmath406 is an arbitrary upper cutoff , and @xmath407 is now any continuous function defined on the interval @xmath408 $ ] such that @xmath409 > 0 \ , \hbox { exists}\,.\ ] ] for all @xmath218 , the integral defining @xmath410 exists , since the integrand vanishes at the endpoint @xmath401 . for convenience ,
we take @xmath411 and @xmath412 by a suitable choice of parameter .
the proof of the fundamental lemma [ fundlm ] for generic @xmath328 reduces to the following claim .
[ calclm ] let @xmath413 and @xmath407 be defined as in and .
then @xmath414 \,=\ , 0\,.\ ] ] we consider a succession of three cases .
@xmath415we start with the basic example @xmath416 , so that @xmath417 after the substitution @xmath418 , @xmath419 from which the limit follows .
@xmath420next , let @xmath421 and @xmath422 be continuous functions on the interval @xmath423 $ ] obeying bounds @xmath424 for some constants @xmath425 and @xmath426 . set @xmath427}\,.\ ] ] then @xmath428 the function @xmath413 vanishes as @xmath67 by ( i ) .
@xmath429 in the general case of interest , @xmath430 } \,=\ , \varepsilon \int_0 ^ 1 \frac{d\phi}{\phi}\left(\frac{\phi}{f(\phi)}\right ) \ , \exp{\!\left[-\frac{\varepsilon}{\phi}\left(\frac{\phi}{f(\phi)}\right)\right]}\,.\ ] ] because @xmath431 for @xmath432 by assumption , the function @xmath433 is continuous and positive for all @xmath432 .
since the limit @xmath434>0}$ ] is also assumed to exist and be non - zero , @xmath435 is continuous and non - vanishing throughout the unit interval .
hence @xmath436 for some constants @xmath425 and @xmath426 , and the general case follows from ( ii ) .
according to the fundamental lemma [ fundlm ] , the value of the self - linking integral @xmath215 does not depend upon the positive parameter @xmath79 which sets the width of the gaussian in the heat form @xmath184 . to evaluate @xmath215 , and in the process to show that @xmath215 is invariant under legendrian isotopy
, we now analyze the self - linking integral in the limit @xmath83 .
the legendrian knot @xmath68 and its regular parametrization @xmath118 remain fixed throughout . in terms of the heat kernel , this limit is the short - time limit , in which the gaussian generator @xmath169 for the compactly - supported cohomology @xmath437 concentrates to a form with delta - function support at the origin .
more geometrically , by the scaling identity in , the limit @xmath83 is equivalent to the limit @xmath290 in which the contact planes represented by @xmath285 in and the legendrian knot @xmath286 in flatten to the @xmath80-plane .
simultaneously , the planck constant @xmath210 in the heisenberg multiplication scales to zero , and the abelian structure of @xmath0 is restored . for this reason , we refer to the limit @xmath83 as the planar limit . in the planar limit ,
the legendrian self - linking integral simplifies immensely , as can be understood from the formula for the integrand @xmath438,\qquad\qquad \widehat\delta z > 0\ , .
\end{aligned}\ ] ] intuitively , the behavior of the integrand is controlled by the exponential factor in the first line of . when @xmath81 is sufficiently large , the integrand is negligible away from the locus where @xmath439\!\big|_{(\theta_1,\theta_2 ) } \,\ll\ , \frac{1}{\lambda}\,,\qquad\qquad ( \theta_1,\theta_2 ) \,\in\ , t^2-\delta(\varepsilon)\,.\ ] ] because @xmath440 and @xmath315 are given by the differences @xmath441 the asymptotic condition in means that the pair @xmath442 map under the embedding @xmath118 to points @xmath220 which are nearly coincident under the lagrangian projection to the @xmath80-plane .
thus the point @xmath24 either lies near the preimage of a crossing ( aka double - point ) on the lagrangian projection @xmath121 , or @xmath24 lies near the diagonal @xmath49 itself , in the boundary region that we previously analyzed in section [ locald ] . 1 .
estimate the contribution to @xmath215 from each crossing of @xmath121 when @xmath81 is large .
2 . estimate the contribution to @xmath215 from the diagonal @xmath222 when @xmath81 is large .
3 . bound the contributions from elsewhere on the integration domain , as well as the errors in the preceding estimates , by a quantity @xmath443 which can be made arbitrarily small as @xmath83 .
conceptually , the local estimates in the first two steps are most important , because these estimates explain why the thurston - bennequin invariant @xmath146 and the rotation number @xmath161 appear in the formula for @xmath215 .
we therefore begin in section [ local ] with simple , informal computations for the first two steps in the proof . the technical heart of the proof resides in the third step , when we carefully bound the errors in the preceding local computations .
this step is required for a rigorous analysis , but the ideas are standard and offer no surprises . for this reason , sections [ preps ] and [ bounds ] could be omitted on the initial reading of the paper . in section [ preps ]
we introduce various geometric quantities to be used in the error analysis , and in section [ bounds ] we make the necessary bounds .
we compute the contribution to @xmath215 from a right - handed crossing in the lagrangian projection .
we depict such a crossing on the left in figure [ localover ] .
we shall proceed softly , reserving precise inequalities for section [ bounds ] . to first - order over the double - point ,
the curve @xmath40 is approximated by a pair of straight lines . for our local computation ,
we take the lines to be parametrized by maps @xmath444 , where @xmath445 passes over @xmath446 by convention . as in the figure
, we take @xmath447 to describe lines which are perpendicular and lie in parallel planes , @xmath448 here @xmath449 is a positive constant , the height of the overpass . with this choice ,
both @xmath447 satisfy the legendrian condition and so describe a legendrian crossing . because we have yet to establish isotopy - invariance of any kind , our assumptions about even the first - order geometry of @xmath40 require justification .
a significant portion of the analysis in section [ bounds ] will be devoted exactly to this issue . the local contribution to @xmath215 from the right - handed crossing at @xmath450
is now given by @xmath451 where we integrate over all @xmath452 , with the standard orientation @xmath238 .
informally , the error which we make when we extend the range of integration from a small region on @xmath17 to the entire plane @xmath125 vanishes exponentially as @xmath83 , due to the rapid decay of the heat form @xmath184 away from the origin . by extending over all of @xmath125
, we will be able to evaluate the integral in closed form .
for the perpendicular lines @xmath447 in , the differences @xmath440 , @xmath315 , and @xmath322 in become @xmath453 and @xmath454 after a small calculation , one finds for the self - linking integrand in @xmath455}\,d\theta_1\^d\theta_2\,,\ ] ] assuming the positivity condition @xmath456 ( else the integrand vanishes ) .
thus , @xmath457}\,.\ ] ] since @xmath458 is large , let us rescale the integration variables to eliminate the overall factor of @xmath81 from the argument of the exponential , @xmath459}\ , .
\end{aligned}\ ] ] after we expand the integrand of asymptotically in @xmath460 , the local contribution to @xmath215 from the right - handed crossing can be evaluated as a gaussian integral over @xmath125 , @xmath461}\,+\ , \co\big(1/\lambda\big),\\ & = \,1 \,+\ , \co\big(1/\lambda\big)\ , .
\end{aligned}\ ] ] note that all dependence on the homotopy parameter @xmath210 disappears as soon as we perform the asymptotic expansion in @xmath81 . in principle
, the contribution from the right - handed crossing in figure [ localover ] also includes the portion of the integration domain @xmath66 where the roles of @xmath462 and @xmath463 are swapped in , with @xmath464 and @xmath465 . in this case , @xmath466 is negative near the origin , and the self - linking integrand vanishes identically by the definition of the heat form @xmath184 . finally , to evaluate the local contribution from the left - handed crossing in figure [ localover ]
, we simply swap the roles of @xmath445 and @xmath446 .
apparently from , this swap is equivalent to an orientation - reversal on @xmath125 , so the sign of @xmath467 is reversed .
comparing to our conventions for the writhe in figure [ writhe ] , we conclude that @xmath467 is the local writhe of the given crossing in @xmath121 . in total , the local contribution to @xmath215 from the crossings is precisely the thurston - bennequin invariant of @xmath40 , @xmath468 where @xmath30 indexes the set of all crossings in the lagrangian projection .
the localization computation is also consistent with proposition [ hlink ] regarding heisenberg linking , together with the diagrammatic formula for @xmath1 in .
more interesting is the local contribution to @xmath215 from the diagonal @xmath222 .
this contribution does depend ( weakly ) on the value of @xmath210 , a small remnant of the topological anomaly . integrating over a neighborhood of the diagonal
really means integrating over the two boundary regions on the cylinder @xmath63 , as we have already considered in our proof of the fundamental lemma in section [ fundamental ] .
so we do not need to perform any new computations to evaluate the contribution from the diagonal .
we begin with the generic case @xmath328 , for which the local expression for the self - linking integrand appears in . directly for @xmath458 , @xmath469}\mskip-25mu d\phi\,\left[\int_0^\infty \!\!d\eta \,\,\frac{\lambda\left(\dot{\gamma}\times\ddot{\gamma}\right)}{4\pi\ , |1-\hbar|\ , |\gamma\times\dot{\gamma}|}\,\exp{\!\left(-\frac{\lambda\ , ||\dot{\gamma}||^2\,\eta}{2\,|1-\hbar|\ , |\gamma\times\dot{\gamma}|}\right)}\right ] + \,\\ & \qquad-\int_{s^1_-\cap [ ( 1-\hbar)(\gamma\times\dot{\gamma})>0]}\mskip-25mu d\phi\,\left[\int_0^\infty \!\!d\eta \,\,\frac{\lambda\left(\dot{\gamma}\times\ddot{\gamma}\right)}{4\pi\ , |1-\hbar|\ , |\gamma\times\dot{\gamma}|}\,\exp{\!\left(-\frac{\lambda\ , ||\dot{\gamma}||^2\,\eta}{2\,|1-\hbar|\ , |\gamma\times\dot{\gamma}|}\right)}\right]\ , .
\end{aligned}\ ] ] the integrals in the two lines of describe the respective local contributions to @xmath215 from collar neighborhoods of the boundary circles @xmath244 on the cylinder in figure [ cylinder ] . both integrals appear with identical signs , after one takes into account the relative boundary orientations on @xmath244 and the explicit dependence of the integrand on @xmath470 in .
we make a trivial change of variables so that the integral in the neighborhood of @xmath471 also runs over positive , as opposed to negative , values of @xmath304 . by the positivity condition in ,
the integral over @xmath309 runs over the subset where @xmath472 , and the integral over @xmath471 runs over the complement .
finally , we extend the integration range over the normal coordinate @xmath304 to infinity at the cost of an exponentially small error for large @xmath81 , and we set @xmath473 at the lower limit of integration for @xmath304 . after integrating over @xmath304 in , @xmath474}\mskip-10mu d\phi\,\,\frac{\dot{\gamma}\times\ddot{\gamma}}{||\dot{\gamma}||^2 } \,-\,\frac{1}{2\pi}\int_{s^1_-\cap[(1-\hbar)(\gamma\times\dot{\gamma})>0]}\mskip-10mu d\phi\,\,\frac{\dot{\gamma}\times\ddot{\gamma}}{||\dot{\gamma}||^2}\,,\ ] ] or put more succinctly , @xmath475 so long as @xmath328 , all dependence on @xmath210 disappears . from the geometric expression for the rotation number in , we deduce @xmath476 on general grounds , the appearance of the rotation number in this calculation is not so surprising , as one was guaranteed to find the integral of some local geometric quantity on @xmath40 .
however , the integrality of the result comes as a minor miracle , which is far from obvious from the definition of the legendrian self - linking integral in .
remember , the value of the naive gauss self - linking integral @xmath51 is not even a deformation - invariant !
we return to our formula in to evaluate the local self - linking contribution from the diagonal when @xmath212 , @xmath477}\,d\phi\^d\eta \,+\,\cdots\,.\ ] ] unlike the expressions in , which are non - zero for @xmath478 , the self - linking integrand in vanishes exponentially as @xmath335 for any @xmath79 . by inspection
we conclude @xmath479 if the heisenberg symmetry is preserved , the diagonal does not contribute to the legendrian self - linking integral .
the informal localization computation in section [ local ] is useful for developing geometric intuition about the behavior of the legendrian self - linking integral . to prove the main theorem
, we retrace the same route philosophically , but we exercise greater care in analyzing the dependence of the error terms on @xmath81 , at least when @xmath81 is large . before we establish precise inequalities in section [ bounds ] , we need to introduce a bevy of constants related to the geometry of the knot @xmath39 and its lagrangian projection @xmath121 .
these constants are important , as the required bounds fall out naturally from them .
the notion of a collar neighborhood for the boundary of @xmath481 is unambiguous , but we also need a proper notion for the neighborhood of each crossing in @xmath121
. the trick will be to choose these neighborhoods to be small enough so that the geometry of @xmath40 is controlled over each neighborhood . informally , we treated this issue by linearizing @xmath40 in figure [ localover ] , but now we work nonlinearly .
as before , @xmath150 is the regular immersed plane curve which is the lagrangian projection of the embedding @xmath118 , @xmath482 by assumption , @xmath158 has only a finite number @xmath92 of simple double - point singularities , located at positions @xmath483 see figure [ lgtref ] for our canonical trefoil example , with @xmath484 .
each crossing @xmath485 for @xmath486 lies under a pair of corresponding points @xmath487 on the knot @xmath40 , @xmath488 let @xmath489 and @xmath490 be the respective heights of @xmath491 and @xmath492 , so that these points have coordinates in @xmath0 given by @xmath493 as in the informal computation , an important geometric quantity is the absolute difference @xmath494 in the heights of @xmath491 and @xmath492 over the crossing , @xmath495 see figure [ lagproj ] for a local ( nonlinear ) picture of @xmath40 near the points @xmath491 and @xmath492 .
let @xmath496 be the open disc of radius @xmath497 centered at the location @xmath485 of a given crossing in the plane .
the union @xmath498 of these discs , each with the same radius @xmath499 , provides an open neighborhood for all crossings in @xmath121 .
we now choose @xmath497 to be sufficiently small so that the following statements are true at each crossing . by continuity of @xmath62 and compactness of the closure @xmath500 ,
such a choice is always possible . 1 .
the disc @xmath502 intersects the immersed plane curve @xmath158 in two arcs , as shown in figure [ inter ] .
we denote these arcs by @xmath503 and @xmath504 , where ` @xmath505 ' indicate the respective upper and lower strands at the crossing . over the disc ,
the embedding @xmath62 restricts to a pair of maps @xmath506 and @xmath507 , with @xmath508 . here
@xmath447 are nonlinear analogues of the expressions in . 2 . with the same arcs
@xmath509 in mind , let @xmath510 be the two - dimensional difference map @xmath511 consider the composition @xmath512 which maps the region on @xmath17 where @xmath513 is locally defined to another region on the @xmath514-plane . then @xmath515 is a diffeomorphism to a curvy quadrilateral region @xmath516 around the origin in the @xmath514-plane , as in figure [ quadri ] .
+ eventually , we will use @xmath517 to make a change - of - variables to simplify the self - linking integrand in the neighborhood of the crossing .
3 . let @xmath518 be the jacobian for the change - of - variables induced by @xmath517 from @xmath24 to @xmath519 , @xmath520 since @xmath517 is a diffeomorphism , @xmath521 is non - vanishing throughout the domain of @xmath517 , as illustrated in figure [ nonvan ] .
we go slightly further and assume that @xmath518 is uniformly bounded from below by a positive constant @xmath522 4 . for the crossing labelled by
` @xmath501 ' , consider all pairs of points on @xmath68 which lie in the image of @xmath523 over the disc @xmath524 .
then the difference in heights @xmath525 for all such pairs lies in the range @xmath526 here @xmath527 is a positive constant , independent of @xmath501 , bounded by @xmath528 + this assumption in implies that the difference @xmath529 for all pairs of points on @xmath40 projecting to @xmath524 obeys @xmath530 for @xmath527 in the given range .
informally , the constant @xmath527 controls the variation in the vertical separation between the two strands of @xmath40 over the disc @xmath524 , relative to the separation @xmath494 over the crossing itself .
the constant ` @xmath531 ' in the bound is a convenient choice related to other inequalities later .
our assumption about the radius @xmath499 of the discs @xmath524 gives us adequate control on the local geometry of @xmath40 above each crossing @xmath485 .
we now introduce further constants related to the magnitude of the self - linking integrand itself .
first , let @xmath532 be the height of the knot @xmath39 .
more formally , @xmath533 is the maximum vertical separation between any pair of points on @xmath40 , @xmath534.\ ] ] we now fix a small positive constant @xmath535 which will control the errors . in reference to the dependence of the heat form @xmath184 on @xmath536 and @xmath131 in , we note the following lemma . [ boundlm ] given @xmath537 and sufficiently large @xmath81 , there exists a constant @xmath538 , depending on @xmath81 , so that for all @xmath539 , @xmath540<\delta\,,\ ] ] and @xmath541<\delta\,.\end{aligned}\ ] ] the proof of the lemma is elementary , but we wish to gain precise knowledge about how @xmath542 must depend upon @xmath533 and @xmath81 for the bounds to hold .
the bounds in and can be treated similarly ; we start with the bound in .
we differentiate the function in with respect to @xmath131 , @xmath543\,=\,\frac{\lambda}{2\pi z^2}\big(\frac{\lambda r^2}{2z}-1\big)\exp\big(-\frac{\lambda r^2}{2z}\big)\,.\ ] ] the derivative in is positive so long as @xmath544 which in turn is equivalent to @xmath545 we will actually require a stronger bound on @xmath546 in regard to its dependence on @xmath81 .
we set @xmath547 so long as @xmath548 , the condition @xmath539 implies the bound in , so that @xmath549\,>\,0\,.\ ] ] the function in therefore increases with @xmath131 for all @xmath550 , and the supremum is achieved at the value @xmath551 , @xmath552\,=\,\frac{\lambda}{2\pi \rz}\exp\big(-\frac{\lambda r^2}{2\rz}\big)\,,\qquad\qquad r > \rr_\lambda\,.\ ] ] for @xmath539 , the argument of the exponential in satisfies @xmath553 had we imposed the weak inequality in , the quantity @xmath554 would have been bounded from below only by a constant , independent of @xmath81 , and we would have no chance to achieve the bound by @xmath443 in . instead , with the strong inequality @xmath539 in , @xmath555 for any positive @xmath535 , we now take @xmath81 sufficiently large so that @xmath556 implying via and the desired inequality in .
the analysis of the function in is identical , up to a factor of @xmath557 , due to appearance of the same gaussian factor . in this case
, we take @xmath558 . to treat both cases of lemma [ boundlm ] simultaneously
, we set @xmath559 with this choice for @xmath542 , the lemma follows .
we require one additional quantity to make our bounds on the error .
this quantity will be a function of @xmath81 which arises in reference to the magnitude of the gaussian integrand at a crossing .
briefly , for given @xmath443 and crossing index ` @xmath501 ' , we let @xmath560 be the solution to @xmath561 } \,=\,\delta\,.\ ] ] explicitly , @xmath562 as a function of @xmath81 , @xmath563 is decreasing for sufficiently large @xmath81 and vanishes in the limit @xmath83 . the most fundamental bound emerges
trivially from the definition of the constant @xmath542 in lemma [ boundlm ] .
let @xmath564 and @xmath565 be the subsets of the cylinder @xmath481 defined by @xmath566\big|_{(\theta_1,\theta_2 ) } \le \rr^2_\lambda\big\},\ ] ] and @xmath567\big|_{(\theta_1,\theta_2 ) } > \rr^2_\lambda\big\}\,.\ ] ] that is , @xmath564 consists of those pairs of points @xmath24 whose images in the @xmath80-plane under the map @xmath568 lie within the critical distance @xmath542 , and @xmath565 consists of those pairs separated in the @xmath80-plane by a distance greater than @xmath542 .
since @xmath564 and @xmath565 are complementary subsets of the cylinder , the self - linking integral can be written as the sum @xmath569 by the very definition of @xmath542 in lemma [ boundlm ] , the magnitude of the self - linking integrand on @xmath565 is everywhere bounded by @xmath443 .
automatically , @xmath570 here @xmath426 is a constant , depending on the geometry of the knot @xmath68 but independent of @xmath81 . for small @xmath443 ,
the value of @xmath215 is well - approximated by the integral over the subset @xmath564 .
our next task is to characterize the points in the domain of integration that lie in @xmath564 . if we formally set @xmath571 in , then @xmath564 consists of those pairs @xmath24 which become coincident after projection to the @xmath80-plane .
such pairs either lie along the diagonal @xmath572 , or they lie in the preimage of a crossing in the lagrangian projection of @xmath40 . when @xmath573 is positive but sufficiently small , these closed sets fatten , and @xmath564 is contained within the disjoint union of a tubular neighborhood @xmath574 of the diagonal and a collection of open balls @xmath575 associated to the crossings , @xmath576 see figure [ nbigu ] for a schematic picture of the open set containing @xmath564 for sufficiently small @xmath542 . as in the picture , each crossing in @xmath121 has two preimages , which are exchanged when the roles of @xmath462 and @xmath463 swap . thus if @xmath121 has @xmath92 crossings , the index ` @xmath577 ' on the balls @xmath578 runs to @xmath579 . in writing @xmath578 for the open ball in @xmath17 , we abuse notation somewhat .
by assumption , the radius of @xmath578 is fixed so that this ball lies in the preimage of the corresponding disc @xmath581 under the map @xmath582 in figure [ inter ] , @xmath583 therefore the radius of @xmath578 is not necessarily equal to @xmath499 , but it is determined by @xmath499 independently of @xmath81 . once @xmath499 is fixed in terms of the geometry of @xmath40 , we can always take @xmath81 sufficiently large and @xmath584 sufficiently small so that @xmath578 contains the relevant portion of @xmath564 .
similarly , the tubular neighborhood @xmath574 of the diagonal has width @xmath585 , meaning that points in @xmath574 satisfy @xmath586 for the parameters in figure [ nbigu ] .
a crucial step will be to fix the value of @xmath587 , which must be large enough so that @xmath574 contains the piece of @xmath564 near the diagonal , but also small enough so that the local analysis from section [ locald ] is applicable everywhere in @xmath574 .
according to the next lemma , both conditions on @xmath574 can be simultaneously satisfied once we set @xmath588 with a positive constant @xmath589 defined geometrically by @xmath590\,.\ ] ] since @xmath318 is an immersion , the minimum speed in is bounded away from zero , which is essential .
the constant @xmath591 is inessential and could be absorbed into the definition of @xmath425 .
the lemma says that , away from crossings , all points on the projection @xmath121 which lie within the distance @xmath542 of a given point @xmath593 are contained within the image of the interval @xmath594 $ ] under @xmath158 , for all values of @xmath132 .
see figure [ branch ] for an illustration of this claim .
the minimum speed along @xmath158 naturally sets the scale of the required interval , which decreases with increasing @xmath425 . for any fixed @xmath595 ,
define the function @xmath596 then @xmath597 , and the derivative of @xmath598 is @xmath599 since @xmath158 is an immersion , @xmath600 is bounded below , and @xmath601 is bounded above .
so if @xmath602 for sufficiently large @xmath81 and hence sufficiently small @xmath542 , then @xmath603 integrating the inequality in , we obtain @xmath604 integrating once more from the definition of @xmath605 , @xmath606 , then @xmath607 , which is the required bound .
because @xmath564 is contained in the union of @xmath574 and @xmath608 , we have a relation between the corresponding self - linking integrals , @xmath609 this inequality again follows from lemma [ boundlm ] , because points contained in either @xmath574 or @xmath578 but not in @xmath564 lie in @xmath565 , where the magnitude of the self - linking integrand is bounded by @xmath443 .
so for sufficiently large @xmath81 , we just need to evaluate the self - linking integral over the balls @xmath578 and the tubular neighborhood @xmath574 in figure [ nbigu ] .
we first evaluate the self - linking integral over the ball @xmath578 . by the positivity condition on the heat form @xmath184 , the self - linking integrand vanishes in exactly half the balls .
reshuffling indices as necessary , we consider only those @xmath610 for @xmath486 on which @xmath295 and the integrand is non - zero .
with malice aforethought , we have arranged that the image of @xmath610 under the product map @xmath582 lies in the disc @xmath611 , where we have control over the geometry of @xmath40 .
in particular , the map @xmath612 in figure [ quadri ] restricts to a diffeomorphism from @xmath610 to a curvy quadrilateral region @xmath613 about the origin in the @xmath514-plane , @xmath614 our analysis will be performed using the @xmath514-coordinates . in these coordinates , the self - linking integrand simplifies , @xmath615\\ & \qquad\,=\,\frac{\lambda}{2\pi\widehat\delta z(u , v ) } \,\e{\!-\lambda ( u^2 + v^2)/2 \widehat\delta z(u , v ) } \left[du\^dv \,+ \ha \left(u \,dv - v\,du\right)\!\^\frac{d\widehat\delta z(u , v)}{\widehat\delta z(u , v)}\right ] .
\end{aligned}\ ] ] here @xmath440 and @xmath315 are identified with the cartesian coordinates @xmath616 and @xmath617 , and @xmath322 is considered to be a function of @xmath519 .
all unknown functional dependence of the self - linking integrand in is absorbed into @xmath618 .
the integrand in is a sum of two terms , @xmath619 and @xmath620 hence after making the change - of - variables in @xmath613 , @xmath621 the analysis of the two terms on the right in is different .
morally , @xmath622 is higher - order in @xmath616 and @xmath617 so will be irrelevant when @xmath81 is large and @xmath623 .
by contrast , @xmath624 is always relevant .
we analyze the integrals of @xmath624 and @xmath622 over @xmath613 in turn .
explicitly , the integral of @xmath624 is given by a kind of nonlinear gaussian , @xmath626 } \
, du\,dv\,,\qquad\widehat\delta z(u , v ) > 0\,.\ ] ] here @xmath627 depending upon whether the diffeomorphism @xmath628 preserves or reverses orientation .
equivalently , from the expression in , the sign is determined by the jacobian in the expansion @xmath629 by inspection of figure [ jacob ] , @xmath630 is exactly the local writhe at the given crossing , @xmath631 to suppress pernicious signs for the remainder , we assume @xmath632 . * we replace the unknown function @xmath633 by the constant displacement @xmath494 at the crossing , with error @xmath634 } du\ , dv \,-\
, \int_\cq\frac{\lambda}{2\pi\delta z_a}\exp{\!\left[-\frac{\lambda\left(u^2+v^2\right)}{2\,\delta z_a}\right]}du\ , dv \right|<\ , m \delta\,.\ ] ] * we extend the subsequent range of gaussian integration from @xmath613 to @xmath125 so that the gaussian integral can be performed analytically , with error @xmath635 } du\,dv - \int_{\br^2}\frac{\lambda}{2\pi\delta z_a}\exp{\!\left[-\frac{\lambda\left(u^2+v^2\right)}{2\,\delta z_a}\right ] } du\,dv \right|<\,\delta\,.\ ] ] of these steps , only the first is non - trivial . for the second , because the gaussian integral over @xmath125 is normalized to unity independent of @xmath81 , we can always choose @xmath81 sufficiently large so that @xmath636}du\ , dv\,<\,\delta\,.\end{aligned}\ ] ] informally , we performed both these steps in arriving at . for the first step , we use heavily the positive function @xmath563 which satisfies @xmath637 } \,=\,\delta\,,\ ] ] and vanishes monotonically as @xmath83 .
the function @xmath563 sets the minimum distance from the origin for which the gaussian integrand in becomes negligible . to be on the safe side in our bounds , we will have to work with a slightly larger distance @xmath638 .
let @xmath639 be the ball of radius @xmath640 which is centered at the origin in @xmath613 , @xmath641 we shall prove that when @xmath519 lies in @xmath642 , the difference between the nonlinear and the usual gaussian in is small , @xmath643 } du\,dv \,-\ , \int_{b_0}\frac{\lambda}{2\pi\delta z_a}\exp{\!\left[-\frac{\lambda\left(u^2+v^2\right)}{2\delta z_a}\right ] } du\,dv\right| \,<\ , \delta\,.\ ] ] otherwise , when @xmath519 lies outside @xmath642 in @xmath613 , we show that both integrals are separately small , with @xmath644 } du\ , dv \
, < m\,\delta\,,\ ] ] and @xmath645 } du\,dv\,<\ , m\,\delta\,.\ ] ] thus the difference must also be small in @xmath646 .
this trick is the engine of asymptotic analysis .
see ch.@xmath647 in @xcite for further background on this idea .
we begin by establishing some easy bounds when @xmath519 lies inside the ball @xmath648 . from the definition of @xmath640 , @xmath649 so the argument of the gaussian is bounded by @xmath650 on the other hand , consider the difference @xmath651 . as a function of @xmath519 , the difference vanishes at @xmath652 and is differentiable there , so @xmath653 for some constant @xmath426 depending on @xmath40 . immediately ,
since @xmath654 scales like @xmath655 , the relative fluctuations in height about @xmath494 satisfy @xmath656 directly from and , @xmath657 the constant @xmath426 in is not necessarily the same as the constant @xmath426 in ! given the relative similarity between the integrands in , we consider their ratio @xmath658}\biggr/\frac{\lambda}{2\pi\delta z_a}\exp{\!\left[-\frac{\lambda \left(u^2+v^2\right)}{2 \delta z_a}\right]}\,.\ ] ] with some algebra , this ratio can be recast as @xmath659}\,,\\ & = \left[1\,+\,\frac{\widehat\delta z(u , v)-\delta z_a}{\delta z_a}\right]^{-1}\cdot\exp{\!\left[\frac{\lambda\left(u^2+v^2\right)}{2\delta z_a}\cdot\frac{\widehat\delta z(u , v)-\delta z_a}{\delta z_a}\cdot\frac{\delta z_a}{\widehat\delta z(u , v)}\right]}\ , .
\end{aligned}\ ] ] by the estimates in and , the prefactor in @xmath660 approaches unity and the argument of the exponential vanishes as @xmath83 .
hence we can choose @xmath81 sufficiently large so that @xmath661}\,-\,\frac{\lambda}{2\pi\delta z_a}\exp{\!\left[-\frac{\lambda\left(u^2+v^2\right)}{2\delta z_a}\right]}\right| du\ , dv\\ & \qquad\qquad\le\int_{b_0}\frac{\lambda\,|q-1|}{2\pi\delta z_a}\exp{\!\left[-\frac{\lambda\left(u^2+v^2\right)}{2\delta z_a}\right ] } du\ , dv\\ & \qquad\qquad<\,\int_{b_0}\frac{\lambda\,\delta}{2\pi\delta z_a}\exp{\!\left[-\frac{\lambda\left(u^2+v^2\right)}{2\delta z_a}\right ] } du\ , dv \,<\ , \delta\ , .
\end{aligned}\ ] ] we are left to examine what happens when @xmath519 lies outside the ball @xmath662 of radius @xmath663 , meaning @xmath664 to start , the bound on the gaussian in is trivial because the integrand is bounded by @xmath443 for all points @xmath519 outside the ball of radius @xmath563 , and @xmath665 .
so the real task is to establish the bound for the nonlinear gaussian in .
consider the following function on @xmath613 , @xmath666 conceptually , we interpret @xmath667 as a fluctuating , position - dependent version of the parameter @xmath81 , so that the width of the nonlinear gaussian varies from point - to - point on @xmath613 . by the estimate in ,
the relative fluctuation factor is bounded from below everywhere on @xmath613 by @xmath668 consequently , by the definition of @xmath667 , @xmath669 associated to the local parameter @xmath670 we have a local scale @xmath671 , also a function of @xmath616 and @xmath617 .
since @xmath672 is monotonically decreasing for large @xmath81 , the lower bound on @xmath667 in means that @xmath673 thus , again by the definition of @xmath563 in , @xmath674}\,<\,\delta\,,\end{aligned}\ ] ] or by substitution from , @xmath675}\,<\,\delta\,.\ ] ] the bound on the nonlinear gaussian in is exactly what we need to control the integral over @xmath646 , so that @xmath676 } du\ , dv\,<\,m\,\delta\,.\ ] ] combining with the trivial bound in , we deduce that the desired integral of @xmath624 over @xmath613 can be well - approximated for large @xmath81 by the naive gaussian integral , @xmath677 } du\ , dv\right| < \ , m\,\delta\,.\ ] ] we are not finished with our error analysis at the crossing , because we still must consider the integral of @xmath622 over @xmath613 in .
we will show that the contribution of @xmath622 is negligible for large @xmath81 , @xmath678 explicitly , from the formula in , the integral of @xmath622 is given in the @xmath519-coordinates by @xmath679 again , we consider the cases that @xmath519 lies inside the ball @xmath642 and outside the ball @xmath642 separately .
when @xmath519 lies outside the ball @xmath680 , then by the definition of @xmath563 in , @xmath681 here we note that the extra factors of @xmath682 and @xmath683 in are smooth functions bounded independently of @xmath81 on @xmath613 .
these functions do not alter the bound by @xmath443 but are absorbed into the constant @xmath426 .
otherwise , for points inside @xmath642 , we have a bound @xmath684 } du\,dv\ , < m\,,\ ] ] which follows by the same arguments used to produce the estimate in . also , since @xmath683 is bounded in @xmath642 and @xmath685 as @xmath686 , we can always choose @xmath81 so that @xmath687 for all @xmath519 in @xmath642 . combining the bounds in , , and for outside and inside @xmath642
, we obtain the conclusion in .
our final goal is to evaluate the self - linking integral over the tubular neighborhood @xmath574 of the diagonal @xmath688 , where @xmath689 is the width set in lemma [ boundw ] . for the informal localization computation in , we used the leading term in the taylor expansion of the self - linking integrand near the diagonal to approximate the integral .
depending upon the value of the parameter @xmath210 , we abbreviate this leading term by @xmath690}\,d\phi\^d\eta\,,\ ] ] or @xmath691}\,d\phi\^d\eta\,.\ ] ] in both cases we assume that the local positivity condition @xmath692 is satisfied , as in . otherwise , @xmath693 . to justify our localization computation
, we must demonstrate for sufficiently large @xmath81 the bound @xmath694 the behavior of @xmath695 for small @xmath304 depends very much on whether @xmath210 is equal to one or not , so we treat the cases in and separately . because the generic case @xmath328 is the more involved , and the more interesting , we begin with it . in principle , the error in the leading approximation to @xmath696 for small @xmath304 is controlled by the magnitude of the next - order term in the taylor expansion .
we will need this correction term for our analysis .
briefly , by the same computations leading to in section [ locald ] , the argument of the heat kernel admits the second - order expansion @xmath697,\ ] ] where @xmath698 .
similarly for the pullback of the heat form @xmath184 itself , @xmath699\times\\ & \times\,\exp{\!\left[-\frac{\lambda\,||\dot{\gamma}||^2 \,|\eta|}{2 \left|1-\hbar\right|\left|\gamma\times\dot{\gamma}\right|}\left(1 \,+ \left(\frac{\dot{\gamma}\cdot\ddot{\gamma}}{||\dot{\gamma}||^2 } - \ha \frac{\gamma\times\ddot{\gamma}}{\gamma\times\dot{\gamma}}\right)\eta \,+\ , \co\big(\eta^2\big ) \right)\right]}\ , .
\end{aligned}\ ] ] we omit the computation leading to , since the details of this formula will not be so important .
the expansion merely confirms that both the prefactor and the argument of the exponential for @xmath695 in receive further corrections at the next order in @xmath304 , as determined by the geometry of the projection @xmath121 .
validity of the leading approximation @xmath695 requires that the correction terms in be small .
at least informally , for the argument of the heat kernel in we require @xmath700 by assumption , @xmath701 is always small on @xmath574 , and @xmath393 is bounded from below .
the condition in is therefore only violated at points where @xmath702 . at these points , the taylor expansion in breaks down . despite the failure of the taylor expansion at points where @xmath703 , these points cause no difficulty .
recall that points where @xmath702 correspond to critical points of the height function @xmath268 on @xmath40 . by the morse assumption which follows lemma [ fundlm ] ,
the function @xmath704 vanishes at only a finite number of isolated critical points @xmath705 on @xmath235 . according to the analysis at the end of section [ proof ] , these points are precisely the endpoints of the disjoint collection of intervals @xmath706}$ ] in . at the endpoints , @xmath695 vanishes , and the exact integrand @xmath696 in is exponentially small ( since @xmath322 is small ) .
consequently , the troublesome points for the taylor expansion can just be removed from the domain of integration , with negligible error .
technically , about each critical point @xmath708 , @xmath709 we consider a small neighborhood @xmath710 with fixed width @xmath711 .
let @xmath712 be the corresponding closed strip @xmath713 and set @xmath714 see figure [ edge ] for a sketch of @xmath715 near one boundary of the cylinder @xmath66 .
the shaded regions indicate the strips of width @xmath716 which have been excised about a pair of zeroes @xmath406 and @xmath717 of the function @xmath337 .
we choose the width @xmath718 of each strip to be small enough so that @xmath719 both integrands in are bounded at the points @xmath708 , so the integrals over @xmath720 can be made as small as desired by the choice of @xmath716 .
moreover , both integrands are decreasing functions of @xmath81 for sufficiently large @xmath81 , so the width @xmath716 can be chosen independently of @xmath81 , our crucial requirement . by definition , the function @xmath721 is now bounded away from zero everywhere on the new domain @xmath715 , @xmath722 here @xmath425 is a constant which depends upon the curve @xmath40 and the parameter @xmath723 , but not on @xmath81 . because the respective contributions from the excised strips are small by assumption , we are free to replace @xmath574 by @xmath715 in the inequality to be proven . on the other hand , due to the lower bound in , we will also have uniform control of error terms such as in the taylor approximation on @xmath715 .
the remainder of the discussion proceeds in rough correspondence to the asymptotic analysis near a crossing . by analogy to the ball @xmath642 in , we introduce a smaller tubular neighborhood @xmath724 defined by @xmath725 we indicate the coaxial configuration schematically in figure [ edge ] . for points inside the small tube @xmath726 , we will show that the difference between the integrals of @xmath696 and @xmath695 is small , @xmath727 for points outside @xmath726 but inside @xmath715 , we will show that both integrals are separately small , with @xmath728 and @xmath729 the extra factor of @xmath730 in the definition of @xmath726 is simply what is needed to ensure the inequality in .
we first consider the points inside @xmath726 . from the second - order expansion in , the ratio of the self - linking integrand to its approximation @xmath695 satisfies @xmath731\times\\ & \times\,\exp{\!\left[-\frac{\lambda\,||\dot{\gamma}||^2 \,|\eta|}{2 \left|1-\hbar\right|\left|\gamma\times\dot{\gamma}\right|}\left(\frac{\dot{\gamma}\cdot\ddot{\gamma}}{||\dot{\gamma}||^2 } - \ha \frac{\gamma\times\ddot{\gamma}}{\gamma\times\dot{\gamma}}\right ) \eta \,+\ , \co\big(\eta^3\big)\right]}\ , .
\end{aligned}\ ] ] again , to deal with the term @xmath732 in the prefactor of , we assume that any zeroes of @xmath338 are isolated , and we remove small neighborhoods as necessary about those zeroes so that the functions which multiply @xmath304 in both the prefactor and the argument of the exponential in are bounded , independently of @xmath81 . for any point in the small tube @xmath726 ,
the argument of the exponential in is bounded in magnitude by @xmath733 where we apply the conditions @xmath707 as well as @xmath734 in @xmath726 . because @xmath735 , this inequality means that the argument of the exponential vanishes , and the prefactor approaches unity , in the limit @xmath83 .
thus for sufficiently large @xmath81 , the fractional error is small , @xmath736 by the same idea in , @xmath737 since we already know the integral of @xmath695 to be bounded and independent of @xmath81 by the local computation in section [ local ] . from the explicit expression for @xmath695 in , @xmath738}\,,\qquad\qquad a , b\,>\,0\,,\ ] ] for some positive constants @xmath395 and @xmath396 .
so in the allowed range @xmath739 on the complement of @xmath726 , @xmath740}\,,\qquad\qquad \rw \,=\ , m \ , \lambda^{-1/4}\,,\\ & = \ , a \ , \lambda \exp{\!\left[-m\ ,
b \ , \lambda^{1/4}\right]}\ , . \end{aligned}\ ] ] by taking @xmath81 sufficiently large , we can make the magnitude of @xmath695 as small as desired on the complement of @xmath726 inside @xmath715 , from which the bound in follows . to establish a similar bound for the pullback of @xmath184 in ,
observe that the argument of the exponential obeys @xmath741 > \ , b\,|\eta|\,,\qquad\qquad b>0\,,\ ] ] provided that @xmath81 is sufficiently large and @xmath304 sufficiently small . here
@xmath742 is a suitable positive constant .
then according to the expansion in , @xmath743}\,,\qquad\qquad a , b\,>\,0\,,\ ] ] exactly as for the preceding bound on @xmath695 in .
the claim in now follows by an identical argument . for the heisenberg - symmetric value @xmath212 , the localization formula from section [ local ] states @xmath745 .
consistent with this result , we establish the basic bound in by showing individually @xmath746 and @xmath747 our workhorse is the next - order expansion of the self - linking integrand , which behaves differently for @xmath212 . for the argument of the heat kernel , the calculations in section [ locald ] yield @xmath748.\ ] ] similarly , @xmath749\times\,\\ & \times\,\exp{\!\left[-\frac{3\,\lambda}{|\eta|}\cdot\left|\frac{||\dot{\gamma}||^2 \,+ \left(\dot{\gamma}\cdot\ddot{\gamma}\right ) \eta \,+\ , \co\big(\eta^2\big)}{\left(\dot{\gamma}\times\ddot{\gamma}\right ) \,+\ , \ha \left(\dot{\gamma}\times\dddot{\gamma}\right ) \eta \,+\ , \co\big(\eta^2\big)}\right|\right]}\ , .
\end{aligned}\ ] ] for sake of brevity , we omit the calculation leading to .
the details of this formula are not important .
correction to the prefactor in vanishes when @xmath212 . ] for the approximation @xmath695 , recall the formula @xmath750}\,d\phi\^d\eta\,.\ ] ] then @xmath695 vanishes smoothly for @xmath478 and otherwise satisfies @xmath751}\,,\qquad\qquad a , m\,>\,0\,,\ ] ] for some positive constants @xmath395 and @xmath426 . for @xmath81 sufficiently large
, @xmath695 can be made as small as desired everywhere on @xmath574 , and the inequality in holds . to treat the pullback of @xmath184 in the same fashion , note that the denominator in obeys @xmath752 provided @xmath753 is sufficiently small ( with @xmath754 ) , as holds when @xmath81 is sufficiently large .
also in this regime , the numerator in is bounded from below by @xmath755 hence on the tubular neighborhood @xmath574 , @xmath756}\,.\ ] ] this inequality has the same shape as that for @xmath695 in , from which we reach the conclusion in .
j. b. etnyre , `` introductory lectures on contact geometry , '' in _ topology and geometry of manifolds ( athens , ga 2001 ) _ , proc .
sympos . pure math . ,
soc . , providence , ri , 2003 , http://arxiv.org/abs/math/0111118 [ arxiv : math/0111118[math.sg ] ] .
h. geiges , `` contact geometry , '' in _ handbook of differential geometry , vol.2 _ , pp.315382 , ed . by f.dillen and l.verstraelen , north - holland , amsterdam , 2006 , http://arxiv.org/abs/math/0307242[arxiv:math/0307242[math.sg ] ] . | the thurston - bennequin invariant provides one notion of self - linking for any homologically - trivial legendrian curve in a contact three - manifold . here
we discuss related analytic notions of self - linking for legendrian knots in @xmath0 .
our definition is based upon a reformulation of the elementary gauss linking integral and is motivated by ideas from supersymmetric gauge theory .
we recover the thurston - bennequin invariant as a special case . |
three - dimensional ( 3d ) topological insulators ( tis ) are characterized by a novel topological order @xcite which dictates the appearance of spin - filtered massless dirac fermions on the surface.@xcite to experimentally address the peculiar physics of 3d tis , it is desirable to access the dirac point of the surface state ( ss).@xcite this is relatively easy with the surface - sensitive spectroscopies such as the angle - resolved photoemission @xcite and the scanning tunneling microscope,@xcite but it is more challenging for the bulk - sensitive transport experiments because the chemical potential is always pinned to the bulk bands ( including the impurity band ) in real materials . to tune the chemical potential to a desirable position for transport experiments ,
two approaches have been employed : one is the tuning of the chemical compositions upon synthesizing crystals,@xcite and the other is the gating to control the surface carriers.@xcite among the latter approach , the electric - double - layer gating ( edlg ) method is a promising new technique @xcite to allow application of a large electric field.@xcite in the edlg configuration , either cations or anions in a liquid electrolyte are accumulated near the surface of a sample by application of an electric field , and they form an electric double layer which generates a very strong electric field locally on the surface .
this leads to the induced surface carrier density of as high as @xmath110@xmath2 @xmath3.@xcite such a large tunability of the surface carrier density is the merit of the edlg method . in the context of tis , yuan _ et al .
_ showed @xcite that the edlg method allows an ambipolar doping control in ultrathin films of bi@xmath0te@xmath4 ; however , the surface state was obviously gapped in the ultrathin films used in ref . , and consequently the transport properties were likely to be dominated by the bulk state .
in the present paper , we show that the edlg method can achieve ambipolar transport even in bulk single crystals , if one uses the highly bulk - insulating bi@xmath5sb@xmath6te@xmath7se@xmath8 ( bsts ) system.@xcite in our experiment , the control of the chemical potential was possible on the _ whole surface of a bulk 3d sample _ , opening new experimental opportunities for tis .
in addition , our data suggest that the chemical potential is moving not only at the surface but also in the bulk , possibly due to a nearly reversible electrochemical reaction to cause apparent bulk doping during the edlg process .
a series of bsts single crystals were grown by a modified bridgman method.@xcite six gold wires were attached to each sample by a spot welding technique , and magnetotransport measurements were performed by a conventional ac six - probe method by sweeping the magnetic field between @xmath99 t. an electrically insulating cup made of stycast 1266 was used as a sample container , in which the sample with gold wires was submerged into an ionic liquid ( il ) electrolyte , as schematically shown in fig .
we used a specially purified ionic liquid [ etmeim][bf@xmath10 as the electrolyte.@xcite as the gate electrode , an additional piece of gold wire was dipped into the il without touching the sample . in this paper , we adopt the convention of defining the gate voltage by taking the sample as the reference point , as was done in most of the previous works . in this convention ,
@xmath11-type carriers are supposed to be doped to the surface when a positive gate voltage is applied .
the gate voltage was applied in the following procedure : first , the temperature @xmath12 of the sample was stabilized at 220 k ; then , the gate voltage @xmath13 was swept slowly , and the temperature was kept at 220 k for at least 10 min after the voltage reached the set value ; finally , the sample was cooled down slowly . when changing the gate voltage , we cycled the above procedure
. figures 1(b)-1(e ) show an example of the history of @xmath12 , @xmath13 , gate current @xmath14 , and the accumulated charge , during the procedure for the target @xmath13 of @xmath15 v. with the il used in the present experiment , gate current was never detected at temperatures below 200 k , even when the gate voltage was changed .
this is because when the il solidifies , ions are not mobile at all .
the accumulated ions can be released when the gate voltage is set to zero and the system is warmed up to 220 k. such a relaxation process from @xmath13 = @xmath15 v is shown in figs .
1(f)-1(h ) .
one can see that the amount of released charge [
@xmath132 @xmath16c in fig .
1(h ) ] is comparable to that of the accumulated charge [
@xmath134 @xmath16c in fig .
the small difference between the released and accumulated charge is probably an indication of an irreversible electrochemical reaction during the gating processes at a high gate voltage . in the above example ,
the total surface area of the sample was 13.4 mm@xmath17 , so that the @xmath132 @xmath16c of charge accumulated on the surface corresponds to the accumulated ion density of @xmath11.6@xmath18 @xmath3 and the capacitance of the unit area 80 @xmath16f @xmath3 , which is comparable to that previously reported for edlg.@xcite figure 1(i ) shows @xmath13 vs @xmath14 curve at 220 k , which indicates that there is a threshold @xmath19 of @xmath12 v below which little current flows , suggesting that the ions are not mobile below this threshold voltage ; similar behavior was previously reported for a different il.@xcite the origin of this behavior is currently not clear , but it may well be a characteristic of the ionic liquid used here . also , it was difficult to obtain reproducible results of the transport properties for @xmath19 between 2 and 3 v , probably because the formation of the electric double layer is unstable in this gate - voltage region . therefore , we closely measured the transport properties only in the range of @xmath20 v and @xmath13 = 0 v.
sb@xmath22te@xmath23se@xmath24 single crystal for @xmath25 , 0 , and + 3.5 v. ( c ) magnetic - filed dependences of @xmath26 at 1.8 k. in these three figures ( a)-(c ) , the dotted , solid , and broken lines correspond to @xmath13 of @xmath15 , 0 , and + 3.5 v , respectively .
( d ) fitting of the two - band conduction model to the @xmath27 data at @xmath13 = 0 v , which yields @xmath28 = 3.2 @xmath29 , @xmath30 = 9.1 @xmath31 10@xmath32 @xmath33 , @xmath34 = 29 cm@xmath17/v@xmath36s , @xmath37 = 8.1 @xmath38 , @xmath39 = 1.6 @xmath31 10@xmath40 @xmath41 , and @xmath42 = 4.7 @xmath31 10@xmath43 cm@xmath17/v@xmath36s .
the gray thick line represents the experimental result and the solid line is the fitted result.,width=321 ] first , we show the results of the edlg experiments on bi@xmath44sb@xmath22te@xmath23se@xmath24 , which is currently one of the most bulk - insulating ti materials .
@xcite figure 2(a ) shows the temperature dependences of the resistivity @xmath45 with the gate voltage ( @xmath13 ) of @xmath15 , 0 , and + 3.5 v. even though this is a bulk single crystal , at low temperatures @xmath45 presents a clear change upon application of @xmath13 ; namely , below @xmath150 k , @xmath45 changes with changing @xmath13 , and it tends to decrease ( increase ) for positive ( negative ) @xmath13 .
figure 2(b ) magnifies this change for @xmath46 20 k. figure 2(c ) shows the magnetic - field dependences of the hall resistivity , @xmath27 .
the sign of the charge carriers remains negative in this sample , although a significant non - linearity suggests the coexistence of surface and bulk conduction channels .
note that the surface conduction in tis can involve both the topological surface states and topologically - trivial two - dimensional electron - gas ( 2deg ) states that appear as a result of band bending .
@xcite in the present experiment , when the gate voltage is applied , the 2deg states are most likely contributing to the conduction alongside of the topological surface states .
unfortunately , we have not been able to elucidate the contributions of the topological and non - topological surface states because no shubnikov - de haas ( sdh ) oscillation has been observed in our gated samples .
( to successfully separate the contributions of the two , detailed information obtained from sdh oscillations is necessary.@xcite ) we therefore make no claim of the composition of the surface carriers in the present paper . in fig .
2(c ) , one can see a clear tendency that both the absolute value of @xmath26 and the slope of @xmath27 decrease upon increasing @xmath13 from @xmath15 to + 3.5 v ; this means that the apparent electron concentration increases with increasing @xmath13 .
since @xmath11-type carriers are expected to be doped when a positive @xmath13 is applied , the present observation can be understood as a natural consequence of the edlg .
the nonlinear @xmath47-dependence observed in @xmath26 is useful for gaining insights into the respective roles of bulk and surface transport channels , because its analysis based on a simple two - band model @xcite gives a crude idea about the relevant transport parameters .
for example , the @xmath27 data at @xmath13 = 0 v give the following estimate based on the fitting shown in fig .
2(d ) : for the bulk channel , the bulk resistivity @xmath48 3 @xmath49 , the bulk carrier density @xmath50 9 @xmath31 10@xmath32 @xmath33 , and the bulk mobility @xmath51 30 cm@xmath17/v@xmath36s ; for the surface channel , the sheet resistance @xmath52 8 @xmath38 , the surface carrier density @xmath53 2 @xmath31 10@xmath40 @xmath41 , and the surface mobility @xmath54 5 @xmath31 10@xmath43 cm@xmath17/v@xmath36s . in this
fitting , the constraint imposed by the presence of sharp kinks at @xmath551 t helps reduce the ambiguity in the fitting parameters , and in fact , the three parameters , @xmath30 , @xmath56 , and @xmath39 , are in reasonable agreement with our previous transport studies of bsts involving sdh oscillations.@xcite the large value of @xmath42 would imply that the sdh oscillations be observed , but we did not observe any sdh oscillations in this sample ; this is possibly because the surface chemical potential ( and hence @xmath39 ) is not very uniform throughout the sample and the sdh oscillations with various different frequencies add up to smear visible oscillations .
the above result of the two - band analysis suggests that the surface contribution in the total conductance was only @xmath11% , which is reasonable because the measured sample had a considerable thickness of 332 @xmath16 m . nevertheless , if one looks at the edlg effect in resistivity [ figs .
2(a ) and 2(b ) ] , one notices a very puzzling fact : for the negative @xmath13 of @xmath15 v , the number of surface electrons are expected to be reduced and indeed , the slope of @xmath26 gets larger ; however , the resistivity increase is as much as 25% .
since the surface contribution in the total conductance is only @xmath11% , even when the surface conduction is completely suppressed by edlg , one can expect an increase in @xmath45 of @xmath11% at most , as long as the bulk channel is not affected by edlg .
therefore , the observed large increase in @xmath45 strongly suggests that the bulk channel must also be affected by edlg .
indeed , the @xmath27 data shown in fig .
2(c ) presents a clear change in the slope at high fields for different @xmath13 , which implies that not only @xmath39 but also @xmath30 is changing . to corroborate this inference , the two - band analyses of the @xmath27 data at finite
@xmath13 suggests that the bulk carriers decreases ( increases ) by 5% ( 10% ) for @xmath13 of @xmath15 v ( @xmath57 v ) .
it is very surprising that the edlg affects the density of bulk carriers by a noticeable amount in a sample as thick as 332 @xmath16 m , but our transport data can hardly be understood if one does not accept this possibility . given that the electric field generated by @xmath11 @xmath31 10@xmath2 @xmath3 of ions on the surface is shielded in less than 100 nm,@xcite the only possibility is that some bulk doping into the bsts sample is taking place during the edlg process . in this respect , the slow time scale of the change in the measured current during the edlg process seems to support the idea that some electrochemical reaction is taking place .
in passing , we note that we have not successfully measured sdh oscillations in any of the gated samples .
this is likely to be due to an inhomogeneous distribution of the local electric field ( which is conceivable because our samples have a lot of macroscopic terraces on the surface ) or some chemical degradation of the surface caused by the ionic - liquid gating .
hence a more definitive analysis of the transport data beyond the simple two - band analysis is currently unavailable .
nevertheless , the bulk doping due to the edlg seems to be an inevitable conclusion of our result .
it was recently found @xcite that the position of the chemical potential with respect to the dirac point of the ss is tunable within the bulk band gap in bsts when one follows a particular series of @xmath59 and @xmath60 that were identified in ref . .
in particular , the chemical potential was found to be close to both the dirac point of the ss and the middle of the bulk band gap in bisbtese@xmath0 .
therefore , for our edlg experiment , to maximize the possibility of achieving ambipolar transport , we mainly measured the bisbtese@xmath0 system .
( the gating data shown in fig . 1 were taken during the experiment on the bisbtese@xmath0 sample reported below . ) in the bisbtese@xmath61 single crystal used for the edlg experiment at @xmath13 = 0 v in a semi - log plot ; inset shows the @xmath13 dependence of @xmath45 at 1.8 k , where the arrows indicate the order of experiments .
( b),(c ) low - temperature part of @xmath62 for various @xmath13.,width=321 ] .experimental order of applied @xmath13 and resultant transport properties at low temperatures .
@xmath63 is the low field limit of the hall coefficient . [
cols="^,^,^,^ " , ] looking at the results of the two - band analyses of the data at finite @xmath13 , one can see that the hall data strongly suggests that both the bulk and surface carriers change sign simultaneously at @xmath64 v. indeed , we found that it is impossible to fit the @xmath26 data for @xmath64 v by assuming that only one of the two channels changes sign .
therefore , it appears that in the present edlg experiment the chemical potential is swung from the @xmath11-type regime to the @xmath65-type regime not only on the surface but also in the bulk .
most likely , what is happening in the bulk at negative @xmath13 is compensation due to electrochemical @xmath65-type doping , which eventually overwhelms the preexisting @xmath11-type bulk carriers . at 1.7 k.,width=245 ] finally , we show in fig .
6 the magnetoresistance ( mr ) data for various @xmath13 values .
obviously , the data present the weak antilocalization ( wal ) effect at low fields;@xcite since the wal effect is a signature of two - dimensional transport and is not usually observed in bulk ti crystals dominated three - dimensional transport , the mr data give additional evidence for a sizable contribution of the surface transport to the total conductance . at the same time , this wal effect makes it difficult to examine the consistency of the parameters obtained from the two - band analysis of @xmath27 in the mr data .
also , the behavior of mr qualitatively changed after the application of @xmath13 = @xmath66 v , the origin of which is not clear at the moment . since the mr is too complicated and not very reproducible , we did not try to make a detailed analysis .
it is important to mention that there are missing charges in our edlg experiment ; namely , the total amount of charge induced by gating in the sample is much smaller than the total amount of ions accumulated on the surface .
for example , in the case of bisbtese@xmath0 , the total amount of accumulated charges measured by the current is @xmath134 @xmath16c , which corresponds to the ion density on the surface of @xmath11.6 @xmath31 10@xmath2 @xmath3 . on the other hand ,
the change in the surface carrier density in bisbtese@xmath0 was @xmath14 @xmath31 10@xmath67 @xmath3 and its bulk carrier density changed by less than 1 @xmath31 10@xmath68 @xmath33 , which amounts to the total charge of less than 20 @xmath16c .
therefore , obviously the gating is not very efficiently performed . in this regard ,
the doping control of the surface carriers in the present experiment is similar to another edlg experiment on bi@xmath0te@xmath4 thin film,@xcite where @xmath17 @xmath31 10@xmath67 @xmath3 of surface carriers were doped with @xmath69 @xmath70 v. for other materials , the amount of surface carrier doping by edlg is of the order of 10@xmath7110@xmath2 @xmath3 , @xcite and thus the electric - field effect on bi - based topological insulators appears to be exceptionally ineffective.@xcite the mechanism to cause these missing charges is not clear at the moment , but we speculate that some electrochemical redox reaction involving adsorbed molecules on the surface causes a layer of immobile charges that shields some fraction of the electric field created by the ions , leading to a weakening of the electric field for inducing mobile carriers in the sample . also , the bulk doping that accompanies the edlg in our bsts samples is surprising .
remarkably , the data for bisbtese@xmath0 suggests that the bulk doping process is nearly reversible and it takes place in the time scale of the order of 10 min .
the chemical mechanism of this bulk doping is not clear at the moment , but the possible cause might be intercalation of ions into the van - der - waals gap in the bsts crystal .
obviously , there is a lot to understand about the electrochemistry accompanying the edlg on bi - based tetradymite ti materials .
in conclusion , the electric - double - layer gating ( edlg ) using ionic liquid was applied to bulk single crystals of bsts to control the chemical potential , and ambipolar transport was observed in a sample of bisbtese@xmath0 as thick as 181 @xmath16 m .
the gating was successfully applied to tune the chemical potential on the whole surface of a three - dimensional sample , and surprisingly , it appears that the edlg on bsts crystals is accompanied by a nearly reversible electrochemical reaction that caused bulk carrier doping .
it turned out that the edlg is exceptionally inefficient for the bsts system , with the maximum change in the surface carrier density of @xmath14 @xmath31 10@xmath67 @xmath3 despite the ion density on the surface of @xmath11.6 @xmath31 10@xmath2 @xmath3 .
the key to the successful ambipolar carrier control in the present experiment was the use of bisbtese@xmath0 crystals in which the chemical potential is located close to the middle of the bulk band gap @xcite and the residual bulk carrier density was only @xmath11 @xmath31 10@xmath32 @xmath33 . in combination with a technique to open a gap on the surface,@xcite the present experiment paves the way for topological magnetoelectric - effect experiments @xcite that require the chemical - potential control on the whole surface of a bulk topological insulator , although the mechanism of the bulk doping associated with edlg needs to be understood before this technique is comfortably applied .
this work was supported by jsps ( next program ) , mext ( innovative area topological quantum phenomena " kakenhi 22103004 and kakenhi 20371297 ) , and afosr ( aoard 104103 and 124038 ) .
p. cheng , c. song , t. zhang , y. zhang , y. wang , j .- f .
jia , j. wang , y. wang , b .- f .
zhu , x. chen , x. ma , k. he , l. wang , x. dai , z. fang , x. xie , x .-
qi , c .- x .
liu , s .- c .
zhang , and q .- k .
xue , phys .
lett . * 105 * , 076801 ( 2010 ) .
there is a report that @xmath11.3 @xmath31 10@xmath72 @xmath3 of carriers were doped to bi@xmath0se@xmath4 films by edlg [ y. onose _ et al .
express * 4 * , 083001 ( 2011 ) ] , but the 6-nm films studied in that work were heavily se deficient and contained as much as 2.2 @xmath31 10@xmath73 @xmath33 of electrons .
hence , the efficiency of the edlg on ti samples may be improved in more metallic samples . | we report that the ionic - liquid gating of bulk single crystals of a topological insulator can control the type of the surface carriers and even results in ambipolar transport .
this was made possible by the use of a highly bulk - insulating bisbtese@xmath0 system where the chemical potential is located close to both the surface dirac point and the middle of the bulk band gap .
thanks to the use of ionic liquid , the control of the surface chemical potential by gating was possible on the whole surface of a bulk three - dimensional sample , opening new experimental opportunities for topological insulators .
in addition , our data suggest the existence of a nearly reversible electrochemical reaction that causes bulk carrier doping into the crystal during the ionic - liquid gating process . |
spectroscopic observations of solar chromospheric flares show that the line profiles emitted by the flaring plasma almost always exhibit asymmetries or shifts .
these features are surely due to the chromospheric plasma motion and the resulting doppler - shifts effects .
interpretation of the shape of line profiles allows us to understand the nature of plasma flows during solar flares .
plasma flows in the chromosphere are important in the analysis of dynamics and energetics of solar flares as well as these phenomena supply matter to the coronal parts of flares in the process of chromospheric evaporation .
understanding of the mechanisms which generate flows is necessary for complete description of solar flares .
spectral line asymmetries arise only because of the line - of - sight component of plasma velocity which due to the doppler - shift is responsible for the modification of spectral line profiles . therefore , for the flares located on the solar disk center we are able to analyze flows oriented perpendicular to the solar surface . for simplicity , the theoretical analysis of line asymmetries often assume that the emitting region is located on the solar disk center and only vertical flows are present . unfortunately , the interpretation of the shape of line profiles is not trivial and to determine the velocity field we must use advanced methods based on hydrodynamical and radiative transfer techniques .
there are some rare cases when the interpretation of line asymmetries is relatively simple .
when the flow of the whole flaring region is homogeneous then the shape of the line is not disturbed compared to the static symmetric line profile , but the whole line is just shifted towards longer or shorter wavelengths ( fig . [ ab - fig : prof ] left ) .
we can measure this shift and using a simple doppler formula calculate specific value of the line - of - sight velocity .
unfortunately , in most cases the line profiles of solar flares exhibit much more complicated structure ( fig .
[ ab - fig : prof ] right ) which suggests that the velocity field is not homogeneous and different parts of the flare move in different way .
there are two main approaches to the modeling of chromospheric line asymmetries .
one is based on the hydrodynamics calculations , where the time evolution of the solar flare atmosphere is calculated and the radiative transfer formulae are used to calculate the radiation from this evolving atmosphere .
another approach is based on semiempirical models of solar flares . in this paper
i will present a short review on observation and interpretation of the chromospheric line asymmetries observed during solar flares .
the term chromospheric lines is commonly used to describe all spectral lines formed in the solar chromosphere , where the temperature is around @xmath0k .
these lines are formed in strong non - lte conditions and complicated radiative transfer calculations are necessary to describe the formation of these lines .
strong chromospheric lines are usually optically thick what means that the optical thickness of the plasma in these lines is very large ( @xmath1 ) . asymmetric line profiles .
profiles 4 , 5 , and 6 correspond to the flare emission ( @xcite ) .
, width=302 ] there is a wide literature concerning this topic and i provide the readers some references contained the most important results .
i will concentrate on chromospheric parts of solar flares where the emission comes from cool ( @xmath0k ) plasma .
this emission is produced mostly in strong resonance lines of hydrogen , calcium or magnesium ( h@xmath2 , h@xmath3 , h@xmath4 , caiih , k , etc . ) since the most spectroscopic observations performed during past years concern h@xmath2 line , the interpretation of these results will take considerable part of this review .
asymmetries of chromospheric lines emitted by the flaring plasma have been observed for more than half of the century .
after the solar spectrographs were developed in order to produce the solar spectrum with sufficient spectral resolution , observers noticed that the line profiles emitted during solar flares are not symmetric ( fig .
[ ab - fig : ellison - prof ] ) .
it was clear from the beginning that these kinds of modifications of the spectral lines are due to the mass motion driven during solar flares .
however , the mechanisms which could drive the plasma flows were not known yet at those days .
the first analysis of the line asymmetries were concentrated on statistical description of the behaviour of lines .
@xcite presented a qualitative analysis of 244 h@xmath2 and caiik spectra of 92 flares .
they found that the blue asymmetry ( blue wing enhancement ) occurs mainly in the early phase of flares , before flare maximum .
however , only 23% of flares contain at least one region with blue asymmetry .
80% of flares exhibit the red asymmetry which dominates during and after the maximum of flare .
it is worth to notice that only 5% of flares shows blue asymmetry exclusively .
however , because not all flares were observed from their beginning , the occurrence of blue asymmetry may be missed for many flares . in fig .
[ ab - fig : asymm - evol ] the time evolution of the asymmetry is presented .
similar analysis was performed by @xcite . by inspection of off - band filtergrams of 60 flares obtained in @xmath51 and @xmath52 from the h@xmath2 line center he found that 92% of flares show red asymmetry and only 5% show blue asymmetry .
line profiles as a function of time development of flares ( solid line ) .
dashed curve takes positive and negative signs into account .
time of flare maximum = 0 ( @xcite ) . , width=377 ] early statistical analysis gave us an qualitative idea about the line asymmetry but the physical interpretation of the shape of line profiles is not considered .
the asymmetric line profile contains an information about the velocity field in the region where the observed line is formed .
therefore , the first problem which needs to be solved is to determine this velocity field using the observed line profile .
secondly , we have to answer the question why do we observe such a flow of plasma , what can generate the flows ?
determination of the velocity from the observed profiles of the chromospheric lines is not a trivial task due to the complicated processes of line formation and complex velocity field in the chromosphere .
strong chromospheric lines are optically thick and the radiation observed at different parts of the line profile comes from different height @xmath6 across the chromosphere ( @xcite ) .
the core of strong chromospheric lines ( hydrogen balmer lines , ca lines ) is formed much higher than the wings of these lines .
therefore , the asymmetry of the specific line depends on the relation between the height @xmath6 of formation of given part of the line profile and the value of the velocity at this height .
this means that if the function describing the velocity across the height in the chromosphere is complicated , the emergent line profile has also very complicated shape . despite of all difficulties with determining correctly the velocity from the line profile shape , many authors tried to use h@xmath2 , caii and other lines to find the velocity in the flaring chromosphere .
all these determinations were based on the measurements of doppler - shifts of the line cores or , more commonly , on the bisector technique .
using the shift of the line core to obtain the velocity may be misleading and the determined velocity is not correct when the velocity gradient in the chromosphere is significant ( @xcite ) .
unfortunately , the estimation of the doppler - shift obtained with the bisector technique can also give wrong results .
the doppler - shift of the line profile correspond to the shift of the central point of the bisector connecting the two wings of the spectral line .
since the different parts of the spectral line are formed at different height in the chromosphere , using bisectors connecting the wings observed at different frequency , we can estimate the line - of - sight velocity at different layers of the chromosphere . however , because the radiation of specific frequency within the spectral line does not come from one narrow layer of the chromosphere but rather from geometrically thick region , we can not say that the doppler velocity determined form given bisector correspond to plasma flow at given height in the chromosphere . moreover , if the velocity gradient in the chromosphere is large then the bisector method can not be used because the doppler - shift of any bisector results from a superposition of many shifts due to the motion of plasma with different velocity along line - of - sight . in spite of this
the bisector technique was commonly used for many years until it was replaced by more advanced , complicated but much more precise non - lte radiative transfer techniques with velocity field included .
line profiles observed during solar flare . the black arrow indicates the peak of asymmetric lines .
the bottom profile represents the emission of the quiet - sun area ( @xcite ) .
, width=453 ] one of the first interpretation of chromospheric line asymmetry observed during flare was presented by @xcite .
the authors postulated that these asymmetries are due to chromospheric evaporation driven by accelerated electrons or thermal conduction .
@xcite was probably the first who realise that chromospheric plasma heated during solar flare may evaporate .
this evaporated plasma provides material for loop prominences often observed as the so - called post - flare loops ( @xcite ; @xcite ) . for more complete review of chromospheric evaporation
see hudson paper in this book .
@xcite postulated that in the analysed flare the non - thermal electrons heat the chromosphere mainly during the impulsive phase , while thermal conduction from the hot coronal plasma heated earlier dominates during the late , thermal phase .
both mechanisms drive upflow of the cool plasma .
the authors stress that for the first time they observed chromospheric evaporation in h@xmath2 line .
far wings , while crosses to the values obtained from the shift of the line peak .
open circles represent the time evolution of the h@xmath2 intensity . the microwave emission at 3750 mhz is also shown ( @xcite ) .
, width=226 ] contradictory results were published by @xcite who suggested that during the impulsive phase of solar flares the downflow of the cool chromospheric plasma is present .
these results are based on the large red asymmetry of the h@xmath2 line observed during the impulsive phase of many solar flares ( fig .
[ ab - fig : ichim - prof ] ) .
the downward motion increases at the onset of a flare to its maximum velocity of shortly before the impulsive peak of microwave emission , and rapidly decreases before the h@xmath2 reaches its maximum ( fig .
[ ab - fig : veloc - evol ] ) .
the red asymmetry of the h@xmath2 line may be also explained by the attenuation of the blue wing by the rising plasma over the flare but the authors exclude this case because the optical thickness of a rising cloud is too small to explain the emission deficit of the blue wing of h@xmath2 line .
also the high temporal resolution spectroscopic observations of h@xmath2 line performed by @xcite confirm the existence of red asymmetry during the impulsive phase of solar flares .
the largest asymmetry is observed during the maximum of microwave emission what confirms the earlier results that the plasma downflow is driven by the accelerated non - thermal electrons .
kms@xmath7 ) in the high temperature region is obtained ( explosive evaporation ) and downflow ( up to @xmath8kms@xmath7 ) is observed in the chromosphere ( chromospheric condensations ) . for more details see the paper of @xcite .
, width=377 ]
the downflow of cool plasma in the form of chromospheric condensations observed during solar flares was predicted theoretically by @xcite .
if a region of the chromosphere heated by non - thermal electrons is thick enough , then the rapid temperature increase produces an enhanced pressure in the heated region .
this overpressure , besides the evaporation , also drives downward moving cool and dense chromospheric condensations ( @xcite ) which seem to be responsible for red asymmetry of the h@xmath2 line profiles reported by many authors .
@xcite modeled the hydrodynamic and radiative response of the atmosphere to short impulsive injections of non - thermal electron beams ( fig .
[ ab - fig : fisher-1 ] ) .
they showed that a high - energy flux of non - thermal electrons drives explosive evaporation accompanied by the formation of cool chromospheric condensations in the flare chromosphere .
a different situation occurs when the flux associated with non - thermal electrons is very low .
then only a weak chromospheric evaporation takes place .
this kind of evaporation is referred to as gentle evaporation ( @xcite ; @xcite ) and it can be observed in chromospheric spectral lines like h@xmath2 or in caii 8542 .
@xcite suggested that the gentle chromospheric evaporation may also occur after the primary energy release , when the non - thermal electron flux is stopped .
this evaporation could be driven by the large conductive heat flux from a high temperature flare plasma contained in magnetic tubes above the chromosphere .
such physical conditions may appear during the gradual phase of solar flares , when there is no significant flux of non - thermal electrons . in the @xcite model for flare - loop formation by magnetic reconnection
the conduction of the thermal energy generated at the slow - mode shocks drives a gentle evaporative upflow from the ribbons .
@xcite observed small but long - lasting blue - shifts in flare ribbons in the h@xmath2 line during the gradual phase of three solar flares and interpreted them as due to upflows with velocities less than 10kms@xmath7 ( fig .
[ ab - fig : schmieder - gentle ] ) .
these upflows were believed to be caused by gentle chromospheric evaporation driven by the heat conduction along the field lines connecting the chromosphere with a reconnection site in the corona .
line profiles observed during the gradual phase of a solar flare ( solid lines ) . weak blue - shift of these lines suggests slight upflow of the plasma with the velocity of a few kms@xmath7 interpreted as gentle evaporation .
dashed lines represent the reference line profiles of the quiet sun area ( @xcite ) .
, width=302 ] the downflow of cool chromospheric plasma during the impulsive phase of solar flares predicted in the theoretical calculations was reported by many authors .
@xcite observed large red asymmetry of the h@xmath2 line during the period of hard x - ray burst ( fig .
[ ab - fig : zarro-1 ] ) .
these asymmetries were used to determine the downward velocities estimated from the maximum shift of the centroid of the bisectors .
the averaged over all red - shifted pixels during the impulsive phase velocity was of the order of @xmath9kms@xmath7 .
the downflow analysed in the h@xmath2 data and the upflow observed in the x - ray lines allows the authors to analyze the momentum balance of the flow .
they conclude that the momenta of upflowing and downflowing plasma are approximately equal .
@xcite conclude that the downflow velocity measured from red wing enhancement can be used as a diagnostics of impulsive solar flare heating conditions ( fig .
[ ab - fig : zarro-2 ] ) .
the work of @xcite shows that the velocity of chromospheric downflows deduced from the red asymmetry of h@xmath2 line is around 30
40kms@xmath7 with the lifetime of the order of 2 3 minutes .
there are two major problems found by the authors : why is the line center nearly not shifted while the line wing shows great asymmetries ?
the second problem concerns the life time of the downflow which is considerably longer than the life time predicted in @xcite simulations .
recent hydrodynamic and radiative transfer simulations may now explain these two problems .
line profiles observed during the impulsive phase of a solar flare . the red wing enhancement is observed at the same time as strong hard x - ray emission .
note that the central part of the line is slightly blue - shifted ( @xcite ) .
, width=94 ] there are more papers presenting the observations of the red asymmetry related to the chromospheric condensations driven during the impulsive phase of solar flares ( e.g. , @xcite fig .
[ ab - fig : wulser ] ) .
all of them determine more or less consistent observational picture of the chromospheric flows during the impulsive phase of flares .
this picture is based on many spectroscopic observations of the chromospheric line profiles .
an important step was done when it became possible to calculate theoretical line profiles and compare them with observations .
line profiles with the red asymmetry ( bottom ) observed during the impulsive phase of a solar flare .
dotted lines mark the reference quiet sun profile ( @xcite ) .
, title="fig:",width=340 ] + line profiles with the red asymmetry ( bottom ) observed during the impulsive phase of a solar flare .
dotted lines mark the reference quiet sun profile ( @xcite ) .
, title="fig:",width=151 ] @xcite computed time - dependent h@xmath2 line profiles for the dynamic model atmosphere of @xcite .
they simulate the effects of power - law electron beam heated chromosphere .
solving the radiative transfer equations for one - dimensional model atmosphere the evolution of h@xmath2 line profile was estimated .
the time of the electron beam heating was 5 s and for detailed description of other parameters and computational methods see @xcite . in fig .
[ ab - fig : canf - gal ] the time sequence of the h@xmath2 line profile is presented . during the non - thermal heating ,
the red - shifted component is present but after the heating was stopped , the h@xmath2 line exhibit the blue asymmetry although the central absorption feature is shifted towards longer wavelengths .
this behaviour is explained by downflow of the chromospheric condensation .
it is also worth to notice that the response of the h@xmath2 emission to the non - thermal electron beam is very fast ( less than second ) .
line profiles for the electron beam heated model atmosphere of @xcite ( @xcite ) .
, width=377 ] similar but more precise simulations of the dynamics and radiation in a solar flare loop was presented by @xcite . except the non - thermal heating of the chromosphere , they took into account the thermal heating by the soft x - ray irradiation within 1 250 range .
@xcite radiative - hydrodynamic code was used to analyze the response of the lower atmosphere at the footpoint of a flare loop . in the radiative transfer calculations
the important transitions of hydrogen , helium and singly ionized calcium and magnesium were treated in non - lte .
one - dimensional atmospheric model was used in the calculations . as a starting models the authors took two different cases pf1 and pf2
the temperature and electron density stratifications of both preflare atmospheres are shown in fig .
[ ab - fig : abbett - hawley-1 ] .
three levels of the non - thermal heating was considered which correspond to weak ( f9 ) , moderate ( f10 ) , and strong ( f11 ) non - thermal flare heating .
the pf1 atmosphere is heated for 70 s with the f9 and f10 fluxes , and the pf2 atmosphere is heated for a shorter , 4 s burst but with strong f11 heating .
figure [ ab - fig : abbett - hawley-2 ] presents the time evolution of emergent h@xmath2 and caiik line profiles . for the h@xmath2 line separated blue - shifted component is clearly visible while caii line exhibit red - shifted component
the contribution function calculated for these two lines explain why we observe such a two - components and asymmetric profiles ( fig .
[ ab - fig : abbett - hawley-3 ] ) .
this analysis shows that the evolution of non - thermally heated chromosphere progresses through two distinct dynamic phases ( @xcite ) : a gentle phase , where the non - thermal energy input of the flare is essentially radiated away into space , and an explosive phase , where the flare energy rapidly heats the atmosphere and drive large amounts of chromospheric material up into the corona , and down toward the photosphere . during the explosive phase ,
there is significant plasma motion and there are steep velocity gradients .
moreover , the effects of thermal x - ray heating of the chromosphere remain negligible compared to the non - thermal heating in the impulsive phase .
( right ) lines after 50 s of flare heating in model f10 .
line frequencies are in velocity units .
atmospheric velocities ( dashed lines ) are taken to be positive toward the corona .
thus , negative velocities associated with downward - moving material correspond to red - shifts in the profile ( @xcite ) . ,
width=453 ] similar , but much more extended calculations were presented by @xcite .
the basics of computational methods are similar to those described by @xcite but there are some significant improvements .
the authors include the double power - law electron beam energy distributions recently observed in solar flares with the reuven ramaty high - energy solar spectroscopic imager ( rhessi ) satellite .
additionally , the effects of xeuv heating from a large number of high - temperature lines was taken into account using results from the chianti and atomdb databases and a wide range 1
2500 was used for direct thermal heating of the chromosphere .
figure [ ab - fig : allred-1 ] present the comparison of the heating rates used by @xcite and @xcite .
these new calculations confirmed the previous results .
however , the line profiles evolution differs from the @xcite calculations
the line asymmetry is not so significant and the blue and red components of the h@xmath2 and caiik lines , respectively , are not observed separately . instead
, the lines are asymmetric with blue or red wing enhancement ( fig .
[ ab - fig : allred-2 ] ) . as in @xcite
, the authors found that the impulsive flare naturally divides into two phases , an initial gentle phase followed by a period of explosive increases in temperature , pressure and velocity .
, h@xmath2 , heii 304 , and caiik lines for the moderate level of non - thermal heating ( model f10 ) .
times are indicated at the top of each column .
the dotted lines indicate the level of the continuum close to the line center , while the dashed line is the preflare line profile ( @xcite ) .
, width=453 ] there are more papers which treat the problem of plasma flows in the non - thermally heated chromosphere ( @xcite ; @xcite ; ; @xcite ; ) .
all these hydrodynamic simulations predict an upflow of the hot coronal plasma due to the enhanced pressure in the region heated by non - thermal electrons or protons .
this upflow is associated with downflow of chromospheric condensations , but with much lower velocities ( fig .
[ ab - fig : mariska - emslie-1 ] ) .
these condensations disturb the line profiles emitted from the chromosphere and cause significant asymmetries observed e.g. , in h@xmath2 line .
the calculated line profiles emerging from flaring atmosphere show roughly similar behaviour than the observed ones but unfortunately , the appearance and the time evolution of the calculated line profiles was not compared to the spectroscopic observations of particular flares .
the validity of the modeling summarized in this chapter was not confirmed observationally up to now .
asymmetries observed in chromospheric line profiles are also modeled using semiempirical flare models . this approach is based on the idea that the temperature stratification of the atmosphere is determined empirically in the way to reproduce the calculated spectrum in the best agreement with observations .
this means that the energy - balance equation is not considered .
many spectral lines and continua are used to construct such kind of models .
initially , static semiempirical models were developed for a quiet - sun atmosphere ( @xcite ) and then for the static flaring atmosphere ( @xcite ) .
these models were constructed under assumption of 1-dimensional geometry and hydrostatic equilibrium ( fig . [ ab - fig : mavn ] ) .
they are static however the time sequence of many semiempirical models can be used to describe the evolving atmosphere but this method is valid only for slowly - evolving atmospheres
. more detailed description of semiempirical models can be found in the paper of mauas in this book .
non - lte radiative transfer methods applied to semiempirical models allows us to calculate the spectrum emerging from the atmosphere .
in particular , the profiles of the chromospheric optically thick lines may be calculated in details ( @xcite ) .
it is also possible to reproduce asymmetric line profiles solving the transfer equation for a dynamic atmosphere with velocity field , using previously calculated level populations for static model ( ) .
however , this approach is justified only for relatively small velocities ( @xmath10 ) which do not significantly affect the level populations of the static model ( ) .
it can not be used for impulsive phase of flares to model the chromospheric condensations which move quite fast .
therefore , such simplified calculations are used to model e.g. , the gradual phase of solar flares , when the velocities in the chromosphere are low .
one of the first semiempirical modeling of chromospheric flows was performed by @xcite . using the h@xmath2 line profiles observed for two flares the authors constructed the series of semiempirical models with chromospheric condensations .
it was shown that chromospheric condensations are responsible not only for the red asymmetry of the h@xmath2 line , but also for the blue asymmetry of the line with central reversal ( fig .
[ ab - fig : gan-1 ] ) .
chromospheric condensations were assumed to be homogeneous with constant velocity .
the most important conclusion of this paper is that the properties of chromospheric condensations seem to be consistent with the results of hydrodynamical models of solar flares .
comparison of calculated h@xmath2 line profiles with real observations present also a valuable part of this paper .
an interesting work was presented by who simulate the influence of the velocity field on the h@xmath2 line profiles .
the calculations were performed using non - lte model of plane - parallel solar flare atmosphere with stationary velocity field .
this velocity field was applied to different layers of the solar atmosphere and the emergent h@xmath2 line profiles were calculated for two models of solar flare f1 and f2 ( @xcite ) .
figure [ ab - fig : nejezchleba-1 ] present an example of asymmetric line profiles calculated for a weak - flare model atmosphere f1 and for different velocity fields .
the main conclusion of this work is that the velocity field affects the level populations via the increase of the downward radiation .
nevertheless , for velocities that do not exceed the thermal velocity of plasma , one can use the static populations for the formal solution of radiative transfer equation including the velocity to reproduce the observed line asymmetries .
other important point is that application of the bisector method would lead in some cases to reverse velocity , in others to underestimation of the velocity .
it includes , besides the part of the profile directly affected by the moving material , also a `` static '' part of the profile . to use the bisector in terms of doppler - shift
the static part should be somehow eliminated .
this remark makes questionable all estimations of the doppler velocity obtained with the bisector method applied to self - reversed or emission chromospheric lines observed in solar flares .
line profiles calculated for a static and dynamic atmosphere with different velocities 10 , 30 , and 50kms@xmath7 ( ) .
, title="fig:",width=188 ] line profiles calculated for a static and dynamic atmosphere with different velocities 10 , 30 , and 50kms@xmath7 ( ) . ,
title="fig:",width=226 ] the conclusions of suggest that deducing the velocity from flare line profiles is rather difficult and can not be done only by searching for doppler - shifts with the bisector method .
chromospheric line profiles suggest that the flare atmosphere is highly dynamic and stratified with rather complicated plasma motion .
these lines are optically `` thick '' and the only reliable way to analyze the flows is to use the non - lte radiative transfer codes , which enable us to compute the chromospheric models with velocity fields . resulting synthetic line profiles can then be compared with the observed ones .
line profiles and determined velocity field across the atmosphere .
negative value indicates an upward velocity . in the lower panel
the time evolution of the velocity value at two different heights in the atmosphere is presented .
filled circles refer to the height @xmath11 900 and open circles to the height @xmath12 km ( ) .
, width=302 ] the direct comparison of the observed and synthetic line profiles was presented in the paper of .
they study the chromospheric structure of a small flare and construct 5 semiempirical models for different times , which reproduce the profiles of the h@xmath4 , caiik , and sii 3905 lines during the flare evolution . in order to reproduce the asymmetry of the lines
the velocity fields were introduced in the line profile calculations .
the modeling was done using the non - lte pandora code of @xcite .
the trial - and - error method was used to reproduce the observed line profiles by the synthetic ones .
figure [ ab - fig : falchi - mauas-1 ] ( upper panels ) present an example of the observed and fitted h@xmath4 line profiles and determined velocity field across the atmosphere . in the lower panel of fig .
[ ab - fig : falchi - mauas-1 ] the time evolution of the velocity value at two different heights in the atmosphere is presented .
the presence of an upward motion in the flaring atmosphere at 1400 km , might be a signature of the chromospheric evaporation observed at chromospheric levels .
it is interesting to notice that around 13:44:30 ut the downflow is observed at the height of 900 km , while the upward motion is evident at 1400 km above the photosphere .
one possible explanation is that chromospheric evaporation together with condensations is observed . in this case
chromospheric evaporation is observed at chromospheric levels and not , as more common , at coronal levels .
line profile emitted from flaring model atmosphere f1 of machado et al .
( 1980 ) . in the middle panel
the velocity used in the modeling as a function of temperature is shown ( positive sign of the velocity indicates downflow ) .
also an example of the observed h@xmath2 line with red asymmetry is presented ( for details see @xcite ) .
, width=453 ] using the semiempirical f1 model of a weak flare ( @xcite ) , @xcite showed that the blue asymmetry of h@xmath2 line profile is observed due to the downflow of chromospheric plasma .
it is interesting to notice that the blue asymmetry is associated with the red - shift of the central absorption feature .
similar results was shown by @xcite .
the structure of the velocity field used in the non - lte simulations of @xcite was qualitatively consistent with the concept of downward - moving chromospheric condensations ( fig .
[ ab - fig : heinzel-1 ] ) .
these calculations were performed using the non - lte code developed by and modified for flare conditions .
the code uses a 1d plane - parallel geometry and the atmosphere is in hydrostatic equilibrium .
hydrogen excitation and ionization equilibrium have been computed by solving simultaneously the radiative transfer equations , the equations of statistical equilibrium for a five - level plus continuum atomic model of hydrogen and the equations of particle and charge conservation .
the equations of statistical equilibrium have been preconditioned according to .
the preconditioning is based on the lambda - operator splitting technique , where the exact lambda operator is expressed as an approximate operator plus the correction .
then the correction is iteratively applied to a lagged source function by using the so - called accelerated lambda iterations ( ali ) method . for multilevel atoms
this method is referred to as mali multilevel accelerated lambda iterations ( ) .
the preconditioned equations are then linearized with respect to the atomic level populations and electron density and solved iteratively ( ) .
this non - lte code was also used in to analyse the time evolution of the line asymmetry observed during the gradual phase of the solar flare on october 22 , 2002 . in this paper for the first time the evaporative flows in the gradual phase are studied quantitatively by using a non - lte radiative transfer code and spectroscopic observations of the flare ribbons .
first the authors analyse the influence of different velocity fields on the emergent h@xmath2 line profile .
again , it was shown that the downflow of flaring plasma causes blue asymmetry of the self - reversed line while upflow red - asymmetry ( fig .
[ ab - fig : berlicki-1 ] ) . for the modeling of the observed line asymmetry except the changes of the value of the velocity , also the height of the velocity field in
the atmosphere was different .
the procedure of fitting the h@xmath2 line profiles was performed using a grid of many models by varying different parameters .
each observed profile was fitted by the least - square technique to a closest synthetic profile from the grid and the model with the velocity field was found for each analysed line profile .
line - centre optical depth distribution of the velocity field used in the modeling plotted for different values of @xmath13 and for @xmath14 ( upper left panel ) and for different values of @xmath15 and for @xmath16 ( upper right panel ) . in the lower panels
the influence of the velocity field on h@xmath2 line profiles emitted from flare is presented ( upflow defined by @xmath17 and downflow @xmath18 for three values of @xmath19 , and 10.0 ( dotted , dashed and continuous lines , respectively ) ( ) . , title="fig:",width=359 ] line - centre optical depth distribution of the velocity field used in the modeling plotted for different values of @xmath13 and for @xmath14 ( upper left panel ) and for different values of @xmath15 and for @xmath16 ( upper right panel ) . in the lower panels
the influence of the velocity field on h@xmath2 line profiles emitted from flare is presented ( upflow defined by @xmath17 and downflow @xmath18 for three values of @xmath19 , and 10.0 ( dotted , dashed and continuous lines , respectively ) ( ) .
, title="fig:",width=377 ] in the analysis the msdp ( multichannel subtractive double pass ) spectrograph ( mein 1991 ) coupled to the vtt telescope working at the teide observatory ( tenerife , canary islands ) was used .
36 h@xmath2 line profiles ( six areas at six different times ) observed during the m1.0 flare on october 22 , 2002 were taken for the analysis ( fig .
[ ab - fig : berlicki-2 ] upper left panel ) .
as an example we present in fig .
[ ab - fig : berlicki-2 ] ( lower panel ) some profiles observed in chosen area at three times .
these observed profiles ( solid lines ) are fitted with the synthetic ones ( dashed lines ) obtained from the grid . in the right panel of fig .
[ ab - fig : berlicki-2 ] the temporal evolution of the velocity field is presented for two different areas of the flare .
line profiles ( upper and lower left panels ) .
0 5 are the areas used in the analysis .
the time evolution of the velocity in the chromosphere deduced from line asymmetries is plotted for two analysed areas in the right panel .
negative velocities correspond to upflow ( ) .
, width=415 ] the authors interpret the upflows found in the flare ribbons in terms of the @xcite model for gentle evaporation .
this process may occur during the gradual phase of solar flares and it can be driven by conductive heat flux from the high - temperature flare plasma contained in magnetic flux tubes above the photosphere . in the future
it would be interesting to use more spatial points at more times and to use the spectra obtained within a wider range of wavelengths .
other distributions of the velocity field in the chromosphere should also be tested .
in addition , to perform non - lte modeling of the flare structure it would be useful to have other spectral lines formed at different levels of the chromosphere .
in this review i presented some interesting papers concerning plasma flows observed during solar flares in cool chromospheric layers .
these flows are directly responsible for the line - asymmetries and/or line - shifts often observed in chromospheric lines emitted by the flaring plasma . an important work was done to understand the flows and their mechanisms . in order to determine the plasma velocity and flow direction the bisector method was applied for line profiles .
unfortunately , as we could see , this method leads in some cases to misleading estimations of the velocity . recently
, the direct comparison of the observed and synthetic line profiles gives more valuable information about velocity fields in the chromosphere .
all the data support the evaporative model of solar flares where explosive chromospheric evaporation of the hot plasma is associated with the chromospheric condensations observed in `` cool '' chromospheric lines . in the late phases of flares the gentle evaporation
may be observed in chromospheric lines . for the future it is necessary to use large and dense grids of the chromospheric models computed with hydrodynamic and non - lte codes .
they may help us to understand the flows and give more realistic description of the physical processes during the flares , particularly the heating mechanisms and their role at different phases of the flare evolution .
finally , really good spectral observations of flares are needed .
they have to be co - spatial , simultaneous and obtained in different spectral ranges ( x - ray , euv , uv , optical , ir ) .
such observations would be very helpful to construct the full picture of the plasma flows during flares .
there are some data concerning the flows observed in soft x - ray and euv but they are extremely rare and almost never cospatial nor simultaneous with the observations in chromospheric lines .
+ this research was supported by the european commission through the rtn programme ( european solar magnetism network , contract hprn - ct-2002 - 00313 .
the author also would like to thank p. heinzel for helpful comments and valuable remarks . | for many years various asymmetrical profiles of different spectral lines emitted from solar flares have been frequently observed . these asymmetries or line shifts
are caused predominantly by vertical mass motions in flaring layers and they provide a good diagnostics for plasma flows during solar flares .
there are many controversial results of observations and theoretical analysis of plasma flows in solar chromospheric flares .
the main difficulty is the interpretation of line shifts or asymmetries . for many years
, methods based on bisector techniques were used but they give a reliable results only for some specific conditions and in most cases can not be applied .
the most promising approach is to use the non - lte techniques applied for flaring atmosphere .
the calculation of synthetic line profiles is performed with the radiative transfer techniques and the assumed physical conditions correspond to flaring atmosphere .
i will present an overview of different observations and interpretations of line asymmetries in chromospheric flares .
i will explain what we have learnt about the chromospheric evaporation in the frame of hydrodynamical models as well as reconnection models .
a critical review will be done on the classical methods used to derive doppler - shifts for optically thick chomospheric lines . in particular , details on the new approach for interpreting chromospheric line asymmetries based on the non - lte techniques will be presented . |
the dynamical processes by which atoms lose their electrons when subjected to strong laser pulses usually fall into two main categories @xcite : sequential double ionization ( sdi ) in which electrons are removed one after the other , and a less trivial class of processes coined nonsequential double ionization ( nsdi ) . in nsdi processes , the coulomb interaction between two electrons plays an essential role .
a particularly relevant example is afforded by the recollision scenario in nsdi @xcite in which a preionized electron returns to the ion core to excite a bound electron .
this excitation may lead to a double ionization if the energy transfer from the preionized electron to the bound electron is sufficient to overcome the coulomb attraction to the nucleus , while the preionized electron keeps enough kinetic energy to remain ionized .
this so - called `` direct impact ionization '' @xcite is often invoked to explain the additional amount of double ionization at intermediate values of laser intensities .
however the dynamics exhibits a much richer variety of double ionization processes .
it has been observed that , sometimes , the recollision excites the parent ion which is ionized by the field after some delay , ranging from a quarter up to several laser cycles .
these processes mimic a sequential double ionization process even though they fall into the nsdi category since the electron - electron interaction plays a role .
this alternative and less straightforward road to nsdi is called recollision excitation with subsequent ionization ( or resi for short @xcite ) .
these processes are by no means rare and unimportant . for instance
, it was shown that resi plays a central role in high harmonic generation @xcite which is a way to generate short x - ray pulses .
resi processes belong to a category of double ionization involving a delay in the second ionization , and this category is referred to as delayed double ionization .
the precise mechanism responsible for delayed double ionization , is still a subject of study from the theoretical and experimental points of view @xcite .
experimentally , the signature of resi can be found in momentum distributions : the two lateral peaks in the bimodal distribution are attributed to direct impact ionization whereas resi contributes to the central part around zero ion momentum @xcite .
it also turns out that direct impact ionization is expected to be triggered when the electric field is close to zero whereas resi is expected when the electric field is maximum .
there is still some debate on the mechanisms of delayed double ionization @xcite .
the conventional scenario for resi is based on a tunneling argument in the adiabatic approximation .
this scenario is depicted in fig .
[ fig : resimechanism ] : after a recollision at a zero of the field , the remaining ion is in an excited state but not yet ionized .
when the laser reaches an extremum , it creates a potential barrier through which the electron can tunnel .
therefore it raises a natural question if resi is a purely quantum phenomenon . to counter this argument ,
resi has been observed with classical calculations @xcite , for which all ionizations are over - the - barrier , casting some doubts on the validity of the tunneling argument in the adiabatic approximation . in this article
we revisit this conventional picture of the resi mechanism by analyzing the phase space mechanisms behind delayed double ionizations , of which resi is a prime example . we show that after an electron is promoted to a specific region of phase space , it is trapped for some time before ionizing .
this specific region is a chaotic one resulting from the competition ( and resonances ) of the coulomb attraction with the laser field . at a given laser intensity , this thin chaotic region
is organized by a single resonance through the hyperbolic periodic orbit it creates .
the chaotic nature of resi trajectories was already identified in ref .
@xcite through the `` deflection function '' .
the way an electron is ionized in the chaotic region is by following the unstable manifold of some selected periodic orbits . as a consequence ,
the properties of these periodic orbits ( as a function of the laser intensity ) drive the delayed double ionization yields . as a signature of these periodic orbits
, we find oscillations in the resi to double ionization yields as a function of the laser intensity in a range of laser frequency .
some of these results were announced in a recent letter @xcite . in this paper , we restrict our analysis to linearly polarized fields .
interesting experimental and early theoretical results suggest the importance of delayed double ionization with elliptic or circular polarization as well @xcite . in sec .
[ sec : cm ] , we describe the classical models we use for the analysis of delayed double ionization .
we exhibit some relevant reduced models , and show that they are pivotal for the identification of the delayed double ionization mechanism . in sec .
[ sec : ddi ] , we present some peculiar features of resi trajectories in phase space . using the reduced models of sec .
[ sec : cm ] , we provide a detailed analysis of the mechanisms in phase space .
we show the importance of the dimensionality of the reduced models .
remarkably , classical mechanics is able to identify both sequential and nonsequential pathways leading to double ionization and reproduces the experimental and computational observations closely @xcite .
this success was ascribed to the paramount role of the electron - electron interaction which is present both in quantum and classical models .
one of the advantages of classical mechanics is to analyze these systems in phase space in a very detailed way . in refs . @xcite it was shown how invariant structures like periodic orbits rule double ionization processes in strong and short laser pulses .
the methodology here is to follow the same modus operandi in order to provide insights into delayed double ionization . in this section
we detail the various models which will be used later in the article .
the parent model is a two active electron atom with soft - coulomb potentials subjected to an intense and short linearly polarized laser pulse in the dipole approximation @xcite .
the corresponding hamiltonian reads : @xmath0 where @xmath1 is the position of the @xmath2 electron ( @xmath3 ) and @xmath4 its canonically conjugate momentum . here
@xmath5 and @xmath6 belong to @xmath7 for @xmath8 , 2 or 3 .
the nucleus is assumed to be fixed at the origin because of its large mass compared to that of the electrons and the relative short duration of the laser pulse .
the linearly polarized laser field ( along the direction @xmath9 ) is characterized by its amplitude @xmath10 and has a wavelength of 780 ( @xmath11 ) or 460 nm ( @xmath12 ) which is specified for each computation .
its envelope @xmath13 is composed of a two laser cycle linear ramp up and a six laser cycle constant plateau .
the constants @xmath14 and @xmath15 are the electron - nucleus and electron - electron softening parameters respectively and are chosen so as to be compatible with the ground state energy @xmath16 , defined as the sum of the first two ionization potentials @xcite .
the same softening parameters are used independently of the dimensionality @xmath17 for a given atom .
the model ( [ eq : fullhamiltonian ] ) has two and a half degrees of freedom . in order to describe the dynamical processes in phase space
, it is desirable to consider reduced models .
these reduced models are described in what follows .
we consider several kinds of reduced models : dimensionally reduced ones , and single electron models , or a combination of both .
all of these models are used in a complementary way so as to get insights into the dynamical processes of delayed double ionizations . from the dynamics associated with hamiltonian ( [ eq : fullhamiltonian ] ) , we notice that any plane containing the nucleus and the polarization axis is dynamically invariant . more specifically , starting with the initial condition @xmath18 and @xmath19 , the two electrons remain in the @xmath20 plane .
it is then possible to consider a dimensionally reduced model where the @xmath21-direction ( and its canonical momentum ) is left out .
such models are generally referred to as @xmath22 dimensional models , making reference to the dimension of the configuration space in which each electron moves .
the @xmath22 dimensional model can further be reduced to a @xmath23 dimensional model by constraining each electron to move along a line which has a constant angle with the axis of polarization .
we notice that , contrary to the reduction from @xmath24 to @xmath22 dimensions which can be carried out independently of the ellipticity of the field , the one from @xmath22 to @xmath23 can be considered only for linear polarization .
the resulting constrained dynamics is obtained using , e.g. , dirac s theory of constrained hamiltonian systems @xcite and the resulting hamiltonian reads @xmath25 where @xmath26 is the angle with the axis of polarization ( see fig .
[ fig:1dmodels ] ) .
two special cases have been studied in the literature .
they correspond to @xmath27 , where the electrons are aligned with the axis of polarization @xcite , and @xmath28 @xcite . in the first case ,
the reduced hamiltonian reads : @xmath29 we note that the reduced dynamics corresponds to a symmetry of original @xmath22 model . in the second case
, the reduced hamiltonian reads : @xmath30 where @xmath31 is the scaled peak field amplitude . here
the lines along which each electron is moving are not conserved by the flow of hamiltonian ( [ eq : fullhamiltonian ] ) .
the main reason why hamiltonian ( [ eq : eckh : ham2e ] ) has been considered in the literature is to allow the escape of electrons simultaneously with symmetrical momenta . in the numerical calculations , we use the two models independently and
show that the dynamical mechanisms that regulate delayed double ionization are qualitatively the same for both @xmath23 dimensional models .
although the @xmath23 dimensional reduction may seem as a crude approximation , it has shown to be surprisingly accurate in reproducing quantum and experimental results in double ionization qualitatively , such as the characteristic knee shape in double ionization yield as a function of the laser intensity .
this is the simplest model looking at the correlated dynamics of two electrons . in this case , the role of the softening parameters is to allow the charged particles to pass through each other along the polarization axis , as they would have done in a transverse direction with a higher dimensional model .
although the two electrons are treated equally in the previous models , a careful analysis of the dynamics reveals that , given some initial conditions , their dynamics are qualitatively very different : typically one electron is quickly ionized ( the outer electron ) while the other one remains close to the core ( the inner electron ) @xcite .
this observation stands independently of the chosen dimension , and it is at the core of the three step scenario @xcite . as a result , for each electron an effective reduced model can be crafted . when this outer electron is far away from the core the electron - electron interaction can be neglected .
therefore the effective hamiltonian for the inner electron is built from hamiltonians ( [ eq : fullhamiltonian ] ) , ( [ eq : ham2e ] ) and ( [ eq : eckh : ham2e ] ) : @xmath32 where @xmath33 is the scaled laser amplitude in @xmath23 dimensional models ( [ eq : ham2e ] ) and ( [ eq : eckh : ham2e ] ) , and @xmath34 for higher dimensional models .
we notice that , up to a scaling of the field amplitude , the same models for the inner electron stand independently of the chosen angle @xmath35 . to simplify the comparison between the models , in all numerical results we drop the laser amplitude scaling parameter ( i.e. , we choose @xmath34 ) .
this model for the inner electron is only valid after the outer electron has left the nucleus ( i.e. , after the preionization at @xmath36 such that mod@xmath37 and outside of the recollision events ) . for convenience , we shift time by @xmath38 so that the inner electron model is valid for @xmath39 .
since most of the nonsequential double ionization processes happen in the plateau @xcite , the model for the inner electron is further simplified as @xmath40 where @xmath41 denotes the laser phase at which the inner electron model is started ( e.g. , after the preionization @xmath42 or after the end of the final recollision ) . in
what follows , we use these reduced models to highlight the relevant mechanisms at play for the delayed double ionization .
first we begin by analyzing some sample trajectories of the two electron hamiltonian ( [ eq : fullhamiltonian ] ) . for the display ,
we only consider the case @xmath43 .
other dimensional models lead to a similar typology of trajectories and the main differences between the various dimensional models are analyzed in sec .
[ sec : dim ] . among all the two - electron trajectories , we consider the ones which are doubly ionized at the end of the laser pulse , and which present a significant delay between the last time the preionized electron influences the core dynamics and the subsequent ionization .
these trajectories will be referred to as delayed double ionizations .
we use an energy criterion to identify if an electron has ionized or not @xcite .
the energy @xmath44 of an electron is defined as the sum of its kinetic energy plus the ( soft ) coulomb interaction with the nucleus : @xmath45 an electron is considered ionized whenever its energy is positive , i.e. , @xmath46 . as for the electron - electron interaction , there are many ways a preionized electron influences the dynamics of the inner electron .
the most famous one in strong field physics is the recollision .
a recollision is said to have occurred whenever the distance between the two electrons is smaller ( or equivalently , whenever the coulomb interaction between the two electrons is larger ) than some threshold .
of course the definition of the threshold is somewhat arbitrary .
however on inspection of a large sample of trajectories , it leads to the definition of a recollision where there is a significant exchange of energy between the two electrons .
we identify three main types of delayed double ionizations ( to be specified later on ) and give an illustration of them in fig .
[ fig : traj ] , and compare them to the standard direct impact mechanism ( top left panel ) .
the common feature of all the delayed double ionizations shown in fig . [ fig : traj ] is that the inner electron s role is very far from being a passive one , waiting for a recollision to ionize as in the standard picture of nsdi . in what follows
we show that the inner electron is promoted to a specific chaotic region of phase space where it is trapped for some time ( which might last up to several laser cycles ) .
the way the inner electron is promoted to this region varies according to the specific trajectory and to the values of the laser parameters .
it is a combination of several mechanisms : recollision , soft interaction and chaotic diffusion .
the differences between these three types may seem very subtle at first glance .
however , by looking closely at the trajectory of the inner electron right before being hit by the outer electron , we notice that the transfer of energy is very low for the soft interaction ( lower right panel ) while it is significant for the recollision ( upper right panel ) . the chaotic diffusion ( lower left panel ) corresponds to the extreme case where there is no recollision during the plateau ( note that during the brief excursion of the firstly ionized electron around @xmath47 laser cycles , the electron is not ionized ) .
it means that the inner electron is promoted to the chaotic layer by the correlated dynamics of the two electrons during the ramp up , and no recollision is involved in the subsequent ionization ( leaving out the `` r '' or resi ) .
note that chaotic diffusion is not a marginal phenomenon among delayed ionization .
all three types of delayed double ionization share the common featured that the inner electron is trapped in a chaotic region of phase space for some time before it ionizes .
the difference between them is the way it reaches this region , i.e. , through a ( soft or strong ) recollision or the correlated dynamics of the two electrons during the ramp up of the field . among the three types , resi trajectories , where a recollision is involved before ionization ,
are the most famous one . in this case , the significant time delay is compared to direct impact double ionization , where the subsequent ionization immediately follows the recollision ( see the upper panels in fig . [
fig : traj ] ) .
this delay might last between a quarter up to several laser cycles .
as expected , a typical delayed double ionization is the result of a cooperation between the three mechanisms , so even though we have tried to classify these trajectories into three categories , the boundaries between them are not sharp at all . in order to get deeper insights into the chaotic region in which the inner electron is promoted , we consider poincar sections of resi trajectories of hamiltonian ( [ eq : fullhamiltonian ] ) with @xmath8 .
we have chosen resi because it constitutes a significant subset of delayed double ionization , and it is easier to detect numerically . in fig .
[ fig : real_resi ] , we display the positions of the electron which ionizes last at the successive maxima of the laser field after the last recollision . looking at fig .
[ fig : real_resi ] , we note that resi trajectories are restricted to specific parts of phase space . in particular , we note a central region , close to the nucleus where there is no record of resi .
this region is delimited by a thin layer where most of the inner electrons are promoted right after a recollision .
in addition , we observe swirling patterns that extend far away from the nucleus . after a recollision
, the inner electron is left with the nucleus and field interactions .
we use hamiltonian ( [ eq : he+ ] ) for a one - dimensional one - electron model . in the left panels of fig .
[ fig : psos_resi1 ] , we represent poincar sections ( stroboscopic plots at the maxima of the laser , i.e. , @xmath48 ) of all the delayed ionizations ( i.e. , ionizations which take more than one laser cycle ) . for clarity
we left out the two first points on the poincar section for each trajectory ( the missing points are located in the same part of phase space as the remaining ones , but they clutter the fine structures depicted on the figure ) . note the close similarity between the poincar section of trajectories of the two - electron model ( [ eq : ham2e ] ) in fig .
[ fig : real_resi ] and the ones of the one - electron model ( [ eq : he+ ] ) in fig .
[ fig : psos_resi1 ] ( upper left panel ) . in particular the swirling patterns in both figures are nearly identical . as shown in figs .
[ fig : real_resi ] and [ fig : psos_resi1 ] , resis are governed by the inner electron model ( [ eq : he+ ] ) after the last recollision . more specifically , for one - dimensional models they correspond to a thin chaotic layer resulting from a resonance between the coulomb attraction and the laser excitation .
since the one - dimensional one - electron model ( [ eq : he+ ] ) captures the essential features of delayed ionization , this model is studied in more details in the next section .
the motivation for studying one - dimensional models is twofold : first , the laser field drives the ionization dynamics along the polarization axis .
second , we find that this simplified dynamics forms the skeleton of higher - dimensional dynamics which , however , differs from the one - dimensional dynamics in significant ways , to be specified in sec .
[ sec : dim ] . in this section
, we analyze the inner electron model given by hamiltonian ( [ eq : he+ ] ) with only one spatial dimension .
it has one and a half degrees of freedom with no conserved quantity in the presence of a laser field ( @xmath49 ) .
chaos is expected to develop , but only locally in phase space , depending on the parameter values of the laser field . to get a rough idea of the dynamics of hamiltonian ( [ eq : he+ ] ) on a short time scale compared to the period of the laser field
, we consider a static field approximation where we freeze the laser field at peak field amplitude : @xmath50 here we analyze this model for @xmath8 .
it exhibits a stark saddle @xcite where the effective potential ( coulomb plus laser ) is locally maximum and thus acts as a physical barrier to ionization . in phase space ,
the saddle has zero momentum and its position @xmath51 is a solution of @xmath52 .
we denote @xmath53 its energy .
since the time dependence has been removed , the system has one degree of freedom and thus it is integrable . in phase space ,
the curve @xmath54 is a separatrix ( homoclinic connection ) between the nonionizing trajectories and the ionizing ones ( see the outermost curve in fig .
[ fig : staticfieldapproximation ] ) . taking into account the time - dependent potential , the dynamics of hamiltonian ( [ eq : he+ ] )
is slightly complicated by the possibility of chaotic trajectories near the saddle point .
in fact , from kam type arguments , it is possible to show the preservation of invariant tori ( from the integrable dynamics described by @xmath55 ) .
we give an illustration of those tori in fig .
[ fig : staticfieldapproximation ] . looking closer at the dynamics , we notice some discrepancies between the static approximation and the time dependent dynamics .
this is particularly the case close to the separatrix in fig .
[ fig : staticfieldapproximation ] .
the bound region of hamiltonian ( [ eq : he+ ] ) corresponds to a part of phase space , close to the nucleus , filled up with kam tori .
the outermost tori corresponds to a boundary torus [ outermost light gray ( magenta online ) curve in the inset of fig .
[ fig : staticfieldapproximation ] ] outside which trajectories might be ionizing .
this region is referred to as the unbound region , where the dynamics leads to ionization .
not all of the trajectories , though , ionize , as it can be seen for instance in a small elliptic island in the inset of fig .
[ fig : staticfieldapproximation ] .
we notice two main dynamical features of the unbound region : close to the boundary torus , we observe a thin chaotic layer [ outermost dots ( blue online ) in the inset of fig . [
fig : staticfieldapproximation ] ] . further away from the nucleus , the dynamics leads to rapid ionization ( outermost white area in the figure ) . in the bound region of the dynamics defined by hamiltonian ( [ eq : he+ ] ) with one dimension , not all tori survive from the integrable case without the laser field .
the weakest ones are the resonant tori .
they are characterized by @xmath56 , where @xmath57 is the free field rotational frequency of the electron around the nucleus and @xmath58 .
they are later referred to as @xmath59:@xmath60 resonances . according to the poincar - birkhoff theorem @xcite ,
when a resonant torus is broken at least two periodic orbits are created instead . from numerical simulations ,
we find two or four resonant periodic orbits depending on the intensity . for the softening parameters @xmath61 , based on the observation that the rotational frequency depends almost linearly with energy , an analytical approximation for the action has been obtained in ref .
the same formula can be generalized for other softening parameters , so that the approximate free field frequency @xmath57 reads @xmath62 where @xmath63 , @xmath64 and @xmath65 are given by @xmath66 the maximum position @xmath67 experienced by the electron is determined from @xmath68 . as a consequence we obtain the expression of the rotational frequency @xmath57 as a function of @xmath67 . in fig .
[ fig : freefielfreq ] we compare this prediction with the actual frequency , computed numerically .
note the very good agreement between the curves , even relatively far away from the nucleus .
equation ( [ eq : freefieldfrequency ] ) combined with the resonance condition allows us to narrow down the regions of phase space in which to look for resonant periodic orbits . for the simplicity of the discussion we set the laser phase @xmath69 , as most of our figures
are drawn at peak field amplitude , such that @xmath70 corresponds to the peak field . in this case , for all @xmath47:@xmath60 resonances ( the electron oscillates exactly @xmath60 times around the nucleus within one laser cycle ) , two resonant periodic orbits with zero initial momentum are present , one on each side of the nucleus . to predict their location ,
we use an adiabatic approximation : the total energy for the inner electron @xmath71 , in the bound region , is almost constant in time . as a consequence , we replace the energy @xmath72 with @xmath71 in eq .
( [ eq : freefieldfrequency ] ) and the resonance condition becomes @xmath73 where @xmath74 is the position on the periodic orbit with zero momentum .
note that the implicit equation @xmath75 has in general two solutions , one positive and the other negative .
they correspond to two periodic orbits with zero initial momentum located on each side of the nucleus . in fig .
[ fig : resonantpo ] we compare the results given by this prediction with the actual location of the resonant orbits .
the very good agreement with the prediction is worth mentioning .
for a given softening parameter @xmath14 , there is a finite number of primary resonances ( and thus a finite number of resonant 1:@xmath60 orbits ) : from eq .
( [ eq : resonantcondition ] ) we deduce that the number of resonances @xmath76 is equal to @xmath77 where @xmath78 is the floor function .
for example , for @xmath79 , we find @xmath80 primary resonances .
this number decreases as @xmath14 increases , i.e. , as larger or equivalently more loosely bound atoms are modeled . for a given resonance index @xmath60 ,
there are two periodic orbits with zero initial momentum , one on each side of the nucleus .
we denote @xmath81 the one located on the same side as the stark saddle ( see the inset of fig . [
fig : staticfieldapproximation ] ) and @xmath82 the other one . for the intensity range we have considered ,
@xmath83 is found to be always elliptic while the stability of @xmath84 depends on the order of the resonance . for odd @xmath60
, @xmath84 is hyperbolic for the whole range of intensities we have investigated . for even @xmath60 , @xmath84 is elliptic for intensities where it is in the bound region and , at higher intensities , after being released in the unbound region , it bifurcates into a hyperbolic periodic orbit .
this bifurcation comes with the creation of an additional elliptic periodic orbit , through period doubling .
for those resonances , in addition to the periodic orbits @xmath85 we find two additional hyperbolic resonant periodic orbits , which we denote @xmath86 .
they are connected through a symmetry of the system , such that we identify the two and restrict the analysis to only one of them ( see fig .
[ fig : psos_resi1 ] , the diamond in the upper right panel ) . as the laser intensity
is increased , invariant tori are broken and the boundary torus between the bound and unbound region shrinks .
this can also be seen in the static field approximation where this decrease is of the order of @xmath87 ( which can be seen from the position of the saddle as given at the beginning of this section ) . at some point ,
the resonant periodic orbits are released to the unbound region .
the dynamical influence of resonant periodic orbits depends strongly on their location and linear and nonlinear stability properties .
looking at fig .
[ fig : staticfieldapproximation ] ( inset ) , we note the elliptic island surrounding the @xmath47:@xmath88 resonant elliptic periodic orbit while the orbit is in the unbound region . compared to fig .
[ fig : psos_resi1 ] , we note that this elliptic region corresponds to a part of phase space where no delayed ionization is detected . in the same figure ,
the resonant orbit @xmath89 is hyperbolic . to analyze its nonlinear properties , we compute its stable and unstable manifolds and display the resulting pictures in fig . [
fig : psos_resi1 ] ( right panels ) .
the manifolds are the pathways by which the electrons approach or leave the chaotic region .
note the strong similarities of the unstable manifold of @xmath90 with the poincar sections of resi trajectories in fig .
[ fig : real_resi ] .
this similarity confirms the important role played by this unstable manifold ( associated with a one - electron model ) in the delayed double ionization process .
looking at fig .
[ fig : psos_resi1 ] , we see that the stable and unstable manifolds associated with @xmath91 develop in the vicinity of the bound region and branches extend deep into the unbound region .
additionally , the width of the chaotic region as given by those manifolds explains why the transition between the bound and unbound regions is so sharp @xcite .
a comparison between the left and right panels of fig . [ fig : psos_resi1 ] shows that the unstable manifold of @xmath91 regulates the dynamics of the resi trajectories .
even some of the very fine details of the unstable manifold are reproduced by this set of resi trajectories .
we note that some parts of the unstable manifold are missing : they correspond to the intersections between the stable and the unstable manifolds .
the overlap between the stable and unstable manifolds of @xmath91 forms a `` sticky '' region @xcite that traps trajectories for some time before ionizing . for deeper insight into the shape of the unstable manifold of @xmath92
, we represent a projection of this manifold in the plane @xmath93 together with a projection of @xmath92 ( black curve ) in fig .
[ fig : manifold ] .
it shows a central region near @xmath94 and two main branches which depart from this central region .
it is worth noticing that these branches are located near the extrema of the electric field , i.e. , when @xmath95 and @xmath96 .
since a delayed ionization trajectory follows this unstable manifold asymptotically , we see that this trajectory leaves the nucleus at times that correspond to the extrema of the electric field .
this is consistent with what is already observed for resi @xcite .
the unstable manifold has some other branches through which an electron can leave the nucleus and that do not correspond to extrema of the electric field .
however , the probability of those ionization channels is much smaller .
the discussion of the laser intensity considered in figs .
[ fig : psos_resi1 ] and [ fig : staticfieldapproximation ] can be generalized to arbitrary intensity . in this case , the hyperbolic resonant periodic orbit(s ) in the unbound region ( @xmath84 or @xmath97 depending on the resonance index @xmath60 ) dynamically organize the chaotic layer surrounding the bound region , through their stable and unstable manifolds .
the unbound region hosts many resonant orbits , associated with the accessible @xmath59:@xmath60 resonances .
however , because of their short period ( the same as the laser ) and the short duration of the trapping for resi ( at most a couple of laser cycles ) the primary resonances @xmath47:@xmath60 are the ones which organize the resi dynamics .
the other periodic orbits in the vicinity of the bound region influence only the fine details of the chaotic structure . as a signature of the influence of the @xmath47:@xmath60 resonances , we represent the normalized resi statistics with respect to the double ionization yield as a function of the laser intensity for two - electron hamiltonian ( [ eq : fullhamiltonian ] ) with @xmath8 ( lower curve with square markers in fig .
[ fig : resi_to_di_ratio ] ) .
we clearly see some oscillations as a function of the intensity . in order to correlate these oscillations with some features of the 1:@xmath60 resonances ,
we represent the value of the greene s residue @xcite of @xmath81 and @xmath86 as a function of the intensity .
these curves illustrate the strong correlation between the oscillations in the relative resi yields with the successive releases of the resonant orbits in the unbound region . in the adiabatic approximation ,
the one - electron model ( [ eq : hamadia ] ) is integrable for @xmath8 .
a priori , this is not the case for two and three dimensions . in fig .
[ fig:2dadiabaticps ] we display the projection of the poincar section for hamiltonian ( [ eq : hamadia ] ) with @xmath43 , corresponding to the adiabatic approximation of hamiltonian ( [ eq : he+ ] ) .
we see that the dynamics is a mixture of regular and chaotic trajectories , with a large chaotic sea [ light gray ( cyan online ) dots on the poincar section ] for energy values close to the stark saddle .
since the laser is time independent , the separatrix ( red curve ) acts as a barrier to ionization , and trajectories are confined inside the bound region ( on each horizontal plane ) .
compared to hamiltonian ( [ eq : he+ ] ) with @xmath8 , the source of chaotic trajectories is very different : for hamiltonian ( [ eq : hamadia ] ) with @xmath43 , it is the result of a nonlinear coupling between the two spatial dimensions , while for hamiltonian ( [ eq : he+ ] ) with @xmath8 , it is the result of the interaction between the particle and the time - dependent field .
it is expected that for the phase space of hamiltonian ( [ eq : he+ ] ) with @xmath43 these two sources of chaos both guide ionization . as a consequence ,
the dynamics allows for the possibility of an electron relatively deep inside the core region to diffuse chaotically and gain energy to be finally released after some delay .
this phenomenon corresponds to the mechanism we have referred to as chaotic diffusion when no collision or soft interaction is involved in the subsequent ionization .
compared to fig .
[ fig:2dadiabaticps ] , the size of the chaotic region is amplified by the fact that additional tori are broken under the time dependence of the field . as a consequence , it becomes harder to define a bound and unbound region for the inner electron ( see fig . [
fig : resi_mechanism ] ) , because of chaotic diffusion . as we did for the model ( [ eq : he+ ] ) with @xmath8
, the laser excitation can be considered as a perturbation of the free field dynamics for @xmath43 .
then , again , kam type arguments show the persistence of invariant tori in the vicinity of the core .
however , contrary to the one dimensional case , these tori are dimensionally too small to create barriers of transport in phase space and a chaotic diffusion can happen across them .
to aid this view , and in a first ( rough ) approximation , one can imagine all the light gray ( cyan online ) dot regions in fig .
[ fig:2dadiabaticps ] to be connected , such that by moving from one layer to the next it is possible to cruise from one chaotic region to another even though there is a torus in between the two .
more realistically , the manifolds of the resonant orbits , through the transverse direction(s ) to the the field , develop relatively deep inside the core region .
this allows for chaotic diffusion , which is drastically reduced for one dimensional models due to the existence of the boundary torus .
close to the nucleus however , because kam tori form an intricate network , and by continuity of the dynamics , the diffusion is extremely slow ( as stated in nekhoroshev s theorem @xcite ) . as a result , and since we consider very short laser pulses , in practice it is possible to identify a ( finite time ) bound and unbound regions with a chaotic layer surrounding the bound region . compared to the one dimensional case , this chaotic layer is more complicated , with branches developed deeper into the core region . in fig .
[ fig : resi_to_di_ratio ] we notice that the resi yields given by hamiltonian ( [ eq : fullhamiltonian ] ) are higher for @xmath43 than for @xmath8 .
we attribute these higher yields to the aforementioned chaotic dynamics coming from an increase of instability associated with additional resonances in higher dimensional models .
it underscores once more the pivotal role played by the inner electron dynamics [ hamiltonian ( [ eq : he+ ] ) ] in these delayed ionizations .
for two dimensional models a direct consequence of the unstable direction for the resonant periodic orbits in the direction transverse to the laser field is the smoothing of the oscillations in fig .
[ fig : resi_to_di_ratio ] for @xmath98 nm wavelength . increasing the wavelength to @xmath99
nm gives birth to more resonances with a more dense tangle of unstable manifolds so that the oscillations are completely washed out in two - dimensional models whether they are one- or two - electron models , as observed from our classical calculations in fig .
[ fig : oscillations780 ] ( upper panel ) .
a second consequence of the unstable direction in the transverse plane is the increase of delayed double ionization yields as can be seen in fig .
[ fig : oscillations780 ] . in the upper panel ,
the statistics for two dimensions ( diamonds ) is roughly one order of magnitude larger than it is for one dimension ( squares ) .
the same trend is confirmed with two electron models ( compare fig .
[ fig : oscillations780 ] , dots in the upper panel , with fig . [ fig : resi_to_di_ratio_atom ] , filled in dots in the middle panel ) with roughly one order of magnitude of difference between the one and two dimensional models .
looking at typical resi trajectories we notice statistically longer delays for two dimensional models than for one .
again , we attribute these longer delays to the enhanced chaotic dynamics with two dimensions .
going to higher dimensions the difference between two and three dimensional models is not as drastic as between one and two dimensions because the two dimensions perpendicular to the polarization direction are dynamically playing the same role due to the rotational symmetry of the system around the polarization direction . to illustrate the importance on the dynamics played by the direction transverse to the field
, we consider a two - dimensional model where the coulomb potential is modified in the @xmath100-direction .
we consider the following modification of hamiltonian ( [ eq : fullhamiltonian ] ) @xmath101 where we @xmath102 is the matrix @xmath103 tuning the parameter @xmath35 enables one to modify the stability of the relevant resonant periodic orbit in the transverse direction without changing its location and stability in the polarization direction . for numerical simulations , we set @xmath104 and display the result in fig . [
fig : oscillations780 ] ( lower panel ) .
for such a parameter , the resonant periodic orbits become stable in the transverse direction . as a consequence of the stabilization of the transverse dynamics ,
the oscillations in the relative resi yields are recovered ( both with one and two electrons ) .
note also that the relative resi yields are significantly reduced compared to hamiltonian ( [ eq : fullhamiltonian ] ) which has a more chaotic dynamics .
when changing the target atom , there has been shown an apparent strong sensitivity of resi yields to the considered atom @xcite . from the generic hamiltonian ( [ eq : fullhamiltonian ] ) and its associated reduced models ,
these different atoms are modeled by varying the ground state energy and adjusting the softening parameters .
then , the analysis carried out in the previous sections can be applied to the specific atom . in particular , we focus on helium ( @xmath105 ) , neon ( @xmath79 ) and argon ( @xmath106 ) and use the same electron - electron softening parameter for the three of them ( @xmath107 ) .
for instance , using eq .
( [ eq : nbresonances ] ) we find @xmath108 primary resonances for he , @xmath80 for ne and @xmath109 for ar so that the number of primary resonances @xmath47:@xmath60 decreases with @xmath14 as @xmath110 .
for all atoms , we find a similar organization of the delayed ionization dynamics by the resonant orbits ( see fig .
[ fig : comparepoincaresections ] ) .
previous studies have shown differences in the momentum spectra for various atoms @xcite .
our analysis suggests that this difference does not come from the phase space structures which are very similar for all atoms . to investigate the specific impact of the resonant orbits on delayed double ionization , we compare the resi yields for each of the three atoms . the resulting picture is displayed in fig .
[ fig : resi_to_di_ratio_atom ] for hamiltonian ( [ eq : fullhamiltonian ] ) and @xmath8 . as observed previously , in the nonsequential double ionization regime , the proportion of resi exhibits oscillations .
these oscillations are a clear signature of the resonant periodic orbits of the reduced hamiltonian ( [ eq : he+ ] ) which organize the chaotic layer around the bound region .
note the very good correlation between these oscillations and the residue curve of the @xmath47:@xmath60 resonant orbits for the three atoms . to investigate the importance of the aligned versus non - aligned model in delayed double ionization , in the middle panel we compare the resi yields for hamiltonians ( [ eq : ham2e ] ) and ( [ eq : eckh : ham2e ] ) .
as expected , since they share the same inner electron reduced model , we observe the same trends between the two models where only the amplitude for low intensities is different . we also note that while similar oscillations are observed for the three atoms , the relative resi yield is larger for ne than it is it for both he and ar , in agreement with what has been observed experimentally @xcite .
we have shown that delayed double ionization results from hamiltonian chaos .
this chaotic behavior results from well - identified resonances between the free field dynamics and the laser excitation . depending on the specific question at hand , we have shown the limits of the reduced models : * when the preionized electron is far away from the nucleus , the dynamics of the remaining electron is accurately described by the inner electron reduced model ( from which the resonances are identified ) . as a consequence , different two electron reduced models that share the same inner electron model
[ see hamiltonians ( [ eq : ham2e ] ) and ( [ eq : eckh : ham2e ] ) ] lead to qualitatively similar results for delayed double ionization . *
while one - dimensional reduced models have been successful in describing direct impact double ionization , they give a partial description for the mechanisms behind delayed double ionization . in this case
, they exhibit the skeleton that regulates delayed ionization through the 1:@xmath60 resonances .
however , they miss the influence of the transverse dimensions that enhance chaos . as a consequence ,
one dimensional models tend to significantly underestimate both the amount and the delays for delayed double ionization . *
the mechanism of resonance is very robust against changes of parameters ( laser frequency and species ) . in particular , for different atoms , the number of resonances varies , leading to a slight variation in normalized resi yields .
nevertheless , in all cases , the resonances are correlated in the same way to the observed oscillations in the relative resi yields .
all of these limits have been illustrated , e.g. , on the plots of the normalized resi yields versus intensity .
in particular , in some range of parameters , it has been shown that these curves exhibit oscillations which act as a clear signature of hamiltonian chaos in these systems .
the authors acknowledge useful discussions with c. cirelli , r. doerner , n. johnson , r. r. jones , u. keller , m. kling , r. moshammer , and a. pfeiffer .
c.c . and f.m .
acknowledge financial support from the cnrs . f.m .
acknowledges financial support from the fulbright program .
this work is partially funded by nsf . | we analyze the dynamical processes behind delayed double ionization of atoms subjected to strong laser pulses . using reduced models ,
we show that these processes are a signature of hamiltonian chaos which results from the competition between the laser field and the coulomb attraction to the nucleus .
in particular , we exhibit the paramount role of the unstable manifold of selected periodic orbits which lead to a delay in these double ionizations . among delayed double ionizations , we consider the case of `` recollision excitation with subsequent ionization '' ( resi ) and , as a hallmark of this mechanism
, we predict oscillations in the ratio of resi to double ionization yields versus laser intensity .
we discuss the significance of the dimensionality of the reduced models for the analysis of the dynamical processes behind delayed double ionization . |
+ [ toc ] strong electron correlation effects remain a highly relevant topic in the physics and chemistry of carbon - based materials .
extensive experimentation has confirmed strong electron correlation effects in quasi - one - dimensional ( quasi-1d ) carbon - based systems such as linear @xmath0-conjugated polymers , single - walled carbon nanotubes , graphene nanoribbons , etc .
historically , the study of the excited state ordering in finite polyenes @xcite provided the most convincing evidence for strong electron correlation effects in quasi-1d @xmath0-conjugated systems , and formed the basis of correlated - electron theories in the long chain polyacetylenes and polydiacetylenes . @xcite the early calculations @xcite were within the semiempirical pariser@xmath1parr@xmath1pople ( ppp ) @xmath0-electron model hamiltonian,@xcite within which eigenstates of even polyenes with c@xmath4 point group symmetry have even and odd spatial as well as charge - conjugation symmetries ( ccss ) , and allowed one - photon ( two - photon ) transitions from the @xmath5a@xmath6 ground state are to @xmath7b@xmath8 ( @xmath7a@xmath6 ) states with opposite ( same ) symmetries ( the superscript @xmath9 indicates singlet spin state and the `` plus '' and `` minus '' superscripts refer to ccs ) .
the occurrence of the lowest two - photon state , the 2@xmath7a@xmath6 , _ below _ the one - photon 1@xmath7b@xmath8 in polyenes @xcite is opposite to that expected within the one - electron hckel and mean - field hartree@xmath1fock ( hf ) theories and is a consequence of strong short - range electron@xmath1electron ( e@xmath1e ) interactions within the ppp model.@xcite subsequent to these early model calculations , considerable effort has gone into developing more sophisticated quantum chemical correlated _ ab initio _ approaches to polyene spectra .
quantum chemical approaches that are now able to reproduce the correct one - photon versus two - photon state ordering for short linear polyenes ( usually up to octatetraene ) include the complete active space second order perturbation theory ( caspt2 ) and third - order coupled cluster theory , @xcite and the extended algebraic diagrammatic construction ( adc(2)-x ) method .
@xcite correct excited state ordering is also found within a density functional theory - based multiple reference configuration interaction ( dft - mrci ) approach .
@xcite ccs characteristic of the ppp model is absent within the latter approaches ; however , the violation of ccs is weak .
@xcite importantly , the quantum chemical approaches do find the double excitation character of the 2@xmath7a@xmath10 .
theoretical works have also found reversed excited state ordering in long acenes from pentacene onward , within the ppp model , @xcite a semi - empirical dft - mrci method @xcite and the adc(2)-x approach .
@xcite acenes can be considered as coupled polyene chains and are also quasi-1d .
in contrast to the quasi-1d systems , there have been few , if any , systematic studies of electron correlation effects in two - dimensional ( 2d ) molecules . in this work
we take the first step toward addressing electron correlation effects in 2d molecules by conducting a systematic study of excited states ordering in polycyclic aromatic hydrocarbons ( pahs ) . by performing meticulous experimental and computational analyses of the one- and two - photon excited states for the three pah molecules shown in figure [ f1]a@xmath1c ,
we find reversed excited state ordering within these 2d molecules in clear contradiction to the description of one - electron theory .
these correlation effects are found to be overall weaker compared to quasi-1d carbon - based systems , due to the larger one - electron bandwidth in 2d .
possible extension of our study to larger molecules can also shed light on electron correlation effects in graphene as a genuine 2d carbon - based system in the thermodynamic limit . within
the hckel tight - binding @xmath0-electron theory graphene is a semimetal , with linear energy versus momentum relationship near the dirac point where the filled valence and empty conduction bands meet .
although many experimental observations appear to be in agreement with this description , @xcite the neglect of the repulsive e@xmath1e interaction between the @xmath0-electrons is increasingly being questioned .
@xcite h@xmath11 ) ( b ) hexa - peri - hexabenzocoronene , hereafter hbc ( c@xmath12h@xmath13 ) , and ( d ) circumcoronene ( c@xmath14h@xmath13).,width=192 ]
quantum chemical approaches can correctly describe low - lying two - photon states of double excitation character only for relatively small molecules .
@xcite the two larger pah molecules of interest here , namely , hbc and circumcoronene , lie outside the scope of these calculations
. moreover , while we are principally interested in the relative energies of the lowest one- and two - photon states , we will in the following compare experimental and calculated two - photon absorption ( tpa ) spectra over a broad frequency range , which requires that we are able to calculate the energies and wave functions of the higher energy two - photon states .
these requirements restrict us to perform our calculations within the semiempirical ppp hamiltonian , @xcite which is written as @xmath15 where @xmath16 creates a @xmath0-electron of spin @xmath17 on carbon atom @xmath18 , @xmath19 is the number of electrons of spin @xmath17 on atom @xmath18 , and @xmath20 .
the one - electron hopping integral @xmath21 is between nearest neighbor carbon atoms @xmath18 and @xmath22 , @xmath23 is the hubbard repulsion between two electrons occupying the same atomic @xmath24 orbital , and @xmath25 is the long - range intersite coulomb interaction . we choose standard @xmath26 ev , @xcite and obtain @xmath25 from the parametrization , @xcite @xmath27 , where @xmath28 is the distance in @xmath29 between carbon atoms @xmath18 and @xmath22 and @xmath30 is an effective dielectric constant .
we fix carbon@xmath1carbon ( c@xmath1c ) bond lengths to @xmath31 @xmath29 and all bond angles to 120@xmath32 .
the actual bond lengths and angles may deviate slightly from these mean values , but given the aromatic nature of the molecules , we expect these deviations and their effects on the energies of the one- and two - photon states to be small .
this is shown explicitly for coronene ( see appendix ) , the only molecule for which the experimental bond lengths are available in the literature .
we choose @xmath33 ev and @xmath34 based on the excellent fits obtained previously with these parameters for the excited states of poly@xmath1paraphenylenevinylene , @xcite polyacenes , @xcite and single - walled carbon nanotubes .
@xcite the molecules in figure [ f1 ] belong to the @xmath2 point group and possess inversion , reflection , 6-fold rotational , and cc symmetries . since @xmath2 is not abelian , in the following we adopt the abelian @xmath35 point group symmetry to partition excited states into distinct one- and two - photon classes .
the ground state is in the @xmath7a@xmath6 subspace for all the molecules of figure [ f1]a@xmath1c .
dipole selection rules dictate that one - photon ( linear ) absorption ( tpa ) occur only to @xmath7b@xmath36 and @xmath7b@xmath37 ( @xmath7a@xmath6 and @xmath7b@xmath38 ) states with spatial and cc symmetries opposite to ( the same as ) that of the ground state .
in addition pahs also have @xmath7b@xmath39 and @xmath7b@xmath40 states forbidden in both one- and two - photon processes .
our calculations were done using the multiple reference singles and doubles configuration interaction ( mrsdci ) @xcite .
the mrsdci retains the dominant single , double , triple , and quadruple excitations for each targeted state @xcite , with the number of configurations systematically increased until numerical convergence is reached . while excitonic effects due to the correlation between a single electron@xmath1hole pair in optically allowed states have been treated very successfully for graphene @xcite and other carbon nanostructures @xcite within the gw - bethe salpeter technique , application of this first - principles approach to investigate the lowest two - photon eigenstates has not yet been possible . for example , the lowest two - photon states that occur below the optical gap in trans - polyacetylene @xcite is missed within the gw - bethe salpeter approach , @xcite even though the latter gives accurate description of the optical exciton .
thus , the inclusion of ci with multiple electron@xmath1hole excitations is a necessary requirement for obtaining two - photon states below the lowest one - photon optical state .
@xcite the semiempirical approach taken by us can not be easily extended to the thermodynamic limit ( except in one dimension , where the density matrix renormalization group approach can be used @xcite ) , but this disadvantage is offset by the ability to include high order correlation effects , involving triply and quadruply excited configurations with respect to the restricted hartree@xmath1fock ( rhf ) reference state .
next , we describe the procedure involved in performing various levels of ci calculations presented in this work .
the initial step of our ci studies involves determination of the self - consistent rhf solutions of the ppp hamiltonian ( see eq [ ppp_ham ] ) .
the basis functions of the many - body calculations are configurations with all possible electron occupancies of the hf molecular orbitals ( mos ) .
a full ci calculation involves construction of the hamiltonian matrix with all possible excitations from the hf ground state , and this is the procedure we have followed in the text for octatetraene and decapentaene . for select few cases , quadruple - ci ( qci ) , which incorporates all excitations up to quadruples from the hf ground state , is also possible . with a large number of electrons , as in the present case ,
one then encounters the following problem . on one hand ,
straightforward high - order ci such as the qci is beyond today s computational capability . on the other hand , lower order ci such as double - ci
does not give the correct excited state ordering for large molecules .
it is precisely for such large systems that the mrsdci is used .
@xcite we have previously performed calculations as large as the ones reported here for quasi-1d molecules and oligomers .
@xcite the mrsdci is performed in two steps . at the first step ,
a few ( @xmath41 ) singly and doubly excited reference configurations that best describe the excited states in the targeted symmetry subspace are selected on the basis of a trial double - ci calculation .
the second step involves the mrsdci calculation , in which the hamiltonian matrix consists of single and double excitations with respect to the original @xmath41 configurations themselves .
the total number of configurations @xmath42 therefore includes also the subset of the dominant triple and quadruple excitations .
the new set of @xmath42 configurations are now examined , single and double excitations from the hf ground states that contribute significantly to the eigenstates within the targeted symmetry subspace are identified , and these are included to augment the new reference space , following which the mrsdci is performed again .
this iterative approach is continued , with updating of @xmath41 and @xmath42 at every step , until convergence to some tolerance determined by the size of the system is reached .
l|r symmetry & coronenehbccircumcoronene + & @xmath41@xmath42@xmath41@xmath42@xmath41@xmath42 + + @xmath43b@xmath44 : & @xmath45@xmath46@xmath47@xmath48@xmath49@xmath50 + @xmath43b@xmath51 : & @xmath52@xmath53@xmath54@xmath55@xmath56@xmath57 + @xmath43a@xmath10 : & @xmath58@xmath59@xmath60@xmath61@xmath62@xmath63 + @xmath43b@xmath64 : & @xmath65@xmath66@xmath67@xmath68@xmath69@xmath70 + in table [ t1 ] we have given the final @xmath41 and @xmath42 reached in our mrsdci calculations for all four @xmath35 subspaces relevant for the calculations of nonlinear absorptions in coronene , hbc and circumcoronene ( for the @xmath7a@xmath10 subspace of coronene the calculation was done using qci ) . to the best of our knowledge , these are the largest ci calculations performed to date for the excited states of these molecules .
the tpa is calculated from the imaginary component of the third - order nonlinear optical susceptibility @xmath71 , the resonant contribution @xmath72 of which is given by the sum - overstate expression , @xcite @xmath73\times\hspace{0.7cm}\nonumber \\ \frac{1}{(\omega_{ng}-\omega)(\omega_{mg}-2\omega - i\delta)(\omega_{pg}-\omega)}- \sum_{np}\big[\mu_{gn}^{i}\mu_{ng}^{j}\mu_{gp}^{l}\mu_{pg}^{k}\hspace{0.7cm}\nonumber \\+\mu_{gn}^{i}\mu_{ng}^{k}\mu_{gp}^{l}\mu_{pg}^{j}+ \mu_{gn}^{l}\mu_{ng}^{k}\mu_{gp}^{i}\mu_{pg}^{j}+\mu_{gn}^{l}\mu_{ng}^{j}\mu_{gp}^{i } \mu_{pg}^{k}\big]\times\hspace{0.7cm}\nonumber \\\frac{1}{(\omega_{ng}-\omega)(\omega_{pg}-\omega)(\omega_{pg}-\omega)}\bigg)\,,\hspace{1.0 cm } \label{tpa}\end{aligned}\ ] ] where @xmath74 is the number of molecules , @xmath75 is the ground state , @xmath76 and @xmath77 are virtual one - photon states , and @xmath78 is two - photon states .
@xmath79 is the line width of the spectrum when @xmath80 and was set to @xmath81 ev in all our @xmath72 calculations for various pah molecules in this work . here
@xmath82 is the @xmath18th component of the matrix element of the transition dipole operator between states @xmath75 and @xmath76 .
other components of the transition dipole matrix elements are defined similarly , and the prime over the first summation indicates @xmath83 . with @xmath35 symmetry ,
tpas are to @xmath7a@xmath6 and @xmath7b@xmath38 states , which are nondegenerate when the coulomb interactions are nonzero . the matrix element of the @xmath84-component of the transition dipole operator is nonzero only between the @xmath7a@xmath6 and the @xmath7b@xmath37 states . as a consequence ,
two - photon resonances in the @xmath85 , @xmath86 , @xmath87 , and @xmath88 components of @xmath72 in eq [ tpa ] are all to @xmath7a@xmath6 states only .
similarly , the matrix element of the @xmath89-component of the transition dipole operator is nonzero between @xmath7b@xmath37 and @xmath7b@xmath38 states only .
hence , two - photon resonances in the @xmath90 , @xmath91 , @xmath92 , and @xmath93 components of @xmath72 are all to @xmath7b@xmath38 states . in eq [ tpa ] above
, the intermediate one - photon virtual state of @xmath7b@xmath94 and two - photon states of @xmath7b@xmath95 , and @xmath7b@xmath96 do not contribute because they all involve matrix elements of the @xmath97 component ( out of plane ) of the dipole operator ; due to the planar structure of pah molecules studied in this work , these dipole transition matrix elements are vanishingly small .
the experimental solution - phase pa spectrum is compared against the orientational average of the gas - phase tpa susceptibilities @xcite , @xmath98\,.\hspace{1.0 cm } \label{tpa - avg}\end{aligned}\ ] ]
* hckel calculations . * before presenting the results of the correlated - electron calculations , we discuss the results of calculations based on the hckel model which is the @xmath99 limit of the ppp model .
the goal here is to demonstrate that the hckel model fails _ qualitatively _ to give the excited state ordering that is observed experimentally ( see below ) . in figure
[ f2]a@xmath1c we show the frontier mos for coronene , hbc and circumcoronene , respectively , where we have indicated the highest occupied and lowest unoccupied mos ( homo and lumo ) and the lowest one and two - photon allowed one electron - one hole ( 1e@xmath11h ) transitions ( 1e@xmath11h two - photon transitions that originate from mos below the homo and are related to the ones shown in the figure by ccs are not shown ) .
note that all energies are in units of @xmath100 and dimensionless here .
two electron - two hole ( 2e@xmath12h ) two - photon states lie outside the region of experimental interest and are hence not shown . optically bright and dark states of b@xmath101 symmetry are degenerate in the hckel limit .
the two - photon @xmath43a@xmath6 and @xmath43b@xmath38 states are also degenerate .
the hopping integral @xmath21 determines all orbital energies and energy gaps within the hckel model .
it is thus possible to fit the allowed optical gap for each molecule by choosing an appropriate @xmath21 , although this may need slightly different @xmath21 values for different molecules .
furthermore , in the absence of correlation contribution to the optical gap , it is only natural that the absolute value of @xmath100 needed to fit the optical gap within the hckel model will be larger than in the ppp model .
fitting the peak energy @xmath102 ev of the strong optical absorption in coronene within the hckel theory requires @xmath103 ev . the calculated linear absorption spectrum with the larger @xmath100 is shown in figure [ f2]d .
figure [ f2]d shows our plot of im[@xmath104 ( see eq [ tpa - avg ] ) for coronene , which was evaluated using the exact hckel wave functions and the sum - overstates procedure of eq [ tpa ] .
the lowest two - photon state in figure [ f2]d is @xmath105 ev above the optical one - photon state in coronene . in discussions of excited state ordering , the focus is usually on energetics alone .
we emphasize that the density of two - photon states and the intensities of the tpas also reveal correlation effects , @xcite and note that in figure [ f2]d the tpa is most intense at the lowest energy .
figure [ f2]e shows the hckel plots of linear absorption and tpa in hbc , again with the calculated one - photon gap matching the experimental gap .
the required @xmath21 is now @xmath106 ev , and the lowest two - photon state is @xmath107 ev above the one - photon state .
we chose the same @xmath103 ev in our calculations for circumcoronene as in coronene ( hckel calculation of circumcoronene with this @xmath21 reproduces the optical gap obtained with ppp - mrsdci and @xmath26 ev , as given in table [ t2 ] below ) .
the calculated hckel linear and tpa spectra for circumcoronene are shown in figure [ f2]f . once again , the lowest two - photon state is considerably higher than the one - photon state , and as in coronene the higher energy two - photon state has a weaker intensity than the lower one .
the similarities between the spectra of coronene and circumcoronene , and their differences with the spectra of hbc ( smaller energy difference between lowest two - photon and one - photon states , much larger energy difference between the two - photon states ) arise from the different characters of the edges , namely , zigzag versus armchair .
@xcite to summarize , the energy gaps to the lowest one - photon optically allowed states , observed experimentally , can always be reproduced within the hckel model by choosing an appropriate one - electron hopping integral @xmath21 ( usually larger than realistic ) .
however , as seen in the `` tpa measurement '' subsection below , the same procedure , when extended to two - photon states , not only fails to give correct fits to experimentally observed energies of two - photon states , but actually predicts incorrect excited state ordering .
l|r molecule & s@xmath108 ( ev ) 2@xmath43a@xmath109 ( ev)t@xmath110 ( ev)@xmath111 + + + octatetraene & @xmath112@xmath113@xmath114@xmath115 + + decapentaene & @xmath116@xmath117@xmath118@xmath119 + + coronene & @xmath120@xmath121@xmath122^{d})$]@xmath123 + + hbc & @xmath124@xmath125@xmath126@xmath127 + + circumcoronene & @xmath128@xmath129@xmath130@xmath131 + + * experimental linear absorption versus ppp - mrsdci calculations . *
the linear absorption measurements were performed with a uv / visible spectrophotometer ( varian inc . ) in dilute ( @xmath1320.1 mm ) solutions of coronene and hbc in dichloroethane .
the coronene sample was obtained from sigma - aldrich chemical co. and hbc was prepared using the method outlined in ref .
we observed that the main features of absorption spectra do not change for concentrations between 0.01 and 1 mm , indicating that aggregation effects , if any , are minimal .
figure [ f3]a , b present the experimental solution phase linear absorption spectra of coronene and hbc , superimposed on our calculated ppp - mrsdci absorption spectra .
all calculated ppp spectra , here and in the following , are with uniform @xmath26 ev .
we find excellent match between the strong experimental absorption band and the calculated absorption band corresponding to transitions to the degenerate 1@xmath7b@xmath36 and 1@xmath7b@xmath37 states in both coronene and hbc .
the weak splitting in the strong experimental absorption band , seen in both cases , confirms the existence of two underlying transitions that have lost their perfect degeneracy in the real systems .
the much weaker absorptions ( at @xmath133 ev in coronene and @xmath134 ev in hbc ) , missing in the calculated absorption spectra , are to the `` forbidden '' 1@xmath7b@xmath39 and 1@xmath7b@xmath40 states .
loss of degeneracy of the allowed transitions , and nonzero oscillator strengths of the `` forbidden '' transitions can be due to the couplings of electrons with vibronic modes as well as weak second neighbor electron hopping that are absent in our hamiltonian .
neither of these interactions affect the relative ordering of the one- and two - photon states .
* tpa measurement . *
we conducted experimental measurements of tpa in @xmath135 1 mm solutions of coronene and hbc using dichloroethane as a solvent .
high - intensity , 100 fs laser pulses with photon energies between 1.5@xmath13 ev were obtained from titanium - sapphire pumped optical parametric amplifier ( opa ) and used to measure the tpas to states at energies [email protected] ev . the coronene and hbc solutions in 1 mm uv - quartz cuvettes were used for open - aperture `` z - scan '' measurement that allows extraction of tpa coefficient .
@xcite we employed a modified differential z - scan approach similar to the one discussed in ref to enhance the sensitivity of our tpa measurements .
the z - scan measurements with the solvent alone were used as the `` control '' for the data reported here . in figure
[ f3]c we show the experimental tpa spectrum of coronene , where we have included the linear absorption for comparison .
the tpa spectrum of figure [ f3]c is different from the hckel tpa spectrum for coronene in figure [ f2]d in three distinct ways : ( i ) resonances to two - photon states _ below _ the allowed one - photon states are observed , ( ii ) the experimental spectrum shows many more structures , and ( iii ) the intensities of the tpas are higher at higher energy .
the tpa spectrum of hbc , shown in figure [ f3]d is very similar to that of coronene , with once again nonzero tpa below the strong one - photon allowed linear absorption .
the strengths of the tpas , both below the optical gap and at higher energies , are significantly higher for hbc with larger molecular size . a similar increase in the intensities of the tpas for small chain lengths
is also observed in polyenes .
@xcite we again note the apparent high density of two - photon states in the energy region [email protected] ev , in contradiction to the prediction of the hckel theory ( see figure [ f2]e ) .
the energy - dependence of the intensities for the tpas in hbc is also similar to that in coronene . * ppp - mrsdci calculations of tpa .
* figure [ f3]e , f presents the calculated tpa spectra ( eq [ tpa - avg ] ) .
dipole matrix elements were evaluated using the ppp - mrsdci many - body wave functions .
the calculated linear spectra have been included in the figures for comparison .
the @xmath136 spectrum has resonances to both @xmath7a@xmath6 and @xmath7b@xmath38 states ( see arrows in figure [ f3]e , f ) . in agreement with experiments ,
the calculated lowest two - photon state , the 2@xmath7a@xmath6 , occurs below the optical 1@xmath7b@xmath36 and 1@xmath7b@xmath37 states in coronene ( by @xmath137 ev ) as well as in hbc ( by @xmath138 ev ) . comparing figure [ f3]d
, f we find excellent matches between the energy locations of the experimental and calculated two - photon resonances in hbc .
the fit is less impressive for coronene .
the high densities of two - photon states observed in both coronene and hbc agree with the occurrence of many close lying @xmath7a@xmath6 and @xmath7b@xmath38 states in the same energy range within the ppp model .
the calculations also reproduce the intensity profile of the experimental tpas , with the intensity at higher energies larger than that at lower energies .
the calculated lowest two - photon state is slightly above the lowest one - photon state in circumcoronene ( see table [ t2 ] below ) , with , however , an energy separation ( @xmath138 ev ) much smaller than expected within one - electron theory .
the almost `` normal '' excited state ordering in circumcoronene is likely due to a weak breakdown of the mrsdci approximation , which becomes less appropriate with increasing size.@xcite with @xmath139 c - atoms , circumcoronene is more than twice the size of coronene and is near the memory limit of mrsdci , _ viz .
_ , @xmath140 atoms .
correlation effects in 2d , though substantial , are weaker than in linear chains .
table [ t2 ] summarizes the calculated ppp energies of the lowest one- and two - photon spin singlet states , the lowest triplet state ( hereafter t@xmath141 ) , and the wave function characteristics of the lowest two - photon state for 1,3,5,7-octatetraene , 1,3,5,7,9-decapentaene , and the three molecules we have investigated here .
the polyene full ci calculations were done with hopping integrals @xmath142 and @xmath143 ev for the double and single bonds , respectively.@xcite we have also included in table [ t2 ] all experimental energy values that were available to us from literature .
we note the excellent agreements between all theoretical and experimental energies for the two polyenes .
equally remarkable agreement is found between the calculated and experimental triplet energies of coronene .
the 2@xmath7a@xmath6 in the polyenes is known to be a quantum - entangled `` two - triplet '' state @xcite whose energy is nearly twice that of t@xmath141 .
by contrast , 2@xmath7a@xmath6 in the pah molecules in all cases is less than twice the energy of their t@xmath141 .
the low energy of the 2@xmath7a@xmath6 in polyenes , relative to the 1@xmath7b@xmath8 , is best understood within valence bond ( vb ) theory .
@xcite for strong hubbard @xmath23 , the ground state within vb theory is predominantly covalent , _
i.e. _ , its wave function is dominated by vb diagrams in which all carbon atom @xmath144-orbitals are singly occupied .
optical absorption within tight - binding hamiltonians necessarily requires charge - transfer between carbon atoms,@xcite which then implies that the 1@xmath7b@xmath8 is dominated by ionic vb diagrams with at least one pair of c@xmath145c@xmath146 ion pair .
two - photon states are reached by a _
second _ charge - transfer excitation from a virtual ionic state , which can create yet another c@xmath145c@xmath146 ion pair , or annihilate the original ion pair , creating an excited state that is also covalent , but orthogonal to the ground state .
@xcite for nonzero hubbard @xmath23 , the lowest ionic states are necessarily higher in energy than the lowest covalent states .
there is a one - to - one correspondence between this qualitative vb picture and the ci approach within the mo basis , in that the 1@xmath7b@xmath8 , obtained by a single charge - transfer from the ground state in configuration space , is an 1e@xmath11h excitation , while the 2@xmath7a@xmath6 has strong contributions from 2e@xmath12h excitations . @xcite .
since 1e@xmath11h excitations , however , dominate the 2@xmath7a@xmath6 wave function in the 2d compounds , especially in hbc , the analogy with the above vb qualitative picture suggests that the 2@xmath7a@xmath6 in the pah is presumably more ionic than in 1d .
while investigation of the ionicity of excited states of such large molecules is a formidable problem , we note that within vb theory large covalent character ( small ionicity ) of the 2@xmath7a@xmath6 necessarily requires large covalent character of the ground state , which can be investigated even for large molecules within the ppp model using a variety of approximate methods such as quantum monte carlo .
we intend to investigate the relative covalent characters of polyenes and the present pah molecules in the future using one such approximate method , the path integral renormalization group , used previously by one of us in a different context .
@xcite to conclude , we have implemented very high level correlated - electron calculations for the one- and two - photon energy states and absorption spectra of three pah molecules : coronene , hbc and circumcoronene , all possessing @xmath2 point group symmetry . for coronene and hbc
we have also performed precise one- and two - photon optical absorption experiments .
the calculated one - photon energies of the 1@xmath7b@xmath36 and 1@xmath7b@xmath37 give excellent fits to the experimental transition energies of these states in both coronene and hbc .
for the first time , two - photon states below the lowest optically allowed one - photon state are demonstrated in 2d pahs , in complete contradiction to the predictions of the one - electron hckel model .
theoretical determination of subgap two - photon states can be achieved only by taking into account electron correlation effects that are an order higher than the 1e@xmath11h excitonic interactions.@xcite .
our theoretical and experimental works , taken together , indicate that while the consequences of e@xmath1e interactions in the pahs are weaker than in polyenes , these effects are nevertheless significant .
in particular , correlation effects in pahs appear to be stronger than in the phenyl group - containing @xmath0-conjugated quasi-1d polymers poly-(paraphenylene ) and poly-(paraphenylenevinylene ) , in which the 2@xmath7a@xmath6 occurs _ above _ the 1@xmath7b@xmath8 state .
@xcite in the future we are planning to perform similar theoretical as well as experimental investigations of pah molecules larger than the ones considered here .
such analyses can potentially open up an effective means to determine the effective short - range hubbard @xmath23 in graphene , which as we pointed out in the introduction , remains controversial .
@xmath29 assigned to all c@xmath1c bonds .
( b ) actual bond lengths of coronene , from ref .
note that @xmath2 symmetry is maintained .
( c ) ground state linear absorption corresponding to ( a ) and ( b ) , performed using the single - ci approximation.,width=336 ]
in our theoretical analyses of e@xmath1e correlation effects , we have used idealized geometries with uniform c@xmath1c bond lengths of @xmath31 @xmath29 and hopping integrals @xmath26 ev . in reality , the c@xmath1c bonds deviate from this idealized length , and these deviations can in principle influence our overall results .
deviations from the idealized geometry are , however , expected to be small for the aromatic molecules in question .
the experimental bond lengths are available only for coronene . here
we demonstrate explicitly for this case that the expected changes in energies due to geometry modifications are negligible , and in no way affects our main conclusion . in figure [ f4]a ,
b we have shown the coronene molecule with its idealized uniform bond lengths , and with the actual bond lengths , @xcite respectively . we calculated all hopping integrals from the known bond length@xmath1hopping integral relationship , @xmath147 , @xcite where @xmath148 is the c@xmath1c bond length in .
the calculated hopping integrals in descending order are @xmath149 , @xmath150 , @xmath151 , and @xmath152 ev , respectively .
the @xmath25 were calculated as defined in eq [ ppp_ham ] within the text .
figure [ f4]c shows the ground state linear absorption within the single - ci for the geometries of figure [ f4]a , b .
deviation from equal bond lengths blue shifts the main absorption band very weakly ( the energy of the optical state is slightly smaller than what is obtained within the mrsdci , but this is of no concern here ) .
as mentioned above , experimental bond lengths are currently not available for hbc . based on our results for coronene , we believe , however , that it is safe to assume that any deviations in energies due to weak distortions from idealized geometries will not affect our overall qualitative conclusion about the reversed excited state ordering , which can not be obtained without explicit inclusion of strong e@xmath1e interactions , and which is the core theme of our work .
* corresponding author * + @xmath153e - mail : [email protected] ; phone : + 1 - 520 - 621 - 6798 ; fax : + 1 - 520 - 621 - 4721 .
+ * notes * + the authors declare no competing financial interest .
this work was supported by nsf grant no .
a.r . acknowledges the support from the dod smart program .
a.sh . and s.m .
acknowledge support from the indo - us science and technology forum award 37 - 2012/2013 - 14 .
57 hudson , b. s. ; kohler , b. e. ; schulten , k. linear polyene electronic - structure and potential . _ excited states . _ * 1982 , * _ 6 , _ 1@xmath195 .
christensen , r. l. ; galinato , m. g. i. ; chu , e. f. ; howard , j. n. ; broene , r. d. ; frank , h. a. energies of low - lying excited states of linear polyenes .
_ j. phys .
a _ * 2008 , * _ 112 , _ 12629@xmath112636 .
schulten , k. ; karplus , m. on the origin of a low - lying forbidden transition in polyenes and related molecules .
lett . _ * 1972 , * _ 14 , _ 305@xmath1309 .
ramasesha , s. ; soos , z. g. correlated states in linear polyenes , radicals , and ions : exact ppp transition moments and spin densities . _ j. chem . phys . _ * 1984 , * _ 80 , _ 3278@xmath13287 .
tavan , p. ; schulten , k. electronic excitations in finite and infinite polyenes .
b _ * 1987 , * _ 36 , _ 4337@xmath14358 .
baeriswyl , d. ; campbell , d. k. ; mazumdar , s. conjugated conducting polymers ; kiess , h. , ed . ; springer : berlin , 1992 .
barcza , g. ; barford , w. ; gebhard , f. ; legeza , o. excited states in polydiacetylene matrix renormalization group study .
b _ * 2013 , * _ 87 , _ 245116@xmath1245131 .
pariser , r. ; parr , r. g. a semi - empirical theory of the electronic spectra and electronic structure of complex unsaturated molecules .
_ j. chem .
phys . _ * 1953 , * _ 21 , _ 767@xmath1776 .
pople , j. a. electron interaction in unsaturated hydrocarbons . _
faraday soc . _ * 1953 , * _ 49 , _ 1375@xmath11385 .
schreiber , m. ; silva - junior , m. r. ; sauer , s. p. a. ; thiel , w. benchmarks for electronically excited states : caspt2 , cc2 , ccsd , and cc3 .
_ j. chem .
phys . _ * 2008 , * _ 128 , _ 134110@xmath1134134 .
silva - junior , m. r. ; schreiber , m. ; sauer , s. p. a. ; thiel , w. benchmarks for electronically excited states : basis set effects on caspt2 results .
_ j. chem .
phys . _ * 2010 , * _ 133 , _ 174318@xmath1174330 .
starcke , j. h. ; wormit , m. ; schirmer , j. ; dreuw , a. how much double excitation character do the lowest excited states of linear polyenes have ? _ chem .
_ * 2006 , * _ 329 , _ 39@xmath149 .
knippenberg , s. ; rehn , d. r. ; wormit , m. ; starcke , j. h. ; rusakova , i. l. ; trofimov , a. b. ; dreuw , a. calculations of nonlinear response properties using the intermediate state representation and the algebraic - diagrammatic construction polarization propagator approach : two - photon absorption spectra . _ j. chem .
phys . _ * 2012 , * _ 136 , _ 064107@xmath1064121 .
krauter , c. m. ; pernpointner , m. ; dreuw , a. application of the scaled - opposite - spin approximation to the algebraic diagrammatic construction schemes of second order . _ j. chem .
* 2013 * , _ 138 , _ 044107@xmath1044118 .
silva - junior , m. r. ; schreiber , m. ; sauer , s. p. a. ; thiel , w. benchmarks for electronically excited states : time - dependent density functional theory and density functional theory based multireference configuration interaction .
_ j. chem .
phys . _ * 2008 * , _ 129 , _ 104103@xmath1104116 .
schmidt , m. ; tavan , p. electronic excitations in long polyenes revisited .
_ j. chem .
phys . _ * 2012 * , _ 136 , _ 124309@xmath1124321 .
tavan , p. ; schulten , k. correlation effects in the spectra of polyacenes . _ j. chem .
phys . _ * 1979 , * _ 70 , _ 5415@xmath15421 .
raghu , c. ; anusooya pati , y. ; ramasesha , s. density - matrix renormalization - group study of low - lying excitations of polyacene within a pariser@xmath1parr@xmath1pople model .
b _ * 2002 , * _ 66 , _ 035116@xmath1035126 .
marian , c. m. ; gilka , n. performance of the density functional theory / multireference configuration interaction method on electronic excitation of extended @xmath0-systems .
_ j. chem . theory .
comput . _ * 2008 , * _ 4 , _ 1501@xmath11515 .
knippenberg , s. ; starcke , j. h. ; wormit m. ; dreuw , a. the low - lying excited states of neutral polyacenes and their radical cations : a quantum chemical study employing the algebraic diagrammatic construction scheme of second order .
phys . _ * 2010 , * _ 108 , _ 2801@xmath12813 .
geim , a. k. ; novoselov , s. the rise of graphene .
_ nat . mat . _ * 2007 , * _ 6 , _
castro neto , a. h. ; guinea , f. ; peres , n. m. r. ; novoselov , k. s. ; geim , a. k. the electronic properties of graphene .
phys . _ * 2009 , * _ 81 , _ 109@xmath1162 .
elias , d. c. ; gorbachev , r. v. ; mayorov , a. s. ; morozov , s. v. ; zhukov , a. a. ; blake , p. ; ponomarenko , l. a. ; grigorieva , i. v. ; novoselov , k. s. ; guinea , f. ; et al .
dirac cones reshaped by interaction effects in suspended graphene .
phys . _ * 2011 , * _ 7 , _ 701@xmath1704 .
kotov , v. n. ; uchoa , b. ; pereira , v. m. ; guinea , f. ; castro neto , a. h. electron@xmath1electron interactions in graphene : current status and perspectives .
_ * 2012 , * _ 84 , _ 1067@xmath11125 .
chandross , m. ; mazumdar , s. coulomb interactions and linear , nonlinear , and triplet absorption in poly(paraphenylenevinylene ) .
b _ * 1997 , * _ 55 , _ 1497@xmath11504 .
sony , p. ; shukla , a. large - scale correlated calculations of linear optical absorption and low - lying excited states of polyacenes : pariser@xmath1parr@xmath1pople hamiltonian .
b _ * 2007 , * _ 75 , _ 155208@xmath1155229 .
wang , z. d. ; zhao , h. b. ; mazumdar , s. quantitative calculations of the excitonic energy spectra of semiconducting single - walled carbon nanotubes within a @xmath0-electron model .
_ phys . rev .
b _ * 2006 , * _ 74 , _ 195406@xmath1195411 .
yang , l. ; deslippe , j. ; park , c .- h . ;
cohen , m. l. ; louie , s. g. excitonic effects on the optical response of graphene and bilayer graphene .
lett . _ * 2009 , * _ 103 , _ 186802@xmath1186805 .
spataru , c. d. ; ismail - beigi , s. ; benedict , l. x. ; louie , s. g. excitonic effects and optical spectra of single - walled carbon nanotubes .
* 2004 , * _ 92 , _
chang , e. ; bussi , g. ; ruini , a. ; molinari , e. excitons in carbon nanotubes : an ab initio symmetry - based approach .
lett . _ * 2004 , * _ 92 , _ 196401@xmath1196404 .
prezzi , d. ; varsano , d. ; ruini , a. ; marini , a. ; molinari , e. optical properties of graphene nanoribbons : the role of many - body effects .
b _ * 2008 , * _ 77 , _ 041404@xmath1041407
. rohlfing , m. ; louie s. g. optical excitations in conjugated polymers .
_ * 1999 , * _ 82 , _ 1959@xmath11962 .
ramasesha , s. ; pati , s. k. ; shuai , z. ; brdas , j. l. the density matrix renormalization group method : application to the low - lying electronic states in conjugated polymers .
quantum chem .
_ * 2000 , * _ 38 , _ 121@xmath1215 .
bursill , r. j. ; barford , w. large - scale numerical investigation of excited states in poly(paraphenylene ) .
b _ * 2002 , * _ 66 , _ 205112@xmath1205119 .
buenker , r. j. ; peyerimhoff , s. d. individualized configuration selection in ci calculations with subsequent energy extrapolation .
_ theor .
acta . _ * 1974 , * _ 35 , _ 33@xmath158 .
tavan , p. ; schulten , k. the low - lying electronic excitations in long polyenes : a ppp - mrd - ci study . _ j. chem .
phys . _ * 1986 , * _ 85 , _ 6602@xmath16609 .
ghosh , h. ; shukla , a. ; mazumdar , s. electron - correlation - induced transverse delocalization and longitudinal confinement in excited states of phenyl - substituted polyacetylenes .
b _ * 2000 , * _ 62 , _ 12763@xmath112774 .
shukla , a. correlated theory of triplet photoinduced absorption in phenylene@xmath1vinylene chains .
b _ * 2002 , * _ 65 , _ 125204@xmath1125209 .
shukla , a. ; ghosh , h. ; mazumdar , s. theory of excited - state absorption in phenylene - based @xmath0-conjugated polymers .
b _ * 2003 , * _ 67 , _ 245203@xmath1245211 .
shukla , a. theory of two - photon absorption in poly(diphenyl ) polyacetylenes .
phys . _ * 2004 , * _ 300 , _ 177@xmath1188 .
shukla , a. theory of nonlinear optical properties of phenyl - substituted polyacetylenes .
b _ * 2004 , * _ 69 , _ 165218@xmath1165227 .
sony , p. ; shukla , a. photoinduced absorption in disubstituted polyacetylenes : comparison of theory with experiments .
b _ * 2005 , * _ 71 , _ 165204@xmath1165208 .
aryanpour , k. ; sheng , c. x. ; olejnik , e. ; pandit , b. ; psiachos , d. ; mazumdar , s. ; vardeny , z. v. evidence for excimer photoexcitations in an ordered @xmath0-conjugated polymer film .
b _ * 2011 , * _ 83 , _ 155124@xmath1155128 .
boyd , r. w. nonlinear optics ; academic press , inc .
: san diego , ca , 1992 .
heflin , j. r. ; wong , k. y. ; zamani - khamiri , o. ; garito , a. f. nonlinear optical properties of linear chains and electron - correlation effects .
b _ * 1988 , * _ 38 , _ 1573@xmath11576 .
mcwilliams , p. c. m. ; hayden , g. w. ; soos , z. g. theory of even - parity states and two - photon spectra of conjugated polymers .
b _ * 1991 , * _ 43 , _ 9777@xmath19791 .
mazumdar , s. ; guo , f. intensities of two - photon absorptions to low - lying even - parity states in linear - chain conjugated polymers .
b _ * 1994 , * _ 49 , _ 10102@xmath110112 .
allan , m. ; neuhaus l. ; haselbach e. ( all - e)-1,3,5,7-octatetraene : electron - energy - loss and electron - transmission spectra .
acta _ * 1984 , * _ 67 , _ 1776@xmath11782 .
abouaf , r. ; daz - tendero s. electron energy loss spectroscopy and anion formation in gas phase coronene .
phys . _ * 2009 , * _ 11 , _ 5686@xmath15694 .
gl , a. d. ; potasz , p. ; hawrylak , p. excitonic absorption in gate - controlled graphene quantum dots .
b _ * 2010 , * _ 82 , _ 155445@xmath1155449 .
rathore , r. ; burns , c. l. a practical one - pot synthesis of soluble hexa - peri - hexabenzocoronene and isolation of its cation - radical salt .
_ j. org .
chem . _ * 2003 , * _ 68 , _ 4071@xmath14074 .
sheik - bahae , m. ; said , a. a. ; wei , t .- h . ;
hagan , d. j. ; van stryland , e. w. sensitive measurement of optical nonlinearities using a single beam .
j. quantum electron .
_ * 1990 , * _ 26 , _ 760@xmath1769 .
mnard , j. m. ; betz , m. ; sigal , i. ; van driel , h. m. single - beam differential z - scan technique .
. opt . _ * 2007 , * _ 46 , _ 2119@xmath12122 .
dayal , s. ; clay , r. t. ; mazumdar , s. absence of long - range superconducting correlations in the frustrated half - filled - band hubbard model .
b _ * 2012 , * _ 85 , _ 165141@xmath1165148 .
soos , z. g. ; ramasesha , s. ; galvo , d. s. band to correlated crossover in alternating hubbard and pariser - parr - pople chains : nature of the lowest singlet excitation of conjugated polymers .
lett . _ * 1993 , * _ 71 , _
fawcett , j. k. the determination and refinement of the molecular structures of some organic compounds ; http://hdl.handle.net/2429/38382 , 1965 .
ducasse , l. r. ; miller , t. e. ; soos , z. g. correlated states in finite polyenes : exact ppp results .
_ j. chem .
phys . _ * 1982 , * _ 76 , _ 4094@xmath14104 . | strong electron correlation effects in the photophysics of quasi - one - dimensional @xmath0-conjugated organic systems such as polyenes , polyacetylenes , polydiacetylenes , etc . , have been extensively studied .
far less is known on correlation effects in two - dimensional @xmath0-conjugated systems . here
we present theoretical and experimental evidence for moderate repulsive electron@xmath1electron interactions in a number of finite polycyclic aromatic hydrocarbon molecules with @xmath2 symmetry .
we show that the excited state orderings in these molecules are reversed relative to that expected within one - electron and mean - field theories .
our results reflect similarities as well as differences in the role and magnitude of electron correlation effects in these two - dimensional molecules compared to those in polyenes . |
double neutron star ( dns ) systems , i.e. binaries consisting of a radio pulsar and a usually undetected second neutron star , have provided an excellent test bed for general relativity theory .
the orbital decay observed in dnss is the first indirect evidence for gravitational - wave radiation .
the merger of the two neutron stars in dnss is expected to be a major source for gravitational - wave detection by the laser interferometer gravitational wave observatory ( ligo ) , virgo , or the laser interferometer space antenna ( lisa ) .
such mergers are also thought to be potential progenitors of the short - duration gamma - ray bursts ( grbs ) .
the most commonly discussed formation channel of dnss , which we refer to as the standard scenario ( e.g. bhattacharya & van den heuvel 1991 ) , involves a high - mass x - ray binary ( hmxb ) phase at an intermediate stage .
the hmxb eventually undergoes roche - lobe overflow ( rlof ) which leads to a common - envelope ( ce ) phase .
the expected outcome of this phase is a binary consisting of the helium core of the massive star and an x - ray emitting neutron star .
chevalier ( 1993 ) argued that a neutron star which moves in the dense environment of a stellar envelope can accrete a considerable amount of matter , which may push the neutron star above the maximum mass allowed for neutron - star matter and lead to the formation of a black hole .
because of this argument , brown ( 1995 ) suggested that hmxbs passing through a ce generally produce helium - star / black - hole binaries rather than helium - star / neutron - star binaries and hence do not lead to the formation of dnss in this standard scenario . to avoid the problem of hypercritical accretion onto a neutron star , brown ( 1995 ) ( see also wettig & brown 1996 ; bethe & brown 1998 ) suggested that dnss originate from binaries consisting of two helium stars in a close orbit , which we will refer to as the double core ce scenario .
the first - born neutron star accretes part of the stellar wind from the other helium star which may weaken its magnetic field strength and spin up the neutron star .
an important constraint for the formation of a double helium - star binary in a close orbit is that the systems start with two components of almost equal mass ( within 4 per cent ) .
however , strong accretion might not have to occur if the companion is extended ( chevalier 1993 ) , if rotation of the magnetized neutron star is taken into account ( chevalier 1996 ) or if there is a strong outflow from a disc around the neutron star ( armitage & livio 2000 ) .
furthermore , taam , king & ritter ( 2000 ) suggested that , considering the low rates of accretion characteristic of the low density envelopes of more evolved stars in combination with the short duration of the ce phase , accretion may not be important .
these counter - arguments suggest that the formation of dnss via the standard channel can not be ruled out at this stage . moreover , there are good arguments suggesting that at least for some pulsars in binaries the neutron star must have passed through a common envelope ( e.g. tauris , van den heuvel & savonije 2000 )
. we will not argue in this paper whether or not a neutron star in a ce phase will become a black hole . nevertheless , the standard scenario clearly is not valid for the situation where the initial binary has almost equal mass components , and hence it is interesting to investigate whether binary systems with these particular initial conditions could also lead to the formation of dnss , which is the main purpose of this paper .
we refer to this channel with stars of almost equal initial masses as the double - core scenario . using binary population synthesis simulations , we study the expected observational properties as well as the birth and merger rates of dnss formed via the double - core scenario and compare them to the results of the standard scenario .
we describe the detail of the double - core scenario , as well as the basic assumptions and the code used in this work , in sect .
[ doublecore : sec : method ] .
the results are presented in sect .
[ doublecore : sec : result ] and discussed further in sect .
[ doublecore : sec : conclusion ] .
the double - core scenario we consider in this paper is a slight variation of the original proposal by brown ( 1995 ) , i.e. it starts with a binary consisting of two stars of almost equal initial masses .
the main difference to the original scenario is that we put more constraints on the initial masses and periods for the following reason .
the pulsars in dnss , which are the first - born neutron stars , are recycled , and this recycling process is assumed to be due to mass accretion from the progenitors of the pulsar companions . therefore , it is necessary that the first neutron star has sufficient time to accrete matter from the companion .
binary calculations to model the formation of contact binaries have been presented by , e.g. , pols ( 1994 ) , wellstein , langer & braun ( 2001 ) and nelson & eggleton ( 2001 ) .
however , evolution through the contact phase has not been modelled ( webbink 2003 ) . here
we assume that the outcome of a contact phase is a close binary consisting of the cores of the primary and the secondary star .
if the masses of the components are very similar , i.e. differ by less than 0.3 per cent ( podsiadlowski , joss & hsu 1992 ) , the outcome of the contact phase is that the two cores are more or less in the same evolutionary phase , i.e. produces he + he or co + co binaries . in this case , the two cores also have almost equal masses and collapse into neutron stars with a relatively short time between the two collapse phases . in this situation , the first - born neutron star has a very limited possibility to accrete matter from the companion .
moreover , the probability of having a binary system with such similar initial masses is extremely low , and therefore we would not expect to form many recycled dnss binaries in this case .
a channel to form non - recycled dnss from almost - twin binaries has previously been studied by belczyski & kalogera ( 2001 ) . to avoid having two cores of almost equal masses and to allow the first neutron star to accrete matter
, we propose that at the onset of the contact phase , the primary is in a more advanced stage of evolution than the secondary .
this can be achieved , e.g. , if the primary has developed a co core when the secondary starts core helium burning , or if the primary has started core helium burning while the secondary is at the end of main sequence or at the beginning of shell hydrogen burning . in this paper we will consider the first option , i.e. the outcome of the contact phase is the co core of the primary and the helium core of the secondary .
the co core collapses into a neutron star , which then accretes matter from the wind and/or the transferred matter during rlof from the helium star .
the latest stage of this scenario is the same as in the standard scenario , i.e. a helium - star / neutron - star ( hes - ns ) binary in a close orbit .
this formation of dns through the double - core scenario is presented in fig .
[ doublecore : fig : scenario ] .
we performed monte carlo simulations of binary systems with almost equal mass components .
the initial primary mass distribution was assumed to follow a salpeter mass function .
since our main aim is to investigate the formation of dnss , we therefore limited the initial primary masses to be @xmath1 , i.e. the presently favoured mass range in which single stars leave neutron stars as remnants .
the secondary mass is chosen such that the mass ratio of the system , @xmath2 , follows a uniform distribution . because our investigation is on almost equal mass binaries , only systems with @xmath3
are taken into account , where the critical mass ratio @xmath4 is determined from single star models so that the secondary will already have finished its main - sequence phase when the primary has finished its helium - core burning phase .
the value of @xmath5 is therefore a function of @xmath6 , varying from 0.97 for @xmath7 to 0.93 for @xmath8 . for the distribution of orbital periods
we take a uniform distribution in @xmath9 , with @xmath10 . to obtain the co core of the primary and the he core of the secondary at the end of the contact phase , we limit our simulations to binary systems in which the primary undergoes case c mass transfer , i.e. its roche radius follows @xmath11 , where @xmath12 and @xmath13 are the minimum and maximum radius for case c mass transfer from the primary .
the minimum radius for case c mass transfer depends on how the star evolves accross the hertzsprung - russell ( h - r ) diagram , in particular on whether it spends most of its helium core - burning phase as a blue supergiant or whether it first experiences an extended helium core - burning phase as a red supergiant ( see , e.g. , podsiadlowski et al .
1992 ) . in the latter case , the minimum radius for case c
mass transfer would be substantially larger , leading to fewer systems that can experience case c mass transfer .
the details of the radius evolution of massive stars is still a major unresolved problem in the theory of evolution of massive stars ; it depends on the treatment of convection , including the effects of convective overshooting and semi - convection , the treatment of mass loss and other mixing processes ( for further discussion see , e.g. , langer & maeder 1995 ) . in order to be able to assess the importance of these uncertainties , we consider two extreme options , where in one we assume that all stars first experience helium core - burning as red supergiants , leading to a high minimum radius for case c mass transfer , and where in the other we assume that the star spends most of its helium core - burning phase as a blue supergiant , leading to a low minimum radius for case c mass transfer .
the corresponding radii , @xmath12 and @xmath13 were estimated from single - star models and their radii in the respective phases . the separation after the contact phase , @xmath14 , is determined as in the case of the standard single - core common - envelope evolution ( webbink 1984 ) , i.e. the envelopes are ejected on the expense of the change in the orbital energy , @xmath15 , where the envelope binding energy is @xmath16 and the change in orbital energy is @xmath17 at the onset of mass transfer , @xmath18 .
the secondary is within its roche lobe but expands as soon as it accretes matter from the primary , such that at the onset of the contact phase @xmath19 .
prior to the mass transfer phase , we neglect the stellar wind mass loss such that the roche radii do not change . using eqs .
[ doublecore : eq : binding ] and [ doublecore : eq : orbital ] , the reduction in separation can be expressed as @xmath20 here @xmath21 is an efficiency parameter for the envelope ejection , taken to be unity .
the masses @xmath22 , @xmath23 and @xmath24 ( @xmath25 for primary and secondary , respectively ) are the stellar mass , the core mass and the envelope mass .
the co and he core masses were taken from standard single star models for solar metallicity and adopting the schwarzschild criterion for convection and included 0.25 pressure scale heights of convective overshooting .
@xmath26 , where @xmath27 is the roche - lobe radius , obtained as in eggleton ( 1983 ) , and @xmath28 is the separation at the onset of the contact phase .
@xmath29 is the envelope binding energy parameter , which depends on the stellar mass and evolutionary stage ( dewi & tauris 2000 ; tauris & dewi 2001 ) . for the purpose of this work
we take a constant value of @xmath30 , i.e. 0.1 which is the average @xmath30 value for high - mass stars ( dewi & tauris 2001 )
. the exact value of @xmath30 derived from the whole binding energy can be as low as 0.02 in stars more massive than @xmath31 20 @xmath32 and as high as 0.7 in 10 @xmath32 on the giant branch .
as alternative option , we also perform simulations with @xmath33 , a value commonly adopted in population synthesis studies involving ce evolution . for each simulation
we apply the same @xmath30 value for both primary and secondary stars . after the contact phase
, we assume that there is no further episode of mass transfer from the co core to the helium core .
the co core explodes in an asymmetric supernova ( sn ) , in which a natal kick is imparted at the birth of the neutron star .
we apply a maxwellian kick velocity distribution with a dispersion @xmath34 190 @xmath35 ( model hp ; hansen & phinney 1997 ) and a single gaussian distribution with @xmath34 290 @xmath35 ( model acc ; arzoumanian , chernoff & cordes 2002 ) .
the latter distribution is similar to the distribution found in the most recent study by hobbs et al .
( 2005 ) who derived a kick velocity distribution with @xmath34 265 @xmath35 .
the orbital period of the post - sn binary , as well as the system velocity after the explosion , are determined as in brandt & podsiadlowski ( 1995 ) .
the explosion is assumed to leave a 1.4 @xmath32 neutron star .
if the binary survives the sn explosion , it continues its evolution as a hes - ns system .
extensive studies on mass transfer phase in hes - ns binaries have been carried out by dewi et al .
( 2002 ) and dewi & pols ( 2003 ) , taken into account a wide parameter space in helium star masses and orbital periods . during the mass - transfer phase , it was assumed that the neutron star accretes matter up to its eddington limit and the rest of the transferred mass is lost to the system , carrying the specific angular momentum of the neutron star .
we apply the results of the systematic studies by dewi et al .
( 2002 ) and dewi & pols ( 2003 ) to determine the orbital parameters of the system at the end of rlof , which are the same as those prior to the explosion of the second sn .
we apply the same kick velocity distribution as in the first sn .
we also consider the recent suggestion that the second neutron star experiences a much lower kick velocity ( dewi , podsiadlowski & pols 2005 ) , i.e. by combining the above - mentioned kick distributions for the first neutron star and a distribution with @xmath34 20 @xmath35 for the second .
these models are denoted by asterisks in table [ doublecore : tab : dns ] ( models hp * and acc * ) . to calculate the birth rate of dns
, we assume a sn rate of 0.01 @xmath36 .
after a dns is formed , it emits gravitational - wave radiation causing the decay of the orbit .
we calculate the merger timescale of each system due to the emission of gravitational waves by applying the formula in peters ( 1964 ) .
if a system merges in less than the hubble timescale , which we assume to be 10 gyr , it contributes to the total merger rate of dnss .
in table [ doublecore : tab : dns ] we present the birth and merger rates of dnss for each simulation . with lower @xmath30 values ( @xmath37 ) ,
we obtain systems with smaller post - contact separations than applying @xmath33 , and hence we would expect fewer systems to survive the contact phase . in general , applying a lower minimum radius for case c mass transfer means allowing for more systems to be included as dns progenitors and hence we obtain higher birth and merger rates . .birth and merger rates of dnss , and potential rate of long - duration grbs for each simulation . also shown in the table are the disruption rates of dnss .
all rates are in units of @xmath0 .
the first column indicates the kick velocity distribution ( see text for explanation ) .
asterisks mark models applying low kick velocities to the second - formed neutron star .
the second and third columns present the @xmath30 value and the criterion for the minimum radius for case c mass transfer .
[ cols="<,^ , < , > , > , > , > " , ] it is still an open question whether black - hole formation is accompanied by a natal kick imparted to the black hole . the distribution of black - hole x - ray binaries in the galaxy may indicate that black holes on the whole do not receive significant kicks at birth ( white & van paradijs 1996 ) .
however , the high space velocity of nova sco strongly indicates a large natal kick in this particular system ( brandt , podsiadlowski & sigurdsson 1995 ) .
the recent study by jonker & nelemans ( 2004 ) also supports the idea of a substantial natal kick for the majority of black holes . in this study , we apply two assumptions for the kick velocity imparted to the black hole at birth ; one is that black holes are formed without a kick velocity ( @xmath38 ) , and the other that a kick velocity of the same magnitude as in the case of neutron - star formation ( @xmath39 @xmath35 ) is imparted . in the case of almost twin binaries passing through a contact phase
, the more massive primary always collapses first with a more massive core .
consequently , in the case of bh - ns binaries , the neutron star is formed later and does not experience any episode of mass transfer , i.e. it is not recycled and can only be observable as a young pulsar .
studies on the formation of recycled pulsar with black - hole companion have been carried out by , e.g. , voss & tauris ( 2003 ) , sipior , portegies zwart & nelemans ( 2004 ) , and recently pfahl et al .
( 2005 ) .
table [ doublecore : tab : bhns ] shows the birth rates of bh - ns and black - hole / black - hole ( bh - bh ) binaries .
all bh - ns and bh - bh systems merge within a hubble time , and hence the numbers in the table represent both the birth and merger rates .
the rates of bh - ns binaries are significantly lower than those of dnss .
the reason is that only a very narrow range of primary masses , i.e. @xmath40 @xmath32 can produce bh - ns binaries in the framework of the double - core scenario .
the trend found in dnss , in which birth and merger rates are the highest for the combination of the low case - c criterion and the high @xmath30 values , and are the lowest for the high case - c criterion and the high @xmath30 value , is also applicable to bh - ns binaries .
the case where the black - hole mass is equal to its immediate progenitor mass yields higher rates , by typically a factor of 1.5 , than the case where mass is lost in the black - hole collapse .
however , when we apply zero kicks for black - hole formation , the rates for the two cases of black - hole mass are rather similar .
one would expect to form more bh - ns or bh - bh binaries than ns - bh systems ( in which the neutron stars are formed before the black holes ) because one requires fine tuning in the initial masses and the orbital period distribution to allow the initially more massive component to have a less massive core at the end of its evolution .
the formation of ns - bh binaries is similar to pulsars b2303 + 46 and j11416545 , in which white - dwarf companions are formed before the pulsars ( tauris & sennels 2000 ; davies , king & ritter 2002 ) . indeed , sipior et al .
( 2004 ) found that the birth rate of bh - ns is about 27 higher than ns - bh .
this is about 10 times higher than the ratio obtained by voss & tauris ( 2003 ) .
the birth rates of bh - ns binaries from our simulations are typically comparable or lower than the ns - bh rates found by voss & tauris ( 2003 ) or pfahl et al .
the latter authors applied more or less similar assumptions of black - hole masses , kick velocities and @xmath30 values .
the comparable rates between our bh - ns result and their ns - bh binaries is because bh - ns binaries formed by the double - core channel originate from a very narrow mass range , as opposed to the formation of bh - ns through the standard scenario which allows a much wider range of mass to form black hole .
when the helium star in a hes - ns binary collapses into a neutron star , it is possible that the system may not survive the sn explosion and become unbound ; in this case the two neutron stars become runaway neutron stars .
the second neutron star , i.e. the descendant of the helium star , will be observed as a young , single radio pulsar , while the first - born neutron star can be observed as a single , recycled pulsar .
pulsars j2235 + 1506 ( @xmath41 55.7 ms ; camilo , nice & taylor 1993 ) and j0609 + 2130 ( @xmath41 59.7 ms ; lorimer et al .
2004 ) are two single pulsars with characteristics resembling those of dnss , and thus have been proposed to have originated from disrupted dnss . adopting the commonly accepted high kick velocity at neutron star birth , lorimer et al .
( 2004 ) suggested that we should observe more disrupted recycled pulsars than dnss .
table [ doublecore : tab : dns ] displays the disruption rates which , in general , are indeed higher than the birth rates in models hp and acc .
the disruption rates are significantly higher than the birth rates in models using the high case - c criterion and high @xmath30 values because the systems are in relatively wide orbits and hence are more likely to become unbound by the sn .
the difference between disruption and birth rates are higher in simulations using acc kick models due to the large kick velocities .
when we apply a low kick velocity at the second sn , the disruption rates are significantly reduced .
the fact that we observe seven dnss and only two disrupted dnss is consistent with the suggestion that kick velocities imparted at the birth of the second neutron stars in dnss are lower than commonly thought .
we can derive the runaway velocities of the two neutron stars as in appendix b in pfahl , rappaport & podsiadlowski ( 2002 ) .
for the first - born neutron stars , we can calculate the spin period as in dewi et al .
( 2005 ) and plot them against the runaway velocities .
[ doublecore : fig : faildns ] presents this plot for models acc and acc*. as a consequence of the high kick velocities , disrupted dnss from model acc have significantly higher runaway velocities .
the interesting difference can be seen in the distribution of spin periods ; while the spin periods of model acc extends to periods as short as 10 ms , only single , recycled pulsars with spin period larger than 0.2 s are produced by model acc*. these systems originate from hes - ns binaries with relatively high - mass helium stars which transfer little amount of mass ( hence the large spin period , see the discussion in dewi et al .
( 2005 ) ) .
this rather striking difference may allow us in principle to constrain the kick velocity distribution for the second supernova kick from the spin period distribution of single recycled pulsars from disrupted dnss .
we have performed monte carlo simulations for the formation of dnss in the double - core scenario . we found merger rates in this channel of 0.1 12 @xmath0 .
these rates are subject to substantial uncertainties in the binary input physics .
although it is clear that model hp provides larger birth and merger rates than model acc , we found that the rates are not very sensitive to the choice of kick velocity as both kick models yield rates of the same order of magnitude .
the rates are sensitive to the kick velocity only for models hp * and acc * where we assumed a much lower kick velocity for the second supernova .
the birth and merger rates depend mainly on the modelling of the contact phase . in this study
we limit the investigation to a combination of @xmath30 values and the minimum radius for case c mass transfer .
although we include primary masses between 8 and 25 @xmath32 , we found that the dnss mainly originate from primary stars in the range of @xmath31 12 23 @xmath32 .
the @xmath30 values in this mass range depend sensitively on the stellar mass , but in any case are always substantially lower than 0.5 , a standard value used in the previous literature .
our adopted value of @xmath37 , even though too low for stars less massive than @xmath31 14 @xmath32 and too high for @xmath4217 @xmath32 , represents the average @xmath30 value for stars in the range of @xmath31 8 25 @xmath32 and hence should provide a more reasonable value than @xmath33 at least for the primary stars . throughout our simulations
we apply the same @xmath30 for the secondary as for the primary . for stars of @xmath31 16 18 @xmath32 ,
the @xmath30 value is relatively constant for a wide range of radius , then increases steeply once the star had finished core helium burning . since we assume that the primary is in a more advanced stage of evolution and , moreover for this particular study , the primary is at least at the end of the core helium - burning phase , using the same @xmath30 value for both the primary and secondary is not always the best approach . estimating the @xmath30 value for the secondary relative to the value for the primary requires knowledge of the secondary radius at the onset of mass transfer phase and modelling of the early mass - transfer phase , which we were not able to investigate in this study .
using different minimum radii for case c mass transfer , which simulates the effects of different evolutionary post - main - sequence evolutionary tracks , is one of the major sources of uncertainty for deriving the birth and merger rates of dnss .
stars less massive than @xmath31 14 @xmath32 generally go through a blue loop and hence the high case - c criterion fits better for primary stars in this range of mass .
the situation is not so certain for more massive stars .
knowledge of the primary and secondary radii at the onset of the contact phase is very important for determining the outcome of the contact phase since the radii determine the envelope binding energy ( and hence the appropriate value of @xmath30 )
. it will require detailed binary evolution calculations to overcome the uncertainties on the radii and @xmath30 values , which we leave for further study .
even when we know the stellar structure through detailed modelling , we are still left with one important uncertainty , i.e. how to model the outcome of a contact phase . in this study we apply the so - called standard prescription in which the final separation is determined by the envelope binding energy . it is worth noting that our results depend on the initial distributions of the binary parameters , in particular the mass - ratio distribution .
if we bias the initial mass - ratio distribution more towards equal masses of both components , e.g. by simply restricting the mass ratio to a uniform distribution between 0.5 and 1 , the resulting rates in table [ doublecore : tab : dns ] are increased by a factor of @xmath43 . the double ce scenario studied by bethe & brown ( 1998 ) yielded a merger rate of @xmath44 for dnss and @xmath45 for bh - ns binaries , which tends to be somewhat higher than most of the rates obtained in this study .
we suspect that this huge difference is due to our constraint of only including primaries which go through case c mass transfer .
we argue that our additional constraint is required to explain the presence of recycled pulsars .
after discussing the possible uncertainties involved in this study , we will now compare our results with those derived from the standard scenario .
dpvdh found birth rates of 0.95 87 @xmath0 , comparable and somewhat larger than this study .
however , comparison between our results and those of dpvdh is not straightforward as the two studies are done with a very different approach .
dpvdh use the bse binary population synthesis code of hurley , tout & pols ( 2002 ) which , among other factors , includes the loss of angular momentum due to stellar wind mass loss in the orbital evolution .
our double - core study uses a more crude approach .
if we also take into account the wind mass loss in our simulations , we suspect that we will have smaller rates .
this is because wind mass loss drives the expansion of orbital separation such that some systems may not experience mass transfer phase at all .
wind mass loss depends strongly on stellar luminosity ( e.g. de jager , nieuwenhuijzen & van der hucht 1988 ; nugis & lamers 2000 ) and hence for high mass stars investigated in this study is very important .
dpvdh also applied actual @xmath30 value which depends on the evolutionary stage of the star , while we leave it as a constant value .
nevertheless , dpvdh also found that the birth and merger rates are more sensitive to the modelling of the mass transfer phase preceeding the formation of the hmxbs rather than to the choice of the kick velocity distribution .
this suggests that modelling of binary evolution of the stage prior to hes - ns binaries is very crucial to provide the best merger rates of dnss .
jdmd acknowledges a talent fellowship from the netherlands organization for scientific research ( nwo ) for her stay at the university of oxford where this work was initiated , and an ioa theory rolling grant from pparc .
armitage p. j. , livio m. , 2000 , apj 532 , 540 arzoumanian z. , chernoff d. f. , cordes j. m. , 2002 , apj 568 , 289 belczyski k. , kalogera v. , 2001 , apj 550 , l183 bethe h. a. , brown g. e. , 1998 , apj 506 , 780 bhattacharya d. , van den heuvel e. p. j. , 1991 , phys .
203 , 1 brandt w. n. , podsiadlowski ph . , 1995 , mnras 274 , 461 brandt w. n. , podsiadlowski ph . , sigurdsson s. , 1995 , mnras 277 , l35 brown g. e. , 1995 , apj 440 , 270 camilo f. , nice d. j. , taylor j. h. , 1993 , apj 412 , l37 chaurasia h. k. , bailes m. , 2005 , apj 632 , 1054 chevalier r. a. , 1993 , apj 411 , l33 chevalier r. a. , 1996 , apj 459 , 322 davies m. b. , ritter h. , king a. , mnras 335 , 369 de jager c. , nieuwenhuijzen h. , van der hucht k. a. , 1988 , a&as 72 , 259 dewi j. d. m. , podsiadlowski ph . , pols o. r. , 2005 , mnras 363 , l71 dewi j. d. m. , pols o. r. , 2003 , mnras 344 , 269 dewi j. d. m. , pols o. r. , savonije g. j. , van den heuvel e. p. j. , 2002 , mnras , 331 , 1027 dewi j. d. m. , tauris t. m. , 2000 , a&a 360 , 1043 dewi j. d. m. , tauris t. m. , 2001 , asp conf .
229 , 255 eggleton p. p. , 1983 ,
apj 268 , 368 fryer c. l. , heger a. , 2005 , apj 623 , 302 habets g. m. h. j. , 1986 , a&a 167 , 61 hansen b. m. s. , phinney e. s. , 1997 , mnras 291 , 569 hobbs g. , lorimer d. r. , lyne a. g. , kramer m. , 2005 , mnras 360 , 974 hurley j. r. , tout c. a. , pols o. r. , 2002 , mnras 329 , 897 jonker p. g. , nelemans g. , 2004 , mnras 354 , 355 langer n. , maeder a. 1995 , a&a , 295 , 685 lorimer d. r. , et al . , 2004 , mnras 347 , l21 nelson c. a. , eggleton p. p. , 2001
, apj 552 ,
664 nugis t. , lamers h. j. g. l. m. , 2000 , a&a 360 , 227 peters p. c. , 1964 , phys . rev .
136 , 1224 pfahl e. , rappaport s. , podsiadlowski ph . , 2002 ,
apj 573 , 283 pfahl e. , podsiadlowski ph . , rappaport s. , 2005 , apj 628 , 343 podsiadlowksi ph . , joss p. c. , hsu j. j. l. , 1992 , apj 391 , 246 podsiadlowski ph . , mazzali p. a. , nomoto k. , lazzati d , cappellaro e. , 2004 , apj 607 , l17 pols o. r. , 1994 , a&a 290 , 119 pols o. r. , dewi j. d. m. , 2002 , pasp 19 , 233 sipior m. s. , portegies zwart s. , nelemans g. , 2004 , mnras 354 , l49 taam r. e. , king a. r. , ritter h. , 2000 , apj 541 , 329 tauris t. m. , dewi j. d. m. , 2001 , a&a 369 , 170 tauris t. m. , sennels t. , 2000 , a&a 355 , 236 tauris t. m. , van den heuvel e. p. j. , savonije g. j. , 2000 , apj 530 , l93 voss r. , tauris t. m. , 2003 , mnras 342 , 1169 webbink r. f. , 1984 , apj 277 , 355 webbink r. f. , 2003 , asp conf
293 , 76 wellstein s. , langer n. , 1999 , a&a 350 , 148 wellstein s. , langer n. , braun h. , 2001 , a&a 369 , 939 wettig t. , brown g. e. , 1996 , new astr . 1 , 17 white n. e. , van paradijs j. , 1996 , apj 473 , l25 woosley s. e. 1993 , apj , 405 , 273 | we present the results of a systematic exploration of an alternative evolutionary scenario to form double neutron - star binaries , first proposed by brown ( 1995 ) , which does not involve a neutron star passing through a common envelope . in this scenario ,
the initial binary components have very similar masses , and both components have left the main sequence before they evolve into contact ; preferably the primary has already developed a co core .
we have performed population synthesis simulations to study the formation of double neutron star binaries via this channel and to predict the orbital properties and system velocities of such systems .
we obtain a merger rate for dnss in this channel in the range of 0.1 12 @xmath0 .
these rates are still subject to substantial uncertainties such as the modelling of the contact phase .
[ firstpage ] stars : evolution binaries : general stars : neutron |
in this article , we show the following result .
[ theo:0 ] let @xmath0 be a closed riemannian @xmath1-manifold with nonnegative ricci curvature .
there exists an embedded closed minimal hypersurface @xmath2 in @xmath0 with a singular set of hausdorff dimension at most @xmath3 such that @xmath4 where @xmath5 is an explicit positive constant depending only on @xmath1 . observe that both the ricci curvature condition and the inequality are scale invariant in this theorem .
we do not know whether the curvature condition can be dropped in dimension greater than two . in dimension two ,
it is the case according to ( * ? ? ?
* theorem 5.3.1 ) , @xcite and @xcite : every closed riemannian surface has a closed geodesic of length bounded from above in terms of the area of the surface .
thus , theorem [ theo:0 ] can be viewed as a partial generalization of this result on closed geodesics on surfaces in terms of minimal hypersurfaces .
another way to generalize this result would be to find an upper bound on the length of the shortest closed geodesic in a closed riemannian manifold in terms of the volume .
except for essential manifolds , where systolic inequalities hold @xcite , and the two - sphere @xcite , this question is completely open , even on manifolds with nonnegative ricci curvature .
the condition on the size of the singular set is customary in this context . to our knowledge , it is still unknown if there exists a closed _ smooth _ minimal hypersurface in every closed riemannian manifold even with nonnegative ricci curvature .
the existence of a minimal hypersurface in theorem [ theo:0 ] actually derives from a min - max principle due to almgren - pitts in geometric measure theory ( see @xcite ) .
when the ricci curvature is positive , the geometry of the min - max minimal hypersurface @xmath2 has recently been described in @xcite .
in fact , theorem [ theo:0 ] is a consequence of the following sweep - out estimate .
[ theo : a ] let @xmath0 be a closed riemannian @xmath1-manifold with nonnegative ricci curvature .
there exists a pl morse function @xmath6 such that @xmath7 where @xmath5 is an explicit positive constant depending only on @xmath1 .
recall that a function @xmath6 is a pl morse function if there exists a simplicial complex structure on @xmath0 such that @xmath8 is linear on every simplex of @xmath0 and takes pairwise distinct values at the vertices of @xmath0 . as already mentioned , the sweep - out estimate of theorem [ theo : a ] is stronger than theorem [ theo:0 ] .
a slightly more general version is given by theorem [ theo : main ] .
observe again that both the ricci curvature condition and the inequality are scale invariant .
contrary to the previous theorem where the question is open , this result fails without any curvature condition for every closed @xmath1-manifold with @xmath9 ( see proposition [ prop : cex ] ) .
however , the curvature assumption can be dropped for closed riemannian surfaces @xcite . in this case , the multiplicative constant depends on the genus of the surface ( and it has to ) .
none of the multiplicative constants in theorems [ theo:0 ] or [ theo : a ] is optimal , even on two - dimensional convex spheres despite a local extremality result @xcite . for positively curved @xmath10-manifolds ,
related sharp upper bounds have been obtained in @xcite . when @xmath0 is a smooth convex hypersurface in a euclidean @xmath1-space , the result of theorem [ theo : a ] follows from @xcite .
a similar result also holds true for domains of the euclidean @xmath1-space ( see @xcite ) .
thus , our sweep - out estimate can be seen as a partial generalization of these results to a non - euclidean setting .
actually , theorem [ theo : a ] can be formulated in terms of a min - max estimate by taking the infimum over all pl morse functions of the supremum of the volume of their fibers .
further consideration of min - max principles can be found in @xcite .
note that min - max processes are also involved in concentration phenomena .
a related estimate holds for manifolds with nonnegative sectional curvature by replacing the volume of the fibers by the diameter ( see @xcite ) . as a consequence of our sweep - out estimate
, we also derive the following isoperimetric inequality .
[ coro : decomp ] there exists a positive constant @xmath5 such that every closed riemannian @xmath1-manifold @xmath0 with nonnegative ricci curvature decomposes into two connected domains with the same volume whose common boundary @xmath2 satisfies @xmath11 to conclude this introduction , we mention some recent applications of the min - max process on three - dimensional manifolds to emphasize its importance in current research .
( the most recent results mentioned below even appeared after this paper was submitted for publication . ) in @xcite , t. colding and w. minicozzi established min - max estimates through the ricci flow which allowed them to simplify some arguments in perelman s proof of the poincar conjecture through a more conceptual approach . in @xcite and @xcite , the authors established genus bound for minimal surfaces constructed via min - max arguments on every closed riemannian @xmath10-manifold .
recently , f. marques and a. neves proved the willmore conjecture @xcite and a conjecture of yau about the existence of infinitely many minimal hypersurfaces in closed riemannian manifolds with positive ricci curvature @xcite by using a min - max process la almgren - pitts .
we also mention that p. glynn - adey and y. liokumovich recently placed a preprint on arxiv @xcite in which they establish theorem [ theo:0 ] among other results .
the techniques seem somewhat different though they also rely on sweep - out estimates . throughout this paper ,
a domain of a complete @xmath1-manifold with piecewise smooth boundary is an @xmath1-submanifold with piecewise smooth boundary .
in this section , we introduce some key notions for the rest of the article and establish preliminary results which will be used later . the definition of a cw complex can be found in ( * ? ? ? * appendix ) along with some related topological properties and applications .
we simply recall that every @xmath12-cell of a cw complex @xmath13 is attached to the @xmath14-skeleton @xmath15 of @xmath13 . the map from this @xmath12-cell to @xmath13
is referred to as the characteristic map of the @xmath12-cell .
we will denote by @xmath16 the ( open ) @xmath17-ball centered at the origin of @xmath18 .
recall also that a @xmath19-quasi - isometry between two metric space is a homeomorphism which is @xmath19-lipschitz and whose reciprocal map is also @xmath19-lipschitz .
the geometry of a cw complex structure can be measured through the following notion .
[ def : cellstruc ] let @xmath20 and @xmath21 .
a compact riemannian @xmath1-manifold @xmath22 with ( possibly empty ) piecewise smooth boundary admits a cw complex structure with quasi - convex cells of size @xmath23 if 1 .
there exists a @xmath19-quasi - isometry @xmath24 between a piecewise flat simplicial @xmath1-complex @xmath25 and @xmath22 .
[ i ] 2 .
the simplicial @xmath1-complex @xmath25 decomposes into @xmath1-cells @xmath26 formed of a union of simplices @xmath27 such that each @xmath1-cell @xmath28 is @xmath19-quasi - isometric to a convex euclidean polyhedron @xmath29 lying in @xmath30 and containing @xmath31 .
this quasi - isometry is denoted by @xmath32 .
[ ii ] 3 . for every @xmath12 ,
the composite @xmath33 is @xmath19-quasi - isometric .
[ iii ] the decomposition of @xmath25 into @xmath1-cells @xmath28 gives rise to a cw complex structure on @xmath22 where the characteristic maps are given by the quasi - isometries @xmath34 described in .
for our purpose , we can think of @xmath22 as an @xmath1-polyhedron whose @xmath1-faces are quasi - isometric to convex euclidean polyhedra . despite the risk of confusion
, we will often identify the @xmath1-cells of @xmath22 with their images by the characteristic maps .
the @xmath1-simplices composing the piecewise flat simplicial @xmath1-complex @xmath25 are not necessarily regular as their edges may have different lengths .
however the @xmath1-cells of @xmath25 have a uniform size , roughly the size of @xmath16 , while still being almost convex .
( this is the reason why we introduce this definition . ) in particular , the volume of the @xmath1-cells is between @xmath35 and @xmath36 , where @xmath37 represents the volume of the unit ball centered at the origin of @xmath18 .
thus , if @xmath38 represents the number of @xmath1-cells in @xmath22 , we immediately obtain @xmath39 the following result shows the existence of cw complex structures with quasi - convex cells of small size on closed riemannian manifolds . [
prop : existence ] let @xmath22 be a bounded domain of a complete riemannian @xmath1-manifold @xmath0 .
for every @xmath40 small enough , the domain @xmath22 admits a cw complex structure with quasi - convex cells of size @xmath41 with @xmath42 . without loss of generality
, we can assume that the boundary of @xmath22 is smooth ( otherwise we smooth it out ) .
let @xmath43 with @xmath17 smaller than half the focal radius of @xmath44 .
let @xmath45 and @xmath46 .
consider a maximal system of disjoint @xmath47-balls of @xmath0 centered in @xmath48 and denote by @xmath49 the centers of these balls .
note that @xmath50 lies in @xmath22 .
since the system of balls is maximal , the balls @xmath51 cover @xmath48 .
as @xmath17 is smaller than half the focal radius of @xmath44 , every point of @xmath44 is at distance @xmath47 from some point of @xmath52 .
thus , the balls @xmath53 cover the @xmath47-neighborhood @xmath54 of @xmath44 in @xmath22 .
now , we complete the collection of disjoint balls @xmath50 of @xmath0 lying in @xmath22 into a maximal system of disjoint @xmath47-balls of @xmath0 lying in @xmath22 .
we will still denote by @xmath49 the centers of the balls thus obtained .
the voronoi cell @xmath55 centered at @xmath49 clearly contains the ball @xmath50 .
it also lies in the ball @xmath53 .
otherwise , we could find a point @xmath56 at distance at least @xmath57 from any point @xmath58 . from the previous paragraph , this implies that the point @xmath59 is at distance at least @xmath17 from @xmath44 .
thus , the ball @xmath60 lies in @xmath22 and is disjoint from the other balls @xmath61 .
this contradicts the construction of the points @xmath49 .
a consequence of these inclusions and the gauss lemma is that the preimage @xmath62 of @xmath63 by the exponential map @xmath64 based at @xmath49 lies in @xmath30 and contains @xmath31 .
this preimage is quasi - isometric to the convex polyhedral voronoi cell @xmath65 formed of the vectors @xmath66 of @xmath67 whose distance to the origin is less or equal to the distance from @xmath66 to the preimages of the @xmath58 under @xmath64 for every @xmath68 .
furthermore , the factor of the quasi - isometry @xmath69 tends to @xmath70 when @xmath17 goes to @xmath71 , _ cf .
_ definition [ def : cellstruc ] , item .
( in the following , all the quasi - isometries will satisfy this property . )
we would like to glue the convex euclidean polyhedra @xmath65 together to construct a simplicial @xmath1-complex @xmath25 quasi - isometric to @xmath22 .
two adjacent voronoi cells @xmath63 and @xmath72 in @xmath22 meeting along an @xmath73-face give rise to two convex euclidean polyhedra @xmath65 and @xmath74 along with two corresponding @xmath73-faces @xmath75 and @xmath76 in @xmath65 and @xmath74 .
if @xmath75 and @xmath76 were isometric , we could glue the convex euclidean polyhedra @xmath65 and @xmath74 together .
unfortunately , this is not always the case .
indeed , the @xmath73-faces @xmath75 and @xmath76 , though combinatorially equivalent , are only quasi - isometric . to get around this technical problem , we consider a quasi - geodesic triangulation of the convex euclidean polyhedra @xmath65 such that the induced triangulations on the voronoi cells @xmath63 in @xmath22 agree on their common faces .
this gives rise to a triangulation @xmath77 of @xmath22 ( refining the cw complex structure of @xmath22 given by the voronoi cells ) where each simplex @xmath78 is quasi - isometric to a euclidean simplex @xmath79 .
now , we need to deform these euclidean simplices @xmath79 in order to glue them together .
let @xmath80 be an edge of @xmath77 .
denote by @xmath81 , where @xmath82 , the euclidean simplices with an edge @xmath83 corresponding to @xmath80 . replace the edges @xmath83 of @xmath81 with an edge of the same length equal to the average of the lengths of @xmath84 .
as the edges @xmath83 are quasi - isometric to @xmath80 , the same goes for the new edges . by applying this averaging argument to every edge of the triangulation @xmath77
, we obtain new euclidean simplices @xmath85 quasi - isometric to the initial euclidean simplices @xmath79 and so to the simplex @xmath78 .
furthermore , the @xmath73-faces of these new euclidean simplices @xmath85 corresponding to common faces of adjacent simplices in @xmath22 are isometric since the lengths of their edges agree . therefore , we can replace each simplex @xmath78 in @xmath22 with a quasi - isometric euclidean simplex @xmath85 so that the resulting space is a simplicial @xmath1-complex @xmath25 quasi - isometric to @xmath22 , _ cf . _ definition [ def : cellstruc ] , item . by construction ,
the simplices of @xmath25 corresponding to the euclidean simplices of the triangulation of @xmath29 form an @xmath1-cell @xmath28 quasi - isometric to @xmath29 , _ cf . _ definition [ def : cellstruc ] , item . since the factors of all these quasi - isometries tends to @xmath70 when @xmath17 goes to @xmath71 , this yields the desired result .
we will also need the following result .
[ lem : proj ] fix @xmath86 and @xmath87 close enough to @xmath70 .
let @xmath88 be an @xmath1-cell of a cw complex structure with quasi - convex cells of size @xmath23 on a bounded domain of a compact riemannian @xmath1-manifold .
there exists a retraction @xmath89 of @xmath90 with lipschitz constant at most @xmath91 . by definition ,
the @xmath1-cell @xmath88 is @xmath19-quasi - isometric to a convex euclidean polyhedron @xmath92 lying in @xmath30 and containing @xmath31 .
the radial projection from @xmath93 to @xmath94 is lipschitz .
a sharp upper bound on its lipschitz constant is given by the limit from the right at @xmath95 of the length ratio @xmath96 , where the points @xmath97 are defined as follows , _ cf .
_ figure [ fig : proj ] .
* the coordinates of @xmath98 and @xmath99 are @xmath100 and @xmath101 . *
the point @xmath102 lies in the same vertical line as @xmath98 and satisfies @xmath103 .
* the point @xmath104 is the intersection point of @xmath105 with the line tangent at @xmath98 to the circle of radius @xmath17 centered at the origin .
a straightforward computation shows that @xmath106 , @xmath107 and @xmath108 where @xmath109 .
hence , @xmath110 we derive the desired lipschitz constant after pre- and post - composition with a @xmath19-quasi - isometry .
the goal of this section is to prove the following result .
[ theo : split1 ] let @xmath0 be a complete riemannian @xmath1-manifold with nonnegative ricci curvature . for every bounded domain @xmath22 of @xmath0 , there exists a smooth hypersurface @xmath2 of @xmath22 with boundary lying in @xmath44 which decomposes @xmath22 into two domains @xmath111 and @xmath112 with @xmath113 such that @xmath114 where @xmath115 and @xmath116 are explicit positive constants depending only on @xmath1 . before proceeding to the proof of this result ,
let us recall the bishop - gromov inequality , which plays a crucial role in the study of manifolds with nonnegative ricci curvature .
let @xmath0 be a complete riemannian @xmath1-manifold with nonnegative ricci curvature .
then for every @xmath117 and @xmath118 , we have @xmath119 in particular , the volume of every @xmath120-ball in @xmath0 is at most @xmath121 , where @xmath37 is the volume of a unit ball in the euclidean @xmath1-space .
a consequence of the bishop - gromov inequality is the following classical packing estimate .
[ lem : packing ] let @xmath122 .
every ball of radius @xmath123 in a complete riemannian @xmath1-manifold @xmath0 with nonnegative ricci curvature can be covered by @xmath124 balls of radius @xmath125 .
let @xmath126 be a ball of radius @xmath123 in @xmath0 .
consider a maximal system of disjoint @xmath127-balls in @xmath0 with centers @xmath49 lying in @xmath126 .
clearly , the balls @xmath128 are contained in @xmath129 . from the bishop - gromov inequality ,
the number of elements in such maximal system of disjoint balls is at most @xmath130 on the other hand , by maximality , the balls @xmath131 cover @xmath126 .
the following result about the existence of capacitors can be found in ( * ? ? ?
* lemma 2.2 ) ( see also @xcite ) .
we include a proof for the sake of completeness .
[ lem : capa ] let @xmath22 be a bounded domain of a complete riemannian @xmath0 with nonnegative ricci curvature .
let @xmath132 where @xmath124 . then there exist two domains @xmath133 and @xmath134 in @xmath22 with @xmath135 at distance at least @xmath125 from each other in @xmath0
. given a subset @xmath98 of @xmath0 , let @xmath136 fix @xmath137 . for @xmath138 , consider the function @xmath139 defined on the @xmath140-fold product @xmath141 of @xmath0 as @xmath142 since @xmath22 is bounded , the function @xmath143 attains its maximum at some point @xmath144 in @xmath141 .
clearly , @xmath145 from the bishop - gromov inequality and @xmath146 for @xmath140 large enough .
let @xmath147 be the smallest integer such that @xmath148 .
define @xmath149 .
note that @xmath150 we also have @xmath151 by lemma [ lem : packing ] , the set @xmath152 can be covered by @xmath153 balls of radius @xmath125 in @xmath0 .
the union of these @xmath125-balls decomposes into the union of @xmath124 sets of the form @xmath154 with @xmath155 . since @xmath156
we derive @xmath157 thus , for @xmath158 , we obtain @xmath159 by construction , we have @xmath160 . note
that we can also smooth out the boundary of @xmath161 .
we can now derive theorem [ theo : split1 ] .
let @xmath133 and @xmath134 be as in lemma [ lem : capa ] .
denote by @xmath162 the @xmath125-neighborhood of @xmath133 in @xmath22 .
note that @xmath134 lies in the complementary set of @xmath162 in @xmath22 .
define a lipschitz function @xmath163 as @xmath164 the function @xmath8 is not a morse function .
however , it can be approximated by a morse function @xmath165 with @xmath166 such that 1 .
@xmath167 if @xmath168 ; 2 .
@xmath169 if @xmath170 ; 3 .
@xmath171 . hence , from the coarea formula @xcite
, we obtain @xmath172 to avoid burdening the argument by epsilontics , we will assume that @xmath173 . thus , there exists a regular value @xmath174 such that @xmath175 the preimage @xmath176 is a smooth hypersurface of @xmath22 with boundary lying in @xmath44 which decomposes @xmath22 into two domains @xmath177)$ ] and @xmath178 .
since @xmath179 , we obtain @xmath180 the theorem follows with @xmath181 and @xmath182 .
in this section , we establish a version of theorem [ theo : split1 ] which preserves the cw complex structure of the manifold with quasi - convex cells .
[ theo : split2 ] fix @xmath20 and @xmath87 close enough to @xmath70 .
let @xmath0 be a complete riemannian @xmath1-manifold with nonnegative ricci curvature .
consider a bounded domain @xmath22 in @xmath0 endowed with a cw complex structure with quasi - convex cells of size @xmath23 .
suppose @xmath22 has more than one @xmath1-cell .
then there exists an @xmath73-chain @xmath2 of @xmath22 with boundary lying in @xmath44 which decomposes @xmath22 into two @xmath1-dimensional cw subcomplexes @xmath111 and @xmath112 with @xmath183 such that @xmath184 where @xmath185 and @xmath186 are explicit positive constants depending only on @xmath1 . for the sake of simplicity
, we will assume that @xmath187 , that is , the @xmath1-cells @xmath88 of the cw complex decomposition of @xmath22 are flat .
the general case only requires technical changes to keep track of the bi - lipschitz factor . from theorem [ theo : split1 ]
, the domain @xmath22 decomposes into two domains @xmath111 and @xmath112 such that their common boundary @xmath188 satisfies @xmath189 without loss of generality , we can assume that no connected component of @xmath111 and @xmath112 lies in a single @xmath1-cell of the cw complex structure of @xmath22 .
this can be achieved by translating such connected components if necessary .
we want to deform the hypersurface @xmath2 into the @xmath73-skeleton of the cw complex decomposition of @xmath22 while controlling its @xmath73-volume and the @xmath1-volume of the two domains of @xmath22 it bounds . by perturbing the hypersurface @xmath2 if necessary
, we can assume that it intersects transversally the @xmath73-cells of @xmath22 and decomposes its @xmath1-cells into finitely many domains .
thus , the hypersurface @xmath2 decomposes into finitely many pieces @xmath190 obtained as the connected components of the intersection of @xmath2 with the @xmath1-cells of the cw complex decomposition of @xmath22 .
for every @xmath191 , denote by @xmath192 the @xmath1-cell of @xmath22 in which @xmath193 lies . by construction , @xmath194 lies in @xmath195 .
we also have @xmath196 consider a volume - minimizing @xmath73-chain @xmath197 in @xmath192 with the same boundary as @xmath193 .
if @xmath197 is at distance at least @xmath17 from the center of @xmath192 , we replace @xmath193 with the image @xmath198 of @xmath197 by the projection @xmath199 given by lemma [ lem : proj ] .
the new @xmath73-chain @xmath198 lies in @xmath195 , has the same boundary as @xmath193 and satisfies @xmath200 from lemma [ lem : proj ] .
if some point @xmath201 of @xmath197 is at distance at most @xmath17 from the center of @xmath192 , we replace @xmath193 with a volume - minimizing @xmath73-chain @xmath198 in @xmath195 with the same boundary as @xmath193 .
note that @xmath202 . since @xmath197 is a minimal @xmath73-chain of @xmath192 with boundary lying in @xmath195 ,
the volume of the intersection of @xmath197 with the @xmath17-ball of @xmath192 centered at @xmath201 is at least @xmath203 from the monotonicity formula , _ cf . _
that is , @xmath204 hence , @xmath205 in both case , we obtain an @xmath73-chain @xmath198 in @xmath195 with the same boundary as @xmath193 such that @xmath206 where @xmath207 .
replacing @xmath193 with @xmath198 gives rise to an @xmath73-chain @xmath208 lying in the @xmath73-skeleton of the cw complex decomposition of @xmath22 , with boundary @xmath209 lying in @xmath44 .
this @xmath73-chain bounds two @xmath1-dimensional cw subcomplexes @xmath210 and @xmath211 in @xmath22 .
furthermore , the volume of @xmath212 is bounded from above as follows @xmath213 hence , the bound for @xmath214 .
( the reason for the second argument in the max will become clear later . ) it remains to bound from below the volume of the two cw subcomplexes @xmath210 and @xmath211 bounded by @xmath212 . by the isoperimetric inequality ,
the volume of the region @xmath215 of @xmath192 bounded by @xmath193 and @xmath198 satisfies @xmath216 where @xmath217 is the @xmath1-dimensional isoperimetric constant in @xmath18 .
+ since @xmath218 , we also have @xmath219 now , let @xmath220 be the set of all the indices @xmath191 such that @xmath221 and @xmath222 be the set of all the indices @xmath191 such that @xmath223 . by substracting up the volume change of @xmath224 through the replacement process for every @xmath1-cell @xmath192 of the cw complex decomposition of @xmath22 , we obtain @xmath225 where @xmath226 .
suppose the number @xmath38 of @xmath1-cells in the cw complex decomposition of @xmath22 is greater or equal to @xmath227 .
from the volume estimate , we have @xmath228 .
combined with the bound , this implies that @xmath229 thus , @xmath230 if @xmath38 is less than @xmath231 , we can take for @xmath210 any @xmath1-cell of @xmath22 and for @xmath211 the union of the remaining @xmath1-cells . in this case , still from , we deduce that @xmath232 where @xmath233 is the @xmath57-sphere centered at the origin of @xmath18 . by our choice of @xmath185 ,
the bound is still satisfied .
we also derive from that @xmath234 in both cases , the result follows by renaming @xmath235 into @xmath236 with @xmath237 . by the simplicial approximation theorem
, we can assume that @xmath2 is an @xmath73-dimensional simplicial chain of a simplicial subdivision of @xmath22 with boundary lying in @xmath44 which decomposes @xmath22 into two @xmath1-dimensional simplicial subcomplexes @xmath111 and @xmath112 satisfying the same geometric estimates .
for the next result , we need to introduce a min - max value for sweep - outs defined in terms of pl morse functions .
let @xmath22 be a piecewise flat simplicial @xmath1-complex homeomorphic to a compact @xmath1-manifold with ( possibly empty ) boundary .
a function @xmath163 is a _ pl morse function _ if it is linear in restriction to every simplex of @xmath22 and takes pairwise distinct values at the vertices of @xmath22 .
note that a pl morse function is uniquely determined by its values on the set of vertices of @xmath22 .
the @xmath73-volume width of a pl morse function @xmath163 is defined as @xmath238 similarly , the @xmath73-volume width of @xmath22 is defined as @xmath239 where @xmath8 runs over all pl morse functions on some simplicial subdivision of @xmath22 . when @xmath22 is a compact riemannian @xmath1-manifold with ( possibly empty ) piecewise smooth boundary and no fixed underlying simplicial complex structure , we extend the notion of pl morse function as follows .
a function @xmath163 is a _ ( generalized ) pl morse function _ if there exists a simplicial complex structure on @xmath22 for which @xmath8 is a pl morse function .
[ def : width ] let @xmath22 be a compact riemannian @xmath1-manifold with ( possibly empty ) piecewise smooth boundary .
the @xmath73-volume width of a morse function @xmath163 is defined as @xmath238 similarly , the @xmath73-volume width of @xmath22 is defined as @xmath239 where @xmath8 runs over all morse functions on @xmath22 . the following estimate on the @xmath73-volume width of a bounded domain in terms of the @xmath73-volume widths of the subdomains it is made of results from a cut - and - paste argument . [ prop : merging ] let @xmath22 be a piecewise flat simplicial @xmath1-complex homeomorphic to a compact @xmath1-manifold with ( possibly empty ) boundary .
let @xmath2 be an @xmath73-chain of @xmath22 with boundary lying in @xmath44 which decomposes @xmath22 into two @xmath1-dimensional simplicial subcomplexes @xmath111 and @xmath112 .
then @xmath240 with extra care in the proof , it seems likely that the @xmath241 factor can be dropped in the conclusion of proposition [ prop : merging ] .
the boundary of the @xmath73-chain @xmath2 decomposes @xmath44 into two simplicial subcomplexes @xmath242 and @xmath243 with @xmath193 lying in @xmath244 .
therefore , the boundary of @xmath245 decomposes into @xmath246 .
let @xmath247 be a pl morse function defined on a simplicial subdivision of @xmath245 . by adding constants to @xmath248 and @xmath249 if necessary
, we can assume that @xmath250 we want to construct a pl morse function @xmath251 defined on a simplicial subdivision of @xmath22 by deforming @xmath248 and @xmath249 in the neighborhood of @xmath2 into two other pl morse functions which agree on @xmath2 .
the new pl morse function @xmath251 should satisfy @xmath252 for every @xmath253 , where @xmath254 is an arbitrarily small error term .
let @xmath255 small enough . from @xcite
, we can assume that the simplicial subdivisions on @xmath111 and @xmath112 arise from the same simplicial subdivision of @xmath22 .
the @xmath1-simplices of @xmath245 meeting @xmath2 form a ( closed ) neighborhood @xmath256 of @xmath2 in @xmath245 .
let @xmath80 be an edge of this neighborhood with exactly one endpoint , say @xmath257 , lying in @xmath2 .
denote by @xmath258 the point of @xmath80 such that @xmath259 .
note that for @xmath260 small enough , we can assume that the values taken by @xmath261 at the points @xmath258 are pairwise distinct .
now , we truncate every @xmath1-simplex @xmath78 of @xmath256 along the hyperplane passing through the points @xmath258 where @xmath80 is an edge of @xmath78 .
this gives rise to a cw complex structure of @xmath256 ( finer than the one given by the initial triangulation ) whose @xmath1-cells are convex euclidean polyhedra , namely truncated @xmath1-simplices .
denote by @xmath262 the ( closed ) neighborhood of @xmath2 in @xmath256 formed of the @xmath1-cells ( _ i.e. _ , truncated @xmath1-simplices ) of @xmath245 meeting @xmath2 .
every @xmath1-cell of @xmath262 meets @xmath2 along a @xmath12-simplex with @xmath263 .
an @xmath1-cell @xmath88 of @xmath262 is said to be _ big _ if it meets @xmath2 along an @xmath73-simplex @xmath264 and _ small _ otherwise .
observe that every hyperplane intersects a small @xmath1-cell ( viewed as a convex euclidean polyhedron ) of @xmath262 along a region of @xmath73-volume at most @xmath265 , where @xmath266 when @xmath260 goes to zero .
similarly , every hyperplane intersects a big @xmath1-cell @xmath88 of @xmath262 along a region of @xmath73-volume at most @xmath267 , where @xmath268 when @xmath260 goes to zero . without introducing new vertices ,
we define a new triangulation of @xmath245 by subdividing the @xmath1-cells ( convex euclidean polyhedra ) of @xmath256 into @xmath1-simplices . in the process ,
big @xmath1-cells ( which have exactly @xmath241 vertices ) are split into @xmath1 simplices of maximal dimension and small @xmath1-cells ( which have at most @xmath269 vertices ) are split into at most @xmath269 simplices of maximal dimension .
we define a pl morse function @xmath251 with respect to the new triangulation of @xmath22 which agrees with @xmath261 on the vertices of @xmath270 and takes the vertices of @xmath2 to pairwise distinct values close to zero in @xmath271 .
the level sets of @xmath272 are transverse to the @xmath73-faces of the triangulation of @xmath22 and intersect each @xmath1-simplex of @xmath273 along a ( possibly empty ) hyperplane .
the volume estimates on these hyperplanes and the bounds on the maximal number of @xmath1-simplices in an @xmath1-cell yield the following estimate @xmath274 where the first sum is over the @xmath1-simplices lying in a big @xmath1-cell of @xmath275 , the second sum is over the @xmath1-simplices lying in a small @xmath1-cell of @xmath275 and @xmath276 is the number of @xmath1-simplices lying in a small @xmath1-cell of @xmath275 .
the level sets of @xmath272 satisfy the following properties for different values of @xmath277 .
for @xmath278 , the level set @xmath279 lies in @xmath111 and agrees with @xmath280 , while for @xmath281 , it lies in @xmath112 and agrees with @xmath282 .
finally , for @xmath283 $ ] , the level set @xmath284 lies in @xmath275 . from this , we immediately derive @xmath285 combined with , we obtain @xmath286 the desired result follows by letting @xmath260 go to zero , namely @xmath287 the boundary of the @xmath73-chain @xmath2 decomposes @xmath44 into two parts @xmath242 and @xmath243 with @xmath193 lying in @xmath244 .
therefore , the boundary of @xmath245 decomposes into @xmath246 .
let @xmath247 be a morse function .
we want to deform @xmath261 in the neighborhood of @xmath2 into another morse function @xmath288 $ ] such that the level set @xmath289 agrees with @xmath2 and @xmath290 for every @xmath291 $ ] , where @xmath254 is an arbitrarily small error term . without loss of generality
, we can assume that @xmath261 has no critical point in the neighborhood of @xmath244 , that its level sets transversally meet @xmath244 and that the range of @xmath261 lies between @xmath71 and @xmath70 . consider a ( closed ) neighborhood of @xmath2 in @xmath245 quasi - isometric to @xmath292 $ ] with @xmath293 small enough .
up to reparametrisation , we will identify this neighborhood with @xmath294 $ ] and @xmath2 with @xmath295 , _ cf .
_ figure [ fig : boundary ] .
now , let us define @xmath288 $ ] as follows : outside @xmath294 $ ] , the function @xmath296 agrees with @xmath261 , while on @xmath294 $ ] , it is given by @xmath297 for every @xmath298 $ ] . since @xmath261 is nonnegative , the functions @xmath261 and @xmath296 agree on @xmath299 .
furthermore , by construction , we have @xmath300}(t ) \cup \left ( ( f_i)^{-1}_{s \times \{0\}}(t ) \times [ 0,t ] \right ) \cup \left ( ( f_i)^{-1}_{s \times \{0\}}([0,t ] ) \times \ { t \ } \right)\ ] ] for every @xmath291 $ ] , _ cf .
_ figure [ fig : boundary ] . by taking @xmath260 sufficiently small
, the @xmath73-volume of the second term of @xmath301 , namely @xmath302 $ ] , can be made arbitrarily small .
similarly , the @xmath73-volume of the last term of @xmath301 , namely , is bounded from above by the @xmath73-volume of @xmath2 up to an arbitrarily small error term .
now , we can perturb @xmath296 into a morse function , still denoted by @xmath296 , with values in @xmath303 $ ] , without increasing too much the @xmath73-volume of its level sets while making sure that @xmath304 . by construction ,
we have @xmath305 let @xmath163 be a morse function .
consider the functions @xmath288 $ ] previously constructed from the restrictions @xmath261 of @xmath8 to @xmath245 .
note that both @xmath306 and @xmath307 ( and so @xmath308 ) are constant equal to @xmath70 on @xmath2 .
define @xmath251 as @xmath309 slightly perturbing @xmath272 in the neighborhood of @xmath2 , we can assume that it is a morse function . by construction
, we have @xmath310 for some arbitrarily small error term @xmath254 , which yields the desired result .
we can now prove the main theorem of this article .
[ theo : main ] let @xmath0 be a complete riemannian @xmath1-manifold with nonnegative ricci curvature . for every bounded domain @xmath22 in @xmath0 , there exists a pl morse function @xmath163 such that @xmath311 where @xmath5 is an explicit positive constant depending only on @xmath1 . when @xmath0 is a closed riemannian @xmath1-manifold and @xmath22 agrees with the whole manifold @xmath0 , we recover theorem [ theo : a ] from the introduction .
fix @xmath20 small enough so that there exists a cw complex structure on @xmath22 with quasi - convex cells of size @xmath23 with @xmath87 close enough to @xmath70 , _ cf . _
proposition [ prop : existence ] . as in the proof of theorem [ theo : split2
] , we will assume that @xmath187 for the sake of simplicity .
that is , the quasi - convex @xmath1-cells of the cw complex structure of @xmath22 are flat .
in particular , @xmath22 is a piecewise flat simplicial @xmath1-complex .
we argue by induction on the number of quasi - convex @xmath1-cells in the cw complex structure of @xmath22 .
suppose that @xmath22 is composed of a single @xmath1-cell .
every polyhedron lying in @xmath30 can be swept out by parallel hyperplanes of volume at most @xmath312 , which correspond to the level sets of some pl morse function .
now , the volume of @xmath22 is greater or equal to @xmath313 .
thus , @xmath314 in the general case , from theorem [ theo : split2 ] , there exists an @xmath73-chain @xmath2 of @xmath22 with boundary lying in @xmath44 which decomposes @xmath22 into two cw subcomplexes @xmath111 and @xmath112 with @xmath315 such that @xmath316 both @xmath111 and @xmath112 have fewer @xmath1-cells in their cw complex decomposition than @xmath22 .
therefore , we can assume that @xmath317 otherwise the result follows by induction .
thus , since @xmath318 , we deduce that @xmath319 combined with proposition [ prop : merging ] and the bound , we obtain @xmath320 hence , the theorem holds with @xmath321 .
given a closed riemannian @xmath1-manifold @xmath0 , we deduce from the min - max principle of @xcite the existence of an embedded closed minimal hypersurface in @xmath0 with a singular set of hausdorff dimension at most @xmath3 and volume at most @xmath322 . combined with theorem [ theo : a ] , this yields theorem [ theo:0 ] .
+ similarly , corollary [ coro : decomp ] is a consequence of theorem [ theo : a ] . from theorem [ theo :
a ] , a closed riemannian manifold @xmath0 with nonnegative ricci curvature and dimension @xmath9 decomposes into two non - necessarily connected domains @xmath111 and @xmath112 with the same volume such that @xmath323 . slightly perturbing the domains @xmath245 and connecting their connected components with thin tubes if necessary
, we can assume that the domains @xmath245 are connected . in dimension two
, it follows from @xcite that any two - sphere with nonnegative curvature can be swept out by a one - parameter family of disjoint simple loops of length at most the length of the shortest closed geodesic .
since the length of the shortest closed geodesic is bounded in terms of the area of the sphere , we immediately obtain the desired decomposition .
in this section , we show that the curvature condition in theorem [ theo : a ] can not be dropped in general . let @xmath0 be a closed riemannian @xmath1-manifold .
fix @xmath324 if @xmath0 is orientable and @xmath325 otherwise .
denote by @xmath326 the @xmath12-cycle space of @xmath0 over @xmath327 , that is , the space of @xmath12-dimensional integral currents of @xmath0 over @xmath327 with zero boundary .
the mass of a @xmath12-cycle @xmath328 defined as @xmath329 extends the notion of volume for @xmath12-submanifolds .
the homotopy groups of the @xmath12-cycle space endowed with the flat norm topology have been determined by f. almgren @xcite .
more precisely , there is a natural isomorphism between @xmath330 and @xmath331 . in particular , @xmath332 this isomorphism allows us to apply the almgren - pitts min - max principle to the @xmath73-cycle space of @xmath0 as follows .
a sweep - out of @xmath0 by @xmath73-cycles is defined as a one - parameter family of @xmath73-cycles @xmath333 starting and ending at the null - cycle which induces a generator of @xmath334 .
an example of sweep - out is given by the level sets @xmath335 of a pl morse function @xmath336 $ ] . from @xcite ,
theorem [ theo : a ] holds true for every riemannian two - sphere without any curvature assumption .
it actually holds true for every closed riemannian surface without any curvature assumption if one allows the multiplicative constant to depend on the genus of the surface ( and examples show the multiplicative constant has to depend on the genus ) .
[ prop : cex ] let @xmath0 be a closed @xmath1-manifold with @xmath9 .
there exists a riemannian metric on @xmath0 with unit @xmath1-volume and arbitrarily large @xmath73-volume width .
in particular , every pl morse function on @xmath0 has a fiber of arbitrarily large @xmath73-volume for this metric .
consider an @xmath1-sphere @xmath341 endowed with a riemannian metric with unit @xmath1-volume and @xmath342 arbitrarily large . given a domain @xmath22 of @xmath0 diffeomorphic to a ball , we consider a degree one map @xmath343 taking @xmath22 to some point @xmath344 and @xmath345 to @xmath346 .
we also define a metric on @xmath0 such that the volume of @xmath345 is arbitrarily small and the restriction of @xmath347 to @xmath22 is an isometry onto @xmath348 .
now , since the map @xmath347 has degree one , the image by @xmath347 of a sweep - out @xmath338 of @xmath0 by @xmath73-cycles induces a sweep - out of @xmath341 . from the geometry of @xmath341 ,
one of the @xmath73-cycles of this sweep - out of @xmath341 has an arbitrarily large mass . as the map @xmath347 is distance , and so mass , nonincreasing , the same holds for the @xmath73-cycles @xmath338 .
this shows that @xmath349 can be arbitrarily large while @xmath0 has unit @xmath1-volume .
working with level sets does not allow us to conclude in the proof of proposition [ prop : cex ] . in order to connect @xmath0 to the sphere @xmath341
, we need the flexibility provided by the general almgren - pitts min - max principle defined on the @xmath73-cycle space . | we establish a min - max estimate on the volume width of a closed riemannian manifold with nonnegative ricci curvature .
more precisely , we show that every closed riemannian manifold with nonnegative ricci curvature admits a pl morse function whose level set volume is bounded in terms of the volume of the manifold . as a consequence of this sweep - out estimate , there exists an embedded , closed ( possibly singular ) minimal hypersurface whose volume is bounded in terms of the volume of the manifold . |
the seesaw mechanism @xcite seems to provide a simple way to understand the smallness of neutrino masses .
the simplest among the various seesaw mechanisms is the type - i seesaw , which requires the introduction of standard model ( sm ) gauge - singlet right - handed ( rh ) neutrinos ( @xmath3 s ) .
the majorana mass @xmath4 of these rh neutrinos explicitly breaks the global @xmath5 symmetry of the sm by two units and gives rise to a light neutrino mass matrix of the form @xmath6 where @xmath7 is the dirac mass generated due to the neutrino yukawa couplings @xmath8 after the electroweak symmetry breaking .
if one assumes maximal allowed values for the third - generation dirac mass @xmath9 , as is implied by certain grand unified theories such as @xmath10 , then the seesaw scale is pushed up to @xmath11 gev making it virtually inaccessible to both collider experiments at the energy frontier and other low energy searches at the intensity frontier .
such naive ( dimensional counting " ) considerations however can be misleading in the context of specific ultraviolet ( uv ) complete seesaw models where there may be dynamical suppressions of neutrino dirac masses .
one class of such models that are based on the sm gauge group generally go by the name of radiative seesaw models ; see e.g. @xcite . within the type - i seesaw framework
, there exists another possibility to suppress the light neutrino masses by assigning specific textures to the dirac and majorana mass matrices in the seesaw formula @xcite .
the stability of these textures can in principle be guaranteed by enforcing some symmetries in the lepton sector .
extending the sm gauge group to the left - right ( l - r ) symmetric gauge group @xmath0 provides a simple uv - complete seesaw model @xcite . here , the rh neutrinos are a necessary part of the model and do not have to be added ` by hand ' just to implement the seesaw mechanism .
an important point is that the rh neutrinos acquire a majorana mass as soon as the @xmath12 symmetry is broken at a scale @xmath13 .
this is quite analogous to the way the charged fermions get masses in the sm by the higgs mechanism when the @xmath14 gauge symmetry is broken at a scale @xmath15 .
the higgs field that gives mass to the rh neutrinos becomes the analogue of the 125 gev higgs boson discovered at the lhc . in this proceedings , we focus on a new class of l - r seesaw models @xcite , where in addition to the rh gauge bosons as well as other higgs fields , light ( tev scale ) rh neutrinos with sizable mixing with the left - handed ( lh ) neutrinos may be accessible at colliders and other low energy experiments ( for a review , see e.g. @xcite ) . with the lhc ramping up to a higher center of mass energy @xmath16 tev and later to 14 tev
, our model as well as other low scale seesaw models , should provide motivation for testing the origin of neutrino mass in the coming years .
for the collider phenomenology of tev - scale lrsm , see e.g. @xcite , and for the current experimental status , see @xcite .
such considerations should also be of interest @xcite for setting physics goals for future higher energy colliders such as the @xmath17 or 100 tev vlhc @xcite .
it is also important to mention here that the new class of l - r seesaw models with relatively large left - right neutrino mixing can be distinguished from the minimal lrsm using different kinematic variables @xcite .
an attractive consequence of the seesaw mechanism is the possibility that it could solve another important puzzle of cosmology , i.e. the origin of matter - antimatter asymmetry in our universe , via a mechanism called leptogenesis @xcite . at the heart of this mechanism
is the out - of - equilibrium decay of the rh majorana neutrinos via the decay modes @xmath18 ( where the superscript @xmath19 denotes the @xmath2-conjugate ) , which violate @xmath20 , @xmath21 and @xmath2 , thereby dynamically generating an asymmetry in the lepton sector .
this primordial lepton asymmetry undergoes thermodynamic evolution as the universe expands and finally gets converted to a baryon asymmetry via equilibrated electroweak sphaleron interactions @xcite .
the attractive aspect of this mechanism is that the _ same _ yukawa couplings @xmath8 that give rise to neutrino masses via the seesaw mechanism in eq .
are also responsible for the origin of matter , thus implying an intimate connection between two seemingly disparate pieces of evidence for beyond the sm physics .
this beautiful idea can be tested , provided the seesaw scale , and hence , the scale at which leptogenesis takes place , is accessible to current and near future laboratory experiments .
low - scale leptogenesis is possible when at least two rh majorana neutrinos have a small mass difference comparable to their decay widths , thus resonantly enhancing the @xmath22-type @xmath2 asymmetry @xcite due to rh neutrino self - energy effects @xcite .
an important question to explore is whether it is possible to falsify the leptogenesis mechanism by laboratory experiments @xcite . in particular
, it is worth investigating if the leptogenesis mechanism works in a tev - scale realization of l - r seesaw and whether an observation of a rh gauge boson at the lhc could falsify this scenario .
this problem was analyzed in detail in @xcite and it was pointed out that the additional dilution and washout effects on the lepton asymmetry due to @xmath23-mediated @xmath24 scattering processes become important for a tev - scale l - r seesaw . since in generic versions of the seesaw model , sub - ev neutrino masses would require the dirac yukawa couplings to be @xmath25 , the washout effects would lead to an efficiency factor @xmath26 ( @xmath27 being the @xmath12 gauge coupling , which is assumed to be equal to the @xmath14 gauge coupling in the minimal scenario ) which is too small for a low - scale leptogenesis to work . through a detailed analysis of the relevant boltzmann equations , it was concluded in @xcite that even for maximal @xmath2-asymmetry @xmath28 , the observed value of the baryon asymmetry can be explained by leptogenesis in these l - r models , only if @xmath29 tev . turning this argument around , if a positive signal for @xmath23 is observed at the lhc , this will falsify leptogenesis as a mechanism for understanding the origin of matter in l - r seesaw framework .
since this is such an important issue , we have reanalyzed this @xcite to examine the robustness of the lower bound on @xmath30 and to investigate if there exists _ any _ allowed parameter space with successful leptogenesis for smaller values of @xmath30 .
we find that for the generic class of l - r seesaw models with large light - heavy neutrino mixing , successful leptogenesis requires @xmath1 tev for the maximal @xmath2 asymmetry . for illustration
, we will present an explicit neutrino mass fit which satisfies the observed baryon asymmetry .
the main new results presented here should supersede those reported in @xcite .
this proceedings is organized as follows : in section [ sec2 ] , we review the basic features of the generic l - r seesaw models as well as our new l - r seesaw model with special dirac and majorana textures resulting in light neutrinos via type - i seesaw with large light - heavy neutrino mixing generated in a natural manner . in section [ sec3 ] , we present an alternative neutrino mass fit relying on cancellations in the seesaw formula . in section [ sec4 ] ,
we discuss resonant leptogenesis in this model and the impact of a tev - scale @xmath23 on the lepton asymmetry .
our conclusions are given in section [ sec5 ] .
in the minimal l - r symmetric model ( lrsm ) , the fermions are assigned to the gauge group @xmath0 as follows : denoting @xmath31 and @xmath32 as the quark and lepton doublets of the @xmath33th generation respectively , @xmath34 and @xmath35 ( also denoted simply by @xmath36 ) are assigned to doublets under the @xmath14 group , while @xmath37 and @xmath38 ( also denoted by @xmath39 ) as the doublets under the @xmath12 group .
the higgs sector of the model can consist of one or several bidoublets @xmath40 and triplets @xmath41 : @xmath42 the gauge symmetry @xmath43 is broken by the vacuum expectation value ( vev ) @xmath44 to the group @xmath45 of the sm .
the lh counterpart ( @xmath46 ) to @xmath47 is not considered here , assuming that parity and @xmath12 gauge symmetry scales are decoupled . in this case , the @xmath46 fields become heavy when the discrete parity symmetry is broken , and disappear from the low energy theory @xcite .
the vev of the @xmath48 field given by @xmath49 breaks the sm gauge group to @xmath50 .
the fermion masses are obtained from the following yukawa lagrangian : @xmath51 where @xmath52 ( @xmath53 being the second pauli matrix ) .
after electroweak symmetry breaking , the dirac fermion masses are given by the generic formula @xmath54 for up - type fermions , and for down - type quarks and charged leptons , it is the same formula with @xmath55 .
the yukawa lagrangian ( [ yuk ] ) leads to the dirac neutrino mass matrix @xmath56 and the majorana mass matrix for the heavy rh neutrinos @xmath57 which go into the seesaw formula for calculating the neutrino masses and the heavy - light neutrino mixing .
since the higgs sector relates the neutrino yukawa couplings with the charged - lepton ones in the lrsm , it is interesting to see how a tev - scale l - r seesaw can be realized in nature .
there are essentially three ways to do so : ( i ) by choosing one set of yukawa couplings to be @xmath58 for a particular vev assignment for the bidoublet higgs fields ; ( ii ) by choosing particular symmetry - motivated textures for the yukawa couplings which ensure that the leading order seesaw contribution to neutrino masses vanish ; and ( iii ) by choosing large yukawa couplings as allowed by specific textures and invoking cancellations in the seesaw formula .
the strategy ( i ) with small yukawa couplings does not lead to any new interesting phenomenology over the minimal lrsm .
therefore , we will focus on strategies ( ii ) and ( iii ) in the following sections , and in particular , use strategy ( iii ) for our numerical analysis for the calculation of lepton asymmetry . a similar analysis can be performed following strategy ( ii ) as well . for the phenomenological implications of this new class of l - r seesaw models ,
see e.g. @xcite .
as mentioned above , the basic strategy for understanding small neutrino masses with a low - scale type - i seesaw is to have the appropriate textures for @xmath7 and @xmath4 in eq . .
there are several examples of this type @xcite in the context of the minimal seesaw extensions of the sm . here
, we discuss the embedding of one such texture ( from ref .
@xcite ) in the l - r model using an appropriate family symmetry @xcite .
the symmetry must not only guarantee the special leptonic textures but also must be free of light scalar bosons which can result if the effect of the discrete symmetry is to automatically lead to a @xmath59 symmetry . moreover in lrsm , the charged - lepton mass matrix @xmath60 is related to @xmath7 which puts additional constraints on phenomenological viability of the model . therefore , we find it remarkable that the model presented below remains a viable tev - scale l - r type - i seesaw model for neutrinos , and as a result , has interesting phenomenological implications beyond the minimal lrsm .
the dirac and majorana mass matrices @xmath7 and @xmath4 considered here have the following form : @xmath61 with @xmath62 and @xmath63 . in the limit of @xmath64 ,
the neutrino masses vanish , although the heavy - light mixing parameters given by @xmath65 can be quite large .
the light neutrino masses given by the seesaw formula become proportional to either @xmath66 , @xmath67 or their products .
if the smallness of @xmath67 and @xmath66 can be guaranteed by some symmetry , then we have a `` natural '' tev - scale seesaw model with tiny neutrino masses .
as we show below , the mass textures can be successfully embedded into the l - r framework , while satisfying all current experimental constraints .
note that the choice of texture for the rh neutrino mass matrix in eq . is necessary to keep the model compatible with strong bounds from tree - level @xmath68 decay via tev - scale @xmath47 exchange @xcite . in order to obtain the special dirac and majorana textures given in eq .
, we supplement the l - r gauge group @xmath0 with a global discrete symmetry @xmath69 @xcite . for the higgs sector , we choose three bi - doublets ( @xmath70 ) with @xmath71 and two rh triplets ( @xmath72 ) with @xmath73 . the fermion and higgs multiplets are assigned the @xmath74 quantum numbers as shown in table [ tab1 ] .
the leptonic part of the yukawa lagrangian invariant under the @xmath74-symmetry is given by @xmath75 in the symmetry limit , the vevs of @xmath76 have the form @xmath77 .
this is because the terms of the form tr@xmath78 which would change the @xmath48 vev to the form diag@xmath79 are forbidden by the @xmath74-symmetry from appearing in the scalar potential .
thus in the symmetry limit , the @xmath7 elements in one column are big and non - zero , and the charged lepton mass matrix has one eigenvalue zero which can be identified as the electron flavor .
the rh neutrino majorana mass matrix has the form in eq .
( [ eq : texture ] ) with @xmath80 . to make the model realistic
, we add small soft symmetry - breaking terms of the form @xmath81 to the scalar potential .
this will induce the @xmath82 vevs of the form @xmath83 , where @xmath84 with @xmath85 respectively being the scalar self - couplings of the bidoublet and triplet fields in the scalar potential .
choosing appropriately small @xmath86 , we can get very small @xmath87 as required to satisfy the neutrino oscillation data .
.the discrete symmetry assignments for the fermion and higgs fields in our l - r model that lead naturally to the special dirac and majorana textures given in eq .
( [ eq : texture ] ) @xcite . [ cols="^,^",options="header " , ] since @xmath88 and @xmath89 , the yukawa couplings @xmath90 are easily calculated using vev s chosen as above , and they are found to be @xmath91 the leptonic @xmath2 asymmetry generated by the rh neutrino decays is governed by these yukawa couplings . if we assume that the two sm doublets in the lr bi - doublet i.e. @xmath92 and @xmath93 are approximate mass eigenstates with @xmath94 , the dominant contribution to the @xmath2 asymmetry comes solely from the self - energy correction by the @xmath93-loop to the decay process @xmath95 .
therefore , only the yukawa coupling matrix @xmath96 determines the flavor effect in rh neutrino interactions relevant to resonant leptogenesis .
the rh neutrino mixing effect at resonance can be captured by the one - loop resummed yukawa couplings @xcite .
explicitly , in the heavy - neutrino mass eigenbasis , we obtain for the resummed yukawa couplings @xmath97 and their @xmath2-conjugates @xmath98 the following values : @xmath99 the heavy neutrino decay rates for the yukawa - mediated two - body processes @xmath100 are calculated with these resummed yukawa couplings : @xmath101 which are crucial parameters for the calculation of lepton asymmetry , as we will see in the next section .
as eluded to above , for low scale seesaw models , a simple way to generate enough lepton asymmetry is to use resonant leptogenesis @xcite .
it requires at least two rh neutrinos to be nearly degenerate , which is guaranteed in our model by the choice of the texture in the rh neutrino mass matrix given by eq . .
thus , in addition to guaranteeing suppression of lepton flavor violating processes such as @xmath102 , this choice of rh neutrino texture also provides a necessary condition for resonant leptogenesis .
as far as the dirac neutrino mass matrix @xmath7 is concerned , we choose the pattern given in eq .
so that it fits neutrino oscillation data . since our focus is on the feasibility of low scale leptogenesis in a realistic model for neutrinos , we do not stress here on the naturalness of the texture , although a similar exercise can be performed with the fit presented in section [ sec2.1 ] .
the details of leptogenesis in this class of models was discussed in @xcite and we do not repeat the basic equations here . however , we have extended the work of @xcite and also present here some revised numerical results due to a computational error in the code used there .
the basic steps in our analysis are as follows : first we calculate the flavored @xmath2 asymmetry @xmath103 \ ; , \label{epsilon}\end{aligned}\ ] ] where the partial decay widths are given by eq . and the total decay width is given by @xmath104 + 2\ : \gamma(n_\alpha\to l_r q_r\bar{q}'_r ) \ ; . \label{tot_width}\end{aligned}\ ] ] note that in lrsm , there is an additional contribution to the total decay width due to the three - body decay mediated by @xmath23 : @xmath105 where @xmath106 is the total decay width of @xmath23 , assuming that all three heavy neutrinos are lighter than @xmath23 .
the @xmath2 asymmetry in our case is determined by the rh neutrino texture , specifically , the magnitudes of @xmath107 and @xmath108 in eq . as well as
the magnitudes of the yukawa couplings , and these parameters are constrained by the neutrino fit .
the second step in our calculation is to calculate the thermodynamic evolution of the normalized heavy neutrino and lepton doublet number densities @xmath109 and @xmath110 respectively , where @xmath111 and @xmath112 is the photon number density , @xmath113 being the riemann zeta function . for simplicity and for a fair comparison with the results of @xcite , here
we use the flavor - diagonal boltzmann equations @xcite : @xmath114 where @xmath115 is a dimensionless variable ( @xmath116 being the temperature of the universe ) and @xmath117 is the heavy neutrino equilibrium number density , @xmath118 being the @xmath119-th order modified bessel function of the second kind .
the various decay ( @xmath120 ) , scattering ( @xmath121 ) and washout ( @xmath122 ) rates appearing in eqs . and are given by @xmath123 \nonumber \\ & \ \equiv \
\frac{1}{2 \zeta ( 3 ) } z^3 k_1 ( z ) k_l^\text{eff } ( z ) \ ; , \label{wash}\end{aligned}\ ] ] where @xmath124 is the hubble parameter at @xmath125 , assuming only sm degrees of freedom in the thermal bath , @xmath126 gev is the planck mass and @xmath127 is the branching fraction of the rh neutrino decays relevant for the generation of @xmath2 asymmetry : @xmath128 \ ; .
\label{br}\end{aligned}\ ] ] the various @xmath129 s appearing in eqs .
- represent the reaction rates that involve the rh neutrino decays and inverse decays as well as other @xmath130 scattering processes in the model ( see [ app ] ) .
only the two - body decays of the rh neutrinos involving complex yukawa couplings are responsible for building up the asymmetry , whereas all other processes lead to washout effects .
we present in figure [ fig : gnh ] the evolution of the various collision rates as given in [ app ] .
the left panel of the figure is relevant to the first boltzmann equation , while the second one is relevant to .
the vertical dashed line shows the critical value @xmath131 beyond which the sphaleron processes become ineffective .
the horizontal dashed line is shown for easy comparison with the hubble rate .
here we have used @xmath132 tev which is the lowest value of @xmath30 we found in this model with an _ explicit _ neutrino fit that satisfies the leptogenesis constraints .
recently , it was pointed out @xcite that the scattering process @xmath133 mediated by the doubly - charged component of the rh triplet @xmath134 might spoil leptogenesis , if @xmath30 and @xmath135 are relatively small .
we have explicitly calculated the washout effect due to this process for various values of @xmath30 and @xmath136 .
as shown in figure [ fig : glhwe ] and also in figure [ fig : gnh ] ( right panel ) , this effect is negligible compared with the decay rate and other scattering rates when @xmath30 and @xmath136 are in the regime of interest . in the strong washout regime ,
the flavored lepton asymmetry is approximately given by @xcite @xmath137 where @xmath138 is the effective washout parameter as defined in eq . and @xmath139 is the effective dilution factor . for the numerical fit given in section [ sec3 ] , we obtain @xmath140 and the total lepton asymmetry is found to be @xmath141 .
this is well consistent with the observed value of the baryon asymmetry @xmath142 @xcite , after taking into account the sphaleron conversion rate and the entropy dilution factors . here
we present our procedure to derive a lower bound of @xmath30 which is compatible with successful leptogenesis in a lrsm with all rh neutrinos being quasi - degenerate in mass .
the total lepton asymmetry is given by eq .
( [ eq : etal ] ) , which can be rewritten using eq . as @xmath143 for the purpose of deriving a lower bound on @xmath30
, we assume that the dominant washout effect comes from the @xmath23-mediated scattering processes , i.e. , @xmath144 where @xmath145 .
this follows from the fact that gauge interactions are flavor - blind , and hence , @xmath146 has no dependence on lepton flavors .
similarly , when all the rh neutrino masses are quasi - degenerate , we can write @xmath147 .
we also assume that @xmath148 for simplicity .
note that @xmath149 is not summed over lepton flavors while @xmath74 is .
we further define @xmath150 and @xmath151 as the @xmath2 asymmetry and branching ratio , respectively , without the 3-body decay width included in the denominator of eqs . and .
for simplicity , we further assume that the branching ratios of the decay process @xmath152 are the same for all the lepton flavors .
with all the above - mentioned reasonable assumptions , we can approximate the total lepton asymmetry in eq . at the critical temperature as @xmath153 }
\left| \sum_\alpha \varepsilon_{l \alpha } \frac{\widetilde{d}_\alpha ( z_c)}{d_\alpha ( z_c ) + 3 s_{r \alpha } ( z_c ) } \right| \\ & \approx \frac{9}{4 \zeta ( 3 ) } \frac{z_c^2 k_1 ( z_c)}{s_r ( z_c ) } \frac{r_s r_d^2}{(3 + r_s r_d ) ( 3 + r_s ) } \varepsilon^y_\text{tot } \label{eq : etal2}\end{aligned}\ ] ] where the ratios @xmath154 and @xmath155 parametrize the relative washout strengths of the 2-body decay , @xmath23-mediated decay and @xmath23-mediated scattering processes .
the values of @xmath156 and @xmath157 depend on @xmath158 , @xmath30 , and the yukawa coupling @xmath159 . assuming a specific value of @xmath159 and calculating the 2-body decay width simply as @xmath160
, we can evaluate these parameters as functions of @xmath158 and @xmath30 .
figure [ fig : etalcon ] shows the contour plots of @xmath161 for two different yukawa couplings @xmath162 and @xmath163 .
the red curves correspond to @xmath164 which is the total @xmath2 asymmetry in the example fit given in section [ sec3 ] .
any region outside the red curve is incompatible with successful leptogenesis under the assumptions we introduced to obtain it , i.e. @xmath165 , @xmath166 , @xmath164 for the two specific values of @xmath159 . for @xmath162 , we find that the lowest value of @xmath30 allowed for @xmath164 is 13 tev at around @xmath158 = 580 tev .
note that the example fit we have presented in the previous section has @xmath167 tev , @xmath168 gev , and the resummed yukawa coulings of order around @xmath169 .
thus , our example fit , shown as a green dot in figure [ fig : etalcon ] , is very close to the minimum value of @xmath30 obtained here in a simplified manner , which justifies the validity of this approach . in other words , we have derived the lower bound in two different ways : ( i ) in section [ sec3 ] by carefully scanning the parameter space in order to find an explicit fit with the lowest possible @xmath30 compatible with leptogenesis , and ( ii ) using the approximate expression in eq .
( [ eq : etal2 ] ) .
the remarkable agreement between these two approaches shows that the simplified expression in eq .
( [ eq : etal2 ] ) is very effective in finding the region in parameter space compatible with leptogenesis and predicting the lower bound of @xmath30 as well as the position in the parameter space where it exists .
if we use the same expression and take the maximal @xmath2 asymmetry allowed in principle , i.e. @xmath170 , then we obtain the blue curves figure [ fig : etalcon ] . with this maximal @xmath2 asymmetry
, we have found that the yukawa coupling @xmath163 gives the lower bound of @xmath30 = 9.9 tev at @xmath158 = 630 gev .
the yukawa couplings can not be increased arbitrarily without spoiling the lepton asymmetry , since not only the source term due to the two - body decay of the rh neutrinos , but also the washout effects due to inverse decay and @xmath171 scattering increase with the yukawa couplings .
similarly , for very small values of the yukawa couplings , the branching fraction of the two - body decay mode becomes comparable or smaller than the three - body decay mode due to @xmath23 interactions , and therefore , the washout effect again increases .
thus , successful leptogenesis works only in a range of the yukawa coupling parameter space .
this is shown in the three - dimensional plot given in figure [ fig:3d ] , where we see that leptogenesis constraints in our model require the yukawa coupling to be @xmath172 for @xmath173 tev .
the robustness of the lower bound on @xmath30 obtained in figure [ fig : etalcon ] can also be verified from figure [ fig:3d ] .
in this proceedings , we address two issues related to seesaw models for neutrino masses . the first one deals with whether the tev scale can be naturally in the tev range without fine - tuning of parameters .
we present a natural tev scale left - right model which achieves this goal and therefore provides a counterexample to the common lore that either the seesaw scale must be superheavy or the active - sterile neutrino mixing must be tiny in a uv - complete seesaw model .
the second issue addresses the important question : whether in such low scale models , one can have successful leptogenesis , and if so , what constraints are implied by this on the mass of the rh gauge boson . in an explicit tev - scale lr model , with an explicit fermion mass fit , we find the lower bound to be 13.1 tev and for generic models in this class of l - r seesaw with enhanced neutrino yukawa couplings compared to the canonical seesaw case and with maximal possible @xmath2 asymmetry for each flavor , this bounds becomes 9.9 tev .
the work of p. s. b. d. is supported by the lancaster - manchester - sheffield consortium for fundamental physics under stfc grant st / l000520/1 .
the work of c. h. l. and r. n. m. is supported by the nsf grant no . phy-1315155 .
the various collision rates for decay and scattering processes in eqs .
( [ be1 ] ) and ( [ be2 ] ) are given below : @xmath174 the scattering terms involving two heavy neutrinos in the initial state , e.g. induced by a @xmath175-channel @xmath23 or @xmath176 , and by an @xmath177-channel @xmath178 , are not included here since their rates are doubly boltzmann - suppressed and numerically much smaller than the scattering rates given above @xcite .
the decay rates are explicitly given by @xmath179 , \\
\gamma^{n_\alpha}_{l_rq\bar{q } ' } \ & = \ \frac{m_{n_\alpha}^3}{\pi^2 z } k_1(z)\left[\gamma(n_\alpha\to l_r q_r\bar{q}'_r)+\gamma(n_\alpha\to \bar{l}_r\bar{q}_r q'_r)\right ] \ ; .\end{aligned}\ ] ] the various collision terms for the @xmath130 scattering processes @xmath180 are defined as @xmath181 where @xmath182 with the kinematic threshold value @xmath183/m_{n_1}^2 $ ] , and @xmath184 are the relevant reduced cross sections , whose explicit expressions can be found in @xcite .
the reduced cross section of the additional process @xmath185 mediated by @xmath134 is given by @xmath186 .\end{aligned}\ ] ]
99 p. minkowski , phys . lett .
b * 67 * , 421 ( 1977 ) ; r. n. mohapatra and g. senjanovi , phys . rev
. lett . * 44 * , 912 ( 1980 ) ; t. yanagida , conf . proc . * c7902131 * , 95 ( 1979 ) ; m. gell - mann , p. ramond and r. slansky , conf . proc . *
c790927 * , 315 ( 1979 ) ; s. l. glashow , nato sci .
b * 59 * , 687 ( 1980 ) .
a. pilaftsis , z. phys .
* c55 * , 275 ( 1992 ) ; e. ma , phys .
lett . * 81 * , 1171 ( 1998 ) ; e. ma , phys .
d * 73 * , 077301 ( 2006 ) ; p. s. b. dev and a. pilaftsis , phys . rev . * d86 * , 113001 ( 2012 ) ; s. s. law and k. l. mcdonald , int
a29 * , 1450064 ( 2014 ) ; a. j. r. figueiredo , eur .
j. c * 75 * , 99 ( 2015 ) .
j. kersten and a. yu .
smirnov , phys .
* d76 * , 073005 ( 2007 ) .
w. buchmller , c. greub , and p. minkowski , phys . lett . *
b267 * , 395 ( 1991 ) ; j. gluza , acta phys . polon .
b * 33 * , 1735 ( 2002 ) ; a. pilaftsis , phys .
lett . * 95 * , 081602 ( 2005 ) ; z .- z .
xing , prog .
suppl . * 180 * , 112 ( 2009 ) ; x .- g . he , s. oh , j. tandean and c .- c .
wen , phys .
d * 80 * , 073012 ( 2009 ) ; a. ibarra , e. molinaro and s. t. petcov , jhep * 1009 * , 108 ( 2010 ) ; r. adhikari and a. raychaudhuri , phys .
d * 84 * , 033002 ( 2011 ) ; m. mitra , g. senjanovi and f. vissani , nucl .
b * 856 * , 26 ( 2012 ) .
a. ferrari _ et al .
_ , phys .
d * 62 * , 013001 ( 2000 ) ; m. nemevsek , f. nesti , g. senjanovic and y. zhang , phys . rev .
d * 83 * , 115014 ( 2011 ) ; j. chakrabortty , j. gluza , r. sevillano and r. szafron , jhep * 1207 * , 038 ( 2012 ) ; s. p.
das , f. f. deppisch , o. kittel and j. w. f. valle , phys .
d * 86 * , 055006 ( 2012 ) ; j. a. aguilar - saavedra and f. r. joaquim , phys .
d * 86 * , 073005 ( 2012 ) .
m. flanz , e. a. paschos and u. sarkar , phys .
lett . * b345 * , 248 ( 1995 ) ; l. covi , e. roulet and f. vissani , phys . lett . *
b384 * , 169 ( 1996 ) .
f. f. deppisch , j. harz and m. hirsch , phys .
* 112 * , 221601 ( 2014 ) ; f. f. deppisch , j. harz , m. hirsch , w. c. huang and h. ps , arxiv:1503.04825 [ hep - ph ] .
z. g. berezhiani , phys .
b * 129 * , 99 ( 1983 ) ; d. chang and r. n. mohapatra , phys .
lett . * 58 * , 1600 ( 1987 ) ; s. rajpoot , mod .
lett . a * 2 * , 307 ( 1987 ) ; a. davidson and k. c. wali , phys .
* 59 * , 393 ( 1987 ) ; k. s. babu and r. n. mohapatra , phys .
lett . * 62 * , 1079 ( 1989 ) and phys .
d * 41 * , 1286 ( 1990 ) . | contrary to the common lore based on naive dimensional analysis , the seesaw scale for neutrino masses can be _ naturally _ in the tev range , with small parameters coming from radiative corrections .
we present one such class of type - i seesaw models , based on the left - right gauge group @xmath0 realized at the tev scale , which fits the observed neutrino oscillation parameters as well as other low energy constraints .
we discuss how the small parameters of this scenario can arise naturally from one loop effects .
the neutrino fits in this model use quasi - degenerate heavy majorana neutrinos , as also required to explain the matter - antimatter asymmetry in our universe via resonant leptogenesis mechanism .
we discuss the constraints implied by the dynamics of this mechanism on the mass of the right - handed gauge boson in this class of models with enhanced neutrino yukawa couplings compared to the canonical seesaw model and find a lower bound of @xmath1 tev for successful leptogenesis assuming maximal @xmath2 asymmetry for each flavor .
we also present a model with explicit neutrino mass fit , where the lower bound goes up to 13.1 tev due to less than maximal primordial @xmath2 asymmetry predicted by the model . |
for a system of @xmath1 particles in volume @xmath2 at temperature @xmath3 in thermal and mechanical equilibrium the true entropy is the gibbs entropy @xcite @xmath4 where @xmath5 is the probability that an @xmath1 particle system has a particular set of particle positions and velocities @xmath6 .
@xmath7 is the internal energy and @xmath8 is boltzmann s constant .
if the distribution @xmath5 factors into the product of a velocity distribution @xmath9 ( which is gaussian at equilibrium ) and a configurational @xmath1-particle distribution @xmath10 , where the @xmath11 superscript signifies that the system contains @xmath1 particles , then at equilibrium it is usual to define a new configurational correlation function @xcite for a sub - set of @xmath12 particles @xmath13 as @xmath14 ( where @xmath15 is the local number density at the position @xmath16 ) .
the correlation function @xmath13 has the property that it approaches one as both the separation between particles , and the system size , becomes infinite .
the first attempt to calculate the configurational entropy of a dense fluid at equilibrium was by green @xcite . considering a canonical ensemble of @xmath1 particles
, he argued that a hierarchy of correlation function ratios of the form @xmath17 are closer and closer to one as @xmath12 increases .
the product is over all distinct choices of @xmath18 symbols from the set @xmath19 .
as there are @xmath12 choices of the label to leave out , there are @xmath12 terms in the product . taking the logarithm of eqn .
( [ prod ] ) and solving recursively we obtain an expansion for the term @xmath20 implicit in eq .
( [ 1 ] ) .
nettleton and m. s. green @xcite , neglecting correlations between three or more particles , obtain a similar result for the configurational entropy in the grand canonical ensemble .
raveche @xcite extended these results to four particle correlations discussing both open ( grand canonical ) and closed ( canonical ) systems . in the following paper mountain and raveche @xcite present the first modern calculations of the configurational entropy based on the py approximation for hard spheres @xcite .
implicit in many of these early works is the idea that the expressions obtained may be useful for calculating the entropy of nonequilibrium systems @xcite .
later , in a series of papers wallace @xcite applied these methods to the calculation of configurational entropy for real systems such as liquid sodium .
for an equilibrium system in the grand canonical ensemble the entropy per particle @xcite is @xmath21 where @xmath22 is the density which is uniform at equilibrium .
the first two terms arise from kinetic theory .
there is a kinetic contribution from the velocity distribution and the term @xmath23 is the configuration term that appears in kinetic theory and the sackur - tetrode equation from the local equilibrium distribution . the term involving 2-body correlations we will refer to as @xmath24 .
the remaining term is the configurational contributions from 3-body correlations but higher order correlation function contributions also exist .
banayai and evans @xcite have shown that when compared with thermodynamic integration , a numerical calculation at equilibrium using 2-body correlations typically yields approximately @xmath25 of the entropy and including 3-body correlations gives most of the rest . in all results presented here
we consider only 2-body correlations , neglecting contributions from 3 ( or more)-body correlations .
molecular dynamics simulations have proved a very effective means of testing theoretical approaches to the study of fluids both in equilibrium , and nonequilibrium steady states @xcite . for hard particles , in the absence of external forces
, a particle trajectory is linear between collisions , so numerical simulations are only limited in accuracy by wordlength , and the accuracy of averages is only limited by statistical considerations . in kinetic - theory
the basic ingredient is the boltzmann entropy @xmath26 which is defined , up to a constant , to be @xmath27 where @xmath28 is the entropy density at position @xmath29 at time @xmath30 .
the time evolution of the @xmath31-space distribution function @xmath32 can be obtained from the boltzmann equation which takes the following form @xmath33,\ ] ] where @xmath34 is the external force as internal forces are included in the collision integral @xmath35 $ ] . in the absence of external forces , and when the spatial distribution @xmath36 is uniform , the distribution function @xmath37 which is normalized as @xmath38 where @xmath39 is the local number density of the system .
the kinetic contributions to the entropy of the quasi - one - dimensional ( qod ) system have been studied recently @xcite and one of the purposes of this paper is to show that the configurational contributions can also be estimated . the relation between the entropy flux and the heat flux may be viewed as a generalized version of the equilibrium clausius relation @xmath40 .
it was found @xcite that the kinetic fluxes of heat and entropy times the local temperature match locally for the qod system away from equilibrium and the kinetic entropy density agrees very well with the local equilibrium result .
one of the goals of this work is to determine whether this matching extends to the configurational components .
the concept of temperature and entropy away from equilibrium are active open problems @xcite .
the quasi - one - dimensional ( qod ) system of hard disks introduced in @xcite ( see figure ( [ qds ] ) ) can be modified to interact with an idealized _ heat reservoir _ in a deterministic and reversible way .
the deterministic thermal reservoir couples to the qod system of hard disks by changing the collision rule at the reservoir boundary @xcite .
a reservoir collision preserves the tangential component of momentum but the normal @xmath41-component after collision becomes @xmath42 where @xmath43 is the fixed value of the reservoir momentum determined by the reservoir temperature @xmath44 and @xmath45 is a reservoir coupling parameter .
as @xmath46 the system decouples from the reservoir and the boundary becomes a hard wall , and as @xmath47 the incoming momentum is replaced by the reservoir momentum .
recent studies of this system @xcite have shown that when in contact with two reservoirs of the same temperature the active mechanical coupling leads to a kinetic entropy production near each reservoir boundary which then flows into the reservoir .
this effect is local and restricted to a small number of boundary layer particles regardless of the system size .
the system with equal temperature reservoirs is also a dissipative dynamical system due to these reservoir collisions .
in the calculations reported here we use a system width @xmath48 , a density @xmath49 and a fixed value of the right - hand reservoir temperature @xmath50 and then choose the left - hand reservoir temperature @xmath51 to obtain the required temperature gradient of @xmath52 so there is a heat flow from left to right . hard - disk quasi - one - dimensional ( qod ) system .
the height @xmath53 is sufficiently small that the disks can not pass one another .
we choose the coordinate origin to be located at the bottom left corner of the system , and the periodic upper and lower system boundaries at @xmath54 are denoted by dashed lines .
the boundaries at @xmath55 and @xmath56 are the hard walls of the reservoirs . ] as this microscopic model couples mechanically and deterministically to the system it can be studied as a dynamical system and also as a thermodynamic system with energy and kinetic entropy flows calculated numerically without approximations .
these idealized heat reservoirs have been studied by computer simulation and kinetic theory @xcite , to obtain both heat conduction in low dimensional systems @xcite , and the lyapunov spectrum and mode structure @xcite .
the instantaneous local temperatures for each particle are @xmath57 and @xmath58 so the instantaneous system temperature is @xmath59 in the absence of a temperature gradient the average @xmath60 gives the system temperature , but when there is a temperature gradient the local time averages @xmath61 and @xmath62 give the local temperatures which will be used to determine the temperature profile inside the system .
the difference between the components of the local temperature can be used to give a measure of the deviation from local thermodynamic equilibrium .
the deterministic reservoir allows the calculation of the usual dynamical systems properties @xcite as well the thermodynamic properties @xcite .
as the system contains hard disks of diameter @xmath63 ( which we set equal to 1 ) in a narrow channel of width @xmath64 that does not allow the disks to interchange positions , see figure ( [ qds ] ) , we can associate any property of particle @xmath16 with the same _ local _ property of the volume element @xmath65 , centred at the average position of particle @xmath16 , that is @xmath66 .
then for example , the local density is the inverse of the average volume occupied by the particle @xmath67 .
this simple connection between volume element and particle property makes the qod system optimal for these studies . where there is the possibility of a difference between the local values of the @xmath41 and @xmath68 temperatures , @xmath69 and @xmath70
, we can modify the local - equilibrium distribution function as follows ; @xmath71}.\ ] ] here @xmath72 is the local number density , and @xmath73 and @xmath74 are the @xmath41 and @xmath68 components of temperature which are all functions of position @xmath41 . the kinetic entropy density obtained at the level of the local - equilibrium approximation is @xmath75\end{aligned}\ ] ] which is reminiscent of the equilibrium sackur - tetrode equation except that the system is in a nonequilibrium steady state and the hydrodynamic fields are local . for a system of spherical particles the microscopic representation for the instantaneous local heat flux vector at position @xmath29 and at time
@xmath30 is given in @xcite .
we restrict ourselves to the case where the local streaming velocity @xmath76 is zero everywhere . defining the vectors @xmath77 and @xmath78 , the impulse force at collisions between particles @xmath16 and @xmath79 is @xmath80 , where @xmath81 is the collision time and @xmath82 is the unit vector in the direction of @xmath83 .
the total heat flux @xmath84 is obtained as the volume integral of the local heat current @xmath85 over the volume assigned to particle @xmath16 , that is @xmath86 . for the qod system the local heat flux has potential contributions from two sources , either from a collision of particles @xmath16 and @xmath87 or from a collision of particles @xmath88 and @xmath16 .
the result is @xmath89 where @xmath90 is the internal energy of particle @xmath16 . in this form
it is clear that the kinetic contribution is at @xmath91 while there are two potential contributions , one at @xmath91 and the other at @xmath92 .
the time averages of the heat current must satisfy the continuity equation so on average , the same amount of heat passes through any vertical line regardless of its @xmath41 position .
the heat current density in equation ( [ jql ] ) defines the instantaneous heat current at some arbitrary @xmath41 and @xmath30 but there is only a kinetic contribution if there is a particle at @xmath93 , and there is only a potential contribution if two particles collide where one is at @xmath94 and for the other at @xmath95 , so it is the time average of this instantaneous quantity that satisfies the continuity equation .
in the same way as green proposed that the equilibrium expression for the entropy could serve as reasonable approximation for the entropy in nonequilibrium systems , we will generalize equation ( [ greenent ] ) to apply to nonequilibrium qod systems with a steady heat flow .
we consider a steady state system with a temperature gradient along the @xmath41 direction where the average total momentum is zero .
mechanical equilibrium implies that the pressure is constant throughout the system , but the imposed temperature gradient induces a density gradient and this needs to be explicitly included in the nonequilibrium form for the entropy .
the temperature gradient implies a density gradient so @xmath22 is position dependent , and the distribution function @xmath96 for particle @xmath16 , defined using @xmath97 , will ensure that the integrand goes to zero at large separations .
thus we propose the following expression for the nonequilibrium entropy of particle @xmath16 @xmath98 where @xmath99 is measured from the position of particle @xmath16 .
the integral is over the volume of the qod system , @xmath100 and @xmath101 where @xmath102 is a cut - off distance .
key to this form is that the density is position dependent and that the correlation functions @xmath96 are defined so that @xmath103 as @xmath104 , so that @xmath105 goes to zero ensuring the convergence of the integral .
the first two terms in eq .
( [ 2 ] ) are the kinetic contributions to the entropy @xmath106 of particle @xmath16 and the subsequent terms are the configurational contributions @xmath107 .
the entropy density in the volume element associated with particle @xmath16 is then given by @xmath108 this nonequilibrium entropy density varies across the qod system as the velocity distribution changes , the local density @xmath109 changes and the correlation function @xmath110 also changes from particle to particle . in particular
, the nonequilibrium boundary conditions induce an asymmetry in @xmath111 with respect to a sign change in @xmath112 .
we will look more closely at these contribution in the coming sections .
we study nonequilibrium qod systems of @xmath113 , @xmath114 and @xmath115 hard disks at a density of @xmath116 with system width @xmath48 .
the temperature of the cold reservoir on the right - hand side is kept constant at @xmath50 , and different values of the temperature of the left - hand reservoir @xmath117 are used to obtained the required temperature gradient @xmath118 . for @xmath119
we consider three values for the temperature of the hot reservoir , @xmath120 so @xmath121 , @xmath122 a moderate gradient @xmath123 and @xmath124 a high gradient @xmath125 .
for each of the larger system sizes we consider the same values of @xmath118 . it has previously been observed that the numerically calculated local _ kinetic _
entropy density agrees well with that calculated from the local equilibrium approximation @xcite .
while the momentum distributions for the particles can not be exactly gaussian for a nonequilibrium steady state , the deviations from gaussian are at best only subtle and the local kinetic entropy density calculated from the numerical distributions is almost indistinguishable from the local kinetic entropy density obtained from the local equilibrium distribution .
the configurational contribution is calculated from eq .
( [ 2 ] ) using the two - particle correlation function @xmath111 collected on a histogram of @xmath126 bins where @xmath112 is the radial distance @xmath127 .
this corresponds to a cutoff in the integral over @xmath112 of @xmath128 . for a qod system
the correlation function can be written as a function of @xmath112 and @xmath68 so the @xmath129-particle configurational integral becomes @xmath130 where @xmath131 and @xmath132 .
the slope of the density gradient is estimated locally at the position of each particle and a linear approximation to the density gradient is used in the integral .
this can lead to inaccuracies near the boundaries where the density varies quickly and possibly nonlinearly on @xmath112 .
a typical result for @xmath111 for particle @xmath133 in an @xmath113 particle system is shown in fig .
( [ gry20 ] ) with some strong @xmath68 dependence at contact ( @xmath134 ) but this disappears quickly at larger @xmath112 . for particle @xmath135 in the same system @xmath111 is shown in fig .
( [ gry60 ] ) and the configurational entropy from @xmath111 is larger in magnitude and the second neighbour peak varies more strongly with @xmath68 . for particle @xmath135 in a system of @xmath113 disks at a density of @xmath136 .
the color scale is the same as fig .
( [ gry20 ] ) .
the hot reservoir temperature is @xmath137 so @xmath138 . the local temperature at particle
@xmath135 is @xmath139 and the local density is @xmath140 . ]
the integral in eqn .
( [ gry ] ) has a maximum cutoff of @xmath141 particle diameters and the convergence of the result depends on the smoothness of the numerically generated @xmath111 .
again near the reservoir boundaries the cutoff needs to be limited so that the range of the integral remains within the simulation cell . if @xmath142 where @xmath143 is the exact result and @xmath144 is a noise term that goes to zero @xmath145 where @xmath146 is the number of samplings of the correlation function and the integral of @xmath144 over @xmath112 and @xmath68 is zero , then it can be shown that @xmath147 .
this implies that @xmath148 is a lower bound on the exact value @xmath149 . in table
( [ int_converg ] ) we present a calculations of the configurational entropy as a function of the number of histogram samplings @xmath146 and the integral cut - off @xmath102 used in eq .
( [ gry ] ) . in each case
the differences are quite small with the maximum being @xmath150 .
.the total configurational entropy @xmath107 of particle @xmath16 in a qod system of @xmath113 disks at a density @xmath151 with the right - hand reservoir temperature of @xmath50 and a left - hand reservoir temperature @xmath122 with a correlation function cut - off of @xmath128 .
the column headed @xmath152 is the converged result , the column headed @xmath153 is the result with only half the sampling and the last column is the converged result with @xmath154 ( rather than @xmath155 ) .
the largest percentage difference is 0.25 % . [
cols="^,^,^,^",options="header " , ] [ asym ] our purpose here is to identify approximate ( or possibly exact ) scaling relations for the properties of the system a function of system size @xmath1 and temperature gradient @xmath118 , with a view to separating bulk properties from surface properties ( or boundary effects ) . to look more closely at system size scaling we consider three systems with the same temperature gradient @xmath138 with @xmath113 , @xmath114 and @xmath115 disks .
as the right - hand temperature is fixed at @xmath50 , we can take the temperature profile for @xmath113 particles and multiply the @xmath73 axis by two and the @xmath41 axis by @xmath156 to match the @xmath114 particle profile .
similarly , taking the @xmath114 particle profile , multiplying the @xmath73 axis by two and the @xmath41 axis by 1.666 , matches the @xmath115 particle @xmath3 profile .
the average density of the system is @xmath136 so this sets the mean of the density profile which occurs in a volume element close to the centre of the system . scaling the @xmath41 axis for the @xmath113 particle system using @xmath157 maps the @xmath113 particle profile onto the @xmath114 particle profile .
again , using the scaling @xmath158 maps the @xmath114 particle profile onto the @xmath115 particle profile . the results for this scaling are shown in fig .
( [ t_sc ] ) .
( red symbols ) and @xmath114 ( blue symbols ) qod systems particles onto the profile for @xmath115 ( green symbols ) particles at a density @xmath151 . in each case
the temperature profile is @xmath138 . ]
the configurational entropy of each particle can also be scaled in the same way as the density .
the @xmath159 coordinate for @xmath119 is scaled by @xmath157 and then by @xmath158 , while for the @xmath160 system we use only the second of these scalings , to map these results onto those for @xmath161 .
essentially the deviations are only noticeable near the system boundaries .
the results shown in fig .
( [ ds_sc ] ) are as good as those for the density profile in fig .
( [ t_sc ] ) .
this means the volume elements with the same density have the same configurational entropy density regardless of system size and increasing system size adds regions to the left and right hand sides that have larger deviations from the mean density and the central value of configurational entropy .
the @xmath12-particle correlation functions for a system of hard core particles only depends on the density and not on the temperature so the configurational contribution to the entropy scales with density .
by contrast , the local equilibrium kinetic entropy density which is observed to be almost equal to the calculated kinetic entropy density , scales with both the temperature and density , thus the two contributions to the total entropy density scale differently .
and @xmath114 qod systems particles onto the profile for @xmath115 particles at a density @xmath151 and a temperature profile of @xmath138 .
the symbols are the same as fig .
( [ t_sc ] ) , red for @xmath113 , blue for @xmath114 and green for @xmath115 . ] particle qod system with @xmath162 .
the green symbols are the calculated kinetic entropy density and the green crosses are the local equilibrium result .
these two results are indistinguishable on this scale .
the red symbols are the configurational contribution to the entropy density and the blue symbols are the total entropy density . ] combining the previously calculated kinetic entropy densities with the configurational entropy density calculated here we can consider the relative sizes of the components . in fig .
( [ s(x ) ] ) the results for an @xmath113 particle system at a density @xmath136 and system width @xmath163 with @xmath164 are shown .
the kinetic entropy density is always positive and the configurational entropy density is always negative but smaller in magnitude so the total entropy density remains positive .
in this paper we have followed the prescription originally imagined by green @xcite , ( i ) that eq .
( [ 2 ] ) is correct when the fluid is in thermal and mechanical equilibrium , and ( ii ) when the fluid is not in thermal and mechanical equilibrium the entropy is always less than the equilibrium value .
strictly , green was speaking about the canonical equivalent to eq .
( [ 2 ] ) rather than the grand canonical result used here .
the demonstration of point ( i ) has been made previously in the work of baranyai and evans @xcite where both 2 and 3-body correlation functions are considered for equilibrium systems .
here we demonstrate point ( ii ) for nonequilibrium systems and more strongly by calculating the local entropy per particle or alternatively by calculating the local entropy density .
the qod system is particularly useful in connecting the property of a single particle with the density of that property in the average volume element occupied by that particle .
it is interesting to remark that the derivations which lead from the entropy in eq .
( [ 1 ] ) to entropy per particle in eq . ( [ 2 ] ) rely on the canonical and grand canonical ensemble formulations that only apply at equilibrium .
there is no equivalent derivation for a nonequilibrium steady state . applying eq .
( [ 2 ] ) outside of equilibrium is based on the fact that the quantities involved , the local density @xmath22 and the 2-body correlation function @xmath110 and higher order correlation functions , are well defined structural quantities that are clearly calculable in a molecular dynamics computer experiment . as green has cautioned , the quantity we have calculated as the local nonequilibrium entropy has the properties that we expect , it is equal to the entropy at equilibrium , but it may not be the thermodynamically meaningful nonequilibrium entropy .
it is an order parameter that is proportional to the nonequilibrium entropy and its usefulness in a thermodynamic context is yet to be determined .
the author thanks thomas dean , stuart hatzioannou and david webb for discussions and assistance with computations .
p. j. daivis , studying non - equilibrium thermodynamics using non - equilibrium molecular dynamics simulations , in proceedings of the 12th joint european thermodynamics conference , jetc 2013 , eds .
m. pilotelli and g.p .
beretta ( brescia , italy , 2013 ) , pp . | in a quasi - one - dimensional system the particles remain ordered from left to right allowing the association of a volume element to the particle which on average resides there .
thus the properties of that single particle can give the local densities in the volume element . with reservoirs of different temperatures connected to each end of the system a steady heat current with an anomalous thermal conductivity results . a local configurational entropy density
is calculated from two - particle correlation functions which varies locally within the nonequilibrium steady state .
this local configurational entropy is proposed as the configurational component of the local entropy of the nonequilibrium steady state .
@xmath0 e - mail : ( [email protected] ) . |
relativistic heavy ion reactions exhibit dominant collective flow behaviour , especially at higher energies where the number of involved particles , including quarks and gluons , increases dramatically . at intermediate stages
approximate local equilibrium is reached , while the initial and final stages may be far out of local equilibrium .
also , different stages may have different forms or phases of matter , especially when quark gluon plasma ( qgp ) is formed .
the need to describe and match different stages of a reaction was realized by the development of the final freeze - out ( fo ) description in landau s fluid dynamical ( fd ) model @xcite .
then it was improved by milekhin @xcite , and a covariant simple model was given by cooper and frye @xcite .
in all these models the fo happened when the fluid crossed a hypersurface in the spacetime . at early relativistic heavy ion collisions ,
the initial compression and thermal excitation was described by a compression shock in nuclear matter .
this was already pointed out by the first publications of w. greiner and e. teller and their colleagues @xcite , and the shock took place crossing a spacetime hypersurface ( e.g. a relatively thin layer resulting in a mach cone ) .
when sudden large changes happen across a spacetime front the conservation laws and the requirement of increasing entropy should be satisfied : @xmath0~=~0\label{nconserve}~;\\ & [ t^{\mu\nu}d\sigma_{\mu}]~=~0\label{tconserve}~;\\ & [ s^{\mu}d\sigma_{\mu } ] ~\geq~0 \label{entropy}\end{aligned}\ ] ] where @xmath1 is the baryon current , @xmath2 is the entropy current , @xmath3 is the energy momentum tensor , which , for a perfect fluid , is given by @xmath4 where @xmath5 is the energy density , @xmath6 is the pressure , @xmath7 is the entropy density , and @xmath8 is the baryon density of matter .
these are invariant scalars .
the @xmath9 is the normal vector of the transition hypersurface , @xmath10 is the particle four velocity @xmath11 , normalized to @xmath12 .
the square bracket means @xmath13=a_1 - a_0 $ ] , the difference of quantity @xmath14 over the two sides of the hypersurface .
the metric tensor is defined as @xmath15 .
we will also use the following notations : @xmath16 , @xmath17 is the invariant scalar baryon current across the front , @xmath18 is the generalized specific volume , @xmath19 , and @xmath20 , @xmath21
. for a perfect fluid local equilibrium is assumed , thus the fluid can be characterized by an equation of state ( eos ) , @xmath22 .
( [ nconserve],[tconserve ] ) and the eos are 6 equations , and can determine the 6 parameters of the final state , @xmath5 , @xmath8 , @xmath6 , and @xmath23 .
later csernai @xcite pointed out the importance of satisfying energy , momentum and particle charge conservation laws across such hypersurfaces and generalized the earlier description of taub @xcite to spacelike and timelike hypersurfaces ( with spacelike and timelike normals respectively ) .
in this situation the matter both before and after the shock was near to thermal equilibrium , and thus the conservation laws led to scalar equations connecting thermodynamical parameters of the two stages of the matter : the generalized _ rayleigh line _ and _ taub adiabat _
@xcite : @xmath24(d\sigma^{\mu } d\sigma_{\mu } ) / [ x ] \ , \ [ p ] = [ ( e+p)x ] / ( x_1 + x_0 ) \ .
\label{rayligh - taub}\ ] ] at much higher energies , at the first stages of the collision , the matter becomes transparent and the initial state is very far from thermal equilibrium . for this stage
other models were needed to handle the initial development , e.g. refs .
@xcite . the initial non - equilibrium state in this situation can not be characterized by thermodynamical parameters or an eos , so the previous approach , with the generalized _ rayleigh line _ and _ taub adiabat _ is not applicable . nevertheless , the intermediate ( fluid dynamical ) stage is in equilibrium and has an eos , while the initial state has a well defined energy momentum tensor . in this work
we will demonstrate that the final invariant scalar , thermodynamical parameters can be determined in this situation also from the conservation laws .
then , bugaev @xcite observed that fo across hypersurfaces with spacelike normals , has problems with negative contributions in the cooper - frye evaluation @xcite of particle spectra , thus the fo must yield an anisotropic distribution , which he could approximate with a cut - jttner distribution @xcite .
this is not surprising as in the rest frame of the front ( rff ) all post fo particles must move `` outwards '' , i.e. @xmath25 is required .
this condition is not satisfied by any non - interacting thermal equilibrium distribution , which extend to infinity in all directions even if they are boosted in the rff .
subsequently , another analytic form was proposed by csernai and tamosiunas , the cancelling - jttner distribution @xcite , which replaced the sharp cutoff by a continuous cutoff , based on kinetic model results .
parallel to this development , the fo process was analysed in kinetic , transport approaches @xcite , where the fo happened in an outer layer of the spacetime , or in principle it could be extended to the whole fluid ( although , at early moments of a collision / explosion , from the center of the reaction few particles can escape ) .
these transport studies also indicated that the post fo distributions may become anisotropic @xcite even for fo hypersurfaces with timelike normal [ in short : _ timelike surface _ ] , if the normal , @xmath26 , and the velocity four - vector , @xmath27 , are ( very ) different
. these studies led to another fo description , where the initial stages of the collision with strongly interacting matter were described by fluid dynamics , while the final , outer spacetime domain ( or later times ) was described by weakly interacting particle ( and string ) transport models , where the final fo was inherently included , as each particle was tracked , until its last interaction .
it is important to mention , that in these approaches , the transition from the fd stage to the molecular dynamics ( md ) or cascade stage happens when the matter crosses a spacetime hypersurface , thus the conservations laws @xcite have to be satisfied and the post fo particle phase space distributions @xcite have to be used when the post fo distributions become anisotropic . in this work for the first time we present a simple covariant solution for the transition problem and conservation laws for the situations when the matter after the front is in thermal equilibrium ( i.e. it has isotropic phase space distribution ) and has an eos , but the matter before the front must not be in an equilibrium state .
then we discuss the situation where microscopic models are appended to the fluid dynamical model , which are in , or close to thermal equilibrium , but the eos , is not necessarily known .
subsequently , we present the way to generalize the problem to anisotropic matter in final state , which is necessary for fo across spacelike surfaces and also for timelike surfaces if the flow velocity is large in the rest frame of the front ( rff ) .
this problem was solved in kinetic approach for the bugaev cut - jttner approach @xcite and the csernai - tamosiunas cancelling - jttner approach , @xcite by calculating the energy momentum tensors explicitly from the anisotropic phase space distributions , but no general solution is given for post fo matter with anisotropic pressure tensor .
the transition hypersurface between two stages of a dynamical development are most frequently postulated , governed by the requirement of simplicity .
thus , such a hypersurface is frequently chosen as a fixed coordinate time in a descartian frame @xmath28 , or at a fixed proper time @xmath29 from a spacetime point , although in a general 3 + 1 dimensional system the choice of such a point is not uniquely defined .
it is important that the _ transition hypersurface should be continuous _ , ( without holes where conserved particles or energy or momentum could escape through , without being accounted for ) . to secure that one quantity ( e.g. baryon charge ) does not escape through the holes of a hypersurface
is not sufficient , as other quantities may ( e.g. momentum in case if @xmath30 is different on the two sides of a hole ) .
again , to construct such a continuous hypersurface in a general 3 + 1 dimensional system is a rather complex task , although , in 1 + 1 or 2 + 1 dimensions it seems to be easy .
both the initial state models and the intermediate stage , fluid dynamical models may be such that the calculation could be continued beyond the point where a transition takes place . then spacetime location of the transition to the next stage can or should be decided , based on a physical condition or requirement , which may be external to the development itself . as a consequence , in some cases the determination of transition surface may be an iterative process . numerically , the extraction of a freeze out ( fo ) hypersurface is by no means trivial .
one of us , brs , has recently provided a proper numerical treatment regarding the extraction of fo hypersurfaces in two ( 2d ) , three ( 3d ) and four ( 4d ) dimensions @xcite .
for instance , in 2d the history , i.e. , the temporal evolution , of a temperature field of a one - dimensional ( 1d ) relativistic fluid can be represented by a gray - level image ( _ cf .
_ , fig . 1 ) . in the figure
, we use the time @xmath28 and the radius @xmath31 for the temporal and the spatial dimensions , respectively .
let bright pixels ( i.e. , picture elements ) refer to high temperatures and dark ones to low temperatures of the fluid .
in this example , a 2d freeze - out hypersurface is an iso - therme . in fig . 1.a , we also depict the corresponding co - variant normal vectors @xmath32 . in 2d
, the length of each normal vector is equal to the length of each supporting iso - contour vector .
each normal vector has its origin at the contra - variant center , @xmath33 , of a given contra - variant iso - contour vector and points to the exterior of the enclosed spacetime region .
the latter is also indicated in fig .
1.b , where we show that a contra - variant normal vector @xmath34 can be obtained by reflection of the co - variant normal vector @xmath32 at the time axis ( dashed line ) . finally , in fig .
1.c we show the contra - variant fo contour vectors with their corresponding contra - variant normal vectors @xmath34 .
not all of these contra - variant normal vectors point to the exterior of the enclosed spacetime region .
note that the sign conventions of the normals of the transition hypersurface are important , and must be discussed , especially if both timelike and spacelike surfaces are studied . in fact , only the timelike contra - variant normal vectors point outwards , whereas the spacelike contra - variant normal vectors point inwards @xcite .
if we know the fo hypersurface and the local momentum distribution after the transition the total , measurable momentum distribution can be evaluated by the cooper - frye formula @xcite .
let us define the contra - variant and co - variant surface normal four - vectors as @xmath35 where in general @xmath36 , as the surface element can be either spacelike ( - ) or timelike ( + ) .
we can also introduce a unit normal to the surface as : @xmath37 so that @xmath38 furthermore @xmath39 where for timelike surfaces @xmath40 and for spacelike surfaces @xmath41 . for the frequently used timelike , one - dimensional case @xmath42 . in the general case
the conserved energy - momentum current crossing the surface element is @xmath43 @xmath44 must be continuous across the freeze - out surface , as must the baryon current @xmath45 , @xmath46 where @xmath47 is the invariant scalar baryon charge current .
we assume that the initial state , `` @xmath48 '' , and its energy momentum tensor and baryon current before the front is known .
we aim for the characteristics of the final state . in total
there are six unknowns in the equilibrated final state , these are @xmath23 , @xmath5 , @xmath6 and @xmath8 ( here we drop the index `` @xmath49 '' for the final state for shorter notation ) , however the pressure @xmath6 , a function of @xmath5 and @xmath8 , is given by the eos , @xmath22 .
knowing @xmath8 and @xmath5 , the eos , and the particular form of the corresponding equilibrated distribution function , the parameters @xmath50 , and @xmath51 , can also be obtained .
thus , we have to solve 5 equations : @xmath52 the l.h.s .
represents quantities of the initial state of matter and the corresponding conserved quantities are known . equations ( [ eq0],[eq1 ] ) can be solved for @xmath53 in the calculational frame : @xmath54 using now eq .
( [ eq1],[eq2],[eq3 ] ) one obtains @xmath55 , and in a similar fashion @xmath56 and @xmath57 @xmath58 this results for @xmath59 , in @xmath60 where , @xmath61 is an invariant scalar , and @xmath62 transforms as the 0-th component of the 4-vector @xmath63 .
notice that eq .
( [ eq9 ] ) was not used up to this point , thus we can use there results both for the baryon - free and baryon - rich case .
we can have an elegant direct solution for the proper energy density , @xmath5 , and pressure , @xmath6 , as both of these quantities are invariant scalars , and we can express these by the covariant , 4-vector equation ( [ em - curr ] ) . from this 4-vector equation
we can get two invariant scalar equations by ( i ) taking its norm , @xmath64 , and ( ii ) taking its projection to the normal direction , @xmath65 : @xmath66 now expressing @xmath67 from eq .
( [ ads ] ) and inserting it to eq .
( [ aa ] ) , we obtain our final equation @xmath68 which can be solved straightforwardly if the eos , @xmath69 , is known .
the other three elements of the equation , @xmath64 , @xmath65 , and @xmath70 , are known from the normal to the surface and from energy - momentum current from the pre - transition side .
then , eqs .
( [ eqg2a]-[eqg2b ] ) can be used to determine the final flow velocity . at the end , after all conservation law equations are solved , we have to check the non - decreasing entropy condition ( [ entropy ] ) to see whether the solution is physically possible . if the overall entropy is decreasing after transition that would mean that the hypersurface is chosen incorrectly .
one will need to choose more realistic condition for the transition and repeat the calculations .
this result can be used both if the initial state is in equilibrium and if it is not . in case of an ideal gas of massless particles after the front , with an eos of @xmath71 , eq .
( [ aa2 ] ) leads to a quadratic equation , @xmath72 where @xmath73 , is the energy momentum transfer 4-vector across a unit hypersurface element .
if the flow velocity is normal to the fo hypersurface , @xmath74 , then for an initial perfect fluid in the local rest ( lr ) frame the above covariant equation takes a simple form , @xmath75 this has two real roots , @xmath76 ( energy density is conserved ) and @xmath77 which does not correspond to a physical solution , as the energy density should not be negative .
if the eos depends on the conserved baryon charge density also , then we must exploit in addition eq .
( [ eqn ] ) : @xmath78 and inserting @xmath79 from here to eq .
( [ ads ] ) yields @xmath80 where @xmath81 is the generalized specific volume , well known from relativistic shock and detonation theory @xcite .
this equation provides another equation for @xmath82 as @xmath83 \ , \label{ep2nd}\ ] ] which , together with eq .
( [ aa2 ] ) and the eos , @xmath22 , provide three equations to be solved for @xmath84 and @xmath8 .
this evaluation of the post fo configuration is in agreement with the theory of relativistic shocks and detonations @xcite allowing for both spacelike and timelike fo hypersurfaces .
see also @xcite .
this method of evaluation observables is frequently used at the end of fluid dynamical model calculations ( see e.g.@xcite ) .
recently a frequently practiced method to describe the final stages of a reaction is to switch the fd model over to a molecular dynamics ( md ) description at a transition hypersurface .
this is frequently a fixed time , @xmath28 , or fixed proper time , @xmath29 hypersurface .
the generation of the initial state of such an md model is a task , which depends on the constituents of the matter described by the md model .
nevertheless , same principles must be satisfied , like the conservation laws , eqs.([nconserve]-[tconserve ] ) .
let us assume , although not required by physical laws , that we have thermal equilibrium on both sides of the transition and we know explicitly the corresponding final momentum distribution of particles .
then , the fundamental equation to construct the post transition microscopic state , in addition to the conservation laws is the cooper - frye formula , @xmath85 assuming that the local phase space distribution , @xmath86 , is known for all initial components of the md model . if @xmath86 are local equilibrium distributions then ( in principle ) we know the intensive and extensive thermodynamical parameters and the eos of the matter when the md model simulation starts
. these must not be the same as the ones before the transition hypersurface . in the usual transition from fd to md models ,
where the initial state of md is in equilibrium , the eos - s are known on both sides of the transition surface , and thus , both the equations of rayleigh - line and taub - adiabat , eqs .
( [ rayligh - taub ] ) , as well as the invariant scalar equations derived here , eqs.([aa],[ads],[aa2],[ep2nd ] ) can be used to determine all parameters of the matter starting the md simulation .
these then determine the phase space distributions , @xmath86 of all components of the md simulation .
subsequently eq.([cf - f ] ) can be used to generate randomly the initial constituents of the md simulation . as eq.([cf - f ] ) is a covariant equation applicable in any frame of reference , the most straightforward is to perform the generation of particles in the calculational frame of the md model .
this transition is by now performed in many hybrid models combining fluid dynamics with microscopic transport models @xcite .
these models at present are the most effective to describe experimental data and make the need for a modified boltzmann transport equation @xcite less problematic . in some cases the first step of the transition , the determination of the parameters of the final state from the exact conservation laws , is dropped with the argument that both before and after the transition the matter has the same constituents and the same eos , thus the all extensive and intensive thermodynamical parameters as well as the flow velocity must remain the same .
then , using the intensive parameters the final particle distributions in the cooper - frye formula , eq.([cf - f ] ) , can be directly evaluated in a straightforward way .
this procedure is correct , but only if all features of the two states of the matter and their eos are identical . in some cases the pre transition eos assumes effective hadron masses depending on the matter density , while the final eos is that of a hadron ideal gas mixture , but with fixed vacuum masses .
this leads to a difference in the eos , thus the above procedure is approximate . in such cases ,
the method can be used , but the accurate conservation laws can be enforced by a final adjustment step described in the next subsection . the situation is similar if the constituents and the eos are almost identical before and after the transition , but before the transition a weak or weakening mean field potential or compression energy is taken into account .
in addition to the above mentioned approximate methods , even for really identical eos - s across the transition or with generating the final eos parameter based on conservation laws for the final eos , inaccuracies may arise due to other reasons : during the random generation of the initial constituent particles of the md simulation , the exact conservation laws may be violated , due to finite number effects . however , the energy and particle number conservations are usually enforced during the random generation of particles , even if the above procedure of solving the conservation laws beforehand is not fully followed .
this is usually the consequence of the fact that the eos of the md model is not necessarily known if the model has complex constituents and laws of motion . in any case to remedy this random error and make the conservation laws exactly satisfied a final correction step is advisable , and it is not always performed .
if the energy and particle number conservations are enforced then , the last variable to balance is the momentum conservation .
this regulates the flow velocity of the matter after the transition initiating the md simulation .
the energy momentum tensor and baryon current for the generated random set of particle species , " @xmath87 , for each fluid cell ( or group of cells if the multiplicity in a single cell is too low ) can be calculated from the kinetic definition : @xmath88 which , yield the resulting momentum and flow velocity of the matter .
this can be used to adjust the flow velocity to achieve exact conservation of momentum , and modify the velocity of generated particles by the required lorentz boost .
the other conserved quantities may then be affected also , but an iterative procedure to eliminate the error completely is not crucial as the error can be given quantitatively .
if the randomly generated state is not following a thermal equilibrium phase space distribution , @xmath89 , and thus does not have an eos , the above described scalar equations can not be used to generate the initial configuration of the md model .
nevertheless , the second step to check the conservation laws with the kinetic definition , and then correct the parameters of the generated particles can be done . for a required level of accuracy in this case an iterative procedure
may be necessary .
another , easier way to remedy this problem is to choose the transition hypersurface earlier so that the subsequent matter is still in thermal equilibrium .
this can always be done if the requirement of entropy increase is satisfied .
we have mentioned that the assumption for having thermal equilibrium in the final state is neither excluded nor required from transport theoretical considerations .
however , thermal equilibrium distribution is not possible if we have to describe fo across a spacelike hypersurface ( see the discussion in section [ intro ] . ) in the md model description the final post fo momentum distributions develope a local anisotropy if the fo has locally a preferred direction . unless the unit normal of the fo hypersurface is equal to the local flow velocity of the pre fo matter , there is always a selected spatial direction which is the dominant direction of fo
this situation is discussed in several theoretical works , and some general features can be extracted from these studies . in explicit transport models this situation
is handled @xcite : starting from an equilibrium jttner distribution and considering a momentum dependent escape probability in the collision term , - which reflected the direction of the fo front and the distance from the front , - an anisotropic distribution was obtained ( i.e. a distribution , which was anisotropic even in its own lr frame ) .
this anisotropic distribution could be approximated with analytic distribution functions @xcite : the starting point is the un - cut , isotropic , jttner distribution in the rest frame of the gas ( rfg ) , which is centered around the 4-velocity vector , @xmath90 .
this distribution is then cut or cut and smoothed .
the resulting distribution has a different new flow velocity , @xmath91 , which is non - zero in rfg , and is pointing in space in the direction of the normal of the fo hypersurface , @xmath92 , labeled by @xmath93 .
this @xmath91 defines the local rest ( lr ) frame of the post fo matter .
the spatial direction of @xmath92 is not affected by the lorentz transformation from rff to rfg and then to lr , as @xmath92 is the direction of the lorentz transformation from rfg to lr .
, must not be parallel to @xmath94 or @xmath95 but these latter velocities can be decomposed to @xmath93 and @xmath96 components with respect to @xmath92 .
due to the construction @xcite of the cut- or cancelling - jttner distributions @xmath97 or @xmath98 . ] in the general case the boost in the @xmath99 direction leads to a change of the distribution function in the @xmath100 direction , but does not affect the distribution in the @xmath101 direction , or the procedure of cutting or cancelling the distribution in the @xmath93 direction .
( the illustration in fig .
2a shows the spatial momentum distribution where the boost in the orthogonal direction @xmath99 is already performed . ) in the final lr frame , the matter is characterized by a rather complex energy momentum tensor , inheriting some parameters from the original uncut distribution in rfg , like the temperature and chemical potential , but as the resulting distribution is not a thermal equilibrium distribution , these parameters are not playing any thermodynamical role .
one has to determine all parameters numerically from conservation laws ( [ nconserve],[tconserve ] ) , as done in refs .
@xcite .
interestingly , a simplified numerical kinetic fo model @xcite led to a fo distribution satisfying the condition @xmath25 for spacelike fo with a smooth distribution function , which is anisotropic ( also in its own lr frame ) and has a symmetry axis pointing in the dominant fo direction .
this distribution was then approximated with an analytic , `` cancelling - jttner '' distribution @xcite , which can also be used to solve the fo problem .
after fo , the symmetry properties of the energy momentum tensor are the same for the cut - jttner and cancelling - jttner cases @xcite .
the fo leads to an anisotropic momentum distribution and therefore to an anisotropic pressure tensor .
the energy momentum tensor is not diagonal in the rfg frame , there is a non vanishing transport term , @xmath102 @xcite , in the 2-dimensional plane spanned by the 4-vectors , @xmath90 and @xmath103 .
one can , however , diagonalize the energy momentum tensor by making a lorentz boost into the lr frame using landau s definition for the 4-velocity , @xmath91 . in this frame
then the energy momentum tensor becomes diagonal , but the pressure terms are not identical , due to the anisotropy of the distribution : @xmath104 here the energy density , @xmath5 , of course must not be the same as in the case of an isotropic , thermal equilibrium post fo momentum distribution .
this can be seen from the kinetic definition of the energy momentum tensor as shown in refs .
@xcite .
we need the complete post fo momentum distribution and the corresponding energy momentum tensor to determine final observables .
this depends on the transport processes at fo , and can not be given in general ; however , due to the symmetries of the collision integral , the symmetries of the energy momentum tensor are the same irrespectively of the ansatz used ( e.g. cut - jttner , cancelling - jttner or some other distribution ) . in kinetic transport approaches
the microscopic escape probability @xcite is peaking in the direction of @xmath103 , which yields a distribution peaking in this direction , i.e. yielding the same symmetry properties as the previously mentioned analytic ansatzes .
the energy momentum tensor in general takes the form @xmath105 where @xmath106 is the orthogonal projector to @xmath107 , and @xmath108 is the unit 4-vector projection of @xmath109 in the direction orthogonal to @xmath107 , i.e. @xmath110 , where @xmath111 ensures normalization to -1 . in the landau lr frame this returns expression ( [ tdiag ] )
the 4-velocity , @xmath107 , and the other parameters of the post fo state of matter , should be determined from the conservation laws ( [ nconserve],[tconserve ] ) .
the schematic diagram of the asymmetric distributions and the different reference frames can be seen in fig .
[ fab ] .
the fo problem was solved for these configurations and ansatzes , by satisfying the conservation laws explicitly for the full energy momentum tensor .
we do not have a general eos(s ) that would characterize the connection among @xmath5 , @xmath112 and @xmath113 , furthermore the relation connecting these quantities depends on the 4-vectors @xmath114 and @xmath115 .
in addition this connection depends on the details or assumptions of the transport model .
the simple models @xcite provide examples for such a dependence .
if @xmath116 is known , then for baryon free matter we can determine four unknowns : @xmath115 and , an additional parameter of the post fo distribution , from eq .
( [ tconserve ] ) .
( due to normalization only 3 components of @xmath115 are unknowns . ) for baryon - rich matter we can determine one more unknown parameter , since we have one additional equation , the conservation of baryon charge from eq .
( [ nconserve ] ) .
the first step of solution can be done similarly to the isotropic case . then in eq .
( [ em - curr ] ) the enthalpy will change as @xmath117 , and @xmath118 , plus an additive term will appear , @xmath119 .
furthermore , eqs .
( [ eq0]-[eq3 ] ) remain of the same form , with @xmath120 , and @xmath113 , plus the additive term @xmath119 will appear in the r.h.s . of eqs.([eq0]-[eq3 ] ) .
this additive term will also appear in the expression of @xmath121 after eq .
( [ eqg2a ] ) and in the denominator of eq .
( [ eqg2b ] ) also .
the additional term , @xmath119 , in eq . ( [ em - curr ] ) is orthogonal to @xmath27 ( by definition of @xmath122 ) , so when we calculate the scalar product ( [ aa ] ) their cross term vanishes , so @xmath123 now one can express @xmath124 from eq .
( [ ads - p ] ) and inserting it to eq .
( [ aa - p ] ) , we obtain that @xmath125 where this equation is not a scalar equation as it dependes on @xmath126 , where the projector is dependent on @xmath27 .
these equations are similar to the ones obtained for the isotropic case , however , to solve this last equation we need a more complex relation among @xmath5 , @xmath112 , @xmath113 .
as these arise from the collision integral in the bte approach the needed relation may depend on @xmath127 and @xmath109 . on the other hand ,
the escape probability may be simple , or may be approximated in a way , which yields an ansatz for this relation with adjustable parameters , and then the problem is solvable .
this was the case in refs .
@xcite .
the recent covariant formulation of the kinetic freeze - out description @xcite indicates that the relation among the different parameters of the anisotropic energy momentum tensor , should be possible to express in terms of invariant scalars , which may facilitate the solution of the anisotropic fo problem .
when the adjustable parameters of the post fo matter are determined in this way from the conservation laws , we still need the underlying anisotropic momentum distribution of the emitted particles in order to evaluate the final particle spectra using the cooper - frye formula with this anisotropic distribution function . once again
, when all conservation law equations are solved we have to check the non - decreasing entropy condition to see whether the selected fo hypersurface is realistic . in case of an anisotropic final state , due to the increased number of parameters and their more involved relations , the covariant treatment of the problem may not provide a simplification , compared to the direct solution of conservation laws for each component of the energy momentum tensor ( e.g. @xcite ) .
recent viscous fluid dynamical calculations evaluate the anisotropy of the momentum distribution is in the pre fo viscous flow ( see e.g. @xcite . )
this anisotropy is governed by the spacetime direction of the viscous transport .
the pre and post fo matter may still be different , e.g. the pre fo state may be viscous qgp with current quarks and perturbative vacuum , while post fo we may have a hadron gas or constituent quark gas .
the final state will also be anisotropic , not only because of the initial anisotropy but also due to freeze - out .
the two physical processes leading to anisotropy are independent , so their dominant directions are in general different . in this case
the general symmetries are uncorrelated and can not be exploited to simplify the description of the transition . due to the change of the matter properties , the conservation laws , eqs .
( [ nconserve]-[entropy ] ) , are needed to determine the parameters of the post fo matter before the cooper - frye formula with non - equilibrium post fo distribution is applied to evaluate observables .
in this work a new simple covariant treatment is presented for solving the conservation laws across a transition hypersurface .
this leads to a significant simplification of the calculation if both the initial and final states are in thermal equilibrium .
the same method can also be used for the more complicated anisotropic final state , however , this method is only advantageous if the more involved relations among the parameters of the post fo distribution and the distribution itself is given in covariant form , preferably through invariant scalars .
e. molnar , l. p. csernai , v. k. magas , a. nyiri , k. tamosiunas , phys . rev . *
c74 * , 024907 ( 2006 ) ; j. phys .
g 34 ( 2007 ) 1901 ; e. molnar , l. p. csernai and v. k. magas , acta phys . hung . a * 27 * , 359 ( 2006 ) ; v. k. magas , l. p. csernai and e. molnar , acta phys . hung
a * 27 * , 351 ( 2006 ) .
v. k. magas , l. p. csernai and e. molnar , eur .
phys . j. a * 31 * , 854 ( 2007 ) ; int .
j. mod .
e * 16 * , 1890 ( 2007 ) ; v. k. magas and l. p. csernai , phys .
b * 663 * , 191 ( 2008);l.p .
csernai , v.k .
magas , e. molnar et al . ,
j. c 25 , 65 ( 2005);v.k .
magas , l.p .
csernai , e. molnar et al .
phys . a 749 , ( 2005 ) .
note , that we actually use for hypersurface construction in 1 + 1 , 2 + 1 , and 3 + 1 dimensional numerical simulations the corresponding computer codes , i.e. , diconex , vesta , and steve , respectively .
@xcite explains in great detail the extraction of an oriented fo contour which is represented by a set of contra - variant ( so - called `` diconex iso - contour '' ) vectors . in 2d ,
the simplices which represent a hypersurface best are line segments , whereas in 3d and 4d they are triangles @xcite and tetrahedrons @xcite , respectively .
in particular , the contra - variant 2d fo contour vectors are oriented counter - clockwise around the enclosed spacetime regions .
the co - variant normals of the contra - variant simplices are obtained from calculating the mathematical duals of these simplices with respect to a geometric product ( _ cf .
_ , e.g. , ref .
@xcite ) within the n - dimensional multi - linear space under consideration . note , that the co - variant normal vectors do not depend on any given metric tensor , whereas the contra - variant normal vectors do @xcite .
bass , a. dumitru , m. bleicher et al .
c60 , 021902 ( 1999 ) ; d. teaney , j. lauret and e.v .
shuryak , nucl .
phys . a 698 , 479 ( 2002 ) ; s.a .
bass , t. renk , j. ruppert et al . , j. phys .
g 34 , s979 ( 2007 ) ; c. nonaka , m. asakawa and s.a .
bass , j. phys .
g 35 , 104099 ( 2008 ) ; h. petersen , j. steinheimer , g. burau , m. bleicher , h. stcker , phys . rev .
c 78 ( 2008 ) 044901 ; t. hirano and y. nara , phys . rev .
c 79 , 064904 ( 2009 ) . | heavy ion reactions and other collective dynamical processes are frequently described by different theoretical approaches for the different stages of the process , like initial equilibration stage , intermediate locally equilibrated fluid dynamical stage and final freeze - out stage . for the last stage the best known is the cooper - frye description
used to generate the phase space distribution of emitted , non - interacting , particles from a fluid dynamical expansion / explosion , assuming a final ideal gas distribution , or ( less frequently ) an out of equilibrium distribution . in this work
we do not want to replace the cooper - frye description , rather clarify the ways how to use it and how to choose the parameters of the distribution , eventually how to choose the form of the phase space distribution used in the cooper - frye formula .
moreover , the cooper - frye formula is used in connection with the freeze - out problem , while the discussion of transition between different stages of the collision is applicable to other transitions also .
more recently hadronization and molecular dynamics models are matched to the end of a fluid dynamical stage to describe hadronization and freeze - out .
the stages of the model description can be matched to each other on spacetime hypersurfaces ( just like through the frequently used freeze - out hypersurface ) .
this work presents a generalized description of how to match the stages of the description of a reaction to each other , extending the methodology used at freeze - out , in simple covariant form which is easily applicable in its simplest version for most applications . |
over the past few years it has been shown that a lepton asymmetry can be generated by the mixing of heavy majorana neutrinos .
majorana neutrinos have the remarkable property of being mixed states of particles and antiparticles . by definition
they induce @xmath0 and consequently @xmath1 transitions . in addition
they could have couplings to scalar particles , which allow them to decay into higgs particles and leptons , i.e. @xmath2 @xcite . in these models
@xmath3violation is introduced through complex couplings and appears in the self - energy @xcite or vertex corrections @xcite .
we shall classify the effects using the terminology of @xmath4 mesons @xcite . we shall call direct or @xmath5type effects , the @xmath3violation which arises from the interference of tree diagrams with vertex corrections .
similarly we call indirect or @xmath6type effects , those which arise through the self - energies . the self - energies play the role of the box diagram in the @xmath4 system and define the physical states
. a consequence of the observation in @xcite is that the physical majorana states @xmath7 are not @xmath8 eigenstates and that the @xmath3asymmetries produce observable effects .
finally , the mass splitting between the states also plays a role and when the mass difference is of the order of the width there is a resonance enhancement @xcite
. the processes under discussion must satisfy unitarity constraints so that the sum of the probabilities for all transitions to and from a state @xmath9 should sum to one and yields @xcite : @xmath10 for our case this means that when we consider the scatterings @xmath11 equation ( [ uni ] ) turns into @xmath12 so that the probabilities for the direct and inverse process should be equal . for the early universe a summation over all possible physical states with boltzmann factors weighting the inital states
is required .
boltzmann factors are introduced according to the thermal properties of the universe . at thermal equilibrium , for example , the weighting factors are all equal , which implies that the lepton asymmetry averages to zero @xcite .
this issue and its implications are explicitly proved in @xcite and further discussed in @xcite .
the generation of a lepton or baryon asymmetry is a combination of the sakharov effects @xcite ( i ) baryon or lepton violation , ( ii ) @xmath13 and @xmath3violation and ( iii ) the thermodynamic properties of the ensemble of particles in the early universe .
for this reason several early papers studied the non - equilibrium equations of the ensemble @xcite .
these thermal considerations together with several subtleties of the phenomenon are mentioned in @xcite . in this article
we present an explicit calculation of the modification of the unitarity constraints as implied by the thermodynamics of the system and point out the meaning of the various terms .
in fact we shall show that the s - channel physical poles play a special role and are responsible for the development of asymmetries . to begin with we consider _ massive unstable particles _ which interact and decay . as the particles interact
some of them occur in intermediate states .
when the lifetimes of the particles are long relative to the interaction times , the intermediate states interact many times with the thermal bath . for this reason
they should not be considered as virtual states , but as ensembles of particles with their own thermodynamic distributions .
this phenomenon modifies the unitarity sum , i.e. the various terms in equation ( [ wei ] ) are weighted by different factors , and a lepton asymmetry is created .
we estimate the collision of the particles in relation to their lifetimes . the width of the particles with energy @xmath14 and mass @xmath15 is given by @xmath16 their frequent interactions are like @xmath17 with a higgs particle exchanged in the t - channel .
the cross - section at temperatures @xmath18 is @xmath19 we now calculate the interaction of one particle @xmath20 with a density of quarks . at a temperature @xmath21 and assuming boltzmann statistics the particle density is given by @xmath22 therefore the interaction rate for relativistic velocities is @xmath23 thus there are many more interactions per decay width @xmath24 or larger .
since many scatterings are taking place during the lifetime of the particles , each state has a different history and the particle states are incoherent . at this epoch
when we consider many scatterings , decays and recombinations of particles , there is no asymmetry , because the unitarity of the s - matrix guarantees the equality of direct and inverse processes .
an asymmetry is generated when specific channels begin to decouple .
as usual we work on an extension of the standard model where we include one heavy right - handed majorana field @xmath25 per generation of light - lepton @xmath26 the new fields are singlets with respect to the standard model @xcite .
extensions of this model have been constructed in gauge theories @xcite , supersymmetric theories @xcite and theories with electroweak singlet neutrinos @xcite .
we consider the contribution from the interference of the diagrams in figure 1 , which appears in the asymmetry of the reaction @xmath27 in the figure , we have indicated the momenta explicitly and have taken the absorptive part of the self - energy .
in addition we consider many generations indicated by the indices @xmath28 and @xmath29 for the external particles , the index @xmath30 for the loop - momentum and @xmath31 and @xmath32 for the heavy particle propagators .
the contribution to the asymmetry is @xmath33 with @xmath34 is the loop diagram and @xmath35 is the born diagram amplitude ( see figure 1 ) .
carrying out the same calculation for the process @xmath36 we obtain the following expression for the asymmetry : @xmath37 in this expression we kept the propagator field for all intermediate states but omitted their widths .
the two expressions are the same except for the factor of the coupling constants .
a careful comparison of the two expressions shows that they become identical when we sum over the intermediate states and over all external particles .
this can be seen by rearranging of the indices .
it is a consequence of unitarity mentioned above .
the same property persists when we keep all the masses of the propagators or when we replace the propagators with @xmath6functions in the narrow - width approximation .
the two expressions @xmath38 and @xmath39 are different when we select specific components .
this is in fact what happens in the development of the universe , as the temperature reaches the mass of a heavy neutrino specific components are multiplied by special boltzmann factors , which single them out as demonstrated below .
in this section we explicitly discuss the thermodynamic development of the system . in this way
we explain how deviations from thermal equilibrium modify the sum of unitary contributions and render them different from zero .
the lepton asymmetry is a sum of several terms : first , there are the decays of the heavy majorana neutrinos : * @xmath40 which creates an excess of leptons , and * @xmath41 which reduces the amount of leptons and the corresponding recombination terms . in addition
there are two scattering processes : * @xmath42 which creates an excess of leptons , and * @xmath43 which reduces the amount of leptons .
the result of all these terms is the following differential equation for the lepton asymmetry density @xmath44 @xmath45 \lambda^{3}_{12 } \label{bolle } \\ & & + \left [ - f_\ell f_{\phi^\dagger } |{\cal m}(\ell \phi^\dagger \rightarrow \psi_1)|^2 + f_{\ell^c } f_\phi |{\cal m}(\ell^c \phi \rightarrow \psi_1)|^2 \right ] \lambda_{12}^{3 } \nonumber \\ & & + 2 \ , \lambda_{12}^{34 } \left\ { f_{\ell^c } f_\phi \left [ |{\cal m}(\ell^c \phi \rightarrow \ell \phi^\dagger)|^2 - |{\cal m}_{ris}(\ell^c \phi \rightarrow \ell \phi^\dagger)|^2 \right ] \right .
\nonumber \\ & & - \left .
f_\ell f_{\phi^\dagger } \left [ |{\cal m}(\ell \phi^\dagger \rightarrow \ell^c \phi)|^2 - |{\cal m}_{ris}(\ell \phi^\dagger \rightarrow \ell^c \phi)|^2 \right ] \right\ } \nonumber\end{aligned}\ ] ] where @xmath46 is the four - particle phase space factor depending on the momenta @xmath47 to @xmath48 @xmath49 and @xmath50 are the boltzmann distributions for leptons and scalar particles , respectively .
@xmath51 is the distribution for the heavy majorana states .
since the @xmath52s are produced copiously through the multiple scatterings as discussed in the previous section they have their own distribution . for the decays and recombinations
we introduce the three - particle phase space @xmath53 the first two lines already include the decays of the real intermediate states ( ris ) and their recombinations . for this reason they are subtracted from the scattering amplitudes .
the interference of the tree with the one loop graph gives the @xmath3violating factor : @xmath54 = ( \varepsilon^\prime + \delta ) \
, |{\cal m}_0|^2\ ] ] with @xmath55 being the tree level amplitude .
both @xmath56 and @xmath57 are @xmath8 violating parameters with @xmath56 is produced by the vertices ( direct @xmath3violation ) and @xmath57 from the self energies . combining the terms and neglecting those ones of @xmath58 ^ 2 , \left [ \mu ( \varepsilon^\prime + \delta ) \right]/t),$ ]
the first two lines of equation ( [ bolle ] ) give : @xmath59 where the phase space was combined with the amplitude @xmath55 to produce the thermally - averaged decay width @xmath60 next we consider the last two lines of equation ( [ bolle ] ) , where we expand the particle distributions in powers of @xmath61 @xmath62 \label{firle } \\ & = & 2 \ , \lambda_{12}^{34 } \
, f^{eq}_{\psi_1 } \left [ ( 1 - \frac{\mu}{t } ) |{\cal m}(\ell^c \phi \rightarrow \ell \phi^\dagger)|^2 - ( 1 + \frac{\mu}{t } ) |{\cal m}(\ell \phi^\dagger \rightarrow \ell^c \phi)|^2 \right .
\nonumber \\ & & + \left .
( 1 + \frac{\mu}{t } ) |{\cal m}_{ris}(\ell \phi^\dagger \rightarrow \ell^c \phi)|^2 - ( 1 - \frac{\mu}{t } ) \right ] \nonumber \\ & & + o\left ( \frac{\mu^2}{t^2 } \right ) \nonumber \end{aligned}\ ] ] the leading terms proportional to unity involve the difference of the complete amplitudes @xmath63 and vanish by virtue of _
unitarity_. this is the only place where unitarity is effective in bringing a complete cancellation .
the remaining terms arise from interactions which change the number of particles ( chemical potentials ) and through deviations from equilibrium .
the former is the origin of terms proportional to the chemical potentials which are multiplied with @xmath3 conserving amplitudes .
the leading term from the real intermediate states is of special interest , because it is twice as big as the second term in equation ( [ wild ] ) , but has the opposite sign .
its net effect is to change the sign of the second term in equation ( [ wild ] ) .
after some algebra , described in the appendix , equation ( [ firle ] ) leads to : @xmath64 \nonumber \quad .\end{aligned}\ ] ] we can substitute the chemical potential by the density for the lepton asymmetry using the equation @xmath65 combining the various terms we arrive at the final boltzmann equation : @xmath66 \langle \gamma_{\psi_1 } \rangle - \frac{1}{2 } \left ( \frac{n_l}{n_\gamma } \right ) n^{eq}_{\psi_1 } \langle \gamma_{\psi_1 } \rangle \label{abi } \\ & & - n_l \
, n_\gamma \left [ \langle \sigma^\prime(\ell \phi^\dagger \rightarrow \ell^c
\phi ) \cdot v \rangle + \langle \sigma^\prime(\ell^c \phi \rightarrow \ell \phi^\dagger ) \cdot v \rangle \right ] \nonumber \end{aligned}\ ] ] this is coupled with the differential equation for the density of the majorana neutrinos , which reads : @xmath67 the integration is now straight forward
. we can integrate equation([mama ] ) , then introduce the answer in equation ( [ abi ] ) whose numerical integration gives the lepton asymmetry .
the terms proportional to @xmath68 in equation ( [ abi ] ) are multiplied by negative coefficients and after integration , if they were alone , they would wash out any lepton asymmetry exponentially .
following the usual method to evaluate the coupled boltzmann equations @xcite we introduce the dimensionless quantity @xmath69 to change the integration variable from the time @xmath70 to @xmath71 which has a temperature dependence .
furthermore it is convenient to look at the particle density per comoving volume , which is : @xmath72 with @xmath73 being the entropy density and @xmath74 the number of effectively massless degrees of freedom ( particles with mass @xmath75 ) .
now the differential equations ( [ abi ] ) and ( [ mama ] ) read : @xmath76 with @xmath77 @xmath78 where @xmath79 are the modified bessel functions @xcite , and @xmath80 here we approximate the cross - section terms as in @xcite . for a detailed calculation of the cross - section terms
see @xcite .
the overall factor @xmath81 is an important measure for the efficiency to create a lepton asymmetry . in the case
@xmath82 which corresponds to @xmath83 the decay rate is much smaller than the expansion rate of the universe and the particles come out of equilibrium and create a lepton asymmetry . in the other case , where @xmath84 the particle decay fast and recombine and are in thermal equilibrium in comparison to the expansion rate of the universe . in this case
the lepton asymmetry will go to zero .
figure 2 shows this dependence of the lepton asymmetry , which we obtain by integrating the boltzmann equations ( [ horn ] ) numerically .
the shape of the curves is in direct correspondence to @xcite , but the asymptotic value of the curves is now bigger , because of the @xmath3parameter @xmath57 coming out of the interference of the tree with the self energy diagram : @xmath85 if @xmath86 increases the asymptotic value of the lepton asymmetry decreases and finally for @xmath87 goes to zero .
similar numerical results were obtained in @xcite where the energy scale for the generation of the lepton asymmetry was studied .
we have shown explicitly how a lepton asymmetry is created through the decay of majorana neutrinos .
the asymmetry is proportional to the sum of the self energy and the vertex contribution .
we emphasize that at thermal equilibrium , the unitarity of the amplitudes implies the vanishing of the lepton asymmetry , as it is demonstrated in section 1 . on the other hand ,
the thermal development of the majorana densities is the reason for the unitarity cancellation to become ineffective and allow the generation of the lepton asymmetry ( see section 2 ) .
finally , we solve the boltzmann equations numerically and create a final asymmetry , which is proportional to the sum of the @xmath3parameters coming from the vertex contribution @xmath56 and from the self energy @xmath88 the latter one has a resonant behaviour depending on the mass difference of the majorana particles @xcite .
we wish to thank a. pilaftsis for helpful communications on unitarity and the incoherence of the physical states .
helpful discussions with a.d .
dolgov and a. joshipura are gratefully acknowledged .
the problem of mixing of majorana neutrinos has been treated at several places .
one formalism is to discuss the development of the various components in the wavefunction , where the absorptive part has a definite effect @xcite .
another method is to construct the inverse propagator and their renormalization @xcite . in both cases
the absorptive part is a physical observable , which can not be removed by renormalization @xcite .
we follow here the approach of references @xcite , where to @xmath89 the @xmath3effects were computed exactly for all relative values of the masses @xcite .
we assume a mass hierarchy and consider an epoch in the development of the universe , where the majorana neutrinos are decoupled from each other .
when the temperature of the universe becomes of the order of the lightest mass @xmath90 there is only this particle left , which can only decay bringing the universe out - of - equilibrium .
the physical heavy majorana neutrino @xmath91 is a mixed state of the interaction states : @xmath92 it has a definite mass and equation ( [ trudy ] ) gives the couplings of the various components .
the transition amplitudes of the mixed state receive contributions from the vertex corrections through the @xmath56 and from the self energies through the mixing of the states : @xmath93 @xmath94 with @xmath95 these are the relevant amplitudes for the boltzmann equation ( [ bolle ] ) , which are given now by : @xmath96 \nonumber \\ & & + 2 \ , \lambda_{12}^{34 } \left\ { f_{\ell^c } f_\phi \left [ |{\cal m}(\ell^c \phi \rightarrow \ell \phi^\dagger)|^2 - |{\cal m}_{ris}(\ell^c \phi \rightarrow \ell \phi^\dagger)|^2 \right ] \right .
\nonumber \\ & & - \left .
f_\ell f_{\phi^\dagger } \left [ |{\cal m}(\ell \phi^\dagger \rightarrow \ell^c \phi)|^2 - |{\cal m}_{ris}(\ell \phi^\dagger \rightarrow \ell^c \phi)|^2 \right ] \right\ } \nonumber\end{aligned}\ ] ] next we substitute the phase space densities and develop them in the parameter @xmath97
@xmath98 \nonumber \\
f_{\ell^c } f_{\phi } & = & f_{\psi_1}^{eq } e^{-\mu / t } = f_{\psi_1}^{eq } \left [ 1 - \frac{\mu}{t } + o\left ( \frac{\mu^2}{t^2 } \right ) \right ] \end{aligned}\ ] ] as usual we assume that the particles are in kinetic equilibrium . finally , we use the definition of the thermally - averaged decay width @xmath99 which is : @xmath100 where @xmath101 are the degrees of freedom and @xmath102 is the particle density . with all this information we arrive at equation([wild ] ) and equation ( [ firle ] ) . in the next section
we study the scattering terms in the boltzmann equation .
in this section we study the scattering terms in the boltzmann equation : @xmath103 \right\ } \label{fifi}\end{aligned}\ ] ] with @xmath104 for the real intermediate states one can use the narrow - width approximation : @xmath105 and the definition of the thermally - averaged cross section @xmath106 as well as the approximations @xmath107 expression ( [ kai ] ) can be derived from equation ( [ fifi ] ) . | we work out the thermodynamic equations for the decays and scatterings of heavy majorana neutrinos including the constraints from unitarity .
the boltzmann equations depend on the cp asymmetry parameter which contains both , a self - energy and a vertex correction . at thermal equilibrium
there is no net lepton asymmetry due to the cpt theorem and the unitarity constraint .
we show explicitly that deviations from thermal equilibrium create the lepton asymmetry .
further considerations on the cp asymmetry in + heavy majorana neutrino decays .75 in marion flanz , emmanuel a. paschos _ institut fr physik _ , + _ universitt dortmund _ , + _ 44221 dortmund , germany _
+ .75 in |
cosmological models obtained by dimensional reduction of 10-dimensional supergravity theories are an active field of research @xcite@xcite .
since supergravity is the low energy limit of string theory , these models should provide a way towards understanding the quantum effects in cosmology .
so far , there has been much activity in investigating models including fields of the ns - ns - sector .
the discovery , that the conformal field theory of p - brane - like superstring backgrounds charged under rr - gauge fields is that of open strings with dirichlet boundary conditions @xcite , has made the rr - field contents of type ii string theories a very interesting subject for string cosmological models .
recently , in a first study of cosmological models including rr - fields , it was found that inflating phases can occur in a certain range of parameters which label the solutions to the equations of motion @xcite .
the ns - ns-3-form field strength had been set to zero in the model and the compactification ansatz was chosen to contain a number of maximally symmetric flat subspaces .
the two branches of the solutions were separated and curvature singularities appeared in each branch . in this paper
we construct curved 4-dimensional cosmological models containing a ns - ns-3-form field strength as well as nontrivial type iib rr - fields . to construct the 4-dimensional model
, we use a 5-brane - solution of type iib supergravity whose rr - sector consists of a 3-form and a scalar . by sl(2,r)-invariance of the action ,
which is realized as a sl(2,z)-subgroup in iib superstring theory , we generate 5-brane - solutions with nontrivial rr - fields and compactify the transverse coordinates on a 5-torus , giving a black hole in 5 dimensions .
after generalizing the black hole solution to constant curvature solutions we compactify to 4 dimensions , obtaining isotropic frw - metrics together with nontrivial rr - fields .
the outline of the paper is as follows : in section 2 we introduce the sl(2,r)-transformed 5-brane - solution of type iib supergravity and compactify it on a 5-torus . in section 3
we make one further compactification to 4 dimensions and obtain frw - models including nontrivial rr - fields in this way . in section 4 we investigate the influence of the rr - fields on the cosmological models with a detailed analysis of the inflationary phase .
section 5 is devoted to the extremal limit of the model .
our starting point is the bosonic part of the 10-dimensional sl(2,r)-invariant type iib supergravity action @xcite [ iib ] s_10=d^10x \ { e^-2 ( r + 4 ( ) ^2 - h_1 ^ 2 ) - ( ) ^2 - ( h_1 + h_2)^2 } . in this action
the string frame metric @xmath2 , the dilaton @xmath3 and the three form field strength @xmath4 are the familiar fields contained in the ns - ns sector of all 10-dimensional string theories .
the three form field strength @xmath5 and the scalar @xmath6 belong to the rr - sector .
the bosonic field content of iib superstring theory also includes a selfdual five form field strength , which we have set to zero in order to be able to write a covariant action . a 5-brane - solution to this action is @xcite [ sol10 m ] ds_(10)^2= - dt^2 + + r^2 d_3 ^ 2 + _
ijdx^idx^j for the 10-dimensional line element in the string frame and [ sol10f ] = -(1-(r_-/r)^2 ) , h_1=q_3,h_2==0 , for the remaining background fields .
the magnetic charge is @xmath7 and @xmath8 and @xmath9 are the line element and volume element of a 3-sphere , respectively , i.e d_3&=&d^2+^2(d^2+^2d^2 ) + _ 3&=&^2ddd .
the rr - fields @xmath10 and @xmath6 are not excited in this solution .
we use sl(2,r)-invariance of the action ( [ iib ] ) to generate nontrivial rr - fields , making the solution ( [ sol10m],[sol10f ] ) an intrinsic iib solution .
a manifest sl(2,r)-invariant form of ( [ iib ] ) was given in @xcite and also used there to construct sl(2,r)-transformed 5-brane - solutions ( see also @xcite for black hole solutions of type ii stringtheory ) .
we do not hodge - dualize the rr-3-form here , instead we construct a solution possessing two magnetic charges . following @xcite , the action of the sl(2,r)-matrix @xmath11 on the background fields is & = & + e^- & = & + h_1 & = & dh_1 + h_2 & = & -bh_1 .
this transformation clearly includes s - duality for the choice @xmath12 , which interchanges the two 3-forms and inverts the string coupling @xmath13 . in detail , the transformed scalar fields are [ sl2sol10f ] &= & + &=&-(1-(r_-/r)^2)+ ( d^2+c^2(1-(r_-/r)^2 ) ) and the string frame line element changes to [ sl2sol10 m ] ds^2_(10)=ds_(10)^2 .
the einstein frame metric is not changed by the sl(2,r ) transformation .
we compactify this solution to five dimensions by wrapping the five transverse coordinates of the 5-brane over a 5-torus , i.e. we make the ansatz for the string frame metric ds^2=_dx^dx^+ e^a _
ijdx^i dx^j , where @xmath14 is a 5-dimensional lorentz metric , @xmath15 are coordinates on the 5-torus and @xmath16 depends on the @xmath17 only .
we neglect the moduli coming from compactifying the three forms as well as vector fields coming from the metric .
the effective 5-dimensional action is [ 5dstring ] s_5&=&d^5x \{e^- ( + ( ) ^2 - ( ) ^2 - h_1 ^ 2 ) - + & & - e^ ( ( ) ^2 + ( h_1+h_2)^2 ) } , where we have used the rescaled modulus field @xmath18 and the 5-dimensional dilaton @xmath19 is given by @xmath20 . from the transformed 5-brane - solution ( [ sl2sol10f ] , [ sl2sol10 m ] ) we get a 5-dimensional solution [ sl2sol5string ] ds_5 ^ 2&= & + h_1&=&dq_3,h_2=-bq_3 + & = & + & = & ( d^2 + c^2(1-(r_-/r)^2 ) ) + & = & . for @xmath21
the identity transformation , i.e. vanishing rr - fields @xmath10 and @xmath6 , this is just the 5-dimensional black hole solution of @xcite , written in the string frame . in these equations @xmath22 and @xmath23
are the values of the parameter @xmath24 at the outer and inner horizon , respectively .
the metric has a curvature singularity at @xmath25 and also at the spacelike surface @xmath26 whereas @xmath22 is a regular event horizon .
this is a black hole solution for @xmath27 only , in the other case the solution describes a naked singularity .
a particular interesting solution is the extremal limit at @xmath28 .
we will discuss the non - extremal and extremal cases in a cosmological context in the next sections .
following @xcite , where this solution was generalized to constant curvature spaces , we can produce a 4-dimensional cosmological solution from this black hole by kaluza - klein compactification of the time coordinate and re - interpretation of the radius coordinate as 4-dimensional time . in order to get the right signature after compactification we must change signature from @xmath29 to @xmath30 in the 5-dimensional solution .
the generalization of ( [ sl2sol5string ] ) to constant curvature spaces is given by [ sol5gen ] ds_5 ^ 2&= & + h_1&=&dq_k , h_2=-bq_k , q=2t_+t_- + & = & + & = & ( d^2 + c^2(1-(t_-/t)^2 ) ) + & = & , where @xmath31 labels the constant curvature three - spaces with line element and volume element d_k&=&d^2+()^2 ( d^2+^2d^2 ) , + _
k&=&()^2 ddd .
it should be emphasized that the time coordinate in this solution is the radius coordinate of the 5-dimensional black hole and the signature is different from usual black hole physics .
we reduce to 4 dimensions using the ansatz ds_5 ^ 2=y^2dy^2+ds_4 ^ 2 where the @xmath32-modulus @xmath33 turns out to be y^2(t ) = . by introducing @xmath34 we arrive at the 4-dimensional action [ cosstring ] s_4&=&d^4x\ { e^-2(r+4()^2-()^2 - ( d)^2 -h_1 ^ 2 ) - + & & -e^+(()^2 + ( h_1+h_2)^2 ) } . the 4-dimensional dilaton @xmath35 is given by @xmath36 .
the 5-dimensional black hole ( [ sol5gen ] ) reduces to the 4-dimensional frw - solution [ frwstring ] ds_4 ^ 2&= & ( -+ t^2 d_k^2 ) + h_1&=&dq_k , h_2=-bq_k , q=2t_+t_- + & = & ( ) + & = & ( ) + & = & ( d^2+c^2(1-t_-^2/t^2 ) ) + & = & for the special choice of sl(2,r)-parameters @xmath37 , the rr - fields @xmath10 and @xmath6 and the @xmath38-modulus field @xmath39 are zero and we recover the cosmological models described in @xcite .
these models have been studied in detail there , both in the string and einstein frame . since the sl(2,r)-transformation does not affect the einstein metric
, we concentrate on the string frame in the following .
let us briefly recall the properties of the model in the pure ns - ns - case , i.e. without contributions from the rr - fields .
the solution describes a closed @xmath40 , open @xmath41 , or flat @xmath42 cosmology respectively . in the string frame , for @xmath43 , the three space oscillates ) . ] between the extrema @xmath44 . for @xmath45
the solution is not oscillating , the extension of the corresponding three spaces is bounded from below at @xmath46 and tends to infinity as @xmath47 . the dilaton @xmath35 is always divergent at @xmath46 , corresponding to the minimal extension of the three spaces .
for @xmath43 , the dilaton is also divergent at @xmath48 , while for @xmath45 , where the time region has no upper bound , it tends to zero for @xmath47 .
the @xmath32-modulus @xmath49 is divergent at @xmath46 , indicating decompactification to 5 dimensions at this points . for @xmath43 ,
the @xmath32-radius becomes zero at the maximum extension of three - space . in the other cases
@xmath45 , the modulus tends to a constant value as @xmath50 . in the einstein frame ,
the curvature scalar is singular at @xmath46 in all three cases and also at @xmath48 in the case @xmath43 .
the extension of three - space shrinks to zero at these points , indicating a big - crunch or big - bang .
let us now study the qualitative changes caused by adding nontrivial rr - fields to the solution described above .
as can be seen from the dilaton expression in ( [ frwstring ] ) , in the case @xmath51 one divergence cancels leaving the expression for the dilaton _ d=0= ( ) . for @xmath43
, there is still a divergence at @xmath48 , but the divergence at @xmath46 has disappeared in all cases @xmath52 .
the case @xmath51 corresponds to zero ns - ns - charge and a constant rr - scalar @xmath53 .
we want to see how the scale factors of the frw - models behave in this special case . for this
we change to a coordinate system which is usually used to study frw - cosmologies ds^2=-d^2+a^2()d_k^2 .
the comoving time coordinate @xmath54 is defined by a solution to the differential equation [ tau ] d^2=dt^2 , t(=0)=t_- and the scale factor @xmath55 is [ scalefactor ] a^2()=t^2 ( ) .
for @xmath51 the scale factor @xmath55 goes to zero at @xmath56 . calculating @xmath57 we find the two different behaviors _ 0=\ { r@:l 0 & d= 0 + & d=0 .
so by switching off the ns - ns-3-form @xmath58 completely , we produce a singularity in the string frame scale factor , whereas a finite value of the ns - ns - charge smoothes out this singularity . at the same time
we find that the @xmath38-compactification radius is zero for @xmath51 at @xmath46 , whereas the @xmath32-radius is still divergent there .
the dilaton @xmath35 takes the minimum value at @xmath46 for @xmath51 and it diverges there for @xmath59 .
the magnetic charge of the ns - ns-3-form prevents the universe from collapsing , as was already recognized in @xcite . without ns - ns - charge ,
the scale factor exhibits a singularity regardless of the existence of rr - charge .
the qualitative behavior of the dilaton - singularities does not change when rr - fields are added to a non - zero ns - ns - field strength .
the minimum extension of the universe at @xmath46 is @xmath60 and for @xmath43 the maximum extension is given by @xmath61 .
however , the duration and effectivity of a possible initial inflationary phase depend on the parameters c and d as well as on @xmath62 and @xmath63 , as can be seen from the equations ( [ tau ] ) and ( [ scalefactor ] ) .
we will discuss this dependence in detail now . in the presence of a ns - ns - charge the string frame scale factor of
the cosmology is free of singularities , although in the einstein frame the familiar curvature singularity close to the time @xmath64 appears .
it is interesting to investigate how this feature is affected by adding rr - fields and whether a phenomenologically interesting inflationary phase can occur . from a phenomenological point of view
one can argue that the einstein frame is the preferred coordinate system . on the other hand one can treat both frames on equal footing , as has been done in recent years in numerous analyses where the transition of a superinflationary pre - big bang phase into a a frw type cosmology was worked out in the string frame @xcite@xcite . in the following
we will examine in more detail the effect of rr - fields on the string frame scale factor close to the time @xmath64 , where the influence of these fields are most apparent .
it was pointed out in @xcite that in the asymptotic limit @xmath65 the expansion scales like @xmath66 and from the discussion in the previous section we know that the s - dual model contains a singularity in the scale factor at this point .
one can easily imagine that close to @xmath56 a short phase of superluminal expansion might occur in the sense of power - law inflation with @xmath67 and @xmath68 sufficiently large to ensure an expansion that could be of phenomenological interest . as a measure to describe an accelerated expansion of the universe we use the quantity @xmath69 . here
@xmath70 is the hubble parameter @xmath71 , a dot means the derivative with respect to @xmath54 and the last part of the equation holds if the expansion is of power law type , @xmath67 . in the general case
we are not able to find analytical expressions for the quantity @xmath72 , because equation ( [ tau ] ) can not be integrated directly .
instead we will give numerical solutions for a certain choice of parameters in fig .
it is shown how the acceleration changes with transformations that lie within a so(2,r)-subgroup of sl(2,r ) , parametrized by the rotation angle @xmath73 .
an increasing rotation angle corresponds to an enlargement of the fields of the rr - sector and simultaneously a reduction of the ns - ns-3-form field strength .
we observe that the initial acceleration is very much enhanced by exchanging ns - ns - charge by rr - fields but a subsequent decelerating phase becomes also gradually more important and thus taking off much of what was gained in the accelerating phase . at the s - dual point
we encounter again the transition from a non - singular to the singular universe and @xmath74 .
the number of e - foldings @xmath75 produced during the expansion increases only slightly by enlarging the rr - fields and remains of the order of one for almost all values of @xmath73 .
since @xmath75 must exceed a certain limit as a prerequisite for a viable inflationary scenario , none of the cosmologies fulfills this requirement .
an expression for the number of e - foldings produced between the time @xmath76 and @xmath77 can easily be derived from ( [ scalefactor ] ) ( t;c)= ( ) + ( d+ ( 1-()^2 ) ) . since the string frame metric
is only influenced by the sl(2,r)-parameters @xmath78 and @xmath79 , we study the case where for a fixed value of d ( fixed ns - ns - charge ) the parameter @xmath78 is increased . for the specific choice @xmath80 and @xmath78 a free parameter , it becomes apparent that the rr - fields enhances the number of e - folds only logarithmically .
however , for @xmath62 being very close to zero , @xmath75 can become sufficiently large for having cosmological significance .
upon expanding @xmath55 to second order in @xmath54 it can also be seen how the singular case is recovered , a()=t_- - ( k- ( ) ^2 ) ( 1 + ) ^2 + o ( ^4 ) .
the pre - factor of the quadratic term gets large for small @xmath62 and @xmath79 and for large @xmath78 . also , the behavior of @xmath72 with this choice of parameters is qualitatively the same as shown in fig . 1 for the so(2 )
transformed models , as well as for all other sets that have been investigated . moreover , curved spatial sections ( @xmath81 ) have only little influence on the qualitative picture .
it was shown in @xcite that for vanishing rr - fields the 5-dimensional solution ( [ sol5gen ] ) has an extremal limit @xmath82 , where the theory corresponds to a wzw model ) part . for @xmath83 ,
the 2d part corresponds to a @xmath84 coset model and for @xmath85 it is trivial . ] and is therefore exact to all orders in @xmath86 . whereas for p - brane superstring backgrounds charged under rr gauge fields the corresponding conformal field theory is known to be that of open strings with dirichlet boundary conditions @xcite
, the higher order contributions to the tree level equations of motions for non - p - brane like backgrounds still lack an explicit sigma model description @xcite . for this
the extremal limit described above may not remain exact after performing an sl(2,r ) transformation .
the corresponding 4d solution , valid for arbitrary @xmath87 , is ds^2&=&t_-^2 ( -d^2+d^2_k ) + & = & ( ( ) ^2 ) + & = & - ( ) + & = & ( d^2+c^2 ( ) ^2 ) + & = & + h_1&=&2d t_-^2_k , h_2=-2b t_-^2_k in this solution , @xmath88 is the conformal time defined by the transformation [ conftime ] t^2=t^2_-+(t_+^2-kt_-^2)()^2 and a possible constant part of the 5-dimensional dilaton in ( [ sol5gen ] ) has to be adjusted @xcite _ 0 _ 0 + ( t_+^2-kt_-^2 ) in order to be able to perform the extremal limit . for vanishing rr - fields ,
we recover again the solution given in @xcite . in this limit ,
the geometry is simply @xmath89 for all three cases ( see figure 2a ) , where @xmath90 stands for a sphere , pseudo - sphere or flat space respectively .
adding rr - fields to the solution , the universe behaves different for different values of @xmath87 . for @xmath91 ,
the conformal factor is oscillating between @xmath92 and @xmath93 .
in the other cases the universe approaches the minimum size @xmath92 for @xmath94 and then expands for ever . especially for the s - dual case , where the ns - ns - field @xmath58 is switched off , the corresponding solutions exhibit again singularities ( see figures 2b - d ) .
2a 2b + 2c 2d
we constructed 4-dimensional frw - cosmologies from 5-dimensional black hole solutions following @xcite .
our models contain nontrivial rr - fields , a 3-form and a scalar , which are generated by sl(2,r ) transformations of a black 5-brane solution of type iib supergravity .
the effect of this rr - fields on the resulting cosmologies have been studied . for the s - dual model , where the ns - ns - charge is replaced by the corresponding rr - charge , the dilaton singularity at the minimum extension of spatial 3-space disappears , but
the model still decompactifies to 5 dimensions there .
also , other than the ns - ns - field strength , the rr - field strength can not prevent the universe from collapsing , and we find a singularity in the string frame scale factor , which is absent when the ns - ns - charge is different from zero .
it was further shown that the influence of rr - fields on the cosmic expansion results in a large increase of the initial accelerating phase .
it is however followed by a subsequent deceleration phase that increases in a similar way .
for that reason the number of e - folds produced during the combined phases increases only slightly , almost logarithmically , with the rr - fields being turned on .
thus , such models do not seem to be suitable for an explanation of spatial homogeneity and isotropy , especially that of the cmb radiation , based on inflationary cosmology .
however , for a small fraction of parameter space , the superluminal expansion of the physical space is effective enough to solve some of the problems inflationary cosmology was invented for .
we examined also the extremal limit of the model , for which the 5-dimensional black hole is on the verge of becoming a naked singularity . since we generalized
the 5-dimensional solution to a constant curvature solution , it is not clear whether the successful description of extremal black holes as being composed of non - interacting branes , anti - branes and momentum @xcite can be applied to this extremal limit .
the effect of rr - fields added to the extremal limit of the pure ns - ns case is to render dynamics to the steady universe .
for the s - dual case we find again singularities in the string frame scale factor . during writing this paper @xcite appeared , where also cosmological models from toroidally compactification of type ii theories are discussed .
we would like to thank s.frste and r.dick for useful discussions .
this work was supported by the `` sonderforschungsbereich 375 - 95 fr astro - teilchenphysik der deutschen forschungsgemeinschaft '' and the european commission tmr programs erbfmrx - ct96 - 0045 and erbfmrx - ct96 - 0090 .
the work of s.s . is supported by deutsche forschungsgemeinschaft .
99 a.a.tseytlin , int . j. mod
d 1 ( 1992 ) 223 i.antoniadis , j.rizos , k.tamvakis , nucl .
b 415 ( 1994 ) 497 e.j.copeland , a.lahiri , d.wands , phys . rev .
d 51 ( 1995 ) 1569 r.easther , k.maeda , hep - th/9605173 , to be published in phys . rev .
d m.gasperini , g.veneziano , astropart .
phys . 1 ( 1993 ) 317 m.gasperini , g.veneziano , mod .
a(1993 ) 3701 m.gasperini , g.veneziano , phys . rev .
d 50 ( 1994 ) 2519 r.brustein , g.veneziano , phys .
b 329 ( 1994 ) 429 n.kaloper , r.madden , k.a.olive , nucl . phys .
b 452 ( 1995 ) 677 n.kaloper , r.madden , k.a.olive , phys . lett .
b 371 ( 1996 ) 34 r.easther , k.maeda , d.wands , phys . rev .
d 53 ( 1996 ) 4247 n.kaloper , mcgill-96 - 34 , hep - th/9609087 j.polchinski , phys . rev .
lett . 75 ( 1995 )
4724 j.polchinsky , s.chaudhuri , c.v.johnson , nsf - itp-96 - 003 , hep - th/9602052 a.lukas , b.ovrut , d.waldram , upr-711 t , hep - th/9608195 j.h.schwarz , phys .
b 360 ( 1995 ) 13 , erratum - ibid .
b364 ( 1995 ) 252 g.w.gibbons and k.maeda , nucl.phys .
b 289 ( 1988 ) 741 g.t.horowitz , a.strominger , nucl . phys b 360 ( 1991 ) 197 e.bergshoeff , j.boonstra , phys .
d 53 ( 1996 ) 7206 a.a.tseytlin , mod .
a11 ( 1996 ) 689 g.t.horowitz , j.m.maldacena , a.strominger , phys .
b 383 ( 1996 ) 151 m. cveti , c.m.hull , damtp - r-96 - 31 , hep - th/9606193 , + m. cveti , d. youm , nucl . phys .
b 476 ( 1996 ) 118 k. behrndt and s. frste , nucl . phys .
b 430 ( 1994 ) 441 k.behrndt , slac - pub-6640 , hep - th/9408137 g.sengupta , imsc-96 - 09 - 27 , hep - th/9609152 g.t.horowitz , s.strominger , phys . rev .
77 ( 1996 ) 2368 a.strominger , c.vafa , phys .
b 379 ( 1996 ) 99 h.l , s.mukherji , c.n.pope , k.w.xu , ctp - tamu-51 - 96 , hep - th/9610107 | we construct 4-dimensional cosmological frw models by compactifying a black 5-brane solution of type iib supergravity , which carries both magnetic ns - ns - charge and rr - charge .
the influence of nontrivial rr - fields on the dynamics of the cosmological models is investigated .
= -1 cm -0.3 cm .2 cm lmu - tpw 96 - 27 + tum - hep-258/96 + sfb-375/122 + hep - th/9610166 + october 1996 + r. poppe@xmath0 , s. schwager@xmath1 .5 cm _ @xmath0 physik department , technische universitt mnchen + 85748 garching , germany _ .2 cm _
@xmath1 sektion physik , universitt mnchen + theresienstr .
37 , 80333 mnchen , germany _ |
complex networks are natural structures for representing many real - world systems .
their flexible and adaptive features make them ideal for interrelating various types of information and to update them as time goes by .
recently , there has been tremendous interest , from various fields of science , in the study of the statistical properties of such networks @xcite . by now , many interesting features of complex networks have been established .
so far , the main attention in the study of complex networks has been on the properties of individual nodes and how they connect ( link ) to their nearest neighbors .
however , exploring the _ local _ properties of the network outside the nearest neighbor level , has not been well studied .
a noteworthy exception is the recent study by girvan and newman @xcite .
such studies will naturally have to address the cluster ( modular ) structure of the network .
to know if a network is modular or not , is important if one tries to assess its robustness and stability , since only a few critical links are responsible for the inter - modular communication in clustered networks .
hence to make the network more robust , such critical links have to be identified and strengthen . in this work
we will address the large scale topological structures of networks .
this problem will be approached by considering an auxiliary diffusion process on the underlying complex network . as it will turn out ,
it is the slowest decaying diffusive eigenmodes that will be of most interest to us , since such modes will contain , as we will see , information about the weakly interacting modules of the network .
hence , by studying how an ensemble of random walkers slowly reaches a state of equilibrium , one can learn something about the topology of the network .
to study spectral properties of networks is not new ; its variants has previously been applied to random graphs @xcite , to social networks ( the correspondence analysis ) @xcite , random and small - world networks ( the laplace equation analysis ) @xcite , artificial scale - free networks @xcite , and community structures @xcite .
a diffusion approach has also made it to practical applications .
for instance , the analysis of a diffusion process lies at the heart of the popular search engine google @xcite .
the original physical motivation behind the method to be outlined below , was that the relaxation of some arbitrary initial state ( of walkers on the network ) toward the steady state distribution , was expected to be _ fast _ in regions that are highly connected , while _ slow _ in regions that had low connectivity . by definition ,
a module is highly connected internally , but has only a small number of links to nodes outside the module .
hence , it was reasoned , that by identifying the slowly decaying eigenmodes of the diffusive network process , one should be able to obtain information about the large scale topological structures , like modules , of the underlying complex network .
we will now formalize this idea and outline the method in some detail .
let us start by assuming that we are dealing with a ( fully connected ) complex network consisting of @xmath1 nodes and characterized by some degree distribution @xmath2 where @xmath3 denotes the connectivity of node @xmath4 .
imagine now placing a large number ( @xmath5 ) of random walkers onto this network .
the fraction of walkers on node @xmath4 ( relative the total number ) at time @xmath6 we will denote by @xmath7 . in each time step , the walkers are allowed to move randomly between nodes that are directly linked to each other .
since there is no way that a walker can vanish from the network , the total number of walkers must be conserved globally , _
i.e. _ one must have @xmath8 at all times . furthermore ,
locally a continuity equation must be satisfied ; if we consider , say node @xmath4 , that directly links to other nodes @xmath9 , one must require that ( the balance equation ) : @xmath10 in writing this equation , we have introduced the so - called adjacency matrix @xmath11 defined to be @xmath12 if node @xmath4 and @xmath9 are directly linked to each other , and @xmath13 otherwise @xcite , and @xmath14 , we recall , is the degree of node @xmath4 , _
i.e. _ its number of nearest neighbors @xmath15 .
the first term on the right hand side of eq .
( [ eq : cont ] ) describes the flow of walkers into node @xmath4 , while the last term is associated with the out - flow of walkers from the same node . as will be useful later , eq .
( [ eq : cont ] ) can be casted into the following equivalent matrix form : @xmath16 where @xmath17 , and @xmath18 is the _ diffusion matrix _ to be defined below .
( [ eq : diffusion ] ) should be compared to the continuous diffusion equation for the particle density : @xmath19 , where @xmath20 is the diffusion constant .
a complex network obviously has an inherent discreetness , and hence no continuous limit can be taken .
however , if the continuous diffusion equation is understood in its discreet form , one is lead to regard eq .
( [ eq : diffusion ] ) as the _ master equation _ for the random - walk process taking place on the underlying network . by comparing eqs .
( [ eq : cont ] ) and ( [ eq : diffusion ] ) , as well as taking advantage of @xmath21 ( because @xmath22 ) , one is lead to @xmath23 . hence , an equivalent formulation of eq .
( [ eq : diffusion ] ) , is @xmath24 where @xmath25 is the _ transfer matrix _ , related to the diffusion matrix by @xmath26 . in component form one
thus has @xmath27 . since the transfer matrix , `` transfers '' the walker distribution one time step ahead
, it can therefore be thought of as a time - propagator for the process .
formally the time development , from some arbitrarily chosen initial state @xmath28 , can be obtained by iteration on eq .
( [ eq : transfer ] ) with the result @xmath29 . here
@xmath30 means the transition matrix to the @xmath6th power . in general
, the transfer matrix ( or equivalently , the diffusion matrix ) will not be symmetric .
however , @xmath31 can be related to a symmetric matrix @xmath32 by the following similarity transformation @xmath33 with @xmath34 .
thus , @xmath25 and @xmath32 will have the same eigenvalue spectrum , and all eigenvalues @xmath35 ( of @xmath31 ) will be real .
it is this eigenvalue spectrum that will control the time - development of the diffusive process through @xmath36 with @xmath37 .
it should be noted that since the total number of walkers is conserved at all times , one must have @xmath38 , and that at least one eigenvalue should be one .
, must satisfy @xmath39 since @xmath40 . ]
we will here adopt the convention and sort the eigenvalues so that @xmath41 corresponds to the largest eigenvalue @xmath42 , @xmath43 to the next to largest one , and so on .
physically the principal eigenvalue @xmath42 , corresponds to a _ stationary state _ where @xmath44 , the diffusive current flowing from node @xmath4 to node @xmath9 is exactly balanced by that flowing from @xmath9 to @xmath4 .
the stationary state is unique for single component networks , and is fully determined by the connectivities of the network according to @xmath45 .
this can be easily checked by substituting this relation into eq .
( [ eq : transfer ] ) .
all modes corresponding to eigenvalues @xmath46 , are decaying modes since the time dependence of @xmath47 enters through @xmath36 and one recalls that @xmath48 .
notice that @xmath49 represent non - oscillatory modes , while @xmath50 correspond to states where oscillations will take place with time , but this latter possibility will not be considered here .
the large scale topology of a given complex network reflects itself in the statistical properties of its diffusion eigenvectors @xmath51 .
one such property is the participation ratio ( pr ) , that will be defined below .
it quantifies the effective number of nodes participating in a given eigenvector with a significant weight .
since the stationary state of a network depends on the connectivities of its nodes , @xmath52 , it is convenient to introduce a normalized eigenvector @xmath53 observe that @xmath54 is nothing but the walker density per link of node @xmath4 .
hence , in effect , @xmath55 represents the _ outgoing currents _ flowing from node @xmath4 , along each of its links , toward its neighbors . in the steady state , where @xmath52 , it follows that these currents are all the same for any link in the network .
hence , highly connected nodes are not treated differently from less connected nodes .
more formally , @xmath56 are the eigenvectors of the transposed transfer matrix @xmath57 corresponding to the same eigenvalue @xmath35 ( of @xmath31 ) . here
one alternatively could have defined @xmath57 ( instead of @xmath31 ) as the transfer matrix from the very beginning .
doing so , would have resulted in a master equation of the form ( [ eq : transfer ] ) , but for the currents @xmath58 with corresponding eigenmodes @xmath56 .
the physical interpretation of such an alternative equation is as follows : instead of walkers , think of a signal propagating on the underlying network .
the signal at node @xmath4 at time @xmath59 is then the average of the signal at the nearest neighbors at time @xmath6 .
if the link currents are normalized to unity , _ i.e. _ if @xmath60 in the @xmath61-norm , then the participation ratio ( pr ) is defined as @xcite @xmath62^{-1}. \end{aligned}\ ] ] for the stationary state , where all the currents are equal to @xmath63 , one has @xmath64 where we recall that @xmath1 is the total number of nodes in the network . hence , when @xmath65 , the ratio @xmath66 can be regarded as an effective number of nodes participating in the eigenvector @xmath56 with a significant weight @xcite
it should be noted that the participation ratio is a simple and crude measure of size .
strictly speaking , for this ratio to be able to say something with confidence about the size of a given module , the main contribution to @xmath66 should come from nodes within that module .
for instance , if @xmath66 gets major contributions from current elements @xmath55 of different signs , _
i.e. _ from different topological structures , it is not trivial to relate @xmath66 to the size of a single module . for all types of complex networks where a modular structure is of interest
, it is important to try to quantify the number of modules .
such information is of interest since this number may say something about the organizing principles being active in the generating process of the network .
however , the measurement of this number is often hampered by statistical uncertainties due to temporary ( non - robust ) topological structures being counted as modules .
hence the challenge is to obtain the significant ( or robust ) number of modules that are due to the `` rule of creation '' and not just happened to be there by chance . for this purpose ,
it is useful to introduce a null model a randomized version of the network at hand but so that the degree distribution of the original network is not changes @xcite .
a rough estimate of the number of different modules contained in a network could be given by the number of slowly decaying non - oscillatory modes that have a participation ratio , @xmath66 , significantly exceeding the ( ensemble ) averaged participation ration of the corresponding randomized network , @xmath67 .
hence , it is suggested that a module is significant if @xmath68 .
we will close this section by noting that the ( outgoing ) currents , @xmath69 , can also be used for vitalization purposes .
the basis of this approach is the observation that the outgoing currents on links within a module are almost constant , and the size of this constant depends on to which extent the module participates in the given eigenmode .
the constants thus varies from module to module and from eigenmode to eigenmode .
hence , by sorting @xmath55 by size , nodes belonging to the same module will be located close to each .
during a two year period in the mid 1970 s , w. zachary studies the relations among @xmath70 members of a university karate club during a period of trouble @xcite .
a serious controversy existed between the trainer of the club ( node @xmath12 in fig .
[ fig : zachary]a ) and its administrator ( node @xmath70 ) .
ultimately this conflict resulted in the breakup of the club into two new clubs of roughly the same size . in his original study @xcite , w. zachary mapped out the strength of friendship between the various members . in our study , however , we will be considering the unweighted version of his network @xcite ( see fig .
[ fig : zachary]a ) .
recently , this same network was used by girvan and newman @xcite in their study of community structures .
the second network that will be considered is a much larger network taken from the organization of the internet .
the internet consists of a large number of individual computers that are identified by their so - called ip - address , and they are usually grouped together in local area networks ( lan ) .
such computer networks are connected to one another via routers a complex network linking device .
in addition to being able to direct pieces of information to its intended destination , a router also has the ability to determine the best path to a given destination ( routing ) .
this is done by keeping available an updated routing table telling the router how to reach certain destinations specified by the network administrator . in much the same way as computers are being organized into networks , the same is done for routers .
if the router network becomes large and coherent enough , it may make out what is called an _ autonomous system _ ( as ) .
an as is a connected segment of a network that consists of a collection of subnetworks ( with hosts attached ) interconnected by a set of routes .
usually it is required that the subnetworks and the routers should be controlled by a coherent organization like , say , a university , or a medium - to - big enterprise .
importantly for the efficiency of the internet , each autonomous system , identified by a unique as number , is expected to present to the other systems a consistent list of destinations reachable through that as , as well as a coherent interior routing plan .
the particular network we will be considering is an autonomous system network collected by the oregon views project on january 3 , 2000 @xcite . in this network
the as will act as nodes , while their routing plans , _ i.e. _ with which other as a given one directly shares information , will correspond to the links of the network .
in sec . [ sec : method ] , it was argued that from the currents corresponding to the _ slowly _ decaying eigenmodes , one should be able map out the large scale structure of the underlying network .
our hypothesis will now be put to the test , and we start with the small friendship network due to zachary @xcite .
this network is depicted in fig .
[ fig : zachary]a , and it is small enough to enable us to see how it all comes about . by visual inspection of this figure , it is apparent that there are , at least , two large scale clusters one corresponding to the trainer of the karate club ( node @xmath12 ) and his supporters , and one to the administrator ( node @xmath70 ) . in fig .
[ fig : zachary]a , the karate club members ( according to zachary ) supporting the trainer in the ongoing conflict are marked with squares ( @xmath71 nodes ) , while the followers of the administrator are marked with open circles ( @xmath72 nodes ) .
notice that this subdivision was done by zachary @xcite , and no sophisticated clustering techniques were applied to obtain these results .
we will now see if the same clustering structure can be obtained by the method outlined in sec .
[ sec : method ] of this paper .
let us start by considering the two most slowly decaying modes of the network , namely @xmath73 and @xmath74 ( with our ordering ) , corresponding to the eigenvalues @xmath75 and @xmath76 , respectively .
if one sorts the elements of the current vector @xmath77 by size , and group them according to their signs , one recovers two clusters ( see the abscissa of fig . [
fig : zachary]b ) .
this grouping , or topological structure , fits nicely with the original assignments made by zachary ( fig .
[ fig : zachary]a ) and indicated by the open squares and circles in fig .
[ fig : zachary]b .
hence it is the trainer - administrator separation of the karate club members that is mapped out by the @xmath77-currents .
observe that node @xmath74 , which has an equal number of links to supporters of the trainer and administrator , has a current that is almost zero .
our identification of this node is the only one that differs from that made originally by zachary .
interestingly , our classification , including node @xmath74 , fits that previously obtained by girvan and newman @xcite in their hierarchical tree clustering approach .
if the same procedure was repeated , but for the currents @xmath78 , one would map out other topological features of the network . in fig . [ fig : zachary]b , a @xmath79-plot is presented .
such a plot groups the nodes along two axis ; the @xmath80-axis corresponding to the trainer - administrator axis and the @xmath81-axis . the most striking feature of fig . [ fig : zachary]b is the group of nodes corresponding to @xmath82 and @xmath83 , _
i.e. _ to the following group of nodes @xmath84 . in fig .
[ fig : zachary]a , these nodes are located in the upper left corner of the graph .
they are in addition to being connected among themselves , only connected to the rest of the network via the trainer .
thus , they represent a sub - cluster , and this is indeed what is apparent from fig .
[ fig : zachary]b .
the participation ratios for the two slowest decaying diffusive modes were found to be @xmath85 and @xmath86 , respectively . as can be seen from fig .
[ fig : zachary]b , these participation ratios receive substantial contributions from both positive and negative current elements , _
i.e. _ from more then one topological structure .
hence , we suspect that these numbers being close to the size of the administrator and trainer `` clan '' ( of size @xmath72 and @xmath71 nodes , respectively ) is somewhat accidental .
we will now focus our attention on the autonomous system ( as ) network introduced in the previous section .
this network , consisting of @xmath87 nodes and @xmath88 undirected links , is so big that plotting it for the purpose of study its modular structure is not a practical option .
motivated by what we found for the small zachary friendship network , we will now go ahead and use somewhat similar techniques for its study . in fig .
[ fig : internet ] the participation ratio , @xmath66 , of eigenvectors @xmath55 ( top ) and the eigenvalue density ( bottom ) are plotted as functions of the corresponding eigenvalues @xmath89 .
the data for the internet , that is an example of a scale - free network @xcite , are displayed together with the data for its randomized counterpart ( the null model ) . from fig .
[ fig : internet ] it is apparent that while the density of states is rather similar for these two networks , the participation ratios of the slowly decaying modes , especially for @xmath35 close to @xmath12 , are markedly higher in the internet network than in the accompanying randomized network .
these differences signal large scale topological structures @xcite that are real and not accidental . for the most slowly decaying diffusive eigenmode ,
the participation ratio is @xmath90 ( cf .
[ fig : internet]a ) . from fig .
[ fig : internet - currents ] , that depicts the currents of the two slowest decaying diffusive eigenmodes , one observes that the main contribution to @xmath91 comes from current elements @xmath92 of the same sign .
thus , a module has been detected , and @xmath91 roughly measures its size . from , fig .
[ fig : internet - currents ] one also makes the interesting observation that the main contributing nodes to @xmath91 are autonomous systems located in russia ( denoted by open squares in fig .
[ fig : internet - currents ] ) .
hence , the diffusive mode @xmath73 maps out russia , as a russian module @xcite ! in total there are @xmath93 russian nodes in our data set .
moreover , from fig . [
fig : internet - currents ] , modules corresponding to countries like the usa , france and korea are also easily identified .
the edges of the as internet network , defined as the nodes corresponding to the most distant values of the current elements @xmath92 , are in this case represented by a russian and a us military site located in the south pacific .
these nodes can thus be said to represent the extreme edges of the internet .
one has also studies the number of significant modules ( where @xmath94 ) , and found that this number for the as network is roughly @xmath95 @xcite .
the number of different nodes participating in one or more of these @xmath96 modules was found to be about @xmath97 .
hence the modularity of the network is ( at least ) @xmath98 . one common feature of the current - current plots , figs .
[ fig : zachary]b and [ fig : internet - currents ] , is their line or star - like structures .
such structures are in fact rather generic , and we will here try to explain why . we have argued earlier that current elements @xmath55 should almost be the same within a module .
hence , for two nodes @xmath4 and @xmath9 belonging to one and the same module , the fraction @xmath99 is expected to be a constant unique for that module , but independent of the diffusive mode @xmath100 . for two different ( significant ) diffusive modes , say @xmath100 and @xmath101 , the fraction @xmath102 ( for node @xmath4 ) will therefore as a consequence also be constant .
thus , under the assumptions made above , one predicts a straight line in a current - current plot ( that pases through the origin ) @xcite : @xmath103 , for nodes @xmath4 belonging to the same module . from fig .
[ fig : internet - currents ] , this simple argument appears to be valid .
that fig .
[ fig : zachary]b seems to not pass through the origin , but still being straight lines , we believe is due to the diffusive modes being excited over a non - trivial background in this case . before presenting the conclusions of this paper , we will add a few closing remarks on the numerical implementation of the method .
for this work to have any relevance for large real - world networks , a fast , memory saving , and optimized algorithm for the calculation of the largest eigenvalues and the corresponding eigenvectors of a sparse matrix is required .
fortunately , such an algorithm has already been implemented and made available _
e.g. _ through the trlan software package @xcite .
this software is optimized for handling large problems , and it can run on large scale parallel supercomputers .
we have generalized the normal diffusion process to diffusion on ( discreet ) complex networks . by considering such a process
, it has been demonstrated that topological properties like modular structures and edges of the underlying network can be probed .
this is achieved by focusing on the slowest decaying eigenmodes of the network .
the use of the procedure was exemplified by considering a small friendship network , with known modular structure , as well as a routing network of the internet , where the structure was not so well known . for the friendship network the known structure was well reproduced , and the internet was indeed found to be modular .
the detected modules of the internet were consistent with the geographical location of the nodes , and the individual modules corresponded roughly to the national structure .
interestingly , it was observed that a political subdivision of the internet was also one of the predictions of the algorithm presented in this paper ; the two most poorly connected nodes of the internet ( extreme edges ) , where found to be represented by a russian and a us military site located in the south pacific .
work at brookhaven national laboratory was carried out under contract no .
de - ac02 - 98ch10886 , division of material science , u.s .
department of energy .
nodes and @xmath104 links ( figure after ref .
@xcite ) . here
open squares and circles are used to denote the supporters , in the ongoing conflict , of the trainer ( node @xmath12 ) and administrator ( node @xmath70 ) , respectively .
( b ) the @xmath79-plot maps out the large scale topology of the network .
the dashed lines indicate the lines of zero currents . ,
title="fig:",width=377,height=302 ] + nodes and @xmath104 links ( figure after ref .
@xcite ) . here
open squares and circles are used to denote the supporters , in the ongoing conflict , of the trainer ( node @xmath12 ) and administrator ( node @xmath70 ) , respectively .
( b ) the @xmath79-plot maps out the large scale topology of the network .
the dashed lines indicate the lines of zero currents . ,
title="fig:",width=377,height=264 ] ( top , a ) and the eigenvalue density ( bottom , b ) as a function of the eigenvalue of the transfer matrix , @xmath105 , measured in the internet ( filled circles ) and in its randomized counterpart ( open squares ) a random scale - free network . the participation ratio was averaged over @xmath106-bins of size 0.05 excluding eigenmodes @xmath107 , and @xmath42 ( cf . ref . ) . , width=377,height=264 ] -th as in this plot
are its current components @xmath108 of the two slowest decaying non - oscillatory diffusion modes .
the symbols reveals the geographical location of the as : russia squares , france
circles , usa crosses , korea
triangles ) .
note the straight lines corresponding to good country - modules .
, width=377,height=302 ] | a diffusion process on complex networks is introduced in order to uncover their large scale topological structures .
this is achieved by focusing on the slowest decaying diffusive modes of the network .
the proposed procedure is applied to real - world networks like a friendship network of known modular structure , and an internet routing network . for the friendship network ,
its known structure is well reproduced . in case of the internet , where the structure is far less well - known
, one indeed finds a modular structure , and modules can roughly be associated with individual countries .
quantitatively the modular structure of the internet manifests itself in an approximately @xmath0 times larger participation ratio of its slowest decaying modes as compared to the null model a random scale - free network .
the extreme edges of the internet are found to correspond to russian and us military sites .
complex random networks , network modules , statistical physics 89.75.-k , 89.20.hh , 89.75.hc , 05.40.fb |
in materials science and engineering , a class of simple models , known as fiber bundle models ( fbm ) , has proven to be very effective in practical applications such as fiber reinforced composites . in this context ,
such models have a history that goes back to the twenties @xcite , and they constitute today an elaborate toolbox for studying such materials , rendering computer studies orders of magnitudes more efficient than brute force methods . since the late eighties @xcite , these models have received increasing attention in the physics community due to their deceivingly simple appearance coupled with an extraordinary richness of behaviors .
as these models are just at the edge of what is possible analytically and typically not being very challenging from a numerical point of view so that extremely good statistics on large systems are available , they are perfect as model systems for studying failure phenomena as a part of theoretical physics .
fracture and material stability has for practical reasons interested humanity ever since we started using tools : our pottery should be able to withstand handling , our huts should be able to withstand normal weather . as science took on the form
we know today during the renaissance , leonardo da vinci studied five hundred years ago experimentally the strength of wires fiber bundles as a function of their length @xcite .
systematic strength studies , but on beams , were also pursued systematic by galileo galilei one hundred years later , as was done by edme mariotte ( of gas law fame ) who pressurized vessels until they burst in connection with the construction of a fountain at versailles .
for some reason , mainstream physics moved away from fracture and breakdown problems in the nineteenth century , and it is only during the last twenty years that fracture problems have been studied within physics proper .
the reason for this is most probably the advent of the computer as a research tool , rendering problems that were beyond the reach of systematic theoretical study now accessible .
if we were to single out the most important contribution from the physics community with respect to fracture phenomena , it must be the focus on _ fluctuations _ rather than averages .
what good is the knowledge of the average behavior a system when faced with a single sample and this being liable to breakdown given the right fluctuation ?
this review , being written by physicists , reflects this point of view , and hence , fluctuations play an important role throughout it .
even though we may trace the study of fiber bundles to leonardo da vinci , their modern story starts with the already mentioned work by @xcite . in 1945 , daniels published a seminal review cum research article on fiber bundles which still today must be regarded as essential reading in the field @xcite . in this paper ,
the fiber bundle model is treated as a problem of statistics and the analysis is performed within this framework , rather than treating it within materials science .
the fiber bundle is viewed as a collection of elastic objects connected in parallel and clamped to a medium that transmits forces between the fibers .
the elongation of a fiber is linearly related to the force it carries up to a maximum value .
when this value is reached , the fiber fails by no longer being able to carry any force .
the threshold value is assigned from some initially chosen probability distribution , and do not change thereafter .
when the fiber fails , the force it carried is redistributed .
if the clamps deform under loading , fibers closer to the just - failed fiber will absorb more of the force compared to those further away .
if the clamps , on the other hand , are rigid , the force is equally distributed to all the surviving fibers .
daniels discussed this latter case . a typical question posed and answered in this paper
would be the average strength of a bundle of @xmath0 fibers , but also the variance of the average strength of the entire bundle .
the present review takes the same point of view , discussing the fiber bundle model as a _ statistical model .
_ only in section v we discuss the fiber bundle model in the context of materials science with all the realism of real materials considered
. however , we have not attempted to include any discussions of the many experimental studies that have been performed on systems where fiber bundles constitute the appropriate tool .
this is beyond the scope of this statistical - physics based review . after introducing ( section ii ) the fiber bundle model to our readers , in section iii
, we present the _ equal load sharing model , _ which was sketched just a few lines back .
this seemingly simple model is in fact extremely rich .
for example , the load at which catastrophic failure occurs is a _
second order critical point _ with essentially all the features usually seen in systems displaying such behavior .
however in this case , the system is analytically tractable .
in fact , we believe that the equal load sharing fiber bundle may be an excellent system for teaching second order phase transitions at the college level . under the heading of fluctuations , we discuss the burst distribution , i.e. , the statistics of simultaneously failing fibers during loading : when a fiber fails and the force it was carrying is redistributed , one or more other fibers may be driven above their failing thresholds . in this equal load sharing model , the absolute rigidity of the bar ( transmitting forces among the fibers )
suppresses the stress fluctuations among the fibers . as such
, there is no apparent growth of the ( fluctuation correlation ) length scale .
hence , although there are precise recursion relations and their linearized solutions are available near the fixed point ( see sec .
iii ) , no straight forward application of the renormalization group techniques @xcite has been made to extract the exponents through length scaling . in section iv _ local load sharing _ is discussed .
this bit of added realism comes at the added cost that analytical treatment becomes much more difficult .
there are , still , a number of analytical results in the literature .
one may see intuitively how local load sharing complicates the problem , since the relative positions of the fibers now become important . under global load sharing ,
every surviving fibers gets the same excess force and , hence , where they are do not matter .
there are essentially three local load sharing models in the literature .
the first one dictates that the _ nearest surviving neighbors _ of the failing fiber absorb its load .
then there are softer models " where the redistribution follows a power law in the distance to the failing fiber .
lastly , there is the model where the clamps holding the fibers are elastic themselves , and this leads to non - equal redistribution of the forces .
section v contains a review of the use of fiber bundle models in applications such as materials science .
we discuss fatigue , thermal failure , viscoelastic effects and precursors of global failure .
we then go on to review the large field of modeling fiber reinforced composites . here
fiber bundle models constitute the starting point of the analysis , which , by its very nature , is rather complex seen from the viewpoint of statistical physics .
lastly , we review some applications of fiber bundle models in connection with systems that initially would seem quite far from the concept of a fiber bundle , such as traffic jams .
we end this review by a summary with few concluding remarks in section vi .
imagine a heavy load hanging from a rigid anchor point ( say , at the roof ) by a rope or a bundle of fibers .
if the load exceeds a threshold value , the bundle fails .
how does the failure proceeds in the bundle ? unless all the fibers in the bundle have got identical breaking thresholds ( and that never happens in a real rope ) , the failure dynamics proceeds in a typical collective load transfer way .
one can assume that in this kind of situation the load is equally shared by all the intact fibers in the bundle .
however , the breaking threshold for each of the fibers being different , some fibers fail before others and , consequently , the load per surviving fiber increases as it gets redistributed and shared equally by the rest .
this increased load per fiber may induce further breaking of some fibers and the avalanche continues , or it stops if all the surviving fibers can withstand the redistributed load per fiber .
this collective or cooperative failure dynamics and the consequent avalanches or bursts are typical for the failure in any many - body system .
it captures the essential features of failure during fracture propagation ( recorded by acoustic emissions ) , earthquake avalanches ( main and aftershocks ) , traffic jams ( due to dynamic clustering ) , etc .
the model was first introduced in @xmath1 by @xcite in the context of textile engineering .
since then it was modified a little and investigated , mainly numerically , with various realistic fiber threshold distributions by the engineering community @xcite . starting from late eighties , physicists took interest in the avalanche distribution in the model and in its dynamics @xcite .
a recursive dynamical equation was set up for the equal - load - sharing version recently @xcite and the dynamic critical behavior is now solved exactly @xcite .
in addition to the extensive numerical results @xcite on the effect of short - range fluctuations ( local load sharing cases ) , some progress with analytical studies @xcite have also been made .
there are a large number of experimental studies of various materials and phenomena that have successfully been analyzed within the framework of the fiber bundle model .
for example , @xcite have used the fiber bundle model to propose explanations for changes in fibrious collagen and its relation to neuropathy in connection with diabetes .
@xcite propose a method to monitor the structural integrity of fiber - reinforced ceramic - matrix composites using electrical resistivity measurements .
the basic idea here is that when the fibers in the composite themselves fail rather than just the matrix in which they are embedded , the structure is about to fail .
the individual fiber failures is recorded through changes in the electrical conductivity of the material .
acoustic emission , the crackling sounds emitted by materials as they are loaded , provide yet another example where fiber bundle models play and important role , see e.g. , @xcite .
= 1.7 in the simplest and the oldest version of the model is the equal load sharing ( els ) model , in which the load previously carried by a failed fiber is shared equally by all the remaining intact fibers in the system . as the applied load is shared globally , this model is also known as global load sharing ( gls ) model or democratic fiber bundle model . due to the consequent mean - field nature
, some exact results could be extracted for this model and this was demonstrated by @xcite in a classic work some sixty years ago .
the typical relaxation dynamics of this model has been solved recently which has clearly established a robust critical behavior @xcite .
it may be mentioned at the outset that the els or gls models do not allow for spatial fluctuations ( due to the absolute rigidity of the platform in fig
. 1 ) and hence such models belong to the the mean field category of critical dynamics , see e.g. , @xcite . fluctuations in breaking time or in avalanche statistics ( due to randomness in fiber strengths ) are of course possible in such models and are discussed in details in this section . a bundle can be loaded in two different ways : strain controlled and force controlled . in the strain controlled method , at each step the whole bundle is stretched till the weakest fiber fails . clearly ,
when number of fibers @xmath0 is very large , strain is increased by infinitesimal amount at each step until complete breakdown and therefore the process is considered as a _ quasi - static _ way of loading . on the other hand , in the force controlled method , the external force ( load ) on the bundle
is increased by same amount at each step until the breakdown .
the basic difference between these two methods is that the first method ensures the failure of single fiber ( weakest one among the intact fibers ) at each loading step , while in the second method sometimes none of the fibers fail and sometimes more than one fail in one loading step .
let @xmath2 denote the strain of the fibers in the bundle .
assuming the fibers to be linearly elastic up to their respective failure point ( with unit elastic constant ) , we can represent the stresses on each of the surviving fibers by the same quantity @xmath2 .
the strength ( or threshold ) of a fiber is usually determined by the stress value @xmath2 it can bear , and beyond which it fails .
we therefore denote the strength ( threshold ) distribution of the fibers in the bundle by @xmath3 and the corresponding cumulative distribution by @xmath4 .
two popular examples of threshold distributions are the uniform distribution @xmath5 and the weibull distribution @xmath6 here @xmath7 is a reference threshold , and the dimensionless number @xmath8 is the weibull index ( fig .
[ fig : distributions ] ) . in the strain controlled loading , at a strain @xmath2 ,
the total force on the bundle is @xmath2 times the number of intact fibers .
the expected or average force at this stage is therefore @xcite @xmath9 the maximum @xmath10 of @xmath11 corresponds to the value @xmath12 for which @xmath13 vanishes : @xmath14 here the failure process is basically driven by _
fluctuations _ and can be analyzed using extreme order statistics @xcite . in the force
controlled method , if force @xmath15 is applied on a bundle having @xmath0 fibers , when the system reaches an equilibrium , the strain or effective stress @xmath2 is ( see fig . [
fig : fbm - model ] ) @xmath16}. \label{strain}\ ] ] therefore , at the equilibrium state , eq .
( [ load ] ) and eq .
( [ strain ] ) are identical .
it is possible to construct recursive dynamics @xcite of the failure process for a given load and the fixed - point solutions explore the _ average behavior _ of the system at the equilibrium state . fig .
[ fig : fbm - model ] shows a static fiber bundle model in the els mode where @xmath0 fibers are connected in parallel to each other ( clamped at both ends ) and a force is applied at one end . at the first step all fibers that can not withstand the applied stress break .
then the stress is redistributed on the surviving fibers , which compels further fibers to break .
this starts an iterative process that continues until an equilibrium is reached , or all fibers fail .
the average behavior is manifested when the initial load is macroscopic ( very large @xmath0 ) .
the breaking dynamics can be represented by recursion relations @xcite in discrete steps .
let @xmath17 be the number of fibers that survive after step @xmath18 , where @xmath18 indicates the number of stress redistribution steps
. then one can write @xcite @xmath19.\ ] ] now we introduce @xmath20 , the applied stress and @xmath21 , the surviving fraction of total fibers .
then the effective stress after @xmath18 step becomes @xmath22 and after @xmath23 steps the surviving fraction of total fibers is @xmath24 .
therefore we can construct the following recursion relations @xcite : @xmath25 and @xmath26 at equilibrium @xmath27 and @xmath28 .
these equations ( eq .
[ rec - x ] and eq .
[ rec - u ] ) can be solved at and around the fixed points for the particular strength distribution @xmath3 .
let us choose the uniform density of fiber strength distribution ( eq . [ uniform ] ) up to the cutoff @xmath29 .
then the cumulative distribution becomes @xmath30 .
therefore from eq .
( [ rec - x ] ) and eq .
( [ rec - u ] ) we can construct a pair of recursion relations @xmath31 and @xmath32 this nonlinear recursion equations are somewhat characteristic of the dynamics of fiber bundle models and such dynamics can be obtained in many different ways .
for example , the failed fraction @xmath33 at step @xmath23 is given by the fraction @xmath34 of the load shared by the intact fibers at step @xmath18 and for the uniform distribution of thresholds ( fig .
[ fig : distributions]a ) , one readily gets eq .
( [ rec - u - uniform ] ) . at the fixed point
the above relations take the quadratic forms @xmath35 and @xmath36 with the solutions @xmath37 and @xmath38 here @xmath39 is the critical value of applied stress beyond which the bundle fails completely . clearly , for the effective stress ( eq .
[ sol - x - uniform ] ) solution with @xmath40 sign is the stable fixed point and with @xmath41 sign is the unstable fixed point whereas for fraction of unbroken fibers
( eq . [ sol - u - uniform ] ) , it is just the opposite .
now the difference @xmath42 behaves like an order parameter signaling partial failure of the bundle when it is non - zero ( positive ) , although unlike conventional phase transitions it does not have a real - valued existence for @xmath43
. @xmath44 fig .
[ fig : uso ] shows the variation of @xmath45 , @xmath46 and @xmath47 with the externally applied stress value @xmath48 .
one can also obtain the breakdown susceptibility @xmath49 , defined as the change of @xmath50 due to an infinitesimal increment of the applied stress @xmath48 : @xmath51 such a divergence in @xmath49 had already been reported in several studies @xcite . to study the dynamics away from criticality ( @xmath52 from below ) , the recursion relation ( eq .
[ rec - u - uniform ] ) can be replaced by a differential equation @xmath53 close to the fixed point , @xmath54 + @xmath55 , ( where @xmath56 ) and this gives @xmath57 where @xmath58 $ ] .
therefore , near the critical point : @xmath59 at the critical point ( @xmath60 ) , a dynamic critical behavior has been observed in the relaxation of the failure process to the fixed point . from the recursion relation ( eq . [ rec - u - uniform ] )
it can be easily verified that the fraction @xmath61 follows a simple power - law decay : @xmath62 starting from @xmath63 . for large @xmath18 ( @xmath64 ) , this reduces to @xmath65 ; @xmath66 ; indicating _ critical slowing down _ which is a robust characterization of the critical state . the critical properties , obtained above , are for the uniform threshold distribution , and the natural question is how general the results are . to check the universality of the els model
, two other types of fiber strength distributions can be easily considered @xcite : linearly increasing density distribution and linearly decreasing density distribution . for linearly increasing density of fiber strengths in the interval @xmath67 $ ]
, the normalized density function and the cumulative distribution are given by ( illustrated in fig .
[ pratip_fig4 ] ) : @xmath68 and @xmath69 now we introduce the transformed quantities : @xmath70 for an initial stress @xmath71 ( or , @xmath72 ) along with the cumulative distribution given by eq .
( [ eq : prob - lin ] ) , the recursion relations ( eq . [ rec - x ] and eq .
[ rec - u ] ) appear as : @xmath73 and @xmath74 the fixed point equations , ( eq . [ fix - x - uniform ] and eq .
[ fix - u - uniform ] ) , now assume cubic form : @xmath75 where @xmath76 , and @xmath77 consequently each of the recursions ( eq . [ eq : stressrecur - lin ] and eq .
[ eq : fracrecur - lin ] ) have three fixed points
only one in each case is found to be stable .
for the redistributed stress the fixed points are : @xmath78 where @xmath79 and @xmath80 similarly , for the surviving fraction of fibers the fixed points are : @xmath81 where @xmath82 and @xmath83 - 27 \gamma_0 ^ 2 / 2 \over
\left [ \left ( \gamma_l^2 - 1 \right )
^2 + 6 \gamma_l \gamma_0 \right ] ^{3/2}}. \label{eq : theta - defn}\ ] ] of these fixed points @xmath84 and @xmath85 are stable whereas @xmath86 , @xmath87 and @xmath88 , @xmath89 are unstable ( fig .
[ pratip_fig5 ] ) .
the discriminants of the cubic equations ( eq .
[ eq : stressfix - lin ] and eq .
[ eq : fracfix - lin ] ) become zero at a critical value @xmath90 ( or , @xmath91 ) of the initial applied stress : @xmath92 \label{eq : stresscrit - lin}\end{aligned}\ ] ] and then each of the quantities @xmath93 and @xmath94 have one stable and one unstable fixed point .
the critical point has the trivial lower bound : @xmath95 .
the expression of @xmath91 in eq .
( [ eq : stresscrit - lin ] ) shows that it approaches the lower bound as @xmath96 which happens for finite values of @xmath97 and @xmath98 when @xmath99 .
it follows that the upper bound for the critical point is also trivial : @xmath100 .
also , at the critical point we get from eq .
( [ eq : phi - defn ] ) and eq .
( [ eq : theta - defn ] ) : @xmath101 or , @xmath102 the stable fixed points @xmath84 and @xmath103 are positive real - valued when @xmath104 ; thus the fiber bundle always reaches a state of mechanical equilibrium after partial failure under an initial applied stress @xmath105 . for @xmath106 ( or , @xmath107 ) , @xmath84 and @xmath103
are no longer real - valued and the entire fiber bundle eventually breaks down .
the transition from the phase of partial failure to the phase of total failure takes place when @xmath48 just exceeds @xmath90 and the order parameter for this phase transition is defined as in eq .
( [ order - uniform ] ) : @xmath108 close to the critical point but below it , we can write , from eq .
( [ eq : theta - defn ] ) and eq .
( [ eq : phi - theta - crit2 ] ) , that : @xmath109^{3/2 } } \label{eq : theta - nearcrit}\end{aligned}\ ] ] and the expressions for the fixed points in eq .
( [ eq : fracfix - lin1 ] ) and eq .
( [ eq : fracfix - lin2 ] ) reduce to the forms : @xmath110 and @xmath111 where @xmath112 is the stable fixed point value of the surviving fraction of fibers under the critical initial stress @xmath90 . therefore , following the definition of the order parameter in eq .
( [ eq : order - lin - defn ] ) we get from the above equation : @xmath113 on replacing the transformed variable @xmath114 by the original @xmath48 , eq .
( [ eq : order - lin - crit ] ) shows that the order parameter goes to zero continuously following the same power - law as in eq .
( [ order - uniform ] ) for the previous case when @xmath48 approaches its critical value from below .
similarly the susceptibility diverges by the same power - law as in eq .
( [ sus - uniform ] ) on approaching the critical point from below :
@xmath115 the critical dynamics of the fiber bundle is given by the asymptotic closed form solution of the recursion ( eq . [ eq : fracrecur - lin ] ) for @xmath116 : @xmath117 { 1 \over t } , \hspace{1.0 cm } t \to \infty , \label{eq : frac - lin - crit - dyn}\ ] ] where @xmath91 and @xmath118 are given in eq .
( [ eq : stresscrit - lin ] ) and eq .
( [ eq : fracfix - lin12-crit ] ) respectively .
this shows that the asymptotic relaxation of the surviving fraction of fibers to its stable fixed point under the critical initial stress has the same ( inverse of step number ) form as found in the case of uniform density of fiber strengths ( eq . [ eq : fbm - omori ] ) .
we now consider a fiber bundle with a linearly decreasing density of fiber strengths in the interval @xmath67 $ ] .
the normalized density function and cumulative distribution ( illustrated in fig .
[ pratip_fig6 ] ) are : @xmath119 and @xmath120 with the transformed quantities defined in eq .
( [ eq : trans - stress ] ) the recurrences ( eq . [ rec - x ] and eq .
[ rec - u ] ) for @xmath71 appear as : @xmath121 and @xmath122 the fixed point equations are again cubic : @xmath123 @xmath124 and they have the following solutions : @xmath125 where @xmath126 @xmath127 and @xmath128 where @xmath129 @xmath130 + 27 \gamma^2 / 2 \over \left [ \left ( 1 + \gamma_l \right ) ^4 - 6 \left ( 1 + \gamma_l \right ) \gamma_0 \right ] ^{3/2}}. \label{eq : theta'-defn}\ ] ] here @xmath87 and @xmath85 are stable fixed points while the rest are unstable ( fig . [ pratip_fig7 ] ) .
the discriminants of eq .
( [ eq : stressfix - lin ] ) and eq .
( [ eq : fracfix - lin ] ) show that the critical applied stress in this case , @xmath131 ( or , @xmath132 ) , is given by : @xmath133 or , @xmath134 in order to satisfy the condition @xmath135 , it requires from eq .
( [ eq : stresscrit - lin-alter ] ) that @xmath136 which imposes an upper bound : @xmath137 .
like before , for @xmath138 the stable fixed points are real - valued , which indicates that only partial failure of the fiber bundle takes place before a state of mechanical equilibrium is reached ; for @xmath139 the fixed points are not real and a phase of total failure exists .
the order parameter @xmath47 of the transition is given by the definition in eq .
( [ eq : order - lin - defn ] ) . for @xmath140
we get the following properties from eq .
( [ eq : phi-defn ] ) , eq .
( [ eq : fracfix - lin1 ] ) , eq .
( [ eq : fracfix - lin2 ] ) and eq .
( [ eq : theta-defn ] ) : @xmath141 @xmath142 and @xmath143 comparing eq .
( [ eq : phi-crit ] ) and eq .
( [ eq : theta-crit ] ) with eq .
( [ eq : phi - theta - crit2 ] ) we see that the critical values of @xmath144 and @xmath145 are the same whereas those of @xmath146 and @xmath147 differ by @xmath148 radians . near the critical point , but below it , we get from eq .
( [ eq : fracfix - lin1 ] ) and eq .
( [ eq : fracfix - lin2 ] ) : @xmath149 and @xmath150 therefore , by the definition of the order parameter in eq .
( [ eq : order - lin - defn ] ) and that of the susceptibility in ( [ eq : suscept - lin - crit ] ) , we get in this case @xmath151 and @xmath152 , @xmath153 .
these power laws have the same exponents as the corresponding ones in the previous cases and differ from those only in the critical point and the critical amplitude . at the critical point the asymptotic relaxation of the surviving fraction of fibers to its stable fixed point [ obtained as an asymptotic solution to eq .
( [ eq : fracrecur - lin ] ) ] is again found to be a power law decay similar to eq .
( [ eq : fbm - omori ] ) and eq .
( [ eq : frac - lin - crit - dyn ] ) : @xmath154 the two density functions , eq .
( [ eq : dens - lin ] ) and eq .
( [ eq : dens - lin ] ) , can be transformed from one to the other by a reflection on the line @xmath155 ( compare fig .
[ pratip_fig4](a ) and fig .
[ pratip_fig6](a ) ) .
but the fixed point equations and their solutions do not have this symmetry .
this is because the density function @xmath3 does not appear directly in the recursion relations for the dynamics .
it is the cumulative distribution @xmath156 which appears in the recursion relations .
( [ eq : prob - lin ] ) and eq .
( [ eq : prob - lin ] ) show that the cumulative distributions of these two cases are not mutually symmetric about any value of the threshold stress @xmath2 ( compare fig .
[ pratip_fig4](b ) and fig .
[ pratip_fig6](b ) ) .
however a certain relation exists between the critical values of the applied stress for a special case of these two models : if @xmath157 , we get from eq .
( [ eq : stresscrit - lin ] ) and eq .
( [ eq : stresscrit - lin ] ) that @xmath158 and @xmath159 respectively ; therefore we have @xmath160 .
the critical behavior of the models discussed in this section show that the power laws found here are independent of the form of the cumulative distribution @xmath161 .
the three threshold distributions studied here have a common feature : the function @xmath162 $ ] has a maximum which corresponds to the critical value of the initial applied stress .
all threshold distributions having this property are therefore expected to lead to the same universality class as the three studied here .
if the threshold distribution does not have this property we may not observe a phase transition at all .
for example , consider a fiber bundle model with @xmath163 , @xmath164 .
here @xmath165 = 1 $ ] and the evolution of the fiber bundle is given by the recursion relation @xmath166 which implies that there is no dynamics at all for @xmath167 and an exponential decay to complete failure , @xmath168 , for @xmath169 .
there are no critical phenomena and therefore no phase transition .
however this general conclusion may not be true for finite - sized bundles @xcite .
thus the els fiber bundles ( for different fiber threshold distributions ) show phase transition with a well defined order parameter which shows similar power law variation on the way the critical point is approached .
for all the cases , discussed here , the susceptibility and relaxation time diverge following similar power laws and the failure processes show similar critical slowing at the critical point .
this suggests strongly that the critical behavior is universal , which we now prove through general arguments @xcite . when an iteration is close to the fixed point , we have for the deviation @xmath170 to lowest order in @xmath171 .
this guarantees an exponential relaxation to the fixed point , @xmath172 , with parameter @xmath173 criticality is determined by the extremum condition ( eq . [ critical - criteria ] ) , which by the relation ( eq . [ rec - u ] ) takes the form @xmath174 thus @xmath175 at criticality . to study the relaxation at criticality
we must expand eq .
( [ dev-1 ] ) to second order in @xmath171 since to first order we simply get the useless equation @xmath176 . to second order
we obtain @xmath177 with a positive constant @xmath178 .
this is satisfied by @xmath179 hence in general the dominating critical behavior for the approach to the fixed point is a power law with @xmath66 .
the values @xmath180 can be shown to be consequences of the parabolic maximum of the load curve at criticality .
thus all threshold distributions for which the macroscopic strength function has a single parabolic maximum , is in this universality class .
it is clear that at the critical stress value @xmath181 , els fiber bundles show phase transition from partially broken state to completely broken state .
what is the order of this phase transition ?
@xcite considered the fraction of unbroken fibers as the order parameter and as it has a discontinuity at the critical stress value , they suggested , after a mean - field analysis , that it can be seen as a first - order phase transition similar to spinodal instability @xcite .
the additional reason for identifying the transition at @xmath182 as a first - order spinodal point had been @xcite that in the presence of short - range interactions ( as in lls , see section iv ) , the transition becomes discontinuous and first - order like .
it is indeed hard to identify continuously changing order parameter there .
we , however , believe the transition in els to be second - order .
chronologically , a little later , a new parameter was identified @xcite : the branching ratio ( @xmath183 ) , which is defined as the probability of triggering further breaking given an individual failure .
the branching ratio continuously approaches ( fig .
[ fig : zeta ] ) the value @xmath184 at the critical stress ( @xmath181 ) starting from @xmath185 value ( for very small @xmath48 ) .
also it shows a power law variation : @xmath186 , with @xmath187 .
therefore @xmath188 acts as the order parameter showing a continuous transition at the critical point , signaling a second - order phase transition . as mentioned earlier @xcite and
@xcite considered the difference between the fraction of unbroken fibers at any @xmath48 and at @xmath181 , as the order parameter ( @xmath47 ) : it shows a similar continuous variation with the applied stress : @xmath189 , with @xmath187 .
apart from this , the susceptibility and relaxation time diverge at the critical point following power laws having universal exponent values @xcite .
one may therefore conclude that at the critical point the els fiber bundles show a second - order phase transition with robust critical behavior as discussed here .
finally we compare the els fiber bundle model studied here with the mean - field ising model . though the order parameter exponent ( equal to @xmath190 ) of this model is identical to that of the mean - field ising model the two models are not in the same universality class .
the susceptibility in these models diverge with critical exponents @xmath190 and @xmath184 respectively on approaching the critical point .
the dynamical critical exponents are not the same either : in this fiber bundle model the surviving fraction of fibers under the critical applied stress decays toward its stable fixed point as @xmath191 , whereas the magnetization of the mean - field ising model at the critical temperature decays to zero as @xmath192 .
when an external load @xmath15 is applied to a fiber bundle , the iterative failure process continues until all fibers fail , or an equilibrium situation with a nonzero bundle strength is reached . since the number of fibers is finite , the number of steps , @xmath193 , in this sequential process is _
finite_. following @xcite , we now determine how @xmath193 depends upon the applied stress @xmath48 .
the state of the bundle can be characterized as _ pre - critical _ or _ post - critical _ depending upon the stress value relative to the critical stress @xmath194 above which the bundle collapses completely .
the function @xmath195 that we now focus on , exhibits critical divergence when the critical point is approached from _ either _ side . as an example , we show in fig . [
fig : relax - fig1_rmp ] the @xmath195 obtained by simulation for a uniform threshold distribution .
we study the stepwise failure process in the bundle , when a fixed external load @xmath196 is applied .
let @xmath197 be the number of intact fibers at step no .
@xmath18 , with @xmath198 .
we want to determine how @xmath197 decreases until the degradation process stops . when @xmath0 is a large number , we recall the basic recursion ( eq . [ rec - u ] ) to formulate the breaking dynamics : @xmath199 where @xmath200 is considered as a continuous variable . _ ( a ) post - critical relaxation _
we study first the post - critical situation , @xmath201 , with positive values of @xmath202 , and start with the simplest one , uniform threshold distribution ( eq . [ uniform ] ) with the critical point at @xmath203 , @xmath204 . then the basic recursion relation ( eq . [
basic - rec ] ) takes the form @xmath205 this nonlinear iteration can be transformed into a linear relation .
we introduce first @xmath206 into eq .
( [ ut - uni ] ) , with a result @xmath207 then we put @xmath208 , which gives @xmath209 hence we get @xmath210 , with solution @xmath211 in the original variable the solution reads @xmath212 where @xmath63 has been used .
( [ ut - uni ] ) shows that when @xmath213 obtains a value in the interval @xmath214 , the next iteration gives complete bundle failure .
taking @xmath215 as the penultimate value gives a lower bound , @xmath216 , for the number of iterations , while using @xmath217 in eq .
( [ finaluni ] ) gives an upper bound @xmath218 .
adding unity for the final iteration , eq .
( [ finaluni ] ) gives the bounds @xmath219 and @xmath220 fig .
[ fig : relax - fig2_rmp]a shows that these bounds nicely embrace the simulation results . note that both the upper and the lower bound behave as @xmath221 for small @xmath222 .
a rough approximation near the critical point is @xmath223 with @xmath224 . due to the inherent simplicity , uniform distribution is somewhat easy to analyze
. therefore we now discuss how to handle other distributions .
let us start with a weibull distribution ( eq . [ weibull ] ) with index @xmath225 .
the critical parameters for this case are @xmath226 and @xmath227 .
the interesting values of the external stress are close to @xmath181 , because for large super - critical stresses the bundle breaks down almost immediately . for @xmath48 slightly above @xmath181 the iteration function @xmath228 takes the form sketched in fig .
[ fig : relax - fig3_rmp ] .
= 2.5 in the iteration function is almost tangent to the reflection line @xmath229 and a long channel of width proportional to @xmath222 appears .
the dominating number of iterations occur within this channel ( see fig .
[ fig : relax - fig3_rmp ] ) .
the channel wall formed by the iteration function is almost parabolic and is well approximated by a second - order expression @xmath230 here @xmath231 is the fixed point , @xmath229 , of the iteration at @xmath60 . with @xmath232 and @xmath202 eq . ( [ quadr ] )
takes the form @xmath233 with @xmath234 . in the channel
@xmath235 changes very slowly , so we may treat the difference equation as a a differential equation : @xmath236 with solution @xmath237 thus @xmath238 is the number of iterations _ in the channel _ , starting with @xmath239 , ending with @xmath240 .
this treatment is general and can be applied to any threshold distribution near criticality .
although the vast majority of the iterations occur in the channel , there are a few iterations at the entrance and at the exit of the channel that may require attention in special cases .
the situation is similar to type i intermittency in dynamical systems , but in our case the channel is traversed merely once
. for the weibull distribution the expansion ( eq . [ quadr ] ) has the precise form@xmath241 where @xmath231 , @xmath242 , @xmath243 and @xmath244 . for completeness we must also consider the number of iteration to reach the entrance to the channel .
it is not meaningful to use the quadratic approximation ( eq . [ b1 ] ) where it is not monotonously increasing , i.e. for @xmath245 .
thus we take @xmath246 as the entrance to the channel , and add one extra iteration to arrive from @xmath63 to the channel entrance .
( numerical evidence for this extra step : for @xmath60 the iteration ( eq . [ it - w ] ) starts as follows : @xmath247 , @xmath248 , @xmath249 , while using the quadratic function with @xmath250 as the initial value , we get after one step @xmath251 , approximately the same value that the exact iteration reaches after two steps . ) with @xmath252 we obtain from eq .
( [ tf - ti ] ) , in the weibull case , the estimate@xmath253 with @xmath254 and @xmath243 . near the critical point eq .
( [ estimatew ] ) has the asymptotic form @xmath255 with @xmath256 .
the critical index is the same as for the uniform threshold distribution .
the theoretical estimates give an excellent representation of the simulation data ( see fig . [
fig : relax - fig2_rmp]b ) .
.2 in _ ( b ) pre - critical relaxation _ .2 in we now assume the external stress to be pre - critical , @xmath257 , and introduce the positive parameter @xmath258 to characterize the deviation from the critical point . starting with uniform threshold distribution and introducing @xmath259 and @xmath260 into eq .
( [ ut - uni ] ) , one gets @xmath261 in this case we put @xmath262 , which gives @xmath263 thus @xmath264 , i.e. @xmath265 starting with @xmath63 , we obtain @xmath266 and hence @xmath267 this corresponds to @xmath268 in the original variable .
= 2.5 in = 2.5 in apparently @xmath213 reaches a fixed point @xmath269 after an infinite number of iterations .
however , for a bundle with _ finite _ number of fibers , only a finite number of steps is needed for the iteration to arrive at a fixed point @xmath270 which is approximately @xcite @xmath271 as a consequence , we can use @xmath272 as the final value in eq .
( [ ntor ] ) .
consequently we obtain the following estimate for the number of iterations to reach this value : @xmath273 fig .
[ fig : relax - fig4_rmp]a shows that the simulation data are well approximated by the analytic formula ( eq . [ subnorm ] ) . for very large @xmath0 eq .
( [ subnorm ] ) is approximated by @xmath274 with @xmath275 .
the critical behavior is again characterized by a square root divergence .
again we use the weibull distribution ( eq .
[ weibull ] ) as an example threshold distribution . in principle , the iteration , @xmath276 will reach a fixed point @xmath277 after infinite many steps . the deviation from the fixed point , @xmath278 , will decrease exponentially near the fixed point : @xmath279 with @xmath280 for the weibull threshold distribution with index @xmath281 @xmath282 and thus @xmath283 if we allow ourselves to use the exponential formula ( eq . [ exp ] ) all the way from @xmath63 , we obtain @xmath284 for a finite number @xmath0 of fibers the iteration will stop after a finite number of steps .
it is a reasonable supposition to assume that the iteration stops when @xmath285 is of the order @xmath184 .
this corresponds to take the left - hand side of eq .
( [ fixw ] ) equal to @xmath286 . the corresponding number of iterations is then given by @xmath287 in general , and @xmath288 in the weibull case . solving the weibull iteration @xmath289 with respect to @xmath48 and inserting into eq .
( [ twsub ] ) , we obtain @xmath290 these two equations represent the function @xmath291 on parameter form , with @xmath277 running from @xmath231 to @xmath292 . for @xmath293 eq .
( [ par1 ] ) shows that @xmath193 is infinite , as it should be . to investigate the critical neighborhood we put @xmath294 with @xmath295 small , to obtain to lowest order @xmath296 the combination of eq .
( [ 57 ] ) and eq .
( [ 58 ] ) gives , once more , the square root divergence @xmath297 now with the magnitude @xmath298 simulation results for the pre - critical weibull distribution are shown in fig .
[ fig : relax - fig4_rmp]b . which shows good agreement with the analytic solution ( eqs .
[ par1]- [ par2 ] ) . for a general threshold distribution
the divergence and its amplitude are most easily deduced by expanding both the load curve @xmath299 $ ] and the characteristic time @xmath300 around the critical threshold @xmath12 .
to lowest contributing order in @xmath301 we find @xmath302+(x_{c}-x)^{2}\label{y}\ ] ] and @xmath303 inserting for @xmath304 from the equation above , and using ( [ twsub ] ) , we find @xmath305 with @xmath306^{-1/2}\ln(n).\label{generalk}\ ] ] to show how the magnitude of the amplitude @xmath307 depends on the form of the threshold distribution , we consider a weibull distribution @xmath308 with varying coefficient @xmath8 , and constant average strength . with @xmath309 the average strength @xmath310 equals unity , and
the width takes the value @xmath311 here @xmath93 is the gamma function . using the power series expansion @xmath312 we see how the width decreases with increasing @xmath8 : @xmath313 for the weibull distribution ( eq . [ weibullk ] ) we use eq .
( [ generalk ] ) to calculate the amplitude @xmath307 , with the result @xmath314 the last expression for large @xmath8 .
comparison between eq .
( [ width ] ) and eq .
( [ kweibullk ] ) shows that for narrow distributions @xmath315 that narrow distributions give small amplitudes could be expected : many fibers with strengths of almost the same magnitude will tend to break simultaneously , hence the relaxation process goes quicker .
.2 in _ ( c ) universality of critical amplitude ratio _ .2 in as function of the initial stress @xmath48 the number of relaxation steps , @xmath195 , shows a divergence @xmath316 at the critical point , both on the pre - critical and post - critical side .
this is a generic result , valid for a general probability distribution of the individual fiber strength thresholds . on the post - critical side @xmath195
is independent of the system size @xmath0 for large @xmath0 . on the pre - critical side
there is , however , a weak ( logarithmic ) @xmath0-dependence , as witnessed by eqs .
( @xmath317 ) , ( @xmath318 ) and ( @xmath319 ) .
note that the critical amplitude ratio takes the same value @xmath320 for the uniform and the weibull distributions .
this shows the universal nature of the critical amplitude ratio , independent of the threshold distribution .
note the difference with normal critical phenomena @xcite due to the appearance of the @xmath321 in this amplitude ratio here .
fiber bundle model captures correctly the non - linear elastic behavior in els mode @xcite . in case of strain controlled loading , using the theory of extreme order statistics
, it has been shown @xcite that els bundles shows non - linear stress - strain behavior after an initial linear part up to which no fiber fails .
similar non - linear behavior is seen in the force controlled loading case as well . moreover , from the recursive failure dynamics , the amount of stress drop at the breaking point can be calculated exactly @xcite . to demonstrate the scenario we consider an els bundle with uniform fiber strength distribution , having a low cutoff @xmath97 , such that for stresses below the low cutoff , none of the fibers fail .
hence , until failure of any of the fibers , the bundle shows linear elastic behavior .
as soon as the fibers start to fail , the stress - strain relationship becomes non - linear .
this non - linearity can be easily calculated in the els model , using eq .
( [ rec - u ] ) for the failure dynamics of the model .
fibers are here assumed to be elastic , each having unit force constant , with their breaking strengths ( thresholds ) distributed uniformly within the interval @xmath322 $ ] : @xmath323 for an applied stress @xmath324 none of the fibers break , though they are elongated by an amount @xmath325 .
the dynamics of breaking starts when applied stress @xmath48 becomes greater than @xmath326 . for @xmath327 , the basic recursion relation ( eq .
[ rec - u ] ) takes the form : @xmath328,\label{nonlinear-1}\ ] ] which has stable fixed points : @xmath329.\label{nonlinear-2}\ ] ] the model now has a critical point @xmath330 $ ] beyond which the bundle fails completely . at each fixed point
, there will be an equilibrium elongation @xmath331 and a corresponding stress @xmath332 develops in the system ( bundle ) . from eq .
( [ nonlinear-1 ] ) , one gets ( for @xmath327 ) @xmath333 also , from the force balance condition , at each fixed - point @xmath334 .
therefore , the stress - strain relation for the els model finally becomes : @xmath335 .,width=264,height=188 ] the stress - strain relation in a els bundle is shown in fig .
[ fig : stress - strain ] , where the initial linear region has unit slope ( the force constant of each fiber ) .
this hooke s region for the stress @xmath336 continues up to the strain value @xmath337 , until which no one of the fibers breaks .
after this , nonlinearity appears due to the failure of a few of the fibers and the consequent decrease of @xmath50 .
it finally drops to zero discontinuously by an amount @xmath338 $ ] at the breaking point @xmath60 or @xmath339 for the bundle .
it may be noted that in this model the internal stress @xmath340 is universally equal to @xmath341 , independent of @xmath326 at the failure point @xmath60 .
a low cutoff in the fiber threshold distribution excludes the presence of very weak fibers in a bundle .
the weaker fibers mainly reduces the strength of a bundle .
but in practice we always try to build stronger and stronger materials ( ropes , cables etc . ) from the fibrous elements .
therefore this situation ( exclusion of weaker fibers ) is very realistic . in this section
we discuss the effect @xcite of a low cutoff on the failure properties of els bundles .
we follow the weakest fiber breaking approach @xcite : the applied load is tuned in such a way that only the weakest fiber ( among the intact fibers ) will fail after each step of loading .
we first find the extreme condition when the whole bundle fails instantly after the first fiber ruptures .
as the strength thresholds of @xmath0 fibers are uniformly distributed between @xmath326 and @xmath184 ( eq . [ nonlinear - dist ] ) , the weakest fiber fails at a stress @xmath326 ( for large @xmath0 ) .
after this single fiber failure , the load will be redistributed within intact fibers resulting a global stress @xmath342 .
now , the number of intact fibers having strength threshold below @xmath343 is @xmath344 stress redistribution can break at least another fiber if @xmath345 and this ` second ' failure will trigger another failure and so on .
thus the successive breaking of fibers can not be stopped till the complete collapse of the bundle .
clearly , there can not be any fixed point ( critical point ) for such ` instant failure ' situation .
putting the value of @xmath343 we get @xmath346 which gives @xmath347 in the large @xmath0 limit the above condition can be written as @xmath348 . therefore , the condition to get a fixed point in the failure process is @xmath349 . for a els model having @xmath350 fibers .
the dotted line represents the analytic form ( eqn .
[ step no ] ) , triangles are the simulated data for a strictly uniform strength distribution , and the circles represent the data ( averages are taken for @xmath351 samples ) for a uniform on average distribution .
, width=226,height=188 ] we can also calculate how many steps are required to attain the final catastrophic failure for @xmath349 .
let us assume that we have to increase the external load @xmath352 times before the final failure .
at each step of such load increment only one fiber fails . then after @xmath352 step the following condition should be fulfilled to have a catastrophic failure : @xmath353}p(y)dy\geq1.\label{uni - step-1}\ ] ] where @xmath354 the solution gives @xmath355 the above equation suggests that at @xmath356 , @xmath357 .
but in reality we have to put the external load once to break the weakest fiber of the bundle
. therefore , @xmath358 for @xmath348 ( fig .
[ fig : els - step ] ) . to check the validity of the above calculation we take ` strictly uniform ' and uniform on average distributions of fiber strength . in our ` strictly uniform ' distribution
the strength of the @xmath359-th fiber ( among @xmath0 fibers ) is @xmath360 .
we can see in fig .
[ fig : els - step ] that the ` strictly uniform distribution ' exactly obeys the analytic formula ( eq . [ step no ] ) but the uniform on average distribution shows slight disagreement which comes from the fluctuation in the distribution function for a finite system size .
this fluctuation will disappear in the limit @xmath361 where we expect perfect agreement . if the contribution to breakdown phenomena in materials science by statistical physics were to be expressed in one word , that word would have to be fluctuations " . in the context of fiber bundles
, this concept refers to the effects of the fibers each having properties that are statistically distributed around some mean , which can not be reproduced by substituting the fiber bundle by an equivalent one where each fiber is identical to all the others .
intuitively , it is not difficult to accept that fluctuations must play an important role in the breakdown properties of fiber bundles or in fracture and breakdown phenomena in general . a plane ride in turbulent weather
compared to one in smooth weather is a reminder of this .
closely connected to the question of fluctuations is that of phase transitions and criticality @xcite . leaving the fiber bundles for a moment ,
consider a fluid whose temperature is slowly raised . at a well - defined temperature determined by the surrounding pressure , the fluid starts to boil .
each gas bubble that rise to the surface is due a fluctuation being larger than a well - defined size for which the bubble grows rather than shrinks away . at a particular pressure the character of the boiling changes character .
there is no longer any size that determined whether a nascent bubble grows or shrinks .
there are bubbles of all sizes . at this particular point , the system is critical and undergoes a second order phase transitions .
the boiling process at other pressures signals a first order transition .
a brittle material under stress develops microcracks .
these appear where the material is weak or where the local stress field is high . as the stress increases ,
more and more microcracks accumulate until either one or a few microcracks go unstable and grow to macroscopic dimensions causing failure .
the spatial fluctuations of the local material properties cause the appearance of microcracks .
their subsequent growth accentuate these initial fluctuations , but in a highly complex manner due to interactions between the growing cracks .
there are similarities between this scenario and a first order transition @xcite . on the other hand , stable mode i crack growth
as studied experimentally by mly and schmittbuhl @xcite shows all the signs of the advancing crack front showing a dynamics compatible with being at a critical point .
we now turn to the global load sharing fiber bundle model in light of the preceeding remarks .
when a fiber ruptures somewhere , the stress on the intact fibers increases .
this may in turn trigger further fiber failures , which can produce bursts ( avalanches ) that either lead to a stable situation or to breakdown of the whole bundle .
a burst is usually defined as the amount or number ( @xmath362 ) of simultaneous fiber failure during loading .
one may study the distribution @xmath363 of the bursts appear during the entire failure process until the complete breakdown of the bundle .
the property of the fiber bundle model of interest in the present context , is the fluctuation driven _ burst distribution . _ in order to define this property , we again consider a finite bundle containing @xmath0 elastic fibers whose strength thresholds are picked randomly from a probability density @xmath3 .
let @xmath364 be the ordered sequence of failure thresholds : @xmath365 .
then the the external load or force @xmath15 on the bundle ( eq . [ load ] ) at the point where @xmath359th fiber is about to fail can be written as : @xmath366 where elastic constant of the fibers is set equal to unity as before .
note that the sequence of external loads @xmath367 is not monotonously increasing .
this may be readily seen from eq .
( [ load - hh ] ) ; the total load is the product of a monotonously increasing _ fluctuating _
quantity @xmath364 and a monotonously _ decreasing _
quantity @xmath368 .
suppose now that our control parameter is the total load @xmath15 , and that @xmath369 fibers have broken . in order to be in this situation , @xmath370 for all @xmath371 .
the latter inequality ensures that the situation we are studying is not unstable .
we increase @xmath15 until it reaches @xmath367 , at which fiber @xmath359 breaks .
if now @xmath372 , then fiber @xmath373 will also break without the external load @xmath15 being further increased .
the same may be true for @xmath374 and so on until the @xmath375th bond breaks .
thus , @xmath376 for @xmath377 .
if now @xmath378 , the burst of breaking bonds then stops at this point , and we have experienced a burst event of size @xmath362 . the total force @xmath15 expressed as a function of elongation @xmath2 , is shown in fig . [ khh_rmp ] . when the control parameter is elongation @xmath2 , the solid curve is followed
however , when the force @xmath15 is the control parameter , the broken line , given by @xmath379 where lfm designates the _ least monotonic function .
_ as a function of @xmath2 .
however , when our control parameter is @xmath15 rather than @xmath2 , the system will follow the dotted line , eq .
( [ lfm ] ) .
the bursts are the horizontal parts of @xmath380 . here
@xmath381.,width=264 ] _ ( a ) generic case _
it was shown in @xcite that the average number of burst events of size @xmath362 per fiber , @xmath382 , follows a power law of the form @xmath383 in the limit @xmath384 . here
, @xmath385 is the _ universal _ burst exponent . the value ( eq . [ eq10 ] ) is , under very mild assumptions , independent of the threshold distribution @xmath156 : the probability density needs to have a quadratic maximum somewhere in the interval @xmath386 .
we demonstrate this in fig .
[ ah - fig1 ] .
the prefactor @xmath178 in eq .
( [ eq9 ] ) is given by @xmath387}}\;,\ ] ] where @xmath388 is the solution of the equation @xmath389 and is the value of @xmath2 for which the characteristics has a maximum .
( [ eq9 ] ) to ( [ eq12 ] ) were derived in @xcite using combinatorial arguments .
we will , however , in the following , take an alternative route based on a mapping between the global load sharing model and a brownian process @xcite .
before we explain this mapping we quote , for later comparison , the pertinent results of the hemmer - hansen analysis @xcite : the probability @xmath390 that a burst event at elongation @xmath2 will have the size @xmath362 is @xmath391 e^{m(x)-1}\right]^\delta \;,\ ] ] where @xmath392 note in particular that by eq .
( [ eq12 ] ) @xmath393 .
let us now assume that we do not load the fiber bundle until complete collapse , i.e. , until @xmath394 , but stop at a value @xmath395 .
we may then ask for @xmath396 , the expected number of burst events of size @xmath362 during the breakdown process that occurs between @xmath397 and @xmath398 .
this is given by the integral @xmath399e^{m(x)}\right]^\delta\;,\\ \nonumber\end{aligned}\ ] ] where on the right - hand side the stirling approximation @xmath400 for large @xmath362 has been used .
the integrand in eq.([eq15 ] ) is strongly peaked near @xmath401 .
we therefore expand it to second order in @xmath402 to find @xmath403 where we have extended the upper integration limit to @xmath404 .
we may do this integral to get @xmath405 where @xmath178 is defined by eq .
( [ eq11 ] ) . we may write ( [ eq17 ] ) in scaling form , @xmath406 where @xmath407 in particular @xmath408 tends to the constant @xmath178 for @xmath409 . two universal critical exponents appear , @xmath410 , eq . ( [ eq10 ] ) , and @xmath411 it is , thus , in the above sense , that the fracture process of the fiber bundle approaches a critical point at total breakdown : the distribution of burst events follows a power law with an upper cutoff that diverges as the bundle approaches total failure . for the uniform distribution ( a ) and the weibull distribution with index 5 ( b ) .
the dotted lines represent the power law with exponent @xmath410 .
both figures are based on @xmath412 samples of bundles each with @xmath413 fibers.,width=226 ] for the uniform distribution ( a ) and the weibull distribution with index 5 ( b ) .
the dotted lines represent the power law with exponent @xmath410 .
both figures are based on @xmath412 samples of bundles each with @xmath413 fibers.,width=226 ] in @xcite and later on in full detail in @xcite , the burst distribution ( eq . [ eq9 ] ) was derived from the assumption that @xmath367 may be directly interpreted as a biased random walk .
the precise nature of this random walk is elucidated below .
it is a peculiar asymmetric walk with variable step length . in the limit @xmath414 and continuous time variable @xmath415 and @xmath416
, this random walk may be mapped onto a continuous brownian process .
such brownian processes have been studied by @xcite , @xcite and @xcite in connection with the distribution of the strength @xmath336 of fiber bundles .
we will in the following derive eq .
( [ eq15 ] ) by means of a biased random - walk model with variable step length .
we find this an interesting example of universality in statistical physics : the asymptotic behavior of one model is found by using a different model with the same asymptotic behavior as the first one , but which is simpler to solve before the continuum limit is taken .
under increasing load the variation of the force per fiber , @xmath417 , will consist of a systematic nonfluctuating part , given by the average load - elongation characteristics , with a small fluctuation of order @xmath418 superimposed .
the precise value of the force fluctuation depends upon whether one studies the force @xmath419 at given elongation , or the force @xmath420 at which fiber number @xmath359 breaks .
let us for both quantities calculate the variance of @xmath421 , @xmath422 , starting with the constant-@xmath359 ensemble .
the force per fiber when the @xmath359th fiber is about to break is , by @xmath423 , @xmath424 x_k \ ; , \ ] ] where @xmath364 is the elongation when the @xmath359th fiber breaks , and we have defined @xmath425 by @xmath426 for large @xmath0 @xmath425 is essentially the average value of @xmath364 . for a fixed @xmath359
the variance of @xmath420 is by eq .
( [ eq21 ] ) given by the variance of @xmath364 : @xmath427 ^ 2 \sigma_{x_k}^2 \ ; , \ ] ] and we seek therefore the probability @xmath428 that the @xmath359th threshold in the ordered threshold sequence lies in the interval @xmath429 .
this probability equals @xmath430^{n - k } p(x ) dx\;.\nonumber\\ \label{eq24}\end{aligned}\ ] ] for large @xmath359 and @xmath0 , and using eq .
( [ eq22 ] ) , this is close to the gaussian distribution @xmath431}}\right)^{1/2}\nonumber\\ & e^{-np(\overline{x}_k)^2(x-\overline{x}_k)^2 /2p(\overline{x}_k ) [ 1-p(\overline{x}_k ) ] } \ dx\;.\\ \nonumber\end{aligned}\ ] ] this gives the variance of @xmath364 and thus of @xmath420 , @xmath432 ^ 3}\over { np(\overline{x}_k)^2 } } .\ ] ] let us now compare with the force fluctuation at constant elongation .
the force per fiber is the following function of elongation @xmath2 , @xmath433 where @xmath434 is the breakdown threshold for the @xmath435th fiber , and @xmath436 is the heavyside function .
this gives immediately the average force @xmath437\;,\ ] ] i.e. , the characteristics , as well as the variance @xmath438}\over n}\;.\ ] ] although the two types of force fluctuations have different variances , in both cases @xmath439 .
the nonmonoticities of the force @xmath421 within the fluctuation zone produce bursts .
since the fluctuations are so small for large @xmath0 , one can treat the burst events _ locally .
_ we now consider the force sequence @xmath367 as a stochastic process .
since we seek the _ asymptotic _ burst distribution , we are interested in the behavior after many steps of the process .
it is convenient , however , to start with the one - step process .
let us determine the probability distribution of the force increase @xmath440 between two consecutive bursts , the first one taking place at elongation @xmath364 with @xmath441 . since @xmath442
, we have @xmath443 the probability to find the @xmath373th threshold in @xmath444 for given @xmath364 , @xmath445^{n - k-2}}\over{[1-p(x_k ) ] } } p(x_{k+1 } ) dx_{k+1}\ ; , \label{eq31}\ ] ] gives directly by using the connection @xmath446 , the probability density @xmath447 of @xmath448 @xmath449^{n - k-2}}\over{[1-p(x_k)]^{n - k-1}}}\cr & p(x_k+{{\delta f+x_k}\over{n - k}})\;.\\ \nonumber\end{aligned}\ ] ] for large @xmath450 this simplifies to @xmath451}}\right ] & for $ \delta f\ge -x_k$,\cr 0 & for $ \delta f < -x_k$.\cr}\nonumber\\\end{aligned}\ ] ] this one - dimensional random walk is asymmetric in more than one way .
first of all , it has nonzero bias @xmath452 in addition the probability distribution around this average is very asymmetric .
the variance is easily determined , @xmath453 ^ 2 \;.\ ] ] the brownian - motion limit of a one - dimensional random walk is completely determined by the first and second moments of the single - step probability distribution .
the results just obtained enables us therefore to select an `` ordinary '' biased random walk with constant step length @xmath454 which has the same brownian motion limit as the burst process .
we imagine having a one - dimensional random walk along the @xmath455-axis with a constant bias .
each step is of length @xmath454 .
let the probability to take a step in the negative @xmath455 direction be @xmath456 , and let @xmath457 be the probability to take a step in the positive @xmath455 direction .
the walk is biased when @xmath457 is different from @xmath456 .
the probability distribution of the position @xmath458 after one step has the average @xmath459 and variance @xmath460 elimination of @xmath454 yields @xmath461 since @xmath462 , the bias parameters are determined .
after @xmath359 steps a gaussian distribution , @xmath463 ^ 2/8pqa^2 k}\over{\sqrt{8\pi pqa^2k}}}\ ] ] is approached when @xmath359 increases .
the two processes will have the same asymptotic behavior when we make the identification @xmath464 where @xmath465 is defined by eq .
( [ eq14 ] ) . when the bias is small , both @xmath457 and @xmath456 are close to 1/2 , and we have to lowest order @xmath466\;,\nonumber\\ q=\textstyle { 1\over 2}[1-m ( x ) ] \;.\\
\nonumber\end{aligned}\ ] ] we have now made the promised mapping between the fiber bundle problem and a random walk with a constant bias .
a constant bias may be used since bursts can be treated locally . the next step is to calculate the burst distribution for such a biased random walk . since this biased random walk by construction
has the same asymptotic behavior in the limit @xmath467 as the original fiber bundle problem , the two burst distributions will asymptotically be the same . in terms of the biased random walk , a burst event of size @xmath362 at time
" @xmath359 may be defined as follows : ( i ) @xmath468 for @xmath469 and @xmath470 .
( ii ) furthermore , to ensure that we are not counting burst event _ inside _ other burst events , the condition @xmath471 for @xmath472 is necessary .
the first condition is in fact a special case of the gambler s ruin " problem @xcite .
a gambler plays a series of independent games against a bank with infinite resources . in each game , the gambler either looses or wins one euro , and the probability that the bank wins is @xmath473 , while the probability that the gambler wins is @xmath474 .
if the gambler starts out with a capital of @xmath455 rs , the probability that she is ruined after precisely @xmath362 games is @xmath475 the probability that condition i is fulfilled for a biased random walk burst of size @xmath362 is then @xmath476 where we have assumed that @xmath477 .
the probability that a biased random walker returns at least once to the origin is @xcite @xmath478 .
the probability that condition ( ii ) , namely that @xmath479 for all @xmath480 is fulfilled , is then simply @xmath481 , and we have that the probability for having a burst of size @xmath362 happening at time " @xmath359 is @xmath482 where we in addition have assumed that @xmath483 .
returning to the fiber bundle model , the bias @xmath484 .
when @xmath2 is close to @xmath388 , we have @xmath485 where @xmath486 .
thus , the probability to have a burst of size @xmath362 between @xmath487 and @xmath488 is @xmath489 thus , the cumulative burst distribution up the elongation @xmath490 is @xmath491 comparing this expression to eq .
( [ eq16 ] ) , we see that they are identical .
this completes the derivation of the asymptotic burst distribution via the mapping between the random - walk problem and the burst process .
_ ( b ) special cases _ the burst distribution given in eq .
( [ eq15 ] ) is valid when the threshold distribution has a parabolic maximum inside the interval of the thresholds .
we now consider threshold distributions that do _ not _ reach their maximum at the boundaries of the interval @xcite .
model examples of such threshold distributions are @xmath492^{-\alpha_0 } & \mbox{for } \ ; x > x_0 \end{array } \right . \label{brede}\ ] ] here @xmath493 and @xmath494 are positive parameters , and @xmath495 is a reference quantity which we for simplicity put equal to unity in the following .
these distributions are all characterized by diverging moments . when @xmath496 , even the first moment the mean as well as all other moments diverge .
this class of threshold distributions are rich enough to exhibit several qualitatively different burst distributions . for the distribution ( eq . [ brede ] ) , with @xmath497 , and for @xmath498 ( upper curve ) , @xmath190 ( middle curve ) , and @xmath499 ( lower curve ) .
the broken part of the @xmath500-curve is unstable and the macroscopic bundle strength will follow the solid line.,width=226 ] the corresponding macroscopic bundle strength per fiber is @xmath501 in fig .
[ fig-3b-6 ] the corresponding macroscopic force curves @xmath502 are sketched .
we note that when @xmath503 , the plateau in eq .
( [ macro ] ) becomes infinitely wide .
the distribution of burst sizes is given by eq .
( [ eq15 ] ) . in the present case
the function @xmath465 takes the form @xmath504 a simple special case is @xmath505 , corresponding to @xmath506 since then the function ( [ af ] ) is independent of @xmath421 : @xmath507 this gives at once @xmath508^{\delta}\nonumber\\ & \simeq & \frac{1-\alpha_0}{\alpha_0\sqrt{2\pi}}\;\delta^{-\frac{3}{2 } } \left[\alpha_0 e^{1-\alpha_0}\right]^{\delta}. \label{one}\end{aligned}\ ] ] in other cases it is advantageous to change integration variable in eq .
( [ eq15 ] ) from @xmath2 to @xmath509 : @xmath510 the asymptotics for large @xmath362 , beyond the @xmath511 dependence of the prefactor , is determined by the @xmath362-dependent factor in the integrand . the maximum of @xmath512 is unity , obtained for @xmath513 , and the asymptotics depends crucially on whether @xmath514 falls outside the range of integration , or inside ( including the border ) . if the maximum falls inside the range of integration the @xmath515 dependence remains .
a special case of this is @xmath516 , for which the maximum of the integrand is located at the integration limit and the macroscopic force has a `` quadratic '' maximum _ at infinity .
_ another special case is @xmath517 ( and @xmath518 ) , for which again the standard asymptotics @xmath519 is valid . in this instance
the macroscopic force has a quadratic _ minimum _ at @xmath520 ( see fig .
[ fig-3b-6 ] for @xmath521 ) , and critical behavior arises just as well from a minimum as from a maximum . in the remaining cases , in which @xmath514 is not within the range of integration in eq .
( [ integral ] ) , the burst distribution is always a power law with an exponential cut - off , @xmath522 here @xmath523 and @xmath524 depend on the parameter values @xmath494 and @xmath493 , however .
this is easy to understand .
since @xmath525 we see that @xmath465 is a monotonically decreasing function for @xmath526 , so that the maximum of @xmath512 is obtained at the lower limit @xmath520 , where @xmath527 .
the asymptotics @xmath528 follows .
this is true merely for @xmath529 , however . for @xmath530 the macroscopic force @xmath531 _ decreases _ near @xmath520 so that a macroscopic burst takes place at a force @xmath494 per fiber , and stabilization is obtained at a larger elongation @xmath532 ( fig .
[ fig-3b-6 ] ) .
the subsequent bursts have an asymptotics @xmath533 determined by the neighborhood of @xmath534 . for @xmath535 ,
the maximum of @xmath512 is obtained at @xmath536 , leading to the asymptotics @xmath537 reflecting the power - law behavior of the integrand at infinity .
[ ah - tab1 ] .[tab1 ] asymptotic behavior of the burst distribution for strong threshold distributions in the els model . [ cols="^,^",options="header " , ] the results are summarized in table [ tab1 ] .
note that the @xmath505 result ( [ one ] ) can not be obtained by putting @xmath505 in eq .
( [ two ] ) since in ( [ integral ] ) the order of the limits @xmath538 and @xmath539 is crucial . _
( c ) crossover behavior _
when all the bursts are recorded for the entire failure process , we have seen that the burst distribution @xmath363 follows the asymptotic power law @xmath540 . if we just sample bursts that occur near the breakdown point , a different behavior is seen . as an illustration we consider the uniform threshold distribution , and compare the complete burst distribution with what one gets when one samples merely burst from breaking fibers in the threshold interval @xmath541 .
[ fig-3b-2 ] shows clearly that in the latter case a different power law is seen . , with @xmath542 and with @xmath543 .
the figure is based on 50 000 samples , each containing @xmath413 fibers .
the exponents of the distributions are @xmath410 and @xmath544 respectively.,width=226 ] this observation may be of practical importance , as it gives a criterion for the imminence of catastrophic failure @xcite .
this proposal has so far not been tested experimentally .
however , it is enticing to note the recent observation by kawamura of a crossover behavior in the magnitude distribution of earthquakes before large earthquakes appear @xcite .
we return to this result in section v.c and fig .
[ fig : kawamura ] .
we introduce the following notation in eq .
( [ eq15 ] ) , @xmath545^{\delta-1 } \exp\left[\delta\ , r(x)\right]dx\;,\nonumber\\\end{aligned}\ ] ] where @xmath546\ ; , \label{eq-3b-8}\ ] ] and @xmath547 .
we note that @xmath548 vanishes at the point @xmath12 .
if we have a situation in which the weakest fiber has its threshold @xmath549 just a little below the critical value @xmath12 , the contribution to the integral in the expression ( eq . [ eq-3b-1 ] ) for
the burst distribution will come from a small neighborhood of @xmath12 .
since @xmath548 vanishes at @xmath12 , it is small here , and we may in this narrow interval approximate the @xmath362-dependent factors in eq .
( [ eq-3b-1 ] ) as follows @xmath550\nonumber \\ & = \exp[-\delta(r^{2}/2+{\mathcal{o}}(r^{3 } ) ) ] \approx\exp\left[-\delta r(x)^{2}/2\right]\nonumber\\ \label{eq-3b-6}\end{aligned}\ ] ] we also have @xmath551 inserting everything into eq .
( [ eq-3b-1 ] ) , we obtain to dominating order @xmath552_{x_{0}}^{x_{c}}\nonumber \\ & = & \frac{\delta^{\delta-2}\ , e^{-\delta}}{\delta ! } \frac{p(x_{c})}{\left|r'(x_{c})\right| } \left[1-e^{-\delta/\delta_{c}}\right]\ ; , \label{eq-3b-15}\end{aligned}\ ] ] with @xmath553 by use of the stirling approximation @xmath554 , the burst distribution ( eq . [ eq-3b-15 ] ) may be written as @xmath555 with a nonzero constant @xmath556 we can see from eq .
( [ eq-3b-12 ] ) that there is a crossover at a burst length around @xmath557 , @xmath558 fibers .
results with @xmath542 , i.e. , when all bursts are recorded , are shown as squares and data for bursts near the critical point ( @xmath543 ) are shown by circles.,width=264 ] we have thus shown the existence of a crossover from the generic asymptotic behavior @xmath540 to the power law @xmath559 near criticality , i.e. , near global breakdown . the crossover is a universal phenomenon , independent of the threshold distribution @xmath3 .
the simulation results we have shown so far are based on _ averaging _ over a large number of samples . for applications
it is important that crossover signal can be seen also in a single sample .
we show in fig .
[ fig3b-3 ] that equally clear crossover behavior is seen in a _ single _
fiber bundle when @xmath0 is large enough .
also , as a practical tool one must sample finite intervals ( @xmath560 , @xmath343 ) during the fracture process .
the crossover will be observed when the interval is close to the failure point @xcite . .
each bond is a fuse with a burn - out threshold @xmath18 drawn from a probability distribution @xmath561 .
, width=240,height=230 ] the els fiber bundle model is a simple model in that it is analytically tractable .
a step up in complexity from the els fiber bundle model , is the _ random fuse model _ @xcite .
while resisting most analytical treatments , this model retains _ computational _ tractability .
the fuse model consists of a lattice in which each bond is a fuse , i.e. , an ohmic resistor as long as the electric current it carries is below a threshold value . if the threshold is exceeded , the fuse burns out irreversibly .
the threshold @xmath18 of each bond is drawn from an uncorrelated distribution @xmath561 .
the lattice is placed between electrical bus bars and an increasing current is passed through it .
the lattice is a two - dimensional square one placed at @xmath562 with regards to the bus bars , and the kirchhoff equations are solved numerically at each node assuming that all fuses have the same resistance .
we show the model in fig .
[ fig : fusenet ] . the els fiber bundle model may be interpreted as a mean - field version of the random fuse model @xcite .
hence , the random fuse model may be used as a testing ground for results ( see table [ els - tab ] ) found with the els fiber bundle model to explore their robustness when other effect not present in the fiber bundle model enter . to test the crossover phenomenon in a more complex situation than for els fiber bundle model
, we consider the random fuse model @xcite .
when one records all the bursts in the random fuse model , the distribution follows a power law @xmath563 with @xmath564 , which is consistent with the value reported in recent studies .
we show the histogram in fig .
[ fig : fuse - cross ] . with a system size of @xmath565 , @xmath566 fuses blow on the average before catastrophic failure sets in .
when measuring the burst distribution only after the first @xmath567 fuses have blown , a different power law is found , this time with @xmath568 .
after @xmath569 blown fuses , on the other hand , @xmath523 remains the same as for the histogram recording the entire failure process ( fig .
[ fig : fuse - cross ] ) . .
the threshold @xmath18 are uniformly distributed on the unit interval . on the average , catastrophic failure sets in after @xmath566 fuses have blown .
the circles denote the burst distribution measured throughout the entire breakdown process .
the squares denote the burst distribution based on bursts appearing after the first @xmath569 fuses have blown .
the triangles denote the burst distribution after @xmath567 fuses have blown .
the two straight lines indicate power laws with exponents @xmath570 and @xmath568 , respectively.,width=264 ] in fig .
[ fig : fig10 - 11 ] , we show the power dissipation @xmath571 in the network as a function of the number of blown fuses and as a function of the total current .
the dissipation is given as the product of the voltage drop across the network @xmath572 times the total current that flows through it .
the breakdown process starts by following the lower curve , and follows the upper curve returning to the origin .
it is interesting to note the linearity of the unstable branch of this curve . in fig
[ fig : fig12 ] , we record the avalanche distribution for power dissipation , @xmath573 . as a function of the number of broken bonds ( upper ) and as a function of the total current @xmath574 flowing in the fuse model ( lower ) . ,
title="fig:",width=226 ] .2 in as a function of the number of broken bonds ( upper ) and as a function of the total current @xmath574 flowing in the fuse model ( lower ) .
, title="fig:",width=226 ] .1 in recording , as before , the avalanche distribution throughout the entire process as well as recording only close to the point at which the system catastrophically fails , result in two power laws , with exponents @xmath575 and @xmath576 , respectively .
it is interesting to note that in this case there is not a difference of unity between the two exponents .
the power dissipation in the fuse model corresponds to the stored elastic energy in a network of elastic elements .
hence , the power dissipation avalanche histogram would in the mechanical system correspond to the released energy . such a mechanical system could serve as a simple model for earthquakes . for the fuse model .
the slopes of the two straight lines are @xmath577 and @xmath578 , respectively .
the circles show the histogram of avalanches recorded after 1000 fuses has blown , whereas the squares show the histogram recorded after @xmath567 fuses have blown .
this is close to catastrophic failure.,width=264 ] @xcite studied the critical behavior of a bundle of fibers under global load sharing scheme with threshold strength chosen randomly from a distribution which is uniform , but discontinuous .
the form of the distribution is @xmath579 where @xmath580 is the gap in the threshold distribution as shown in fig .
[ ah - dutta1 ] . here ,
a fraction @xmath421 of the fibers belong to the weaker section @xmath581 and the remaining to the stronger section @xmath582 .
the condition of uniformity of the distribution demands @xmath583 so that fixing @xmath532 and @xmath421 immediately settles the value of @xmath584 . to study the dynamics of this model , once again the recursive equation approach was used .
importantly , the redistributed stress must cross @xmath584 for the complete failure of the bundle to take place .
when the external load is such that the redistributed stress at an instant @xmath18 , i.e. , @xmath585 is greater than @xmath584 , the fixed point solution has the form @xmath586\ ; , \label{dd3}\ ] ] so that the critical stress @xmath90 is @xmath587 } \label{dd4}\ ] ] and the redistributed stress at the critical point is found to be @xmath341 as in the uniform distribution @xcite .
this immediately restricts the value of @xmath584 to be less than @xmath341 and therefore , @xmath588 is defined as the critical distribution in this model .
on the other hand , uniformity condition , equation ( [ dd2 ] ) , puts another restriction , namely @xmath589 .
interestingly , the critical stress is a function of the gap @xmath580 and reduces to one fourth when the gap goes to zero .
the exponents related to the order parameter and susceptibility stick to their mean field values .
however , the existence of a forbidden region shows a prominent signature in the avalanche size distribution of the mixed model .
the expression for the total avalanche size distribution @xmath363 in this model includes two terms , one due to the contribution from thresholds between @xmath185 to @xmath532 and the other from the stronger section of fibers .
hence , the total avalanche size @xmath363 is @xmath590 where @xmath591^{\delta}\nonumber\\ \label{dd5}\end{aligned}\ ] ] and @xmath592^{\delta}\;. \label{dd6}\end{aligned}\ ] ] the leading behavior of @xmath593 is given by @xcite @xmath594 where @xmath595 which clearly indicates a rapid fall of the contribution of weaker fibers .
on the other hand , @xmath596 resembles the imminent failure behavior studied by @xcite , where the avalanche size exponent shows a crossover from @xmath597 to @xmath598 as @xmath599 . for the mixed model
, the total avalanche size distribution @xmath363 shows a nonuniversal behavior for small @xmath362 values , though eventually there is a crossover to the universal mean field value .
the most fascinating observation is the following : though the gap in the distribution is always present , nonuniversality is only prominent in the limit @xmath600 .
divakaran and dutta showed that this nonuniversal behavior stems from the avalanche of fibers in the weak section and only in the vicinity of the critical distribution , the contribution of @xmath593 overcomes @xmath596 .
otherwise , the faster fall of @xmath593 and large value of @xmath596 together force the avalanche size exponent to be @xmath597 .
it is to be noted that the nonuniversal behavior is most prominent at a critical distribution where the avalanche size exponent crosses over to @xmath598 in the asymptotic limit .
the typical behavior of @xmath363 is shown in fig .
[ ah - dutta2 ] for two different distributions highlighting the increase in nonuniversal region as @xmath584 approaches 0.5 .
however , for many discontinuities in the threshold distributions , avalanche size distribution shows a nonuniversal , non power - law behavior @xcite for small - size avalanches , although the large avalanches still exhibit similar crossover behavior as we discuss here .
divakaran and dutta also looked at the model where fibers from two different weibull distributions are mixed @xcite .
though an interesting variation of the critical stress with the mixing parameter was obtained using a probabilistic method introduced by @xcite , there is no deviation in the avalanche size exponent . in a recent paper ,
@xcite studied the infinite gap limit of the discontinuity model . here
, they considered a fraction @xmath601 of the fibers having infinite threshold strength mixed with fibers having threshold chosen from a distribution @xmath3 .
they observed a critical fraction @xmath602 such that for @xmath603 , the avalanche size exponent switches from the well known mean field exponent @xmath604 to a lower value @xmath605 .
it was also showed that such a behavior is observed for those distributions where the macroscopic constitutive behavior has a maxima and a point of inflexion .
it is also claimed that below a critical gap , the hidalgo et al .
model reduces to the discontinuity model of divakaran and dutta .
@xcite studied the global load sharing fiber bundle model in a wedge - shaped geometry .
that is , the fibers are connected to two rigid blocks placed at an angle with respect to each other .
the fibers are loaded by rotating the blocks with respect to each other , resulting in a linear loading gradient on them . in the limit of a threshold distribution tending towards zero width ,
the fibers break in an orderly fashion according to the load , and hence , position in the wedge . as the width
is increased , a process zone i.e. , a zone where some fibers fail whereas others stay intact develops .
when the width of the threshold distribution is wide enough , the process zone spans the entire bundle . in this limit a burst size exponent
@xmath410 is recovered .
however , with a narrower distribution so that a well - defined process zone smaller than the size of the bundle develops , the burst exponent @xmath606 is found .
when the bundle is stretched continuously from zero , fluctuation plays crucial role and the generic result is a power law @xcite @xmath607 , for large @xmath362 , with @xmath410 .
however , experiments may be performed in a different manner , where the load is increased in finite steps of size @xmath608 . the value of the exponent increases @xcite then to @xmath609 : @xmath610 .
the basic reason for the difference in the power laws is that increasing the external load in steps reduces the fluctuations in the force .
the derivation ( see section iii.b.1 ) of the asymptotic size distribution @xmath611 of avalanches , corresponding to stretching by infinitesimal steps , shows the importance of force fluctuations @xcite .
an effective reduction of the fluctuations requires that the size @xmath608 of the load increase is large enough so that a considerable number of fibers break in each step .
here is an analytic derivation , following @xcite , how to calculate the burst distribution in such situation . for the uniform distribution of thresholds ( eq . [ uniform ] ) ,
the load curve is parabolic , @xmath612 so that the expected critical load equals @xmath613 . with a sufficiently large @xmath608 we may use the macroscopic load equation ( eq . [ loaduniform ] ) to determine the number of fibers broken in each step .
the load values are @xmath614 , with @xmath509 taking the values @xmath615 for the uniform threshold distribution . by eq .
( [ loaduniform ] ) the threshold value corresponding to the load @xmath614 is @xmath616 the expected number of fibers broken when the load is increased from @xmath614 to @xmath617 is close to @xmath618 here the minimum number of @xmath362 is @xmath608 , obtained in the first load increase .
the integral over all @xmath509 from @xmath185 to @xmath619 yields a total number @xmath620 of broken fibers , as expected , since the remaining one - half of the fibers burst in one final avalanche .
the number of avalanches of size between @xmath362 and @xmath621 , @xmath622 , is given by the corresponding interval of the counting variable @xmath509 : @xmath623 .
since @xmath624 we obtain the following distribution @xcite of avalanche sizes : @xmath625 for consistency , one may estimate the total number of bursts by integrating @xmath363 from @xmath626 to @xmath404 , with the result @xmath619 , as expected .
[ fig : els - per - rmp1 ] shows that the theoretical power law ( eq . [ guniform ] ) fits the simulation results perfectly for sufficiently large @xmath362 .
the simulation records also a few bursts of magnitude less than @xmath608 because there is a nonzero probability to have bundles with considerably fewer fibers than the average in a threshold interval . however , these events will be of no importance for the asymptotic power law in the size distribution . in order to see whether the asymptotic exponent value @xmath570 is general ,
simulations for another threshold distribution has been performed , the weibull distribution ( eq . [ weibull ] ) with index @xmath225 , which confirms similar asymptotic behavior ( fig .
[ fig : els - per - rmp2 ] ) . for a general threshold distribution
@xmath156 a load interval @xmath608 and a threshold interval are connected via the load equation @xmath627 . since @xmath628 $ ] , an increase @xmath608 in the load corresponds to an interval @xmath629}\label{dx}\ ] ] of fiber thresholds .
the expected number of fibers broken by this load increase is therefore @xmath630 note that this number diverges at the critical point , i.e. at the maximum of the load curve , as expected . following the similar method , as in case of uniform distribution , we can determine @xcite the asymptotic distribution for large @xmath362 : @xmath631 with a nonzero constant @xmath632 where we have used that at criticality @xmath633 .
thus the asymptotic exponent value @xmath570 is universal . for the weibull distribution considered in fig .
[ fig : els - per - rmp2 ] we obtain @xmath634 this burst distribution must be given on parameter form , the elimination of @xmath2 can not be done explicitly .
the critical point is at @xmath635 and the asymptotics is given by eq .
[ dasymp ] , with @xmath636 .
if we let the load increase @xmath608 shrink to zero , we must recover the asymptotic @xmath611 power law valid for continuous load increase .
thus , as function of @xmath608 , there must be a crossover from one behavior to the other .
it is to be expected that for @xmath637 the @xmath611 asymptotics is seen , and when @xmath638 the @xmath610 asymptotics is seen .
so far we have discussed in detail the statistical distribution of the _ size _ of avalanches in fiber bundles @xcite .
sometimes the avalanches cause a sudden internal stress redistribution in the material , and are accompanied by a rapid release of mechanical energy . a useful experimental technique to monitor
the energy release is to measure the acoustic emissions ( ae ) , the elastically radiated waves produced in the bursts @xcite .
experimental observations suggest that ae signals follow power law distributions . what is the origin of such power laws ?
can we explain it through a general scheme of fluctuation guided breaking dynamics that has been demonstrated well in els fiber bundle model ?
we now determine the statistics of the energies released @xcite in fiber bundle avalanches .
as the fibers obey hooke s law , the energy stored in a single fiber at elongation @xmath2 equals @xmath639 , where we for simplicity have set the elasticity constant equal to unity .
the individual thresholds @xmath560 are assumed to be independent random variables with the same cumulative distribution function @xmath156 and a corresponding density function @xmath3 . _
( a ) energy statistics _ let us characterize a burst by the number @xmath362 of fibers that fail , and by the lowest threshold value @xmath2 among the @xmath362 failed fibers .
the threshold value @xmath640 of the strongest fiber in the burst can be estimated to be @xmath641 since the expected number of fibers with thresholds in an interval @xmath642 is given by the threshold distribution function as @xmath643 . the last term in ( eq . [ xmax ] ) is of the order @xmath286 , so for a very large bundle the differences in threshold values among the failed fibers in one burst are negligible . hence the energy released in a burst of size @xmath362 that starts with a fiber with threshold @xmath2
is given as @xmath644 following @xcite the expected number of bursts of size @xmath362 , starting at a fiber with a threshold value in the interval @xmath429 , is @xmath645 where @xmath646 the expected number of bursts with energies less than @xmath571 is therefore @xmath647 with a corresponding energy density @xmath648 explicitly , @xmath649 with @xmath650^{\delta}.\label{gn}\end{aligned}\ ] ] here @xmath651 with a critical threshold value @xmath12 , it follows from ( eq . [ en ] ) that a burst energy @xmath571 can only be obtained if @xmath362 is sufficiently large , @xmath652 thus the sum over @xmath352 starts with @xmath653 $ ] , where @xmath654 $ ] denotes the integer part of @xmath454 .
_ ( b ) high energy asymptotics _ bursts with high energies correspond to bursts in which many fibers rupture . in this range
we use stirling s approximation for the factorial @xmath655 , replace @xmath656 $ ] by @xmath657 , and replace the summation over @xmath362 by an integration . thus @xmath658^{\delta}\ ; d\delta.\label{integral1}\end{aligned}\ ] ] by changing integration variable from @xmath362 to @xmath659 we obtain @xmath660^{\delta}\ ; ds\nonumber \\ & = & \frac{n}{2e^{3/2}\pi^{1/2}}\int\limits _ { 0}^{x_{c}}(1-p(s)-sp(s))e^{-eh(s)}\ ; ds,\nonumber\\ \label{integral2}\end{aligned}\ ] ] with @xmath661\frac{2}{s^{2}}.\label{h}\ ] ] for large @xmath571 the integral ( eq .
[ integral2 ] ) is dominated by the integration range near the minimum of @xmath662 . at the upper limit @xmath663
we have @xmath664 , since @xmath633 .
this is also a minimum of @xmath662 , having quadratic form , @xmath665 inserting these expressions into ( eq . [ integral2 ] ) and integrating , we obtain the following asymptotic expression , @xmath666 where @xmath667}.\label{c}\ ] ] in fig . [
fig : energy_bursts_high ] we compare the theoretical formula with simulations for the uniform distribution ( eq . [ uniform ] ) , which corresponds to @xmath668 , and @xmath669 , and for the weibull distribution ( eq . [ weibull ] ) with index @xmath670 , which corresponds to @xmath671 and @xmath672 .
characterizing energy bursts in fiber bundles with ( a ) the uniform threshold distribution ( eq . [ uniform ] ) and ( b ) the weibull distribution ( eq . [ weibull ] ) of index @xmath673 .
the graphs are based on 1000 samples with @xmath413 fibers in each bundle .
open circles represent simulation data , and dashed lines are the theoretical results ( eqs . [ as ] - [ c ] ) for the asymptotics . , width=240,height=230 ] characterizing energy bursts in fiber bundles with ( a ) the uniform threshold distribution ( eq . [ uniform ] ) and ( b ) the weibull distribution ( eq . [ weibull ] ) of index @xmath673 .
the graphs are based on 1000 samples with @xmath413 fibers in each bundle .
open circles represent simulation data , and dashed lines are the theoretical results ( eqs . [ as ] - [ c ] ) for the asymptotics . , width=240,height=230 ] the corresponding asymptotics ( eq . [ as ] )
are also exhibited in fig .
[ fig : energy_bursts_high ] . for both threshold distributions
the agreement between the theoretical asymptotics and the simulation results is very satisfactory .
the exponent @xmath674 in the energy burst distribution is clearly universal .
note that the asymptotic distribution of the burst magnitudes @xmath362 is governed by the same exponent @xcite . _
( c ) low - energy behavior _
the low - energy behavior of the burst distribution is by no means universal : @xmath675 may diverge , vanish or stay constant as @xmath676 , depending on the nature of the threshold distribution . in fig .
[ fig : energy_bursts_low ] we exhibit simulation results for the low - energy part of @xmath675 for the uniform distribution and the weibull distributions of index @xmath673 and index @xmath225 . , in the low - energy regime , for the uniform threshold distribution ( circles ) , the weibull distribution with @xmath670 ( triangles ) and weibull distribution with @xmath677 ( squares ) .
the graphs are based on 1000 samples with @xmath413 fibers in each bundle .
, width=240,height=230 ] we see that @xmath675 approaches a finite limit in the weibull @xmath670 case , approaches zero for weibull @xmath677 and apparently diverges in the uniform case .
all this is easily understood , since bursts with low energy predominantly correspond to single fiber bursts ( @xmath678 , _
i.e. _ @xmath679 ) and to fibers with low threshold values . the number of bursts with energy less than @xmath571 therefore corresponds to the number of bursts with @xmath680 , which is close to @xmath681 .
this gives @xmath682 for the uniform distribution @xmath675 should therefore diverge as @xmath683 for @xmath676 .
the simulation results in fig . [ fig : energy_bursts_low ] are consistent with this divergence . for the weibull distribution of index 2 , on the other hand
, ( eq . [ enull ] ) gives @xmath684 when @xmath676 , a value in agreement with simulation results in the figure .
note that for a weibull distribution of index @xmath8 , the low - energy behavior is @xmath685 .
thus the weibull with @xmath670 is a borderline case between divergence and vanishing of the low - energy density .
the same lowest - order results can be obtained from the general expression ( eq .
[ g ] ) , which also can provide more detailed low - energy expansions . for high energies
the energy density obeys a power law with exponent @xmath674 .
this asymptotic behavior is universal , independent of the threshold distribution .
a similar power law dependence is found in some experimental observations on acoustic emission studies @xcite of loaded composite materials .
in contrast , the low - energy behavior of @xmath675 depends crucially on the distribution of the breakdown thresholds in the bundle . @xmath675
may diverge , vanish or stay constant for @xmath676 .
[ els - tab ] ccc exponent for & value & comment + order parameter @xmath686 & @xmath341 & @xmath687 + + breakdown susceptibility @xmath688 & @xmath341 & @xmath687 + + relaxation time @xmath689 & @xmath341 & amplitude ratio @xmath690 + + avalanche size distribution @xmath691 & @xmath609 & discrete load increase + & @xmath597 & continuous load increase + + energy burst distribution @xmath692 & @xmath597 & in the asymptotic limit .
+ & & for the low energy limit , + & & distribution is non universal .
so far we have studied fiber bundles where the force once carried by a failing fiber is spread equally among all the surviving fibers .
this may often be a very good approximation .
however , intuitively it is natural that fibers being closer to a failing fiber feel more of an effect than fibers further away an effect reminiscent of stress enhancement around cracks .
we will in this section discuss three classes of models where there are local effects in how the forces carried by failed fibers are distributed .
we start with the most extreme where the forces are totally absorbed by the nearest surviving fibers .
we then move on to models where the stress is distributed according to a power law in the distance from the failing fiber , and lastly a model where we assume the _ clamps _ the fibers are attached to are soft and therefore deform due to the loading of the fibers as can be seen in pulling on the hairs on one s arm .
the extreme form for local load redistribution is that all extra stresses caused by a fiber failure are taken up by the _ nearest - neighbor _ surviving fibers @xcite .
the simplest geometry is one - dimensional , so that the @xmath0 fibers are ordered linearly , with or without periodic boundary conditions . in this case
precisely _ two _ fibers , one on each side , take up , and divide equally , the extra stress , see fig .
[ ah - fig4a-0 ] .
when the strength thresholds take only two values , the bundle strength distribution has been found analytically @xcite . at a total force @xmath15 on the bundle the force on a fiber surrounded by @xmath693 previously failed fibers on the left - hand side , and @xmath694 on the right - hand side , is then @xmath695 here @xmath696 is one - half the force - per - fiber , is a convenient variable to use as the driving _ force parameter . _ is the control parameter.,width=151 ] @xcite and @xcite studied numerically the burst distribution in the local load - sharing model . in fig .
[ ah - fig4a-1 ] we exhibit simulation results similar to those first appearing in @xcite using threshold strengths randomly distributed on the unit interval .
again a power - law distribution seems to appear .
however , the burst exponent @xmath523 seems much larger than in the global lad - sharing model , eq .
[ eq10 ] , @xmath697 thus the relative frequency of long ( non - fatal ) bursts is considerably reduced .
it was concluded from the numerics that systems with local load - sharing are _ not _ in the universality class of fiber bundles with global load redistribution .
@xcite set out to analytically calculate the burst distribution in the local load - sharing model , finding the surprising result that there is no power law distribution of bursts at all ; it is exponential .
fibers ( crosses ) , and calculated from eq . ( [ 850 ] ) ( boxes ) .
the straight line shows the power law @xmath698 and the broken curve the function @xmath699 with @xmath700 .
note the small value of @xmath701.,width=226 ] @xcite did not find the burst distribution for general threshold distribution .
rather , they limited their study to the uniform threshold distribution in the force parameter @xmath421 , given by @xmath702 bursts in the local and the global models have different characters . in the local model a burst develops with one failure acting as the seed . if many neighboring fibers have failed , the load on the fibers on each side is high , and if they burst the load on the new neighbors will be even higher , etc . in this way
a weak region in the bundle may be responsible for the failure of the whole bundle . for a large number @xmath0 of fibers the probability of a weak region
somewhere is higher .
this hints in a qualitative way that the maximum load the bundle are able to carry does not increase proportional to @xmath0 , but slower than linear .
the result of a calculation based on combinatorics @xcite was the burst distribution @xmath703\;df.\nonumber\\\label{850}\end{aligned}\ ] ] where @xmath704 is the probability that a selected region of @xmath705 consecutive fibers have all failed whereas the two fibers at both boundaries are still intact at force parameter @xmath487 , @xmath706 is the probability that a force increase from @xmath421 to @xmath707 leads to a burst of length @xmath705 and magnitude @xmath454 .
@xmath708 is the probability at force parameter @xmath421 that among the @xmath352 first fibers there is no fatal burst , and that the last @xmath709 fibers of these have all failed .
the load distribution rule ( eq . [ hp ] ) implies that burst of size @xmath362 does necessarily lead to a complete breakdown of the whole bundle if the external force is too high , i.e. , if @xmath2 exceeds a critical value @xmath710 . since here
a fiber can at most take a load of unity , we have @xmath711 let us now attempt to find an simple estimate for the maximal force per fiber that the fiber bundle can tolerate . in order to do
that we assume that the fatal burst occurs in a region where no fibers have previously failed so that the burst has the same magnitude and length . we know that a single burst of length @xmath712 is fatal , eq .
( [ max ] ) , so our criterion is simply @xmath713 if we take into account that the two fibers adjacent to the burst should hold , and ignore the rest of the bundle , the gap distribution would be @xmath714 ^ 2p(\delta , \delta;f)\;df\nonumber\\ & = \frac{2p(\delta,\delta)}{\delta(\delta+1)(\delta+2)^{\delta+1}}.\nonumber\\\end{aligned}\ ] ] with the abbreviation @xmath715 we have @xmath716 using stirling s formula . taking logarithms we have @xmath717\nonumber\\ & \simeq - ( \delta + 2)\;,\nonumber\\ \label{950}\end{aligned}\ ] ] using that @xmath718 for @xmath719 when @xmath362 is large .
the failure criterion ( eq . [ fatal ] ) then takes the form @xmath720 since @xmath721 we have the following estimate for the maximum force @xmath15 that the fiber bundle can tolerate before complete failure : @xmath722 due to the assumption that the fatal burst occurs in a region with no previously failed fibers , the numerical prefactor is an overestimate .
the size dependence @xmath723 shows that the maximum load the fiber bundle can carry does not increase proportionally to the number of fibers , but slower .
this is to be expected since the probability of finding somewhere a stretch of weak fibers that start a fatal burst increases when the number of fibers increases .
the @xmath724 dependence agrees with a previous estimate by @xcite for a uniform threshold distribution .
the bimodal distribution used in @xcite also shows this behavior . the burst distribution eq .
( [ 950 ] ) is exponential .
the probability of a single burst zipping through the fiber bundle grows with the system size @xmath0 .
this contrasts strongly with the global load sharing model whose strength ( maximum force it can sustain ) grows linearly with @xmath0 , and whose burst distribution follows a universal power law . if the latter behavior is reminiscent of a second order transition with a critical point , the local load sharing model behaves more as if moving towards a first order phase transition .
let us discuss briefly the effect of a low cutoff ( in fiber strength distribution ) on the failure properties of lls model .
as in case of els model ( see section iii.a ) , we consider uniform fiber threshold distribution having a low cutoff @xmath326 ( eq . [ nonlinear - dist ] ) .
we present a probabilistic argument to determine the upper limit of @xmath326 , beyond which the whole bundle fails at once .
following the weakest fiber breaking approach the first fiber fails at an applied stress @xmath326 ( for large @xmath0 ) . as we are using periodic boundary conditions ,
the @xmath725 nearest neighbors ( @xmath725 is the coordination number ) bear the terminal stress of the failing fiber and their stress value rises to @xmath726 .
now , the number of nearest neighbors ( intact ) having strength threshold below @xmath343 is @xmath727 ( see eq .
[ ad-1 ] ) . putting the value of @xmath728 and @xmath343
we finally get @xmath729 if @xmath730 , then at least another fiber fails and this is likely to trigger a cascade of failure events resulting complete collapse of the bundle .
therefore , to avoid the ` instant failure ' situation we must have @xmath731 , from which we get the upper bound of @xmath326 : @xmath732 as the above condition does not depend on the coordination number @xmath352 , at any dimension the whole bundle is likely to collapse at once for @xmath348 .
it should be mentioned that lls model should behave almost like els model at the limit of infinite dimensions and therefore the identical bound ( of @xmath326 ) in both the cases is not surprising .
a numerical study @xcite confirms ( fig . [
fig : lls - step ] ) the above analytic argument in one dimension .
when average step value goes below @xmath733 , one step failure is the dominating mode then .
one can find the extreme limit of @xmath326 when all the nearest neighbors fail after the weakest fiber breaks .
then the lls bundle collapses instantly for sure . setting @xmath734
one gets the condition @xmath735 , where stress level of all the nearest neighbors crosses the upper cutoff 1 of the strength distribution .
clearly such failure is very rapid ( like a chain reaction ) and does not depend on the shape of the strength distributions , except for the upper cutoff . also as @xmath725 increases ( els limit ) , @xmath326 for instant failure assumes the trivial value 1 .
similar sudden failure in fbm has been discussed by @xcite in the context of a ` one sided load transfer ' model . in lls model
: for @xmath736 the average step values go below @xmath733 , i.e. , the bundle fails at one step in most of the realizations . , width=226,height=188 ] the local load sharing ( lls ) scheme introduces stress enhancement around the failed fiber , which accelerate damage evolution .
therefore , a few isolated cracks can drive the system toward complete failure through growth and coalescence .
the lls model shows zero strength ( for fiber threshold distributions starting from zero value ) at the limit @xmath361 , following a logarithmic dependence on the system size ( @xmath0 ) @xcite .
now for threshold distributions having a low cutoff ( @xmath326 ) , the ultimate strength of the bundle can not be less than @xmath326 .
for such a uniform distribution ( eq .
[ nonlinear - dist ] ) , numerical simulations show ( fig .
[ fig : lls - strength ] ) that as @xmath326 increases the quantity ( strength-@xmath326 ) approaches zero following straight lines with @xmath286 , but the slope gradually decreases which suggests that the system size dependence of the strength gradually becomes weaker .
is plotted against @xmath286 for different @xmath326 values : @xmath737 ( square ) , @xmath738 ( circle ) , @xmath739 ( up triangle ) , @xmath740 ( down triangle ) , @xmath741 ( star ) .
all the straight lines approach @xmath185 value as @xmath361 . , width=226,height=188 ] in @xcite a model interpolating between the global load sharing fiber bundle and a variant of a local load sharing model was introduced and studied . @xcite and @xcite followed up this work . in the model studied by @xcite a fraction @xmath742 of the load
a failing fiber carries would be distributed among its surviving neighbors and a fraction @xmath743 among all surviving fibers .
hence , for @xmath744 the model would be purely local load sharing , whereas for @xmath745 it would be purely global load sharing .
we show in fig .
[ ah - fig24 ] space time diagrams of the one - dimensional version of the model for different values of @xmath742 . for increasing values of @xmath742 , there is increasing localization .
is increased , there is increasing localization of the failing fibers.,width=340 ] both @xcite and @xcite found a phase transition when interpolating between the global load sharing model and the local load sharing model discussed earlier in this section . in the one - dimensional model shown in fig .
[ ah - fig24 ] , the critical value of @xmath742 , is @xmath746 for a flat threshold distribution @xcite .
a crucial mechanism in brittle fracture is the stress enhancement that occurs at crack tips .
the stress field has a @xmath747 singularity , where @xmath748 is the distance to the crack tip , in this region .
it is the interplay between fracture growth due to this singularity and due to weak spots in the material that drive the development of the fracture process @xcite .
clearly , there is a cutoff in the stress field as @xmath749 .
this may be caused by non - linearities in the material constitutive relations or by microstructure in the material such as the presence of crystallites .
@xcite has introduced a fiber bundle that contains a power - law dependence on the distance from a failing fiber on the force redistribution in order to model the stress singularity seen around crack tips .
the fiber bundle is implemented as a regular two - dimensional grid of parallel fibers clamped between two stiff blocks .
assuming that fiber @xmath750 has just failed , a force transfer function @xmath751 where @xmath752 and @xmath574 is the set of intact fibers , redistributes the forces .
@xmath753 is treated as a parameter on the unit interval .
there are two limiting cases , @xmath754 which recovers the global load - sharing fiber bundle and @xmath755 which recovers the local load - sharing model with nearest - neighbor stress alleviation ( section [ sec:4a ] ) .
the load increase on fiber @xmath435 is hence given by @xmath756 where @xmath757 is the set of failed fibers up to that point .
this model is too complex for analytical treatment and numerical simulations must be invoked . around @xmath758
there is a transition in behavior between essentially global load sharing and local load sharing as described in section [ sec:4a ] : for @xmath759 , the maximum sustainable force scales the number of fibers in the bundle at the outset , @xmath0 , whereas for @xmath760 , a @xmath724-behavior is observed as in eq .
( [ lffatal ] ) .
this is seen in figs .
[ ah - fig18 ] and [ ah - fig19 ] .
the burst distribution shows a power - law distribution with exponent @xmath410 again signaling global load sharing behavior for smaller values of @xmath753 .
as @xmath753 is increased , deviations from this behavior is seen .
this must be interpreted as a crossover towards local load sharing behavior as described in the previous section , see fig .
[ ah - fig20 ] . in the variable range fiber bundle of @xcite .
the data points where @xmath761 can be fitted to a power law with exponent @xmath410 . for larger @xmath753 ,
the slope becomes steeper and resembling the behavior seen in the local load - sharing model , see fig .
[ ah - fig4a-1].,width=302 ]
lastly , the structure of the clusters of failed fibers at breakdown is studied . in the global load sharing model
implemented in two dimensions does not yield anything particular .
there is a percolation transition in the cluster size distribution when the relative density of failed fibers reaches the percolation threshold , but this has no particular significance in the evolution of the model .
@xcite find the cluster distribution having two distinct behaviors , depending on whether @xmath753 is smaller than or larger than 2 .
there is no clear power law behavior , see fig.[ah - fig21 ] .
@xcite have studied an anisotropic version of this model .
the force transfer function ( [ forcetransferhidalgo ] ) is in this work generalized to @xmath762 where @xmath763 is an anisotropy parameter .
the behavior of this model turns out to be quite similar to that found in the isotropic model .
@xcite introduced a low cutoff in the threshold distribution @xcite in the variable range fiber bundle model of @xcite .
they studied the burst distribution as a function of @xmath753 the burst distribution as a function of @xmath753 and cutoff in the failure thresholds in terms of deformation , @xmath764 .
[ ah - fig22 ] summarizes their findings : for the explored values of @xmath753 in the range @xmath765 , a crossover from burst exponent @xmath410 to @xmath544 is seen for small @xmath764 , whereas for larger @xmath766 , the burst distribution may be fitted to a value @xmath767 for small @xmath764 . and @xmath764 values : a ) @xmath758 , b ) @xmath768 , c ) @xmath769 and d ) @xmath770 . from @xcite.,width=302 ]
@xcite and @xcite introduced a fiber bundle model where the fibers are hierarchically organized .
the fibers are paired two by two .
each of these pairs is seen as an order one " fiber .
this pairing is repeated for the order one fibers , creating order two fibers and so on . within each sub - bundle ,
the fibers are subject to equal load sharing .
if the cumulative threshold distribution for the fibers at order zero is @xmath771 , then the threshold distribution at level one is @xmath772\;,\ ] ] a result which is readily generalized to any level . even though the starting point here is global load sharing , the approximation introduced by treating the fibers at each level as fibers with given thresholds , leads to the introduction of spatial load dependence in the model . to our knowledge
, bursts have not been studied within this framework . in this section
we generalize the fiber bundle problem to include more realistically the elastic response of the surfaces to which the fibers are attached .
so far , these have been assumed to be infinitely stiff for the equal load sharing model , or their response has been modeled as very soft , but in a fairly unrealistic way in the local load sharing models , see section [ sec:4a ] .
we will end up with a description that is somewhat related to the models of the previous section , in particular the model of @xcite . in @xcite , a realistic model for the elastic response of the clamps
was studied .
the model was presented in the context of the failure of weldings . in this language , the two clamps were seen as elastic media glued together at a common interface . without loss of generality , one of the media may be assumed to be infinitely stiff whereas the other is soft . when a force is applied to a given fiber , the soft clamp responds by a deformation falling off inversely to the distance from the loaded fiber .
hence , the problem becomes one of solving the response of the surface with respect to a given loading of the fibers .
fibers exceeding their maximally sustainable load fail , and the forces and deformations must be recalculated .
the two clamps can be pulled apart by controlling ( fixing ) either the applied force or the _ displacement . _
the displacement is defined as the change in the distance between two points , one in each clamp positioned far from the interface .
the line connecting these points is perpendicular to the average position of the interface . in our case ,
the pulling is accomplished by controlling the displacement . as the displacement is increased slowly
, fibers will fail , eventually ripping the two surfaces apart .
we now concretize these ideas in a model .
it consists of a two - dimensional square @xmath773 lattices with periodic boundary conditions .
the lower one represents the hard , stiff surface and the upper one the elastic surface .
the nodes of the two lattices are matched , ( i.e. , there is no relative lateral displacement ) .
the fibers are modelled as in the previous sections : elastic up to a threshold value which has been individually chosen for each fiber from some threshold distribution .
the spacing between the fibers is @xmath454 in both the @xmath2 and @xmath487 directions .
the force that each fiber is carrying is transferred over an area of size @xmath774 to the soft clamp : as the two clamps are separated by controlling the displacement of the hard clamp relative to the zero level , @xmath775 , the forces carried by the fibers increases from zero .
when the force carried by a fiber reaches its breaking threshold , it breaks irreversibly and the forces redistribute themselves through the deformation of the soft clamp .
hence , the fibers are broken one by one until the two clamps are no longer in mechanical contact .
the force , @xmath776 , carried by the @xmath435th fiber is given by @xmath777 where @xmath359 is the spring constant and @xmath778 is the deformation of the elastic clamp at site @xmath435 .
all unbroken fibers have @xmath779 while a broken fiber has @xmath780 .
the quantity @xmath781 is , therefore , the length , and since @xmath779 , also the force carried by fiber @xmath435 .
the deformation of the soft clamp is described by the coupled system of equations , @xmath782 where the elastic green function , @xmath783 is given by @xcite @xmath784 in this equation , @xmath659 is the poisson ratio , @xmath785 the elastic constant , and @xmath786 the distance between sites @xmath435 and @xmath750 .
the indices @xmath435 and @xmath750 run over all @xmath787 sites .
the integration over the area @xmath774 is done to average the force from the fibers over this area . as remarked by @xcite , the green function ( eq . [ m3 ] )
applies for a medium occupying the infinite half space .
however , with a judicious choice of elastic constants , it may be used for a finite medium if its range is small compared to @xmath709 , the size of the system . by combining eq .
( [ m1 ] ) and eq .
( [ m2 ] ) , one obtains @xmath788 where matrix - vector notation is used .
@xmath789 is the @xmath790 identity matrix , and @xmath791 is the green function represented as an @xmath792 dense matrix .
the constant vector @xmath793 is @xmath787 dimensional .
the _ diagonal _ matrix @xmath794 is also @xmath792 .
its matrix elements are either 1 , for unbroken fibers , or 0 for broken ones . once eq .
( [ m4 ] ) is solved for the force , @xmath795 , eq . ( [ m2 ] )
yields the deformations of the elastic clamp .
( [ m4 ] ) is of the familiar form @xmath796 . since the green function connects all nodes to all other nodes , the @xmath792 matrix @xmath797 is dense which puts severe limits on the size of the system that may be studied .
the simulation proceeds as follows : one starts with all springs present , each with its stochastic breakdown threshold .
the two media are then pulled apart , the forces calculated using the conjugate gradient algorithm ( cg ) @xcite , and the fiber which is the nearest to its threshold is broken , i.e. , the matrix element corresponding to it in the matrix @xmath794 is zeroed .
then the new forces are calculated , a new fiber broken and so on until all fibers have failed . however , there are two problems that render the simulation of large systems extremely difficult from a numerical point of view .
the first is that since @xmath791 is @xmath792 _ dense _ matrix , the number of operations per cg iteration scales like @xmath798 .
even more serious is the fact that as the system evolves and springs are broken , the matrix @xmath799 becomes ill - conditioned . to overcome the problematic @xmath798 scaling of the algorithm , the matrix - vector multiplications are done in fourier space since the green function is diagonal in this space
symbolically , these multiplications may be written as follows , @xmath800 where @xmath801 is the fft ( fast fourier transform ) operator and @xmath802 its inverse ( @xmath803 ) .
since @xmath789 and @xmath794 are diagonal , operations involving them are performed in real space . with this formulation ,
the number of operations / iteration in the cg algorithm now scales like @xmath804 rather than @xmath798 . to overcome the ill - conditioning of the matrix
@xmath799 we need to precondition the matrix @xcite .
this means that instead of solving eq .
( [ m5 ] ) , one solves the equivalent problem @xmath805 where we simply have multiplied both sides by the arbitrary , positive definite preconditioning matrix @xmath806 .
clearly , the ideal choice is @xmath807 which would always solve the problem in one iteration .
since this is not possible in general , we look for a form for @xmath806 which satisfies the following two conditions : ( 1 ) as close as possible to @xmath808 , and ( 2 ) fast to calculate .
the choice of a good @xmath806 is further complicated by the fact that as the system evolves and fibers are broken , corresponding matrix elements of @xmath794 are set to zero .
so , the matrix @xmath809 evolves from the initial form @xmath810 to the final one @xmath789 .
@xcite did not find a fixed @xmath806 that worked throughout the entire breakdown process .
they therefore chose the form @xmath811 which is the taylor series expansion of @xmath812 .
for best performance , the number of terms kept in the expansion is left as a parameter since it depends on the physical parameters of the system .
it is important to emphasize the following points .
( a ) as fibers are broken , the preconditioning matrix evolves with the ill - conditioned matrix and , therefore , remains a good approximation of its inverse throughout the breaking process .
( b ) all matrix multiplications involving @xmath791 are done using ffts .
( c ) the calculation of @xmath806 can be easily organized so that it scales like @xmath813 where @xmath352 is the number of terms kept in the taylor expansion , equation ( [ m7 ] ) .
the result is a stable accelerated algorithm which scales essentially as the volume of the system .
[ ah - fig5 ] shows the force - displacement curve for a system of size @xmath814 and elastic constant @xmath815 . whether we control the applied force , @xmath15 , or the displacement , @xmath775 ,
the system will eventually suffer catastrophic collapse . however , this is not so when @xmath816 as shown in fig .
[ ah - fig6 ] . in this case , only controlling the force will lead to catastrophic failure . in the limit when @xmath817 , the model becomes the equal load - sharing fiber bundle model , where @xmath818 . in this limit
there are no spatial correlations and the force instability is due to the decreasing total elastic constant of the system making the force on each surviving bond increase faster than the typical spread of threshold values .
no such effect exists when controlling displacement @xmath775
. however , when the elastic constant , @xmath785 , is small , spatial correlations in the form of localization , where fibers that are close in space have a tendency to fail consecutively , do develop , and these are responsible for the displacement instability which is seen in fig .
[ ah - fig5 ] .
systems with @xmath815.,width=226 ] systems with @xmath816.,width=226 ] we now turn to the study of the burst distribution . figs .
[ ah - fig7 ] and [ ah - fig8 ] show the burst distribution for @xmath815 and 100 . in both cases
we find that the burst distribution follows a power law with an exponent @xmath819 .
it was argued in ref .
@xcite that the value of @xmath523 in this case is indeed @xmath597 as in the global load sharing model .
these two figures should be compared with fig .
[ ah - fig20 ] showing the burst distribution in the variable range fiber bundle model of hidalgo et al . , @xcite , where @xmath410 is recovered as long as the range exponent @xmath753 is small , rendering the forces long range among the fibers .
, @xmath815 .
the slope of the straight line is @xmath820.,width=226 ] , @xmath816 .
the slope of the straight line is @xmath820.,width=226 ] as the failure process proceeds , there is an increasing competition between local failure due to stress enhancement and local failure due to local weakness of material .
when the displacement , @xmath775 , is the control parameter and @xmath785 is sufficiently small ( for example @xmath815 ) , catastrophic failure eventually occurs due to localization .
the onset of this localization , i.e. , the catastrophic regime , occurs when the two mechanisms are equally important .
this may be due to self - organized criticality @xcite occurring at this point . in order to test whether this is the case , @xcite measured the size distribution of broken bond clusters at the point when @xmath775 reaches its maximum point on the @xmath821 characteristics , _
i.e. _ the onset of localization and catastrophic failure .
the analysis was performed using a hoshen - kopelman algorithm @xcite .
the result is shown the in fig .
[ ah - fig10 ] , for 56 disorder realizations , @xmath822 and @xmath815 .
the result is consistent with a power law distribution with exponent @xmath823 , and consequently with self organization .
if this process were in the universality class of percolation , the exponent would have been 2.05 .
hence , we are dealing with a new universality class in this system .
this behavior should be contrasted to the one seen in the variable range fiber bundle model studied by @xcite where no power law distribution was found ( see fig .
[ ah - fig21 ] in section [ sec:4b ] ) . and elastic constant @xmath815 .
the straight line is a least square fit and indicates a power law with exponent @xmath823.,width=226 ]
the aim of this section is to demonstrate how the fiber bundle model may be used as a tool for studying both important phenomena occuring in materials such as fatigue and for studying important classes of materials , notably fiber reinforced composites .
a general review has recently been written by @xcite on this subject .
we also discuss applications of fiber bundle models in other contexts , ranging from traffic modeling to earthquake dynamics .
materials may undergo time dependent deformation under steady load .
sometimes when a load is applied , though surviving at first stage , the system fails after long time .
this type of failure is referred as the fatigue - failure or creep rupture @xcite .
the fatigue - failure is basically a thermally activated process @xcite and originates at the atomic level of the fibers where the molecules are in random thermal vibrations .
eventually a molecule acquires sufficient thermal energy to overcome the local energy barrier and slips relative to other molecules .
the frequency of such events is greatly enhanced by increases in temperature , stress and impurity level . after a molecular slip or rupture ,
neighboring molecules become overloaded and the failure rate increases .
these molecular failures accumulate locally and produce microcracks within the material .
also micro - cracks can grow at the crack - tips with time due to chemical diffusion in the atmosphere @xcite which helps the growth of fractures .
these failures nucleate around the defects in the solid and the failure behavior and its statistics therefore crucially depends on the disorder or impurity distribution within the sample .
the system then fails under a stress less than its normal strength ( @xmath181 ) and the failure time ( @xmath300 ) depends on both , the applied load and the impurity level .
fatigue - failure in fiber bundle model was first studied by @xcite considering different classes of fibers and several breakdown rules .
the probabilistic analysis gives the lifetime distribution under various loading conditions : constant load , loads proportional to time and periodic loads .
the ` time dependent fatigue ' and ` cycle dependent fatigue ' both are addressed in this work introducing the concept of ` memory ' effect , i.e. , the load history can affect the failure of fibers .
such time dependent failure in fiber bundles have been considerably extended and generalized by phoenix et al .
@xcite for equal load sharing ( els ) and local load sharing ( lls ) bundles .
the approximate fatigue life time distributions have been achieved through probabilistic analysis introducing power law breakdown rule and exponential breakdown rule at the molecular level @xcite .
when dry fibers are replced by viscoelastic elements having time dependent deformation properties , fiber bundle model exhibits creep behavior @xcite in terms of the macroscopic response under constant external load .
there exists a critical load ( or stress ) below which the deformation attains a constant value ( infinite lifetime ) and above the critical load , deformation increases monotonically resulting global failure ( finite lifetime ) .
another extension of classical fiber bundle model , the continuous damage model @xcite , captures similar creep behavior assuming that a fiber can fail more than once and at each failure its stiffness gets reduced by a constant factor . in both of these models
the lifetime of the bundle diverges at the critical load following robust power law variation with the applied load . also , few experiments @xcite have been performed to observe the failure time of materials and its statistics .
the effect of thermal activation and disordered noise on the failure have recently been measured for material breakdown and approximate fatigue behavior has been obtained @xcite using fiber bundle models . in this section ,
we discuss several approaches of achieving fatigue - failure behavior in equal load sharing fiber bundle models .
the approaches differ basically in the way how time - dependence has been incorporated in the failure process .
the influence of noise on macroscopic failure in fiber bundle model , has been studied numerically @xcite , using both disorder noise and thermal noise .
the strength of each fiber is characterized by a _
critical stress _
@xmath824 which is a random variable that follows a normal distribution of mean @xmath825 and variance @xmath826 : @xmath827 where @xmath828 is the boltzmann constant and @xmath829 is the disorder noise .
again each fiber is subjected to an additive time dependent random stress @xmath830 which follows a zero mean normal distribution of variance @xmath831 : @xmath832 here @xmath833 is the thermal noise . of an homogeneous bundle ( @xmath834 ) in a creep test . ( a ) @xmath835 as a function of the normalized force @xmath836 for several values of thermal noise variance @xmath831 : ( cross ) @xmath837 ; ( circle ) @xmath838 ; ( triangle )
( b ) @xmath835 as a function of @xmath840 for several values of @xmath494 : ( box ) @xmath841 ; ( circle ) @xmath842 ; ( triangle ) @xmath843 .
continuous lines in ( a ) and ( b ) are the fits with eq .
( [ thermal_eq4 ] ) and eq .
( [ thermal_eq5 ] ) , respectively . adapted from @xcite . , title="fig:",width=226,height=226 ] of an homogeneous bundle ( @xmath834 ) in a creep test . ( a )
@xmath835 as a function of the normalized force @xmath836 for several values of thermal noise variance @xmath831 : ( cross ) @xmath837 ; ( circle ) @xmath838 ; ( triangle ) @xmath839 .
( b ) @xmath835 as a function of @xmath840 for several values of @xmath494 : ( box ) @xmath841 ; ( circle ) @xmath842 ; ( triangle ) @xmath843 .
continuous lines in ( a ) and ( b ) are the fits with eq .
( [ thermal_eq4 ] ) and eq .
( [ thermal_eq5 ] ) , respectively . adapted from @xcite .
, title="fig:",width=226,height=226 ] due to the equal load sharing scheme , if a number @xmath844 of fibers are broken at time @xmath18 after force @xmath15 is applied on the bundle , the local force on each of the remaining fibers will be : @xmath845 where @xmath0 is the total number of fibers in the intact bundle and @xmath846 is the initial force per fiber .
if @xmath847 , the model reduces to the static one . in that case the applied force is increased linearly from zero to the critical value @xmath848 above which the whole bundle breaks .
therefore , at a constant force @xmath15 , the bundle breaks in a single avalanche only if @xmath849 , otherwise it will never break .
if @xmath850 then the system can break at an applied force @xmath851 due to the thermal effect .
such thermal failure of the model has been studied numerically as a function of @xmath494 , @xmath831 and @xmath826 @xcite . of an heterogeneous bundle ( @xmath852 ) in creep test .
the life time @xmath835 is plotted as a function of @xmath840 at @xmath841 .
different symbols correspond to different values of @xmath853 : ( circle ) @xmath834 ; ( triangle ) @xmath854 ; ( box ) @xmath855
. adapted from @xcite.,width=226,height=226 ] when @xmath856 , the failure time @xmath857 as a function of @xmath836 follows an exponential law for any fixed value of @xmath826 : @xmath858 , \label{thermal_eq4}\ ] ] where @xmath763 is a fitting parameter [ fig .
[ thermal1](a ) ] , which is a function of @xmath831 . at constant stress ,
the failure time depends on thermal noise @xmath831 as : @xmath859 where a is a function of @xmath2 [ fig . [ thermal1](b ) ] .
similar result has also been observed in case of a heterogeneous fiber bundle [ fig .
[ thermal2 ] ] .
one can compare these results with pomeau s theory @xcite for the failure time of solids : @xmath860 where @xmath861 is the imposed stress , @xmath862 is a constant , @xmath863 the surface energy , @xmath864 the young modulus , @xmath865 is a constant which depends on the geometry , @xmath866 is an effective temperature and @xmath867 the dimensionality of the system .
this theory is based on the physical argument that thermal activation of micro cracks @xcite is responsible for the macroscopic failure of the material .
it can be noted that the functional dependence of @xmath835 on stress is different for fiber bundle model and for solids .
the main reason of which is the different geometry of the stress distribution in the fiber bundle and the solids .
these numerical studies suggest that disorder noise amplifies the effect of thermal noise and reduces the dependence of @xmath835 on the temperature and this can explain the recent experimental observations on microcrystals @xcite , gels @xcite and macroscopic composite materials @xcite .
the numerical observations described above have been confirmed later through an analytic investigation by @xcite . in a homogeneous ( no disorder in fiber strengths )
fiber bundle model , force or stress on each fiber is @xmath868 where @xmath869 and @xmath870 is the random noise with a gaussian distribution @xmath871 with zero mean and variance @xmath831 . now
the probability that one fiber survives after @xmath18 time step is @xmath872^t , \ ] ] where @xmath161 is the cumulative probability .
then the probability that the entire bundle can survive after @xmath18 time step is @xmath873^{nt } .\ ] ] therefore the average failure time is @xmath874 } .\ ] ] after the first fiber breaks , the situation remains the same with a smaller bundle and larger stress .
thus the average failure time after @xmath875 broken fiber is @xmath876\ } } } .\ ] ] now the total failure time can be obtained by taking the sum over all @xmath435 as @xmath877\ } } } .\ ] ] when @xmath0 is large , one can replace the sum by a continuous integral : @xmath878\}}}dy\\ & = & \int_{x_0}^\infty \frac { -1}{\ln{[1-p(1-z ) ] } } \frac { dz}{z}.\end{aligned}\ ] ] to achieve a closed form equation it has been considered that the above sum is dominated by the time required for breaking the first fiber when @xmath494 is much smaller than the maximum load @xmath388 and when @xmath879 .
then @xmath161 can be considered to be much smaller than @xmath184 . now
the derivative of the average time with respect to @xmath494 gives @xmath880 } } \approx \frac { 1}{x_0 p(1-x_0 ) } .\ ] ] when @xmath881 the error function can be expanded as @xmath882 .\ ] ] finally taking into account the dominating terms , one gets @xmath883 these analytic expressions are identical to what have been observed earlier @xcite in numerical simulations .
a similar analysis @xcite shows that when fiber strengths are distributed ( heterogeneous case ) , the average first failure time can be expressed as @xmath884 where @xmath144 is an effective temperature that is added to the temperature @xmath885 due to the disorder .
hence , the disorder leads to an effective temperature @xmath886 , and this is what @xcite proposed : time independent heterogeneities of the system modify the effective temperature .
@xcite estimate the total time to failure for the thermally activated fiber bundle model and find the same behavior as in eq .
( [ roux_disorder ] ) : the disorder adds a constant to the temperature of the fiber bundle . in @xcite ,
these results are generalized to the two - dimensional fuse model . not only the temperature , but several other factors can result in fatigue - failure in materials : weather effects , chemical effects etc . @xcite .
recently , there has been an attempt @xcite to incorporate all the noise effects through a single parameter @xmath831 ( @xmath828 is boltzmann constant ) which can directly influence the failure probability of the individual elements . such a failure probability @xmath887 ) at any applied
stress @xmath48 , induced by a non - zero noise @xmath831 , has been formulated as : @xmath888 , & 0\leq\sigma\leq x\\ 1 , & \sigma > x\end{array}\right\ } , \label{fat - prob}\ ] ] where @xmath2 is the failure strength of an element .
clearly , the failure probability increases as @xmath48 and @xmath831 increases . without any noise ( @xmath847 )
the model is trivial : the bundle remains intact for stress @xmath257 and it fails completely for @xmath889 where @xmath90 is the critical stress value . against ( a ) external stress @xmath48 and ( b ) against noise @xmath885 ( @xmath890 here ) , for a homogeneous bundle containing @xmath891 fibers .
the insets show the variation of the fraction @xmath94 of unbroken fibers with time @xmath18 for different @xmath885 values [ 1.2 ( cross ) and 1.0 ( plus ) ] in ( a ) and different @xmath48 values [ 0.15 ( cross ) and 0.12 ( plus ) ] in ( b ) .
the dotted and dashed lines represent the theoretical result eq .
( [ fat - time2 ] ) .
, width=226,height=188 ] against ( a ) external stress @xmath48 and ( b ) against noise @xmath885 ( @xmath890 here ) , for a homogeneous bundle containing @xmath891 fibers .
the insets show the variation of the fraction @xmath94 of unbroken fibers with time @xmath18 for different @xmath885 values [ 1.2 ( cross ) and 1.0 ( plus ) ] in ( a ) and different @xmath48 values [ 0.15 ( cross ) and 0.12 ( plus ) ] in ( b ) .
the dotted and dashed lines represent the theoretical result eq .
( [ fat - time2 ] ) . ,
width=226,height=188 ] at @xmath892 and under any stress @xmath48
( @xmath893 ) , some fibers ( weaker fibers ) fail and the total load has to be supported by the surviving fibers , which in turn enhances their stress value , inducing further failure .
the bundle therefore fails at @xmath257 after a finite time @xmath894 . in case of homogeneous bundle
( all fibers have the same strength @xmath2 ) , the critical stress value for the bundle is @xmath895 .
then the time dynamics at an applied stress @xmath48 can be written @xcite as @xmath896,\label{fat - recur}\ ] ] where @xmath213 is the fraction of total fibers remains intact after time step @xmath18 . in the continuum limit
, we can write the above recursion relation in a differential form @xmath897.\label{fat - diff}\ ] ] the solution gives the failure time @xmath898du.\label{fat - time1}\ ] ] hence , for @xmath257 @xmath899.\label{fat - time2}\ ] ] again , for @xmath889 , one gets @xmath900 from eq .
( [ fat - recur ] ) , giving @xmath901 . for small @xmath831 and as @xmath52 ,
@xmath902 $ ] .
these results agree qualitatively with the recent experimental observations @xcite . in order to investigate the fatigue behavior in heterogeneous fiber bundles ,
uniform distribution of fiber strengths has been considered @xcite .
the noise - induced failure probability has the similar form : @xmath903 $ ] for @xmath904 and @xmath905 for @xmath906 , where @xmath2 denotes the strength of the individual fibers in the bundle .
now it is difficult to tackle the problem analytically .
however , monte carlo simulations @xcite show ( fig .
[ fig : fat_hetero ] ) the variations of average failure time ( @xmath894 ) with noise ( @xmath831 ) and stress level ( @xmath48 ) : @xmath907,\label{fat - het - time } \label{eq : fati - nonhomo}\ ] ] where , @xmath90 is the critical stress .
this phenomenological form eq .
( [ fat - het - time ] ) is indeed very close to the analytic result eq .
( [ fat - time1 ] ) for the homogeneous fiber bundle .
@xcite considered the breaking dynamics in a fiber bundle where each fiber has a failure probability @xmath457 determined by the loading time @xmath18 and the load @xmath48 on it as @xmath908 $ ] , and analyzed the life time @xmath909 distribution . for @xmath910 , els and
lls models have identical gaussian distribution for @xmath911 . for @xmath912 ,
lls shows extreme statistics , while els gives gaussian behavior , see e.g. , @xcite .
@xcite also studied similar @xmath911 distribution and their scaling property with load and temperature variations .
.3 in creep behavior has been achieved @xcite in a bundle of viscoelastic fibers , where a fiber is modeled by a kelvin - voigt element ( see fig .
[ fig : kelvin ] ) and results in the constitutive stress - strain relation @xmath913 here @xmath914 is the applied stress , @xmath915 is the corresponding strain , @xmath916 denotes the damping coefficient , and @xmath864 is the young modulus of the fibers . in the equal load sharing mode
, the time evolution of the system under a steady external stress @xmath917 can be described by the equation @xmath918 as one can expect intuitively , there is a critical load @xmath90 for the system and eq .
( [ eq : eom ] ) suggests two distinct regimes depending on the value of the external load @xmath919 : when @xmath919 is below the critical value @xmath920 eq .
( [ eq : eom ] ) has a fixed - point solution @xmath921 , which can be obtained by setting @xmath922 in eq .
( [ eq : eom ] ) @xmath923.\label{eq : stationary}\end{aligned}\ ] ] in this case the strain value converges to @xmath921 when @xmath924 , and no macroscopic failure occurs .
but , when @xmath925 , no fixed - point solution exists . here , @xmath926 remains always positive , that means in this case , the strain of the system @xmath927 monotonically increases until the system fails globally at a finite time @xmath193 @xcite . the solution of the differential equation eq .
( [ eq : eom ] ) gives a complete description of the failure process . by separation of variables ,
the integral becomes @xmath928}+c,\label{eq : integ}\end{aligned}\ ] ] where @xmath178 is integration constant . below the critical point @xmath929 the bundle slowly relaxes to the fixed - point value @xmath921 .
the characteristic time scale of such relaxation process can be obtained by analyzing the behavior of @xmath927 in the vicinity of @xmath921 . after introducing a new variable @xmath930 as @xmath931
, the differential equation can be written as @xmath932\delta_0.}\label{eq : delta}\end{aligned}\ ] ] clearly , the solution of eq .
( [ eq : delta ] ) has the form @xmath933}$ ] , with @xmath934},\label{eq : tau}\end{aligned}\ ] ] where @xmath300 is the characteristic time of the relaxation process .
the variation of the relaxation time @xmath300 with the external driving near the critical point @xmath920 is crucial for any dynamical system .
since @xmath935 has a maximum of the value @xmath920 at @xmath936 , in the vicinity of @xmath936 one can use the approximation : @xmath937 where the multiplication factor @xmath524 depends on the cumulative distribution @xmath161 . using the approximation eq .
( [ eq : series ] ) , it can be shown from eq .
( [ eq : tau]),that @xmath938 therefore , the relaxation time diverges following a universal power law with an exponent @xmath939 .
note that a dry fiber bundle model , under constant load shows similar power law divergence ( see section iii.a ) . for uniform and weibull distributions with two different weibull index for the els case .
all the three curves are parallel to each other on a double logarithmic plot with an exponent close to @xmath741 , in agreement with the general result eq .
( [ critical ] ) . from @xcite .
, width=226,height=188 ] how does the system behave above the critical point ?
the behavior can be analyzed in the same way when @xmath919 is close to @xmath920 .
then one can write @xmath940 , where @xmath941 .
it is obvious that the relaxation steps are too many when @xmath927 becomes close to @xmath936 .
therefore , the integral in eq .
( [ eq : integ ] ) , is dominated by the region close to @xmath936 .
using eq .
( [ eq : series ] ) , the integral in eq .
( [ eq : integ ] ) becomes @xmath942 evaluating the integration over a small @xmath915 interval in the vicinity of @xmath936 , one gets @xmath943 thus , @xmath193 has a two - sided power law divergence at @xmath920 with a universal exponent @xmath944 independent of the specific form of the disorder distribution @xmath945 , similarly to @xmath300 in case of dry fiber bundle ( see section iii.a ) . to check the validity of the universal power law behavior of @xmath193 , simulations were performed @xcite with various disorder distributions , _
i.e. _ uniform distribution and the weibull distribution of the form @xmath946}$ ] , where @xmath947 is the characteristic strain and @xmath8 is the shape parameter . the simulation results ( fig .
[ fig : sigma ] ) are in excellent agreement with the analytic results .
a slow relaxation following fiber failure can also lead to creep behavior @xcite . in this case , the fibers are linearly elastic until they break , but after breaking they undergo a slow relaxation process .
therefore when a fiber breaks , its load does not drop to zero instantaneously .
instead it undergoes a slow relaxation process and thus introduces a time scale into the system .
as the intact fibers are assumed to be linearly elastic , the deformation rate is @xmath948 where @xmath2 denotes the stress , @xmath949 denotes the strain and @xmath864 is the young modulus of the fibers .
in addition , to capture the slow relaxation effect , the broken fibers with the surrounding matrix material are modeled by maxwell elements ( fig .
[ fig : maxwell ] ) , _ i.e. _ they are assumed as a serial coupling of a spring and a dashpot .
such arrangement results in the following non - linear response @xmath950 where @xmath951 is the time dependent stress and @xmath952 is the time dependent deformation of a broken fiber .
the relaxation of broken fibers is characterized by few parameters , @xmath953 and @xmath509 , where @xmath954 is the effective stiffness of a broken fiber , the exponent @xmath509 characterizes the strength of non - linearity and @xmath757 is a constant . in equal load sharing mode , when external stress @xmath914 is applied ,
the macroscopic elastic behavior of the composite can be represented by the constitutive equation @xcite @xmath955+x_{{\rm b}}(t)p(x(t)).\label{eq : macro}\end{aligned}\ ] ] here @xmath956 is the amount of stress carried by the broken fibers and @xmath957 and @xmath958 denote the fraction of broken and intact fibers at time @xmath18 , respectively . by construction ( fig .
[ fig : maxwell ] ) , the two time derivatives have to be always equal @xmath959 now the differential equation for the time evolution of the system can be obtained using eq .
( [ eq : macro ] ) , eq .
( [ eq : broken ] ) and eq .
( [ eq : condit ] ) as @xmath960\right\ } \nonumber\\ & & = b\left[\frac{\sigma_{{\rm o}}-x\left[1-p(x)\right]}{p(x)}\right]^{m}.\label{eq : eom_max } \end{aligned}\ ] ] similar to the viscoelastic model , two different regimes of @xmath585 can be distinguished depending on the value of @xmath917 : if the external load is below the critical value @xmath920 a fixed - point solution @xmath961 exists which can be obtained by setting @xmath962 in eq .
( [ eq : eom_max ] ) @xmath963.\label{eq : statio_max}\end{aligned}\ ] ] this means that the solution @xmath585 of eq .
( [ eq : eom_max ] ) converges asymptotically to @xmath961 resulting in an infinite lifetime @xmath193 of the system . when the external load is above the critical value , the deformation rate @xmath964 remains always positive , resulting in a macroscopic failure in a finite time @xmath193 .
now we focus on the universal behavior of the model in the vicinity of the critical point . below the critical point the relaxation of @xmath585 to the stationary solution @xmath961 can be presented by a differential equation of the form @xmath965 where @xmath966 denotes the difference @xmath967 .
hence , the characteristic time scale @xmath300 of the relaxation process only emerges if @xmath514 .
also , in this case relaxation time goes as @xmath968 when approaching the critical point from below .
however , for @xmath969 the situation is different : relaxation process is characterized by @xmath970 , where @xmath971 with @xmath972 .
again , close to the critical point , it can also be shown that the lifetime @xmath193 shows power law divergence when the external load approaches the critical point from above @xmath973 the exponent is universal in the sense that it does not depend on the disorder distribution .
however it depends on the exponent @xmath509 , which characterizes the non - linearity of broken fibers . as a function of the distance from the critical point , @xmath974 , for two different values of the parameter @xmath509 .
the number of fibers in the bundle is @xmath975 . from @xcite.,width=226,height=188 ] as a check , numerical simulations have been performed @xcite for several different values of the exponent @xmath509 ( fig .
[ fig : sigma_maxwell ] ) .
the slope of the fitted straight lines agrees well with the analytic predictions ( eq . [ eq : tau_max ] ) .
an interesting experimental and theoretical study of fatigue failure in asphalt was performed by @xcite .
the experimental set - up is shown in fig .
[ ah - fig13 ] .
the cylindrical sample was subjected to cyclic diametric compression at constant load amplitude @xmath914 , and the deformation @xmath222 as a function of the number of cycles @xmath976 was recorded together with the number of cycles @xmath977 at which catastrophic failure occurs .
[ ah - fig15 ] shows deformation @xmath222 as a function of number of cycles @xmath976 for two different load amplitudes @xmath978 and 0.4 . here
@xmath90 is the tensile strength of the asphalt .
fig.[ah - fig16 ] shows the number of cycles to catastrophic failure as a function of the load amplitude @xmath979 .
this curve shows three regimes , the middle one being characterizable by a power law , @xmath980 this is the _ basquin regime _ @xcite .
@xcite find @xmath981 for the asphalt system . as a function of number of cycles @xmath976 , showing both experimental data and theoretical curves based on the fiber bundle model . from @xcite.,width=302 ] in order to model the behavior found in figs .
[ ah - fig15 ] and [ ah - fig16 ] , @xcite introduce equal load - sharing fiber bundle model as illustrated in fig .
[ ah - fig13]a .
each fiber @xmath982 is subjected to a time dependent load @xmath983 .
there are two failure mechanisms present .
fiber @xmath435 fails instantaneously at time @xmath18 when @xmath983 for the first time reaches its failure threshold @xmath434 . however , there is also a damage accumulation mechanism characterized by the parameter @xmath984 . in the time interval
@xmath985 , fiber @xmath435 accumulates a damage @xmath986 where @xmath987 is a scale parameter and @xmath988 is an exponent to be determined .
hence , the accumulated damage is @xmath989 when @xmath984 for the first time exceeds the damage accumulation threshold , @xmath990 , fiber @xmath435 fails . the thresholds @xmath434 and @xmath990 are chosen from a joint probability distribution @xmath991 .
@xcite make the assumption that this distribution may be factorized , @xmath992 . as a function of load amplitude @xmath979 .
the two curves show experimental and fiber bundle model data respectively . from @xcite.,width=340 ] in addition to damage accumulation , there is yet another important mechanism that needs to be incorporated in the model : damage _ healing _
a time scale @xmath300 is associated with this mechanism , and the els average force - load equation @xmath993x(t)\;,\ ] ] where @xmath994 is the cumulative instantaneous breaking threshold probability , is generalized to @xmath995 \left[1-p_t(x(t))\right]x(t)\;,\ ] ] where @xmath996 .
@xcite show that the basquin law ( [ kun_basquin ] ) may be derived analytically from eq.([kun_general_break ] ) leading to @xmath997 .
the solid curves in fig .
[ ah - fig15 ] are fits of the theoretical curves based on eq.([kun_general_break ] ) to the experimental data for the @xmath222 vs.@xmath976 .
likewise , fig .
[ ah - fig16 ] shows the fit of the theoretical @xmath977 vs. @xmath979 to the experimental data . for this fit , @xmath758 and @xmath998 .
a fundamental question in strength considerations of materials is when does it fail .
are there signals that can warn of imminent failure ?
this is of uttermost importance in e.g. the diamond mining industry where sudden failure of the mine can be extremely costly in terms of lives .
these mines are under continuous acoustic surveillance , but at present there are no tell - tale acoustic signature of imminent catastrophic failure .
the same type of question is of course also central to earthquake prediction and initiates the search for precursors of global ( catastrophic ) failure events , see e.g. , @xcite , @xcite , @xcite .
the precursor parameters essentially reflect the growing correlations within a dynamic system as it approaches the failure point .
as we show , sometimes it is possible to predict the global failure points in advance .
needless to mention that the existence of any such precursors and detailed knowledge about their behavior for major catastrophic failures like earthquakes , landslides , mine / bridge collapses , would be of supreme value for our civilization . in this sub - section
we discuss some precursors of global failure in els models .
we also comment on how can one predict the critical point ( global failure point ) from the precursor parameters .
.1 in _ ( a ) divergence of susceptibility and relaxation time _ .1 in and @xmath999 with applied stress for a bundle having @xmath1000 fibers .
uniform distribution of fiber thresholds is considered and averages are taken over @xmath569 sample .
the dotted straight lines are the best linear fits near the critical point.,title="fig:",width=226,height=188 ] and @xmath999 with applied stress for a bundle having @xmath1000 fibers .
uniform distribution of fiber thresholds is considered and averages are taken over @xmath569 sample .
the dotted straight lines are the best linear fits near the critical point.,title="fig:",width=226,height=188 ] as we discussed earlier ( section iii.a ) in case of els fiber bundles the susceptibility ( @xmath49 ) and the relaxation time ( @xmath300 ) follow power laws ( exponent @xmath1001 ) with external stress and both diverge at the critical stress .
therefore if we plot @xmath1002 and @xmath999 with external stress , we expect a linear fit near critical point and the straight lines should touch @xmath1003 axis at the critical stress .
we indeed found similar behavior ( fig .
[ fig : xi - tau ] ) in simulation experiments after taking averages over many sample . for application ,
it is always important that such prediction can be done in a single sample . for a single bundle having very large number of fibers , similar response of @xmath49 and @xmath300
have been observed .
the estimation ( through extrapolation ) of the failure point is also quite satisfactory ( fig .
[ fig : xi - tau - single ] ) . and @xmath999 with applied stress for a single bundle having @xmath1004 fibers with uniform distribution of fiber thresholds .
straight lines represent the best linear fits near the critical point . , title="fig:",width=226,height=188 ] and @xmath999 with applied stress for a single bundle having @xmath1004 fibers with uniform distribution of fiber thresholds .
straight lines represent the best linear fits near the critical point .
, title="fig:",width=226,height=188 ] .1 in _ ( b ) pattern of breaking - rate _ .1 in different stress values .
circles are for stresses below the critical stress and triangles are for stresses above the critical stress .
the simulation has been performed for a single bundle with @xmath1004 fibers having uniform distribution of fiber thresholds .
the dotted straight line has a slope @xmath1005.,width=226,height=188 ] when we apply load on a material body , it is important to know whether the body can support that load or not .
the similar question can be asked in fiber bundle model .
we found that if we record the breaking - rate , i.e , the amount of failure in each load redistribution -then the pattern of breaking - rate clearly shows whether the bundle is going to fail or not . for any stress below the critical state , the breaking - rate follows exponential decay ( fig .
[ fig : breaking - rate ] ) with the step of load redistribution and for stress values above critical stress it is a power law followed by a gradual rise ( fig .
[ fig : breaking - rate ] ) . clearly , at critical stress it follows a robust power law with exponent value @xmath1005 that can be obtained analytically from eq .
( [ eq : fbm - omori ] ) . as we can see from fig .
[ fig : breaking - rate - min ] that when the applied stress value is above the critical stress , breaking - rate initially goes down with step number , then at some point it starts going up and continues till the complete breakdown . that means if breaking - rate changes from downward trend to upward trend - the bundle will fail surely
-but not immediately after the change occurs -it takes few more steps and number of these steps decreases as we apply bigger external stress ( above the critical value ) .
therefore , if we can locate this minimum in the breaking - rate pattern , we can save the system ( bundle ) from breaking down by withdrawing the applied load immediately .
we have another important question here : is there any relation between the breaking rate minimum and the failure time ( time to collapse ) of the bundle ?
there is indeed a universal relationship which has been explored recently @xcite through numerical and analytical studies : for slightly overloaded bundle we can rewrite eq .
( [ finaluni ] ) as @xmath1006 where @xmath1007 from eq .
( [ un ] ) follows the breaking rate @xmath1008 @xmath1009 has a minimum when @xmath1010 which corresponds to @xmath1011 when criticality is approached , i.e. when @xmath1012 , we have @xmath1013 , and thus @xmath1014 , as expected .
we see from eq .
( [ un ] ) that @xmath1015 for @xmath1016 this is an excellent approximation to the integer value at which the fiber bundle collapses completely .
thus with very good approximation we have the simple connection @xmath1017 .
when the breaking rate starts increasing we are halfway ( see fig .
[ fig : breaking - rate - min ] ) to complete collapse ! + vs. step @xmath18 ( upper plot ) and vs. the rescaled step variable @xmath1018 ( lower plot ) for the uniform threshold distribution for a bundle of @xmath1019 fibers .
different symbols are used for different excess stress levels @xmath1020 : 0.001 ( circles ) , 0.003 ( triangles ) , 0.005 ( squares ) and 0.007 ( crosses ) .
, title="fig:",width=226,height=226 ] vs. step @xmath18 ( upper plot ) and vs. the rescaled step variable @xmath1018 ( lower plot ) for the uniform threshold distribution for a bundle of @xmath1019 fibers .
different symbols are used for different excess stress levels @xmath1020 : 0.001 ( circles ) , 0.003 ( triangles ) , 0.005 ( squares ) and 0.007 ( crosses ) .
, title="fig:",width=226,height=226 ] .1 in _ ( c ) crossover signature in avalanche distribution _ .1 in in els model with uniform fiber strength distribution .
here bundle size @xmath1000 and averages are taken over @xmath1021 sample .
two power laws ( dotted lines ) have been drawn as reference lines to compare the numerical results .
, width=226,height=226 ] the bursts or avalanches can be recorded from outside -without disturbing the ongoing failure process .
therefore , any signature in burst statistics that can warn of imminent system failure would be very useful in the sense of wide scope of applicability . as discussed in section iii.b ,
when the avalanches are recorded close to the global failure point , the distribution shows ( fig .
[ fig : els - cut - dist ] ) a different power law ( @xmath544 ) than the one ( @xmath410 ) characterizing the size distribution of all avalanches .
this crossover behavior has been analyzed analytically in case of els fiber bundle model and similar crossover behavior is also seen @xcite in the burst distribution and energy distribution of the fuse model which is an established model for studying fracture and breakdown phenomena in disordered systems .
the crossover length becomes bigger and bigger as the failure point is approached and it diverges at the failure point ( eq . [ eq-3b-16 ] ) . in some sense ,
the magnitude of the crossover length tells us how far the system is from the global failure point .
most important is that this crossover signal does not hinge on observing rare events and is seen also in a single system ( see fig .
[ fig3b-3 ] ) . therefore , such crossover signature has a strong potential to be used as a useful detection tool .
it should be mentioned that a recent observation @xcite suggests a clear change in exponent values of the local magnitude distributions of earthquakes in japan , before the onset of a mainshock ( fig .
[ fig : kawamura ] ) .
this observation has definitely strengthened the possibility of using crossover signals in burst statistics as a criterion for imminent failure .
, much smaller than the average value @xmath1022 . from @xcite.,width=302 ] as we have seen , fiber bundle models provide a fertile ground for studying a wide range of breakdown phenomena . in some sense , they correspond to the ising model in the study of magnetism . in this section , we will review how the fiber bundle models are generalized to describe composites containing fibers .
such composites are of increasing practical importance , see e.g. , fig .
[ ah - fig11 ] .
the status of modeling fiber reinforced composites has recently been reviewed @xcite .
these materials consist of fibers embedded in a matrix . during tensile loading
the main part of the load is carried by the fibers and the strength of the composite is governed to a large extent by the strength of the fibers themselves .
the matrix material is chosen so that its yield threshold is lower than that of the fibers which are embedded in it .
common materials used for the fibers are aluminum , aluminum oxide , aluminum silica , asbestos , beryllium , beryllium carbide , beryllium oxide , carbon ( graphite ) , glass ( e - glass , s - glass , d - glass ) , molybdenum , polyamide ( aromatic polyamide , aramid ) , kevlar 29 and kevlar 49 , polyester , quartz ( fused silica ) , steel , tantalum , titanium , tungsten or tungsten monocarbide .
most matrix materials are resins as a result of their wide variation in properties and relatively low cost .
common resin materials are epoxy , phenolic , polyester , polyurethane and vinyl ester .
when the composite is to be used under adverse conditions such as high temperature , metallic matrix materials such as aluminum , copper , lead , magnesium , nickel , silver or titanium , or non - metallic matrix materials such as ceramics may be used . when the matrix material is brittle , cracks open up in the matrix perpendicular to the fiber direction at roughly equal spacing . in metallic matrix materials , plasticity sets in at sufficient load .
lastly , in polymer matrix composites , the matrix typically responds linearly , but still the fibers carry most of the load due to the large compliance of the matrix .
when a fiber fails , the forces it carried are redistributed among the surviving fibers and the matrix . if the matrix - fiber interface is weak compared to the strength of the fibers and the matrix themselves , fractures develop along the fibers .
when the matrix is brittle , the fibers bridging the developing crack in the matrix will , besides binding the crack together , lead to stress alleviation at the matrix crack front .
this leads to the energy necessary to propagate a crack further increases with the length of the crack , @xcite i.e. , so - called _ r - curve behavior _ @xcite .
when the bridging fibers fail , they typically do so through debonding at the fiber - matrix interface .
this is followed by pull - out , see fig .
[ ah - fig12 ] .
the cox shear lag model forms the basis for the standard tools used for analyzing breakdown in fiber reinforced composites @xcite .
it considers the elastic response of a single fiber in a homegeous matrix only capable of transmitting shear stresses . by treating the properties of the matrix as
effective and due to the self - consistent response by the matrix material and the rest of the fibers , the cox model becomes a mean - field model @xcite .
extensions of the cox single - fiber model to debonding and slip at the fiber - matrix interface have been published @xcite . in 1961 the single - fiber calculation of cox
was extended to two - dimensional unidirectional fibers in a compliant matrix , i.e. , a matrix incapable of carrying tensile stress , by @xcite . in 1967 ,
this calculation was followed up by @xcite for three - dimensional unidirectional fibers placed in a square or hexagonal pattern .
they found the average _ stress intensity factor _
( i.e. , the ratio between local stress in an intact fiber and the applied stress ) to be @xmath1023 after @xmath359 fibers failing . in his ph.d .
thesis , @xcite extended these calculations to aligned arrays of broken fibers mixed with intact fibers .
this approach was subsequently generalized to systems where the matrix has a non - zero stiffness and hence is able to transmit stress @xcite .
viscoelasticity of the matrix has been included by @xcite and @xcite .
@xcite demonstrated that when the fibers respond under global load - sharing conditions , a mean - field theory may be constructed where the breakdown of the composite is reduced to that of the failure of a single fiber in an effective matrix @xcite .
@xcite studied the redistribution of forces onto the neighbors of a single failing fiber within a two - dimensional uni - directional composite using the shear - lag model , finding that within this scheme , the stress - enhancement is less pronounced than earlier calculations had shown .
@xcite introduced a multi - fiber failure model including debonding and frictional effects at the fiber - matrix interface , finding that the stress intensity factor would decrease with increasing interfiber distance .
an important calculational principle , the _ break influence superposition technique _ was introduced by @xcite based on the method of @xcite in order to handle models with multiple fiber failures .
the technique consists in determining the transmission factors , which give the load at a given position along a given fiber due to a unit negative load at the single break point in the fiber bundle .
the multiple failure case is then constructed through superposition of these single - failure transmission factors .
this method has proven very efficient from a numerical point of view , and has been generalized through a series of later papers , see @xcite .
@xcite , @xcite , @xcite and @xcite analyzed the interaction between multiple breaks in uni - directional fibers embedded in a matrix using a lattice green function technique @xcite to calculate the load transfer from broken to unbroken fibers including fiber - matrix sliding with a constant interfacial shear resistance @xmath300 , given by either a debonded sliding interface or by matrix shear yielding .
the differential load carrying capacity of the matrix is assumed to be negligible . in the following
we describe the zhou and curtin approach in some detail .
the load - bearing fibers have a strength distribution given by the cumulative probability @xmath1024 of failure over a length of fiber @xmath709 experiencing a stress @xmath48 , where @xmath1025 where @xmath8 is the weibull index . when a fiber breaks , the load is transferred to the unbroken fibers .
we will return to the details henceforth .
the newly broken fiber slides relatively to the matrix . the shear resistance @xmath300 provides an average axial fiber stress along the single broken fiber @xmath1026 at a distance @xmath455 from the break , where @xmath748 is radius of the fibers and @xmath1027 is the axial fiber stress prior to the failure at point @xmath455 .
this defines a length scale @xmath1028 the total stress change within a distance @xmath1029 of the break is distributed to the other fibers .
a key assumption in what now follows is that the total stress in each plane @xmath455 is conserved : the stress difference @xmath1030 is distributed among the other intact fibers at the same @xmath455-level . in order to set up the lattice green function approach
, the system must be discretized .
each fiber , oriented in the @xmath455 direction , of length @xmath1031 is divided into @xmath1032 elements of length @xmath1033 .
the fibers are arranged on the nodes of a square lattice in the @xmath1034 plane so that there is a total of @xmath1035 fibers .
the lattice constants in the @xmath2 and @xmath487 directions are @xmath1036 and @xmath1037 , respectively .
each fiber is labeled by @xmath352 where @xmath1038 .
this is shown in fig.[ah - fig14 ] .
the stress on fiber @xmath352 in layer @xmath509 along the @xmath455 direction is given by @xmath1039 .
fiber @xmath352 at layer @xmath509 may be intact .
it then acts as a hookean spring with spring constant @xmath1040 responding to the stress @xmath1039 .
if fiber @xmath352 has broken at layer @xmath509 , it carries a stress equal to zero .
the third possibility is that fiber @xmath352 has broken elsewhere at @xmath1041 , and layer @xmath509 is within the slip zone .
it then carries a stress @xmath1042 which is the discretization of eq .
( [ zhou_curtin_slide ] ) with zero spring constant .
each element @xmath509 of fiber @xmath352 has two end nodes associated with it . at all such nodes
, springs parallel to the @xmath1034 plane are placed linking fiber @xmath352 with its nearest neighbors .
these springs have spring constant @xmath1043 .
the displacement of the nodes is assumed confined to the @xmath455 direction only .
zhou and curtin denote the displacement of node connecting element @xmath509 with element @xmath1044 of fiber @xmath352 , @xmath1045 , and the displacement of node linking element @xmath509 with element @xmath1046 of fiber @xmath352 , @xmath1047 .
the force on element @xmath509 of fiber @xmath352 from element @xmath509 of fiber @xmath1048 is @xmath1049 the reader should compare the following discussion with that which was presented in section [ sec:4c ] .
we now assume that it is only layer @xmath1050 that carries any damaged or slipped elements , the rest of the layers @xmath1051 are perfect .
let @xmath1052 .
if a force @xmath1053 is applied to the nodes , the response is @xmath1054 where @xmath1055 is the lattice green function .
given the displacements from solving this equation combined with eq.([zhou_curtin_force_on_n ] ) , the force carried by each broken element is found .
the inverse of the lattice green function is @xmath1056 .
the elements of @xmath1057 are either zero , @xmath1043 or @xmath1040 , reflecting the status of the springs ; undamaged , slipping or broken . when there are no breaks in layer @xmath1050 , we define @xmath1058 , and @xmath1059 .
hence , @xmath1060 plays a rle somewhat similar to the matrix @xmath1061 defined in eq .
( [ m4 ] ) . by combining these definitions ,
zhou and curtin find @xmath1062 the matrices @xmath1055 and @xmath1057 have dimension @xmath1063 where @xmath1064 . by appropriately labeling the rows and columns ,
the matrices @xmath1065 and @xmath1055 may be written @xmath1066 where the @xmath1067 matrix @xmath1068 couples elements _ within _ the layer @xmath1050 , where all the damage is located .
the matrix @xmath1069 couples elements within the rest of the layers . these are undamaged perfect " . the two matrices @xmath1070 and @xmath1071 provide the cross couplings .
the matrix @xmath1072 becomes in this representation @xmath1073 combining this equation with eq .
( [ zhou_curtin_g_and_d ] ) gives @xmath1074 where the intact green function @xmath1075 may be found analytically by solving eq .
( [ zhou_curtin_response ] ) for the intact lattice in fourier space @xmath1076 , @xmath1077^{-1}\;. \nonumber\\\end{aligned}\ ] ] by using that @xmath1078 , zhou and curtin find that @xmath1079 f^+_{n',0}\;,\ ] ] where @xmath1080 and @xmath1081 refer to the upper and lower node attached to element @xmath352 is layer @xmath1050 . before completing the model , the weibull strength distribution , eqs.([zhou_curtin_cumulative ] ) and ( [ zhou_curtin_phi ] ) , must be discretized .
each element @xmath1082 is given a maximum sustainable load @xmath1083 from the cumulative probability @xmath1084 where @xmath1085 .
the breakdown algorithm proceeds as follows : 1 .
a force per fiber set equal to the smallest breaking threshold , @xmath1086 , is applied to the system .
2 . the weakest fiber or fibers are broken by setting their spring constants to zero . 3 . decrease the stresses in the element below and above the just broken fibers according to eq .
( [ zhou_curtin_disc_slip ] ) .
4 . solve eq .
( [ zhou_curtin_dyson ] ) for the layers in which the breaks occurred .
calculate the spring displacements in the layers where the breaks occurred using eq .
( [ zhou_curtin_u_u ] ) and an effective applied force @xmath1087 .
the force on each intact spring in such a layer is then @xmath1088 .
6 . with the new stresses , search for other springs that carry a force beyond their thresholds @xmath1083 .
if such springs are found , break these and return to ( 2 ) .
otherwise proceed .
search for the spring which is closest to it breaking threshold .
this spring is the one with @xmath1089 .
increase the load by a factor @xmath1090 , where @xmath870 is equal to or somewhat larger than unity .
this factor is present to take into account the non - linearities introduced in the system due to the slip of the fibers .
proceed until the system no longer can sustain a load . by changing the ratio @xmath1091 between the moduli of the springs in the discretized lattice , it is possible to go from fiber bundle behavior essentially evolving according to equal load sharing ( els ) to local load sharing ( lls ) . whereas the computational cost of finite - element calculations on fiber reinforced composites scales with the volume of the composite , the break influence superposition technique and the lattice green function technique scale with the number of fiber breaks in the sample .
this translates into systems studied by the latter two techniques can be orders of magnitude larger than the former @xcite .
after this rather sketchy tour through the use of fiber bundle models as tools for describing the increasing important fiber reinforced composite materials , we now turn to the use of fiber bundle models in non - mechanical settings .
the typical failure dynamics of the fiber bundle model captures quite faithfully the failure behavior of several multicomponent systems like the communication or traffic networks .
similar to the elastic networks considered here , as the local stress or load ( transmission rate or traffic currents ) at any part of the network goes beyond the sustainable limit , that part of the system or the network fails or gets jammed , and the excess load gets redistributed over the other intact parts .
this , in turn , may induce further failure or jamming in the system . because of the tectonic motions stresses develop at the crust - tectonic plate resting ( contact ) regions and the failure at any of these supports induces additional stresses elsewhere .
apart from the healing phenomena in geological faults , the fiber bundle models have built - in features to capture the earthquake dynamics . naturally , the statistically established laws for earthquake dynamics can be easily recast into the forms derived here for the fiber bundle models .
we will consider here in some more details these three applications specifically .
the fiber bundle model has been applied @xcite to study the cascading failures of network structures , like erdos - renyi networks @xcite , known as er networks and watts - strogatz networks @xcite , known as ws networks , to model the overloading failures in power grids , etc . here , the nodes or the individual power stations are modelled as fibers and the transmission links between these nodes are utilized to transfer the excess load ( from one broken fiber or station to another ) .
the load transfer of broken fibers or nodes through the edges or links of the underlying network is governed by the lls rule @xcite . under a non - zero external load @xmath1092
, the actual stress @xmath1093 of the intact fiber @xmath435 is given by the sum of @xmath48 and the transferred load from neighboring broken fibers .
the local load transfer , from broken fibers to intact fibers , depends on the load concentration factor @xmath1094 with @xmath1095 , where the primed summation is over the cluster of broken fibers directly connected to @xmath435 , @xmath1096 is the number of broken fibers in the cluster @xmath750 , and @xmath1097 is the number of intact fibers directly connected to @xmath750 .
let the external stress @xmath48 be increased by an infinitesimal amount @xmath1098 starting from @xmath1099 .
fibers for which strength @xmath1100 break iteratively until no more fibers break . for each increment of @xmath48
, the size @xmath1101 of the avalanche is defined as the number of broken fibers triggered by the increment .
the surviving fraction @xmath1102 of fibers can be written as @xmath1103 one can also measure directly the response function @xmath49 , or the generalized susceptibility , denoted as @xmath1104 the critical value @xmath181 of the external load , can be defined from the condition of the global breakdown @xmath1105 .
the critical value @xmath181 and the susceptibility @xmath49 have been calculated numerically for the model under the lls rule on various network structures , such as the local regular network , the ws network , the er network , and the scale - free networks ( fig .
[ fig : sigmac ] ) .
the results suggest that the critical behavior of the model on complex networks is completely different from that on a regular lattice . more specifically , while @xmath181 for the fbm on a local regular network vanishes in the thermodynamic limit and is described by @xmath1106 for finite - sized systems ( see the curve for @xmath1107 in fig .
[ fig : sigmac ] , corresponding to the ws network with the rewiring probability @xmath1107 ) , @xmath181 for _ all _ networks except for the local regular one does not diminish but converges to a nonzero value as @xmath0 is increased .
moreover , the susceptibility diverges at the critical point as @xmath1108 , regardless of the networks , which is again in a sharp contrast to the local regular network [ see fig . [
fig : sigmac](b ) ] .
the critical exponent @xmath341 clearly indicates that the fbm under the lls rule on complex networks belongs to the same universality class as that of the els regime @xcite although the load - sharing rule is strictly local .
the observed variation for @xmath90 is only natural for lls model . in an lls model
if @xmath352 successive fibers fail ( each with a finite probability @xmath1109 ) , then the total probability of such an event is @xmath1110 as the probability is proportional to the bundle size @xmath0 . if this probability is finite , then @xmath1111 ( for any finite @xmath1109 , the failure probability of any fiber in the bundle ) . for a failure of @xmath352
successive fibers , the neighboring intact fibers get the transferred load @xmath1112 which if becomes greater than or equal to their strength , it surely fails giving @xmath1113 ( see section iv.a ) .
the evidence that the lls model on complex networks belongs to the universality class of the els model is also found @xcite in the avalanche size distribution @xmath363 : unanimously observed power - law behavior @xmath1114 ( a ) for all networks except for the local regular one ( the ws network with @xmath1107 ) is in perfect agreement with the behavior for the els case @xcite . on the other hand ,
the lls model for a regular lattice has been shown to exhibit completely different avalanche size distribution @xcite .
also one can observe a clear difference in terms of the failure probability @xmath1115 defined as the probability of failure of the whole system at an external stress @xmath48 . while @xmath1115 values for lls on complex networks fall on a common line , lls on regular network shows a distinctly different trend @xcite , see also @xcite .
one can apply the equal load sharing fiber bundle model to study the traffic failure in a system of parallel road network in a city .
for some special distributions , like the uniform distribution , of traffic handling capacities ( thresholds ) of the roads , the critical behavior of the jamming transition can be studied analytically .
this , in fact , is exactly comparable with that for the asymmetric simple exclusion process in a single channel or road @xcite .
traffic jams or congestions occur essentially due to the excluded volume effects ( among the vehicles ) in a single road and due to the cooperative ( traffic ) load sharing by the ( free ) lanes or roads in multiply connected road networks ( see e.g. , @xcite ) . using fbm for the traffic network ,
it has been shown @xcite that the generic equation for the approach of the jamming transition in fbm corresponds to that for the asymmetric simple exclusion processes ( asep ) leading to the transport failure transition in a single channel or lane ( see e.g. , @xcite ) .
let the suburban highway traffic , while entering the city , get fragmented equally through the various narrower streets within the city and get combined again outside the city ( see fig .
[ bkc - traffic - model ] ) .
if @xmath1116 denotes the input traffic current and @xmath1117 is the total output traffic current , then at the steady state , without any global traffic jam , @xmath1118 . in case @xmath1117
falls below @xmath1116 , the global jam starts and soon @xmath1117 drops to zero .
this occurs if @xmath1119 , the critical traffic current of the network , beyond which global traffic jam occurs .
let the parallel roads within the city have different thresholds for traffic handling capacity : @xmath1120 for the @xmath0 different roads ( the @xmath352-th road gets jammed if the traffic current @xmath435 per road exceeds @xmath1121 ) .
initially @xmath1122 and increases as some of the roads get jammed and the same traffic load @xmath1116 has to be shared equally by a lower number of unjammed roads .
next , we assume that the distribution @xmath1123 of these thresholds is uniform up to a maximum threshold current ( corresponding to the widest road traffic current capacity ) , which is normalized to unity ( sets the scale for @xmath1124 ) .
gets fragmented into uniform currents @xmath435 in each of the narrower roads and the roads having threshold current @xmath1125 get congested or blocked .
this results in extra load for the uncongested roads .
we assume that this extra load per unconjested roads gets equally redistributed and gets added to the existing load , causing further blocking of some more roads .
, width=240,height=240 ] the jamming dynamics in this model starts from the @xmath352-th road ( say in the morning ) when the traffic load @xmath435 per city roads exceeds the threshold @xmath1126 of that road . due to this jam ,
the total number of unconjested roads decreases and the rest of these roads have to bear the entire traffic load in the system .
hence the effective traffic load or stress on the uncongested roads increases and this compels some more roads to get jammed .
these two sequential operations , namely the stress or traffic load redistribution and further failure in service of roads continue till an equilibrium is reached , where either the surviving roads are strong ( wide ) enough to share equally and carry the entire traffic load on the system ( for @xmath1127 ) or all the roads fail ( for @xmath1128 ) and a ( global ) traffic jam occurs in the entire road network system .
this jamming dynamics can be represented by recursion relations in discrete time steps .
let @xmath1129 be the fraction of unconjested roads in the network that survive after ( discrete ) time step @xmath18 , counted from the time @xmath1130 when the load ( at the level @xmath1131 ) is put in the system ( time step indicates the number of stress redistributions ) .
as such , @xmath1132 for all @xmath18 and @xmath1133 for @xmath1130 for any @xmath435 ; @xmath1134 for @xmath924 if @xmath1135 , and @xmath1136 for @xmath924 if @xmath1119 .
here @xmath1129 follows a simple recursion relation @xmath1137 @xmath1138 the critical behavior of this model remains the same as discussed in section iii.a in terms of @xmath435 and the exponent values remain unchanged : @xmath1139 , @xmath66 for all these equal ( traffic ) load sharing models . in the one dimensional lane or road
is possible if , say , the @xmath705-th site is occupied and the @xmath1140-th site is vacant .
the inter - site hopping probability is indicated by @xmath1141 .
, width=240,height=144 ] in the simplest version of the asymmetric simple exclusion process transport in a chain ( see fig .
[ aspe - model ] ) , the transport corresponds to movement of vehicles , which is possible only when a vehicle at site @xmath705 , say , moves to the vacant site @xmath1142 .
the transport current @xmath574 is then given by @xcite @xmath1143 where @xmath1144 denotes the site occupation density at site @xmath705 and @xmath1141 denotes the inter - site hopping probability .
the above equation can be easily recast in the form @xmath1145 where @xmath1146 .
formally it is the same as the recursion relation for the density of uncongested roads in the fbm model discussed above ; the site index here in asep plays the role of time index in fbm .
such exact correspondence indicates identical critical behavior in both the cases .
the same universality for different cases ( different threshold distributions ) in fbm suggests similar behavior for other equivalent asep cases as well @xcite . for extension of the model to scale free traffic networks ,
see @xcite .
the earth s outer crust , several tens of kilometers in thickness , rests on tectonic shells . due to the high temperature - pressure phase changes and consequent ionizations of the metallic ores , powerful
magneto - hydrodynamic convective flows occur in the earth s mantle at several hundreds of kilometers in depth .
the tectonic shell , divided into about ten mobile plates , have got relative velocities of the order of a few centimeters per year , see e.g. , @xcite .
the stresses developed at the interfaces between the crust and the tectonic shells during the ( long ) _ sticking _ periods get released during the ( very short ) _ slips _ , causing the releases of the stored elastic energies ( at the fault asperities ) and consequent earthquakes .
two well known phenomenologically established laws governing the earthquake statistics are ( a ) the gutenberg - richter law @xmath1147 relating the number density ( @xmath1148 ) of earthquakes with the released energy greater than or equal to @xmath571 ; and ( b ) the omori law @xmath1149 where @xmath1150 denotes the number of aftershocks having magnitude or released energy larger than a preassigned small but otherwise arbitrary threshold value . as mentioned already ,
because of the tectonic motions , stresses develop at the crust - tectonic plate contact regions and the entire load is supported by such regions .
failure at any of these supports necessitates load redistributions and induces additional stresses elsewhere .
in fact the avalanche statistics discussed in section iii.b can easily explain the gutenberg - richter law ( eq . [ gr - law ] ) with the identification @xmath1151 .
similarly , the decay of the number of failed fibers @xmath1152 at the critical point , given by @xmath1153 ( see e.g. , eq . [
eq : fbm - omori ] ) , crudely speaking , gives in turn the omori law behavior ( eq .
[ omori - law ] ) for the fiber bundle model , with the identification @xmath1154 .
for some recent discussions on further studies along these lines , see e.g. , @xcite .
the stick - slip motion in the burridge - knopoff model ( see e.g. , @xcite ) , where the blocks ( representing portion of the solid crust ) connected with springs ( representing the elastic strain developed due to tectonic motion ) are pulled uniformly on a rough surface , has the same feature of stress redistribution as one or more blocks slip and the dynamics was mapped onto a els fiber bundle model by @xcite .
the power law distributions of the fluctuation driven bursts around the critical points have been interpreted as the above two statistical laws for earthquakes .
the fiber bundle model enjoys a rare double position in that it is both useful in a practical setting for describing a class of real materials under real working conditions , and at the same time being abstract enough to function as a model for exploring fundamental breakdown mechanisms from a general point of view .
very few models are capable of such a double life .
this means that a review of the fiber bundle model may take on very different character depending on the point of view .
we have in this review emphasized the fiber bundle model as a model for exploring fundamental breakdown mechanisms .
failure in loaded disordered materials is a collective phenomenon .
it proceeds through a competition between disorder and stress distribution .
the disorder implies a distribution of local strength .
if the stress distribution were uniform in the material , it would be the weakest spot that would fail first .
suppose now that there has been a local failure at a given spot in the material .
the further away we go from this failed region , the weaker the weakest region within this distance will be .
hence , the disorder makes local failures repel each other : they will occur as far as possible from each other .
however , as the material fails locally , the stresses are redistributed .
this redistribution creates hot spots where local failure is likely due to high stresses .
since these hot spots occur at the boundaries of the failed regions , the effect of the stress field is an attraction between the local failures @xcite .
hence , disorder and stress has opposite effect on the breakdown process ; repulsion vs. attraction , and this leads to competition between them . since the disorder in the strength of the material , leading to repulsion , dimishes throughout the breakdown process , whereas the stresses create increasingly important hot spots , it is the stress distribution that ends up dominating towards the end of the process .
the fiber bundle model catches this essential aspect of the failure process . depending on the load redistribution mechanism , the quantitative aspects change .
however , qualitatively it remains the same . in the els case , there are no _ localized _ hot spots , all surviving fibers are loaded the same way .
geometry does not enter into the redistribution of forces , and we may say that the hot spots " include all surviving fibers .
this aspect gives the els fiber bundle model its mean field character , even though all other fluctuations are present , such as those giving rise to bursts .
as shown in section iii , the lack of geometrical aspects in the redistribution of forces in the els model enables us to construct the recursion relations ( e.g. , eq . [ rec - u ] ) which capture well the failure dynamics .
we find that the eventual statistics , governed by the fixed points for @xmath1155 , the average strength of the bundle , essentially shows a normal critical behavior : order parameter @xmath1156 , breakdown susceptibility @xmath1157 and relaxation time @xmath1158 with @xmath1159 and @xmath1160 for a bundle of @xmath0 fibers , where subscripts @xmath1161 and @xmath687 refers to post and pre critical cases respectively .
the statistics of fluctuations over these average behavior , given by the avalanche size distributions @xmath1162 with @xmath1163 for discrete load increment and @xmath1164 for quasi - static load increment in such els cases .
the critical stress @xmath90 of the bundle is of course nonuniversal and its magnitude depends on the fiber strength distribution . for the lls model , we essentially find ( see section iv.a ) , the critical strength of the fiber bundle @xmath1165 which vanishes in the macroscopic system size limit . the avalanche size distribution is exponential for such cases . for range - dependent redistribution of load ( see section iv.b ) one recovers the finite value of @xmath90 and the els - like mean field behavior for its failure statistics
. extensions of the model to capture creep and fatigue behavior of composite materials are discussed in section v.a .
precursors of global failure are discussed in section v.b .
it appears , a detailed knowledge of the critical behavior of the model can help very precise determination of the global failure point from the well defined precursors .
section v.c provides a rather cursory review of models of fiber reinforced composites .
these models go far beyond the simple fiber bundle model in complexity and represent the state of the art of theoretical approaches to this important class of materials . however , as complicated as these models are , the philosophy of the fiber bundle model is still very much present .
finally we discussed a few extensions of the model to failures in communication networks , traffic jams and earthquakes in section v.d .
as discussed here in details , the fiber bundle model not only gives an elegant and profound solution of the dynamic critical phenomena of failures in disordered systems , with the associated universality classes etc , but also offers the first solution to the entire linear and non - linear stress - strain behavior for any material up to its fracture or rupture point .
although the model had been introduced at about the same time ( @xmath1 ) as the ising model for static critical phenomena , it is only now that the full ( mean - field ) critical dynamics in the fiber bundle model is solved .
apart from these , as already discussed , several aspects of the fluctuations in this model are now well understood .
even from this specific point of view , the model is not only intuitively very attractive , its behavior is extremely rich and intriguing
. it would be surprising if it did not offer new profound insights into failure phenomena also in the future .
we thank p. bhattacharyya and p. c. hemmer for important collaborations at different parts of this work .
we acknowledge the financial support from norwegian research council through grant no .
nfr 177958/v30 .
thanks sintef petroleum research for providing partial financial help and moral support toward this work . | the fiber bundle model describes a collection of elastic fibers under load .
the fibers fail sucessively and for each failure , the load distribution among the surviving fibers changes .
even though very simple , this model captures the essentials of failure processes in a large number of materials and settings .
we present here a review of the fiber bundle model with different load redistribution mechanisms from the point of view of statistics and statistical physics rather than materials science , with a focus on concepts such as criticality , universality and fluctuations .
we discuss the fiber bundle model as a tool for understanding phenomena such as creep , and fatigue , how it is used to describe the behavior of fiber reinforced composites as well as modelling e.g. network failure , traffic jams and earthquake dynamics . |
the supersymmetrized ( susy ) standard model ( sm ) , _ e.g. _ , the minimal supersymmetric standard model ( mssm ) , has for a long time intrigued particle theorists in that it is free of the scalar field quadratic divergences that plague non - supersymmetric theories @xcite .
in addition , the mssm has made three predictions which have since been verified by experiment : 1 .
the value of @xmath16 which arises from unified gauge couplings at @xmath17 gev that evolve via renormalization group ( rg ) evolution down to the weak scale within the context of the mssm @xcite , 2 .
the large top quark mass @xmath18 gev @xcite is exactly what is needed to initiate a radiative breakdown of electroweak symmetry in the mssm @xcite , and 3 .
the measured value of the higgs boson mass @xmath19 gev @xcite which falls squarely within the narrow window required by the mssm @xcite .
in contrast , so far no evidence for direct production of superpartners has emerged at lhc , leading to mass limits @xmath20 gev @xcite and @xmath21 gev @xcite in the context of various simplified models .
the latter lower bound has been particularly disconcerting since it is in direct conflict with an oft - repeated mantra that one or more light third generation squarks ( @xmath22 gev ) are required for a _
natural _ susy solution to the little hierarchy ( lh ) problem . here
, the lh is characterized by the growing gap between the weak scale , as represented by @xmath23 gev , and the superparticle mass scale @xmath24 which apparently lies within the multi - tev range .
the light top squark narrative has lead to an `` all hands on deck '' call for exploring every conceivable gap of allowed masses and decay modes in the simplified model @xmath25 vs. @xmath26 ( the lsp , lightest susy particle ) plane .
the impression has been made that by covering every possibility for existence of light top squarks , then one may be ruling out weak scale susy or else showing that whatever form susy takes , it is not as `` we '' understood it @xcite .
the top squark mass bound is also being invoked to justify costly decisions regarding future experimental facilities : if weak scale susy as we know it is ruled out , and the sm remains valid well into the multi - tev range , then perhaps a 100 tev hadron collider is the way to go as all bets from theory would be off . alternatively ,
if susy remains just beyond the energy horizon , then perhaps ilc and an energy upgrade lhc ( he - lhc ) operating with @xmath2733 tev are the right machines to build . given the stakes involved , it is becoming critical to ensure the validity of our reasoning regarding the notions of electroweak naturalness and fine - tuning . to address this issue , in sec .
[ sec : nat ] we briefly review several estimates of electroweak naturalness in the sm and in susy . we believe that several common measures are technically mis - applied in the susy case . when corrected to allow for the fact that
the soft parameters should be correlated , they reduce to the model independent measure @xmath28 , where ew denotes electroweak@xcite .
the latter measure also leads to bounds on top squarks and gluinos , but instead allows for @xmath29 tev and @xmath30 tev at little cost to naturalness since these masses enter into the value of @xmath31 as finite one- and two - loop corrections respectively . in sec .
[ sec : bm ] , we present a top squark benchmark model from the two - extra - parameter non - universal higgs model @xcite ( nuhm2 ) which allows for highly natural susy spectra with @xmath19 gev .
this leads to a grand overview plot of expectations for populating the @xmath25 vs. @xmath32 plane in sec .
[ sec : plane ] .
this plot presents a guide for top squark hunters at the lhc as to where in the plane their quarry of natural susy solutions lies for low values of @xmath28 . here
, we find @xmath331.8 tev for @xmath34 while @xmath29 tev for @xmath35 .
hardly any solutions lie in the highly scrutinized _ compressed _ region where @xmath36 . in sec . [
sec : bsg ] , we evaluate expectations for the flavor - changing decay @xmath37 versus @xmath25 and find for @xmath38 gev that one always expects large deviations from the measured value whereas for @xmath39 tev , then the susy loops decouple and one gains accord with experiment : in this sense , it comes as no great surprise that lhc top squark hunters have yet to sight their trophy . in sec .
[ sec : prod ] , we outline top squark production and decay rates for natural susy and in sec . [ sec : prospects ] we outline a more realistic proposal for future top squark searches in a semi - simplified model which corresponds more closely with predictions from theory .
a summary and conclusions are given in sec .
[ sec : conclude ] .
for any observable @xmath40 , if the contributions to @xmath40 are given by = a + b + f(b ) + c , then we would claim the value of @xmath40 is _ natural _ if each contribution on the right - hand - side is comparable to or less than @xmath41 . if this were not the case ,
if say one contribution @xmath42 were far larger than @xmath40 , then some other contribution would have to be fine - tuned to large opposite - sign values such as to maintain the measured value of @xmath40 .
thus , the naturalness measure = ||/|o| would be vindicated ( here , rhs stands for right - hand - side ) . in the case of the quantity
@xmath43 , if as a consequence of @xmath44 getting large , then @xmath43 becomes large negative , these two quantities are _ dependent _ and should be _ combined _ before evaluating naturalness .
this is embodied by the fine - tuning rule articulated in ref .
@xcite : _ in evaluating fine - tuning , it is not permissible to claim fine - tuning of dependent quantities one against another_. for illustration , in the case of the sm with a scalar potential given by v=-^2_sm|^|+|^|^2 , the physical higgs boson mass is given by m_h_sm^2 2 ^ 2_sm+m_h_sm^2 , where the largest contribution to @xmath45 comes from the famous quadratic divergences : m_h_sm^2(-_t^2
+ + + ) ^2 , where @xmath46 is the sm top - quark yukawa coupling , @xmath47 is the @xmath48 gauge coupling and @xmath49 represents the energy scale cut - off on the quadratically divergent one - loop mass corrections . since @xmath50 is _ independent _ of @xmath45 , then @xmath51 can be freely dialed , or fine - tuned , to maintain the measured value of @xmath52 gev @xcite .
a valid measure of fine - tuning here would be @xmath53 .
requiring @xmath54 implies an upper bound on the sm effective theory energy cutoff of @xmath55 tev .
the situation in the mssm is quite different @xcite . in this case ,
the well - known quadratic divergences all cancel but there remains a variety of intertwined logarithmic divergent contributions to @xmath56 . in the mssm , we have m_h^2 - 2\ { ^2 ( ) + m_h_u^2 ( ) } ~-2\ { ^2 ( ) + m_h_u^2 ( ) + m_h_u^2 ( ) } , where now @xmath57 is the superpotential higgsino mass term and @xmath58 is the up - higgs soft susy breaking squared mass . the quantity @xmath59 is properly evaluated by integrating the renormalization group equation : = ( -g_1 ^ 2m_1 ^ 2 - 3g_2 ^ 2m_2 ^ 2+g_1 ^ 2s+3f_t^2x_t ) , where @xmath60 with @xmath61 the renormalization scale , @xmath62 ( @xmath633 ) are the various gaugino masses , @xmath64 are the corresponding gauge coupling constants , @xmath65 is the top yukawa coupling , s = m_h_u^2-m_h_d^2+tr , and x_t = m_q_3 ^ 2+m_u_3 ^ 2+m_h_u^2+a_t^2 , [ eq : xt ] where @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath70 , @xmath71 are the soft masses for the down - type higgs , left - handed squarks , left - handed sleptons , right - handed up - type squarks , right - handed down - type squarks , and right - handed charged sleptons , respectively , and @xmath72 is the @xmath73-term for the top yukawa coupling . to evaluate @xmath59 , it is common in the literature to set the gauge couplings , the @xmath74 parameter and @xmath58 equal to zero so that a simple one step integration can be performed leading to m_h_u^2 ( ) ~- ( m_q_3 ^ 2+m_u_3 ^ 2+a_t^2 ) ( ) .
[ eq : delmhus ] the fine - tuning measure @xmath75 requires at least one ( and actually three ) third generation squarks with mass less than 650 gev@xcite .
the issue here is that , unlike the sm case , @xmath76 is _ not independent _ of the high scale value of @xmath77 .
in fact , the larger @xmath78 is , the larger is the cancelling correction @xmath59 .
this violates the fine - tuning rule .
instead , one ought to first _ combine _ dependent contributions , then evaluate the independent contributions to the observed value of @xmath56 to check whether they exceed its value . upon regrouping @xmath79 , where @xmath80 .
then it is seen that the criteria for naturalness is that the _ weak scale _ values of @xmath81 and @xmath58 are each comparable to @xmath56 .
this corrected measure allows for _ radiatively - driven naturalness _ : large , unnatural values of @xmath58 at the high scale @xmath49 may be driven to natural values at the weak scale via radiative corrections@xcite .
the measure @xmath82 was proposed by ellis _
_ @xcite and investigated more thoroughly by barbieri and giudice @xcite . here ,
the @xmath83 are fundamental parameters of the theory labeled by index @xmath84 .
to begin , one may express @xmath85 in terms of weak scale susy parameters m_z^2 -2 ^ 2 -2m_h_u^2 , [ eq : mzsapprox ] where the partial equality holds for moderate - to - large @xmath86 values ( @xmath87 is the ratio of the higgs vevs ) and where we assume for now the radiative corrections are small .
next , one needs to know the explicit dependence of @xmath58 and @xmath81 on the fundamental parameters .
semi - analytic solutions to the one - loop renormalization group equations for @xmath58 and @xmath81 can be found for instance in ref s @xcite .
for the case of @xmath88 , it is found that @xcite m_z^2 & & -2.18 ^ 2 + 3.84 m_3 ^ 2 + 0.32m_3m_2 + 0.047 m_1m_3 - 0.42 m_2 ^ 2 + & & + 0.011 m_2m_1 - 0.012m_1 ^ 2 - 0.65 m_3a_t-0.15 m_2a_t + & & -0.025m_1 a_t+0.22a_t^2 + 0.004 m_3a_b + & & -1.27 m_h_u^2 -0.053 m_h_d^2 + & & + 0.73 m_q_3 ^ 2 + 0.57 m_u_3 ^ 2 + 0.049 m_d_3 ^ 2 - 0.052 m_l_3 ^ 2 + 0.053 m_e_3 ^ 2 + & & + 0.051 m_q_2 ^ 2 - 0.11 m_u_2 ^ 2 + 0.051 m_d_2 ^ 2 - 0.052 m_l_2 ^ 2 + 0.053 m_e_2 ^ 2 + & & + 0.051 m_q_1 ^ 2 - 0.11 m_u_1 ^ 2 + 0.051 m_d_1 ^ 2 - 0.052 m_l_1 ^ 2 + 0.053 m_e_1 ^ 2 , [ eq : mzsparam ] where all terms on the right - hand - side are understood to be @xmath89 scale parameters .
the conundrum is then : what constitutes fundamental parameters ? if all gut scale parameters on the rhs of eq .
are fundamental , then for the doublet top squark soft term we would find @xmath90 and so @xmath91 would imply @xmath92 gev in accord with eq . . if instead we assume scalar mass universality as in the cmssm , then the fourth and fifth lines of eq .
combine to @xmath93 and instead @xmath94 would require @xmath95 tev : multi - tev scalars are natural as in focus - point susy @xcite .
in fact , in more fundamental supergravity theories with sugra breaking in a hidden sector , then _ all _ soft terms are computable as multiples of the more fundamental gravitino mass @xmath96 @xcite .
then all soft terms on the rhs of eq
. are _ dependent _ and must be combined according to the fine - tuning rule . in this case , eq .
collapses to a simpler form @xcite : m_z^2 - 2.18 ^ 2
+ am_3/2 ^ 2 , [ eq : mzsm32 ] and instead low fine - tuning requires @xmath97 and also @xmath98 . equating eq . with eq .
shows that @xmath99 and so we are led to consistency with the corrected implication of @xmath100 : the criteria for electroweak naturalness is that the weak scale values of @xmath101 and @xmath102 are @xmath103 gev .
the corrected versions of @xmath100 and @xmath104 are consistent with requiring low _ electroweak _
fine - tuning in @xmath85 .
minimization of the scalar potential in the minimal supersymmetric standard model ( mssm ) leads to the well - known relation @xcite & = & -^2[eq : mzs1 ] + & & -m_h_u^2-_u^u-^2 , [ eq : mzs ] where @xmath105 and @xmath106 denote the 1-loop corrections ( expressions can be found in the appendix of ref .
@xcite ) to the scalar potential , @xmath58 and @xmath66 are the higgs soft masses at the weak scale
. the second line is obtained from moderate to large values of @xmath107 ( as required by the higgs mass calculation@xcite ) .
susy models requiring large cancellations between the various terms on the right - hand - side of eq . to reproduce the measured value of @xmath85
are regarded as unnatural , or fine - tuned .
in contrast , susy models which generate terms on the rhs of eq . which are all less than or comparable to @xmath108 are regarded as natural .
thus , the _ electroweak _ naturalness measure @xmath109 is defined as @xcite radiative corrections , over 40 terms contribute .
the measure @xmath28 is programmed in the isajet spectrum generator isasugra @xcite .
neglecting radiative corrections , and taking moderate - to - large @xmath110 , then @xmath111 so the main criterion for naturalness is that _ at the weak scale _ @xmath112 and @xmath113 @xcite .
the value of @xmath66 ( where @xmath114 with @xmath115 being the mass of the cp - odd higgs boson ) can lie in the tev@xmath14 range since its contribution to the rhs of eq . is suppressed by @xmath116 .
the largest radiative corrections typically come from the top squark sector : _
u^u ( _ 1,2)= f(m__1,2 ^ 2 ) , [ eq : siguu ] where @xmath117 is the weak mixing angle , @xmath118 , @xmath119 , and @xmath120 , with @xmath121 . requiring highly mixed tev - scale top squarks
minimizes @xmath122 whilst lifting the higgs mass @xmath123 to @xmath124 gev @xcite . using @xmath125 or better than 3% fine - tuning , high fine - tuning sets in and
is displayed visually in fig . 2 of ref .
] then instead of earlier upper bounds , it is found that @xmath30 tev , @xmath29 tev and @xmath126 gev .
thus , gluinos and squarks may easily lie beyond the current reach of lhc at little cost to naturalness while only the higgsino - like lighter charginos and neutralinos are required to lie near the weak scale .
the lightest higgsino @xmath127 comprises a portion of the dark matter and would escape detection at lhc .
the remaining dark matter abundance might be comprised of _
e.g. _ axions@xcite .
owing to their compressed spectrum with mass gaps @xmath12820 gev , the heavier higgsinos are difficult to see at lhc owing to the rather small visible energy released from their three body decays @xmath129 and @xmath130 ( where the @xmath131 stands for sm fermions ) .
in this section , we illustrate some aspects of top squark and higgs boson masses and mixings for a sample susy benchmark model from the two - extra - parameter non - universal higgs model ( nuhm2@xcite ) with parameter space given by m_0 , m_1/2 , a_0 , , , m_a , where the non - universal gut scale parameters @xmath58 and @xmath66 have been exchanged for the more convenient weak scale values of @xmath57 and @xmath115 . here
, we will adopt parameter choices @xmath132 tev , @xmath133 gev , @xmath88 , @xmath134 gev and @xmath135 tev . in fig .
[ fig : mvsa0 ] frame _ a _ ) , we plot the values of the various third generation sfermion masses versus variation in the @xmath136 parameter . it is seen that for @xmath137 , then the various sfermion masses range between 3 and 5 tev .
as @xmath136 becomes large positive or negative , the @xmath138 contributions to the mssm rg equations tend to drive the soft masses @xmath139 and @xmath140 to lower values due to the @xmath141 ( eq . )
contribution to the rg running , which is amplified by the large top - quark yukawa coupling @xmath65 .
the @xmath142 and @xmath143 mass values hardly change since their rg equations include @xmath144 and @xmath145 which are only amplified by the much smaller @xmath146- and @xmath44-yukawa couplings . also , the large @xmath72 term causes large mixing in the top squark sector which enhances the splitting of the stop eigenstates .
only the value of @xmath25 is driven to sub - tev values for @xmath147 tev . in fig .
[ fig : mvsa0]_b _ ) , we show the value of @xmath123 vs. @xmath136 .
the higgs mass at one loop is given by m_h^2m_z^2 ^ 2 2 + , where now @xmath148 and @xmath149
. for a given value of @xmath150 , this expression is maximal for large mixing in the stop sector with @xmath151 .
we see from the plot that @xmath123 is maximal for large negative @xmath136 .
this is because the _ weak scale _ value of @xmath72 is large negative leading to large mixing in the stop sector . for large positive @xmath136
, then the value of @xmath72 largely cancels against gauge contributions in the @xmath72 running so @xmath72 runs to small values at the weak scale leading to small mixing and too small a value of @xmath123 : see fig .
[ fig : mvsa0]_c _ ) .
in frame [ fig : mvsa0]_d _ ) , we show the calculated value of @xmath28 . here , we see that @xmath152 for @xmath137 , but for this value of @xmath136 , the value of @xmath123 is too small . for @xmath153 tev , then we have large mixing leading to @xmath154 gev ( shown by the red - shaded part of the curve ) , but also some suppression in the @xmath155 values leading to very natural solutions with @xmath156 . for @xmath157 tev
, then @xmath28 drops below 30 , but unfortunately @xmath123 is too low at @xmath158 gev .
in this section , we present a grand overview of the locus of natural susy models in the @xmath25 vs. @xmath32 mass plane .
this plane was initially proposed as a template for top squark searches in ref .
@xcite and has now served for several years to give a panoramic view of top squark search results in various simplified models from lhc data . in fig .
[ fig : mt1mz1 ] , we present the results of the scan over nuhm2 parameter space from ref .
@xcite where upper bounds on sparticle masses were derived from requiring not - to - large values of @xmath28 .
the scan values were @xmath15920 tev , @xmath1603 tev , @xmath161 , @xmath1621.5 tev , @xmath16320 tev and @xmath16460 .
it was required that 1 .
electroweak symmetry to be radiatively broken , 2 .
the @xmath127 was lsp , 3 .
the lightest chargino obeyed the lep2 limit @xmath165 gev , 4 .
lhc8 bounds on @xmath166 and @xmath167 were respected and 5 .
@xmath168 gev . from the figure
, we see that solutions with @xmath34 are clustered with @xmath1691.3 tev while if we allow for @xmath125 then @xmath25 can range up to 3 tev .
gives an upper bound on the mass of @xmath170 as @xmath171 tev , which is much lower than our result .
this apparently severe bound results from the different strategy of the parameter scan , which turns out to be more restricted than ours .
for example , they scan parameters in the ranges of @xmath172 tev and @xmath173 tev at the weak scale . as can be seen from fig .
[ fig : mvsa0]_c _ ) , however , @xmath174 tev can give a very small value of @xmath109 , which is out of the range of the parameter scan in ref .
top - squark masses can also be as large as @xmath175 tev for @xmath176 . ]
the black - dotted line shows where @xmath36 which is the compressed region , in which laborious searches for top squark production are taking place .
notice that essentially no highly natural solutions lie in this region .
it is also important to note that the lsp is mainly higgsino - like in this region in order to satisfy naturalness with low @xmath28 .
we also present for comparison several search contours from the atlas collaboration .
the region within the solid - black contour represents the area ruled out by current atlas searches at lhc13 for @xmath177 : for @xmath178 or @xmath179 @xcite and for @xmath180 @xcite .
these search results range up to @xmath181 gev which covers only a fraction of the expected range from natural susy .
we note , however , that some of these limits might be significantly relaxed in the present case as they are obtained on the assumption of a specific decay channel in a simplified setup .
for example , the limit from the atlas one lepton , jets plus missing energy search @xcite , in which all of the produced top squarks are assumed to decay into @xmath182 , would be relaxed since the @xmath183 decay branch is about 50% in the natural susy parameter space , as we will see below . considering this ,
we also show in fig .
[ fig : mt1mz1 ] as the red contour the limits presented in ref .
@xcite , which are obtained by recasting the cms top - squark mass limits @xcite for models with light higgsinos ; the resultant upper bound on the top - squark mass is again found to be about @xmath184 gev . in fig .
[ fig : mt1mz1 ] , we show as well projected contours of what hl - lhc can achieve via the top - squark search in the 0-lepton channel ; the @xmath185 discovery and 95% cl exclusion contours with 3000 fb@xmath186 integrated luminosity data are shown in the green and orange solid lines , respectively @xcite . here
, we see that hl - lhc with 3000 fb@xmath186 of integrated luminosity may be able to probe up to @xmath187 tev .
this can be compared with a recent theory study@xcite finding hl - lhc may probe top squark pair signatures to @xmath187 tev .
a combination of the 0-lepton and 1-lepton search results may further push its reach by @xmath188 gev @xcite . in either case , hl - lhc probes perhaps less than half the natural susy parameter space via top squark pair searches . before concluding this section , we comment on the excess events observed in the atlas top - squark searches based on the one lepton , jets plus missing energy final states @xcite , where @xmath189 , @xmath190 , and @xmath191 excesses are observed in the signal categories , sr1 , bc2x_diag , and dm_low , respectively . as discussed in ref .
@xcite , these excesses may be explained with a top squark with a mass of @xmath192 gev and light higgsinos with masses of @xmath193 gev .
however , such parameter region has already been excluded by other searches @xcite as shown in fig .
[ fig : mt1mz1 ] , and thus these excesses are not accounted for in the present setup .
here , we examine expectations for the rare branching fraction @xmath195 which takes place via @xmath196 loops in the sm and via @xmath197 and @xmath198 loops in susy @xcite ( other susy loops also contribute but typically with much smaller amplitudes ) . the sm value for this decay is found to be @xcite @xmath199 which is to be compared to the recent belle measurement@xcite that @xmath200 . for the susy @xmath37 calculation
, we use the nlo results from @xcite which is encoded in isatools @xcite . in fig .
[ fig : bsg ] , we show the predicted value of @xmath201 from our scan over nuhm2 model parameters for points satisfying @xmath34 ( red ) and 30 ( blue ) versus @xmath25 . the various constraints from above , including lhc search and compatibility with @xmath123 , are included .
we also indicate the belle central value and @xmath202 bounds by the dashed and dot - dashed lines , respectively . from the plot
, we see a large deviation between the predicted and measured values of @xmath37 for light @xmath25 values .
especially noteworthy is that _ no _ values of @xmath37 lie within the @xmath202 measured band for @xmath38 gev . recall that this range of stop masses is often considered generally natural @xcite before amending the calculations of @xmath100 and @xmath104 . as @xmath25 increases , the predicted range of @xmath37 rises asymptotically to be within the measured range : this occurs especially for @xmath39 tev .
the intermediate region with @xmath203 tev contains points in agreement with the measured value , where the various @xmath204 amplitudes , which can occur with either positive or negative values , cancel one - with - another .
but even in this region of @xmath25 values , the bulk of points tend to deviate severely from the measured value .
this is because the loop contributions can always be large since the higgsino - like charginos and stops are both light . from examining the confrontation between predicted and measured values of @xmath37
, it comes as no surprise that light stops have yet to be detected at lhc .
in this section , we consider top squark pair production and decay rates at the lhc .
top squark pair production proceeds dominantly through the qcd @xmath205 and @xmath206 annihilation channels .
the nlo production rates for lhc with @xmath207 and 14 tev are calculated using prospino @xcite and shown in fig .
[ fig : sigt1 ] versus top squark mass @xmath25 .
we also show production rates for future proposed @xmath208 colliders operating with @xmath209 , 33 , 50 and 100 tev .
the vertical dashed line shows the approximate locus of the atlas / cms bounds on @xmath25 from searches within the context of simplified models with a low value of @xmath32 .
the dotted vertical line denotes the projected reach of hl - lhc for top - squarks .
we see that the total production cross section for @xmath181 gev at lhc14 are in the @xmath21020 fb range . by moving up to @xmath211 gev , the cross section drops by about an order of magnitude to about @xmath212 fb . at @xmath213 tev
, @xmath214 drops by another order of magnitude to about 0.1 fb .
these total cross sections may be compared to the upper limit on @xmath25 from requiring @xmath125 whereupon @xmath215 tev is required . for such large values of @xmath25 ,
the total cross sections are in the @xmath216 fb range .
probing such massive top squarks will likely require an lhc energy upgrade ( he - lhc with @xmath2733 tev ) or else a future circular collider ( fcc ) with @xmath217100 tev @xcite . in fig .
[ fig : bft1]_a _ ) , we show the expected top squark branching fractions versus @xmath136 along the top squark model line .
the branching fractions are from isajet @xcite . in the plot
, the black curve denotes @xmath218 where for our model line @xmath4 is the lighter , mainly higgsino - like , chargino .
this mode occurs at the @xmath219 rate and is rather model independent ( within the context of natural susy with light higgsinos ) .
the @xmath4 further decays via 3-body mode into @xmath220 where @xmath127 is the higgsino - like lsp .
since @xmath221 ( and @xmath222 ) are @xmath22320 gev , most of the decay energy goes into making the @xmath127 rest mass and is undetected .
the @xmath224 energy is rather soft leading to a few soft tracks .
thus , both the @xmath4 and @xmath225 are only quasi - visible .
meanwhile , the @xmath44-jet from @xmath2 decay may be quite hard , typically in the hundreds of gev .
the red and blue curves denote the @xmath226 and @xmath227 respectively .
both these branching fractions come in at the 2025% level thus covering the bulk of the remaining decays . while the @xmath127 is invisible ( it presumably comprises a portion of the dark matter ) ,
again the @xmath225 and @xmath4 are quasi - visible .
meanwhile , the top quarks are produced at large @xmath228 and also their rest mass leads to energetic decay products .
in addition , there is a non - negligible decay rate @xmath229 where @xmath230 is bino - like and yields visible decays .
these decays occur at the few percent level .
furthermore , @xmath170 decays into wino - like @xmath231 and @xmath232 can occur but at the sub - percent level .
the dip in branching fractions at the center of the plot is due to turn on of @xmath233 . in fig .
[ fig : bft1]_b _ ) , we show the same branching fractions versus @xmath25 along the model line .
the branching fractions are again seen to be rather model independent except for @xmath234 ( in the excluded range ) where the decays into top - quarks become kinematically forbidden .
the branching fractions in this plot are double - valued since certain top squark mass values can occur for both large positive and large negative values of @xmath136 .
these mainly affect the tiny branching fractions into wino - like electroweakinos .
the most direct implication of naturalness is the existence of light higgsinos of mass @xmath235300 gev , the lighter the better .
given these expectations on @xmath26 , the lhc lower bound @xmath21 gev applies and we expect top squarks to lie in the mass range @xmath1813000 gev at little cost to naturalness . this mass range is consistent with expectations from comparing the predicted @xmath37 to its measured value .
then , the highly scrutinized @xmath236 degeneracy rarely if ever applies and we expect instead a rather large @xmath237 mass difference . in this case , the top squark branching fraction predictions from sec . [ sec : prod ] are rather robust : they result over a huge range of nuhm2 parameter space and also under the natural general mirage mediation parameter space found in ref .
we would then expect , quite generally , the following collider signatures to obtain : a. @xmath238 , b. @xmath239 , c. @xmath240 . these signatures should be accompanied by the usual initial state radiation plus perhaps additional semi - soft tracks from associated light higgsino @xmath4 and @xmath225 decays .
the first signal channel a. includes rather hard @xmath44-jets plus hard @xmath241 and should be plagued by backgrounds including @xmath242 production where @xmath243 .
one might create distributions using the @xmath244 variable applied to the @xmath9 final state to try to extract a kinematic upper edge which could yield an estimate of the top squark mass . for signal channel b. , we expect a hard @xmath245-jet along with a hard @xmath44-jet and @xmath241 .
this channel would include @xmath9 along with an added @xmath246 where in the case of hadronic @xmath247 decays , the @xmath247 mass may be reconstructed .
the dominant backgrounds would include @xmath248 production , @xmath249 production and @xmath250 production where @xmath251 and @xmath252 , single top production and @xmath253 production .
this `` mixed top - squark decay channel '' has previously been emphasized by graesser and shelton@xcite .
signal channel c. contains a hard @xmath248 pair plus large @xmath241 .
major backgrounds would include @xmath254 production .
the hard @xmath245-jets may benefit from a top - tagger@xcite .
a credible semi - simplified model could be presented in the @xmath25 vs. @xmath255 mass plane where the several dominant decay branching fractions would be allowed to take place .
physically , this is what is expected to happen and one would then include the dominant mixed decay mode where one @xmath170 decays to @xmath256 while the other decays to @xmath3 . finally , we comment on indirect searches for top squarks at the lhc . since @xmath2573 tev is predicted in the radiatively - driven natural susy , one may expect that its signature can be probed indirectly via the precise measurements of the higgs decay branching ratios , as top squarks affect the @xmath258 and @xmath259 decay channels at one - loop level . as it turns out , however , the deviations of these decay branches from the sm prediction are too small to be detected even at the hl - lhc @xcite .
this observation again leads to the conclusion that future colliders such as ilc or an energy upgraded lhc are required for a thorough coverage of ( just the top - squark sector of ) natural susy .
in this paper we have re - examined the phenomenology of top squarks expected from natural susy . we first noted that older expectations of very light top squarks based on requiring small @xmath260 are technically flawed in that they neglect the contribution of @xmath58 to its own running . by properly including this contribution
, then the @xmath100 measure reduces to @xmath28 .
the @xmath125 requires light higgsinos @xmath261300 gev while much heavier top squarks @xmath2623 tev are allowed at little cost to naturalness . in the latter case , the radiative corrections to @xmath58 aid in driving it from large unnatural high scale values to natural values at the weak scale
a situation known as radiatively - driven natural susy or rns . for the case of bg naturalness ,
if @xmath104 is evaluated in multi - soft - parameter effective theories , then one obtains an overestimate of fine - tuning as compared to the calculation for a more fundamental theory wherein the soft terms are all correlated . in the latter case , @xmath104 reduces to @xmath28 . using @xmath28
, it is found that current lhc top squark search constraints have probed only a fraction of the allowed @xmath25 vs. @xmath32 parameter plane .
the compressed region , which has been heavily searched , admits few or no solutions .
further , values of @xmath38 gev lead to typically large deviations in @xmath37 .
top squark production and decay rates are calculated in natural susy and lead to comparable mixtures of @xmath241 plus @xmath263 , @xmath248 and @xmath264 signatures .
it is emphasized that a semi - simplified model containing the major admissible final states would be most helpful to truly constrain the natural susy parameter space or to discover top squarks .
nonetheless , plenty of perfectly natural susy solutions exist with @xmath25 values well beyond the reach of hl - lhc
. to probe the entire expected natural susy top squark parameter space will likely require an energy upgrade of lhc to the @xmath2733 tev regime . to this end , in fig .
[ fig : bar ] , we show the current exclusion limit on top squark masses from atlas / cms for a light @xmath265200 gev .
we also show the hl - lhc projected reach and exclusion limits for 3000 fb@xmath186 of integrated luminosity @xcite along with the projected reach of future @xmath208 colliders with @xmath266 and 100 tev@xcite .
in contrast to common notions , the display shows that hl - lhc has a very limited reach for natural susy in the top - squark pair production channel .
even if no top - squark signal is seen at hl - lhc , then there will be little impact on excluding natural susy ( other channels such as same - sign diboson or soft dilepton plus jets appear more lucrative to hl - lhc)@xcite .
however , an energy upgrade to he - lhc with @xmath266 tev will have a @xmath185 discovery reach to @xmath267 tev and a 95% cl exclusion reach to 4 tev .
such a reach will either discover or exclude natural susy in the top squark sector .
we also show the snowmass projected reach@xcite for top - squark pairs for a 100 tev collider .
such a machine is projected to probe up to @xmath268 tev .
this reach probes beyond a 33 tev machine only further into unnatural regions of parameter space .
we thank x. tata for comments on the manuscript . this work was supported in part by the us department of energy , office of high energy physics .
99 e. witten , nucl .
b * 188 * , 513 ( 1981 ) ; r. k. kaul , phys .
b * 109 * , 19 ( 1982 ) .
s. dimopoulos , s. raby and f. wilczek , phys .
d * 24 * ( 1981 ) 1681 ; u. amaldi , w. de boer and h. furstenau , phys .
b * 260 * , 447 ( 1991 ) ; j. r. ellis , s. kelley and d. v. nanopoulos , phys . lett .
b * 260 * ( 1991 ) 131 ; p. langacker and m. x. luo , phys .
d * 44 * ( 1991 ) 817 .
l. e. ibaez and g. g. ross , phys . lett .
* b110 * , 215 ( 1982 ) ; k. inoue _
. phys . * 68 * , 927 ( 1982 ) and * 71 * , 413 ( 1984 ) ; l. ibaez , phys . lett . * b118 * , 73 ( 1982 ) ; h. p. nilles , m. srednicki and d. wyler , phys .
b * 120 * ( 1983 ) 346 ; j. ellis , j. hagelin , d. nanopoulos and m. tamvakis , phys .
b125 * , 275 ( 1983 ) ; l. alvarez - gaum .
j. polchinski and m. wise , nucl
* b221 * , 495 ( 1983 ) ; b. a. ovrut and s. raby , phys .
b * 130 * ( 1983 ) 277 ; for a review , see l. e. ibanez and g. g. ross , comptes rendus physique * 8 * ( 2007 ) 1013 .
g. aad _ et al . _ [ atlas collaboration ] , phys .
b * 716 * ( 2012 ) 1 .
s. chatrchyan _ et al . _
[ cms collaboration ] , phys .
b * 716 * ( 2012 ) 30 .
g. aad _ et al . _ [ atlas and cms collaborations ] , phys .
* 114 * , 191803 ( 2015 ) .
h. e. haber and r. hempfling , phys .
* 66 * ( 1991 ) 1815 ; j. r. ellis , g. ridolfi and f. zwirner , phys .
b * 257 * ( 1991 ) 83 ; y. okada , m. yamaguchi and t. yanagida , prog .
* 85 * ( 1991 ) 1 ; for a review , see _ e.g. _
m. s. carena and h. e. haber , prog .
* 50 * ( 2003 ) 63 .
the atlas collaboration , atlas - conf-2016 - 052 .
cms collaboration , cms - pas - sus-16 - 015 .
g. aad _ et al .
_ [ atlas collaboration ] , eur .
j. c * 75 * ( 2015 ) no.10 , 510 erratum : [ eur .
j. c * 76 * ( 2016 ) no.3 , 153 ] .
the atlas collaboration , atlas - conf-2016 - 077 .
cms collaboration , cms - pas - sus-16 - 028 .
n. craig , arxiv:1309.0528 [ hep - ph ] .
h. baer , v. barger , p. huang , a. mustafayev and x. tata , phys .
* 109 * ( 2012 ) 161802 .
h. baer , v. barger , p. huang , d. mickelson , a. mustafayev and x. tata , phys .
d * 87 * ( 2013 ) 11 , 115028 .
d. matalliotakis and h. p. nilles , nucl .
b * 435 * ( 1995 ) 115 ; p.
nath and r. l. arnowitt , phys .
d * 56 * ( 1997 ) 2820 ; j. r. ellis , t. falk , k. a. olive and y. santoso , nucl .
b * 652 * ( 2003 ) 259 ; h. baer , a. mustafayev , s. profumo , a. belyaev and x. tata , jhep * 0507 * ( 2005 ) 065 . i. i. y. bigi and s. rudaz , phys .
* 153b * ( 1985 ) 335 . h. baer and x. tata , phys .
* 167b * ( 1986 ) 241 .
k. i. hikasa and m. kobayashi , phys . rev .
d * 36 * ( 1987 ) 724 .
m. drees and k. i. hikasa , phys .
b * 252 * ( 1990 ) 127 .
h. baer , m. drees , r. godbole , j. f. gunion and x. tata , phys .
d * 44 * ( 1991 ) 725 .
k. a. olive and s. rudaz , phys .
b * 340 * ( 1994 ) 74 .
p. meade and m. reece , phys .
d * 74 * ( 2006 ) 015010 .
m. graesser and j. shelton , jhep * 0906 * ( 2009 ) 039 .
m. drees , m. hanussek and j. s. kim , phys .
d * 86 * , 035024 ( 2012 ) .
j. cao , c. han , l. wu , j. m. yang and y. zhang , jhep * 1211 * ( 2012 ) 039 .
y. bai , h. c. cheng , j. gallicchio and j. gu , jhep * 1207 * ( 2012 ) 110
. z. han , a. katz , d. krohn and m. reece , jhep * 1208 * ( 2012 ) 083 . c. han , k. i. hikasa , l. wu , j. m. yang and y. zhang , jhep * 1310 * ( 2013 ) 216 . y. bai , a. katz and b. tweedie , jhep * 1401 * ( 2014 ) 040 .
d. a. demir and c. s. n , phys .
d * 90 * ( 2014 ) 095015 .
m. czakon , a. mitov , m. papucci , j. t. ruderman and a. weiler , phys .
* 113 * ( 2014 ) no.20 , 201803 .
a. kobakhidze , n. liu , l. wu , j. m. yang and m. zhang , phys .
b * 755 * ( 2016 ) 76 .
h. abe , j. kawamura and y. omura , jhep * 1508 * , 089 ( 2015 ) .
g. belanger , d. ghosh , r. godbole and s. kulkarni , jhep * 1509 * ( 2015 ) 214 . h. an and l. t. wang , phys .
lett . * 115 * ( 2015 ) 181602 .
k. i. hikasa , j. li , l. wu and j. m. yang , phys .
d * 93 * ( 2016 ) no.3 , 035003 .
k. rolbiecki and j. tattersall , phys .
b * 750 * ( 2015 ) 247 .
a. belyaev , v. sanz and m. thomas , jhep * 1601 * ( 2016 ) 102 .
s. macaluso , m. park , d. shih and b. tweedie , jhep * 1603 * ( 2016 ) 151 . j. kawamura and y. omura , phys .
d * 93 * , no . 5 , 055019 ( 2016 ) .
d. gonalves , k. sakurai and m. takeuchi , phys .
d * 94 * , no . 7 , 075009 ( 2016 ) . c. han , m. m. nojiri , m. takeuchi and t. t. yanagida , arxiv:1609.09303 [ hep - ph ] . c. han , j. ren , l. wu , j. m. yang and m. zhang , arxiv:1609.02361 [ hep - ph ] . h. c. cheng , c. gao , l. li and n. a. neill , jhep * 1605 * ( 2016 ) 036 .
m. r. buckley , d. feld , s. macaluso , a. monteux and d. shih , arxiv:1610.08059 [ hep - ph ] .
a. pierce and b. shakya , arxiv:1611.00771 [ hep - ph ] .
y. bai , j. berger , j. osborne and b. a. stefanek , arxiv:1611.05046 [ hep - ph ] .
g. h. duan , k. i. hikasa , l. wu , j. m. yang and m. zhang , arxiv:1611.05211 [ hep - ph ] .
a. cici , z. kirca and c. s. un , arxiv:1611.05270 [ hep - ph ] . h. baer , v. barger , d. mickelson and m. padeffke - kirkland , phys .
d * 89 * , 115019 ( 2014 ) .
h. baer , v. barger and d. mickelson , phys .
d * 88 * ( 2013 ) no.9 , 095013 .
r. kitano and y. nomura , phys .
d * 73 * , 095004 ( 2006 ) ; m. papucci , j. t. ruderman and a. weiler , jhep * 1209 * ( 2012 ) 035 ; k. ishiwata , n. nagata and n. yokozaki , phys .
b * 710 * , 145 ( 2012 ) ; c. brust , a. katz , s. lawrence and r. sundrum , jhep * 1203 * ( 2012 ) 103 ; j. a. evans , y. kats , d. shih and m. j. strassler , jhep * 1407 * ( 2014 ) 101 .
j. r. ellis , k. enqvist , d. v. nanopoulos and f. zwirner , mod .
phys . lett .
a * 1 * , 57 ( 1986 ) .
r. barbieri and g. f. giudice , nucl .
b * 306 * , 63 ( 1988 ) .
l. e. ibanez , c. lopez and c. munoz , _ nucl .
* b 256 * ( 1985 ) 218 ; a. lleyda and c. munoz , _ phys .
* b 317 * ( 1993 ) 82 .
h. abe , t. kobayashi and y. omura , _ phys .
* d 76 * ( 2007 ) 015002 .
s. p. martin , _ phys .
* d 75 * ( 2007 )
115005 . j. l. feng , ann .
nucl . part .
* 63 * ( 2013 ) 351 .
j. l. feng , k. t. matchev and t. moroi , phys .
* 84 * ( 2000 ) 2322 .
s. k. soni and h. a. weldon , phys .
b * 126 * ( 1983 ) 215 ; v. s. kaplunovsky and j. louis , phys .
b * 306 * ( 1993 ) 269 ; a. brignole , l. e. ibanez and c. munoz , nucl .
b * 422 * ( 1994 ) 125 [ erratum - ibid .
b * 436 * ( 1995 ) 747 ] .
h. baer and x. tata , `` weak scale supersymmetry : from superfields to scattering events , '' cambridge , uk : univ .
( 2006 ) 537 p. f. e. paige , s. d. protopopescu , h. baer and x. tata , hep - ph/0312045 .
k. l. chan , u. chattopadhyay and p. nath , phys .
d * 58 * ( 1998 ) 096004 ; s. akula , m. liu , p. nath and g. peim , phys .
b * 709 * ( 2012 ) 192 ; m. liu and p. nath , phys .
d * 87 * ( 2013 ) 9 , 095012 .
h. baer , v. barger and m. savoy , phys .
d * 93 * ( 2016 ) no.3 , 035016 .
k. j. bae , h. baer and e. j. chun , phys .
d * 89 * ( 2014 ) no.3 , 031701 .
h. baer , j. sender and x. tata , phys .
d * 50 * ( 1994 ) 4517 .
the atlas collaboration , atlas - conf-2016 - 050 .
the atlas collaboration , atlas - conf-2016 - 076 .
m. aaboud _ et al . _ [ atlas collaboration ] , phys .
d * 94 * ( 2016 ) no.3 , 032005 .
cms collaboration , cms - pas - sus-16 - 029 .
atlas collaboration , atlas - phys - pub-2013 - 011 .
j. s. kim , k. rolbiecki , r. ruiz , j. tattersall and t. weber , arxiv:1606.06738 [ hep - ph ] .
s. bertolini , f. borzumati , a. masiero and g. ridolfi , nucl .
b * 353 * ( 1991 ) 591 .
j. l. hewett , phys .
* 70 * ( 1993 ) 1045 ; v. d. barger , m. s. berger and r. j. n. phillips , phys .
* 70 * ( 1993 ) 1368 .
m. misiak _ et al .
_ , phys .
* 114 * ( 2015 ) no.22 , 221801 ; m. czakon , p. fiedler , t. huber , m. misiak , t. schutzmeier and m. steinhauser , jhep * 1504 * ( 2015 ) 168 .
a. abdesselam _ et al .
_ [ belle collaboration ] , arxiv:1608.02344 [ hep - ex ] .
h. baer and m. brhlik , phys .
d * 55 * ( 1997 ) 3201 ; h. baer , m. brhlik , d. castano and x. tata , phys .
d * 58 * ( 1998 ) 015007 .
w. beenakker , r. hopker and m. spira , hep - ph/9611232 .
t. cohen , r. t. dagnolo , m. hance , h. k. lou and j. g. wacker , jhep * 1411 * ( 2014 ) 021 . n. arkani - hamed , t. han , m. mangano and l. t. wang , phys . rept .
* 652 * ( 2016 ) 1 .
t. golling _ et al .
_ , arxiv:1606.00947 [ hep - ph ] .
h. baer , v. barger , h. serce and x. tata , arxiv:1610.06205 [ hep - ph ] .
m. l. graesser and j. shelton , phys .
* 111 * ( 2013 ) no.12 , 121802 .
d. e. kaplan , k. rehermann , m. d. schwartz and b. tweedie , phys .
* 101 * , 142001 ( 2008 ) ; j. thaler and l. t. wang , jhep * 0807 * , 092 ( 2008 ) ; l. g. almeida , s. j. lee , g. perez , i. sung and j. virzi , phys .
d * 79 * , 074012 ( 2009 ) ; t. plehn , g. p. salam and m. spannowsky , phys .
* 104 * , 111801 ( 2010 ) ; t. plehn , m. spannowsky , m. takeuchi and d. zerwas , jhep * 1010 * , 078 ( 2010 ) ; v. barger and p. huang , phys . lett .
b * 708 * ( 2012 ) 296 .
k. j. bae , h. baer , n. nagata and h. serce , phys .
d * 92 * , no . 3 , 035006 ( 2015 ) . y. gershtein _ et al .
_ , arxiv:1311.0299 [ hep - ex ] . h. baer , v. barger , m. savoy and x. tata , phys . rev .
d * 94 * ( 2016 ) no.3 , 035025 . | in supersymmetric models with radiatively - driven naturalness and light higgsinos , the top squarks may lie in the 0.53 tev range and thus only a fraction of natural parameter space is accessible to lhc searches .
we outline the range of top squark and lightest susy particle masses preferred by electroweak naturalness in the standard parameter space plane .
we note that the branching fraction for @xmath0 decay favors top squarks much heavier than 500 gev .
such a range of top - squark mass values is in contrast to previous expectations where @xmath1 gev had been considered natural . in radiative natural susy ,
top squarks decay roughly equally via @xmath2 and @xmath3 where @xmath4 and @xmath5 are higgsino - like electroweak - inos .
thus , top squark pair production should yield all of @xmath6 , @xmath7 , @xmath8 and @xmath9 signatures at comparable rates .
we propose that future lhc top squark searches take place within a semi - simplified model which corresponds more closely to expectations from theory .
ou - hep-161104 + ut-16 - 32 + * a top - squark hunter s guide * + howard baer@xmath10 , vernon barger@xmath11 , natsumi nagata@xmath12 and michael savoy@xmath13 + _ @xmath13department of physics and astronomy , + university of oklahoma , norman , ok 73019 , usa + _ _ @xmath14department of physics and astronomy , + university of wisconsin , madison , wi 53706 , usa + _ _ @xmath15department of physics , university of tokyo , bunkyo - ku , tokyo 113 - 0033 , japan + _ |
there are several motives to find the explicit form of conserved densities of differential - difference equations ( ddes ) .
the first few conservation laws have a physical meaning , such as conservation of momentum and energy .
additional ones facilitate the study of both quantitative and qualitative properties of solutions @xcite .
furthermore , the existence of a sequence of conserved densities predicts integrability .
yet , the nonexistence of polynomial conserved quantities does not preclude integrability .
indeed , integrable ddes could be disguised with a coordinate transformation in ddes that no longer admit conserved densities of polynomial type @xcite .
another compelling argument relates to the numerical solution of partial differential equations ( pdes ) . in solving integrable discretizations of pdes
, one should check that conserved quantities indeed remain constant .
in particular , the conservation of a positive definite quadratic quantity may prevent nonlinear instabilities in the numerical scheme .
the use of conservation laws in pde solvers is well - documented in the literature @xcite .
several methods to test the integrability of ddes and for solving them are available .
solution methods include symmetry reduction @xcite and solving the spectral problem @xcite on the lattice . adaptations of the singularity confinement approach @xcite , the wahlquist - estabrook method @xcite , and symmetry techniques @xcite allow one to investigate integrability .
the most comprehensive integrability study of nonlinear ddes was done by yamilov and co - workers ( see e.g. @xcite ) .
their papers provide a classification of semi - discrete equations possessing infinitely many local conservation laws . using the formal symmetry approach ,
they derive the necessary and sufficient conditions for the existence of local conservation laws , and provide an algorithm to construct them . in @xcite
, we gave an algorithm to compute conserved densities for systems of nonlinear evolution equations .
the algorithm is based on the concept of invariance of equations under dilation symmetries .
therefore , it is inherently limited to polynomial densities of polynomial systems .
recently an extension of the algorithm towards ddes was outlined in @xcite . here
we present a full description of our algorithm , which can be implemented in any computer algebra language .
we also provide details about our software package * diffdens.m * @xcite in _ mathematica _ @xcite , which automates the tedious computation of closed - form expressions for conserved densities . following basic definitions in section 2 ,
the algorithm is given in section 3 . to illustrate the method we use the toda lattice @xcite , a parameterized toda lattice @xcite , and the discretized nonlinear schrdinger ( nls ) equation @xcite . in section 4
, we list results for an extended lotka - volterra equation @xcite , a discretized modified kdv equation @xcite , a network equation @xcite , and some other lattices @xcite .
the features , scope and limitations of our code * diffdens.m * are described in section 5 , together with instructions for the user . in section 6
, we draw some conclusions .
consider a system of ddes which is continuous in time , and discretized in the single space variable , @xmath0 where @xmath1 and @xmath2 are vector dynamical variables with any number of components .
for simplicity of notation , the components of @xmath3 will be denoted by @xmath4 etc .
we assume that @xmath2 is polynomial with constant coefficients .
if ddes are non - polynomial or of higher order in @xmath5 , we assume that they can be recast in the form ( [ multisys ] ) . a local _ conservation law _
is defined by @xmath6 which is satisfied on all solutions of ( [ multisys ] ) .
the function @xmath7 is the _ conserved density _ and @xmath8 is the associated _ flux .
_ both are assumed to be polynomials in @xmath9 and its shifts . obviously , @xmath10 and this telescopic series vanishes for a bounded periodic lattice or a bounded lattice resting at infinity . in that case , @xmath11 is constant in time .
so , we have a conserved quantity .
consider the one - dimensional toda lattice @xcite @xmath12 where @xmath13 is the displacement from equilibrium of the @xmath14th unit mass under an exponential decaying interaction force between nearest neighbors . with the change of variables , @xmath15 lattice ( [ orgtoda ] ) can be written in polynomial form @xmath16 system ( [ todalatt ] ) is completely integrable @xcite .
the first two density - flux pairs @xmath17 can be easily be computed by hand .
we introduce a few concepts that will be used in our algorithm .
let @xmath18 denote the _ shift - down _ operator and @xmath19 the _ shift - up _ operator .
both are defined on the set of all monomials in @xmath9 and its shifts .
if @xmath20 is such a monomial then @xmath21 and @xmath22 for example , @xmath23 and @xmath24 it is easy to verify that compositions of @xmath18 and @xmath19 define an _ equivalence relation _ on monomials .
simply stated , all shifted monomials are _ equivalent _ , e.g. @xmath25 this equivalence relation holds for any function of the dependent variables , but for the construction of conserved densities we will apply it only to monomials . in the algorithm
we will use the following _ equivalence criterion _ : if two monomials @xmath26 and @xmath27 are equivalent , @xmath28 then @xmath29 $ ] for some polynomial @xmath30 that depends on @xmath9 and its shifts .
for example , @xmath31 since @xmath32 = u_{n-1 } u_{n+1 } + [ m_n - m_{n+1 } ] , $ ] with @xmath33 the _ main _ representative of an equivalence class is the monomial of that class with @xmath34 as _ lowest _ label on @xmath35 ( or @xmath36 ) . for example , @xmath37 is the main representative of the class with elements @xmath38 etc .
lexicographical ordering is used to resolve conflicts .
for example , @xmath39 ( not @xmath40 ) is the main representative in the class with elements @xmath41 etc .
scaling invariance , which results from a special lie - point symmetry , is an intrinsic property of many integrable nonlinear pdes and ddes . indeed , observe that ( [ todalatt ] ) , and the couples @xmath42 and @xmath43 in ( [ todapair ] ) ( after inserting them in ( [ conslaw ] ) ) , are invariant under the dilation ( scaling ) symmetry @xmath44 where @xmath45 is an arbitrary parameter .
stated differently , @xmath46 corresponds to one derivative with respect to @xmath47 denoted by @xmath48 similarly , @xmath49 our three - step algorithm exploits the symmetry ( [ symtoda ] ) to find conserved densities .
_ step 1 : determine the weights of variables _ the _ weight _ , @xmath50 of a variable is by definition equal to the number of derivatives with respect to @xmath51 the variable carries .
weights are non - negative , rational , and independent of @xmath52 without loss of generality we set @xmath53 the _ rank _ of a monomial is defined as the total weight of the monomial .
for now we assume that all the terms ( monomials ) in a particular equation have the same rank .
this property is called _ uniformity in rank_. the uniformity in rank condition leads to a system of equations for the unknown weights .
different equations in the vector equation ( [ multisys ] ) may have different ranks . requiring uniformity in rank for each equation in ( [ todalatt ] )
allows one to compute the weights of the dependent variables .
indeed , @xmath54 yields @xmath55 which is consistent with ( [ symtoda ] ) .
_ step 2 : construct the form of the density _
this step involves finding the building blocks ( monomials ) of a polynomial density with prescribed rank @xmath56 .
all terms in the density must have the same rank @xmath56 . since we may introduce parameters with weights ( see example 6 )
, the fact that the density will be a sum of monomials of uniform rank does not necessarily imply that the density must be uniform in rank with respect to the dependent variables .
let @xmath57 be the list of all the variables with positive weights , including parameters with weight .
the following procedure is used to determine the form of the density of rank @xmath56 : * form the set @xmath58 of all monomials of rank @xmath56 or less by taking all appropriate combinations of different powers of the variables in @xmath57 . * for each monomial in @xmath58 , introduce the appropriate number of derivatives with respect to @xmath5 so that all the monomials exactly have weight @xmath56 . gather in set @xmath59 all the terms that result from computing the various derivatives . *
identify the monomials that belong to the same equivalence classes and replace them by the main representatives . call the resulting simplified set @xmath60 , which consists of the _ building blocks _ of the density with desired rank @xmath56 .
* linear combination of the elements in @xmath61 with constant coefficients @xmath62 gives the form of polynomial density of rank @xmath63 continuing with ( [ todalatt ] ) , we compute the form of the density of rank @xmath64 from @xmath65 we build @xmath66 next , introduce @xmath5-derivatives , so that each term exactly has rank @xmath67 thus , using ( [ todalatt ] ) , @xmath68 gather the resulting terms in set @xmath69 since @xmath70 and @xmath71 in @xmath59 , @xmath72 is replaced by @xmath73 and @xmath74 by @xmath75 hence , we obtain @xmath76 linear combination of the monomials in @xmath61 gives the form of the density : @xmath77 _ step 3 : determine the unknown coefficients in the density _
the following procedure simultaneously determines the constants @xmath62 and the form of the flux @xmath8 : * compute @xmath78 and use ( [ multisys ] ) to remove all the @xmath51-derivatives . * regarding ( [ conslaw ] ) , the resulting expression must match the pattern @xmath79 .
use the equivalence criterion to modify @xmath78 .
the goal is to introduce the main representatives and to identify the terms that match the pattern . *
the part that does not match the pattern must vanish identically for any combination of the components of @xmath9 and their shifts .
this leads to a _ linear _ system @xmath80 in the unknowns @xmath62 .
if @xmath80 has parameters , careful analysis leads to conditions on these parameters guaranteeing the existence of densities .
see @xcite for a description of this compatibility analysis .
* the flux is the first piece in the pattern @xmath81.$ ] carrying on with ( [ todalatt ] ) , we determine the coefficients @xmath82 through @xmath83 in ( [ formrho3toda ] ) by requiring that ( [ conslaw ] ) holds .
simultaneously , we determine @xmath84 compute @xmath78 using ( [ formrho3toda ] ) .
use ( [ todalatt ] ) to remove @xmath85 etc .
after grouping the terms @xmath86 use the equivalence criterion to modify @xmath87 .
for instance , replace @xmath88 by @xmath89 $ ] . in terms of main representatives ,
\nonumber \\ & & + c_2 u_{n } u_{n+1 } v_{n } + [ c_2 u_{n-1 } u_n v_{n-1 } - c_2 u_{n } u_{n+1 } v_{n } ] \nonumber \\ & & + c_2 v_{n}^2 + [ c_2 v_{n-1}^2 - c_2 v_{n}^2 ] - c_3 u_n u_{n+1 } v_n - c_3 v_n^2 . \end{aligned}\ ] ] next , group the terms outside of the square brackets and move the pairs inside the square brackets to the bottom . rearrange the latter terms so that they match the pattern @xmath91 $ ] .
hence , @xmath92.\end{aligned}\ ] ] the first piece inside the square brackets determines @xmath93 the terms outside the square brackets must all vanish piece by piece , yielding @xmath94 the solution is @xmath95 since densities can only be determined up to a multiplicative constant , we choose @xmath96 and substitute this into ( [ formrho3toda ] ) and ( [ formflux3toda ] ) .
the explicit forms of the density and the flux follow : @xmath97 to illustrate how the algorithm works for ddes with parameters , consider the parameterized toda lattice @xmath98 where @xmath99 and @xmath100 are _ nonzero _ parameters without weight .
in @xcite it was shown that ( [ partodalatt ] ) is completely integrable if and only if @xmath101 ; but then ( [ partodalatt ] ) is ( [ todalatt ] ) . using our algorithm , one can easily compute the _ compatibility conditions _ for @xmath99 and @xmath100 , so that ( [ partodalatt ] ) admits a polynomial conserved densities of , say , rank 3 .
the steps are the same as for ( [ todalatt ] )
. however , ( [ todasystem ] ) must be replaced by @xmath102 a non - trivial solution @xmath103 will exist if and only if @xmath104 analogously , ( [ partodalatt ] ) has density @xmath105 of rank 1 if @xmath106 and density @xmath107 of rank 2 if @xmath108 only when @xmath109 will ( [ partodalatt ] ) have conserved densities of rank @xmath110 3 : @xmath111 ignoring irrelevant shifts in @xmath112 these densities agree with the results in @xcite . in @xcite , ablowitz and ladik studied the following integrable discretization of the nls equation : @xmath113 where @xmath114 is the complex conjugate of @xmath115 instead of splitting @xmath46 into its real and imaginary parts , we treat @xmath46 and @xmath116 as independent variables and augment ( [ orgabllad ] ) with its complex conjugate . absorbing @xmath117 in the scale on @xmath118 @xmath119 since @xmath120 we have @xmath121 neither of the equations in ( [ abllad ] ) is uniform in rank . to circumvent this problem
we introduce an auxiliary parameter @xmath99 with weight , and replace ( [ abllad ] ) by @xmath122 this extra freedom allows us to impose uniformity in rank : @xmath123 which yields @xmath124 or ,
@xmath125 we show how to get the building blocks of the density of rank @xmath126 . in this case @xmath127 and @xmath128 the monomials @xmath129 and @xmath130 have already rank @xmath126 , so no derivatives are needed .
the monomials @xmath46 and @xmath131 will have rank @xmath126 after introducing @xmath132 .
there is no way for the remaining monomials @xmath133 and @xmath134 to have rank @xmath126 after differentiation with respect to @xmath5 .
therefore , they are rejected .
the remaining intermediate steps lead to @xmath135 although uniformity in rank is essential for the first two steps of the algorithm , after the second step , we may already set @xmath136 the computations now proceed as in the previous example .
the trick of introducing one or more extra parameters with weights can always be attempted if any equation in ( [ multisys ] ) lacks uniformity in rank .
we list some conserved densities of ( [ abllad ] ) : @xmath137 \nonumber \\ \!\!\!\!\!\!\!\!\!\!\!\!\ ! & + & c_2 [ { \textstyle{1 \over 3 } } u_n^3 v_{n+1}^3 + u_n u_{n+1 } v_{n+1 } v_{n+2 } ( u_n v_{n+1 } + u_{n+1 } v_{n+2 } + u_{n+2 } v_{n+3 } ) \nonumber \\ \!\!\!\!\!\!\!\!\!\ ! & + & \ !
u_n v_{n+2 } ( u_n v_{n+1 } \!+\
! u_{n+1 } v_{n+2 } ) \!+\
! u_n v_{n+3 } ( u_{n+1 } v_{n+1 } \!+\
! u_{n+2 } v_{n+2 } ) \!+\ !
u_n v_{n+3}].\end{aligned}\ ] ] our results confirm those in @xcite .
also , if defined on an infinite interval , ( [ orgabllad ] ) admits infinitely many independent conserved densities @xcite . although it is a constant of motion , we can not find the hamiltonian of ( [ orgabllad ] ) , @xmath138,\ ] ] for it has a logarithmic term @xcite
in @xcite , hu and bullough considered an extended version of the lotka - volterra equation : @xmath139 for @xmath140 ( [ extvolterra ] ) is the well - known lotka - volterra equation , for which the densities were presented in @xcite .
we computed five densities of ( [ extvolterra ] ) for the case @xmath141 .
the first three are : @xmath142 for @xmath143 , we also computed five densities of ( [ extvolterra ] ) .
the first three are : @xmath144 we computed four densities of ( [ extvolterra ] ) for @xmath145 to save space we do not list them here .
the integrability and other properties of ( [ extvolterra ] ) are discussed in @xcite . in @xcite , we found the following integrable discretization of the mkdv equation : @xmath146 we computed four densities of ( [ discmkdv ] ) . the first three are : @xmath147 the integrable nonlinear self - dual network equations @xcite can be written as : @xmath148 we computed the first four densities of ( [ dual ] )
. the first three are @xmath149 shabat and yamilov
@xcite studied the following integrable volterra lattice : @xmath150 with our program we computed the first four densities for this system : @xmath151 in @xcite , the following hamiltonian lattice is also listed : @xmath152 four densities of ( [ ravilsys ] ) are : @xmath153
we describe the features , scope and limitations of our program * diffdens.m * , which is written in _ mathematica _ syntax @xcite .
the program has its own _ menu _ interface with a dozen data files .
users should have access to _ mathematica_. the code * diffdens.m * and the data files @xcite must be in the same directory . after launching _ mathematica _ , type .... in[1]:=
< < diffdens.m .... to read in the code * diffdens.m * and start the program . via its menu interface , the program will prompt the user for answers .
the density is available at the end of the computations . to view it in standard _
notation , type ` rho ` . to display it in a more elegant subscript notation , type ` subscriptform[rho ] ` . to test systems that are not in the menu
, one should prepare a data file in the format of the data files that are provided with the code .
of course , the name for a new data file should not coincide with any name already listed in the menu , unless one intended to modify those data files .
for the parameterized toda lattice ( [ partodalatt ] ) the data file reads : .... ( * start of data file d_ptoda.m with parameters . * ) ( * toda lattice with parameters aa and bb * ) u[1][n_]'[t]:= aa*u[2][n-1][t]-u[2][n][t ] ; u[2][n_]'[t]:= u[2][n][t]*(bb*u[1][n][t]-u[1][n+1][t ] ) ; noeqs = 2 ; name = " toda lattice ( parameterized ) " ; parameters = { aa , bb } ; weightpars = { } ; ( * the user may give the rank of rho * ) ( * and a name for the output file . * ) ( * rhorank = 3 ; * ) ( * myfile = " ptodar3.o " ; * ) ( * the user can give the weights of u[1 ] and u[2 ] , * ) ( * and of the parameters with weight ( if applicable ) . * ) ( * weightu[1 ] = 1 ; weightu[2 ] = 2 ; weight[aa ] = 1 ; * ) formrho = 0 ; ( * the user can give the form of rho .
* ) ( * for example , for the density of rank 3 : * ) ( * formrho = c[1]*u[1][n][t]^3+c[2]*u[1][n][t]*u[2][n-1][t]+ c[3]*u[1][n][t]*u[2][n][t ] ; * ) ( * end of data file d_ptoda.m * ) .... a brief explanation of the lines in the data file now follows . `
u[i][n_]'[t ] : = ... ; ` give the @xmath154 equation of the system in _ mathematica _ notation .
1.5pt ` noeqs = 2 ; ` specify the number of equations in the system .
1.5pt ` name = " toda lattice ( parameterized ) " ; ` pick a short name for the system under investigation .
the quotes are essential .
1.5pt ` parameters = { aa , bb } ; ` give the list of parameters in the system . if none , set _ parameters = \ { } _ .
1.5pt ` weightpars = { } ; ` give the list of the parameters that are assumed to have weights .
note that weighted parameters are _ not _ listed in _
parameters _ , the latter is the list of parameters without weight .
1.5pt ` ( * rhorank = 3 ; * ) ` optional .
give the desired rank of the density , if less interactive use of the program is preferred ( batch mode ) .
1.5pt ` ( * myfile = " ptodar3.o " ; * ) ` optional . give a name of the output file , again to bypass interaction with the program .
1.5pt ` ( * weightu[1 ] = 1 ; weightu[2 ] = 2 ; * ) ` optional .
specify a choice for _ some or all _ of the weights .
the program then skips the computation of the weights , but still checks for consistency .
particularly useful if there are several free weights and non - interactive use is preferred .
1.5pt ` formrho = 0 ; ` if _ formrho _ is set to zero , the program will _ compute _ the form of @xmath155 1.5pt .... formrho = c[1]*u[1][n][t]^3+c[2]*u[1][n][t]*u[2][n-1][t]+ c[3]*u[1][n][t]*u[2][n][t ] ; .... alternatively , one could give a form of @xmath156 ( here of rank 3 ) .
the density must be given in expanded form and with coefficients c[i ] .
if form of @xmath156 is given , the program skips both the computation of the weights and the form of the density .
instead , the code uses what is given and computes the coefficients c[i ] .
this option allows one , for example , to test densities from the literature .
1.5pt anything within ` ( * ` and ` * ) ` is a comment and ignored by _
mathematica_. once the data file is ready , pick the choice ` tt ) your system ` " in the menu .
our program can handle systems of first order ddes that are polynomial in the dependent variables . only one independent variable ( @xmath5 )
is allowed .
no terms in the ddes should have coefficients that _ explicitly _ depend on @xmath5 or @xmath34 .
the program only computes polynomial conserved densities in the dependent variables and their shifts , without explicit dependencies on @xmath5 or @xmath34 .
theoretically , there is no limit on the number of ddes . in practice ,
for large systems , the computations may take a long time or require a lot of memory .
the computational speed depends primarily on the amount of memory . by design ,
the program constructs only densities that are uniform in rank .
the uniform rank assumption for the monomials in @xmath157 allows one to compute _ independent _ conserved densities piece by piece , without having to split linear combinations of conserved densities . due to the superposition principle ,
a linear combination of conserved densities of unequal rank is still a conserved density .
this situation arises frequently when parameters with weight are introduced in the ddes .
the input systems can have one or more parameters , which are assumed to be nonzero .
if a system has parameters , the program will attempt to compute the compatibility conditions for these parameters such that densities ( of a given rank ) exist .
the assumption that all parameters in the given dde must be nonzero is essential . as a result of setting parameters to zero in a given system of ddes , the weights and the rank of @xmath157 might change .
in general , the compatibility conditions for the parameters could be highly nonlinear , and there is no general algorithm to solve them .
the program automatically generates the compatibility conditions , and solves them for parameters that occur linearly .
grbner basis techniques could be used to reduce complicated nonlinear systems into equivalent , yet simpler , non - linear systems . for ddes with parameters and
when the linear system for the unknown coefficients @xmath62 has many equations , the program saves that system and its coefficient matrix , etc .
, in the file _
worklog.m_. independent from the program , the worklog files can later be analyzed with appropriate _ mathematica _
functions .
the assumption that the ddes are uniform in rank is not very restrictive .
if the uniform rank condition is violated , the user can introduce one or more parameters with weights .
this also allows for some flexibility in the form of the densities .
although built up with terms that are uniform in rank in the dependent variables and parameters , the densities do no longer have to be uniform in rank with respect to the dependent variables alone .
conversely , when the uniform rank condition _ is _ fulfilled , the introduction of extra parameters ( with weights ) in the given dde leads to a linear combination of conservation laws , not to new ones . in cases where it is not clear
whether or not parameters with weight should be introduced , one should start with parameters without weight .
if this causes incompatibilities in the assignment of weights ( due to non - uniformity ) , the program may provide a suggestion .
quite often , it recommends that one or more parameters be moved from the list _ parameters _ into the list _ weightpars _ of weighted parameters . for systems with two or more free weights , the user will be prompted to enter values for the free weights .
if only one weight is free , the program will automatically compute some choices for the free weight , and continue with the lowest integer or fractional value .
the program selects this value for the free weight ; it is just one choice out of possibly infinitely many .
if the algorithm fails to determine a suitable value , the user will be prompted to enter a value for the free weight .
negative weights are not allowed .
zero weights are allowed , but at least one of the dependent variables must have positive weight .
the code checks this requirement , and if it is violated the computations are aborted .
note that _ fractional weights _ and densities of _ fractional rank _ are permitted .
our program is a tool in the search of the first half - dozen conservation laws .
an existence proof ( showing that there are indeed infinitely many conservation laws ) must be done analytically .
if our program succeeds finding a large set of independent conservation laws , there is a good chance that the system of ddes has infinitely many conserved densities .
if the number of conserved densities is 3 or less , the dde may have other than polynomial conserved densities , or may not be integrable ( in the chosen coordinate representation ) .
we offer the scientific community a _ mathematica _ package to carry out the tedious calculations of conserved densities for systems of nonlinear ddes .
the code * diffdens.m * , together with several data and output files , is available via internet url @xcite .
the existence of a large number of conservation laws is an indicator of integrability of the lattice .
therefore , by generating the compatibility conditions , one can analyze classes of parameterized ddes and filter out the candidates for complete integrability .
nal gkta thanks wolfram research , inc . for an internship .
we acknowledge helpful comments from drs .
b. herbst , s. mikhailov , w .- h .
steeb , y. suris , p. winternitz , and r. yamilov .
we also thank g. erdmann for his help with this project .
hohler and k. olaussen , int .
a 11 ( 1996 ) 1831 .
shabat and r.i .
yamilov , leningrad math .
j. 2 ( 1991 ) 377 .
hickernell , stud .
appl . math .
69 ( 1983 ) 23 .
leveque , numerical methods for conservation laws ( birkhuser verlag , basel , 1992 ) .
sanz - serna , j. comput .
( 1982 ) 199 .
d. levi and p. winternitz , j. math .
34 ( 1993 ) 3713 .
d. levi and o. ragnisco , lett .
nuovo cimento 22 ( 1978 ) 691 .
a. ramani , b. grammaticos and k.m .
tamizhmani , j. phys . a : math . gen .
25 ( 1992 ) l883 .
b. deconinck , phys .
lett . a 223 ( 1996 ) 45 .
i. cherdantsev and r. yamilov , in : symmetries and integrability of difference equations , eds .
d. levi , l. vinet and p. winternitz ( american mathematical society , providence , rhode island , 1996 ) 51 .
d. levi and r. yamilov , j. math .
38 ( 1997 ) 6648 .
v.v . sokolov and a.b .
shabat , in : soviet sci .
4 ( harwood academic publishers , new york , 1984 ) 221 .
r. yamilov , in : proc .
workshop on nonlinear evolution equations and dynamical systems ( world scientific publishing , singapore , 1993 ) 423 .
. gkta and w. hereman , j. symb .
( 1997 ) 591 .
. gkta and w. hereman .
the program condens.m is available via internet url : http://www.mines.edu / fs_home / whereman/. m.j .
ablowitz and j.f .
ladik , j. math .
17 ( 1976 ) 1011 .
ablowitz and j.f .
ladik , stud .
55 ( 1976 ) 213 .
ablowitz and b.m .
herbst , siam j. appl .
50 ( 1990 ) 339 .
hu and r.k .
bullough , j. phys .
a : math . gen .
30 ( 1997 ) 3635 .
ablowitz and p.a .
clarkson , solitons , nonlinear evolution equations and inverse scattering ( cambridge university press , cambridge , u.k . , 1991 ) . | an algorithm to compute polynomial conserved densities of polynomial nonlinear lattices is presented .
the algorithm is implemented in _
mathematica _ and can be used as an automated integrability test . with the code * diffdens.m * , conserved densities are obtained for several well - known lattice equations . for systems with parameters ,
the code allows one to determine the conditions on these parameters so that a sequence of conservation laws exist . , conservation law ; integrability ; semi - discrete ; lattice |
the error function @xmath4 is defined by @xcite @xmath5 and its inverse @xmath6 , which we will denote by @xmath7 satisfies @xmath8 = \operatorname{erf}\left [ \mathfrak{i}(x)\right ] = x.$ ] the function @xmath9 appears in several problems of heat conduction @xcite . in @xcite
we considered the function @xmath10 and its inverse @xmath11 satisfying @xmath12 = n\left [ s(x)\right ] = x.\ ] ] it is clear from ( [ erf ] ) and ( [ n ] ) that@xmath13\ ] ] and therefore@xmath14 in @xcite we showed that@xmath15 \label{ds}\ ] ] and @xmath16 where @xmath2 is a polynomial of degree @xmath17 satisfying the recurrence @xmath18 the same approach was employed by carlitz in @xcite . from ( [ 2.3 ] ) , it follows easily that for a fixed value of @xmath17 @xmath19 from ( [ sj ] ) and ( [ sn ] ) , we conclude that@xmath20 since@xmath21 we have@xmath22 it follows from ( [ ider ] ) that estimating @xmath23 for large values of @xmath17 is equivalent to finding an asymptotic approximation of the polynomials @xmath2 when @xmath24 .
the objective of this work is to study @xmath2 asymptotically as @xmath1 for various ranges of @xmath25 we shall obtain different asymptotic expansions for @xmath1 and ( i ) @xmath26 ( ii ) @xmath27 and ( iii ) @xmath28 the paper is organized as follows : in section [ section2 ] we approach the problem using a singularity analysis of the generating function @xcite of the polynomials @xmath29 in section [ section3 ] we apply the wkb method to the differential - difference equation ( [ 2.3 ] ) . in @xcite
, we used this approach in the asymptotic analysis of computer science problems and in @xcite to study the krawtchouk polynomials .
finally , in section [ section4 ] we analyze ( [ 2.3 ] ) again using the ray method @xcite and obtain an asymptotic approximation valid in various regions of the @xmath30 domain . in @xcite , @xcite , @xcite , we employed the same technique to analyze asymptotically other families of polynomials and in @xcite , @xcite to study some queueing problems .
in @xcite we obtained the exponential generating function @xmath31-\frac{x^{2}}{2}\right\ } , \ ] ] which implies that@xmath32\right\ } \frac{dz}{z^{n+1}},\ ] ] where the integration contour is a small loop around the origin in the complex plane . using ( [ ds ] ) , we have@xmath33\frac { dz}{z^{n+1}}\\ & = e^{-x^{2}/2}\frac{1}{\sqrt{2\pi}n^{\prime}(x)}\frac{n!}{2\pi\mathrm{i}}\oint\limits_{\left\vert z\right\vert < r}\frac{1}{z^{n+1}}ds[n(x)+zn^{\prime } ( x)]\\ & = \frac{n!}{2\pi\mathrm{i}}\oint\limits_{\left\vert z\right\vert < r}\frac{n+1}{z^{n+2}}s[n(x)+zn^{\prime}(x)]dz\end{aligned}\ ] ] and therefore@xmath34\frac{dz}{z^{n+2}}. \label{pn1}\ ] ] since @xmath35 has singularities at @xmath24 and @xmath36 we consider the functions@xmath37 we have@xmath38 and@xmath39 changing variables to @xmath40 n^{\prime}(x)\ ] ] in ( [ pn1 ] ) , we obtain@xmath41 ^{-(n+2)}dw,\ ] ] or , @xmath42 ^{n+2}}dw , \label{contour}\ ] ] where @xmath43 is a small loop about @xmath44 in the complex plane , with @xmath45 to expand ( [ contour ] ) for @xmath1 with a fixed @xmath46 we employ singularity analysis .
the function @xmath47 has singularities at @xmath48 by ( [ erf ] ) , we have@xmath49 , \quad\mathcal{j}\rightarrow\infty,\ ] ] so that@xmath50 and by symmetry we have@xmath51 the integrand in ( [ contour ] ) thus has singularities at @xmath52 and @xmath53 but for @xmath54 the former is closer to @xmath55 we expand ( [ contour ] ) around @xmath52 by setting @xmath56 and using@xmath57 ^{-(n+2)}\sim\left [ z_{1}(x)\right ] ^{-(n+2)}\exp\left [ -\sqrt{\frac{\pi } { 2}}e^{x^{2}/2}\frac{\delta}{z_{1}(x)}\right ] .
\label{13}\ ] ] then , we deform the contour @xmath43 in ( [ contour ] ) to a new contour @xmath58 that encircles the branch point at @xmath52 ( see figure [ fig12 ] ) .
this leads to@xmath59 ^{-(n+2)}\frac{1}{n}\label{14}\\ & \times\frac{1}{2\pi\mathrm{i}}\int\limits_{0}^{\infty}\left ( \upsilon ^{+}-\upsilon^{-}\right ) \exp\left [ -\sqrt{\frac{\pi}{2}}\frac{e^{x^{2}/2}}{z_{1}(x)}\delta\right ] d\delta,\nonumber\end{aligned}\ ] ] where@xmath60 here @xmath61 corresponds to the approximation of @xmath62 for @xmath63 above or below the right branch cut in figure [ fig12 ] .
for @xmath17 large we have@xmath64 and then evaluating the elementary integral in ( [ 14 ] ) leads to @xmath65 as @xmath1 with@xmath66 ^{-\left ( n+1\right ) } = \frac{n!}{\sqrt{2\ln\left ( n\right ) } } \left [ \frac{e^{-x^{2}/2}}{\zeta\left ( x\right ) } \right ] ^{n+1}\label{15}\ ] ] and @xmath67 \sim\exp\left ( -\frac{x^{2}}{2}\right ) \left [ x^{-1}+o\left ( x^{-3}\right ) \right ] , \quad x\rightarrow\infty.\label{zeta}\ ] ] in figure [ xo1 ] we plot @xmath68 $ ] and @xmath69 $ ] .
we see that the approximation is very good for @xmath70 but it becomes less precise as @xmath71 this is because our previous analysis assumes that @xmath1 with @xmath72 if either @xmath73 or @xmath74 we must modify it , which we will do next . when @xmath75 or more generally when @xmath76 the singularities at @xmath52 and @xmath77 are nearly equidistant from @xmath55 on the scale @xmath78 @xmath79 we have@xmath80 and ( [ 15 ] ) simplifies to @xmath81 but , to this we must add the contribution from @xmath53 which corresponds to replacing @xmath82 by @xmath83 and multiplying by @xmath84 the right hand side of ( [ 17 ] ) .
we note from ( [ z1- ] ) that @xmath85 so that the integrand in ( [ contour ] ) is antisymmetric with respect to the map @xmath86 thus , for @xmath1 and @xmath27 we get @xmath87 with@xmath88 .\label{18}\ ] ] as @xmath89 the alternating term becomes negligible and ( [ 18 ] ) matches to ( [ 15 ] ) , as @xmath90 in figure [ x=0 ] we plot the ratio @xmath91 /\ln\left [ \psi_{2}(y,40)/40!\right ] $ ] and verify the accuracy of ( [ 18 ] ) . letting @xmath92 in ( [ 14 ] ) using ( [ 15 ] )
yields@xmath93 which differs from ( [ large x ] ) .
this suggests that another scale must be analyzed , where @xmath94 and @xmath17 are both large .
thus , we consider the case of @xmath92 , with @xmath95 now the singularity at @xmath52 in ( [ contour ] ) becomes close to @xmath96 since @xmath97 we use the form ( [ pn1 ] ) and expand @xmath98 $ ] for @xmath99 and @xmath71 setting @xmath100 with @xmath101 we obtain @xmath102=\sqrt{2}\mathcal{j}\left\ { 1+\sqrt{\frac{2}{\pi}}\left [ ze^{-x^{2}/2}-\zeta\left ( x\right ) \right ] \right\ }
\\ = \sqrt{2}\mathcal{j}\left [ 1+\sqrt{\frac{2}{\pi}}e^{-x^{2}/2}\left ( z-\frac{1}{x}+o\left ( x^{-3}\right ) \right ) \right ] \\ \sim\sqrt{2}\sqrt{-\ln\left [ \sqrt{\frac{2}{\pi}}e^{-x^{2}/2}\left ( \frac { 1}{x}-z\right ) \right ] } \\ \sim\sqrt{2}\sqrt{\frac{x^{2}}{2}+\ln(x)-\frac{1}{2}\ln\left ( \frac{2}{\pi } \right ) -\ln\left ( 1-\xi\right ) } .\end{gathered}\ ] ] thus , we have@xmath103 here the contour @xmath58 is a small loop about @xmath104 now we again employ singularity analysis , with the branch point at @xmath105 determining the asymptotic behavior for @xmath106 a deformation similar to that in figure [ fig12 ] leads to@xmath107 for @xmath108 this collapses to ( [ large x ] ) . by examining ( [ 15 ] ) and
( [ 20 ] ) , we can obtain the following approximation@xmath109 ^{n+1},\label{21}\ ] ] which is more uniform in @xmath94 , since it holds both for @xmath70 and @xmath110 for @xmath17 large and for @xmath92 with @xmath17 fixed .
however , we must still use ( [ 18 ] ) if @xmath17 is large and @xmath94 is small . in figure [ xlarge ]
we plot @xmath68 $ ] and @xmath111 $ ] and confirm that ( [ 21 ] ) is a better approximation than ( [ 15 ] ) for large values of @xmath25
we shall now rederive the results in the previous section by using only the recurrence ( [ 2.3 ] ) and ( [ large x ] ) .
we apply the wkb method to ( [ 2.3 ] ) , seeking solutions of the form @xmath112
with@xmath113 b(x , n),\quad n\rightarrow\infty .
\label{3.1}\ ] ] thus , we are assuming an exponential dependence on @xmath17 and an additional weaker ( e.g. , algebraic ) dependence that arises from the function @xmath114 using ( [ 3.1 ] ) in ( [ 2.3 ] ) leads to @xmath115 + o\left ( \frac{\partial^{2}b}{\partial n^{2}}\right ) \\ & = \left [ x+a^{\prime}(x)\right ] b(x , n)+\frac{1}{n+1}\frac{\partial } { \partial x}b(x , n).\end{aligned}\ ] ] expecting that @xmath116 and @xmath117 we set@xmath118 and@xmath119 to solve ( [ 3.3 ] ) we let@xmath120 to find that@xmath121 solving this separable ode leads to@xmath122 , \label{3.7}\ ] ] where @xmath123 is a constant of integration . to fix @xmath124
we assume that expansion ( [ 3.1 ] ) , as @xmath74 will asymptotically match to ( [ large x ] ) , when this is expanded for @xmath106 in view of ( [ 3.1 ] ) this implies that@xmath125 , \quad x\rightarrow\infty,\ ] ] so that@xmath126 in view of ( [ 3.7 ] ) this is possible only if @xmath127 and then from ( [ zeta ] ) we have@xmath128 \sim\ln(x),\quad x\rightarrow\infty . \label{3.8}\ ] ] we next analyze ( [ 3.4 ] ) . using ( [ 3.8 ] )
to compute @xmath129 we obtain@xmath130 solving this first order pde by the method of characteristics , we obtain@xmath131 , \ ] ] where @xmath132 is at this point an arbitrary function .
however , since @xmath17 is large and @xmath133 we need only the behavior of @xmath132 for large values of its argument .
we again argue that by matching to ( [ large x ] ) we have @xmath134 b(x , n)\sim x^{n},\quad x\rightarrow\infty,\ ] ] and using ( [ 3.8 ] ) we get@xmath135 and thus@xmath136 so that@xmath137 combining our results , we have found that @xmath138 } } \left [ \frac{e^{-x^{2}/2}}{\zeta\left ( x\right ) } \right ] ^{n+1},\quad n\rightarrow\infty . \label{3.11}\ ] ] this applies for @xmath70 and @xmath139 where we can regain the results of the singularity analysis ( [ 15 ] ) by simply using@xmath140 } \sim\sqrt{\ln ( n)},\quad n\rightarrow\infty.\ ] ] formula ( [ 3.11 ] ) is also valid for @xmath141 and @xmath74 where it reduces to ( [ large x ] ) .
however , ( [ 3.1 ] ) breaks down for @xmath90 we thus consider the scale @xmath78 @xmath142 and set@xmath143 with which ( [ 2.3 ] ) becomes@xmath144 for fixed @xmath145 we seek an asymptotic solution of ( [ 3.13 ] ) in the form@xmath146 where @xmath147 will have a weaker ( e.g. , algebraic or logarithmic ) dependence on @xmath148 from ( [ 3.13 ] ) we obtain , using ( [ 3.14 ] ) , @xmath149 \label{3.15}\\ & = \left [ 1-\frac{1}{n}+o\left ( n^{-2}\right ) \right ] \frac{\partial } { \partial y}q(y , n)+\frac{y}{n}q(y , n).\nonumber\end{aligned}\ ] ] if @xmath147 has an algebraic dependence on @xmath17 , then @xmath150 should be roughly @xmath151 relative to @xmath152 @xmath153 roughly @xmath154 and so on .
thus , we expand @xmath147 as@xmath155 where @xmath156 have a very weak ( e.g. , logarithmic ) dependence on @xmath17 and balance terms in ( [ 3.15 ] ) of order @xmath157 and @xmath158 to obtain@xmath159 and@xmath160 \label{3.18}\\ & = \frac{\partial}{\partial y}q_{1}(y , n)-\frac{\partial}{\partial y}q_{0}(y , n)-yq_{0}(y , n).\nonumber\end{aligned}\ ] ] solving ( [ 3.17 ] ) yields @xmath161 where @xmath162 must be determined . we could solve ( [ 3.18 ] ) , using ( [ 3.19 ] ) , but its solution would involve another arbitrary function of @xmath17 .
thus , considering higher order terms will not help in determining @xmath163 instead , we employ asymptotic matching to ( [ 3.11 ] ) . expanding ( [ 3.11 ] ) for @xmath73 and comparing the result to ( [ 3.12 ] ) as @xmath89 with ( [ 3.14 ] ) , ( [ 3.16 ] ) and ( [ 3.19 ] )
, we conclude that @xmath164 but then our approximation for @xmath165 is not consistent with @xmath166 for odd @xmath148 we return to ( [ 3.13 ] ) and observe that the equation also admits an asymptotic solution of the form@xmath167 where , analogously to ( [ 3.17 ] ) , we find that @xmath168 so that another asymptotic solution to ( [ 3.13 ] ) is@xmath169 we argue that any linear combination of ( [ 3.14 ] ) and ( [ 3.21 ] ) is also a solution and that the combination which vanishes at @xmath170 for odd @xmath17 has @xmath171 and @xmath172 as in ( [ 3.20 ] ) .
we have thus obtained , for @xmath173 @xmath174 , \quad n\rightarrow\infty.\ ] ] this agrees with ( [ 18 ] ) , obtained by singularity analysis in section 2 . to summarize ,
we have shown how to infer the asymptotics of @xmath2 using only the recursion ( [ 2.3 ] ) and the large @xmath94 behavior ( [ large x ] ) .
our analysis does need to make some assumptions about the forms of various expansions and the asymptotic matching between different scales .
we shall now find a uniform asymptotic approximation for @xmath2 using a discrete form of the ray method @xcite .
this approximation will apply for @xmath94 and/or @xmath17 large .
we seek an approximate solution for ( [ 2.3 ] ) of the form @xmath175 , \label{anszat}\ ] ] where @xmath176 as @xmath106 since @xmath177 @xmath178 we see that we must have@xmath179 as @xmath180 using ( [ anszat ] ) in ( [ 2.3 ] ) , we have@xmath181 as @xmath139 where we have used@xmath182 to solve ( [ eikonal ] ) , we use the method of characteristics , which we briefly review .
given the first order partial differential equation@xmath185 we search for a solution @xmath186 by solving the system of characteristic equations@xmath187 with initial conditions@xmath188 = 0 , \label{initial1}\ ] ] and@xmath189 where we now consider @xmath190 to all be functions of the variables @xmath191 and @xmath192 for the eikonal equation ( [ eikonal ] ) , we have@xmath193 and therefore the characteristic equations are@xmath194 and@xmath195 solving ( [ charac1 ] ) subject to the initial conditions@xmath196 we obtain@xmath197 , \nonumber\\ p=\sqrt{\frac{\pi}{2}}\exp\left ( \frac{s^{2}}{2}+b\right ) \left ( t+s\right ) \left [ \operatorname{erf}\left ( \frac{t+s}{\sqrt{2}}\right ) -\operatorname{erf}\left ( \frac{s}{\sqrt{2}}\right ) \right ] \label{char0}\\ + \exp\left ( -\frac{1}{2}t^{2}-st+b\right ) -t - e^{b}+a,\quad\quad q=-\frac { 1}{2}t^{2}-st+b\nonumber\end{gathered}\ ] ] from ( [ initial ] ) we have@xmath198 which is consistent with ( [ initial1])@xmath199 therefore,@xmath200 , \label{rays}\]]@xmath201 \label{pq}\\ + s\exp\left ( -\frac{1}{2}t^{2}-st\right ) -\left ( t+s\right ) , \quad\quad q=-\frac{1}{2}t^{2}-st+\ln\left ( s\right ) .\nonumber\end{gathered}\ ] ] in figure [ rays ] we sketch the rays @xmath202 for @xmath203 .$ ] using ( [ pq ] ) in ( [ eqf ] )
we have@xmath204 \label{eqf1}\\ & + s\left [ 1+\frac{1}{2}t^{2}+st-\ln\left ( s\right ) \right ] \exp\left ( -\frac{1}{2}t^{2}-st\right ) -\left ( t+s\right ) .\nonumber\end{aligned}\ ] ] using ( [ a , b ] ) in ( [ initial2 ] ) , we get @xmath205 and solving ( [ eqf1 ] ) subject to ( [ f(0,s ) ] ) , we obtain@xmath206 \label{f(t , s)}\\ & \times s\left [ 1+\frac{1}{2}t^{2}+st-\ln\left ( s\right ) \right ] -\left ( \frac{1}{2}t^{2}+st\right ) + f_{0}\nonumber\end{aligned}\ ] ] or , using ( [ rays]),@xmath207 n+\frac{1}{2}\left ( s^{2}-x^{2}\right ) \left ( n+1\right ) + f_{0}. \label{f(x , n , s)}\ ] ] to solve the transport equation ( [ transport ] ) , we need to compute @xmath208 and @xmath209 as functions of @xmath191 and @xmath192 use of the chain rule gives@xmath210 and hence,@xmath211 where the jacobian @xmath212 is defined by @xmath213 using ( [ rays ] ) , we can show after some algebra that@xmath214 using @xmath215 in ( [ transport ] ) , we have@xmath216 or@xmath217 and using ( [ charac1 ] ) , we obtain@xmath218 since @xmath219 we have@xmath220 where we have used ( [ inversion ] ) and ( [ j ] ) .
thus,@xmath221 and therefore@xmath222 for some function @xmath223 since from ( [ initial ] ) we have @xmath224 while ( [ j1 ] ) gives @xmath225 we conclude that @xmath226 and hence @xmath227 .
\label{g(t , s)}\ ] ] using ( [ j1 ] ) we can write ( [ g(t , s ) ] ) as@xmath228 .
\label{g(n , s)}\ ] ] replacing @xmath229 and @xmath230 in ( [ anszat ] ) by ( [ f(x , n , s ) ] ) and ( [ g(n , s ) ] ) , we obtain @xmath231 as @xmath139 with @xmath232 \frac{s^{n}}{\sqrt{n+1+ns^{-2 } } } , \label{phi}\ ] ] where @xmath233 is still to be determined . eliminating @xmath191 from ( [ rays ] ) we get@xmath234 = 0,\label{eqs}\ ] ] which defines @xmath235 implicitly . for every @xmath236 there
exist only two solutions @xmath237 and @xmath238 of ( [ eqs ] ) ( see figure ) .
since @xmath4 is an odd function , it follows that @xmath239 although we have @xmath240 $ ] for @xmath241 and @xmath242 $ ] for @xmath243 the two approximations are comparable when @xmath94 is small and therefore we must add their contributions .
we shall now find the constant @xmath244 in ( [ phi ] ) by using ( [ large x ] ) .
we rewrite ( [ eqs ] ) as@xmath245 ^{-2}\label{s1}\ ] ] and for a fixed value of @xmath246 consider the limit @xmath71 it follows from ( [ s1 ] ) that @xmath247 and therefore we consider an expansion of the form@xmath248 using ( [ s3 ] ) in ( [ s1 ] ) , we obtain@xmath249 and therefore@xmath250 solving ( [ s4 ] ) we have@xmath251 } , \quad x\rightarrow\infty,\label{spxlarge}\ ] ] where @xmath252 denotes the lambert function , defined by @xcite@xmath253 = z,\quad\forall z\in\mathbb{c}\ ] ] and having the asymptotic behavior@xmath254 ^{2}\right ) , \quad z\rightarrow\infty.\label{wasympt}\ ] ] using ( [ spxlarge ] ) and ( [ wasympt ] ) in ( [ phi ] ) , we obtain@xmath255 from stirling s formula@xmath256 n^{n}e^{-n},\quad n\rightarrow\infty,\label{stirling}\ ] ] and @xmath257 n^{n},\quad n\rightarrow\infty,\ ] ] we conclude that@xmath258 and thus@xmath259 \sqrt{\frac{2\pi s^{2}}{\left ( n+1\right ) s^{2}+n}}.\label{phi1}\ ] ] using ( [ s(0,n ) ] ) , we have @xmath260 as @xmath261 with@xmath262 + \phi\left [ x , n ;-
s_{p}(-x , n)\right ] , \quad n\rightarrow\infty.\label{pfinal}\ ] ] in figure [ ray0 ] we compare @xmath263 $ ] and @xmath264 $ ] for @xmath265 and in figure [ raylarge ] for @xmath266 we note that the asymptotic approximation ( [ pfinal ] ) is more uniform than ( [ 15 ] ) , ( [ 18 ] ) and ( [ 20 ] ) but it is less explicit since @xmath267 must be obtained numerically . next , we compare the results of this section with those in the previous two sections .
we first consider @xmath54 with @xmath70 and @xmath106 from ( [ eqs ] ) , we have@xmath268 } , \label{snlarge}\ ] ] where @xmath269 was defined in ( [ zeta ] ) . using ( [ snlarge ] ) and ( [ wasympt ] ) in ( [ phi1 ] )
, we get@xmath270 ^{n+1},\ ] ] which agrees with ( [ 15 ] ) after taking ( [ stirling ] ) into account .
now we consider the limit @xmath139 with @xmath271 and @xmath272 from ( [ eqs ] ) , we have@xmath273 using ( [ sp ] ) , ( [ s(0,n ) ] ) and ( [ wasympt ] ) in ( [ phi1 ] ) , we find that@xmath274 and therefore@xmath275\ ] ] agreeing with ( [ 18 ] ) . finally , we consider the limit @xmath1 with @xmath276 @xmath277 @xmath278 from ( [ eqs ] ) , we have@xmath279 } .
\label{s2}\ ] ] using ( [ s2 ] ) and ( [ wasympt ] ) in ( [ phi1 ] ) , we have@xmath280 which agrees with ( [ 20 ] ) . | we analyze the sequence of polynomials defined by the differential - difference equation @xmath0 asymptotically as @xmath1 .
the polynomials @xmath2 arise in the computation of higher derivatives of the inverse error function @xmath3 .
we use singularity analysis and discrete versions of the wkb and ray methods and give numerical results showing the accuracy of our formulas .
msc - class : 33b20 ( primary ) 34e20 , 33e30 ( secondary ) .
keywords : inverse error function , differential - difference equations , singularity analysis , discrete wkb method . |
the _ hubble space telescope _ ( hst ) has been used to undertake a systematic survey of extragalactic radio sources , the 3cr snapshot survey ( de koff et al .
1996 ; martel et al . 1998,1999
; mccarthy et al . 1997 , de vries et al . 1997 ) selected from the _ revised third cambridge catalogue of radio sources _ ( 3cr ) ( bennett 1962a , b ; spinrad et al .
1985 ) . these data , taken with the wide field planetary camera ( wfpc2 ) using the f702w , f555w and ramp narrow - band filters ,
allow us to investigate the relationships between the radio and optical morphologies in a large sample of powerful radio galaxies over redhshifts of @xmath1 .
imaging of nearby seyfert galaxies with hst , such as mrk 3 , mrk 6 , mrk 573 , ngc 1068 , ngc 2992 and ngc 4151 , ( capetti et al .
1995a , 1995b , 1996 , 1997 ; winge et al .
1997 ; allen et al . 1998 , 1999 ;
axon et al .
1998 ) has shown an intimate connection between the radio structure and the extended nlr ( enlr ) .
these studies show that the interaction of the radio jet with the ism is the main source of uv photons that ionize the enlr in seyfert galaxies . in previous work using hst data of 3c 299 , feinstein et al .
( 1999 ) found that the the ne radio lobe lies within a shell - like structure of the enlr , suggesting a physical connection between the jet and this enlr .
evidence of this interaction comes from the values of the line ratio diagnostics and the [ o ii]@xmath2/[o iii]@xmath0 ratio , an estimator of the change of the ionization parameter @xmath3 , over the enlr .
we are extending this work by investigating whether this scenario is also applicable to other powerful radio sources . in this paper
, we discuss the radio galaxy 3c 244.1 3c 244.1 is a frii radio source with two lobes with angular separation of 52@xmath4 at position angle ( p.a . ) of @xmath5 .
both lobes have the same flux density ( fernini et al .
( 1997 ) @xmath6 w / hz ) .
fernini et al .
( 1997 ) also detected the central component of the source that coincides with the optical identification of spinrad et al .
( 1985 ) , showing that the radio lobes are located asymmetrically with respect to this central compact source . at optical wavelengths ,
the host galaxy of 3c 244.1 was shown to have strong emission lines and a redshift of @xmath7 ( spinrad et al .
from ccd images , mccarthy et al .
( 1995 ) note that the galaxy lies in a cluster , and their [ o iii ] image shows a high surface brightness fan of emission extending @xmath8 to the northwest , with a position angle similar to the radio structure . in this paper
we discuss the morphology of 3c 244.1 as shown by the high spatial resolution images of wfpc2 , and we investigate the possible scenarios that give rise to the observed emission morphology .
the hst / wfpc2 observations of 3c 244.1 were taken as part of the 3cr imaging snapshot survey ( pi : sparks ) which was conducted in cycles 4 to 8 , in the f555w and f702w broad bands , and narrow emission - line bands ( de koff et al .
1996 ; martel et al . 1998,1999
; mccarthy et al . 1997 , etc ) .
the 3c 244.1 images were obtained with the wfpc2/pc in april 1994 and may 1995 for the f702w filter and in july 1996 for the f555w filter .
narrow - band images with the wfpc2/wf2 and the fr680n ramp filter were taken in november 1995 .
table 1 shows the observation log . at a redshift of @xmath7 ,
the galaxy is at a distance of 1.5 gpc with a projected linear scale of 5.1 kpc arcsec@xmath9 ( assuming @xmath10=65 km sec@xmath9 mpc@xmath9 and @xmath11 ) .
the wfpc2/pc scale is @xmath12 pixel@xmath9 and for the wfpc2/wf2 the scale is @xmath13 pixel@xmath9 , and therefore , the physical scales for the images are 232 pc pixel@xmath9 for the pc mode and 492 pc pixel@xmath9 for the wf2 mode . the reduction procedure for the f702w filter data , which is close to cousins _ r _ filter , is fully discussed in martel et al .
this reduction includes the standard wfpc2 pipeline processing followed by cosmic - ray removal .
the data reduction was carried out using iraf and the stsdas package . at the redshift of 3c 244.1
, the f702w filter includes the h@xmath14 , [ o iii]@xmath15 , he ii@xmath16 , h@xmath17 , [ o iii]@xmath184959,5007 and [ n i]@xmath19 emission lines .
the f555w filter data , which is close to johnson _ v _ , was reduced in the same manner and then registered and added to produce one final image with higher s / n .
due to the redshift , the f555w filter includes the flux from the emission lines of [ ne v]@xmath20 , [ o ii]@xmath2 , [ ne iii]@xmath21 , h@xmath22 and h@xmath14 ( the latter considerably attenuated by the filter response ) . in what follows we will define f702w as _ r _ , and f555w as _ v_. the fr680n is a narrow - band ramp filter , which , given the redshift of the galaxy and its position on the ccd , produces an [ o iii]@xmath0 image .
these images were reduced as described above and the final image was re - scaled to the resolution of the wfpc2/pc .
the f702w and f555w filter data were flux - calibrated using values for the inverse sensitivities ( photflam ) of @xmath23 ergs @xmath24 @xmath9 dn@xmath9 , @xmath25 @xmath24 @xmath9 dn@xmath9 , and zero - points for the vega system of m@xmath26 and m@xmath27 , respectively .
non - rotated images were used for these flux calibrations to minimize interpolation errors .
the fr680n filter was calibrated using the ramp filter calculator ( biretta et al , 1996 ) .
an approximate coordinate frame for the wfpc2 data is provided by the image header information , based on the hst guide stars .
the accuracy for the coordinates is approximately @xmath28 ( biretta et al .
. images obtained with different filters were registered to a common reference frame using cross - correlation of the brighter elliptical structures .
the sky background in each image was determined through the statistical analysis of a @xmath29 clipping of the average background value in several regions in each of the images .
the optical broadband images show an elliptical galaxy with blobs and filaments ( fig . 1 and 2 ) . both the _ v _ and _
r _ images show these structures , and some are also present in the ramp image .
we have named the most prominent of these structures as a , b , c , d and e ( fig .
3 ) . to compare with radio data , we superimposed a 8.4 ghz radio map from the observations of leahy ( 1999 ) in fig .
the total extension of the radio structure is @xmath30 ( from lobe to lobe ) , larger that the size of these hst images .
the object named a is bright and extended , and is located @xmath31 from the nucleus ( p.a .
@xmath32 ) , and has the same shape in all filters .
the flux in the narrow - filter implies a large emission of [ o iii]@xmath0 emission .
the structure named b , is located @xmath33 away from the nucleus at p.a .
@xmath34 , and is similar in shape to structure a albeit fainter . both of these structures ( a and b ) seem to be connected with the galaxy by very faint filaments . the object named c is located to the west at a separation of @xmath35 from the nucleus at p.a .
@xmath36 . a filament appears to connected it directly to the center of the galaxy . superimposed over this line at @xmath28 from the nucleus
, there is another blob ( c ) .
the filament just described and those associated with blobs a and b , can be interpreted as the edges of a biconic structure with an opening angle of @xmath37 .
this kind of morphology has been found to be common in a variety of active galaxies ranging from nearby spirals ( for example ngc 1068 , evans et al .
1991 ; ngc 5728 , wilson et al .
1993 ) to powerful 3cr objects ( e.g. 3c 299 feinstein et al .
the southern radio - jet is located in the middle of this bi - conical structure .
the structure d is located inside the elliptical profile of the galaxy , ne of the nucleus . after the subtraction of a @xmath38 profile ( using the iraf / stsdas task _ ellipse _ ) from the image , a filamentary link emerges connecting d to the galaxy nucleus ( see fig . 1 . ) .
this structure seems not to be as large as the others and is very difficult to measure in the broad - band images because of the steep shape of the elliptical galaxy profile .
this structure is also evident in the [ o iii]@xmath0 image , but as for the broad - band data , it is not possible to measure its flux reliably . the structure named e , at @xmath39 from the nucleus at p.a .
@xmath40 , is not present in the narrow - band filter .
its profile is compact and steep , showing all the typical characteristics of a field galaxy with a redshift different from that of 3c 244.1 .
we have carried out photometry of the structures surrounding 3c 244.1 using the iraf task _ polyphot _ ( from the noao apphot package ) .
this iraf task computes the magnitude of an object in an image inside a polygonal aperture .
the sky background is also computed and subtracted and the errors are calculated . to measure the bulk of the emission for each structure a polygon that closely matches the shape of the region was careful chosen .
the same polygons were used on all images from the different filters to make the data comparable between these filters . however , for the _ v _ filter , where the signal is lower , some of these structures were difficult to measure above the background noise .
table 2 shows the result of the photometry using the polygonal measurement technique and the subsequent calibration . for the broad - bands filters ,
the data were calibrated using the inverse sensitivity ( photflam ) and the filter width ( see wide field and planetary camera instrument handbook - biretta , 1996 ) .
for the fr680n filter , the calibration was done using the iraf / stsdas _ synphot _ , and checked with the wfpc2 exposure time calculator ( etc ) .
the flux observed at this filter , basically the flux of the [ o iii]@xmath0 line , was corrected by a small contamination ( @xmath41% ) of the [ o iii]@xmath42 line ( asumming that the flux of the [ o iii]@xmath0 line is 2.88 times the [ o iii]@xmath42 flux ) .
this correction was calculated by computing the central wavelength at the position on the ccd , and assuming that the filter is gaussian , centered at that wavelength with a fwhm of 1.3% ( value obtained from the etc ) of this central wavelength .
columns 2 , 3 and 4 of table 2 are the results of these calibrations . from fig .
1 , structures a and b are very close the jet flow , so the first attempt was to test if these objects were optical synchrotron knots on the radio jet .
so we fit a power law to flux from the broad band filters f555w and f702w .
if the main source of flux is synchrotron radiation , the flux must follow a power law with an exponent in the range -0.5 to -1.5 . to make the test as more accurately as we can , we used the task _ calcphot - synphot _
( from the synphot / stsdas iraf package ) to integrate the power law under a better simulation of each filter .
we found that the observed flux is not a power law and the best fit is a flat distribution with a null slope .
so , we conclude that the observed flux is inconsistent with power - law distributions of optical synchrotron radiation . from the the broad band filters
, we can estimated the ( v - i ) colors of the elliptical host galaxy , a and b regions .
colors of the other structures can not be measured unless we have a better modeling of the underline galaxy flux , for which we need deeper observations and better spatial resolution .
objects a and b are far away from the elliptical profile and are clearly not contaminated . for the elliptical galaxy
we obtain ( v - i)=1.2 @xmath430.3 which is close to the value predicted by rnnback et al .
( 1996 ) for an elliptical galaxy at the redshift of 3c 244.1 ( z=0.428 ) meanwhile the objects a and b , seems to be bluer with values -0.2 @xmath43 0.2 and -0.6 @xmath43 0.5 respectively . both structures a and b , given the large flux in the narrow ramp filter , are obviously at the 3c 244.1 redshift .
also both show the same morphology shape as the broad - band images . since the _ r _
filter data also includes the flux of h@xmath14 , [ o iii]@xmath15 , he ii@xmath16 , h@xmath17 , [ o iii]@xmath184959,5007 and [ n i]@xmath19 emission lines , the total flux measured in this broad band must be larger than the ramp filter which only includes [ o iii]@xmath0 .
the fact that they are so similar can be explained if we assume than the flux originates in a pure emission line region , without any significant continuum contribution .
as the narrow - band filter data shows , there is a large amount of flux in the emission line [ o iii]@xmath0 .
thus it is very reasonable to suggest that the structures a and b are extended narrow line regions ( enlr ) .
the enlr have been found to be associated with a wide variety of agn , from seyfert 2 to radio galaxies .
column 5 ( table 2 ) shows the flux calibration of the broad - band _ r _ filter assuming that all the flux is from a source that has a spectrum dominated by bright emission lines .
the shape of the f702w filter is practically flat ( see wide field and planetary camera 2 instrument handbook , page 217 ) , and so it is easy to compute the calibration of the flux of all the lines by calculating the total flux for only one emission line ( that accounts for all the line emission flux ) . to make this calculation , the etc was used and the quantum efficiencies ( system + filter + ccd , hereafter qt ) for each wavelength of the conspicuous emission lines included in the _ r _ filter band were computed .
table 3 shows the results of this calculation and confirm that all these emission lines are practically attenuated by similar amounts .
therefore we used h@xmath17 as an equivalent line ( i.e. a line that has all the flux of all the lines included in the filter ) for the calibration .
the error of this procedure is extremely small .
a key question is to identify the mechanism responsible for the ionization of these structures and filaments .
the most likely mechanisms are : photoionization by the agn , photoionization by the agn with matter - bounded clouds and shock - ionization by the radio jet .
we discuss each of these mechanism in detail .
this mechanism assumes that the uv flux produced by the agn s nuclear source photoionizes the line emission region .
the physical status of the gas at any place can be described with the ionizing parameter ( @xmath3 ) , which considers the dilution of the uv photons as the distance from the nucleus increases . for regions
a and b , it is very unlikely that the agn is the only source of photons because the large distances . to account for this scenario ,
some reference numbers of the total photon budget can be calculated .
rawlings et al . ( 1989 ) performed spectroscopy of the nuclear zone of 3c 244.1 and show a relation of [ o iii]@xmath0 /h@xmath44 . using the results of ferland and netzer ( 1983 ) , which assume that the main source of photoionization continuum is a power law ( @xmath45 and that the gas has a density of @xmath46 , they calculate several line emission ratios .
using these results , it is straightforward to compute the ionization parameter ( @xmath3 ) .
this leads @xmath47 for solar abundances and @xmath48 for subsolar abundances .
this spectrum covers only light coming from the nuclear zone that is observed in the narrow band filter ( regions a and b are not included in this spectrum ) , so it is possible to calculate the total uv photon emission of the nucleus . because @xmath49 and @xmath50 can be estimated from the wfpc2 emission line image , this gives @xmath51 photons sec@xmath9 for solar abundances and @xmath52 photons sec@xmath9 for sub - solar abundances . due to the large distances , these two values ( for each this different abundances ) are not enough to ionize the blobs a and b. therefore , we conclude that direct photoionization by the nucleus is unlikely to be the dominant mechanism responsible for the observed enlr emission .
this idea was developed by binette et al .
( 1996 ) in order to explain the discrepant high ionization line ratios and high electron temperatures observed in many active galaxies .
these models are based on a parameter @xmath53 which is the solid angle covered by the `` matter - bounded '' component relative to that covered by the `` ionization - bounded''component . in another paper , binette et al .
( 1997 ) refined their models . taking into account the possibility of having a larger @xmath3 parameter , they derived three models with values of @xmath54 , 0.05 , 0.02 and a density @xmath55 @xmath56 for the `` matter bounded '' clouds , which are exposed to ionization radiation from the agn . as we only have one line and the integrated total flux of several lines from the _ r _ filter , we computed the flux ratio @xmath57/[o iii]@xmath0 noting that the bulk of the r filter flux is the sum of h@xmath14 , [ o iii]@xmath15 , he ii@xmath16 , h@xmath17 , [ o iii]@xmath184959,5007 and [ n i]@xmath19 fluxes .
note that the contribution from the stellar host galaxy is not significant at the location of the blobs .
the line ratio can be derived from table 2 and are @xmath58 for structure a and @xmath59 for structure b. we computed @xmath60 ( the ratio of lines include in the _ r _ filter to [ o iii]@xmath0 ) for the results of mixed models of the papers mentioned above .
table 4 shows these results of @xmath60 as a function of the @xmath53 parameter for the model of binette et al .
( 1996 ) and models h , m , l of ( binette et al . 1997 ) .
models m and l seem to fit the observed @xmath60 for structures a and b , for @xmath61 and 0.34 for each object . but both of them have the same problem that we have discuss for the first case of photoionization by the agn , @xmath62 for model m and @xmath63 for model l , values that imply a very high @xmath64 ( @xmath65 ) due to the distances to the galaxy nucleus .
this calculation makes the mixed models very unlikely .
fig 1 shows that the enlr structures a and b are located close to the edges of the radio - jet and it is possible that there is an interaction between the jet and these regions .
it is important to study whether there is a physical relationship between the radio - jet and the optical line emission .
taylor et al .
( 1992 ) proposed that fast bowshocks resulting from the interaction of the radio jet and the ism were the source of ionizing photons of the emission - line gas in a number of sources .
capetti et al .
( 1995a , 1995b , 1996 , 1997 ) and winge et al .
( 1997 ) were the first to show that this mechanism best explains the optical emission in the nlr in nearby seyfert galaxies ( mrk 3 , mrk 6 , mrk 573 , ngc 1068 , ngc 4151 and ngc 7319 ) .
recently , this work has been confirmed by aoki et al .
( 1999 ) and kukula et al .
( 1999 ) . in the case of the powerful radio galaxies of the 3cr catalogue ,
feinstein et al . ( 1999 ) showed that this interaction also occurs in 3c 299 , where clearly the ne radio - jet and the enlr have similar morphologies , and where there is further evidence of this interaction from the values of the different emission - line ratios and the evolution of the line ratio [ o ii]@xmath2/[o iii]@xmath0 , as an estimator of the changes of @xmath3 , over the region .
dopita & sutherland ( 1995a , b ) have modeled in detail the ionization of the enlr due to shocks . in one scenario , which has been shown to work for seyfert galaxies , the radio jet interacts with the local interstellar medium and shocks the gas . in this scenario ,
the hot post - shock plasma gas produces photons that can diffuse upstream and downstream of the jet .
photons diffusing upstream can encounter the preshocked gas and produce an extensive precursor hii region , while those traveling downstream will influence the ionization and temperature structure of the recombination of the shock . to check the validity of this interpretation
, we use the line diagnostic test of dopita & sutherland ( 1995b ) .
we applied the same procedure as the case above and use the flux ratio @xmath57/[o iii]@xmath0 . on the other hand
, we can compute the value of this ratio for emission lines arising in a shocked enlr . using table 1 of dopita & sutherland ( 1995b )
we calculated the total integrated flux for the lines involved and the ratio to [ o iii]@xmath0 flux as : @xmath66\lambda 5007\ qt_{[o iii]\lambda 5007 } } $ ] where _ i _ represents each of the lines mentioned above and @xmath67 their respective flux .
table 5 shows the results of this computation , for shock speeds of 100 , 150 , 200 and 500 km sec@xmath9 , with and without a precursor h ii region . from this table ,
the values observed for structures a ( @xmath58 ) and b ( @xmath59 ) are easily explained with shock models of @xmath68 km sec@xmath9 .
we conclude that shock interactions between the expanding jet and the local denser medium are responsible for the observed enlr emission in 3c 244.1
we have shown that radio galaxy 3c 244.1 has a filamentary structure and some blobs bright in emission line ( [ o iii]@xmath0 ) . to the south
, this structure seems to be an emission - line cone with an opening angle of @xmath37 ( from p.a .
@xmath32 to p.a @xmath36 ) .
the radio - jet is located at the center of this cone ( p.a .
@xmath5 ) . the two brighter blobs ( named a and b ) are likely part of the enlr ( by comparing the flux between broad - band and narrow - band ) , similar to those associated with agns . these structures are larger and located far away from the nucleus ( @xmath31 ) . to the north more of these structures were found , but near and associated with the nucleus .
we tested several scenarios to explain the source of energy for the emission lines observed and we found that the direct photoionization by the agn can only be possible if an unlikely large amount of uv photons is provided by agn .
the same is true in more sophisticated mixed medium models ( binette et al .
1996,1997 ) .
the observations seem to fit well the behavior of a model material shocked by the radio jet ( dopita et al .
a shock velocity of 200 km sec@xmath9 can explain the ratio of the lines included in the _ r_-band filter to the [ o iii]@xmath0 line - emission .
therefore , from the morphology ( location of the radio - jet ) and these physical arguments ( the flux measured is consistent with emission from shocked gas ) , we conclude that the enlr structures of 3c 244.1 are the result of the interaction of the radio jet with ism gas . c c c c f702w & broad band & 300 & 1994 apr 17 + f702w & broad band & 300 & 1995 may 28 + fr680n & [ o iii]@xmath0 & 300 & 1995 nov 13 + fr680n & [ o iii]@xmath0 & 300 & 1995 nov 13 + f555w & broad band & 300 & 1996 sep 7 + f555w & broad band & 300 & 1996 sep 7 + c r r r r a & 17.26 @xmath43 1.37 & 20.57 @xmath43 2.69 & 9.36 @xmath43 1.24 & 24.12 @xmath43 1.91 + b & 5.25 @xmath43 0.82 & 5.04 @xmath43 1.62 & 3.58 @xmath43 0.77 & 7.34 @xmath43 1.14 + c & 2.6 @xmath43 0.59 & & & 3.62 @xmath43 0.83 + c & 2.3 @xmath43 0.52 & & & 3.24 @xmath43 0.74 + c c h@xmath14 & 0.133 + @xmath69\lambda4363 $ ] & 0.139 + @xmath70\lambda4686 $ ] & 0.132 + h@xmath17 & 0.124 + @xmath69\lambda4959 $ ] & 0.113 + @xmath69\lambda5007 $ ] & 0.112 + @xmath71\lambda5199 $ ] & 0.092 +
l c c c c 0.04 & 1.79 & 1.83 & 4.95 & 6.77 + 0.14 & 1.74 & 1.76 & 2.99 & 3.19 + 0.24 & 1.71 & 1.71 & 2.46 & 2.50 + 0.34 & 1.68 & 1.67 & 2.21 & 2.21 + 0.44 & 1.66 & 1.65 & 2.06 & 2.05 + 0.54 & 1.65 & 1.63 & 1.97 & 1.95 + 0.64 & 1.64 & 1.61 & 1.90 & 1.88 + 0.74 & 1.63 & 1.60 & 1.85 & 1.83 + 0.84 & 1.62 & 1.59 & 1.81 & 1.79 + 0.94 & 1.61 & 1.58 & 1.78 & 1.76 + 1.04 & 1.60 & 1.57 & 1.76 & 1.73 + | we present results of _ hubble space telescope _ ( hst ) observations of the radio galaxy 3c 244.1 .
the broadband f702w ( _ r _ ) and f555w ( _ v _ ) images ( wfpc2/pc ) show an elliptical galaxy and gaseous filaments and blobs surrounding it . in the narrow - band ramp filter , dominated by [ o iii]@xmath0 , these filaments are bright and have the same morphology as the broad band images . to the south , the filaments have a cone - shaped structure and the radio jet is located at the center of this cone . to the north of the galaxy
, the structure is found near the nucleus of the galaxy within its elliptical profile . from the photometry ,
the two brighter structures seem to be extended narrow line emission regions ( enlrs ) .
the comparison with diagnostic line ratios shows that the observed emission is consistent with interactions between the expanding radio - jet and the local denser medium . |
pelupessy et al . ( 2000 ) formulated the wind momentum equation for sectorial line driven winds including the finite disk correction factor for an oblate rotating star with gravity darkening for both the continuum and the lines .
they calculated models with line force parameters around the bi stability jump at 25000 k. in this case , from the pole to the equator , the mass flux increases and the terminal velocity decreases .
their results showed a wind density contrast @xmath1 ( hereafter @xmath2 ) of about a factor 10 independent of the rotation rate of the star . in this work
, we implement an approximative version for the oblate finite disk correction factor , @xmath3 , allowing us to solve numerically the non linear m cak momentum .
we solve then this equation for a _ classical be _ star , for polar and equatorial directions . in this study
we do not take into account the bi stability jump .
in order to incorporate the oblate distortion of the shape of the star due to the stellar rotation to the m cak hydrodynamic model , we implement an approximative function . in view of the behaviour of @xmath3 and @xmath0 we approximate its ratio @xmath4 via a sixth order polynomial interpolation in the inverse radial variable @xmath5 , i.e. , @xmath6 . in this form
, we assure that the topology found by cur ( 2004 ) is maintained by the @xmath0 term , but it is modified by the incorporation of the @xmath7 polynomial . with this approximation we can solve numerically the non linear m cak differential equation .
we solve the oblate m
cak equations for a classical be star with the following stellar parameters : @xmath8 k , @xmath9 , and @xmath10 ( slettebak et al . 1980 ) and line force parameter @xmath11 . for the other parameters , we have used two different values for @xmath12 , namely : @xmath13 and @xmath14 , and three values for @xmath15 , i.e. , @xmath16 . in this study , we have considered _ the same value of @xmath15 , @xmath17 and @xmath12 _ for the pole and equator , i.e. without taking into account the bi stability jump .
we show only solutions for @xmath18 at the equator and pole . for the cases where @xmath19 and @xmath20 , the density contrasts exceed a factor 100 , values which are in agreement with observations ( lamers & waters 1987 ) . for equator
( _ dashed line _ ) and pole ( _ solid line _ ) with @xmath19 and @xmath13 .
b ) solid lines : density constrasts for @xmath18 and @xmath13 .
dashed lines : spherical cases.,title="fig:",width=249 ] for equator ( _ dashed line _ ) and pole ( _ solid line _ ) with @xmath19 and @xmath13 .
b ) solid lines : density constrasts for @xmath18 and @xmath13 .
dashed lines : spherical cases.,title="fig:",width=249 ] + fig .
the oblate correction factor has been implemented in an approximative form .
the factor @xmath21 in the oblate correction factor certainly modifies the topology of the hydrodynamical differential equation and we suspect from first calculations that other critical points may exists and , therefore , more solutions might be present .
we recover the observed density contrast only when the rotational velocity of the star is near the break - up velocity , confirming other theoretical works ( see e.g. , townsend et al , 2004 ) .
the use of a set of self consistent line force parameters is necessary to understand the wind dynamics of these rapid rotators .
the full version of the oblate correction factor will be implemented to study the topology of the wind . | we have incorporated the oblate distortion of the shape of the star due to the stellar rotation , which modifies the finite disk correction factor ( @xmath0 ) in the m - cak hydrodynamical model .
we implement a simplified version for the @xmath0 allowing us to solve numerically the non linear m - cak momentum equation .
we solve this model for a classical be star in the polar and equatorial directions .
the star s oblateness modifies the polar wind , which is now much faster than the spherical one , mainly because the wind receives radiation from a larger ( than the spherical ) stellar surface . in the equatorial direction we obtain slow solutions , which are even slower and denser than the spherical ones . for the case
when the stellar rotational velocity is about the critical velocity , the most remarkable result of our calculations is that the density contrast between the equatorial density and the polar one , is about 100 .
this result could explain a long - standing problem on be stars . |
two of the most influential models of the interstellar medium ( ism ) to date are the two- and three - phase models of field , goldsmith & habing ( 1969 ) and of mckee & ostriker ( 1977 , hereafter , mo ) .
these models relied on the then known atomic and radiative heating and cooling processes to provide a self - consistent picture of the ism in which the concepts of thermal and pressure equilibria played a fundamental role .
another important model was the time - dependent model of the ism by gerola , kafatos & mccray ( 1974 ) , which was presented as an alternative to the pressure equilibrium two - phase model of field et al.(1969 ) but assumed radically different conditions : a constant - density medium under the influence of stochastic , local heating events that should cause strong local fluctuations of pressure and temperature , because the cooling and recombination times are comparable or shorter than the time between exposure of a given gas parcel to one of those heating events .
note that the mo model also recognized the existence of local fluctuations in the pressure , although it was still based on the premise of `` rough pressure balance '' .
nevertheless , both the equilibrium and the time dependent models left out a number of important aspects in the ism physics .
the multiphase equilibrium models essentially neglected the possibility of large pressure fluctuations in the ism . the time dependent model instead included this possibility as a fundamental premise , but still neglected the fact that such fluctuations should induce motions , which should in general be turbulent and cause strong density fluctuations .
the turbulence involves gas motions at all scales which not only provide ram `` pressure '' , but also mixing , and can produce compressions rather than `` support '' ( e.g. , elmegreen 1993 ; , & scalo 1999a ) .
moreover , both the time - dependent and the mo models omitted other sources of pressure in the ism , such as magnetic fields and cosmic rays .
the pressure from these agents is in fact significantly larger than the thermal pressure ( e.g. , boulares & cox 1990 ) .
elmegreen ( 1991 , 1994 ) has performed a combined instability analysis including self - gravity , cooling and heating , and magnetic fields , but the full nonlinear behavior can only be dealt with by means of numerical simulations of the gas dynamics in the galactic disk . in this review
, i will summarize a variety of results in this area , pioneered by bania & lyon ( 1980 ) , concerning the spatial distribution both on the galactic plane and perpendicular to it ( 2 ) , the one - point and correlation statistics of the physical fields ( 3 ) , the comparison between `` synthetic '' observations of the simulations and actual observations of the ism ( 4 ) , and several dynamical aspects , including the role of thermal and ram pressures , the virial balance of clouds , the nonlinear q parameter and the star - gas connection ( 5 ) .
i conclude ( 6 ) with an attempt to present a comprehensive scenario , which should constitute the first step towards a full dynamical theory of the ism that can take over where the time - dependent and the mo models left off .
given the focus on hi gas of the present conference , this review concentrates on hydro- and magnetohydrodynamic numerical simulations of the multi - temperature gas in the galactic disk .
thus , the vast literature existing on numerical simulations of isothermal gas , aimed at the molecular - cloud regime , will necessarily be excluded .
the reader is directed to the review by et al .
( 2000 ) for a discussion of that area current up to 1999 , and that by mac low & klessen ( 2002 ) for the more recent results .
non - hydrodynamical models of the dynamical systems kind have also been excluded ( see , e.g. , the review by shore & ferrini 1995 ) .
one of the first concerns of numerical simulations was to show that stellar energy injection through ionization heating , is indeed capable of carving `` holes '' containing hot ( @xmath1k ) or warm ( @xmath2k ) dilute gas , and of producing dense , cool ( @xmath3k ) clouds , or in general , what has often been referred to as a `` multiphase '' medium .
however , it is important to note that in all simulations , the temperature spans a continuum of values , and the so - called `` phases '' are generally defined simply in terms of ranges within this continuum .
studies both parallel and perpendicular to the galactic plane were conducted .
it is also important to note that a general feature of the dynamic , yet statistically - stationary regime into which simulations settle , and which is evident from animations , is that the `` clouds '' have no long - lasting identity , but instead are continually `` morphing '' ( i.e. , changing shape , stretching , merging with other structures , and splitting apart ; see 5.1 ; shadmehri , & , this volume ) .
early attempts to show the generation of a multi - temperature medium , even at very low resolutions in two dimensions ( 2d ) parallel to the galactic plane , were those by bania & lyon ( 1980 ) , chiang & prendergast ( 1985 ) , and chiang & bregman ( 1988 ) .
in particular , chiang & prendergast also identified a `` star - formation '' instability , similar to a thermal instability ( field 1965 ; field et al .
1969 ) around an equilibrium state in which star formation balances mass loss , and stellar heating of the gas balances radiative cooling ( see 5 ) . on the plane , self - gravity and later magnetic fields ,
the coriolis force and shear were included ( , passot & pouquet 1995 ; passot , & pouquet 1995 ) , although still without supernova energy input , and without a thermal instability - inducing cooling function
. moreover , et al . ( 1999a ) suggested that , since clouds are really sites of larger - scale converging flows , they should often contain shocks within them , implying that the magnetic field can develop large fluctuations and even reversals .
supernova - like energy input at the highest - density sites was considered by gazol - patio & passot ( 1999 ) , who found a filling factor of the hot gas at the midplane of typically a few percent , with occasional excursions of up to @xmath4 20% .
gerritsen & icke ( 1997 ) gave the radial profiles and mass fractions of the gas and star surface density on the plane , finding that the warm gas becomes the dominant gas component beyond 2 kpc from the center , and that between 2 and 7 kpc the mass fractions remain roughly constant , at @xmath5% for the warm gas , @xmath4 15% for the cold gas , and @xmath6% for what they called the `` lukewarm '' gas , with temperatures intermediate between those of the cold and warm `` phases '' .
in general , in 2d simulations , the cloud morphology is extremely filamentary , the filaments arising where large - scale motions collide and shock . in turn , `` knots '' form at sites where the filaments collide ( see also wada & norman 1999 , 2001 ; chappell & scalo 2001 ) .
wada & koda ( 2001 ) furthermore find that in simulations with a weak bar - like gravitational potential , the pc - scale filaments and clumps may organize into large - scale spirals .
interestingly , the filamentary morphology appears to persist in the 3d simulations of wada ( 2001 ) . in these simulations , which do not include the magnetic field
, the filamentary structure appears to be the consequence of tidal interactions between clumps formed by the combined gravitational and thermal instabilities .
inclusion of the magnetic field may provide another channel for forming filamentary density structures .
perpendicular to the disk , two main paths of numerical study have been pursued : one assumes that the stellar energy injection maintains a turbulent , convective regime in which the vertical structure of the disk is a consequence of both the buoyancy of the hot gas and the bulk motions induced in the cold medium by the injection of thermal and kinetic energy by the stellar sources .
the time average of the instantaneous gas distribution is then reported as the resulting vertical distribution .
most of the simulations in this category have neglected the magnetic field .
the other assumes perturbations about an initial magneto - hydrostatic equilibrium , as initially proposed by parker ( 1966 ) , and has studied the nonlinear evolution of the perturbations . in the first category ,
rosen & bregman ( 1995 ) investigated the average filling factor and density profile of the various `` phases '' as a function of height above the midplane and of the energy injection rate from supernovae , finding that only injection rates comparable to the observed galactic rate reproduce the observed vertical distribution of the various `` phases '' , implying that it is the stellar energy injection rate which is responsible for the vertical distribution . also , they found that more than one exponential component is needed to model the average vertical density profile , and that their cold gas profile compares favorably to the observational profile of dickey and lockman ( 1990 ) . using 3d simulations ,
similar results have been obtained more recently by gerritsen & icke ( 1997 ) , de avillez ( 2000 ) and wada ( 2001 ) .
the former authors used sph simulations of the whole galactic disk to find the scale heights for the cold and warm gas and for two stellar populations as a result of the stellar energy injection , while de avillez found that , without stellar injection , the disk collapses vertically to a few tens of pc .
wada found that the supernova explosions form a plume - like halo .
in the second , magnetostatic category , recent simulations of the parker instability in a thick disk have been performed by santilln et al .
( 2000 ) in 2d , and by kim , ryu & jones ( 2001 ) and franco et al .
( 2001 ) in 3d .
santilln et al . and kim et al .
concluded that the parker instability alone can not be the main formation mechanism of giant molecular clouds in the general ism , because of the low final density enhancements obtained .
however , franco et al . find that the instability can trigger the formation of such clouds inside spiral arms .
the effect of the passage of a spiral arm has been modeled by martos & cox ( 1998 ) , who suggested that it may induce a `` hydraulic jump '' in which the gas is pushed to higher altitudes as it it is shocked upon entering the arm and acquires a higher pressure .
martos et al .
( 1999 ) have furthermore suggested that this jump may trigger star formation at high altitude over the midplane ( over 500 pc ) .
however , the simulations of martos & cox explored only the isothermal and adiabatic cases ( with a model term for the magnetic pressure but see 3.1 ) and did not include self - gravity , neglecting the possible triggering of thermal and gravitational instabilities by the shock compression ( c.f .
exploring this scenario in the presence of those agents will be of great interest . at a more local level , de avillez & mac low ( 2001 )
have shown that mushroom - shaped structures arise naturally and rather frequently in their 3d simulations including supernova energy injection , as a consequence of the buoyancy of the hot gas in supernova remnants , and proposed that these may correspond to similar structures observed in recent hi surveys ( e.g. , higgs 1999 ; taylor 1999 ) three - dimensional simulations addressing the vertical structure of the galactic disk including both the magnetic field _ and _ the energizing role of the massive stars have been scarce so far .
korpi et al .
( 1999 ) placed supernovae randomly in regions with densities higher than the average .
they discussed the filling factor of the hot gas as a function of height above the plane , finding values @xmath4 2030% at the midplane , but this factor is highly sensitive to the degree of spatial correlation of the model supernova explosions , becoming larger as the supernovae are less correlated .
this probably explains the larger values of the midplane filling factor they found compared to the values reported by gazol - patio & passot ( 1999 ) , since the latter authors placed the sne exclusively at density maxima , making them more strongly correlated . in the simulations of korpi
, the magnetic field is initially taken very small , and grows in time , but at the time at which these authors analyzed their simulations , the magnetic pressure was still much smaller than the equipartition value with the turbulent component ( @xmath7 ) .
thus , they did not discuss the relative roles of the stellar energy injection and the magnetic field in maintaining the disk thickness .
this remains an open issue .
given the turbulent , yet statistically stationary nature of the ism suggested by the numerical models , it is important to understand the statistical descriptors . in this section
i discuss work related to various such descriptors .
the density pdf was found by ( 1994 ) to have a lognormal shape in numerical simulations of isothermal flows , which are reasonable approximations to the flow regime in molecular clouds .
passot & ( 1998 ; see also & passot 1999 ) proposed that the origin of the lognormality is the constancy of the sound speed in isothermal flows .
indeed , in this case , the probability of a given density jump is determined simply by the distribution of velocity jumps in the flow , because the local mach number , which determines the density jump , is independent of the local density , and depends only on the velocity difference across the shock .
thus , all density jumps belong to the same distribution , and , assuming each jump is independent of the previous and later ones ( i.e. , that shocks are independent from each other ) , the central limit theorem can be applied to the logarithm of the density , since the density jumps constitute a multiplicative process .
the width of the pdf is proportional to the rms mach number .
those authors also found that in the more general case of polytropic flows , in which @xmath8 , where @xmath9 is the ( total ) pressure , @xmath10 is the density , and @xmath11 is in general a parameter reflecting the net behavior of the pressure with density , the density pdf approaches a power law at high ( resp .
low ) densities for @xmath12 ( resp .
@xmath13 ) .
they understood this in terms of the dependence of the sound speed on density , which , when introduced in the previously - lognormal pdf , gives an asymptotic power law behavior on one side of the pdf ( see nordlund & padoan 1999 for a related discussion ) . in more ism - like simulations in 2d , including magnetic fields , heating and cooling , the coriolis force and shear , and localized stellar energy injection , scalo et al .
( 1998 ) and kritsuk & norman ( 2001 ) have reported pdfs with power - law tails at high densities , suggestive of a total pressure which behaves as if having an effective exponent satisfying @xmath14 , possibly due to the contribution of the magnetic field to the total pressure . however , in the purely magnetic , isothermal case , several workers ( padoan & nordlund 1999 ; ostriker , stone & gammie 2001 ; passot & 2002 ) have found that for cases with intermediate - to - large alfvnic mach numbers , the magnetic field appears quite uncorrelated with the density .
passot & ( 2002 ) suggest that the lack of correlation implies that the magnetic field does not act as a restoring force , and thus does not efficiently act as a pressure , instead acting more as a forcing .
this implies that modeling the magnetic pressure by means of a @xmath11 may be inadequate in the turbulent case .
padoan & nordlund ( 1999 ) have suggested that the lack of correlation is a signature of the intermittency of the field , and pointed out that this is consistent with observations of the magnetic field strength in molecular clouds .
apparently , the lack of correlation is also seen observationally ( e.g. , crutcher , heiles & troland 2002 ; troland & heiles , this volume ) . in non - magnetic , multi - temperature simulations , wada & norman ( 2001 ) have reported pdfs with a lognormal tail at high densities , and a gaussian one at moderately low densities . wada ( 2001 , private communication )
has furthermore found a power - law range at very low densities .
this could seem at odds with the results of scalo et al . and
kritsuk & norman , but upon closer inspection , there is little discrepancy , because most of the pdfs of wada & norman can be understood in terms of the various polytropic regimes to which the gas in their simulations is subject . in any case ,
a detailed study is in order .
the pdf of total projected ( _ column _ ) density has been discussed by ostriker et al . (
2001 ) , & garca ( 2001 ) and burkert & mac low ( 2001 ) .
although these works were aimed at isothermal gas , the results from & garca should be readily extendable to the non - isothermal polytropic case .
they suggested that the column density pdf should not have a unique functional form , but that instead it should transit from the shape of the underlying 3d density distribution ( a lognormal for isothermal flows ) to a gaussian , as dictated by the central limit theorem when the contributing sample is large enough .
the former ( resp .
latter ) case occurs in the limit of few ( resp .
many ) independent `` density '' events along the line of sight ( los ) , i.e. , when the extension of the observed object along the los is smaller ( resp .
much larger ) than the correlation length . under this interpretation ,
the result of ostriker et al . that the pdf is close to lognormal , suggests that molecular clouds may have sizes comparable to the turbulent correlation length , a result used by burkert & mac low ( 2001 ) to suggest that molecular clouds receive their energy injection predominantly at the large scales .
observational results on the column density pdf for the hi gas are in need to shed light in this direction for this important component of the ism .
the kinetic energy spectrum is one of the most extensively discussed properties of turbulent flows .
simulations of weakly compressible hydrodynamic cases ( e.g. , porter , pouquet & woodward 1992 ) and of magnetized incompressible regimes ( cho , lazarian & vishniac 2001 ) give a spectral slope close to the kolmogorov value of @xmath15 . in the magnetic case , this is consistent with the goldreich & sridhar ( 1995 ) theory .
however , in highly compressible cases , no unique slope has been found , and in fact it appears to be dependent on the degree of compressibility of the flow , given by both the rms mach number and the effective exponent @xmath11 .
indeed , for isothermal cases , slopes between @xmath16 ( padoan & nordlund 2000 , 3d ) and @xmath17 ( gammie & ostriker 1996 , 1 + 2/3 d ) have been found ; the latter value has also been reported for multi - temperature cases in 2d by passot et al .
( 1995 ) , and is the one expected for a velocity field dominated by shocks ( e.g. , saffman 1968 [ sec . 6 ] ; kadomtev & petviashvili 1973 ) .
this value is consistent with the observed scaling linewidth - size relation for molecular clouds ( see , e.g. , 1999 ; et al .
2000 ) found by larson ( 1981 ; see also blitz 1993 for a summary of more recent results ) .
observational determinations of the energy spectrum in the hi gas are often suggestive of spectra closer to the kolmogorov value ( e.g. , minter & spangler 1996 ; stanimirovic & lazarian 2001 ; dickey et al .
2001 ) , but sometimes they are not ( e.g. , deshpande , dwarakanath & goss 2000 ) .
it is indeed quite possible that the warm neutral and ionized gas exhibit a behavior closer to incompressible kolmogorov turbulence while the molecular gas is much more compressible , because the cooling time in the former is comparable or larger than the dynamical time , while it is much shorter in the molecular regime .
this implies that the behavior in the diffuse gas may be closer to adiabatic ( and therefore only weakly compressible ) ( kritsuk & norman 2001 ; snchez - salcedo , & gazol 2002 ) . an important statistical descriptor that has been discussed mostly in the context of molecular clouds , but that is in principle equally important in other regimes in the ism , in particular the hi gas , is the pdf of the velocity ( vector ) field . in the case of incompressible turbulence , the velocity pdf is known to be nearly gaussian , while that of the velocity _ gradient _ ( or velocity difference across positions in a map ) is closer to exponential ( e.g. , frisch 1995 ) .
observationally , only the line - of - sight component of the velocity is available , and only in projection , and so both line profiles ( falgarone & phillips 1990 ; falgarone et al .
1994 ) and pdfs of the line velocity centroids ( miesch & scalo 1995 ; lis et al . 1996 , 1998 ;
miesch , scalo & bally 1999 ) have been proposed as estimators of the line - of - sight velocity pdf .
falgarone et al .
( 1994 ) showed that the line profiles of weakly compressible turbulence have similar moments to those of @xmath18co and @xmath19co observational data , while lis et al.(1998 ) found that the line centroid _ difference _ pdfs of weakly compressible turbulence are also similar ( non gaussian ) to those of co data . from these results ,
this group has suggested that the regime in molecular clouds is only weakly compressible , and dominated by vortical motions . on the other hand , miesch & scalo ( 1995 ) and miesch et al .
( 1999 ) have found that the pdfs of the line centroid themselves ( not of the differences ) in both archival hi data and in molecular - line observations of several star forming regions are also non - gaussian , with nearly exponential tails , in sharp contrast with the gaussian pdf for the velocity characteristic of incompressible or weakly compressible turbulence .
miesch et al .
further pointed out that images of the largest - magnitude centroid velocity difference are less filamentary than expected for weakly compressible cases . from these results ,
this group concluded that the turbulence in molecular clouds is highly compressible .
the rationale here is that the non - gaussian velocity increment pdfs would be common to both the compressible and incompressible cases , and therefore not a good discriminator between the two , while apparently the non - gaussian centroid pdfs are exclusive of the ( proposed to be highly compressible ) molecular cloud turbulence .
the 2d numerical simulations of a pressureless gas with ( small - scale ) stellar energy injection of chappell & scalo ( 2001 ) support this view , giving exponential velocity pdfs .
however , the 3d simulations of self - gravitating , isothermal turbulence forced at large scales of klessen ( 2000 ) give nearly gaussian velocity centroid pdfs .
the reason for this discrepancy is unclear , and may reside in the different scales of energy injection in the two sets of simulations , or in the greater compressibility of the pressureless simulations ( which may actually be a better model of the cool atomic gas in the ism ) .
clearly , this remains an open issue , both numerically and observationally , especially concerning the hi gas in the ism .
an important application of numerical models of the ism consists of `` observing '' them in a manner similar to how actual observations of the ism are performed .
much of the work in this direction has been oriented towards the molecular cloud regime , and thus is out of the scope of this review , but the interested reader is referred to the recent work of padoan , mac low , e. ostriker , stutzki and collaborators . here
we discuss some applications to the global ism , and molecular - cloud work only when it has a direct impact on the global ism .
one early result found by burton ( 1971 ) and adler & roberts ( 1992 ) , albeit somewhat ignored until recently , is that structures in position - velocity ( pv ) space often do not correspond to actual , connected structures in physical , 3d space , but simply to chance superpositions of disconnected structures along the los ( see also et al .
1999a ; pichardo et al .
2000 ; ostriker et al . 2001 ; & mac low 2001 ; wada & koda 2001 ) .
this effect is expected to be especially important in the hi gas , given that its emmision is mostly optically thin and widespread .
pichardo et al . (
2000 ) have recently found two other related results concerning the structure seen in velocity channel column density maps .
first , as foreseen by burton ( 1971 ) , those authors found that the morphology in the channel maps shows a somewhat greater resemblance to that of the _ los - velocity _ field than to that of the density field .
this result , together with that of lazarian & pogosyan ( 2000 ; see also lazarian , pogosyan & esquivel , this volume ) , that the spectral index of the `` emissivity '' ( strictly speaking , of the column density ) in velocity channels of spectroscopic observations depends on _ both _ the spectral indices of the density and velocity fields , constitutes strong evidence that the structure of the los - velocity field plays an important role in determining the structure in velocity channel maps .
the numerical simulations were also used by lazarian et al.(2001 ) to test the predictions of lazarian & pogosyan ( 2000 ) , confirming them to within 10% .
second , pichardo et al . found that the `` emissivity '' power spectrum in velocity channels continues to have substantial power even at scales small enough that both the density and the velocity spectra have started to decay , suggesting that the velocity segregation imposed on the column density by the spectroscopic observation procedure introduces some spurious small - scale power of its own , unrelated to the structure existing in the 3d physical space .
henney & ( 2002 ) interpret this as the result of the formation of _ caustics _
( surfaces of large intensity in pv data cubes ) due to the contribution of finite - extent regions along the los to vanishingly thin velocity channels .
this effect may constitute an alternative or complementary explanation to the one proposed by deshpande ( 2000 ) for the reported observations of structures at very small scales ( 10 - 100 au ) in hi absorption studies ( see references in deshpande s paper ) , which was based on the suggestion that neighboring lines of sight contain contributions from even the largest scales ( in the los direction ) , which account for the variability from one los to a neighboring one .
a different kind of synthetic observation of numerical simulations was given by rosen , bregman & kelson ( 1996 ) , who compared strip scans of the integrated `` emission '' ( total column density ) of cold and hot gas in different locations in their simulations to strip scans of hi and x - ray emission .
they concluded that the best matches were provided from locations in hot bubbles , reinforcing the idea of the sun being in such a location .
they also found examples of both correlation and anticorrelation between the hi and x - ray emission , with a slight statistical preference for anticorrelations .
one of the main advantages of the numerical simulations is that all the physical variables are known everywhere in space and at all times .
this allows for a detailed study of the forces acting to create the structures , the degree to which the structures are transient or bound , and the feedback effects between the stars and the gas , among many other issues .
below , we discuss some of them , after noting that , in general , the regime suggested by the simulations is analogous to kolmogorov turbulence , in the sense that it is the _ statistics _ of the flow that remain constant , even though the state is highly dynamical , and individual structures are transient .
note , however , that that is probably where the analogy ends , as interstellar turbulence is highly compressible ( at least in cool and lukewarm regions ) , magnetized , and is forced over a wide range of scales , rather than only at the largest scales ( see , e.g. , scalo 1987 ; norman & ferrara 1996 ; et al .
moreover , in the compressible case , the turbulent cascade , if present , is likely to have `` leaks '' from all scales down to the dissipative scales via the shocks ( kadomtsev & petviashvili 1973 ) , contrary to the energy - conserving cascade of incompressible turbulence .
one of the fundamental premises of the two- and three - phase models of field et al .
( 1969 ) and mo was that of pressure balance in the ism , so that cool , dense hi clouds are confined by the pressure of their warmer , more diffuse surroundings , even though in the mo model clouds form in the compresed layers of expanding supernova remnants . however ,
ever since the work of bania & lyon ( 1980 ) , hydro- and magnetohydrodynamical numerical simulations have shown that many features of the ism , including the thermal distribution , can be reproduced solely by the action of stellar kinetic energy injection on the ambient gas .
in other words , it is important to determine whether ( or when ) the relatively low variability ( a few orders of magnitude , compared to the variations by many dex of the density and temperature ) of thermal pressure is the cause or the effect of the interstellar density and velocity structure . a first piece of evidence was provided by the simulations of chiang & bregman ( 1988 ) , rosen & bregman ( 1995 ) , et al .
( 1995 ) and passot et al .
( 1995 ) , which contained no thermally - unstable temperature range ( although the passot et al .
simulations did contain a constant - pressure temperature range with thermal @xmath20 ) , yet they produced realistic clouds and intercloud structures that formed as a consequence of turbulent compressions induced by the stellar energy input .
the main differences with the mo model are that the clouds in the simulations do not only arise in shells around stellar sources , but instead the whole medium is in a turbulent state in which compressions are not necessarily the direct result of an expanding shell , and that , being formed by dynamical compressions , the clouds are continually changing shape , merging , getting stretched and disrupted ( `` morphing '' ) , rather than being `` confined '' in any way .
this is reflected in virial - balance analyses of numerical simulations , which show that the second time derivative of the moment of inertia of clouds in the simulations , far from being near zero , is generally much larger than the contributions from the thermal , kinetic , magnetic and gravitational energies , and instead has its major contribution from moment of inertia flux through cloud boundaries ( & 1997 ; shadmehri et al .
, this volume ) .
of course , simulations including self - gravity ( passot et al.1995 ; wada & norman 1999 , 2001 ) show the existence of gravitationally contracting large - scale structures ( especially since a small @xmath11 reduces the jeans length ; tohline , bodenheimer & christodolou 1988 ; elmegreen 1991 ; , passot & pouquet 1996 ) , as well as of collapsing small - scale ones .
meanwhile , the internal structure of the giant cloud complexes is continually reshaped by the stellar energy injection .
one could argue that the clouds are confined by turbulent pressure , but this would be misleading , as turbulence contains large - scale chaotic motions ( at the scale of the whole cloud ) which involve the distortion , and in general , transient character of the clouds ( et al.1999a ) .
an important consequence of this turbulent scenario is that it raises the possibility of molecular clouds having formation times ( of order a few myr ) shorter by factors of at least half an order of magnitude than previous estimates ( e.g. , blitz & shu 1980 ) , assuming they form from turbulent accumulation of neutral gas ( , hartmann & 1999b ; elmegreen 2000 ) . in the scenario of et al .
( 1999b ) , molecular clouds form from larger - scale converging flows in the hi gas that therefore have larger velocity dispersions than those normally associated to the molecular gas ( because larger scales have larger velocity dispersions in most turbulent flows ) . a key issue
, however , is that most of the accumulation process may occur in the atomic phase , and only when the column density has reached high enough values does the molecular phase appear ( hartmann , & bergin 2001 ) , eliminating concerns that in such short times not much accumulation can be achieved unless the density of the initial medium is already ( too ) large ( pringle , allen & lubow 2001 ) .
the short `` formation '' time scales refer only to the time after molecular gas appears . in this dynamic scenario ,
the relatively weak variability of the thermal pressure in the global ism is simply a consequence of cooling and heating functions that imply a slow variation of the thermal - equilibrium pressure with density , while the density field is driven by the turbulence , which in turn is powered by the presence of many local heating stellar sources . indeed , et al .
( 1996 ) noticed that , if the cooling times are short compared to the dynamical times and the cooling can be approximated as a ( possibly piece - wise ) power law , the gas follows a nearly polytropic behavior in which @xmath21 for a power - law cooling function , with @xmath22 for @xmath23 ( scalo et al .
note that a thermally unstable range would have @xmath24 .
certainly , the functional form of the pressure feeds back on the flow , but mainly to determine the response of the density field to the turbulent compressions , as evidenced by the dependence of density the pdf on @xmath11 ( passot & 1998 ; see 3.1 ) . of course
, the pressure near stellar injection sites is much larger than that on the average ism , as essentially they may be regarded as either lying on a different equilibrium between the cooling and the local stellar heating ( which is much larger than the diffuse background heating ) , or completely outside of thermal equilibrium , if the dynamical times are short enough . in the simulations of passot
( 1995 ) , which included only ob - star ionization - like heating , the pressure in `` hii '' regions was @xmath6 times larger than the average .
the simulations of mac low et al .
( 2001 ) , which included supernova energy input , exhibited pressure variations of up to three orders of magnitude .
however , the stellar heating is a relatively local phenomenon , which serves mostly as the forcing for the global ism ( avila - reese & 2001 ) , while the large fraction of the volume that is not under the direct influence of the sources has a pressure determined essentially by the turbulent density fluctuations through the applicable cooling and heating laws . on the other hand ,
ism simulations including a thermally unstable temperature range between the cold and warm `` phases '' show density pdfs without clear signs of bimodality , as would be expected for actual phase segregation ( , gazol & scalo 2000 ; wada & norman 2001 ; kritsuk & norman 2001 ) , and temperature pdfs which , although bimodal , contain significant amounts of gas in the thermally unstable range ( gazol et al .
2001 ; kritsuk & norman 2001 ; see also gerritsen & icke 1997 ) ; the same effect is seen in the plots of korpi et al .
( 1999 ) , although they did not discuss it .
the multimodality of the temperature pdf does not necessarily signal the existence of phases , but simply of heating and cooling processes which favor certain values of the temperature as a function of density .
the simulations of korpi et al .
( 1999 ) , which include supernova heating , and thus hot ( @xmath25 k ) gas , have roughly flat density pdfs except for a pronounced peak at the density of the hot gas , indicative of the relatively large relative filling factor of this temperature range , but still do not show significant signs of `` phase '' segregation .
for there to exist real phase segregation , a discontinuity must occur in one or more of the thermodynamic variables .
however , in the simulations there are no such discontinuities ( within the numerical resolution and smoothing of discontinuities ) , and instead the flow appears as a continuum , with the only discontinuities being dynamical shocks .
an exception is the case of the simulations by hennebelle & prault ( 1999 , 2000 ) and koyama & inutsuka ( 2000 ) , who studied the triggering of thermal instability by shocks passing through initially stable gas .
however , these authors did not consider the case of a fully turbulent ism in the presence of other forces besides the thermal pressure gradient .
instead , although further testing at higher resolution is still necessary , the simulations of the fully turbulent ism strongly suggest that the `` unstable '' range is actually quite populated in the ism .
a number of observational works point in the same direction ( e.g. , dickey , salpeter & terzian 1977 ; heiles 2001 , this volume ) , although these require further confirmation themselves ( see miville - deschenes , hennebelle & prault , this volume , for an opposing view ) .
the mechanism responsible for the at least partial inhibition of the thermal instability is not quite understood so far .
magnetic pressure possibly plays a role ( but see 3.1 ) , and velocity fluctuations may inhibit the instability if they have shorter time scales than the cooling time , a phenomenon which is most likely to occur at small spatial scales and/or low densities ( snchez - salcedo , & gazol 2002 ) .
it is interesting to note that a similar result has been recently reported for the parker instability , which apparently can be partially or completely suppressed by the presence of fluctuations in the magnetic field ( kim & ryu 2001 ) . also
, simple continuous recycling from one phase to another may imply the existence of a persistent population of gas traversing the unstable regime , as suggested by lioure & chize ( 1990 ) .
a complementary point of view is that of wada , spaans & kim ( 2000 ) , who have recently proposed that the pure gravitational instability , aided by thermal instability , is capable of the formation of large low density regions ( `` voids '' ) in the global ism , which , in their picture , correspond to ( some of ) the observed hi holes in galactic disks .
this is not inconsistent with the previous view of shaping the interiors of giant cloud complexes through the turbulence powered by stellar energy injection .
indeed , the largest cloud complexes may be gathered together by large - scale processes such as gravitational instabilities or the passage of spiral density waves , necessarily leaving behind , by simple mass conservation , large cavities which in fact must occupy large volumes , due to their lower densities . on the other hand , the stellar energy input ,
which is most likely dissipated within relatively short distances from the injection sites ( avila - reese & 2001 ) , can have mainly the role of shaping the complexes interiors .
wada et al .
distiguished between stellar - carved cavities , which should be filled with hot gas , and instability - carved ones , which should be filled mostly with warm gas .
the nonlinear development of gravitational instability in magnetized galactic disks has been recently studied by kim & ostriker ( 2001a , b ) .
these authors note that the amplification of non - axisymmetric perturbations in the presence of shear saturates in linear theory , and that thus any @xmath26 threshold for nonaxisymmetric gravitational runaway must originate from nonlinear effects . in this context , they numerically estimate the threshold values of @xmath26 in the nonlinear case .
they also distinguish between the mechanisms operating behind the modified swing amplification and the magneto - jeans instability . an interesting point to note
is that even in works without stellar energy injection , the outcome of the development of the various instabilities is often a chaotic , or turbulent , medium ( wada et al.2000 ; kim , ryu & jones 2001 ; kim & ostriker 2001a , b ; koyama & inutsuka 2001 ; kritsuk & norman 2001 ) , and so these large - scale instabilities can also be sources of the internal turbulence of the cloud complexes ( see also sellwood & balbus 1999 for an analytic discussion of the magnetorotational instability powering turbulence in hi disks ) . in general , most simulations including non - random prescriptions for star formation ( sf ) show that it self - propagates and , when other agents such as self - gravity and cooling are also included , spontaneous sf occurs too ( e.g. , chiang & prendergast 1985 ; et al .
moreover , as has been noted repeatedly , the stars in the simulations provide the energy feeding the gas motions , and a feedback cycle , analogous to that of oort ( 1954 ) , is established . however , the cycle is highly chaotic locally ( et al .
1995 , passot et al.1995 ; gazol - patio & passot 1999 ; wada & norman 2001 ; see also shore & ferrini 1995 ) , and depends on global quantities such as the mean magnetic field strength , and on stellar properties such as the energy deposited per stellar source ( passot et al . 1995 ; gerritsen & icke 1997 ; gazol - patio & passot 1999 ) .
other properties of the sf process , such as the slope of the two - point correlation function describing the clustering of young stars , can also be accounted for by this type of numerical models ( scalo & chappell 1999 ) .
additionally , chappell & scalo ( 2001 ) have studied the ratio of present to past - average sf rate as a function of various parameters in their simulations , finding , among other results , that starbursts can only occur when the past - average rate is low or the system is small , that the broad distribution of this ratio in late - type systems can be understood as a result of either a small size or a small metallicity , which imply that larger expanding shell column densities are required for gravitational instability , and that exponential tails in the velocity distribution are due to multiple shell interactions , not individual stellar winds .
it is worth noting here that the `` star formation instability '' identified by chiang & prendergast ( 1985 ) , which was one of the few attempts to investigate analytically the star - gas interplay within a hydrodynamic approach , has not been discussed in recent years , even though its effect may be present in all models with self - consistent prescriptions for star formation , possibly explaining its chaotic evolution .
another important connection that has not been considered in numerical models is the excitation of density and velocity perturbations in the gas by the gravitational effect of the stars ( kegel & vlk 1983 ) .
these issues certainly deserve further investigation .
in this review i have summarized a large body of results derived mainly from numerical simulations of the ism , which have allowed workers to capture many of the complex dynamical aspects of ism structure and evolution .
the numerical models have suggested a much more complex ism than a simple three - phase medium in pressure equilibrium . instead ,
a turbulent continuum has emerged , apparently without sharp phase segregation , and in which turbulent ram pressure is mostly responsible for cloud formation within the large complexes , while strong thermal pressure imbalances due to local stellar energy injection are responsible for powering the turbulence through expanding shells , although otherwise the near constancy of the thermal pressure has little effect in confining density structures , since _ the absence of a pressure gradient does not imply that inertial motions can not exist_. the medium is not simply a collection of overlapping shells with clouds forming in their compressed layers , but instead is globally turbulent , with shells and all structures in general `` morphing '' and merging into the global turbulence . at the largest scales ,
combined gravitational , magnetic and thermal instabilities appear to contribute , together with spiral density waves and supershells , to the formation of the largest complexes and perhaps voids , as well as possibly feeding the turbulence from the largest scales .
the individual structures being transient , the appropriate description for the global structure is statistical
. some statistical measures of the flow ( the density pdf and the energy spectrum ) depend on the cooling ability of the flow , which in turn determines its compressibility . since this in turn determines the ability to form gravitationally bound structures through compressions , and therefore stars which feed the turbulence , a full feedback loop clearly exists analogous to the old oort ( 1954 ) cycle .
however , as this review probably makes evident , the body of results is highly scattered , and a coherent , global theory of the ism is still lacking .
such a theory should be able to predict fundamental statistical indicators of the ism such as its topological properties ( i.e. , the statistical distribution of the mass and other physical quantities ) , the rate of production of collapsing objects ( the star formation rate and its efficiency ) and the mixing rate of the processed chemical elements , as well as the thermal and radiative properties of the various temperature and density regimes , all as a function of simple input physics , such as the total available mass and angular momentum , the atomic processes determining the cooling rates , and the energy injection per source . note that , in principle , even `` parameters '' such as the energy injection rate and the scale of injection , should be derivable from the theory ( or model ) , because the sources in this case are linked to the flow dynamics , as is the case of large - scale instabilities and of the star formation rate .
many of the physical variables are expected to be highly fluctuating locally , and the prediction of both average ( see , e.g. , blitz , this volume ) and typical fluctuation values is crucial .
simulations including all relevant physical ingredients ( self - gravity , disk rotation , magnetic fields , cosmic rays , stellar energy input , spiral density waves , chemistry ) and that can perform the necessary radiative transfer , all at high enough resolution , are needed .
so are careful experiments which allow disentangling the effects of all those ingredients , and of the average quantities versus those of the fluctuations .
consideration of the gravitational effect of the stars is also needed , which implies that hybrid gas+stars similations are needed .
then , continuous feedback to / from observations is essential through the production of synthetic observations of the numerical models .
a long , exciting way still lies ahead to a comprehensive theory that grasps the dynamics , thermodynamics and statistics of the ism with the same level of detail with which the time - dependent and the multi - phase models grasped the thermal and radiative issues .
i gratefully acknowledge pepe franco , alex lazarian , marco martos , thierry passot and keiichi wada for a critical reading of the manuscript and useful comments and precisions .
very specially , i thank john scalo , who pointed out various important topics , and helped in making sense of some others . this work has received partial financial support from conacyt grant 27752-e to the author , and has made extensive use of nasa s ads service .
adler , d.s . &
roberts , w.w . ,
1992 , apj 384 , 95 avila - reese , v. & , e. 2001 , apj 553 , 645 , j. & , e. 1997 , in star formation near and far : seventh astrophysics conference , eds .
holt & l.g .
mundy ( new york : aip press ) , p. 81
, j. , , e. & scalo , j. 1999a , apj 515 , 286 , , j. , hartmann , l. & , e. 1999b , apj 527 , 285 , j. & mac low , m .- m .
2001 , astro - ph/0108136 bania , t.m . &
lyon , j.g .
1980 , apj 239 , 173 blitz , l. & shu , f.h .
1980 , apj 238 , 148 blitz , l. 1993 , in protostars and planets iii , eds.e.h .
levy & j.i . lunine ( tucson : univ . of arizona press ) , p. 125 boulares , a. & cox , d.p .
1990 , apj 365 , 544 burkert , a. & mac low , m .- m .
2001 , astro - ph/0109447 burton , w.b .
1971 , 10 , 76 chappell , d. & scalo , j. 2001 , mnras 325 , 1 chiang , w .- h . & prendergast , k.h . 1985 ,
apj 297 , 507 chiang , w .- h . & bregman , j.n .
1988 , apj 328 , 427 cho , j. , lazarian , a. & vishniac , e. 2001 , astro - ph/0105235 cox , d.p . & smith , b.w . 1974 ,
apj 189 , l105 crutcher , r. , heiles , c. & troland , t. 2002 in simulations of magnetohydrodynamic turbulence in astrophysics , eds . e. falgarone & t. passot ( dordrecht : springer ) , in press de avillez , m.a .
& mac low , m .-
2001 , apj 551 , l57 deshpande , a a. 2000 , mnras 317 , 199 deshpande , a a. , dwarakanath , k.s .
& goss , w.m . 2000 , apj 543 , 227 dickey , j.m .
, salpeter , e.e . & terzian , y. 1977 , apj 211 , l77 dickey , j.m .
& lockman , f.j 1990 , araa 28 , 215 dickey , j.m .
, mcclure - griffiths , n. m. , stanimirovic , s. , gaensler , b. m. , green , a. j. 2001 , apj 561 , 264 elmegreen , b.g .
1991 , apj 378 , 139 elmegreen , b.g .
1993 , apj 419 , l29 elmegreen , b.g .
1994 , apj 433 , 39 elmegreen , b.g .
2000 , apj 530 , 277 falgarone , e. & phillips , t.g .
1990 , apj 359 , 344 falgarone , e. , lis , d. c. , phillips , t. g. , pouquet , a. , porter , d. h. , woodward , p. r. 1994 apj 436 , 728 field , g.b .
1965 , apj 142 , 531 field , g.b . ,
goldsmith , d.w . &
habing , h.j .
1969 , apj 155 , l149 franco , j. , kim , j. , alfaro , e.j . &
hong , s.s .
2001 , astro - ph/0111406 frisch , u. 1995 , turbulence ( cambridge : cambridge university press ) gammie , c.f .
& ostriker , e.c .
1996 , apj 466 , 814 gazol - patio , a. & passot , t. 1999 , apj 518 , 748 gazol , a. , , e. , snchez - salcedo , f.j . &
scalo , j. 2001 , apj 557 , l121 gerola , h. , kafatos , m. & mccray , r. 1974 , apj 189 , 55 gerritsen , j.p.e . &
icke , v. 1997 , 325 , 972 goldreich , p. & sridhar , s. 1995 , apj 438 , 763 hartmann , l. , j. & bergin , e.a .
2001 , astro - ph/0108023 heiles , c. 2001 , apj 551 , l105 hennebelle , p. & prault , m. 1999 , 351 , 309 hennebelle , p. & prault , m. 2000 , 359 , 1124 henney , w.j . & , e. 2002 , in preparation higgs , l.a .
1999 , in new perspectives on the interstellar medium , eds .
taylor , t.l .
landecker & g. joncas ( san francisco : asp ) , p. 15
kadomtsev & petviashvili 1973 , sov .
18 , 115 kegel , w.h .
& vlk , h.j .
1983 , 119 , 101 kim , j. & ryu , d. 2001 apj 561 , l135 kim , j. , ryu , d. & jones , t. 2001 , apj 557 , 464 kim , w.t . & ostriker , e.c .
2001a , apj 559 , 70 kim , w.t . & ostriker , e.c .
2001b , astro - ph/0111398 klessen , r. 2000 , apj 535 , 869 korpi , m. j. , brandenburg , a. , shukurov , a. , tuominen , i. & nordlund , .
1999 , apj 514 , l99 koyama , h. & inutsuka , s .-
2000 , apj 532 , 980 koyama , h. & inutsuka , s .- i .
2001 , astro - ph/0112420 kritsuk , a. & norman , m.l .
2001 , astro - ph/0112437 lazarian , a. & pogosyan , d. 2000 , apj 537 , 720 lazarian , a. , pogosyan , d. , , e. & pichardo , b. 2001 , apj 555 , 130 lioure , a. & chize , j .-
p . 1990 , , 235 , 379 lis , d. c. , pety , j. , phillips , t. g. & falgarone , e. 1996 , apj 463 , 623 lis , d. c. , keene , j. , li , y. , phillips , t. g. & pety , j. 1998 , apj 504 , 889 mac low , m .-
m . , balsara , d. , de avillez , m.a .
& kim , j. 2001 , astro - ph/0106509 mac low , m .-
& klessen , r. 2002 , araa , in preparation martos , m. & cox , d. p. 1998 , apj 509,703 martos , m. , allen , c. , franco , j. & kurtz , s. , apj 1999 , 526 , l89 mckee , c.f . & ostriker , j.p .
1977 , apj 218 , 148 ( mo ) mckee , c.f .
1995 , in the physics of the interstellar medium and the intergalactic medium , eds .
a. ferrara , c.f .
mckee , c. heiles & p.r .
shapiro ( san francisco : asp ) , p. 292
miesch , m.s .
& scalo , j. 1995 , apj 450 , l27 miesch , m.s . ,
scalo , j. & bally , j. 1999 , apj 524 , 895 minter , a.h . , spangler , s.r . , 1996 , apj 458 , 194 nordlund , .
& padoan , p. 1999 , in interstellar turbulence , eds .
j. franco & a. carramiana ( cambridge : cambridge university press ) , p. 218
norman , c.a . &
ferrara , a. 1996 , apj 467 , 280 oort , j.h .
1954 , bull .
12 , 177 ostriker , e.c . ,
stone , j.m . &
gammie , c.f .
2001 , apj 546 , 980 padoan , p. & nordlund , .
1999 , apj 526 , 279 padoan , p. & nordlund , . 2000 , astro - ph/0011465 parker , e.n .
1966 , apj 145 , 811 passot , t. , e. & pouquet , a. 1995 , apj 455 , 536 passot , t. & , e. 1998 , phys .
e 58 , 4501 passot , t. & , e. 2002 , in preparation pichardo , b. , , e. , gazol , a. , passot , t. & , j. 2000 , apj 532 , 353 porter , d.h . ,
pouquet , a. & woodward , p.r .
1992 , 68 , 3156 pringle , j. e. , allen , r.j .
& lubow , s. h. 2001 , mnras 327 , 663 rosen , a. & bregman , j.n .
1995 , apj 440 , 634 rosen , a. , bregman , j.n . & kelson , d.d .
1996 , apj 470 , 839 saffman , p.g .
1968 , in topics in nonlinear physics , ed.n.j .
zabusky ( new york : springerverlag , p. 485
snchez - salcedo , f.j .
, e. & gazol , a. 2002 , apj submitted santilln , a. , kim , j. , franco , j. , martos , m. , hong , s.s . & ryu , d. 2000 , apj , 545 , 353 scalo , j. 1987 , in interstellar processes , eds.d.j .
hollenbach & h.a .
thronson ( dordrech : reidel ) , p. 349 scalo , j. , , e. , chappell , d. & passot , t. 1998 , apj 504 , 835 scalo , j. & chappell , d. 1999 , apj 510 , 258 sellwood , j.a . , balbus , s.a . 1999 ,
apj 511 , 660 shore , s.n . &
ferrini , f. 1995 , fund .
16 , 1 stanimirovic , s. & lazarian , a. 2001 , apj 551 , l53 taylor , a.r . 1999 , in new perspectives on the interstellar medium , eds .
taylor , t.l .
landecker & g. joncas ( san francisco : asp ) , p. 3 tohline , j. e , bodenheimer , p. h. , & christodoulou , d. m. 1988 , apj 322 , 787 , e. 1994 , apj 423 , 681 , e. , passot , t. & pouquet , a. 1995 , apj 441 , 702 , e. , passot , t. & pouquet , a. 1996 , apj 473 , 881 , e. & passot , t. 1999 , in interstellar turbulence , eds.j .
franco & a. carramiana ( cambridge : cambridge university press ) , p. 223 , e. 1999 , in millimeter - wave astronomy : molecular chemistry & physics in space , eds .
w. f. wall , a. carramiana & l. carrasco ( dordrecht : kluwer ) , p. 161 , e. , ostriker , e. c. , passot , t. , gammie , c. f. & stone , j. m. 2000 , in protostars and planets iv , eds mannings , v. , boss , a.p . & russell , s. s. ( tucson : university of arizona press ) , p. 3
, e. , gazol , a. & scalo , j. 2000 , apj 540 , 271 wada , k. & norman , c.a .
1999 , apj 516 , l13 wada , k. , spaans , m. & kim , s. 2000 , apj 540 , 797 wada , k. 2001 , apj 559 , l41 wada , k. & norman , c.a .
2001 , apj 547 , 172 wada , k. & koda j. 2001 , astro - ph/0110228 wolfire , m. g. , hollenbach , d. , mckee , c. f. , tielens , a. g. g. m. , bakes , e. l. o. 1995 , apj 443 , 152 | i review recent results from numerical simulations on the structure and dynamics of the ism , and attempt to put together a coherent dynamical scenario . in particular , i discuss results on 1 ) the spatial distribution of the gas components , showing that reasonable agreement between simulations and observations exists , but noting that in most models the components are simply defined as temperature intervals , because distinct thermodynamic `` phases '' do not arise ; 2 ) some statistical issues of the physical fields , like the dependence of the one - point statistics of the density field on the effective equation of state of the gas , the poor correlation of magnetic strength with density , the energy spectrum in weakly and highly compressible cases , and the one point statistics of the velocity field ; 3 ) the effects of spectroscopic observation on distorting the physical structures and results from synthetic observations of the simulations , and 4 ) several dynamical and thermodynamical issues , such as the ( apparently minor ) role of the thermal instability in forming and confining clouds , the continuous , rather than abrupt , transition between `` phases '' , which in turn may be consequences of the dynamics rather than the agents controlling it , the possibility of short time scales ( @xmath0 a few myr ) for molecular cloud formation , and the star - gas connection , mentioning that the models generally exhibit self - propagating star and cluster formation , while the stars may drive the medium- and small - scale gas motions , and that a `` star formation instability '' may induce chaotic behavior of the star formation rate locally .
i conclude with a round - up view , and a discussion of the work needed ahead .
# 1_#1 _ # 1_#1 _ = # 1 ' # 1#1i"13i 1.25 in .125 in .25 in |
universal spectral bounds for laplace and schrdinger operators , i.e. , bounds that control eigenvalues with expressions that do not depend on the specific geometry of the domain or on details of the potential ( cf .
@xcite ) , can be derived from fundamental identities involving traces of operators and their commutators @xcite .
this insight has proved useful both for unifying numerous previously known inequalities of this kind and for discovering new ones @xcite .
related methods have also been used to obtain control on the spectrum of laplace and schrdinger operators in terms of curvature @xcite .
many of the universal inequalities related to trace identities are sharp in the sense that they are saturated for particular examples : for the schrdinger operators treated in @xcite the upper bounds on eigenvalue gaps @xmath5 become identities for all @xmath6 in the case of the harmonic oscillator , while for laplacians on embedded manifolds discussed in @xcite all of the gap bounds become identities for embedded spheres .
more recently , in some circumstances ( e.g. , @xcite ) universal bounds on moments of eigenvalues have been connected to `` semiclassical '' theorems about the spectrum such as asymptotic behavior as the index @xmath7 and nonasymptotic bounds in the spirit of the berezin - li - yau inequality @xcite .
one of the motivations of this work is to sharpen the understanding of moments of eigenvalues and of riesz means of the spectrum , which plays something like the role of a dual version of moments . among the applications of our analysis
will be a family of differential inequalities for functions determined by the spectrum , extending the analysis of @xcite . by legendre duality as in @xcite
these imply bounds on ratios of averages of eigenvalues .
we also introduce a novel type of inequality relating arithmetic and geometric means of eigenvalues .
a second motivation is to better unify the subject of universal bounds with formally analogous semiclassical spectral theorems . in the next section we present some more abstract versions of the essential trace identity of @xcite for a class of self - adjoint operators @xmath8 enjoying algebraic properties modeled on those of of schrdinger operators .
we also identify a special family of functions for which @xmath9 can be sharply controlled , _ viz . _ : [ sj ] let @xmath8 be a self - adjoint operator and let @xmath10 be a distinguished subset of the spectrum .
we let @xmath11 denote the smallest closed interval containing @xmath12 .
a @xmath13 function @xmath14 belongs to the set @xmath15 of _ trace - controllable functions _ provided that on @xmath11 , h1 .
: : @xmath16 ; h2 .
: : @xmath17 ; h3 .
: : @xmath18 is concave ; h4 .
: : if @xmath19 , there exists @xmath20 such that @xmath21 is nondecreasing in @xmath22 ; h5 .
: : @xmath23 , where @xmath24 denotes the spectral projector for the set @xmath12 . for reasons of parsimony we shall sometimes assume only a subset of the hypotheses in the statements of some theorems .
the model situation is that at least the lower part of the spectrum consists of eigenvalues @xmath25 and @xmath26 . among familiar functions in @xmath15 we mention @xmath27 and @xmath28 with @xmath29 .
note that conditions h1h4 are preserved by multiplication @xmath30 and that conditions h1h3 are preserved by compositions in the form @xmath31 .
the only condition in definition [ sj ] that may not be familiar is h4 , so we observe some sufficient conditions for its validity , depending on some elementary facts about concavity , in particular , [ cvxprop ] if the function @xmath32 is concave for @xmath33 , then @xmath34 is concave on the same interval .
we observe that this is immediate when @xmath35 by a calculation of the second derivative .
we give a proof without this hypothesis .
recall that a function f is concave on an interval @xmath36 iff its right and left derivatives exist at all interior points of @xmath36 and @xmath37 is nonincreasing , in the extended sense that if the right and left derivatives differ at @xmath38 , then @xmath39 . ( for this and other basic facts about concave functions see chapter 5 of @xcite . )
therefore we compare the derivative @xmath40 at @xmath41 ( right derivative ) and @xmath42 ( left derivative ) , for @xmath43 .
( we shall not complicate the notation by distinguishing right and left derivatives in the following calculation . )
@xmath44 by the mean value theorem of convex functions , for some @xmath45 , @xmath46 , and thus the final term is @xmath47 .
[ gfunction ] suppose that @xmath48 satisfies hypotheses h2 and h3 of definition [ sj ] and that h4@xmath49 there exists @xmath20 such that @xmath50 then @xmath48 satisfies hypothesis h4 .
we study the function @xmath51 occurring in hypothesis h4 .
with the change of variable @xmath52 , @xmath53 satisfies the conditions of proposition [ cvxprop ] , so @xmath54 is concave for positive @xmath38 .
but this expression evaluates to @xmath55 , establishing that @xmath51 is concave for @xmath56 .
therefore @xmath57 is nonincreasing . at the same time we know by h4@xmath49 that @xmath58 , so it follows that @xmath59 on @xmath11 .
we consider a self - adjoint operator @xmath8 with domain @xmath60 on a hilbert space @xmath61 with scalar product @xmath62 .
we suppose that @xmath8 has nonempty point spectrum , and that @xmath63 is a finite - dimensional subspace of @xmath61 spanned by an orthonormal set @xmath64 of eigenfunctions of @xmath8 .
we further let @xmath65 denote the spectral projector associated with @xmath8 and a borel set @xmath66 , and @xmath67 .
we refer to @xcite for terminology , notation , and details about the spectral theorem .
[ main theorem 1 ] let @xmath8 and @xmath68 be self - adjoint operators with domains @xmath60 and @xmath69 such that @xmath70 . then , for any real - valued @xmath13-function @xmath48 the derivative @xmath71 of which is a concave function @xmath72i.e .
, hypothesis h3 of definition [ sj ] ) , @xmath73\phi_j,[h , g]\phi_j\rangle + f(\lambda_j)\,\langle[g , [ h , g],]\phi_j,\phi_j\rangle\\ & \qquad \leq \sum_{\lambda_j\in j } \int { \big(f(\lambda_j)+\frac1{2}f'(\lambda_j)(\kappa-\lambda_j)\big)(\kappa-\lambda_j)|\langle g\phi_j , dp_{\kappa } p_{{\mathcal j}^c } g\phi_j \rangle|^2}\nonumber\end{aligned}\ ] ] in case the spectrum of @xmath8 is purely discrete , we may write the inequality as @xmath74\phi_j,[h , g]\phi_j\rangle + f(\lambda_j)\,\langle[g , [ h , g],]\phi_j,\phi_j\rangle\\ & \qquad \leq \sum_{\lambda_j\in j}\sum_{\lambda_k\in j^c } \big(f(\lambda_j)+\frac1{2}f'(\lambda_j)(\lambda_k-\lambda_j)\big)(\lambda_k-\lambda_j)|\langle g\phi_j,\phi_k\rangle|^2.\nonumber\end{aligned}\ ] ] the proof will use the following lemma : [ cvxlem ] let @xmath75 such that @xmath71 is a concave function .
then for all @xmath76 @xmath77 by the fundamental theorem of calculus and the concavity of @xmath71 we have @xmath78 we begin with an observation that is an abstract version of what is known in quantum theory as the oscillator - strength sum rule of thomas , reiche , and kuhn @xcite : by a straightforward calculation , the self - adjoint operators @xmath79 satisfy @xmath80\phi_j,\phi_j\rangle = 2 \langle ( h - \lambda_j ) g\phi_j , g\phi_j\rangle,\ ] ] which , with the spectral resolution , equals @xmath81 thus @xmath82\phi_j,\phi_j\rangle = \int_{}{(\kappa-\lambda_j)dg_{j \kappa}^2},\ ] ] where @xmath83 .
when @xmath84 for @xmath85 , we also write the discrete matrix elements as @xmath86 .
multiplying by @xmath87 and summing over @xmath88 in @xmath12 , we get @xmath89\phi_j,\phi_j\rangle = \sum_{\lambda_j\in j}\int _ { } f(\lambda_j)(\kappa-\lambda_j)dg_{jd}^2\\ & \qquad = \sum_{\lambda_j\in j}\sum_{\lambda_k\in j } f(\lambda_j)(\lambda_k-\lambda_j)g_{jk}^2+\sum_{\lambda_j\in j}\int_{\kappa \in j^c } f(\lambda_j)(\kappa-\lambda_j)dg_{j \kappa}^2.\nonumber\end{aligned}\ ] ] using the symmetry of the matrix elements @xmath90 we rewrite the first double sum as follows : @xmath91 applying lemma [ cvxlem ] and once again using the symmetry of the @xmath90 we get @xmath92 at the same time , the pair @xmath79 satisfies the trace formula @xmath93\phi_j,[h , g]\phi_j\rangle.\ ] ] multiplying by @xmath94 and summing over @xmath95 we get @xmath96\phi_j,[h , g]\phi_j\rangle \\ & \quad = -\frac1{2}\sum_{\lambda_j\in j}\int { f'(\lambda_j)(\kappa-\lambda_j)^2 dg_{j \kappa}^2}\nonumber\\ & \quad = -\frac1{2}\sum_{\lambda_j\in j}\sum_{\lambda_k\in j } f'(\lambda_j)(\lambda_k-\lambda_j)^2g_{jk}^2\nonumber\\ & \qquad-\frac1{2}\sum_{\lambda_j\in j}\int_{\kappa \in j^c } f'(\lambda_j)(\kappa-\lambda_j)^2 dg_{j \kappa}^2.\nonumber\end{aligned}\ ] ] combining , and yields the statement of the theorem .
[ quadratic function ] if @xmath97 , then holds with equality .
in particular , for any @xmath98 we have @xmath99\phi_j,\phi_j\rangle - 2(z-\lambda_j)\,\langle[h , g]\phi_j,[h , g]\phi_j\rangle } \\ & \qquad = 2 \sum_{\lambda_j\in j}\int_{\kappa \in j^c } ( z-\lambda_j)(z-\kappa)(\kappa-\lambda_j ) d g_{j \kappa}^2.\nonumber\end{aligned}\ ] ] equation is a particular case of a more general trace identity that will be explored in a future work .
we now add for the first time the assumption that the lower part of the spectrum of @xmath8 is discrete and denote this set @xmath100 .
the following theorem captures a universal relationship between the lower part of the spectrum for @xmath101 and the values of @xmath102 and @xmath103 .
[ inequalities for bottom of spectrum ] let @xmath8 and @xmath68 be self - adjoint operators with domains @xmath60 and @xmath69 such that @xmath70 . let the subset @xmath100 lie below the rest of the spectrum of @xmath8 . then for any @xmath104 , with @xmath105 , @xmath106\phi_j,[h , g]\phi_j\rangle + f(\lambda_j)\,\langle[g , [ h , g]]\phi_j,\phi_j\rangle\right)\\ & \qquad \leq \frac1{2}\big(f(\lambda_n)+\frac1{2}f'(\lambda_n)(\lambda_{n+1}-\lambda_n)\big)\sum_{j=1}^n\langle[g , [ h , g]]\phi_j,\phi_j\rangle.\nonumber\end{aligned}\ ] ] consider the right side of . since @xmath107 and @xmath108 , we have @xmath109 to prove it suffices to show that @xmath110 is nondecreasing so @xmath111 can be replaced with @xmath112 . as a concave function
, @xmath113 is nondecreasing on @xmath11 , so this is true because of the assumption that @xmath114 . under slightly weakened assumptions on @xmath48
we get the following ( weaker ) inequality : [ weaker inequalities for bottom of spectrum ] let @xmath8 and @xmath68 be self - adjoint operators with domains @xmath60 and @xmath69 such that @xmath70 .
let the subset @xmath100 lie below the rest of the spectrum of @xmath8 .
then , for any real @xmath13-function @xmath48 satisfying hypothesis h2 with @xmath115 , @xmath116\phi_j,[h , g]\phi_j\rangle + f(\lambda_j)\,\langle[g , [ h , g]]\phi_j,\phi_j\rangle\right)\\ & \qquad \leq \frac1{2}f(\lambda_{n+1})\sum_{j=1}^n\langle[g , [ h , g]]\phi_j,\phi_j\rangle.\nonumber\end{aligned}\ ] ] in applications to laplace , schrdinger , and similar differential operators , the commutators typically simplify as follows : there are constants @xmath117 with @xmath118 , such that @xmath119,\quad \beta h+\alpha \,\ , \ge\ , -[h , g]^2.\ ] ] ( recall that @xmath120 ^ 2 = [ h , g ] [ h , g]^ * \ge 0 $ ] . ) for instance , see @xcite .
therefore we also may assume without loss of generality that @xmath8 has only nonnegative eigenvalues .
indeed , if @xmath121 for some real constant @xmath122 , then @xmath123 for the model case of the dirichlet laplacian @xmath124 on a domain @xmath125 in @xmath126 there is a choice of cartesian system coordinate system for such that with @xmath127 , @xmath128 , and @xmath129 . in the literature these same effective constants are often obtained by averaging over all the coordinates , but the coordinate system can always be chosen to make this unnecessary .
values of @xmath130 arise for several reasons . in the case of a schrdinger operator @xmath131
, the potential energy disappears from all commutators , and the term @xmath132 ^ 2 $ ] is typically dominated by the kinetic energy term , i.e. , @xmath120 ^ 2 \le \beta ( - \delta)$ ] rather than @xmath133 . in an elementary way , the addition of @xmath134 can compensate for the absence of @xmath135 if , say , the negative part of @xmath135 is bounded .
even if the negative part of the potential @xmath135 is unbounded , if it lies in certain function classes , there are constants @xmath136 and @xmath137 such that for all functions @xmath138 in the quadratic - form domain of @xmath8 , @xmath139 in which case @xmath140 , and consequently @xmath141 examples of function classes guaranteeing the estimate are that @xmath142 is a rollnik potential in three dimensions or that @xmath143 with @xmath144 when @xmath145 .
for discussion of these conditions refer to @xcite .
another instance where @xmath130 is of interest is in the case of laplace or schrdinger operators on hypersurfaces @xmath146 in @xmath147 . by letting @xmath68
be the cartesian coordinate @xmath148 in the ambient space , and choosing the orientation of the coordinate system appropriately ( or averaging over all coordinates ) , @xmath149 ^ 2 = - 4\delta + h^2(x),\ ] ] where @xmath32 is the sum of the principal curvatures at the point @xmath150 and @xmath151 now denotes the laplace - beltrami operator on @xmath146 @xcite .
our estimates therefore apply to laplace - beltrami operators on @xmath146 with a @xmath152 .
schrdinger operators on @xmath146 will require @xmath134 to be the sum of this curvature effect and any contribution owing to the negative part of @xmath135 .
the situation is analogous for laplace or schrdinger operators on manifolds immersed in other symmetric spaces @xcite . under these conditions
theorem [ inequalities for bottom of spectrum ] is simplified : [ corollary to inequalities for bottom of spectrum ] if in addition to the assumptions of theorem [ inequalities for bottom of spectrum ] the relations hold , then @xmath153 if , in addition , the spectrum of @xmath8 is purely discrete and all sums over the full spectrum @xmath154 are finite , then @xmath155 in the following sections we apply corollary [ corollary to inequalities for bottom of spectrum ] for appropriate functions @xmath156 .
in this section we prove various inequalities for eigenvalues under the assumptions of corollary [ corollary to inequalities for bottom of spectrum ] for appropriate functions @xmath156 , and we restrict ourselves to operators @xmath8 with purely discrete spectrum . as a first result we generalize the result of @xcite on the partition function @xmath157 .
let @xmath48 and @xmath8 satisfy the assumptions of corollary [ corollary to inequalities for bottom of spectrum ] and suppose that @xmath158 and @xmath159 are finite for all @xmath160 , then @xmath161 is nonincreasing .
if @xmath156 satisfies the assumptions of theorem [ main theorem 1 ] , then @xmath162 satisfies the same assumptions for any @xmath160
. then inequality of corollary [ corollary to inequalities for bottom of spectrum ] reads as follows : @xmath163 which proves the proposition .
the proposition applies to @xmath164 , for any @xmath165 and therefore [ weigthed heat kernel ] if @xmath166 is finite for all @xmath160 and @xmath8 satisfies the assumptions of corollary [ corollary to inequalities for bottom of spectrum ] , then @xmath167 is nonincreasing .
in particular , corollary [ weigthed heat kernel ] shows that @xmath168 for all @xmath169 . as a second application
, we shall show that certain moments of eigenvalues are dominated by their geometric mean .
let @xmath170 be a parameter to be chosen later and @xmath171 such that @xmath172 and @xmath173 .
for @xmath174 $ ] the function @xmath175 defined by @xmath176 satisfies the assumptions of theorem [ inequalities for bottom of spectrum ] and corollary [ corollary to inequalities for bottom of spectrum ] for all @xmath177 $ ]
. indeed , with @xmath178 we see that @xmath179 if @xmath180 , using the estimate @xmath181 .
furthermore , @xmath182 since @xmath183 . as in the proof of theorem [ inequalities for bottom of spectrum ]
we show that @xmath184 if @xmath185 , and hence @xmath186 provided that @xmath177 $ ] . we define @xmath187 applying corollary [ corollary to inequalities for bottom of spectrum ] with @xmath188 as an upper bound , we have @xmath189 for all
@xmath177 $ ] . since @xmath190 where @xmath191 we see that @xmath192 attains a global nonnegative minimum at @xmath193 .
therefore we have the following result : [ main theorem on moments ] let @xmath171 such that @xmath172 and @xmath173 .
then for all @xmath6 we have @xmath194 in particular , for @xmath195 the function @xmath196 is nonincreasing , and for all @xmath197 we have @xmath198 where @xmath199 denotes the geometric mean of the first @xmath6 eigenvalues .
inequality is obtained from by taking the limit @xmath200 .
we note that @xmath201 for all @xmath6 .
recently , it was shown in @xcite that riesz means of eigenvalues of the laplacian on bounded domains satisfy differential inequalities , which in turn imply universal eigenvalue bounds that are sharp in the sense of having the correct behavior as expected from the weyl law for @xmath102 as @xmath4 .
we note here how some of the results of @xcite can be extended as a consequence of theorem [ main theorem 1 ] .
let @xmath48 be a function in the class @xmath15 such that @xmath202 .
we define a generalized riesz mean by @xmath203 where @xmath204 if @xmath205 and zero otherwise and @xmath48 satisfies the hypotheses of theorem [ main theorem 1 ] . for simplicity
we consider the case @xmath206 , which can always be arranged by shifting @xmath207 .
[ corollary to generalized riesz means ] if in addition to the assumptions of theorem [ inequalities for bottom of spectrum ] the relations hold , then @xmath208 in @xcite the riesz means @xmath209 for @xmath210 have been studied for the dirichlet problem .
we obtain these means from corollary [ corollary to generalized riesz means ] choosing @xmath211 and putting @xmath212 .
we therefore have [ generalization of hahe ] suppose that the pair @xmath79 satisfy the relations .
then @xmath213 is a nondecreasing function for @xmath214 .
formula is identical in form to an inequality in @xcite ; the constant @xmath215 has simply replaced @xmath216 in the earlier article . as a consequence
we obtain a weyl - sharp bound , @xmath217 provided that @xmath218 .
( in case the constant @xmath134 has not been set to @xmath188 , the averages @xmath219 on the left side are both replaced by @xmath220 . )
for the dirichlet laplacian @xmath124 on a domain @xmath125 in @xmath126 such that @xmath8 has only eigenvalues , we put @xmath221 , the multiplication operator by a suitable cartesian coordinate .
as shown for example in @xcite we then have @xmath206 , @xmath222 and @xmath223 . for the dirichlet laplacian
the inequality of our main theorem [ inequalities for bottom of spectrum ] reads @xmath224 we claim that all estimates of the form are sharp in the semiclassical limit . indeed , recall that according to the weyl law , on any bounded domain the semiclassical limit of the eigenvalue @xmath102 is given by @xmath225 as @xmath226 , where @xmath135 denotes the volume of @xmath125 . in terms of the counting function
@xmath227 this is equivalent to @xmath228 now , for any function @xmath48 , we have @xmath229 from which it easily follows that @xmath230 consequently , theorem [ main theorem on moments ] for the moments @xmath231 is sharp . since the semiclassical limit of @xmath232 does not depend on @xmath233 and is given by @xmath234 as @xmath226 , phase - space bounds on @xmath235 for @xmath236 follow from the berezin - li - yau bound @xcite for @xmath237 by the monotonicity property
. we can further refine theorem [ main theorem on moments ] for the dirichlet laplacian by exploiting the right side of inequality of theorem [ inequalities for bottom of spectrum ] , or respectively corollary [ corollary to inequalities for bottom of spectrum ] , in order to relate the arithmetic and geometric means of eigenvalues to the sizes of the eigenvalue gaps .
we begin by choosing @xmath238 for @xmath197 , which corresponds to the choice @xmath239 in .
therefore @xmath175 in satisfies the assumptions of theorem [ inequalities for bottom of spectrum ] and corollary [ corollary to inequalities for bottom of spectrum ] for all @xmath98 in an interval of the form @xmath240 $ ] for some @xmath241 which is determined by the condition @xmath242 . defining @xmath243 we find @xmath244 defining @xmath245 and @xmath246 we have @xmath247 for all @xmath248 $ ] .
we see that @xmath249 has a global minimum at @xmath250 . as
the global minimum of @xmath192 is below the global minimum of @xmath249 , we obtain the inequality @xmath251 we note that the left side of can be bounded above by @xmath252 which yields an explicit upper bound on the gap @xmath253 . however , we find it more convenient to optimize the inequality @xmath254 with respect to @xmath98 .
the right side has then a global minimum at @xmath255 , while the left side has its global minimum at @xmath256 . after taking the exponential on both sides we therefore obtain the inequality @xmath257 extending the above discussion to all pairs @xmath258 of theorem [ main theorem on moments ] we obtain the following refinement for the dirichlet laplacian : [ main theorem on moments dirichlet laplacian ]
let @xmath171 such that @xmath172 and @xmath173 .
then for all @xmath6 we have @xmath259 in particular , for @xmath195 the function @xmath260 is nonincreasing .
ashbaugh , m.s . : the universal eigenvalue bounds of payne - plya - weinberger , hile - protter , and h. c. yang . in spectral and inverse spectral theory ( goa , 2000 ) , proc .
indian acad .
112 , 330 ( 2002 ) | we prove trace inequalities for a self - adjoint operator on an abstract hilbert space .
these inequalities lead to universal bounds on spectral gaps and on moments of eigenvalues @xmath0 that are analogous to those known for schrdinger operators and the dirichlet laplacian , on which the operators of interest are modeled . in addition
we produce inequalities that are new even in the model case .
these include a family of differential inequalities for generalized riesz means and theorems stating that arithmetic means of @xmath1 for @xmath2 are universally bounded from above by multiples of the geometric means , @xmath3 . for schrdinger operators and the dirichlet laplacian these bounds are weyl - sharp , i.e. , saturated by the standard semiclassical estimates as @xmath4 . |
a number of brief accounts of the modern application of dyson - schwinger equations ( dses ) to hadron physics and qcd have recently appeared @xcite .
they have recapitulated on aspects of dynamical chiral symmetry breaking and confinement , and the expression of these phenomena in the hadron spectrum ; and contain original contributions regarding : comparisons between dse results and lattice simulations , and the mutual benefits of such comparisons @xcite ; the pion and its excited states @xcite ; and states with vacuum quantum numbers @xcite .
we have a modicum to add .
we begin with qcd s renormalised gap equation @xmath1 wherein : @xmath2 is the dressed - gluon propagator ; @xmath3 is the dressed - quark - gluon vertex ; @xmath4 is the @xmath5-dependent current - quark bare mass ; and @xmath6 represents a translationally - invariant regularisation of the integral , with @xmath5 the regularisation mass - scale .
we employ a pauli - villars scheme , which is implemented in eq.([gendse ] ) by considering the quarks as minimally anticoupled ( @xmath7 ) to additional massive gluons ( @xmath8 ) .
this effects a tempering of the integrand that is expressed via a modification of the gluon propagator s ultraviolet behaviour : @xmath9 through which the integral s superficial linear divergence is regulated .
the regularising mass - scale is naturally removed to infinity as the last stage in the calculation of any renormalisable quantity . in eq.([gendse ] ) , @xmath10 are respectively the quark - gluon - vertex and quark wave function renormalisation constants , which depend on @xmath5 and the renormalisation point , @xmath11 , as does the mass renormalisation constant @xmath12 .
the gap equation s solution has the form @xmath13 , \label{sinvp}\ ] ] where @xmath14 is the running quark mass , . ] and is obtained subject to the renormalisation condition @xmath15 the renormalisation scheme should be understood to yield mass - independent renormalisation constants .
the gap equation illustrates the features and flaws of each dse .
it is a nonlinear integral equation for @xmath16 and hence can yield much - needed nonperturbative information .
however , the kernel involves the two - point function @xmath2 and the three - point function @xmath3 .
the gap equation is therefore coupled to the dses these functions satisfy .
those equations in turn involve higher @xmath17-point functions and hence the dses are a tower of coupled integral equations with a tractable problem obtained only once a truncation scheme is specified .
it is unsurprising that the best known truncation scheme is the weak coupling expansion , which reproduces every diagram in perturbation theory .
this scheme is systematic and valuable in the analysis of large momentum transfer phenomena because qcd is asymptotically free but it precludes any possibility of obtaining nonperturbative information , which we identified as a key aspect of the dses .
the gap equation describes the manner in which the propagation of a quark is affected by the medium being traversed , and the same is true of the gauge sector gap equations satisfied by gluon and ghost schwinger functions .
the bethe - salpeter equations are the necessary complement as they provide the means by which one may understand how qcd s elementary excitations form the colour singlet bound states that are the observable manifestation of strong interaction dynamics .
an important example is the inhomogeneous bethe - salpeter equation for the axial - vector vertex : @xmath18_{tu } = z_2 \left[\gamma_5 \gamma_\mu \frac{\tau^j}{2}\right]_{tu } + \int^\lambda_q \left [ \chi_{5\mu}(q;p ) \right]_{sr } \
, k_{tu}^{rs}(q , k;p ) \,,\ ] ] where @xmath19 with @xmath20 $ ] and , plainly , we focus on @xmath21 so that @xmath22 are pauli matrices in flavour space . the conventions of eq.([avbse ] ) describe a vertex with total momentum @xmath23 and relative momentum @xmath24 , whose outgoing quark and antiquark legs carry momenta @xmath25 and @xmath26 , respectively .
the appearance here of @xmath27 $ ] expresses the arbitrariness in defining the relative momentum in a poincar covariant framework : physical observables will be independent of @xmath28 in a symmetry preserving truncation , as seen , e.g. , in ref.@xcite . of particular importance in eq.([avbse ] ) is @xmath29 , the fully amputated dressed quark - antiquark scattering kernel .
this four - point schwinger function also appears implicitly in qcd s gap equation because it is one piece of the kernel in the inhomogeneous integral equation satisfied by @xmath30 .
a primary measure of our comprehension of strong qcd is the possession of a veracious understanding of the pion , and for this a clear grasp of the nature of chiral symmetry and its dynamical breaking are vital . in asymptotically free theories , chiral symmetry and its breaking are completely expressed via the axial - vector ward - takahashi identity : @xmath31 wherein the new elements are the current - quark mass matrix : @xmath32 $ ] , and the pseudoscalar vertex @xmath33_{tu } = z_4 \left[\gamma_5 \frac{\tau^j}{2}\right]_{tu } + \int^\lambda_q \left [ \chi_{5}(q;p ) \right]_{sr } \ , k_{tu}^{rs}(q , k;p ) \,.\ ] ] the last two terms of eq.([avwtim ] ) vanish in the chiral limit .
the identity is then a concise statement of chiral symmetry , a consequence of which is plainly an intimate relation between the three - point function in eq.([avbse ] ) and the two - point function in eq.([gendse ] ) . since dynamical chiral symmetry breaking is an essentially nonperturbative phenomenon , satisfying our measure of comprehension requires that there exist a nonperturbative and systematic truncation scheme which preserves identities such as eq.([avwtim ] ) .
at least one such scheme exists @xcite : a dressed - loop expansion of the dressed - quark - gluon vertices that appear in the half - amputated dressed - quark - antiquark scattering kernel : @xmath34 , and that is sufficient to arrive via the dses at exact results in qcd .
for example , it is a general property that the axial - vector vertex , obtained via eq.([avbse ] ) , exhibits a pole whenever @xmath35 , where @xmath36 is the mass of any pseudoscalar @xmath37-@xmath38 meson ; viz .
, @xmath39 with @xmath40 being the @xmath41 bound state s bethe - salpeter amplitude : @xmath42 , \label{genpibsa}\ ] ] and @xmath43 , its leptonic decay constant @xmath44 where the trace is over colour , flavour and spinor indices . equation ( [ fpin ] ) is the precise expression in quantum field theory for the pseudo__vector _ _ projection of the meson s bethe - salpeter wave function onto the origin in configuration space .
the appearance of @xmath45 guarantees that @xmath43 is gauge invariant , and independent of the regularisation scale and renormalisation point . the lowest mass pole contribution to eq.([genavv ] ) , denoted by @xmath46 , is the ground state pion , which receives the spectroscopic assignment @xmath47 in the naive @xmath48 quark model .
the @xmath49 pseudoscalar meson poles correspond to those excited states of the pion that would receive the spectroscopic assignments @xmath50 in the quark model .
the pseudoscalar vertex , @xmath51 , which appears in eq.([avwtim ] ) , also exhibits such a pole : @xmath52 and here the residue @xmath53 expresses the pseudo__scalar _ _ projection of the meson s bethe - salpeter wave function onto the origin in configuration space .
the appearance of @xmath54 guarantees that @xmath55 is gauge invariant , independent of the regularisation scale , and evolves correctly under the renormalisation group . in qcd
it therefore follows from eqs.([genavv ] ) and ( [ genpvv ] ) , as a necessary consequence of chiral symmetry and its dynamical breaking , that for any pseudoscalar @xmath37-@xmath38 meson @xcite @xmath56 \ , \rho_{\pi_n}(\zeta)\,.\ ] ] ( this expression is valid independent of the magnitude of the current - quark mass and , following ref.@xcite , the generalisation to @xmath57-flavour is straightforward . )
the so - called gell - mann oakes renner relation for the ground state pion appears as a corollary of eq.([gmorgen ] ) @xcite , and another important corollary of eq.([gmorgen ] ) , valid for pseudoscalar mesons containing at least one heavy - quark , is described in refs.@xcite .
in addition , the proof of eq.([gmorgen ] ) establishes @xcite that in the chiral limit @xmath58 namely , in the chiral limit the leptonic decay constant vanishes for _ every _ pseudoscalar meson _ except _ the ground state pion .
this result in particular is a constraint on models and theoretical methods used to search for exotics and hybrids .
the leading - order term in the scheme of refs.@xcite is the renormalisation - group - improved rainbow - ladder truncation , which is expressed in the bethe - salpeter equations via @xmath59_{ts } \left[\rule{0mm}{0.7\baselineskip } \frac{\lambda^a}{2}\gamma_\sigma\right]_{r u}\!,\ ] ] with @xmath60 an _ operative _ coupling , and in the truncated gap equation through : @xmath61 this truncation can be used to correlate lattice data on dressed - quark and -gluon schwinger functions @xcite .
it is also a reliable tool in the study of vector and flavour nonsinglet pseudoscalar channels @xcite , and has been refined and exploited by maris and tandy in a series of articles via a one - parameter model for the operative coupling : @xmath62 } \ , { \cal f}(q^2 ) \ , , \label{gk2}\ ] ] wherein @xmath63)]/q^2 $ ] , @xmath64 @xmath65 @xmath66gev ; @xmath67 ; @xmath68 ; and @xmath69gev .
this form expresses the interaction as a sum : the second term ensures that perturbative behaviour is preserved at short - range ; and the first makes provision for enhancement ( strong coupling ) at long - range .
the true parameters in eq.([gk2 ] ) are @xmath70 and @xmath71 , which together determine the integrated infrared strength of the rainbow - ladder kernel .
however , they are not independent : in fitting to a selection of observables , a change in one is compensated by altering the other ; e.g. , on the domain @xmath72\,$]gev , the fitted observables are approximately constant along the trajectory @xmath73 @xcite . hence eq.([gk2
] ) is a one - parameter model .
this correlation : a reduction in @xmath70 compensating an increase in @xmath71 , ensures a fixed value of the interaction s integrated infrared strength .
it also defines a single dressed - glue mass - scale that characterises infrared gluodynamics : @xmath74 using maris and tandy s model for the operative coupling , the rainbow - ladder truncation yields @xcite @xmath75gev ; @xmath76gev ; @xmath77 , at a current - quark mass @xmath78mev .
this model is currently being employed in a rainbow - ladder analysis of @xmath79 pseudoscalar mesons , and it has been reported @xcite that when only the @xmath80 term in eq.([genpibsa ] ) is retained , , @xmath81 that are approximately @xmath82% smaller than experiment . ]
one obtains : @xmath83 cf .
@xmath84 ; and @xmath85 at the physical ( nonzero ) current - quark mass .
the latter result indicates that eq.([fpiniszero ] ) is a good approximation even at real - world current - quark masses and thereby emphasises its usefulness as a constraint .
a correct description of the neutral pion s two - photon decay is another stringent test of one s understanding of strong qcd , and that understanding again relies fundamentally on a true comprehension of chiral symmetry and its dynamical breakdown . to illustrate this we consider the renormalised impulse approximation to the axial - vector vector vector ( avv ) vertex : @xmath86 where @xmath87 and the electric charge matrix @xmath88=e\,{\rm diag}[2/3,-1/3]$ ] . in eq.([tmnr ] ) , the dressed - quark propagator is determined by the rainbow gap equation , and the dressed vector and axial - vector vertices are obtained from the dressed - ladder truncation of the relevant inhomogeneous bethe - salpeter equations .
it is apparent from eq.([genavv ] ) that for @xmath89 ; namely , in the neighbourhood of the pion pole , @xmath90 is dominated by the @xmath91 coupling and therefore plays a role in describing that decay .
the bare avv vertex exhibits a superficial linear divergence and as with all other schwinger functions it must be rigorously defined via a translationally invariant regularisation scheme . in this case
the appropriate pauli - villars prescription corresponds to minimally anticoupling the photon to additional flavoured quarks with a large mass @xmath92 . to elucidate we
introduce @xmath93 where @xmath94 is the renormalisation point invariant current - quark mass and every element in the integrand is calculated using this current - quark mass in the appropriate rainbow - ladder truncated dse .
in this case eq.([tmnr ] ) is explicitly @xmath95 with @xmath96 as the last step in the calculation .
we now focus on the chiral limit , @xmath97 , and consider the divergence of the avv vertex at @xmath98 ; viz .
, in the neighbourhood of the massless pion s pole : @xmath99 equation([ptmnr ] ) follows from the vector vertex analogue of eq.([avbse ] ) and the axial - vector ward - takahashi identity depicted in fig.[figavwti ] , , where @xmath100 is the unamputated renormalised rainbow - ladder - dressed quark - antiquark scattering matrix . ] which is an extension of @xmath101 and can be derived following the method in ref.@xcite : the identity is valid if , and only if , every dressed - quark propagator that appears therein is obtained from the rainbow dse , eq.([rainbowdse ] ) , and the accompanying dressed vertices are determined from the consistent ladder bethe - salpeter equation ; i.e. , the equation constructed using eq.([ladder ] ) and the rainbow quark propagator . it is plain that in the chiral limit eq.([ptmnr ] ) does not contain terms which depend on the light - quark mass .
this axial - vector ward - takahashi identity is an extension of eq.([avwti0 ] ) .
it is valid if , and only if : the dressed - quark propagator , @xmath102 , is obtained from eq.([rainbowdse ] ) ; the axial - vector vertex , @xmath103 , is obtained from eq.([avbse ] ) with the kernel constructed from @xmath102 and eq.([ladder ] ) ; the pseudoscalar vertex is constructed analogously ; and the unamputated renormalised quark - antiquark scattering matrix : @xmath104 $ ] , is constructed from the elements just described.,scaledwidth=80.0% ] consider the first term on the r.h.s . of eq.([ptmnr ] ) .
the complete polarisation integral is convergent , by virtue of our regularisation , and must yield a rank - two lorentz pseudotensor depending on only one four - vector argument , @xmath105 .
a nonzero value is impossible .
the second term vanishes for the same reason .
let s turn to the last term .
the only relevant contribution appears in the limit @xmath106 , in which case all dressing is suppressed and we are left to evaluate @xmath107 recall now that at any order in the nonperturbative , systematic , symmetry - preserving dse truncation scheme of refs.@xcite , dynamical chiral symmetry breaking entails that in the chiral limit the divergence of the axial - vector vertex is dominated by the pion pole in the neighbourhood of @xmath108 .
this result is expressed in eq.([genavv ] ) .
it follows that in the rainbow - ladder truncation , which is the leading - order in that scheme , @xmath109 consequently , in the chiral limit , the bose - symmetrised amplitude for @xmath0 decay @xmath110 in eq.([apigg ] ) we have arrived at the textbook result with no dependence on model details ; viz . , independent of the detailed form of the operative coupling , eq.([alphamt ] ) , and thereby provided an explanation of the numerical analysis reported in ref.@xcite .
in doing so we have made plain that the impulse approximation expressed in eqs.([tmnr ] ) and ( [ pi0gg ] ) is the precise and only truncation of the six - point schwinger functions relevant to this decay which is consistent with the rainbow - ladder truncation .
it follows that the `` seagull terms '' identified in ref.@xcite are absent at this order but may appear in a consistent treatment of the process at an higher order of truncation .
moreover , it is generally true that seagull terms are absent in the consistent rainbow - ladder analysis of any six - point quark schwinger function .
there are many additional applications of interest to this audience , of which a variety are reviewed in refs.@xcite , and one may say that the keystones of success in modern dse usage are : an appreciation and expression of the momentum - dependence of dressed - parton propagators at infrared length - scales ; and the identification and employment of a nonperturbative , systematic , symmetry preserving truncation scheme .
one topic of material importance , worth mentioning explicitly here , is the pion s valence quark distribution .
the dse calculation @xcite agrees with perturbative qcd s prediction @xcite ; i.e. , @xmath111 in the valence region , it must be borne in mind that the pion s bethe - salpeter amplitude is momentum dependent at every regularisation scale .
this is an essential feature of qcd and any calculation which overlooks it is immaterial . ] but this is inconsistent with extant data @xcite and a recent transverse lattice calculation @xcite .
the discrepancy between data and the expectation from perturbative qcd can not be resolved using contemporary simulations of lattice - regularised qcd @xcite .
it raises difficult questions and has prompted proposals for remeasurement of the distribution function @xcite .
we have omitted a discussion of contemporary challenges confronting the application of dses in hadron physics and qcd .
important amongst these are : the development of a veracious picture of the infrared behaviour of the bethe - salpeter kernel ; and the extension of the framework to processes involving baryons .
this list is not exhaustive .
nonetheless , comments on the former may be found in refs.@xcite and the latter in refs.@xcite .
cdr warmly thanks the organisers of this light cone workshop : `` hadrons and beyond , '' at the uk institute for particle physics phenomenology , for their hospitality and support , and particularly simon dalley and linda wilkinson .
this work was also supported by : _
fwf erwin - schrdinger - auslandsstipendium nr .
_ j2233-n08 ; department of energy nuclear physics division contract no .
w-31 - 109-eng-38 ; national science foundation contract no
. int-0129236 ; and the a.v .
humboldt foundation via a f.w
. bessel research award .
tandy , prog . part .
* 50 * ( 2003 ) 305 .
a. krassnigg and c.d .
roberts , `` dses , the pion , and related matters , '' nucl - th/0308039 .
a. krassnigg and c.d .
roberts , `` dyson - schwinger equations : an instrument for hadron physics , '' nucl - th/0309025 .
r. alkofer and c.s .
fischer , `` the kugo - ojima confinement criterion and the infrared behavior of landau gauge qcd , '' hep - ph/0309089 .
pennington , `` in the debris of hadron interactions lies the beauty of qcd , '' hep - ph/0309228 ; `` in the debris of hadron interactions lies the beauty of qcd ( part ii ) , '' hep - ph/0310186 .
p. maris and c.d .
roberts , phys . rev .
c * 56 * ( 1997 ) 3369 .
a. bender , c.d .
roberts and l. von smekal , phys .
b * 380 * ( 1996 ) 7 .
a. bender , w. detmold , c.d .
roberts and a.w .
thomas , phys .
c * 65 * ( 2002 ) 065203 .
p. maris , c.d .
roberts and p.c .
tandy , phys .
b * 420 * ( 1998 ) 267 .
k. langfeld , h. markum , r. pullirsch , c.d .
roberts and s.m .
schmidt , phys .
c * 67 * ( 2003 ) 065206 .
p. maris and c.d .
roberts , `` qcd bound states and their response to extremes of temperature and density , '' in proc . of the _ workshop on nonperturbative methods in quantum field theory _ , edited by a.w .
schreiber , a.g .
williams and a.w .
thomas ( world scientific , singapore , 1998 ) pp . 132 - 151 ; m.a .
ivanov , yu.l .
kalinovsky and c.d .
roberts , phys .
d * 60 * ( 1999 ) 034018 .
bhagwat , m.a .
pichowsky , c.d .
roberts and p.c.tandy , phys .
c * 68 * ( 2003 ) 015203 .
p. maris and p.c .
tandy , phys .
c * 61 * ( 2000 ) 045202 ; _
ibid_. , c * 62 * ( 2000 ) 055204 .
j. volmer _ et al . _ [ jlab @xmath112 collaboration ] , phys .
* 86 * ( 2001 ) 1713 .
p. maris and p.c .
tandy , phys .
c * 60 * ( 1999 ) 055214 .
d. jarecke , p. maris and p.c .
tandy , phys .
c * 67 * ( 2003 ) 035202 .
p. maris and c.d .
roberts , int .
j. mod .
e * 12 * ( 2003 ) 297 .
p. maris , a. raya , c.d .
roberts and s.m .
schmidt , eur .
j. a * 18 * ( 2003 ) 231 .
p. bicudo , phys .
c * 67 * ( 2003 ) 035201 .
p. maris and p.c .
tandy , phys . rev .
c * 65 * ( 2002 ) 045211 .
lepage , s.j .
brodsky , t. huang and p.b .
mackenzie , `` hadronic wave functions in qcd , '' clns-82/522 , _ invited talk , banff summer inst . on particle physics , banff , alberta , canada , aug 16 - 28 , 1981_. c.d .
roberts and s.m .
schmidt , prog.part .
* 45 * ( 2000 ) s1 ; r. alkofer and l.von smekal , phys .
* 353 * ( 2001 ) 281 .
hecht , c.d .
roberts and s.m .
schmidt , phys .
c * 63 * ( 2001 ) 025213 ; c.d .
roberts , nucl .
* 108 * ( 2002 ) 227 .
brodsky , m. burkardt and i. schmidt , nucl .
b * 441 * ( 1995 ) 197 .
conway _ et al .
_ , phys .
d * 39 * ( 1989 ) 92
. s. dalley and b. van de sande , phys .
d * 67 * ( 2003 ) 114507 .
w. detmold , w. melnitchouk and a.w .
thomas , phys .
d * 68 * ( 2003 ) 034025 .
holt and p.e .
reimer , aip conf .
proc . * 588 * , 234 ( 2001 ) ; k. wijesooriya , _
et al_. , `` the @xmath113 reaction and the pion structure function , '' proposal no
. jlab pr01 - 110 .
hawes , p. maris and c.d .
roberts , phys .
b * 440 * ( 1998 ) 353 .
alekseev , few body syst .
* 32 * ( 2003 ) 193 .
bloch , few body syst .
* 33 * ( 2003 ) 111 .
hecht , m. oettel , c.d .
roberts , s.m .
schmidt , p.c .
tandy and a.w .
thomas , phys .
c * 65 * ( 2002 ) 055204 ; j.c.r .
bloch , a. krassnigg and c.d .
roberts , `` regarding proton form factors , '' nucl - th/0306059 , to appear in few body syst . | an hallmark of present - day dyson - schwinger equation applications in hadron physics is the existence of a systematic and symmetry preserving truncation scheme .
this enables the proof of exact results ; e.g. , the leptonic decay constant of every pseudoscalar meson except the pion vanishes in the chiral limit .
calculations using the scheme s leading - order truncation are reliable in the vector and flavour nonsinglet pseudoscalar channels . in this rainbow - ladder truncation
, an impulse approximation provides the consistent current for all six - point quark schwinger functions .
that is well illustrated via the anomalous process @xmath0 .
using two- , three- and four - point schwinger functions calculated in rainbow - ladder truncation , the textbook value of the width is obtained algebraically and independent of model details if , and only if , the impulse approximation is used to describe the associated matrix element . |
the world changes .
this basic fact is expressed in the second law of thermodynamics : _ the entropy of a closed system never decreases _ : @xmath1 where the entropy is denoted by @xmath2 . in the 1870s ,
ludwig boltzmann @xcite@xcite provided a description of the entropy relating the states of macro and micro - systems .
he argued that any system would evolve toward the macrostate that is consistent with the largest possible number of microstates .
the number of microstates and the entropy of the system are related by the fundamental formula : @xmath3 where @xmath4 jk@xmath5 is boltzmann s constant and @xmath6 is the volume of the phase - space that corresponds to the macrostate of entropy @xmath2 .
the physical processes in any thermodynamical system are irreversible .
consequently , any system always tends to its state of maximum entropy . if we consider the maximal system , the universe , the situation is not so clear . in the past , the universe was hotter and at some point matter and radiation were in thermal equilibrium ( i.e. in a state of maximum entropy ) ; then , how can entropy still increase if it was at a maximum at some past time ?
some attemps were made @xcite to answer this question , invoking the expansion of the universe : the universe actually began in a state of maximum entropy , but because of the expansion , it was still possible for the entropy to continue growing .
the main problem with this type of explanation is that the effects of gravity can not be neglected in the early universe @xcite .
roger penrose suggested that entropy might be assigned to the gravitational field itself .
though locally matter and radiation were in thermal equilibrium , the gravitational field should have been quite far from equilibrium , since gravity is an atractive force and the universe was initially structureless @xcite .
consequently , the early universe was globally out of equilibrium , being the total entropy dominated by the entropy of the gravitational field . in absence of a theory of quantum gravity , a statistical measure of the entropy of the gravitational field is not possible .
the study of the gravitational properties of macroscopic systems through classic general invariants , however , might be a suitable approach to the problem .
penrose proposed that the weyl curvature tensor can be used to specify the gravitational entropy .
the weyl tensor is a 4-rank tensor that contains the independent components of the riemann tensor not captured by the ricci tensor .
it can be considered as the traceless part of the riemann tensor .
the weyl tensor can be obtained from the full curvature tensor by substracting out various traces .
the riemann tensor has 20 independent components , 10 of which are given by the ricci tensor and the remaining by the weyl tensor .
the weyl tensor is given by : @xmath7\beta}-g_{\beta[\gamma}r_{\delta]\alpha})+\\ \nonumber & + & \frac{2}{(n-1)(n-2 ) } r~g_{\alpha[\gamma}g_{\delta]\beta},\\\end{aligned}\ ] ] where @xmath8 is the riemann tensor , @xmath9 is the ricci tensor , @xmath10 is the ricci scalar , @xmath11 $ ] refers to the antisymmetric part , and @xmath12 is the number of dimensions of space - time .
the behaviour of the weyl tensor follows what is expected for the gravitational entropy throughout the history of the universe : it is zero in the ( homogeneous ) friedmann - robertson - walker model and it is large in the schwarzschild space - time .
rudjord , gr@xmath0n and sigbj@xmath0rn @xcite made a recent attempt to develop a description of the gravitational entropy based on the construction of a scalar derived from the contraction of the weyl tensor and the riemann tensor .
their phenomenological approach is based on matching their description of the entropy of a black hole with the hawking - bekenstein entropy @xcite . in particular , they calculated the entropy of schwarzschild black holes and the schwarzschild - de - sitter space - time .
the main goal of the present study is to extend the calculation of the gravitational entropy to other space - time geometries and to analyse whether the proposal still works in more complex objects .
in this section we provide a brief review of the work done by rudjord et .
al . they suggested an estimator for the gravitational entropy taking into account the hawking - bekenstein entropy .
the entropy of a black hole can be described by the surface integral : @xmath13 where @xmath14 is the surface of the horizon of the black hole and the vector field @xmath15 is : @xmath16 with @xmath17 a unitary radial vector .
the scalar @xmath18 is defined in terms of the weyl scalar ( w ) and the krestchmann scalar ( r ) .
it takes the form : @xmath19 in order to find an acceptable description of the entropy of a black hole , it is required that equation [ sint ] be equal at the horizon to the hawking - bekenstein entropy : @xmath20 the latter equation allows to calculate the constant @xmath21 : @xmath22 where @xmath23 is boltzmann s constant , @xmath24 is the planck area , @xmath25 @xmath26@xmath27 @xmath28 is the constant of gravitation , and @xmath29 . as usual , @xmath30 is the planck s constant .
in what follows we will set the constant @xmath21 equal to 1 since it plays the role of a scale factor in equation [ sint ] .
this simplifies the notation .
the entropy density can be then determined by means of gauss s divergence theorem , rewriting equation ( [ sint ] ) as a volume integral : @xmath31 here , the absolute value brackets were added to avoid negative or complex values of entropy .
rudjord et .
al . calculated the entropy density in schwarzschild , de sitter , and schwarzschild - de - sitter ( sds ) space - times .
their investigations of the de sitter space - time led them to conclude that the entropy of the cosmological horizon can not be gravitational .
they suggested two different interpretations : either the horizon entropy is in general of different nature from gravitational entropy or , if a cosmological constant is introduced , there is a thermodynamical factor related to it .
the schwarzschild space - time seems to have only gravitational entropy .
then , the schwarzschild and de sitter space - times were interpreted as two extreme cases of the sds space - time .
this result is what is expected from a reasonable description of the gravitational entropy : large thermodynamical entropy in the early universe and large gravitational entropy around black holes . in the next section we present the results of our calculations of the gravitational entropy and density for more complex objects : reissner - nordstrm , kerr , kerr - newman black holes , and a special case of wormhole using rudjord et .
al.s conjecture .
the reissner - nordstrm metric @xcite is a spherically symmetric solution of the einstein field equations @xcite .
it is not a vacuum - solution since the source has electric charge @xmath32 , and hence there is an electromagnetic field .
the energy - momentum tensor of this field is : @xmath33 where @xmath34 is the electromagnetic field strength tensor and @xmath35 is the electromagnetic 4-potential . outside the charged object
the 4-current @xmath36 is zero , so the maxwell equations are : @xmath37 the solution for the metric is given by : @xmath38 where @xmath39 is the schwarzschild radius and : @xmath40 is related to the electric charge @xmath32 .
the study of the metric shows the presence of two different horizons : the outer ( @xmath41 ) and the inner ( @xmath42 ) horizons : @xmath43 if @xmath44 , the outer horizon becomes schwarzschild s and the inner horizon collapses into a central singularity . we shall discuss a formulation of the gravitational entropy in terms of both horizons . in order to find @xmath18
, we calculate the weyl scalar and the kretschmann scalar .
the results are : @xmath45 and @xmath46 therefore , @xmath18 is given by : @xmath47 according to rudjord et .
@xcite , in order to find the gravitational entropy we must operate in a 3-space .
therefore , the spatial metric is defined as : @xmath48 where @xmath49 is the 4-dimensional space - time metric and the latin indices denote spatial components , @xmath50 , @xmath511,2,3
. then , the infinitesimal surface element is given by : @xmath52 from equations [ 3dst ] , and [ infele ] in the reissner- + nordtrm space - time , the space metric and the infinitesimal surface element are : @xmath53,\ ] ] @xmath54 the integration over the two different horizons must be done carefully since there is a singularity at the origin .
the simple method that rudjord et .
al . developed in the case of the schwarzschild black hole was to integrate over a small sphere with radius @xmath55 around the origin , and to substract this from integral [ sint ] .
this yields : @xmath56 the scalar @xmath18 does not depend on @xmath57 and @xmath58 , so we have : @xmath59\int_0^{2\pi } \int_0^\pi \sin \theta d \theta d \phi \notag . \\\end{aligned}\ ] ] when @xmath60 , the entropy of the horizons is obtained : @xmath61 by replacing equation [ rnh ] into equation [ s+ ] we can see that when @xmath44 , the entropy of the outer horizon becomes the schwarzschild entropy @xcite . for the inner horizon
we obtain an indeterminate form : @xmath62 a possible interpretation of the last limit is the following : when the electric charge goes to zero , the inner horizon collapses into the singularity .
in other words , the entropy of the inner horizon goes to the entropy of the singularity .
general relativity is not an epistemologically well - determined theory : the language of the theory says nothing about the singularity because continuum space - time does not exist there .
we now calculate the entropy density using equation [ den ] .
we get : @xmath63 where : @xmath64 and @xmath65.\ ] ] again , if @xmath44 , we obtain the same result as in the schwarzschild case . figure [ fig : rn_entropy_density ] shows a plot of @xmath66 for different values of the charge . , @xmath67 .
, width=340,height=226 ] figure [ radiusrn ] shows the value of the radius of the outer horizon for a given range of the charge .
the radius is only well defined for @xmath68 $ ] . outside this interval
the horizon does not exist and the singularity is naked . according to the cosmic censorship conjecture @xcite , singularities only occur if they are hidden by an event horizon .
if this conjecture is valid , the value of the charge has to be restricted . , @xmath67 .
, width=264,height=188 ] figure [ 2drn ] shows a plot of the entropy density for a fixed value of the charge .
the function @xmath66 is not defined at @xmath69 and it tends asymptotically to zero for large values of the radius .
this is actually what it is expected of a good description of the entropy density for reissner - nordstrm black holes : @xmath66 can not be defined at the origin because of the presence of a singularity , and goes to zero as the field decreases .
furthermore , one of the minima of the function @xmath66 is at @xmath70 where the inner cauchy horizon is located .
this horizon is near two relative maxima and an other relative minimum of the function @xmath66 .
a possible explanation is that in the case of reissner - nordstrm black hole , near the cauchy horizon , there is a small region of space - time where the curvature is high and potentially unstable against external perturbations , as shown by poisson and israel @xcite .
the behaviour of the entropy density for @xmath71 $ ] could then be reflecting the accumulation of gravitational energy and entropy in that region , and the instability of the inner horizon . . here , @xmath72 , @xmath67
. , width=302,height=226 ] a wormhole is a region of space - time with non - trivial topology .
it has two mouths connected by a throat .
the mouths are not hidden by event horizons as in the case of black holes , and , in addition , there is no singularity to avoid the passage of particles from one side to the other .
contrary to black holes , wormholes are holes in space - time , i.e. , their existence implies a multiple - connected space - time @xcite .
there are many types of wormhole - like solutions of the einstein field equations @xcite .
let us consider the static spherically symmetric line element : @xmath73 here , @xmath74 and @xmath75 are two arbitrary functions to be constrained by the required properties of the wormhole .
the function @xmath75 is called `` the shaped function '' because it determines the spatial shape of the wormhole , whereas @xmath76 determines the gravitational redshift and is called the `` redshift function '' .
the radial coordinate @xmath77 has a special significance : @xmath78 is the circumference of a circle centered on the wormhole s throat .
it decreases from @xmath79 to a minimum value @xmath80 , the radius of the throat ; then it increases from @xmath80 back to @xmath79 as distance to the throat increases .
in order to have a wormhole which is transversable in principle , we demand @xcite @xcite : 1
. the function @xmath76 must be finite everywhere ( to be consistent with the absence of event horizons ) .
2 . in order for the spatial geometry to tend to an appropiate asymptotically flat limit
, it must happen that : @xmath81 and , @xmath82 we can impose certain restrictions on the stress - energy that generates the wormhole s curvature .
these `` junction conditions '' allow to determine the specific form of the functions @xmath83 and @xmath76 .
one possible case is to confine the exotic material , that keeps the throat open , to the interior of a sphere of surface radius @xmath84 .
the vacuum region outside @xmath84 constrains the external space - time geometry to have the standard schwarzschild form @xcite : @xmath85 there are some other solutions in which the exotic material is limited to the throat vecinity . morris and
thorne @xcite gave an example of a wormhole whose interior solution in and around the throat has exotic material and it is joined onto non - exotic matter at a radius @xmath86 .
there is a surface layer at @xmath84 of thickness @xmath87 where the tension @xmath88 drops to zero . finally , the external region has the schwarzschild form for @xmath89 .
the complete wormhole solution reads : @xmath90+r_{s}/100 & \textrm{at $ r_{\rm{s}}\leq r \leq r_{{\rm l}}$},\\ b\equiv r_{\rm{s}}/100 & \textrm{at $ r_{\rm{l } } \leq r$ } , \end{array}\right.\ ] ] @xmath91 where @xmath92 , @xmath93 , and @xmath94
. since one of the main features of wormholes is the absence of event horizons , the formulation of the entropy in terms of a surface integral on the event horizon is meaningless . as we shall see , the calculation of the entropy density using the rudjord et .
al.s conjecture , however , is still possible .
we begin the calculus in the region closer to the throat and with exotic matter .
the metric takes the form : @xmath95 where @xmath96 and @xmath97 is the radius at the throat .
the calculation of the weyl and kretschmann scalars gives : @xmath98 and @xmath99 respectively .
the scalar @xmath18 takes the simple form : @xmath100 from equation ( [ wm ] ) the spatial metric is : @xmath101 in terms of @xmath18 and the covariant derivative the entropy density is : @xmath102 } \frac{1}{r}\right|.\ ] ] the plot of @xmath66 as a function of @xmath77 is shown in figure [ fig : wormhole ] .
.,width=302,height=226 ] we observe in figure [ fig : wormhole ] that the entropy density is zero at the throat ( where the exotic matter is confined ) and then increases smoothly to then to tend asymptotically to zero at large distances from the object . for @xmath103
the solution obtained has no physical meaning because the wormhole metric is defined for @xmath104 .
this result is what we expect , interpreting the wormhole as a surgery construction out from two symmetric schwarzschild space - times .
the solution of the field equations for a rotating body of mass @xmath105 and angular momentum per unit mass @xmath106 was found by kerr @xcite . since the boyer - lindquist coordinates present
coordinate singularities in the event horizons , we work in the kerr coordinates where these pathologies have been removed . the boyer - lindquist and the kerr coordinates
are related by the following coordinate transformation : @xmath107 the metric can be written in matrix form as : @xmath108 where : @xmath109}{\rho^2}\sin^2\theta,\end{aligned}\ ] ] and : @xmath110 here @xmath111 represents the angular momentum of the black hole and the inner and outer horizons occur at : @xmath112 figure [ radiuskerr ] shows a plot of the radius of the outer horizon as a function of the angular momentum . again , assuming the cosmic censorship conjecture , the values of the angular momentum are restricted to @xmath113 $ ] .
, @xmath67.,width=264,height=188 ] just as the schwarzschild solution is the unique vacuum solution of einstein s field equations ( as result of israel s theorem ) , the kerr metric is the unique stationary axisymmetric vacuum solution ( carter - robinson theorem ) .
the calculation of the scalars yields : @xmath114 since @xmath115 , the result of the calculation of the entropy in the outer event horizon is equal to the area of this horizon .
the metric in this horizon takes the form : @xmath116}{\rho^2}\sin^2\theta d\chi^2+\rho^2d\theta^2.\end{aligned}\ ] ] the @xmath117 function can be written as @xmath118 , where @xmath42 and @xmath41 are the radii of the inner and outer horizon respectively . from equation
[ sup ] , replacing @xmath119 , we obtain : @xmath120 finally , the calculation of the area gives : @xmath121 @xmath122 @xmath123 in the same way we can make the calculation for @xmath124 : @xmath125 the entropy for both horizons takes the simple form : @xmath126 for @xmath127 , the correct schwarzschild limit is obtained . in the case of axisymmetric space - times ,
it is not possible to calculate the spatial metric ( see equation [ 3dst ] ) because of the presence of the metric coefficient @xmath128 : as the body is rotating the spatial position of each event in the kerr space - time depends on time .
the covariant divergence , then , is obtained from the determinant of the full metric : @xmath129 where @xmath130 is the determinant of the metric and is given by : @xmath131 ^ 2 \sin^2\theta.\ ] ] by introducing equation [ g ] into [ for ] , we calculate the entropy density .
the result is : @xmath132 we can see that the entropy density does not depend explicitly on the mass of the black hole .
a 3-dimensional plot of equation [ densitykerr ] is shown in figure [ densitykerr1 ] for @xmath133 and different values of the angular momentum .
the entropy density is everywhere well defined except for @xmath134 and @xmath69 , where the ring singularity is located . at large distances , @xmath135
, the function @xmath66 goes to zero , as expected . , and @xmath136.,width=340,height=188 ]
the kerr - newman metric of a charged spinning black hole is the most general black hole solution .
it was found by newman et al . in 1965
@xcite . as in the case of the kerr black hole , in order to avoid the coordinate singularities in the event horizons , we make the following change of coordinates : @xmath137 where : @xmath138 the metric takes the matrix form : @xmath139 here , @xmath140 , and : @xmath141}{\rho^2}\sin^2\theta.\end{aligned}\ ] ] the kerr - newman solution depends on three parameters : the mass , the angular momentum and the charge .
we show , in figure [ rayonkn ] , that the radius of the outer horizon is well defined for @xmath142 $ ] and @xmath143 $ ] .
this is clearly a combination of the previous cases . , @xmath67 .
, width=302,height=226 ] the calculation of the determinant @xmath144 gives : @xmath145 which has the same form as in the kerr space - time . since the kerr - newman space - time is a non - vacuum solution of einstein s field equations because of the presence of an electromagnetic field , the weyl and + kretschmann scalars are not equal .
the calculation of the scalar @xmath18 yields : @xmath146 where : @xmath147 @xmath148 and , @xmath149 we compute the entropy density using equation [ for ] .
the function we find , however , is singular for certain values of the radius : @xmath150,\ ] ] since @xmath10 , the kretschmann scalar , is a polynomial of order six with at least one positive real root .
for example , if we solve the equation @xmath151 for @xmath152 , @xmath153 , @xmath154 , and @xmath155 , the roots of the kretschmann scalar are : @xmath156 since the only difference between kerr space - time and kerr - newman space - time is the @xmath18 scalar ( the determinant @xmath130 is the same ) , we conclude that rudjorn et .
al.s proposal for the entropy scalar ( see equation [ p1 ] ) can not be used to calculate the entropy density in the kerr - newman space - time .
hence , we propose this alternative definition for @xmath18 : @xmath157 for a kerr - newman black hole equation [ p2 ] yields : @xmath158 where : @xmath159 @xmath160 we compute the entropy density using equation [ for ] .
the result is : @xmath161 and : @xmath162 in figures [ kn13d ] and [ kn23d ] it is shown the plots of the entropy density of kerr - newman black hole as a function of the radial coordinate and the angular momentum or the charge respectively . in both cases the entropy density diverges for @xmath69 and @xmath163 as expected . ,
@xmath134 , and @xmath164.,width=302,height=188 ] , @xmath134 , and @xmath165.,width=340,height=188 ] figure [ tabla ] shows a table with values of the entropy density at different @xmath166 , in the core of the object and in both horizons .
it can be seen that only in the equatorial plane the entropy density diverges , where the ring singularity is located . for @xmath167 ,
the entropy density is everywhere well defined .
[ cols="^,^,^,^",options="header " , ] in order to check if our definition of the @xmath18 scalar ( equation [ p2 ] ) can be used in other non spherically symmetric space - times , we will test it in kerr black holes .
the calculation of the @xmath18 scalar by equation [ p2 ] yields : @xmath168 where : @xmath169 and the entropy density gives : @xmath170 . here , @xmath171 , @xmath134 , and @xmath172.,width=302,height=188 ] .
here , @xmath171 , @xmath173 , and @xmath165.,width=302,height=188 ] towards the center of the object the entropy density goes to zero , except for @xmath134 and @xmath69 where the ring singularity is located ( see figures [ kerrn1 ] and [ kerrn2 ] ) .
we conclude that this new scalar also gives a reasonable description of the gravitational entropy density in the kerr space - time .
we remark that in axisymmetric space - times , the vector field @xmath174 could have an angular component : @xmath175 where @xmath176 is a unitary vector . according to equation [ den ] ,
the entropy density then takes the form : @xmath177\right|.\ ] ] the second term of this equation is singular for @xmath178 and @xmath179 , that is , at the poles of both kerr and kerr - newman black holes .
the vector field @xmath174 is not regular over the whole horizon and consequently can not be used as a reliable tool to estimate the gravitation entropy .
this a direct consequence of the coupling of space and time components in rotating systems .
the entropy density for the spherically symmetric compact objects , reissner - nordstrm black holes and wormholes , shows the expected features for a good measure of the gravitational entropy density in terms of rudjord et .
al.s proposal : entropy density is everywhere well - defined , except in reissner - nordstrm black holes where the singularity is located , and goes to zero as the field decreases . in the case of a simple wormhole space - time , @xmath180 at the throat because of the presence of exotic matter . for axisymmetric space - times ,
however , rudjord et .
al s formulation has to be modified because the entropy density presents divergences . in this work
we conclude that in the case of rotating bodies , the entropy density has to be calculated from the full metric form .
furthermore , the @xmath18 scalar is redefined as the weyl scalar .
these changes give a reasonable description of the gravitational entropy density : the function @xmath66 is well - defined everywhere except where space - time is singular and it tends asymptotically to zero outside the black hole .
+ we remark , however , that classical estimators as discussed here , should be tested in cosmological scenarios in order to determine their validity in more general space - times
. a complete characterization of the gravitational entropy will only be possible with a quantum theory of gravitation , which should be a singularity - free theory . in the meanwhile
, classical estimators can be helpful for some applications .
the general validity of the second law of thermodynamics in black hole interiors remains an open issue , that will be addressed in a forthcoming work .
boltzmann , l. : @xmath183ber die beziehung zwischen dem zweiten hauptsatz der mechanischen w@xmath182rmetheorie und der wahrscheinlichkeitsrechnung respektive den s@xmath182tzen @xmath181ber das w@xmath182rmegleichgewicht . _ wiener berichte _ * 76 * , 373 - 435 ( 1877 ) | pure thermodynamical considerations to describe the entropic evolution of the universe seem to violate the second law of thermodynamics .
this suggests that the gravitational field itself has entropy . in this paper
we expand recent work done by rudjord , gr@xmath0n and sigbj@xmath0rn where they suggested a method to calculate the gravitational entropy in black holes based on the so - called ` weyl curvature conjecture ' .
we study the formulation of an estimator for the gravitational entropy of reissner - nordstrm , kerr , kerr - newman black holes , and a simple case of wormhole .
we calculate in each case the entropy for both horizons and the interior entropy density .
then , we analyse whether the functions obtained have the expected behaviour for an appropriate description of the gravitational entropy density .
example.eps gsave newpath 20 20 moveto 20 220 lineto 220 220 lineto 220 20 lineto closepath 2 setlinewidth gsave .4 setgray fill grestore stroke grestore |
the exchange interaction is of increasing importance in the study of controllable quantum mechanics using solid - state systems .
as well as being fundamentally important in manybody physics , it is this interaction which is often used to mediate spin flips or entanglement in spintronics and quantum information processing ( qip ) @xcite . for these reasons
there has been considerable study recently on the origin and control of the exchange interaction @xcite . for applications involving the time varying control of the exchange interaction , such as qip , the stability in time of this interaction is of crucial importance .
as the origin of the exchange interaction is essentially the overlap of electron wave functions , the interaction strength is sensitive to the local charge environment .
recent work @xcite has shown that the dependence of the exchange interaction is approximately linearly dependent on fluctuations in the local electric field . in this paper
we develop a general framework with which the effect of these fluctuations can be analytically included in time dependent calculations of the state of a spin system . using the superoperator formalism @xcite
, we derive an expression for a pair of spins as a function of time , depending on the exchange fluctuation amplitude and rate . the extension to multiple fluctuators , in particular , spectral distributions of fluctuators
is also considered .
the formalism can be used to investigate the effect of exchange fluctuations on spintronics , quantum control schemes and specifically quantum error correction ( qec ) .
we begin by studying the exchange coupling hamiltonian in the presence of a single charge fluctuator with the aim of understanding the decohering effects of the fluctuator .
the hamiltonian for the process is @xmath0 the exchange coupling @xmath1 varies in time due to a random telegraph noise ( rtn ) process , @xmath2 and we assume a net effect of the form @xmath3 where @xmath2 describes the fluctuator .
this rtn process couples with strength @xmath4 ( ultimately dependent on the distance between the coupled spins and the fluctuator ) to the bare exchange term , @xmath5 .
the time evolution of the system can then be described by the density matrix master equation @xmath6 , \label{eqn : density_matrix_evolution}\ ] ] where @xmath7 is the density matrix of the system .
additional terms can be added to this master equation to also model non - unitary evolution , such as decohering processes . as a matter of convenience we may re - express the system evolution in superoperator form . in superoperator form the density matrix
is given a vector representation , denoted by @xmath8 , by transforming the matrix into a single column , one row at a time @xcite .
a superoperator @xmath9 contains all the evolution of the system ( both unitary and non - unitary ) @xmath10 for purely hamiltonian evolution the superoperator @xmath9 can be written down in terms of @xmath11 and the identity operator @xmath12 .
\label{eqn : form_of_superoperator_p}\ ] ] the superoperator simplifies to @xmath13 for the hamiltonian we consider .
here , @xmath14 is the heisenberg interaction in superoperator form .
if the hamiltonian is time independent , then the superoperator * p * is also time independent and the density matrix at some time @xmath15 is @xmath16 given an initial state @xmath17 .
we show how this time independent formalism is relevant to our problem shortly .
the rtn process , @xmath2 , is modelled as in ref .
the noise fluctuates randomly between -1 and 1 with the frequency of the fluctuations controlled by the correlation time @xmath18 . here
, @xmath19 is the typical frequency of jump times , where the jump time instants are , @xmath20 and the @xmath21 are random numbers such that @xmath22 . the noise process @xmath2 is described as @xmath23 where @xmath24 is the heaviside step function , and @xmath2 can fluctuate between @xmath25 .
we choose @xmath26 and control the coupling strength via @xmath4 as in eq .
[ eqn : exchange_coupling ] .
the density matrix evolution for our system can be found by numerically averaging over many such noise histories @xmath2 to obtain the correct system dynamics .
for an initial state @xmath27 , @xmath28 where the @xmath29 are the evolution operators for trajectories @xmath30 .
since the hamiltonian ( eq . [ eqn : exchange_hamiltonian ] ) commutes with itself at all times , @xmath31=0 , \label{eqn : commutation_relation}\ ] ] the @xmath29 may be expressed as @xmath32 where @xmath33 and @xmath34 describe the total time the fluctuator exists in the -1 and + 1 states respectively for a particular noise history , and @xmath35 using the result of eq .
[ eqn : rho_at_time_t ] for the superoperator form of a density matrix governed by a time independent hamiltonian we re - express eq .
[ eqn : rho_in_terms_of_u ] such that @xmath36 the ensemble averaged superoperator , @xmath37 , is the average of all the individual trajectory superoperators @xmath38 , @xmath39 this implies that @xmath37 may be constructed by numerically averaging over many noise histories .
the averaging is crucial in obtaining the correct system dynamics , as the rtn is a stochastic process and so there are many unique noise trajectories .
averaging over these noise trajectories results in non - unitary evolution despite the hamiltonian being strictly unitary .
conversely , it is possible to derive @xmath37 analytically by describing the stochastic rtn using an appropriate probability density function ( pdf ) . by considering all unique @xmath38 as a function of the average fluctuator state @xmath40 , weighted by a pdf giving the occurrence likelihood of the average fluctuator state , and integrating this over all possible @xmath41 ,
the resulting expression for @xmath37 is @xmath42 here , @xmath43 is the unique individual superoperator corresponding to a particular value of @xmath41 and @xmath44 is the pdf , which determines the probability that during the time interval @xmath45 , the average fluctuator state is @xmath41 . in section [
sect : pdf ] we show how to specify the pdf , so that we can use it to analytically determine @xmath37 in section [ sect : determining_q ] .
the statistical properties of an rtn process have been studied extensively in the context of reliability theory , alternating renewal processes and queueing theory @xcite . in our case
, we are specifically interested in the probability of the rtn spending a certain fraction of the observation period in a particular state .
the pdf for an rtn signal fluctuating between the states 0 and + 1 is given by @xcite as @xmath46 , \label{egn : pdf_derived from zacks_paper}\ ] ] where @xmath47 is the modified bessel function of the first kind .
this pdf assumes the initial state is + 1 , and that _ at least _ a single fluctuation occurs . here , @xmath48 is used to describe the time spent in the state 0 and @xmath45 is the duration of the process we are considering . the parameter @xmath19 characterises the fluctuator rate as before . properly normalised the pdf is , @xmath49}{\sinh^2\left(\frac{\lambda t}{2}\right)}. \label{egn : pdf_normalised}\ ] ] we could equally describe a process which begins in the state 0 , with @xmath50 describing the time spent in this state . assuming at least a single fluctuation occurs , the full pdf is obtained by averaging over both possible starting states @xmath51 .
\label{eqn : pdf_averaging}\ ] ] we may re - express this pdf in terms of the mean fluctuator state @xmath41 , where @xmath52 $ ] . taking care to preserve the normalisation , the pdf for an rtn process of duration @xmath45 assuming at least one fluctuation occurs is @xmath53,\\ & = & \frac{\lambda t}{4}\frac{i_1\left(\lambda t\sqrt{1-\xi^2}\right)}{\sqrt{1-\xi^2}\sinh^2\left(\frac{\lambda t}{2}\right)}. \label{eqn : pdf_general}\end{aligned}\ ] ] the case where no fluctuations occur
must be treated separately . in this case
we expect @xmath41 to be either of @xmath54 .
the properly normalised pdf for this case can be described using two delta functions , @xmath55 .
\label{eqn : pdf_no_fluctuations}\ ] ] the full , general pdf is constructed by appropriately weighting @xmath56 and @xmath57 , with the fluctuation probability given by the poisson distribution @xmath58 where @xmath59 denotes the number of fluctuations , such that @xmath60 and @xmath61 .
after simplification , the resulting pdf is @xmath62\\ + \frac{\lambda t}{e^{\lambda t}-1}\frac{i_1\left(\lambda t\sqrt{1-\xi^2}\right)}{\sqrt{1-\xi^2}}. \label{eqn : full_pdf_general}\end{gathered}\ ] ] in what follows we examine the three limiting cases of the pdf and use these to construct an approximate pdf .
the approximate pdf provides greater physical insight when working within these limits .
examining the two limiting cases of the pdf @xmath57 , the fast and slow fluctuator limits and combining them with @xmath56 , leads to a simplified expression which approximates @xmath44 .
we begin by considering the slow fluctuator limit @xmath63 for the distribution describing at least one fluctuation , @xmath57 .
this is the regime where no more than one fluctuation occurs . in this limit @xmath64 and @xmath65
this reduces the pdf to @xmath66 this uniform distribution implies that a fluctuation is just as likely to occur at any time during the system evolution .
the limit @xmath67 represents a fast fluctuator . in this regime @xmath68 and @xmath69 which reduces the pdf to @xmath70 in this limit @xmath41
will be small , therefore making the substitution @xmath71 , we find the pdf to be gaussian about the origin , @xmath72 which we expect intuitively .
we note that this approach is similar to that used by happer and tam when considering the gaussian limit of rapid spin exchange in alkali vapors @xcite .
weighting these two limiting cases and the pdf describing no fluctuations using the poisson distribution as before , allows us to construct an approximate pdf @xmath73 where @xmath74 .
this approximate @xmath44 provides nice analytic solutions for @xmath37 in each of the three interesting fluctuator regimes .
while this is only an approximation to the exact solution ( eq . [ eqn : full_pdf_general ] ) , it can provide more physical insight , as will become apparent later .
the superoperator @xmath37 can be derived analytically via eq . [
eqn : integral_over_average_fluctuator_state ] using the pdfs determined in the previous section . of particular interest is the non - unitary part of the superoperator .
the non - unitary superoperator can be found by expanding the superoperator into a unitary and non - unitary part , such that @xmath75 .
the evolution in the absence of a fluctuator is contained within the unitary part , @xmath76 whilst the effect of the charge fluctuator is contained within the non - unitary part @xmath77 .
note that these two parts can be factored out due to the commutation relation ( eq .
[ eqn : commutation_relation ] ) .
we now determine the non - unitary parts of the superoperator for @xmath44 and its various approximations . beginning with the case where no fluctuations occur and the pdf is given by @xmath56 , as in eq .
[ eqn : pdf_no_fluctuations ] , we find @xmath78 where @xmath14 is the heisenberg superoperator introduced earlier . when there is at least one fluctuation ( see eq .
[ eqn : pdf_general ] ) the resulting form of the superoperator is @xmath79-\cos\left(\alpha\mbox{\boldmath$\sigma$\unboldmath}_{\rm h}t\right)}{2\sinh^2\left(\frac{\lambda t}{2}\right)}. \label{eqn : qnu_general}\ ] ] examining the limiting cases of the general pdf we find that for the slow fluctuator ( see eq .
[ eqn : pdf_lambda_0 ] ) @xmath80 and in the fast fluctuator limit ( see eq .
[ eqn : pdf_lambda_infinity_gaussian ] ) @xmath81 it should be noted that this final superoperator corresponds exactly to that which would be obtained using the lindbladian formalism @xcite if a decoherence operator of the form @xmath82 was included .
from this one can deduce that the fast fluctuator limit is equivalent to purely markovian decoherence due to interaction with the environment via an exchange like two - qubit decoherence channel .
this is in contrast to conventional dephasing which is modelled using two independent @xmath83 channels , one for each qubit .
this distinction is particularly important as it implies that exchange fluctuations due to environmental charge fluctuations introduce _ correlated _ errors which can have important implications for fault - tolerant qec @xcite . using the previous results
, we can determine @xmath77 for the full weighted pdf s in both the approximate and exact cases .
the exact pdf given in eq .
[ eqn : full_pdf_general ] yields @xmath84-\cos\left(\alpha\mbox{\boldmath$\sigma$\unboldmath}_{\rm h}t\right)\right\ } , \label{eqn : qnu_full_exact}\end{gathered}\ ] ] whilst for the approximate pdf given in eq .
[ eqn : approx_full_pdf_general_construction ] we find @xmath85 in general it is difficult to graphically compare these analytic forms of the superoperator to the numerical result . however , it is possible in this case , as the superoperator @xmath77 is a sparse matrix with the same underlying structure of the @xmath14 superoperator which defines it .
it follows from the definition of @xmath14 ( see eq . [ eqn : form_of_our_superoperator_p ] ) that the only non - zero matrix elements of the heisenberg superoperator are @xmath86 .
consequently , a comparison of the resulting non - zero matrix element of @xmath77 , denoted @xmath87 , proves effective in determining the agreement between the analytic ( exact and approximate ) superoperators and exact numerical solution for the superoperator . ) and approximate ( eq . [ eqn : qnu_approximate ] ) ] and exact numerical ( simulated from eq .
[ eqn : big_q ] ) solutions of @xmath77 .
plotted is the non - zero matrix element of @xmath77 , denoted @xmath87 , as a function of time for a range of fluctuator rates , which span each of the three limiting regimes .
the results show very good agreement between all three solutions , with the analytic approximation deviating only slightly when the fluctuator rate is on the time scale of @xmath45 . ]
the results , as a function of time for a range of fluctuator rates , shown in fig .
[ fig : qnu_analytic_vs_numeric ] , reveal very good agreement between the exact analytic and numerical results for all rates @xmath19 .
the approximate solution also matches closely , particularly in the slow and fast fluctuator limits .
slight deviations from the exact solution can be seen when the fluctuations occur on the time scale of the process we are considering @xmath88 . in this regime
the contribution from the uniform distribution @xmath89 is at its maximum and approximately on par with contributions from the other two distributions .
the deviation from the exact results do not come as a surprise , as the approximate pdf is constructed from contributions due to 0 , 1 or many fluctuations .
adding contributions from 2 , 3 and more fluctuations would reduce this discrepancy .
we now present the generalisation of the single fluctuator formalism to multiple fluctuators in the following section .
extending this formalism to multiple fluctuators is straight forward and provides a method for the treatment of many physically realistic scenarios .
the total ensemble averaged superoperator , @xmath90 , for @xmath91 fluctuators is just the product of all the individual ensemble averaged superoperators , @xmath37 , such that @xmath92 this result is useful for a finite number of fluctuators each with known strength and rate .
however , in most instances only the spectral distribution in strength and rate will be known and therefore eq . [ eqn : multiple_fluctuators ] offers no further insight . by considering all possible unique superoperators
@xmath93 weighted by their probability of occurrence @xmath94 ( where @xmath95 $ ] ) , in a similar way to the method used to construct @xmath37 in eq .
[ eqn : integral_over_average_fluctuator_state ] , we may re - express eq .
[ eqn : multiple_fluctuators ] as @xmath96^{np_i } , \label{eqn : multiple_fluctuators_weighted}\ ] ] where in general there are @xmath97 possible fluctuator types and @xmath91 fluctuators .
we would like to interpret the @xmath94 as a spectral distribution function in @xmath98 and @xmath99 . by expressing @xmath100 as a sum of logarithms @xmath101\right\ } , \label{eqn : multiple_fluctuators_sum_log}\ ] ] and extending the definition of @xmath100 to the continuum
, we may replace the @xmath94 with a spectral distribution function @xmath102 such that @xmath103\,d\alpha\,d\lambda\right\ } , \label{eqn : multiple_fluctuators_integral_log}\end{gathered}\ ] ] ensuring that the spectral distribution function is properly normalised @xmath104 the effects of a region of charge noise can now be modelled using either approach ( eq .
[ eqn : multiple_fluctuators ] or eq .
[ eqn : multiple_fluctuators_integral_log ] ) .
the choice will depend on exactly what information is known about the system . in the following section we outline how to implement the superoperators and provide a discussion of some important limitations of this approach .
the superoperators in sections [ sect : determining_q ] and [ sect : multiple_fluctuators ] were constructed on the basis of the commutation relation for our heisenberg hamiltonian ( see eq .
[ eqn : commutation_relation ] ) .
the commutation relation meant we could express the total evolution operator as a product of two evolution operators , each describing the total time spent in one of the fluctuator states ( see eq .
[ eqn : evolution operator_for given_noise_history ] ) .
the superoperators themselves also commute as a result . without further approximation the superoperators
can be used to model individual processes satisfied by the hamiltonian in eq .
[ eqn : commutation_relation ] ( or a similar commuting hamiltonian , such as the ising interaction ) . in more complex superoperator applications , for example when
multiple applications of the superoperators are separated by non - commuting operations , it may be necessary to make a further approximation . when modelling these more complex processes ( see fig .
[ fig : non_commuting_noisy_process ] for an example where two superoperators are separated by a non - commuting gate operation , @xmath105 ) a problem arises with the formalism when considering the slow fluctuator limit
. separated by a non - commuting gate operation .
attempting to use the superoperators to determine the effects of charge noise in a process like this can lead to the introduction of errors in the slow fluctuator regime . ] specifically , the superoperator terms which should describe no fluctuations at all , actually account for the possibility of a fluctuation occurring between superoperator applications .
we refer to these terms as _
cross terms_. as the fluctuation rate increases the poissonian weighting of these cross terms in the overall superoperator reduces , hence reducing the cross terms significance .
we now consider a simple example which illustrates how these cross terms manifest themselves , before showing how an approximate solution can be constructed for the slow fluctuator regime by removing the cross terms . when the superoperators do not commute as in the example shown in fig .
[ fig : non_commuting_noisy_process ] , two or more applications of the superoperators leads to the introduction of unphysical cross terms in the slow fluctuator limit .
this becomes apparent when we consider the action of the superoperator describing no fluctuations @xmath106 , which may be expanded in terms of the superoperators , @xmath107 , each describing one of the two fluctuator states @xmath108 , in the no fluctuator limit @xmath109 .
\label{eqn : no_fluctaution_components}\ ] ] the cross terms from the product of two ( or more ) of these superoperators which sandwich non - commuting operations results in the description of a single ( or multiple ) fluctuation(s ) .
for example , consider the system in fig .
[ fig : non_commuting_noisy_process ] .
the gate operation @xmath105 does not commute with the superoperators @xmath110 where @xmath111 is the superoperator representation of the gate operation @xmath105 . expanding out each of the fluctuator superoperators using @xmath112 , \label{eqn : q_breakdown}\end{gathered}\ ] ] and eq .
[ eqn : no_fluctaution_components ] , with some rearranging we find @xmath113}{4}\mathbf{q}^{\left(u\right)}\left(t_2\right)\big[\mathbf{q}^+_0\left(t_2\right)\mathbf{g}\mathbf{q}^+_0\left(t_1\right)\\ + \mathbf{q}^+_0\left(t_2\right)\mathbf{g}\mathbf{q}^-_0\left(t_1\right)+\mathbf{q}^-_0\left(t_2\right)\mathbf{g}\mathbf{q}^+_0\left(t_1\right)\\ + \mathbf{q}^-_0\left(t_2\right)\mathbf{g}\mathbf{q}^-_0\left(t_1\right)\big]\mathbf{q}^{\left(u\right)}\left(t_1\right)+\ldots~ , \label{eqn : no_fluctuations_cross_terms}\end{gathered}\ ] ] where we have only shown the terms which should describe no fluctuations .
this entire expression should represent the total superoperator describing no fluctuations .
however , careful inspection shows the presence of two cross terms , which actually imply the occurrence of a fluctuation during the non - commuting gate operation .
cross terms of this form are actually a manifestation of this superoperator formalism and should be removed without also removing any unitary evolution .
it should be emphasised that this problem only occurs in the slow fluctuator limit , where there is a significant probability of there being no fluctuations during a two - qubit operation . as the fluctuation rate increases , the probability of a fluctuation occurring during the single qubit gate increases , which means that each application of the two - qubit gate becomes statistically independent . in this limit , the formalism as presented so far is exact and does not require any attention to cross terms .
we now wish to remove cross terms describing processes which should not occur , such as those in eq .
[ eqn : no_fluctuations_cross_terms ] .
when two or more applications of the @xmath106 superoperator occur in succession , the introduction of these unphysical terms also occurs .
it is possible to construct an approximate solution by carefully removing these cross terms .
cross terms in the large @xmath19 limit do not pose a problem as the @xmath114 superoperator provides the dominant contribution to @xmath115 in this limit .
it will therefore be most important to remove the cross terms due solely to the @xmath106 superoperator - the zeroth order cross terms .
in general , when many applications of the superoperators are required , there will be higher order cross terms .
for example , the first order cross terms would contain the @xmath106 superoperator and a single instance of the @xmath114 superoperator .
successive improvements to the approximate solution are achieved by removing these higher order cross terms . as more orders of these terms
are removed the approximation improves for larger @xmath19 , with the actual region of improvement dependent on the poissonian weighting of the cross terms being removed .
as an example , the zeroth order cross terms , denoted @xmath116 , are removed by firstly re - weighting @xmath115 using the poisson distribution over the total process duration @xmath117 ( i.e. , including non - commuting operations ) , @xmath118 the cross terms can now be removed @xmath119 taking care to not also remove any unitary evolution .
the cross terms may also be removed from the second term , however the improvement from doing this is minimal due to the small contribution from the zeroth order terms at large @xmath19 .
the exchange interaction is of fundamental importance for controllable quantum mechanics in solid - state systems .
its application to mediate spin flips or entanglement has particular importance in spintronics and qip , hence the stability of the exchange interaction is crucial for precise time varying control . in solid - state spin systems
this stability can be affected by the local charge environment , in particular charge fluctuators , due to the exchange couplings dependence on the electron wave function overlap .
we have developed a model to describe the effect of charge fluctuators on the exchange interaction as a function of time , using superoperators dependent on the noise amplitude and rate .
these superoperators can be included in time - dependent calculations of the state of the spin system to model the effect of the charge noise .
furthermore this analysis holds for other spin couplings , like the ising interaction , where a commutation relation analogous to eq .
[ eqn : commutation_relation ] exists . in the fast fluctuator limit
we demonstrated how interaction with the environment via an exchange like decoherence channel leads to purely markovian decoherence , although the decoherence operator leads to correlated noise across the two spins .
the generalisation to multiple fluctuators means that the effect of charge fluctuators distributed according to a spectral distribution function can also be modelled . in the simpler case where only a small number of well defined fluctuators exist ,
the total superoperator is just the product of the individual fluctuator superoperators .
as our model is completely analytic , the effects of exchange fluctuations can in most instances be included trivially in more sophisticated analyses , without the need to explicitly sum over noise histories .
this is important for analysing the operation of spintronic devices as well as qec and fault - tolerance for qip .
this work was supported by the australian research council , the australian government and the us national security agency ( nsa ) and the army research office ( aro ) under contract number w911nf-08 - 1 - 0527 .
jhc acknowledges the support of the alexander von humboldt foundation and lclh is the recipient of an australian research council australian professorial fellowship ( dp0770715 ) | we describe how the effect of charge noise on a pair of spins coupled via the exchange interaction can be calculated by modelling charge fluctuations as a random telegraph noise process using probability density functions .
we develop analytic expressions for the time dependent superoperator of a pair of spins as a function of fluctuation amplitude and rate .
we show that the theory can be extended to include multiple fluctuators , in particular , spectral distributions of fluctuators .
these superoperators can be included in time dependent analyses of the state of spin systems designed for spintronics or quantum information processing to determine the decohering effects of exchange fluctuations . |
how is the critical behaviour affected by the introduction of disorder ( usualy dilution or bond - randomness ) into a model ?
the harris criterion @xcite states that @xmath5 , the scaling index of the operator corresponding to randomness at the pure system fixed point ( also called the crossover exponent ) is equal to @xmath6 ( @xmath7 and @xmath8 are the specific - heat and correlation length exponents of the pure model ) .
thus the critical behaviour of the pure system is unaltered by disorder if @xmath9 , and altered when @xmath10 .
if @xmath11 the situation is marginal .
renormalization - group methods ( namely expansion in the parameter @xmath12 , where @xmath13 is the dimensionality ) , applied to n - component continuous spin models with weak quenched bond disorder@xcite , confirmed the harris criterion @xmath14 .
moreover , it was found that for @xmath15 a new stable ` random ' fixed point with new exponents appears .
for the case @xmath16 ( ising model ) the stability of an @xmath17 random fixed point was shown by jayaprakash and katz@xcite . for the @xmath18 ising model @xmath11 , and the operator corresponding to randomness is expected to be marginally relevant . two partially conflicting theories by dotsenko and dotsenko@xcite ( dd ) and by shalaev@xcite , shankar@xcite , and ludwig@xcite ( ssl )
were suggested for the @xmath18 ising model with small randomness .
while both theories agreed on the divergence of the specific heat , e.g. @xmath19 conflicting predictions were made for other quantities like the magnetization @xmath20 and susceptibility @xmath21 .
extensive monte - carlo ( mc ) simulations of dilute and random - bond ising models ( see a review in ref .
and references therein ) helped to confirm the prediction on the specific heat .
the other quantities were seen to behave ( as predicted by ssl ) , in essence , as in the pure model .
the disordered @xmath22 ising model has been studied by means of mc simulations as well@xcite .
less attention has been paid in recent times to other disordered systems such as random q - state potts models .
kinzel and domany @xcite used a real - space migdal - kadanoff renormalization transformation for the random - bond potts model on the square lattice in two dimensions and found ( when @xmath23 ) a stable random fixed point with a reduced ( with respect to the pure model ) but positive @xmath24 . using methods of the same spirit , andelman and
berker @xcite also found a random fixed point .
for @xmath25 their calculation suggests @xmath26 and @xmath27 .
other workers@xcite too have suggested a negative @xmath28 for @xmath29 potts models in two dimensions .
mc work on random q - state potts models has not been carried out , except for recent work on the bond - random two dimensional 8-state potts model@xcite , where randomness was shown to change the first order phase transition of the pure model into a second order one@xcite .
novotny and landau ( nl)@xcite used mc simulations to study the two dimensional site - dilute baxter - wu@xcite model which is of the same universality class as the 4-state potts model .
nl measured the exponents @xmath30 , @xmath31 , and @xmath32 which did not seem to depend on the amount of dilution .
they could not determine whether the new stable fixed point is a @xmath18 random ising fixed point ( which is very similar to the pure ising fixed point ) or a new impure baxter - wu fixed point . in this paper
we address the effect of randomness on a critical line which connects the critical points of the @xmath18 ising and 4-state potts models . such is the critical line of the two dimensional ashkin - teller model@xcite on a square lattice .
a convenient representation of the pure ashkin - teller ( at ) model is in terms of two ising spin variables , @xmath33 and @xmath34 , placed on every site of a lattice .
denoting by @xmath35 a pair of nearest neighbor sites , the hamiltonian is given by @xmath36 \label{eq : at1}\ ] ] here @xmath37 is the strength of the interactions between neighboring @xmath38 and @xmath39 spins , and @xmath40 is a four - spin coupling ( throughout this article we absorb a factor of @xmath41 into the coupling constants ) .
the phase diagram of the ferromagnetic at model is known in two dimensions from duality transformations and renormalization group studies@xcite .
the three dimensional model has been studied as well@xcite .
the phase diagram which we reviewed previously@xcite includes a line of critical points which connects the ising critical point at one end to the 4-state potts critical point at the other end . along this line
the exponents vary continuously , and have been determined analytically @xcite , interpolating between their ising and four - state potts values . for instance , the crossover exponent connected with randomness @xmath14 changes smoothly from 0 at the decoupled ising point , to 1 at the 4-state potts point .
therefore , according to the harris criterion , we expect randomness to be a relevant operator of varying strength in this regime of the at model , and expect the critical behaviour of the disordered model to differ from that of the pure model .
the critical behaviour of the disordered at model has not been studied before , apart from a conjecture by alcaraz and tsallis @xcite as to the location of the critical manifold of the bond dilute at model .
none the less the effect of disorder on the critical line of the baxter ( or symmetric 8-vertex ) model@xcite , which is isomorphic to the critical line of the at model@xcite , was considered .
dd@xcite have extended their study of the disordered ising model to the baxter model , and found , for small disorder , the specific heat divergence to change from its pure form to the form of eq .
( [ eq : c(t)ising ran ] ) .
matthews - morgan , landau and swendsen ( mls)@xcite have studied the site - dilute baxter model by means of a mcrg ( mc renormalization - group ) method .
mls found @xmath42 and @xmath43 slightly below the ising value .
both studies indicate that continuous variation of critical exponents of the pure model is substituted by flow to a random ising fixed point . in a previous article@xcite we described an efficient mc cluster algorithm@xcite for the at model . in this work
we have used this algorithm to perform extensive mc simulations@xcite of the random - bond at model .
the importance of using a cluster algorithm for the acquirement of reasonably accurate data can not be overlooked .
this is especially true in the highly random regime where local algorithms are extremely slow@xcite while cluster algorithms are most [email protected] partial elimination of critical slowing down allows us to go to large lattice sizes ( up to l=256 ) and obtain accurate results on the finite size scaling of the thermodynamic quantities at criticality .
our results seem to favor an ising like critical behaviour . this work is organized as follows . in sec .
[ sec : rbat ] we define the random - bond at model and a related anisotropic at model .
as will be explained , the anisotropic at model is used as a non - disordered reference model for the random - bond model .
we use the duality transformation of the at model to locate exactly critical surfaces of the two models , which are related to the line of critical points of the pure model . in sec .
[ sec : details ] we describe the mc cluster algorithm , give definitions of the measured quantities and describe our choice of simulation points.we have performed simulations of the anisotropic and random - bond at models at several points on their critical surface . in sec . [
sec : results ] we give our results which consist of the finite size dependence of thermodynamic quantities of the two models at criticality .
analysis of our results shows that the anisotropic model exhibits varying critical exponents , i.e. the specific heat diverges as @xmath44 where at different points of the critical surface of the anisotropic at model we measured different values of @xmath28 . on the other hand
, the random - bond model seems to exhibit a single type of behaviour , i.e. the specific heat diverges as @xmath45 this is identical with a pure ising behaviour and not with a random ising one . in sec .
[ sum ] we summarize our results .
the model we wish to study is the _ random - bond _ at model ( rbat ) on a square lattice , which is defined by the hamiltonian @xmath46 \;.
\label{eq : rbat}\ ] ] the coupling constants @xmath47 and @xmath48 are chosen according to @xmath49 it seems quite natural ( the _ need _ for this comparison will be made clear at the end of this section ) to compare the rbat with the _ anisotropic _ at model ( aat ) with the same hamiltonian ( [ eq : rbat ] ) but with the couplings distributed as follows : @xmath50 spatial anisotropy is usualy known to be an irrelevant operator , that is , the critical behaviour of a model is unaltered by introduction of anisotropy .
thus one expects that continuous variation of critical exponents will continue to exist on some critical manifold of the aat , as is the case for the isotropic ( or pure ) model .
a duality transformation can be used to locate critical manifolds of the aat and rbat in a way we now present .
consider an at system on a square lattice with spatially varying coupling constants ( the spatial variation can be due to anisotropy , or it can be due to randomness as we consider later ) .
the dual lattice @xmath51 of the original square lattice @xmath52 is the square lattice whose sites are at the centers of the plaquettes of the original lattice . in figure
[ fig : duality exp ] sites of @xmath52 are denoted by dots , while sites of @xmath51 are denoted by crosses .
consider a bond of strength @xmath53 between two sites in @xmath52 . under duality
it is mapped onto a bond of strength @xmath54 $ ] , where @xmath55 is the duality transformation of the at model@xcite which we list in the appendix .
the dual bond lies in the dual lattice @xmath51 , intersecting the original bond .
this situation is depicted in figure [ fig : duality exp ] .
the original bond of strength @xmath53 is depicted by the solid line while its dual is depicted by the dashed line .
suppose the only spatial variation of the couplings is one of anisotropy .
then an at system with couplings of strength @xmath56 in the horizontal direction and strength @xmath57 in the vertical direction , transforms under duality into an at system on the dual lattice @xmath51 with couplings of strength @xmath58 in the vertical direction and strength @xmath59 in the horizontal direction .
the free energies of the original system and its dual have the same singular part , and therefore we can write @xmath60= f_{s}[(\widetilde{k^{2}},\widetilde{\lambda^{2}}),(\widetilde{k^{1 } } , \widetilde{\lambda^{1 } } ) ] \ ; , \label{eq : fs anidual}\ ] ] where the first pair of couplings corresponds to the horizontal bonds , and the second pair to the vertical ones .
the duality transformation @xmath55 is self - inverse .
thus , an anisotropic at system with couplings @xmath61 $ ] such that @xmath62 , transforms under duality onto itself .
so , there exists a two dimensional self - dual manifold ( it would help the reader to view the duality transformation as a mathematical mapping in the four dimensional space of couplings @xmath61 $ ] ) .
the line of critical points of the isotropic at model lies within the subspace @xmath63 , defined by @xmath64 .
@xmath63 is located between a decoupled ising model where @xmath65 and a four - state potts model with @xmath66 .
two sub - areas of @xmath63 which map onto each other under duality are separated by a self dual line ( fulfilling @xmath67 ) .
the assumption that @xmath63 contains only two phases , with a single phase transition between them(where @xmath68 is non - analytic ) leads to the conclusion that this phase transition must occur on the self - dual line . in this way
the line of critical points of the isotropic at was located exactly@xcite .
let us now consider the corresponding subspace , @xmath69 , of the aat for which @xmath70 and @xmath71 .
@xmath69 lies between an anisotropic decoupled ising subspace where @xmath72 and an anisotropic four - state potts subspace with @xmath73 .
@xmath69 is invariant under duality since for @xmath74 we get ( see appendix ) @xmath75 .
it follows also that the self dual plane @xmath70 , @xmath62 is included in the subspace we are considering .
thus , again assuming that @xmath69 contains only two phases , with a single phase transition between them , we conjecture that the self dual plane is a plane of phase transitions ( this kind of argument was first used by fisch@xcite for the random - bond ising model , and later for potts models by kinzel and domany@xcite ) . since the space of couplings is four dimensional , while the self dual plane is only two dimensional , the self dual plane is only _ part _ of the three dimensional critical manifold in @xmath69 .
notice that this is unlike the case of the potts model or any model with a single coupling constant where the self dual line is _ identical _ with the critical line .
the argument that leads to locating the corresponding critical plane of the rbat is very similar to the one presented above , up to a subtle difference . consider a specific realization of a rbat on a square lattice with couplings @xmath76 distributed at random with probabilities @xmath77 and @xmath78 , respectively .
replacing each bond by its dual we get ( on the dual lattice ) a system with couplings @xmath58 and @xmath59 , distributed at random with probabilities @xmath77 and @xmath78 .
this defines a one to one mapping under duality from the ensemble of realizations of @xmath79 $ ] rbat systems , onto the ensemble of the dual systems @xmath80 $ ] . under duality
each bond is mapped onto the bond on the dual lattice which intersects the original bond ( see fig . [
fig : duality exp ] ) .
thus , the spatial distribution of the two types of bonds on the dual lattice is different from that of the original lattice .
but since all spatial distributions have the same weight in their ensemble , there is a one to one correspondence of bond configurations in the two ensembles , and we can write : @xmath81= f_{s}[(\widetilde{k^{1}},\widetilde{\lambda^{1}}),(\widetilde{k^{2 } } , \widetilde{\lambda^{2}}),p ] \ ; , \label{eq : fs rbdual}\ ] ] where it is understood that in calculating @xmath68 we average over all spatial bond distributions .
choosing @xmath82 we find that there is a self dual manifold at @xmath62 .
our ability to locate a self dual manifold for the case @xmath82 makes this choice of @xmath77 a convenient case for studying the rbat . viewing the duality transformation as a mathematical mapping in the space of couplings @xmath83 $ ] , it is identical to the duality transformation in the space of couplings @xmath84 $ ] of the anisotropic at model .
thus all conclusions which were based on duality in sec .
[ sec : duality ani ] apply here equally . following the logic of the previous subsection
we conjecture that the subspace of the rbat for which @xmath70 , @xmath85 and @xmath82 includes in it a critical plane defined by @xmath70 , @xmath62 . in order to study the effect of randomness on the at model
we need a reference non - disordered system for comparison .
when studying a specific rbat model with couplings @xmath83 $ ] it is not clear with which specific non - random at model with couplings @xmath53 should one compare .
none the less a comparison with the corresponding aat model with couplings @xmath84 $ ] seems to be quite natural .
consider the anisotropic model with bonds @xmath56 in the horizontal direction and bonds @xmath57 in the vertical direction .
next , consider as a perturbation , changing a small fraction @xmath77 of the bonds in the horizontal direction into bonds of strength @xmath57 and the same fraction of bonds in the vertical direction into bonds of strength @xmath56 .
when @xmath77 increases all the way to @xmath82 the anisotropy disappears and the model is completely random .
so , in a sense , the perturbation of randomness can take us continuously from the critical plane of the anisotropic model to the critical plane of the random model . in view of this simple picture
we have performed simulations of the anisotropic and random - bond at models at several points on their self - dual planes described above .
we indeed find the two planes to be critical . in the next section
we give some details of the simulations .
in ref . we described a cluster algorithm for the at model and tested its performance ( on the isotropic model ) .
as explained there , the algorithm can be regarded from two different points of view .
the conceptually simpler and faster ( though less general ) one is that of the wolff embedding idea@xcite .
it is faster in the sense that its implementation on the computer runs faster since the freeze - delete procedure is less complicated .
the main idea is to embed into the at model an ising model of space dependent couplings @xmath86 and simulate it using the sw@xcite or wolff@xcite procedure for the ising model . to be explicit ,
consider the hamiltonian ( [ eq : rbat ] ) , and take the @xmath39 variables as fixed , so we can write @xmath87 @xmath88 represented by the second sum is a constant , and remembering that we are considering @xmath89 for all ( i , j ) , @xmath90 is a ferromagnetic ising model in the @xmath38 variables with couplings @xmath91 . simulating this ising model with any procedure that will maintain detailed balance with respect to @xmath90 and will not change the value of @xmath88 , will also maintain detailed balance with respect to @xmath92 .
so we can use for example the sw@xcite or wolff@xcite procedure for the ising model .
this by itself will maintain detailed balance but will not be ergodic since the @xmath39 variables will not be updated . obviously to update
the @xmath39 variables the same process should be repeated , holding the @xmath38 variables fixed and simulating an ising hamiltonian with the @xmath39 variables . to summarize , our procedure goes as follows : choose at random whether to embed into the at hamiltonian an ising hamiltonian in the @xmath38 ( or @xmath39 ) spins .
pick a random site in the lattice , grow a cluster of @xmath38 ( or @xmath39 ) spins , using the wolff single cluster procedure with the ising hamiltonian ( ( [ eq : at - is ] ) or the opposite one in case of a @xmath39 embedding ) , and flip it .
all our simulations were performed at the critical plane defined by @xmath93 : \ ; ( k^{2},\lambda^{2})=(\widetilde{k^{1}},\widetilde{\lambda^{1 } } ) \;\;,\;\ ; 0 \leq \lambda^{1 } \leq k^{1 } \;\ } \ ; , \label{eq : def rbs}\ ] ] as explained in sec .
[ sec : rbat ] ( for the anisotropic model just omit the @xmath82 ) . for convenience
we also use the following notation for the coupling constants : @xmath94 the first series of measurements were performed at five points @xmath95 , which we label as @xmath96 .
these were chosen so as to interpolate between @xmath97 , which is an anisotropic ( or random - bond ) decoupled ising critical point , and @xmath98 , which is an anisotropic ( or random - bond ) four - state potts critical point .
the points @xmath99 interpolate in a similar manner to the way in which the critical line of the pure at connects the pure decoupled ising critical point with the pure four - state potts critical point .
thus we chose the five points @xmath100 to lie equidistantly on the line @xmath101 the slope @xmath102 was chosen so that the line ( [ eq : line ] ) lies parallel ( in the @xmath103 plane ) to the critical line of the pure at model , @xmath104 .
note that once @xmath105 are defined , @xmath106 are defined through ( [ eq : def rbs ] , [ eq : def xz ] ) .
we clearly have some freedom in our choice of @xmath107 , the decoupled ising couplings ; setting @xmath108 in ( [ eq : def rbs ] ) one gets @xmath109 as well .
the remaining parameter , @xmath110 , is chosen so that ( see ( [ eq : def rbs ] ) ) @xmath111 since the extent of deviation from pure behavior is obviously determined not only by @xmath77 but also by the difference between the two sets of couplings , the ratio of @xmath112 was chosen so that randomness will be pronounced@xcite . while ( [ eq : ratio1 ] ) defines the first simulation point @xmath113 , the fifth simulation point @xmath114 is chosen to be a four - state potts point lying at the intersection of the line ( [ eq : line ] ) with the four - state potts line @xmath115 ( or @xmath116 ) .
the other three simulation points @xmath117 lie equidistantly on the line ( [ eq : line ] ) between @xmath118 and @xmath119 : @xmath120 thus the first series of points @xmath99 at which we made measurements @xmath121 is defined through equations ( [ eq : def rbs],[eq : def xz],[eq : ratio1],[eq : x1place ] ) and the definition of @xmath119 .
two additional measurement points were intended to monitor the effect of varying the amount of randomness in the model on the critical behaviour . in order to define these two points
we define two series of five points each , @xmath122 and @xmath123 .
similarly to the series @xmath121 , these two series are defined through equations ( [ eq : def rbs],[eq : def xz],[eq : x1place ] ) .
but in contrast with the series @xmath121 , eq . ( [ eq : ratio1 ] ) is substituted by @xmath124 for the series @xmath122 , and by @xmath125 for the series @xmath123 .
though we defined here ten additional points we actually performed measurements only at the points @xmath126 and @xmath127 .
the points @xmath126 , @xmath127 , @xmath128 represent three rbat ( or aat ) models with coupling ratios @xmath129 and @xmath130 respectively .
we calculated the energy@xcite per site : @xmath131=-\frac{1}{l^{2}}[{\langle \sum_{<i , j>}\{k_{i , j}(\sigma_{i}\sigma_{j}+\tau_{i}\tau_{j})+ \lambda_{i , j}\sigma_{i}\tau_{i}\sigma_{j}\tau_{j}\ } \rangle } ] \label{eq : e}\ ] ] where @xmath132 is the linear lattice size ; the angular brackets denote the usual thermal mc average whereas the square brackets denote the quenched average over random - bond configurations .
the specific heat per site follows from the energy fluctuations @xmath133 .
\label{eq : cv}\ ] ] we define @xmath134 to compute the magnetization per site , @xmath135 , we used @xmath136 \ , , \label{eq : m}\ ] ] taking advantage of the symmetry between the @xmath38 and @xmath39 spins to increase accuracy .
another order parameter of the transition plane we are considering is the polarization@xcite , @xmath137 , defined as @xmath138 \
, , \label{eq : p}\ ] ] which measures the magnetization of the @xmath139 spins .
we calculated the magnetic susceptibility of the @xmath38 spins using@xcite @xmath140 \label{eq : xi}\ ] ] and also measured @xmath141 , the size of the cluster flipped at each step , and calculated @xmath142 .
there is a connection between the size of the clusters and the susceptibility @xcite which is common for all algorithms that generate non - interacting clusters of spins ( if all spins in a cluster have the same value ) . for the wolff single cluster procedure it has the simple form @xcite @xmath143
lastly we measured the susceptibility of the @xmath139 spins , @xmath144 : @xmath145 \label{eq : xip}\ ] ] for each sample we estimated the effective autocorrelation times@xcite of the energy , @xmath146 and the magnetization , @xmath147 using the method outlined in ref . .
the length of an individual run was often too short for a precise estimate , but averaging over all samples gave a rough estimate of @xmath148 $ ] .
these estimates of @xmath149 allowed us to verify that all samples were equilibrated well enough before we started collecting data .
typically we discarded at least @xmath150 spin configurations .
the fluctuations in the random bond configurations introduce , in addition to the usual mc error , another source of statistical error .
thus the total error for the specific heat , @xmath151 , for example , is given by @xmath152 where @xmath153 is the number of random bond samples , @xmath154 is the length of the mc runs and @xmath39 is the autocorrelation time of the algorithm ( which we can estimate by @xmath149 ) .
@xmath155 is the variance of @xmath1 within a given sample and is due to thermal fluctuations , while @xmath156 is the variance of the exact @xmath157 within the ensemble of random bond samples .
thus both the thermal fluctuations and the sample to sample fluctuations contribute to the error .
we found that both terms in ( [ eq : p toterr ] ) imply that simulations of the at models and the potts model are much more expensive in computer time than simulations of the ising model .
first there is an increase in @xmath39 in consistence with our findings in the pure models@xcite .
e.g. at @xmath158 we find for the random ising model @xmath159 compared with @xmath160 for the random four - state potts model .
second we found that for all thermodynamic quantities @xmath137 the relative variance @xmath161 was larger for the random four - state potts model than for the random ising model .
this effect was strongest for the specific heat , e.g. for @xmath158 @xmath162 of the random four - state potts model was approximately an order of magnitude larger than that of the random ising model . in accordance with these findings , the number of samples simulated for the ising model varied between @xmath163 for @xmath164 and @xmath165 for @xmath158 , while for the four - state potts model it varied between @xmath166 for @xmath164 and @xmath167 for @xmath158 .
we expect the critical specific heat of the aat to follow the finite size scaling@xcite form of ( [ eq : c(t)ani ] ) , @xmath168 the exponent ratio @xmath6 is predicted analytically@xcite to vary continuously along the critical line of the pure at model between @xmath169 at the decoupled ising critical point and @xmath170 at the four - state potts critical point .
one expects the same variation to hold for the corresponding critical manifold of the aat at which the simulations were performed . in figure
[ fig : cvacft1 ] we plot the critical specific heat , @xmath1 as a function of @xmath171 for the five critical aat models @xmath172 .
fits are made to the form ( [ eq : c(l)ani ] ) , and the estimated values of @xmath173 are listed in table [ tab : aat1 ] . since the model @xmath174 is an anisotropic ising model , its specific heat does not follow the form ( [ eq : c(l)ani ] ) , but rather @xmath175 .
the anisotropic four - state potts model @xmath176 should have @xmath177 with a logarithmic correction@xcite to scaling @xmath178 . in the range of @xmath132
considered , the logarithmic correction results in an estimate for @xmath173 which is too low .
the results for the specific heat of the aat models clearly demonstrate the variation of @xmath173 from its ising value , @xmath179 to its four state potts value .
as will be seen shortly this is quite different from the results for the rbat models . in order to analyze the specific heat results of the rbat we first tried to fit the results to ( [ eq : c(l)ani ] ) , but unlike the specific heat of the aat ,
this was not possible for the full lattice size range @xmath180 .
thus in fig .
[ fig : cvrcfc1 ] we show the specific heat as a function of @xmath171 for the five critical rbat models @xmath172 along with fitting curves to ( [ eq : c(l)ani ] ) , where only data for @xmath181 were used to fit .
the estimated exponents @xmath173 obtained from these fits are 0 within errors indicating a possible logarithmic behaviour .
in addition , fig .
[ fig : cvrcfc1 ] indicates a crossover of the effective exponents ( the slope between every two successive data points ) from higher to lower values .
physically we expect that this is due to a crossover from the pure models exponents to some new random critical behaviour .
we are thus led to the following finite size scaling ansatz for the specific heat @xmath182\ ; , \label{eq : c fss cross}\ ] ] where @xmath183 is the critical exponent ratio of the corresponding pure model and @xmath184 .
an important point to note is that for each random bond at model , following the discussion at the end of sec .
[ sec : duality ran ] , @xmath183 is taken from our results ( listed in table [ tab : aat1 ] ) for the corresponding anisotropic at model .
@xmath185 is a measure of the randomness , and is related through the relation @xmath186 to a crossover length @xmath187 , at which crossover from the pure model s power law behaviour to the random logarithmic behaviour occurs .
thus for @xmath188 our eq .
( [ eq : c fss cross ] ) reduces to ( [ eq : c(l)ani ] ) while for @xmath189 and @xmath190 a logarithmic behaviour is attained , @xmath191 apart from crossing over to the correct pure result ( [ eq : c(l)ani ] ) when @xmath192 , eq .
( [ eq : c fss cross ] ) was constructed so that in the ising model limit , @xmath193 , it becomes @xmath194 this is the finite size scaling form of ( [ eq : c(t)ising ran ] ) which was shown by wang et al.@xcite to fit the random bond ising model well .
this form will be examined below as another candidate for describing the rbat models . in fig .
[ fig : cvrcaat ] the specific heat of the five critical rbat models @xmath172 are plotted again but with fits to ( [ eq : c fss cross ] ) using the full lattice size range @xmath180 , on a semilogarithmic scale .
the fitting parameters are listed in table [ tab : cv fit ] . for these points ( with large randomness , @xmath195 ) , the crossover lengths @xmath187 are found to be 1 . in fig .
[ fig : cvrdaat ] the specific heats of three critical rbat models @xmath196 are plotted with the fitting functions of ( [ eq : c fss cross ] ) .
@xmath183 is taken from the corresponding anisotropic models and the fitting parameters are listed in table [ tab : cv fit ] .
@xmath183 of the three models @xmath196 is very similar ( .40 , .37 , .37 respectively ) but they differ in their amount of randomness , @xmath130 respectively . the form ( [ eq : c fss cross ] ) seems to describe all three models adequately with the main change being in the crossover length @xmath187 which decreases as randomness increases .
the estimates for @xmath187 can be compared with results from random bond ising models@xcite with the same coupling ratios .
for @xmath197 we obtain @xmath198 compared with @xmath199 and @xmath200 for the respective ising models .
according to the harris criterion one expects the crossover length to scale roughly as @xmath201 where @xmath202 is the crossover exponent and @xmath203 ( @xmath204 , @xmath205)is a measure of the randomness .
for the rbat models considered here @xmath206 , while for the random bond ising models randomness is marginal ( @xmath207 ) , so that the much smaller crossover lengths that we find are consistent with the harris criterion .
since there is no theoretical prediction for the critical behaviour of the rbat model , we examined possible scaling forms for the specific heat , other than ( [ eq : c fss cross ] ) .
a natural candidate is a double logarithmic behaviour as in eq .
( [ eq : c(t)ising ran ] ) .
this form was predicted by dd for the disordered baxter@xcite and ising@xcite models and confirmed for the latter model . in its asymptotic limit , the finite size scaling form of ( [ eq : c(t)ising ran ] ) is @xmath208 we compare this form with ( [ eq : c fss log ] ) as both these forms are asymptotic and do not include crossover from the pure behaviour . in figure [ fig : cvrclg1 ] we plot again our specific heat results for all the models tested except for @xmath126 which is not considered due to its large crossover length .
the results are fitted by ( [ eq : c fss log ] ) in fig .
[ fig : cvrclg1 ] and by ( [ eq : c fss log log ] ) in fig .
[ fig : cvrcll1 ] , for @xmath209 . as found by wang et al .
, for the random bond ising model @xmath174 , agreement with ( [ eq : c fss log log ] ) is much better than with ( [ eq : c fss log ] ) , according to our error analysis . for the model @xmath210
too agreement with ( [ eq : c fss log log ] ) is better but this can be attributed to slow crossover behaviour from the small power law @xmath211 ( the condition @xmath212 is not fulfilled for such small @xmath183 and for the range of @xmath132 considered ) . for the models @xmath213 the quality of both fits is comparable while for the random bond four state potts @xmath176 the fit ( [ eq : c fss log ] ) is better . for the model @xmath127
we have obtained larger statistics ( more than 900 samples for @xmath214 and 300 samples for @xmath158 ) and the result is conclusively in favor of ( [ eq : c fss log ] ) .
last we considered the possibility that the crossover from the power law behaviour to a logarithmic behaviour is yet followed at even larger @xmath132 by a crossover to a double logarithmic behaviour .
this was done by fitting the specific heat data for @xmath215 according to ( [ eq : c fss dotz ] ) .
though the fits were of good quality , the estimated crossover lengths @xmath216 did not make sense physically .
for the @xmath127 model we found @xmath217 which actually implies a logarithmic behaviour for all accessible values of @xmath132 .
this should be compared against our estimate @xmath218 and the value @xmath219 for the corresponding ising model ( see above ) . for the models
@xmath121 we find that @xmath220 increases monotonically from the ising model @xmath174 where @xmath221 to the four state potts model @xmath176 where @xmath222 .
this is contrary to the expectation that the crossover length should decrease due to the increase in the crossover exponent @xmath202 , as implied from ( [ eq : lc harris ] ) .
we conclude that our results are in favor of a logarithmic divergence of the specific heat and not a double logarithmic one for the rbat model , excluding the random bond ising model .
the finite size scaling form ( [ eq : c fss cross ] ) gives the best agreement with our results for all the models , including those with weak randomness , and for the full lattice size range @xmath180 .
thus we predict for @xmath223 the scaling form : @xmath224\ ; , \label{eq : c scaling cross}\ ] ] a possibility which we tend to rule out is that both types of critical behaviors ( that is ( [ eq : c scaling cross ] ) and ( [ eq : c(t)ising ran ] ) ) occur and that there is some bifurcation plane which separates the two types of behaviors .
according to finite size scaling theory one expects that the critical susceptibility diverges as @xmath225 for large enough @xmath132 .
the exponents ratio @xmath226 is predicted analytically@xcite to be constant , @xmath227 , all along the critical line of the pure at model .
this is expected also to hold for the corresponding critical manifold of the aat .
figure [ fig : xtacall ] shows results for the critical susceptibility as a function of @xmath171 for the five critical models , @xmath172 , of the aat . so that the points wo nt fall on top of each other , @xmath21 for the model @xmath99 has been multiplied by @xmath228 .
the solid lines are linear fits to the form ( [ eq : xfss ] ) for @xmath229 , yielding estimates for critical exponents @xmath43 which are listed in table [ tab : aat1 ]
. all results seem to be consistent with the analytic prediction , giving a combined best estimate of @xmath230 . an effective exponent analysis ( or a trend analysis )
shows that as @xmath171 increases , @xmath231 approaches the value @xmath232 from above .
the finite size scaling form ( [ eq : xfss ] ) and the value of @xmath233 have been predicted@xcite for the disordered decoupled ising model and confirmed in mc simulations@xcite .
figure [ fig : xtrcall ] shows results for the critical susceptibility as a function of @xmath171 for the five random critical models @xmath172 , of the rbat .
the same analysis has been carried out as for the aat and the estimated critical exponents @xmath43 are listed in table [ tab : rand1 ] .
the estimates for the decoupled ising critical point @xmath174 and for the models @xmath234 are consistent with @xmath233 .
on the other hand for the model @xmath235 and the random - bond four - state potts model @xmath176 we obtain @xmath236 and @xmath237 respectively .
we also tried fitting the results with the form @xmath238 for @xmath229 .
the obtained estimate for @xmath239 is listed in table [ tab : rand1 ] . for the random - bond four state potts
we find @xmath240 while for the models @xmath241 the error in @xmath239 is at least of the same order as the estimate for @xmath239 itself . the critical magnetization of the five anisotropic models is expected@xcite to follow the form @xmath242 with @xmath243 .
applying straightforward finite size scaling analysis , as described above for the susceptibility , to the magnetization results , we obtained estimates for @xmath244 which are listed in table [ tab : aat1 ] .
the results are slightly below the expected value , but an effective exponent analysis ( or a trend analysis ) shows that as @xmath171 increases @xmath245 approaches the value @xmath246 from below .
the finite size scaling form ( [ eq : mfss ] ) and the value of @xmath247 have been predicted@xcite for the disordered decoupled ising model and confirmed in mc simulations@xcite . figure [ fig : m1rcall ] shows results for the critical magnetization as a function of @xmath171 for the five critical models , @xmath172 , of the rbat . a fit to the form ( [ eq : mfss ] ) yields the estimated critical exponents @xmath244 listed in table [ tab : rand1 ] .
the estimates for the decoupled ising critical point @xmath174 and for the models @xmath234 are consistent with @xmath247 .
on the other hand for the model @xmath235 and the random - bond four - state potts model @xmath176 we obtain @xmath248 and @xmath249 respectively .
we also tried fitting the results with the form @xmath250 for @xmath229 .
the obtained estimate for @xmath251 is listed in table [ tab : rand1 ] . for the random - bond four state potts model
we find @xmath252 while for the models @xmath241 the error in @xmath253 is at least of the same order as the estimate for @xmath251 itself .
since the polarization is an order parameter of the transition , the critical polarization , @xmath137 , and @xmath144 should follow the forms [ eq : pxp1 ] @xmath254 @xmath255 the critical exponents @xmath256 and @xmath257 are known analytically@xcite all along the critical line of the pure at model .
for example the ratio @xmath258 varies continuously between its value at the decoupled ising critical point , @xmath259 , and its value at the four state potts critical point , @xmath260 .
these two extreme values can be obtained , through their relations with @xmath244 , for any ( pure , anisotropic or disordered ) decoupled ising or four - state potts model . for the decoupled ising models
@xmath72 ( or @xmath65 for the pure model ) , the @xmath38 spins system and the @xmath39 spins system are decoupled , hence @xmath261 this implies @xmath262 for all the decoupled ising models .
for the four state potts models @xmath263 and @xmath264 ( or @xmath265 for the pure four state potts ) so that @xmath38 spins , @xmath39 spins and @xmath139 spins are equivalent , and @xmath266 it follows that @xmath267 . thus interpreting our results as @xmath247 for the critical plane of the rbat implies @xmath268 for the random - bond decoupled ising model , and @xmath269 for the random - bond four - state potts model .
the question still remains whether the variation between these two values along the critical plane of the rbat will be continuous , as is the case for the pure model , or whether it will become discontinuous under the influence of disorder .
the results presented below are in favor of a continuous variation .
the values of @xmath270 can be obtained through the scaling relation , @xmath271 which our results fully confirm .
this leads to @xmath272 for the ising models and @xmath273 for the four - state potts models .
figures [ fig : xpacall ] and [ fig : p3acall ] show the critical @xmath144 and polarization , @xmath137 , respectively , as a function of @xmath171 for the five critical models , @xmath172 , of the aat .
the solid lines are linear fits to the forms ( [ eq : pxp1 ] ) for @xmath229 , yielding estimates for critical exponents @xmath270 and @xmath258 which are listed in table [ tab : aat1 ] .
as expected we find that @xmath270 varies between @xmath274 for the decoupled ising model and @xmath275 for the four state potts model .
@xmath258 varies accordingly fulfilling the scaling relation ( [ eq : scaling1 ] ) .
figures [ fig : xprcall ] and [ fig : p3rcall ] show the critical @xmath144 and polarization , @xmath137 , respectively , as a function of @xmath171 for the five critical models , @xmath172 , of the rbat .
the same analysis has been carried out as for the aat and the estimated critical exponents @xmath270 and @xmath258 are listed in table [ tab : rand1 ] .
the results for the random - bond four - state potts , @xmath176 , agree with @xmath276 and @xmath277 to a high precision and thus need no further explanation .
there is a clear mismatch between the estimated values of @xmath270 and @xmath258 and their expected values for the random - bond ising model @xmath174 .
even the left side of the relation ( [ eq : pising ] ) , @xmath278 , which implied @xmath279 , is far from being met , but rather we find @xmath280 both for the anisotropic and for the random - bond ising models .
this is probably connected to the fact that we are measuring , as is usual in a mc calculation , @xmath281 and @xmath282 which are larger than @xmath283 and @xmath284 respectively .
both in the random - bond and in the anisotropic ising model the relative variance of @xmath137 is larger than the relative variance of @xmath135 .
this makes @xmath282 a worse estimator of @xmath285 than @xmath281 is of @xmath283 .
the discrepancy is a finite size effect@xcite and should decay as @xmath132 increases . indeed
if we estimate @xmath258 from lattices of size @xmath286 we obtain @xmath287 which is consistent with the exact value @xmath259 .
an estimate of @xmath270 from lattices of size @xmath288 yields @xmath289 which is consistent with the exact value @xmath272 .
the results for @xmath144 and @xmath137 seem to favor a continuous variation of @xmath258 and @xmath270 for the rbat models as is the case for the aat models .
as explained above , for symmetry reasons one expects that if @xmath290 then @xmath291 .
this is consistent with the variation of exponents we find .
none the less since we found the specific heat of all models to diverge logarithmically one could expect the rbat models to be more universal . a possible scenario to expect could be @xmath292 for all rbat models but the ising model . then assuming @xmath293 with ( [ eq : scaling1 ] ) would imply @xmath294 for the ising model and @xmath295 for all the other rbat models .
such a behaviour would of course be masqueraded by crossover , but none the less we do not find evidence for such a behavior in our results .
further investigation of this problem is clearly needed .
we ve examined the effect of bond - disorder on the line of critical points of the at model .
a duality transformation was used in order to locate a critical manifold of the random bond at model which corresponds to the line of critical points of the pure model .
an anisotropic at model was used as a convenient reference pure model .
our results consist of the finite size dependence of @xmath296 at criticality of several random bond and anisotropic at models .
the models @xmath297 interpolate between ising and four - state potts models with large disorder , thus also monitoring the effect of changing the crossover exponent @xmath183 .
the models @xmath196 serve to monitor the effect of changing the amount of disorder while keeping the crossover exponent approximately the same ( @xmath298 ) .
the results of the critical specific heat of all the rbat models agree very well with the crossover behaviour ( [ eq : c fss cross ] ) , according to which for @xmath299 , @xmath1 diverges with the same exponent @xmath183 as the corresponding aat model , while for @xmath300 , @xmath1 grows as @xmath171 . our estimates for the crossover lengths of the models @xmath301 are much smaller than estimates of wang et al . for random bond ising models with the same amount of disorder .
this is consistent with the harris criterion and the crossover exponent @xmath302 of the former models and the crossover exponent @xmath207 for the latter models . for the random bond ising model
we find a double logarithmic divergence ( [ eq : c fss dotz ] ) , or ( [ eq : c fss log log ] ) asymptotically , as was found by wang et al .
we view the adequacy of ( [ eq : c fss cross ] ) for all the rbat models ( with different @xmath183 and different amounts of randomness ) for the full lattice range @xmath303 , together with the consistency of the crossover lengths with physical understanding ( harris criterion ) , as a strong indication that the rbat models ( including the four state potts model ) exhibit a a logarithmic divergence of @xmath1 .
we tend to rule out the possibility of a double logarithmic divergence of @xmath1 as predicted by dd@xcite for the baxter model , for reasons explained in sec .
[ sec : results ] .
although @xmath304 and @xmath305 do not vary for the pure model , our results indicate a possibility that @xmath304 and @xmath305 do vary for the random models . none the less this very small variation might be an artifact of a correction to scaling , e.g. of a logarithmic type .
the exponents connected with the polarization @xmath137 and the susceptibility @xmath144 , @xmath306 and @xmath307 which vary continuously for the pure at model seem to do so also for the rbat model . but on the basis of universality the possibility that these exponents are actually the same ( with some non - universal corrections to scaling and a crossover effect ) for all rbat models excluding the ising model , seems to be a plausible scenario in need of further investigation .
very recently an experimental study of a two dimensional four - state potts system with quenched impurities was done@xcite and the values of @xmath308 , @xmath309 , and @xmath310 were obtained .
these are consistent with our findings for the four - state potts system @xmath176 but with a value of @xmath311 which is smaller than ours .
sw would like to thank h.j .
heermann and the many particle group of the hlrz jlich for their warm hospitality and the generous allocation of computer time on the intel ipsc/860 and the paragon .
this research has been supported by the us - israel bi - national science foundation ( bsf ) , and the germany - israel science foundation ( gif ) .
we list here the duality transformation @xmath312 $ ] of the at model and a few of its properties . for convenience
we use instead of @xmath53 different variables for the coupling constants , @xmath313 ; @xmath314 under a duality transformation @xmath55 , a bond of strength @xmath315 transforms into a bond on the dual lattice of strength @xmath316 given by @xcite @xmath317 we list several properties of the transformation .
it is self - inverse , i.e. , @xmath318=\vec{x}$ ] .
it maps the zero temperature point ( @xmath319 ) onto @xmath320 ( @xmath321 ) and vice versa .
the ising subspace @xmath322 and the 4-state potts subspace @xmath323 are invariant under @xmath55 .
the subspace @xmath63 , defined by @xmath324 ( or @xmath325 ) , is also invariant under @xmath55 .
the line @xmath104 is a self dual line and its intersection with @xmath63 is the line of critical points of the at model .
it is also true that if @xmath326 then @xmath327 and vice versa .
99 email address : [email protected] email address : [email protected] a.b .
harris _ j.phys .
c_,*7*:1671,(1974 ) . r.b .
stinchcombe , in _ phase transitions and criticalphenomena_,edited by c. domb and j.l .
lebowitz(academic , new york , 1983 ) , vol
w.kinzel and e. domany _ phys .
b_,*23*:3421 , ( 19801 ) .
t.c . lubensky and a.b .
_ , * 24*:311,(1974 ) .
c. jayaprakash and h.j .
rev . b_,*16 * : 3987,(1977 ) .
vik s. dotsenko and vl s. dotsenko , adv .
* 32 * , 129(1983 ) .
shalaev , sov . phys .
solid state * 26 * , 1811 ( 1984 ) .
r. shankar , phys rev .
lett . * 58 * , 2466 ( 1987 ) ; phys rev . lett . * 61 * , 2390 ( 1988 ) .
ludwig , phys rev .
lett . * 61 * , 2388 ( 1988 ) .
w. selke , in computer simulations in condensed matter physics iii , edited by d.p .
landau , k.k .
mon , and h.b .
schttler ( springer , heidelberg , 1991 ) .
d. andelman and a. nihat berker _ phys .
b_,*29*:2630,(1984 ) . b. derrida and e. gardner , _
j. phys a_,*17 * , 3223 ( 1984 ) .
ludwig , _ nuc .
b_*285 * , 97 ( 1986 ) .
s. chen , a.m. ferrenberg and d.p .
lett._,*69*:1213,(1992 ) .
a.n . berker _
physica a _ * 194 * , 72 ( 1993 ) .
novotny and d.p .
rev . b_,*24 * : 1468,(1981 ) .
baxter and f.y .
lett._,*31*:1294,(1973 ) .
j. ashkin and e. teller , _ phys .
rev._*64 , * 178 ( 1943 ) .
f.y . wu and k.y .
lin , _ j.phys .
c_*7 , * l181 ( 1974 ) .
e. domany and e.k .
riedel , _ phys .
b_*19 , * 5817 ( 1979 ) .
ditzian , j.r .
banavar , g.s . grest and l.p .
kadanoff , _ phys .
b_*22 , * 2542 , ( 1980 ) . s. wiseman and e. domany , _ phys .
rev e_*48 , * 4080 ( 1993 ) .
. baxter _ exactly solved models in statistical mechanics _ , academic press ( 1982 ) .
b. nienhuis , in _ phase transitions and critical phenomena _ , edited by c. domb and j.l .
lebowitz(academic , new york , 1987 ) , vol .
f.c . alcaraz and c. tsallis , j.phys . a * 15 * , 587 ( 1982 ) .
baxter , ann .
* 70 * , 193 ( 1972 ) .
vik s. dotsenko and vl s. dotsenko _ j.phys .
a_*17*:l301,(1984 ) . d. matthews - morgan , d.p .
landau and r.h .
lett._,*53*:679,(1984 ) .
for a review see : a.d .
sokal , _ nuc .
suppl.)_,*20,*55 , ( 1991 ) k. binder , d.w .
heermann , _ monte carlo simulations in statistical physics .
_ springer - verlag , berlin , ( 1988 ) .
heuer , phys .
b * 42 * , 6476 ( 1990 ) .
wang , w. selke , vi.s .
dotsenko and v.b .
physica a_,*164*:221,(1990 ) .
m. hennecke and u. heyken , _ j. stat .
phys._*72 * , 829 ( 1993 ) .
f.y . wu and y.k .
wang , j. math .
* 17 * , 439 ( 1976 ) .
r. fisch _ j. stat .
phys._,*18*:111,(1978 ) .
u. wolff , _ phys .
lett._*62 , * 361 ( 1989 ) .
s. caracciolo , r.g .
edwards , a. pelissetto , and a.d .
sokal , _ nuc .
suppl.)_*20 , * 72 ( 1991 ) .
swendsen and j.s .
wang , phys .
lett . * 58 * , 86 ( 1987 ) .
andreichenko , vl.s .
dotsenko , w.selke , and j .- s .
wang , nucl . phys .
b * 344 * , 531 ( 1990 ) .
a factor of @xmath328 has been omitted , but is irrelevant since all measurements are at a single temperature .
b_,*300*:501,(1988 ) .
sokal , in _
computer simulation studies in condensed matter pysics : recent developments , _ ed .
landau , k.k . mon and h
barber , in _ phase transitions and critical phenomena _ , edited by c. domb and j.l .
lebowitz , vol .
li and a.d .
sokal , _ phys .
lett._*63 , * 827 ( 1989 ) .
l. schwenger , k. budde , c. voges , and h. pfnr , _ phys .
rev . lett._*73 * , 296 ( 1994 ) . .
critical exponents ratios and fitting parameters from the first series of simulations , @xmath172 , and the second series @xmath329 of the anisotropic at model . the exponent ratios and parameters
are defined in the text . [ cols="<,<,<,<,<,<",options="header " , ] | the critical behaviour of a bond - disordered ashkin - teller model on a square lattice is investigated by intensive monte - carlo simulations .
a duality transformation is used to locate a critical plane of the disordered model .
this critical plane corresponds to the line of critical points of the pure model , along which critical exponents vary continuously . along this line
the scaling exponent corresponding to randomness @xmath0 varies continuously and is positive so that randomness is relevant and different critical behaviour is expected for the disordered model .
we use a cluster algorithm for the monte carlo simulations based on the wolff embedding idea , and perform a finite size scaling study of several critical models , extrapolating between the critical bond - disordered ising and bond - disordered four state potts models .
the critical behaviour of the disordered model is compared with the critical behaviour of an anisotropic ashkin - teller model which is used as a refference pure model .
we find no essential change in the order parameters critical exponents with respect to those of the pure model .
the divergence of the specific heat @xmath1 is changed dramatically .
our results favor a logarithmic type divergence at @xmath2 , @xmath3 for the random bond ashkin - teller and four state potts models and @xmath4 for the random bond ising model . |
can a local excitation ( source ) in classical field theories be invisible to observers outside the region of excitation ?
this question has recently received renewed interest .
berry _ et al . _
@xcite described a peculiar excitation case for the one - dimensional wave - equation of a perfectly elastic string under tension .
they show that the response of the string can be made to be confined to a bounded region by carefully choosing a forced excitation of oscillatory type .
this means that the excitation will not propagate away along the string .
denardo gives a simple and intuitive explanation by using a wave interference argument @xcite .
et al _ @xcite discuss conditions of finite string length and dissipation .
other recent work investigated non - propagating excitations include marengo and ziolkowski @xcite who discuss the generalization of non - propagating conditions of dalembertian ( @xmath0 ) operators and its temporally reduced version the helmholtz operator ( @xmath1 ) on various related classical scalar and vector fields .
marengo , devaney and ziolkowski @xcite give the condition for time - dependent but not necessarily time - harmonic non - radiating sources and for selective directional radiation for the inhomogeneous wave equation in three spatial dimensions .
marengo and ziolkowski @xcite generalize these conditions to more general scalar and vector field dynamics .
marengo , devaney and ziolkowski @xcite also give examples in one and three spatial dimension for the time - harmonic case .
hoenders and ferwerda @xcite discuss the relationship of non - radiating and radiating parts of the case of the reduced helmholtz equation , which can be derived from the string equation by assuming general oscillatory time solutions ( see @xcite ) .
denardo and miller @xcite discuss the related case of leakage from an imperfect non - propagating excitation on a string .
gbur @xcite provides a comprehensive recent review of this topic and the reader is referred to this review for more detailed historical context . of the earlier work
the following contributions are particularly relevent for the discussion here : schott @xcite gave the condition for non - radiation of a spherical shell on a circular orbit .
bohm and weinberg @xcite extended this result to more general spherical charge distributions and goedecke @xcite showed how an asymmetrical charge distribution with spin is non - radiating .
all of these works are concerned with the case of spatially moving sources .
finally it is noteworthy , that non - radiating sources play an important role in inverse problems and have been investigated in a one - dimensional electrodynamic situation by habashy , chow and dudley @xcite . in this paper
our purpose is to describe this phenomena in the case of a lattice string in one dimensions by discretizing dalembert s solution .
this approach is used extensively to simulate vibrating strings and air tubes of musical instruments .
see @xcite and references therein .
this leads to explicit dynamical constructions of previously reported non - propagating excitations .
its simplicity allows for additional insight into the mechanism that allows for the local confinements and the conditions under which they occur .
i will show how the basic mechanisms that provide a time - harmonic stationary non - propagating excitation in one dimension as studied by berry _
et al . _ and gbur _ et al .
_ @xcite allows for a much wider class of excitations .
for instance can such an excitation be relieved from the time - harmonic assumption beyond one period allowing for non - propagating sources that are short - lived .
directional excitations can easily be achieved using very simple bidirectional excitation patterns .
these are explicit constructions of such waves in one spatial dimension whose general condition of existence in the three - dimensional case has been derived by marengo , devaney and ziolkowski @xcite .
wave propagation can be virtually slowed down .
in general i will show that non - propagating excitations can be extended to steered excitation regions with basic physical restrictions imposed by the underlying field dynamics .
first i will give a quick derivation of the simple lattice model from the wave equation as can also be found in @xcite
. then i will give a new argument and construction of the berry _ et al .
_ type non - propagating excitation purely based on discrete string dynamics .
this will then be compared to the original approach .
then i will extend the discussion to examples of additional types of non - propagating waves , including directional and slowed waves .
finally i will discuss very general constraints on such `` steered '' localized excitations .
the lattice string model can easily be derived from the wave - equation by discretizing the dalembert solution .
hence the continuous case will be discussed first .
the one - dimensional homogeneous wave equation of the perfectly elastic string under tension is : @xmath2 where @xmath3 is derived from mass density @xmath4 and tension @xmath5 .
the dalembert solution of the homogeneous `` free field '' case has the well known form ( * ? ? ?
( 4 ) ) : @xmath6 hence the solution of the general of the homogeneous wave - equation are two propagating waves whose content is restricted by initial and boundary conditions . as wave - equation
is linear we have a connection between initial conditions and external driving forces .
driving forces can be seen as infinitesimal time frames that act on the wave dynamics by imposing a new initial condition at each point in time .
hence we need to consider the initial value problem to gain insight into both processes at once . at a given time frame @xmath7
let the following initial conditions hold : @xmath8 equation ( [ eq : in1 ] ) with ( [ eq : we ] ) gives a particular solution @xmath9 : @xmath10 taking the first temporal derivative of ( [ eq : we ] ) and satisfying equation ( [ eq : in2 ] ) we get : @xmath11 integrating with respect to @xmath12 we get ( * ? ? ?
* eq . ( 10 ) p. 596 ) : @xmath13 from equations ( [ eq : satin1 ] ) and ( [ eq : satin2 ] ) we can solve for the traveling wave components : @xmath14 we see that forced displacement @xmath15 splits evenly between left and right traveling waves and the integrated forced velocity @xmath16 splits with a sign inversion . for
our current discussion i will share the assumption of no initial velocity of berry _ et al .
_ @xcite and hence the integral over @xmath16 will vanish . for the infinite string this is already the complete solution for any twice differentiable function of free solutions and external forced displacements .
to arrive at lattice equations we discretize the solution of the wave - equation ( [ eq : we ] ) in time via the substitution @xmath17 where @xmath18 is the discrete time - step and @xmath19 is the discrete time index .
this automatically corresponds to a discretization in space as well , because in finite time @xmath18 a wave will travel @xmath20 distance according to ( [ eq : we ] ) .
the spatial index will be called @xmath21 .
the free - field discrete dalembert solution : @xmath22 in general we can always express all discrete equations in terms of finite time steps or finite spatial lengths .
we chose a temporal expression and substitute @xmath20 and suppress shared terms in @xmath23 to arrive at the index version of the discrete dalembert solution @xcite : @xmath24 by equations ( [ eq : force1 ] ) and ( [ eq : force2 ] ) we see that at an instance @xmath25 the discrete contribution of external forced displacements splits evenly between the traveling waves and we arrive at the discrete field equations including external forced displacements : @xmath26
next we will construct the non - propagating excitation from the lattice string dynamics directly . for simplicity and without loss of generality , we will assume a region aligning with the discretization domain throughout .
we want to construct an excitation which is confined to a length @xmath27 .
for now we will assume that the string should otherwise stay at rest .
this implies that there are no incoming waves into the region @xmath28 $ ] from the outside .
we are interested in a non - trivial excitation within the region .
first we consider the contributions to the position @xmath29 . as there are no incoming external waves we get : @xmath30 we do expect non - trivial wave @xmath31 to reach the boundary but we require the total outgoing wave to vanish we have : @xmath32 the necessary external forced displacement contribution to for cancellation needs to be : @xmath33 the complete incoming wave ( [ eq : fullwg1 ] ) will see the same forced contribution ( [ eq : forceeq ] ) and with equation ( [ eq : noin ] ) we get : @xmath34 hence the matched forced displacement leads to a reflection with sign inversion at the region boundary at @xmath29 .
following the same line of argument at point @xmath35 we get the related condition : @xmath36 with these two conditions we can study the permissible form of excitations .
first we assume an initial forced displacement impulse from a position @xmath37 in the interior of the domain @xmath38 .
hence @xmath39 and @xmath40 with @xmath41
. it will take half the impulse @xmath42 steps to reach the left boundary and the other half @xmath43 steps to reach the right one . at each boundary
the respective condition ( [ eq : cond1],[eq : cond2 ] ) needs to be satisfied and we get : @xmath44 the impulse will then reflect back and create periodic matching conditions . @xmath45 with @xmath46 .
hence we see that a single impulse will necessitate an infinite periodic series of forced external displacements at the boundaries to trap the impulse inside as each `` annihilation '' of a half - pulse reaching the boundary leads to a `` creation '' of a reflected one . the required impulse response of a boundary forced function @xmath47
can easily be observed from equations ( [ eq : boundf1][eq : boundf4 ] ) to be spatially periodic in @xmath48 with an initial phase factor dictated by the starting position @xmath37 .
additionally the functional shape of the impulse responses @xmath47 is completely defined for all time steps as @xmath49 for all times that equations ( [ eq : boundf1][eq : boundf4 ] ) do nt apply . a condition for stopping a non - propagating excitation can be derived from the fact that a impulse will return to its initial position every @xmath48 time steps . additionally it is easy to see that the traveling impulses will occupy the same spatial position every odd multiple of @xmath50 with a sign inversion .
hence an impulsive forced displacement @xmath51 with @xmath52 will cancel an initial impulse @xmath40 . from this
we can immediately deduce the following property : the shortest possible single impulse finite non - propagating excitation takes @xmath50 time - steps.[th : shortest ] and more generally : the time of any single impulse excitation finite non - propagating excitation has to be @xmath53 .
more importantly we observe the property : _ non - propagating excitations can be finite in duration_. this is an extension beyond berry _ et al . _
@xcite which assumes infinitely periodic temporal progressions in their derivations .
the general solution for discrete non - propagating wave functions can be derived by observing that any initial `` phase '' @xmath54 is orthogonal to other phases @xmath55 for @xmath56 , i.e. @xmath57 for @xmath58 . within a @xmath50 period @xmath59
is well - defined by @xmath60 .
interestingly though this provides the only restriction to the forced boundary functions .
this can be seen by theorem [ th : shortest ] .
after @xmath50 each @xmath54 will find constructive interference and can be annihilated or rescaled to an arbitrary other value @xmath61 .
hence any arbitrary succession of @xmath62 force distributions with a @xmath50 termination is permissible .
hence periodicity is not necessary .
the time harmonic case can be derived if the initial force distribution within the domain is not modified over time .
then a configuration will repeat after traveling left and right , being reflected at the domain boundary twice , traversing the length of the region twice .
hence the lowest permissible wave - length is @xmath48 . by reflecting twice the wave will have gone through a @xmath63 phase shift , but we note that the periodicity condition is also satisfied if any number of additional @xmath63 shifts have been accumulated . hence we get for permissible wave - numbers : @xmath64 or @xmath65 by allowing only even @xmath19 we get the berry _ at al . _ condition @xcite for an even square distribution .
the odd @xmath19 situation corresponds to the odd - harmonic out - of - phase construction proposed by denardo @xcite .
many of these properties can be seen visually in the numerical simulation depicted in figure [ fig : nonprop ] .
it is interesting to observe that two synchronous point - sources oscillating with the above phase condition will not be completely non - propagating
. they will only be non - propagating after waves created at the wave onset have escaped .
this is a refinement of the argument put forward by denardo @xcite and can intuitively be described as _ non - interference of the first trap period_. hence the first pairs of pulses will have half - amplitude components escaping in either direction but every subsequent period will be trapped .
this behavior , which could be called imperfect trapping or trapping with transient radiation , is depicted in figure [ fig : pointpair ] .
sources presented by berry _ et al .
_ and denardo @xcite do not display this behavior because the force is assumed to be oscillatory at all times and hence has no onset moment .
non - propagating excitations can be used as generic building blocks for other unusual excitation induced behavior on the string .
in particular i will next describe how to construct an uni - directional emitter , and a virtually slowed propagation .
in fact a non - propagating excitation can be seen as virtually stopping a wave at a particular position .
a one - sided open trap immediately suggests another unusual excitation type , namely the directional excitation . the string is to be excited in such a way that a traveling wave in only one direction results .
we start with a one - sided open trap .
this is a trap that uses a reflection condition ( [ eq : cond1 ] ) and ( [ eq : cond2 ] ) only on one side of an initial excitation .
evidently the wave then can only travel in the opposite direction .
for the discussion we will describe a right - sided propagator ( i.e. a propagator traveling with increasing negative index ) .
the trapping condition then reads : @xmath66 hence the trapping excitation point is a @xmath37 time - step lagging negative copy of the original excitation .
the emitted wave will have the form @xmath67 the emitting wave will show self - interference at a phase of @xmath68 time - steps , as can be seen in the simulation depicted in figure [ fig : onesided ] .
in general the self - interference phase can be chosen by the distance @xmath37 between the wave creation point and the trapping point .
it is worth noting that it is possible to eliminate interference by trapping the lagging contribution and hence create a wave non - interference directional wave left of the trapping region .
virtual slow waves can be achieved by alternating directional wave propagation with trapping .
the slowness of the wave propagation can be controlled by the number and and duration times of the traps along a propagation .
the propagation characteristics of the dynamic operator has not changed at all , hence we call the this state `` virtually slow '' as opposed to the case where the field itself induces a change in wave propagation speed .
this also means that within a slowed or `` steered '' region the wave propagation is the one prescribed by the dynamic operator @xmath69 on the string @xmath70 .
the amount of time spend in traps determines the overall slowness .
one example of slow wave consists of an immediate alteration between one stage of trapping and one step of one - sided propagation is illustrated in figure [ fig : slowwave ] .
the effective propagation speed of the wave can easily be read from the diagram to give @xmath71 . as is evident from theorem [ th : shortest ]
, a unit @xmath72 trap will last @xmath73 time - steps and will not propagate spatially and one step of free propagation will last one time - step and and make one spatial step , hence resulting in a spatial to temporal ratio of @xmath74 .
the trapping relations are : @xmath75 with @xmath76 .
the generalized interpretation of the excitation interaction lead to the general dynamical confinement of waves by external excitation . for instance following very similar arguments as for virtual slow waves a construction is possible which gives a slowed `` cone of influence '' by successively widening the trap boundaries at a speed slower than the the wave speed @xmath77 . by this argument
it is sufficient for the trap boundaries change to be less than @xmath77 for it to be trapping the wave .
this is not a necessary condition by the following counter - example : let the trap width be @xmath35 and change rapidly by some slope @xmath78 to some new constant width @xmath79 at which it becomes constant .
obviously the wave will then be able to reach the new boundary even though a local change of the boundary exceeded the dynamical speed @xmath77 .
the necessary condition can be seen from our previous construction . at a trap boundary a wave
is reflected and will propagate in the opposite direction of the domain following the linear characteristic @xmath77 .
only if this characteristics intersects with the dynamic trapping boundary will there be another externally forced reflection as illustrated in figure [ fig : steering ] .
these may in fact have regions where no trapping is necessary and possible .
it is important to note that while we assumed that the incoming wave vanishes , see equation ( [ eq : noin ] ) , the outgoing wave condition ( [ eq : noout ] ) does not change if there is in fact an incoming wave .
the `` reflection wave '' ( [ eq : cond1 ] ) and ( [ eq : cond2 ] ) can be rewritten for a non - zero incoming field without affecting the trapping : @xmath80 and @xmath81 these conditions are `` absorbing '' in the sense that an external field entering the trapping region will not leave it .
the `` non - interacting '' property of a trap defined by the periodic matching conditions ( [ eq : boundf1][eq : boundf4 ] ) can be seen by assuming a non - zero incoming wave at one point of the trap boundary @xmath82 .
then the total wave entering the trapping region the sum of the wave created by the trapping condition and the incoming wave value @xmath83 , where @xmath84 denotes the first trap boundary reached .
when reaching the second trapping boundary @xmath85 the now outgoing wave will see a matching force @xmath86 leaving an outgoing wave contribution @xmath87 to escape the trapping region @xmath88 . in order to achieve selective radiation , only part of the content of a trapped region
are trapped at the boundary as can be achieved by using a reduced force at the trapping boundary or by selectively omitting certain phases in the trapping force pattern .
marengo and ziolkowski @xcite present ideas very much related to ideas presented here and in berry _
. however , they arrive at a definition of non - radiating ( nr ) sources that is not obviously similar to the traps presented here .
in particular they define nr sources as being non - interacting . while @xcite note that a central property of nr sources is that they store non - trivial field energy , traps described here can not only store , but accumulate and selectively radiate waves .
the difference can be understood by observing that for example berry _
_ assume a simple time - harmonic driver ( * ? ? ?
* eq . ( 3 ) ) throughout their discussion : @xmath89 by our earlier discussion we see that the temporal progression of the boundary has to match the content of the interior domain .
hence once the boundary is defined to be oscillatory the interior of the domain needs to be spatially harmonic as derived in @xcite and has been rederived here .
hence a nr source as noted in literature , with exception of the general orthogonality formulation for time - varying sources given by marengo , devaney and ziolkowski @xcite , can be thought of as a time - oscillatory trap .
the arguments made here use a formalism that is discrete in nature .
however , the discreteness of the arguments are not necessarily restrictive .
the continuous case can be imagined with the discrete time - step made small ( @xmath90 ) or alternatively , discrete pulses can be substituted with narrow distributions of compact support . in neither case
are the results of interest derived here altered . as has already been derived in @xcite
the critical condition for non - propagating waves lie at the boundary of the domain - range that the wave ought not to leave . in the discrete case it is easy to see how this insight can be used and generalized .
in fact , the boundaries of the confining domain need not be static , nor need the condition be used in a two - sided fashion .
in summary , this paper presented constructions of a broad class of non - propagating sources on a string lattice model using trapping conditions .
in particular this includes numerical demonstrations of finite - duration non - propagating excitations , directional excitations , as well as virtually slowed waves .
these examples help explain the extension of non - propagating sources beyond the time - periodic case and include treatment of onset , annihilation and spatial steering .
these properties ought to be observable in experiments well - described by the wave equation .
this equation often arises in problems in acoustics , elasticity , optics and electromagneticsm . and
hence the results presented here apply to these domains of application .
while here i discussed the forward problem , these results also relate to the inverse problem of finding source contributions from the one - dimensional field state as occur for example in acoustical , optical and electromagnetic detection problems .
i am grateful for reprints provided by bruce denardo and greg gbur , who also brought relevant references to my attention .
i m also thankful for helpful comments and pointers to relevant literature by an anonymous referee . | using a lattice string model , a number of peculiar excitation situations related to non - propagating excitations and non - radiating sources are demonstrated .
external fields can be used to trap excitations locally but also lead to the ability to steer such excitations dynamically as long as the steering is slower than the field s wave propagation .
i present explicit constructions of a number of examples , including temporally limited non - propagating excitations , directional excitation and virtually slowed propagation . using these dynamical lattice constructions
i demonstrate that neither persistent temporal oscillation nor static localization are necessary for non - propagating excitations to occur . |
superfluids are peculiar states of matter in which , at the cost of losing part of their individuality , particles gain the ability of cooperative lossless motion @xcite .
the occurrence of superfluidity can be attributed to the formation of certain non - local correlations within the system . for two - dimensional ( 2d ) systems , mermin and
wagner showed that heisenberg models can be neither ferromagnetic or anti - ferromagnetic at finite temperatures @xcite .
hohenberg further ruled out the existence of long - range ordering in 2d bose and fermi systems @xcite .
these rigorous results imply that bose - einstein condensation ( bec ) does not exist at any finite temperature in uniform , interacting 2d systems in the thermodynamic limit , since spontaneous long - range ordering is prevented by long - wavelength fluctuations .
hence one might expect that conventional superfluidity would not occur in 2d systems .
nevertheless , a different path to superfluid behavior is possible in 2d systems . at low temperatures quasi - long - range correlations
may form with an associated power - law decay that eventually reaches zero instead of extending throughout the system .
it was theoretically shown by berezinskii @xcite and by kosterlitz and thouless @xcite ( bkt ) that in 2d a transition to a superfluid state may occur at finite temperatures .
qualitatively the physics of such low - temperature 2d system is conveniently described using the notion of vortex - antivortex pairs ( vaps ) . at temperatures below the bkt transition ,
long - wavelength fluctuations destroy true long - range order and yield spontaneous creation and annihilation of bound vaps at the boundaries of the local domains of the resulting `` quasicondensate '' @xcite .
these locally coherent blocks contribute to the power - law decay of the two - point correlation function resulting in a superfluid response of the system . on increasing the temperature , fluctuations increase and vaps unbind at the bkt transition .
the breaking of vaps results in the proliferation of free vortices and an exponentially decaying correlation length , and hence the system loses its superfluid properties .
this was further quantified by nelson and kosterlitz who predicted a universal jump in the superfluid density at the critical point , which may be used to empirically detect the bkt transition @xcite . in a
trapped ultra - cold bose gas the situation is rather complicated due to the inhomogeneity arising from the confining potential .
bagnato and kleppner showed that a trapped , _
ideal _ bose gas undergoes bec at finite temperatures @xcite , implying that in principle a trapped , _ interacting _ bose gas may exist in a coherent bec phase and/or in a bkt - type phase .
the low temperature structure of a real 2d bose gas in a trap has therefore attracted a fair amount of theoretical discussion in the recent literature and it has been debated whether the superfluid transition in such systems is of bec - type or bkt - type @xcite .
experimentally , the superfluid bkt transition in a bulk was first realized in liquid helium thin films by bishop and reppy @xcite .
et al . _ reported an observation of the transition in superconducting josephson junction arrays @xcite , followed by safonov _
et al . _ who measured a kink in the three - body loss rate in spin - polarized hydrogen @xcite .
these experiments relied on indirect methods of observation where as in the trapped atomic gases vaps and their dynamics can be directly imaged .
the quasi-2d regime in trapped quantum degenerate gases has been approached experimentally by using tight axial confinement with the aid of optical potentials and through centrifugal expansion in rapidly rotated condensates @xcite . however , probing the details of the quasicondensation transition has only recently become experimentally accessible @xcite . indeed , dalibard and co - workers observed phase defects in the interference patterns of multiple quasi-2d gases trapped in the valleys of an optical lattice @xcite .
these phase defects were evidently caused by unbound free vortices @xcite .
further observations on spatial phase correlations in the system provided evidence for the cross - over between the bkt quasicondensate and normal state @xcite . while the observed correlations were shown to be consistent with the bkt quasicondensation picture , the question whether the system is superfluid or not still remained .
recently , an observation of the bkt cross - over has been achieved in a 2d lattice of josephson - coupled becs @xcite and in a single 2d dipole trap @xcite .
the interplay between interactions and inhomogeneous effects arising from the trapping potential have made theoretical predictions for the low temperature phases of dilute , atomic bose gas difficult .
the main point we address in this paper is the question of superfluidity in this system at low temperatures .
there is not a single observable that categorically defines superfluidity and we present results of microscopic calculations for a variety of observables , including off - diagonal long - range order ( condensate ) , fluctuations , scissor mode dynamics and presence of vortices . from the combined analysis of these quantities
we are able to infer a cross - over temperature , @xmath0 , below which the system exhibits superfluidity .
the paper is organized as follows : in sec .
ii we discuss various measurables useful in examining the superfluid properties of our system .
our computational approach is explained in sec .
iii , and the results are presented in sec .
iv followed by the concluding remarks in sec . v.
a substance which has the ability to flow without dissipation is superfluid . although the difference between a superfluid and a classical fluid may seem intuitively clear , it is difficult , if not impossible , to find a single universal definition for superfluidity against which any material could be tested .
indeed , the complete description of superfluidity is not a single feature but a complex of phenomena @xcite . in an interacting 3d atomic gas ,
the formation of bec is essentially equivalent to the emergence of a macroscopic wave function , which inherently exhibits long - range order throughout the system .
furthermore , the system attains finite superfluid fraction at the bec transition .
the situation is more subtle in 2d where the condensation process is plagued by long - wavelength phase fluctuations . in the following ,
we introduce measurables relevant for providing evidence of superfluidity in a quasi - two - dimensional sample of trapped ultra - cold atoms and apply these definitions to discuss the superfluidity of quasicondensates .
quantized vortices are the hallmark of superfluids .
the flow @xmath1 of a superfluid described by a macroscopic wave function with a phase @xmath2 obeys the condition of irrotationality , @xmath3 , and the onsager - feynman quantization of circulation , @xmath4 where @xmath5 is an integer and @xmath6 is the mass of an atom
. therefore rotation ( @xmath7 ) is only possible around the phase singularity at the core of a quantized vortex where the superfluid density vanishes . if normal fluid is present , it occupies the volume in the vortex core .
the quantized vortices can therefore be seen to play a two - fold role in superfluid systems . on the one hand
their coherent role is vital in enabling superfluids to rotate , while on the other hand they can be viewed as ( topological ) defects that are a source of incoherence causing a reduction in the superfluid fraction .
it is worth noting that the observation of a vortex alone is not a sufficient criterion from which superfluidity may be deduced .
for instance , while a persistent vortex in a zero temperature condensate is readily accepted as _ proof of superfluidity _ , in a 2d system near bkt cross - over an observation of a transient isolated vortex would be more likely to signify unbinding of vaps and hence _ loss of superfluidity_. therefore , special care must be taken in the treatment of a situation where both spontaneously ( thermally activated ) and actively ( by external rotation ) created vortices may exist simultaneously .
a particularly vivid example of a classical vortex which has many of the characteristics of a superfluid vortex has been realized in a recent experiment in which the @xmath8 vortex phase winding was imprinted into a non - superfluid cloud of atoms and was observed to persist for extended times due to the cancellation of spins in the diffusion process @xcite .
consider a system described by the usual one - body reduced density matrix @xmath9 where @xmath10 is the second quantized bosonic field operator and the brackets denote quantum mechanical ensemble averaging . if @xmath11 has a macroscopic eigenvalue @xmath12 where @xmath13 is the number of particles in the system , it is said to possess bose - einstein condensation in the state determined by the corresponding eigenvector , also known as the macroscopic condensate wave function .
the concept of long - range order is often encountered in the context of superfluidity . in a homogeneous bose - einstein condensed system ,
true long - range order exists in the sense that @xmath14 where as in the normal state the off - diagonal correlations decay exponentially with the spatial separation @xmath15 where @xmath16 characterizes the length scale over which the correlations decay .
although the existence of bose - einstein condensation does not _ a priori _ imply the system to be superfluid , the existence of off - diagonal order in the system can be considered as a prerequisite for superfluidity . in the case of finite systems , such as trapped atomic gases , eq .
( [ eq3 ] ) generalises by understanding that the boundary of the system is mapped to infinity . two - dimensional systems lacking true long - range order may attain superfluidity through the berezinskii - kosterlitz - thouless mechanism . in this case the correlations decay algebraically with distance @xmath17 and are characterised by the exponent , @xmath18 , where @xmath19 is the 2d superfluid density , and @xmath20 is the thermal de broglie wavelength . for temperatures below the bkt transition temperature ( @xmath21 ) the first order approximation to
the critical exponent is @xmath22 , corresponding to the universal jump at @xmath23 in the superfluid density @xcite .
further insight into the state of the system is obtained by studying the higher - order coherence properties of the system @xcite .
the second - order coherence function @xmath24 yields information about the local coherence of the bose field , @xmath25 , and in general for @xmath26th order coherence @xmath27 for @xmath28 , while for the corresponding thermal state @xmath29 . by decomposing the field into ` coherent ' ( @xmath30 ) and ` incoherent ' ( @xmath31 ) parts , @xmath32 , eq .
( [ eq6 ] ) becomes @xmath33 where @xmath34 . in a similar fashion @xmath35 which essentially measures the probability of three - particle coincidences .
the total three - body recombination rate @xmath36 where @xmath37 is the rate constant and @xmath38 is the total density , is an experimentally measurable quantity yielding overall information of the third order coherence properties of the system . while it has been used experimentally to infer the quasicondensation transition point in helium thin films @xcite , obtaining similar information in trapped 2d gases is more problematic since the local gas density increases as the transition is crossed from normal to quasicondensed state and this partially compensates for the corresponding decrease in @xmath39 .
the moment of inertia , @xmath40 , of a superfluid about a chosen axis is reduced from its classical value , @xmath41 , due to the irrotational motion of superfluid matter .
the temperature dependent superfluid fraction @xmath42 where @xmath43 is the total particle number , is a quantity of special interest .
a finite value of this macroscopic measurable may be used as an evidence of superfluidity .
microscopically , this information about superfluidity is encoded in the elementary excitation spectrum of the system .
the collective scissors mode oscillation has been employed to prove that the occurrence of bec in 3d implies superfluidity @xcite .
essentially , the scissors mode may be viewed as an oscillation of an ellipsoidal cloud of atoms about its semi - axis in the plane . in the collisionless regime ,
a gas in a normal state has two prominent undamped scissors mode eigenfrequencies @xmath44 where @xmath45 and @xmath46 are the planar trapping frequencies . in eq .
( [ eq11 ] ) @xmath47 corresponds to an irrotational quadrupole oscillation and @xmath48 is related to a classical rotational motion .
if superfluid component is present , it oscillates at an additional characteristic frequency @xmath49 whose existence thus provides a clear sign for superfluidity of the system in the collisionless regime .
it is to be noted , however , that in the hydrodynamic limit @xmath48 becomes over - damped and both the remaining thermal mode , @xmath47 , and the superfluid scissors mode , @xmath50 , attain the same value .
in such situation , the damping rate of this mode may in principle be used to reveal superfluid response , although this may prove to be difficult to achieve in practice .
the scissors mode excitations of the system are directly related to the reduced moment of inertia @xcite @xmath51 where @xmath52 is the fourier transform of the time - dependent quadrupole moment @xmath53 .
substitution of eq .
( [ eq13 ] ) in eq .
( [ eq10 ] ) , yields a formula for the superfluid fraction in terms of the scissors mode excitations . we emphasize that the presence of a superfluid scissors mode , @xmath50 , implies non - classical rotational inertia , @xmath54 , finite superfluid fraction , @xmath55 , and hence superfluidity .
we employ the method of classical fields as detailed in refs .
essentially , this amounts to propagating the projected gross - pitaevskii equation @xmath56 in time for the field , @xmath57 , restricted in the subspace determined by an energy cut - off in the harmonic oscillator basis states .
the projector , @xmath58 , serves to constrain the evolution of the field within the subspace of highly occupied states .
each simulation corresponds to the evolution of a single trajectory through the phase space and therefore , in order to construct thermodynamic observables , one should ensemble average over many different but equivalent trajectories
. however , when considering an equilibrium quantities , we may assume the system to be ergodic and replace such ensemble averages by time - averages over the instantaneous field configurations taken from a single trajectory .
the field , @xmath57 , is normalized to the number of particles , @xmath59 , described by the restricted basis .
the total number of particles , @xmath60 , is obtained by using the semiclassical hartree - fock approximation for the @xmath61 above cut - off particles as in refs .
@xcite and as described below .
the in - plane phase function , @xmath62 , of the complex field , @xmath57 , allows for an explicit detection of the locations of vortices and antivortices .
the classical field is completely described by the conserved total energy , @xmath63 , an energy cut - off for the restricted basis , @xmath64 , the dimensionless nonlinearity constant , @xmath65 , and the harmonic trap frequencies , @xmath66 and @xmath67 . here
the spatial length scale is @xmath68 . from these simulations
we can also compute the equilibrium temperature , @xmath69 , and chemical potential , @xmath70 , as described in refs .
@xcite .
an inherent feature in our numerical method requires that the total particle number , @xmath43 , must be computed _
a posteriori_. this is done within the self - consistent hartree - fock approximation by computing the particles not included in the simulated field , @xmath30 , from the semiclassical density .
the form of the semiclassical integral reflects the quasi-2d nature of the trap .
for the temperatures considered here , @xmath71 and therefore several of the lowest axial oscillator states , @xmath72 , contribute significantly to the total number of particles .
however , the temperature is too low for the equipartition theorem to apply and therefore these axial levels need to be treated discretely in the semiclassical integral .
the hartree - fock energy for this system is given by @xmath73 with @xmath74 the kinetic energy .
the thermal densities are given by @xmath75 where @xmath76 and @xmath70 is the chemical potential .
the interaction term in eq .
( [ hfegy ] ) contains the multilevel coupling constant @xmath77 which accounts for the interactions between the particles in different axial levels .
we have denoted @xmath78 and @xmath79 are the normalized harmonic oscillator eigenstates . finally , the number of above cut - off particles @xmath80 is obtained by integrating over the 2d densities and summing the contributions from different axial energy levels . to obtain data points for a fixed number of particles @xmath43 for a range of temperatures
, we estimate the cutoff energy @xmath64 , total number of particles @xmath59 and energy @xmath63 of the classical region of the system according to the prescription of ref .
we then simulate this within the pgpe and calculate the temperature and total number of thermal particles using the procedure described above .
we use this knowledge to make any necessary adjustment to the initial guesses to end up with a target @xmath43 . in principle
we could construct an approximation to the full green s function from our time - dependent classical field simulations allowing us to extract the collective excitation frequencies and their damping rates for the system .
it would be , however , a formidable task .
furthermore , an identification of near - degenerate modes would become cumbersome .
instead , we concentrate on a specific class of excitations
the so - called scissors modes which may be selectively excited and consequently their oscillation frequency can be individually measured from our dynamical simulations . in order to accurately compute the scissors mode collective oscillation frequencies , an ensemble averaging over many equivalent trajectories is required .
therefore we first prepare a large set of initial field configurations by time - sampling a single equilibrium simulation . these instantaneous field configurations
are then rotated 11 degrees in the @xmath81 -plane with respect to the semi - axis of the anisotropic trapping potential . to facilitate lossless rotation of the state
, we must first project the classical field into a larger eigenbasis in order to account for the increase in the energy after rotation due to the anisotropic trapping geometry .
subsequently the seed states thus prepared are propagated in time and the quadrupole moment , @xmath82 , is measured .
the above procedure is repeated for @xmath83 microstates for each temperature point .
the scissors mode frequencies are then obtained from the fourier transformation of the ensemble averaged quadrupole signal @xmath84 , from which also the damping rates of those modes can be estimated . in order to allow practical comparison with future experiments ,
we choose experimentally realistic system parameters which allow us to study the scissors - mode collective oscillation frequencies .
the strength of particle interactions is determined by a constant , @xmath85 , and the harmonic confining potential , @xmath86 , is characterized by the cartesian frequencies @xmath87 hz .
the trap is chosen to be anisotropic in the @xmath81 -plane in order to lift the degeneracy between the quadrupole and scissors modes . however , the planar anisotropy is kept moderate , in contrast to ref .
@xcite , in order to separate the frequency of the superfluid scissors - mode from both of the two classical modes .
we consider @xmath88 @xmath89rb atoms interacting with the @xmath90-wave scattering length @xmath91 nm .
several results we present are generated by performing an average along elliptical trajectories in the 2d plane about the trap centre .
this is done in order to utilize the full information contained in the simulated fields . since our system
does not possess cylindrical symmetry ( in order to facilitate computation of scissors modes ) , we perform this averaging by considering elliptical shells of constant ( 2d ) trap potential energy and average over all spatial points falling on such strips . in what follows , @xmath92 , denotes the distance from origin to such ellipse along the weakly trapped @xmath93-axis of the trap .
in this section we present our numerical results and analysis for the superfluid indicators described above . taking the compendium of results described below , we claim that the quasicondensate studied here is superfluid .
all measurables extracted from our simulations point to a cross - over temperature @xmath0 at which our system attains superfluidity . in order to lay down the qualitative features of the system we have plotted instantaneous 2d classical field densities , @xmath94 , in fig .
[ f1](a ) and ( b ) , and the corresponding phases @xmath95 , fig . [ f1](c ) and ( d ) . the temperatures are @xmath96k for fig .
[ f1](a ) and ( c ) and @xmath97k for fig .
[ f1](b ) and ( d ) . at low temperatures
the density and phase are relatively uniform , while at high temperatures both exhibit strong fluctuation and and vortices and antivortices are nucleated .
this observation is in striking contrast to the usual situation in 3d and highlights the main qualitative difference between 2d and 3d systems . to further quantify the emergence of vortices and antivortices due to the phase fluctuations , we have measured at each temperature point the probablility , @xmath98 , of finding a vortex or antivortex at radius @xmath92
this is done by locating all phase singularities in an instantaneous classical field configuration and averaging over 1000 different microstates .
the classical field area is then divided into ellipsoidal strips of equal width and the vortex occupation probability is obtained by counting the number of phase singularities detected within each strip divided by the number of microstates sampled ( this is the _ radial _ averaging discussed earlier ) .
thus obtained probabilities for a set of temperatures , @xmath99 nk , are plotted as functions of radial distance , @xmath92 , in fig .
the bullets denote the coherence length , measured as @xmath100 radius of @xmath101 ( defined in sec .
c below ) and the red curve is for @xmath102k . there is a sudden jump in the vortex occupation probability at the cross - over . in the superfluid phase there is a vortex - free region at small radii .
vortex pairs are observed in a narrow band near the edge of the coherent region of the system , i.e. the spatial region that has a flat phase in fig [ f1](c ) .
thus the system can be divided in three concentric regions : central coherent and vortex free bec - like region , coherent bkt - like region where vortices are bound , and an incoherent thermal outmost region where vortices are free .
we compute the one - body density matrix , eq .
( [ eq1 ] ) , for the classical field by assuming ergodicity , which allows us to replace the ensemble average by a time average .
the number of condensed particles , @xmath103 , is obtained by computing the largest eigenvalue of the density matrix .
figure [ f3 ] displays the condensate fraction as a function of temperature .
the curve is plotted to provide comparison with the pure 2d ideal gas relation @xmath104 where @xmath105 is the critical temperature calculated for a quasi-2d ideal - gas in our trap geometry containing @xmath106 particles .
the vertical line is the temperature @xmath107 nk , below which this system is superfluid and the markers are the simulation data .
unlike in the recent experiment of krger _ et .
@xcite we find the interactions to cause only a minor shift in the critical temperature from the non - interacting boson result . in terms of the definition based on the eigenvalues of the density matrix , the system may be claimed to show bose - einstein condensation at all temperatures below @xmath0 .
nevertheless it turns out that the system is best described in terms of phase fluctuating quasicondensate apart from the very lowest temperatures .
coherence is an essential feature of superfluidity .
the density matrix , eq .
( [ eq1 ] ) , provides an useful probe for the global coherence between two spatially separated points in the system . in particular
it conveys the knowledge of the correlation length and the information on the possible presence of long - range order .
figure [ f4a ] shows the two - point function , @xmath108 for different temperatures . for low temperatures , @xmath109 , and small radii we witness power - law decay of @xmath101 in accordance with eq .
( [ eq5 ] ) , where as for temperatures , @xmath110 , and/or near the edge of the coherent region , exponential decay is observed .
the qualitative behavior changes at the cross - over temperature , @xmath0 , denoted by the red line . the second - order coherence function , eq .
( [ eq6 ] ) , measures local coherence in the gas .
particularly , for a purely thermal sample @xmath111 and for completely coherent state @xmath112 . in our inhomogeneous system @xmath113 interpolates between these two values as shown in fig .
[ f4b ] where @xmath114 is displayed for different temperatures as functions of the radial distance from the trap centre . at the lowest temperatures , @xmath114 , shows a flat part in accordance with the presence of a nearly phase coherent bec . at large enough radii and for high enough temperatures @xmath114 approaches its thermal value .
the plateau disappears near the temperature where the macroscopic wave function vanishes .
the thicker red line indicates @xmath114 for a result with @xmath115 . in fig .
[ f5 ] , we have plotted the coherence length as function of temperature . the values are obtained by measuring the @xmath100 width of the two - point correlation function .
we have also plotted a function , @xmath116 , where , @xmath117 , ( solid curve ) for @xmath118 and @xmath109 .
the value for the exponent , @xmath26 , may be crudely explained in terms of the thomas - fermi radius , @xmath119 of an isotropic 2d condensate , since @xmath120 varies linearly with the temperature in the vicinity of the cross - over point , @xmath0 .
we have computed the scissors mode frequencies according to the description in the methods section .
the obtained oscillation frequencies and their relative intensities are displayed in fig .
the blue bullets are the mean oscillation frequencies obtained by fitting double gaussian functions to the normalised fourier spectrum at each temperature , which is indicated in gray in the background .
the horizontal dashed red lines are the analytical predictions eqs .
( [ eq11 ] ) and ( [ eq12 ] ) . above the cross - over temperature , @xmath0 , indicated by the red vertical line
, we obtain signal for two thermal modes whose frequencies are found to agree with the predictions of eq .
( [ eq11 ] ) . at temperatures well below the cross - over
, only one scissors mode persists with a frequency corresponding to that of a superfluid , see eq .
( [ eq12 ] ) .
this is the key feature verifying the quasicondensate to be superfluid .
it is interesting to notice that we only observe two different scissors modes at all temperatures .
the upper ` irrotational ' mode simply experiences frequency shift across the cross - over associated with the change in superfluid density of the system .
this is to be contrasted with 3d systems where in general three different scissors modes exist and the superfluid scissors mode experiences downward ( as opposed to the upward shift seen in fig .
[ f7 ] ) frequency - shift on increasing temperature across the cross - over @xcite .
we have also computed the scissors modes for different systems of varying particle interaction strengths which could be experimentally realized using feshbach resonances .
we have verified that in the strongly interacting systems only a single scissors mode survives at all temperatures , making it difficult to distinguish the superfluid and thermal response of the system from one another .
in conclusion , we have studied the problem of superfluid - normal cross - over in a real , trapped quasi-2d bose gas . in such systems the formation of pure bose - einstein condensation is challenged by the long - wavelength phase fluctuations and this fact has made the characterization of such systems difficult both theoretically and experimentally . by performing classical field simulations for these systems ,
we have shown that such quasicondensates are superfluid below the cross - over temperature @xmath0 .
this conclusion is based on observations of the global coherence properties and scissors mode excitations of the system , which constitute the two major results of this article proving superfluidity of quasicondensates . in particular , the emergence of condensate scissors - mode below the cross - over temperature provides an unequivocal microscopic evidence of non - classical rotational inertia and thus the superfluidity of the system .
we have not found signs of fragmentation in terms of the eigenvalues of the density matrix below the cross - over temperature .
a similar conclusion is obtained from the fact that only a single condensate peak is observed in momentum space .
this indicates that despite of the prevailing phase fluctuations , the superfluid state of these systems resembles more closely that of a single - mode trapped bose - einstein condensate than the bulk berezinskii - kosterlitz - thoules superfluid phase .
quasi-2d quantum gases are currently under active experimental investigation .
while experiments have verified aspects of first - order coherence in this system consistent with the berezinskii - kosterlitz - thouless transition , there has yet to be any direct evidence of superfluidity . in this paper
we have reported a series of tests that can be realized in an experiment and should cast light on the formation of superfluidity in these systems .
this research was supported by the marsden fund of new zealand , the university of otago , and the new zealand foundation for research science and technology under the contract nerf - uoox0703 : quantum technologies .
mjd acknowledges the financial support of the australian research council and the queensland state government .
90 a. j. leggett , _ quantum liquids _ , ( oxford university press , oxford , 2006 ) n. d mermin , h. wagner , phys .
* 17 * , 1133 ( 1966 ) .
p. c. hohenberg , phys .
158 * , 383 ( 1967 ) . v. l. berezinskii , sov .
jetp * 32 * , 493 ( 1971 ) ; * 34 * , 610 ( 1972 ) .
j. m. kosterlitz , d. j. thouless , j. phys .
c : solid state phys .
* 6 * , 1181 ( 1973 ) . v. n. popov , functional integrals in quantum field theory and statistical physics , ( reidel , dordrecht , 1983 ) .
d. r. nelson and j. m. kosterlitz , phys .
lett . * 39 * , 1201 ( 1977 ) .
v. bagnato , d. kleppner , phys .
a * 44 * , 7439 ( 1991 ) .
d. s. petrov , m. holzmann , g. v. shlyapnikov , phys .
lett . * 84 * , 2551 ( 2000 ) .
n. prokofev , o. ruebenacker , b. svistunov , phys .
lett . * 87 * , 270402 ( 2001 ) .
n. prokofev and b. svistunov , phys .
a * 66 * , 043608 ( 2002 ) .
j. o. andersen , u. al khawaja , h. t. c. stoof , phys .
lett . * 88 * , 070407 ( 2002 ) . c. gies , b. p. van zyl , s. a. morgan , d. a. w. hutchinson , phys . rev .
a * 69 * , 023616 ( 2004 ) .
a. trombettoni , a. smerzi , p. sodano , new j. phys .
* 7 * , 57 ( 2005 ) .
t. p. simula , m. d. lee , d. a. w. hutchinson , philos .
lett . * 85 * , 395 ( 2005 ) .
t. p. simula and p. b. blakie , phys .
lett . * 96 * , 020404 ( 2006 ) .
a polkovnikov , e. altman , and e demler , _ proc .
usa _ * 103 * , 6125 ( 2006 ) .
m. holzmann , g. baym , j .-
blaizot , f. lalo , proc .
usa * 104 * , 1476 ( 2007 ) .
i. bloch , j. dalibard and w. zwerger , arxiv:0704.3011 , ( 2007 ) .
m. holzmann and w. krauth , arxiv:0710.5060 , ( 2007 ) .
d. j. bishop , j. d. reppy , phys .
40 * , 1727 ( 1978 ) .
d. j. resnick , j. c. garland , j. t. boyd , s. shoemaker , r. s. newrock , phys .
lett . * 47 * , 1542 ( 1981 ) .
a. i. safonov , s. a. vasilyev , i. s. yasnikov , i. i. lukashevich , s. jaakkola , phys .
lett . * 81 * , 4545 ( 1998 ) .
a. grlitz , et al . phys
87 * , 130402 ( 2001 ) .
v. schweikhard , i. coddington , p. engels , v. p. mogendorff , e. a. cornell , phys .
lett . * 92 * , 040404 ( 2004 ) .
d. rychtarik , b. engeser , h .- c .
ngerl , r. grimm , phys .
lett . * 92 * , 173003 ( 2004 ) .
n. l. smith , w. h. heathcote , g. hechenblaikner , e. nugent , c. j. foot , j. phys .
b : at . mol .
phys . * 38 * , 223 ( 2005 ) .
s. stock , z. hadzibabic , b. battelier , m. cheneau , and j. dalibard , phys .
lett . * 95 * , 190403 ( 2005 ) .
z. hadzibabic , z. krger , m. cheneau , b. battelier , and j. dalibard , nature * 441 * , 1118 ( 2006 ) .
p. krger , z. hadzibabic , and j. dalibard , phys .
lett . * 99 * , 040402 ( 2007 ) .
v. schweikhard , s. tung , and e. a. cornell , phys .
lett . * 99 * , 030401 ( 2007 ) .
p. clad , c. ryu , a. ramanathan , k. helmerson , and w. d. phillips , ( private communication ) .
a. j. leggett , rev . mod . phys . * 71 * , 318 ( 1999 ) .
r. pugatch , m. shuker , o. firstenberg , a. ron and n. davidson , phys .
lett * 98 * , 203601 ( 2007 ) .
r. j. glauber , phys .
rev . * 130 * , 2529 ( 1963 ) .
o. m. marag , s. a. hopkins , j. arlt , e. hodby , g. hechenblaikner , and c. j. foot , phys .
lett . * 84 * , 2056 ( 2000 ) .
o. m. marag , g. hechenblaikner , e. hodby , and c. j. foot , phys .
86 * , 3938 ( 2001 ) .
d. guery - odelin and s. stringari , phys .
lett . * 83 * , 4452 ( 1999 ) .
f. zambelli and s. stringari , phys .
a * 63 * , 033602 ( 2001 ) .
m. j. davis , r. j. ballagh and k. burnett , j. phys .
b * 34 * , 4487 ( 2001 )
. m. j. davis , s. a. morgan and k. burnett , phys .
lett . * 87 * , 160402 ( 2001 ) .
m. j. davis , s. a. morgan and k. burnett , phys .
a * 66 * , 053618 ( 2002 ) .
p. b. blakie and m. j. davis , phys .
a * 72 * , 063608 ( 2005 ) .
m. j. davis and p. b. blakie , phys . rev .
* 96 * , 060404 ( 2006 ) . m. j. davis and s. a. morgan , phys .
a * 68 * , 053615 ( 2003 ) .
m. j. davis and p. b. blakie , j. phys .
a * 38 * , 10259 ( 2005 ) .
p. b. blakie and m. j. davis , j. phys .
b * 40 * , 2043 ( 2007 ) . | we investigate the harmonically trapped interacting bose gas in a quasi-2d geometry using the classical field method .
the system exhibits quasi - long - range order and non - classical rotational inertia at temperatures below the berezinskii - kosterlitz - thouless cross - over to the superfluid state .
in particular , we compute the scissors - mode oscillation frequencies and find that the irrotational mode changes its frequency as the temperature is sweeped across the cross - over thus providing microscopic evidence for the emergence of superfluidity . |
investigating many applied problems , we can consider a second - order parabolic equation with mixed derivatives as the basic equation .
an example is diffusion processes in anisotropic media . in desining various approximations for the corresponding boundary - value problems , we focus on the inheritance of the primary properties of the differential problem during the construction of the discrete problem .
locally one - dimensional difference schemes are obtained in a simple enough way for second - order parabolic equations without mixed derivatives @xcite .
mixed derivatives complicate essentially the construction of unconditionally stable schemes of splitting with respect to the spatial variables for parabolic equations with variable derivatives , even for two - dimensional problems . in some problems
, it is convenient to use the fluxes ( derivatives with respect to a spatial direction ) as unknow quantities .
this idea may be implemented in the most simple manner for one - dimensional problems @xcite . to introduce fluxes ,
mixed and hybrid finite elements are applied @xcite . the original parabolic equation with mixed derivatives
may be written as a system of equations for the fluxes .
the basic peculiarity of this system is that the time derivatives for the fluxes in separate equations are interconnected to each other . for the problem in the flux variables ,
unconditionally stable schemes with weights are developed .
locally one - dimensional schemes are proposed for problems without mixed derivatives .
in a bounded domain @xmath0 , the unknown function @xmath1 , @xmath2 , satisfies the equation @xmath3 assume that the coefficients @xmath4 satisfy the conditions @xmath5 for any @xmath6 with constant @xmath7 .
consider the boundary value problem for equation ( [ 2.1 ] ) with homogeneous dirichlet boundary conditions @xmath8 and the initial conditions in the form @xmath9 we introduce a vector quantity @xmath10 ( the index @xmath11 denotes transposition ) such that @xmath12 where @xmath13 is a square matrix @xmath14 ( @xmath15 ) with elements @xmath16 . using this notation , equation ( [ 2.1 ] ) may be written as @xmath17 we can write the above problem ( [ 2.3])([2.5 ] ) in the operator form .
scalar functions are considered in the hilbert space @xmath18 with the scalar product and norm defined by the rules @xmath19 for vector functions , we use the hilbert space @xmath20 , where @xmath21 taking into account ( [ 2.2 ] ) , we can treat the matrix @xmath22 as a linear , bounded , self - adjoint , and positive definite operator in @xmath23 : @xmath24 where @xmath25 is the identity operator in @xmath23
. suppose @xmath26,i.e .
, @xmath27 on the set of functions that satisfy the boundary conditions ( [ 2.3 ] ) , for the gradient and divergence operators , we have @xmath28 it follows from this that @xmath29 , i.e. , @xmath30 in the above notation ( [ 2.7])([2.9 ] ) , from ( [ 2.3])([2.5 ] ) , we obtain the cauchy problem for the system of operator - differential equations @xmath31
@xmath33 for the problem ( [ 2.1])([2.4 ] ) , the following equation corresponds @xmath34 wich is supplemented by the initial condition ( [ 2.12 ] ) .
taking into account that @xmath35 it is possible to eliminate @xmath36 from the system of equations ( [ 2.10 ] ) , ( [ 2.11 ] ) that gives @xmath37 in view of ( [ 2.11 ] ) and ( [ 2.12 ] ) , we put @xmath38 in constructing locally one - dimensional schemes ( schemes based on splitting with respect to spatial directions ) , we focus on the coordinatewise formulation of equations ( [ 2.10 ] ) , ( [ 2.11 ] ) , ( [ 2.14 ] ) ) and ( [ 2.14 ] ) .
let @xmath39 then the basic system of equations ( [ 2.10 ] ) , ( [ 2.11 ] ) takes the form @xmath40 @xmath41 the equation ( [ 2.13 ] ) for @xmath36 is reduced to @xmath42 for the flux components ( see ( [ 2.14 ] ) ) , we obtain @xmath43 the equations of the system ( [ 2.19 ] ) are connected with each other , and , moreover , the time derivatives are interconnected .
the problem ( [ 2.12 ] ) , ( [ 2.18 ] ) seems to be much easier we have a single equation instead of the system of @xmath44 equations
. nevertheless , some possibilities to design locally one - dimensional schemes for the system of equations are still there . here
we present elementary a priori estimates for the solution of the above cauchy problems for operator - differential equations , which will serve us as a checkpoint in the study of discrete problems .
multiplying equation ( [ 2.13 ] ) scalarly in @xmath45 by @xmath36 , we obtain @xmath46 taking into account ( [ 2.7 ] ) and @xmath47 we arrive at @xmath48 this inequality implies the estimate @xmath49 for the solution of the problem [ 2.12 ] ) , ( [ 2.18 ] ) .
now we investigate the problem ( [ 2.14 ] ) , ( [ 2.15 ] ) . by the properties ( [ 2.7 ] ) of the operator @xmath22 , for @xmath50 , we have @xmath51 in view of ( [ 2.21 ] ) , we define the hilbert space @xmath52 , where the scalar product and norm are @xmath53 multiplying equation ( [ 2.15 ] ) scalarly in @xmath23 by @xmath54 , we obtain @xmath55 in view of @xmath56 we arrive at a priori estimate @xmath57 for the solution of the problem ( [ 2.14 ] ) , ( [ 2.15 ] ) .
we conduct a detailed analysis using a model two - dimensional parabolic problem in a rectangle @xmath58 in @xmath0 , we introduce a uniform rectangular grid @xmath59 and let @xmath60 be the set of interior nodes ( @xmath61 ) . on this grid ,
scalar grid functions are given .
for grid functions @xmath62 , we define the hilbert space @xmath63 with the scalar product and norm @xmath64 to determine vector grid functions , we have two main possibilities
. the first approach deals with specifying vector functions on the same grid as it used for scalar functions . the second possibility ,
which is traditionally widely used , e.g. , in computational fluid dynamics , is based on the grid arrangement , where each individual component of a vector quantity is referred to its own mesh . here
we restrict ourselves to the use of the same grid for all quantities , in particular , for setting the coefficients @xmath16 .
consider approximations for the differential operators @xmath65 we apply the standard index - free notation from the theory of difference schemes @xcite for the difference operators : @xmath66 if we set the coefficients of the elliptic operator at the grid points , then @xmath67 more opportunities are available in approximation of operators with mixed derivatives . as the basic discretization @xcite
, we emphasize @xmath68 @xmath69 instead of @xmath70 , we can take their linear combination .
in particular , it is possible @xcite to put @xmath71 in the general case , we set @xmath72 the introduced discrete operators approximate the corresponding differential operators with the second order : @xmath73 where @xmath74 .
we define a grid subset @xmath75 , where the corresponding components of vector quantities are defined .
let @xmath76 @xmath77 @xmath78 @xmath79 and @xmath80 for the grid vector variables , instead of two components , we will use four components , putting @xmath81 for the grid functions defined on grids @xmath82 , we define the hilbert spaces @xmath83 , where @xmath84 for the grid vector functions in @xmath85 , we set @xmath86 now we construct the discrete analogs of differential operators @xmath87 introduced according to ( [ 2.8 ] ) , ( [ 2.9 ] ) . using the above difference derivatives in space , we set @xmath88 so that @xmath89 .
similarly , we define @xmath90 , where @xmath91 thus @xmath92 for the adjoint operator , we have @xmath93 and @xmath94 @xmath95 the above discrete operators approximate the corresponding differential operators with the first order : @xmath96 for the operator - differential equation ( [ 2.13 ] ) , we put into the correspondence the equation @xmath97 where , e.g , @xmath98 . for equation ( [ 3.14 ] ) , we consider the cauchy problem @xmath99 the construction of the operator @xmath100 is associated with the approximations ( [ 3.1])([3.5 ] ) .
the most important properties are self - adjointness and positive definiteness of the operator @xmath100 .
the equation ( [ 3.14 ] ) approximates the differential equation ( [ 2.13 ] ) with the second order .
the system of equations ( [ 2.10 ] ) , ( [ 2.11 ] ) is attributed to the system @xmath101 @xmath102 for the flux problem ( [ 2.14 ] ) , ( [ 2.15 ] ) , we put into the correspondence the problem @xmath103 @xmath104 similarly to ( [ 2.20 ] ) , we prove the following estimate for the solution of the problem ( [ 3.14 ] ) , ( [ 3.15 ] ) : @xmath105 for the estimate ( [ 2.22 ] ) , we put into the correspondence the estimate @xmath106 for the solution of the problem ( [ 3.23 ] ) , ( [ 3.24 ] ) .
we introduce a uniform grid in time with a step @xmath107 and let @xmath108 , @xmath109 . for numerical solving the problem ( [ 3.14 ] ) , ( [ 3.15 ] ) , we apply the standard two - level scheme with weights , where equation ( [ 3.14 ] ) is approximated by the scheme @xmath110 where @xmath111 and , e.g. , @xmath112 . taking into account ( [ 3.15 ] ) , the operator - difference equation ( [ 4.1 ] ) is supplemented with the initial condition @xmath113 the truncation error of the difference scheme ( [ 4.1])([4.3 ] ) is @xmath114 .
the study of the difference scheme is conducted using the general theory of stability ( well - posedness ) for operator - difference schemes @xcite .
let us formulate a typical result on stability of difference schemes with weights for an evolutionary equation of first order . from ( [ 4.4 ] ) , in the standard way
, we get the desired stability estimate @xmath117 which may be treated as a direct discrete analogue of the a priori estimate ( [ 2.20 ] ) for the solution of the differential problem ( [ 2.12 ] ) , ( [ 2.18 ] ) . schemes with weights for a system of semi - discrete equations ( [ 3.21 ] ) , ( [ 3.22 ] ) are constructed in a similar way .
we put @xmath118 @xmath119 the scheme ( [ 4.7 ] ) , ( [ 4.7 ] ) is equivalent to the scheme ( [ 4.1 ] ) . in view of theorem 1 , it is stable under the restriction @xmath115 , and for the solution of difference problem ( [ 4.2 ] ) , ( [ 4.7 ] ) , ( [ 4.7 ] ) , the a priori estimate ( [ 4.4 ] ) holds .
the computational implementation of the unconditionally stable operator - difference schemes ( [ 4.1])([4.3 ] ) for the parabolic equation ( [ 2.1 ] ) with mixed derivatives is based on solving discrete elliptic problems at every time step . for the problem ( [ 3.14 ] ) , ( [ 3.15 ] ) , it seems more convenient to employ additive schemes ( operator - splitting schemes ) that provide the transition to a new time level using simpler problems associated with the inversion of the individual operators @xmath124 rather then their combinations . by the nature of the operators @xmath125 , in this case
, we speak of locally one - dimensional schemes .
the issues of designing unconditionally stable locally one - dimensional schemes for a parabolic equation without mixed derivatives have been studied in detain . for parabolic equations with mixed derivatives ,
locally one - dimensional schemes were constructed in several papers ( see , e.g. , @xcite ) .
strong results on unconditional stability of operator - splitting schemes can be proved only in a uninteresting case with pairwise commutative operators ( the equation with constant coefficients ) . for our problems ( [ 2.1])([2.4 ] )
, the construction of locally one - dimensional schemes requires separate consideration .
let us investigate approaches to constructing locally one - dimensional schemes for the problem ( [ 3.23 ] ) , ( [ 3.24 ] ) .
the computational implementation of the scheme with weights ( [ 4.9 ] ) , ( [ 4.10 ] ) , which is unconditionally stable for @xmath115 , is associated with solving the system of difference equations for four components of the vector @xmath126 .
the equations of this system are strongly coupled to each other , and this interconnection does exist not only for the spatial derivatives ( operators @xmath127 , @xmath128 ) , but also for the time derivatives ( @xmath129 ) .
thus , we need to resolve the problem of splitting for the operator at the time derivative , too
. the simplest case is splitting of the spatial operator without coupling the time derivatives .
such a technique is directly applicable for the construction of locally one - dimensional schemes for parabolic equations without mixed derivatives , where @xmath130 in equation ( [ 2.1 ] ) .
assume that @xmath131 i.e. , @xmath132 is the diagonal part of @xmath133 . for numerical solving the problem ( [ 3.23 ] ) , ( [ 3.24 ] ) , we employ the difference scheme , where only the diagonal part of @xmath133 is shifted to the upper time level . in our notation , we set @xmath134 with the initial conditions according to ( [ 4.10 ] ) .
the scheme ( [ 4.10 ] ) , ( [ 5.2 ] ) has the first - order approximation in time .
it seems more preferable , in terms of accuracy , to apply the scheme that is based on the triangular decomposition of the self - adjoint matrix operator @xmath138 : @xmath139 for the problem ( [ 3.23 ] ) , ( [ 3.24 ] ) , we construct the additive scheme with the splitting ( [ 5.5 ] ) , where @xmath140 the main result is formulated in the following statement .
the alternating triangle operator - difference scheme ( [ 4.10 ] ) , ( [ 5.5])([5.7 ] ) belongs to the class of schemes that are based on a pseudo - time evolution process the solution of the steady - state problem is obtained as a limit of this pseudo - time evolution .
it has the second - order accuracy in time if @xmath142 , and ony the first order for other values of @xmath143 .
hout , k.j . , mishra , c. : stability of the modified craig - sneyd scheme for two - dimensional convection - diffusion equations with mixed derivative term .
mathematics and computers in simulation * 81*(11 ) ( 2011 ) 25402548 | to solve numerically boundary value problems for parabolic equations with mixed derivatives , the construction of difference schemes with prescribed quality faces essential difficulties . in parabolic problems ,
some possibilities are associated with the transition to a new formulation of the problem , where the fluxes ( derivatives with respect to a spatial direction ) are treated as unknown quantities . in this case , the original problem is rewritten in the form of a boundary value problem for the system of equations in the fluxes .
this work deals with studying schemes with weights for parabolic equations written in the flux coordinates .
unconditionally stable flux locally one - dimensional schemes of the first and second order of approximation in time are constructed for parabolic equations without mixed derivatives .
a peculiarity of the system of equations written in flux variables for equations with mixed derivatives is that there do exist coupled terms with time derivatives . |
recent cosmological observations suggest that there is a dark energy component to the energy density of the universe , which should be added to the matter component so as to reach the critical density .
theorists have considered various possibilities for the nature of this dark energy , notably a cosmological constant and quintessence , a dynamical scalar field leading to a time - varying equation of state parameter , @xmath3 .
these models most often involve a single field @xcite or , in some cases , two coupled scalar fields @xcite .
other possibilities for the origin of dark energy include the generalized chaplygin gas proposal @xcite and cardassian models @xcite . in order to unravel the nature of dark energy ,
it is crucial to use observations so as to be able to discriminate among different models . in particular
, the existence of a dark energy component affects the structure of the cmb power spectrum , which is particularly sensitive to the amount of dark energy at different epochs in cosmology .
for instance , the locations of peaks and troughs depend crucially on the amount of dark energy today and at last scattering as well as the dark energy time - averaged equation of state , which are model - dependent quantities @xcite .
hence , one can use the high precision measurements recently obtained by the boomerang @xcite , maxima-1 @xcite , archeops @xcite and , in particular , wmap @xcite observations to constrain dark energy models .
this study has already been performed for the case of a cosmological constant @xcite , the generalized chaplygin gas @xcite and for some of the quintessence models that have been used in the literature , e.g. for the pure exponential potential @xcite , a `` leaping kinetic term '' model @xcite , ratra - peebles potential @xcite , a class of sugra potentials @xcite and cosine - type quintessence @xcite .
the goal of this paper is to study the effect of a dark energy component defined by the quintessence potential proposed in ref .
@xcite v()=e^- [ eq : pot ] , on the location of the first three peaks and the first trough of the cmb power spectrum ( for a first study of cmb anisotropies for this model see ref .
we also analyze the consequences of cluster abundance constraints on @xmath4 , the _ rms _ mass fluctuation on scales of 8 @xmath5 mpc . the m - theory motivated potential of eq .
( [ eq : pot ] ) leads to a model with some interesting features , namely there are two types of attractor solutions giving rise to an accelerating universe today , corresponding to permanent or transient acceleration @xcite .
transient vacuum acceleration is a particularly appealing scenario that would also solve the apparent incompatibility between an eternally accelerating universe and string theory , at least in its present formulation , given that string asymptotic states are inconsistent with spacetimes that exhibit event horizons @xcite . for both types of solutions ,
there is scaling of the densities early in the expansion history , with @xmath6 in the radiation dominated era and @xmath7 in the matter dominated epoch , followed by vacuum domination and accelerated expansion .
transient vacuum dominated solutions arise for @xmath8 , in which case the potential has no local minimum or , for @xmath9 , if the field @xmath10 arrives at the local minimum with enough kinetic energy to roll over the potential barrier and continue rolling down the potential .
permanent vacuum domination occurs for @xmath9 , if @xmath10 gets trapped in the local potential minimum .
our study of the dependence of the first three peaks and first trough locations on the potential parameters in the @xmath11 cosmological parameters space , in view of wmap s results , reveals that , for @xmath1 , the quintessence model of eq .
( [ eq : pot ] ) is favoured as compared to the @xmath0cdm model , provided @xmath12 , which determines the average fraction of dark energy before last scattering , satisfies @xmath13 . for @xmath14 and @xmath13 ,
we find the opposite _ i.e. _ @xmath0cdm is favoured as compared to quintessence .
as @xmath12 increases , @xmath15 , the model s results become comparable to @xmath0cdm s , independently of @xmath16 , as should be expected since the average fraction of dark energy before last scattering becomes negligible .
moreover , the model presents a non - negligible fraction of dark energy at last scattering and during structure formation , typical of early quintessence models , leading to suppressed clustering power on small length scales as suggested by wmap / cmb / large scale structure combined data @xcite .
we consider a spatially - flat friedmann - robertson - walker ( frw ) universe containing a perfect fluid with barotropic equation of state @xmath17 , where @xmath18 is a constant ( @xmath19 for radiation and @xmath20 for dust ) , together with a scalar ( quintessence ) field with potential given by eq .
( [ eq : pot ] ) .
the evolution equations for a spatially - flat frw model with hubble parameter @xmath21 are & = & - 12 ( 43_r + _
m+ ^2 ) , + _ r & = & -4 h_r , + _ m & = & -3 h_m , + & = & -3 h - v , [ eq : fried ] subject to the friedmann constraint h^2 = 13 ( _ r+_m + 12 ^ 2 + v ) , [ friedcon ] we work in units where @xmath22 .
the cmb peaks arise from acoustic oscillations of the primeval plasma just before the universe becomes transparent .
the angular momentum scale of the oscillations is set by the acoustic scale @xmath23 which for a flat universe is given by @xcite [ eq : la ] _ a = , where and @xmath24 is the conformal time , @xmath25 and @xmath26 being , respectively , the conformal time today and at last scattering ; @xmath27 is the average sound speed before decoupling _ s & & _ ls^-1 _ 0^_ls c_s d , + c_s^-2&=&3 + 94_b(t)_(t ) , [ eq : bcs ] with @xmath28 the ratio of baryon to photon energy density . in an idealised model of the primeval plasma
, there is a simple relation between the location of the @xmath29-th peak and the acoustic scale , namely @xmath30 .
however , the location of the peaks is slightly shifted by driving effects and this can be compensated by parameterising the location of the @xmath29-th peak , @xmath31 , as in @xcite _ p_m _ a ( m - _ m ) _ a ( m -|-_m ) , [ eq : lm ] where @xmath32 is the overall peak shift and @xmath33 is the relative shift of the m - th peak relative to the first .
( [ eq : lm ] ) is correct also for the location of troughs if we set , for instance , @xmath34 for the first trough and @xmath35 for the second trough .
although is not in general possible to derive analytically a relationship between the cosmological parameters and the peak shifts , one can use fitting formulae that describe their dependence on these parameters .
doran and lilley @xcite give accurate analytic approximations for the first three peaks and first trough , which can be found in the appendix for convenience .
notice that , as the authors of ref .
@xcite point out , although these formulae were obtained using a standard exponential potential , one expects the results to be fairly independent of the form of the potential ( since the shifts are almost independent of post recombination physics ) unless it is qualitatively very different from the exponential potential before last scattering .
the locations of the first two acoustic peaks and the first trough , from the wmap measurements of the cmb temperature angular power spectrum are @xcite _ p_1 & = & 220.10.8 , + _ p_2 & = & 54610 , + _ d_1 & = & 411.73.5 ; [ eq : wmap ] notice that all uncertainties are 1@xmath36 and include calibration and beam errors .
the location of the third peak is given by boomerang measurements @xcite _ p_3 = 825^+10_-13 .
[ eq : l3 ] we have studied the location of the first three peaks and first trough in the @xmath37 cosmological parameter space for the potential eq .
( [ eq : pot ] ) , for different values of @xmath38 and @xmath12 , slicing through the @xmath39 an @xmath40 planes . for each value of @xmath12 and
a , @xmath41 is chosen such that @xmath42 . given the rather strict bound @xmath43 @xcite ,
we assume hereafter @xmath44 .
our analysis in the @xmath39 plane shows that , for @xmath45 , the strongest constraints arise from the positions of the first and second peaks and , for clarity , we plot only the @xmath46 contours corresponding to the wmap bounds on these quantities , eq .
( [ eq : wmap ] ) , for different values of @xmath12 and @xmath47 ( see fig . [
fig : neq1 ] ) ; also shown are the @xmath46 contours corresponding to the @xmath0cdm model for the same values of the cosmological parameters , where the dashed box represents wmap s bounds on @xmath48 and @xmath49 , namely @xcite _ m h^2=0.140.02 , h=0.720.05 [ eq : wmapb ] the full box corresponds to the bounds obtained on these quantities from a combination of wmap data with other cmb experiments ( acbar and cbi ) , 2dfgrs measurements and lyman @xmath50 forest data @xcite _ m h^2=0.135^+0.008_-0.009 , h=0.71^+0.04_-0.03 .
[ eq : wmapc ] for @xmath51 , the strongest constraints come from wmap s bounds on @xmath52 and @xmath53 , see fig .
[ fig : neq95 ] . finally , the results of our analysis in the @xmath40 plane , with @xmath54 , are shown in fig .
[ fig : h71 ] , where we plot the same contours as for fig .
[ fig : neq95 ] .
we have also studied constraints resulting from @xmath4 , the _ rms _ density fluctuations averaged over @xmath55 spheres .
@xcite gives an estimate of the cmb - normalized @xmath4-value for a very general class of quintessence models from @xmath56 and the @xmath4-value of the @xmath0cdm model with the same amount of dark energy today @xmath57 : [ eq : main ] _
8^_8^ ( a_eq)^ 3 |_^sf/ 5 ( 1-_^0 ) ^-(1 + |w ^-1)/5 . where @xmath58 is the scale factor at matter - radiation equality
, @xmath59 is an average value for the fraction of dark energy during structure formation _
i.e. _ during the matter dominated era , before @xmath60 starts growing rapidly at scale factor @xmath61 _
^sf ^-1 _ a_eq^a_tr _ ( a ) d a .
the effective equation of state @xmath62 is an average value for @xmath63 during the time in which @xmath60 is growing rapidly : ^-1=. in order to compute @xmath4 for the @xmath0cdm model , we use the definition _ 8 ^ 2 _
0^ ^2(k ) ( 3j_1(kr)kr)^2 , [ eq : sigma8 ] with @xmath64 mpc and ^2(k ) = _ h^2 ( kh_0)^3+n t^2(k ) , [ eq : delta ] where @xmath65 is the transfer function describing the processing of the initial fluctuations , for which we use the fitting function @xcite t(q ) & = & + & & ^-1/4 , [ eq : trans ] with @xcite q & = & kh , + & = & _ m h[eq : q ] and @xmath66 is the density perturbation at horizon crossing . a fit to the four - year cobe data gives @xcite 10 ^ 5_h=1.94_m^-0.785-.05_m[eq : delh ] where @xmath67 and @xmath68 ( assuming there are no gravitational waves ) .
taking into account reionization effects , eq .
( [ eq : delh ] ) should be corrected ; we use the fitting formula of ref .
@xcite = 1 + 0.76 -1.96 ^ 2 + 1.46 ^ 3 , [ eq : delhtau ] where @xmath69 is the optical depth .
this formula is reliable up to @xmath70 ; we use @xmath71 , which is within the range of wmap s bound , @xmath72 @xcite .
we compare our results for @xmath4 with large scale structure data .
the recent study of the mass function of 300 clusters at redshifts @xmath73 using early sdss data yields @xmath74 @xcite . other cluster analysis
( using eg .
bcs , reflex and rosat data ) yield different values : @xmath75 @xcite , @xmath76 @xcite , @xmath77 @xcite and @xmath78 @xcite . the results of our study of @xmath4 constraints for the model of eq .
( [ eq : pot ] ) are shown in figs .
1 @xmath79 3 ( dotted curves ) , where we have plotted the largest @xmath4 contours that are compatible both with the large scale structure fits mentioned above and the cmb computation , eqs .
( [ eq : main ] ) - ( [ eq : delhtau ] ) .
[ fig : neq1 ] shows that , as @xmath12 increases , the allowed region becomes smaller and more similar to the @xmath0cdm model results ; in particular , for @xmath80 , the allowed region is only marginally larger than the one for @xmath0cdm model , which is quite small .
notice that a lower bound on @xmath12 already exists from standard big bang nucleosynthesis ( bbn ) and the observed abundances of primordial nuclides , which implies @xmath81 , or , considering a possible underestimation of systematic errors , the more conservative result @xmath82 @xcite ; these bounds require , respectively , @xmath83 and @xmath84 , for the model we are considering . for @xmath51 , fig .
[ fig : neq95 ] shows that the allowed region becomes larger until @xmath85 but does not change significantly for @xmath86 .
again , we find that , as @xmath12 increases , @xmath87 , the model s results become more similar to the ones for the @xmath0cdm model . regarding the results of our analysis in the @xmath40 plane , with @xmath54 , fig .
[ fig : h71 ] , we see that the allowed region becomes smaller and shifts towards smaller values of @xmath16 and @xmath88 as @xmath12 increases , in which case there is again a similarity with the @xmath0cdm model results .
we have also studied the dependence of the first three peaks and first trough locations on parameter a. we find that this dependence is extremely small and can safely be neglected .
we notice that the values of @xmath38 and @xmath12 we have previously considered all correspond to the permanent acceleration regime ; however , as should be expected , changing @xmath38 so as to reach the transient acceleration regime does not alter the analysis since this regime is not significantly different from the one where acceleration is permanent until the present , and therefore peak positions should not be affected . hence , the model s behaviour depends essentially on parameter @xmath12 , which measures the amount of `` early quintessence '' ( using the terminology of ref .
@xcite ) since @xmath89 , where @xmath90 is the average fraction of dark energy before last scattering we conclude that , for @xmath1 , the @xmath0cdm model becomes increasingly disfavoured compared with the quintessence model of eq .
( [ eq : pot ] ) as the amount of early quintessence becomes higher ( @xmath13 ) . for @xmath14 ,
the opposite is true _
i.e _ @xmath0cdm is favoured as compared to quintessence if @xmath13 .
notice that , as @xmath12 increases ( @xmath15 ) , independently of the value of @xmath16 , the model s results become comparable to @xmath0cdm s , as should be expected since @xmath93 decreases .
moreover , we see that quintessence is distinguishible from @xmath0cdm only for @xmath94 and @xmath1 .
finally , we would like to mention that the non - negligible values we obtain for @xmath95 ( e.g. for @xmath91 , we get @xmath96 ) , typical of early quintessence models , will lead , as shown in ref .
@xcite , to suppressed clustering power on small length scales as suggested by wmap / cmb / large scale structure combined data .
the work of t. barreiro is fully financed by the fundao para a cincia e a tecnologia ( fct ) grant ppd/3512/2000 .
bento acknowledges the partial support of fct under the grant pocti/1999/fis/36285 .
the work of a.a .
sen is fully financed by the same grant .
santos is supported by fct grant sfrh / bd/4797/2001 .
g_1 & = & 0.0783 w_b^-0.238 [ 1 + 39.5 w_b^0.763]^-1 , + g_2 & = & 0.56 [ 1 + 21.1 w_b^1.81]^-1 .
[ eq : coef ] the relative shift of the first acoustic peak is zero , @xmath100 , and the relative shifts of the second and third peaks are given by _
2=c_0-c_1r_*-c_2r_*^-c_3 + 0.05(n_s-1 ) , [ eq : delphi2 ] with c_0 & = & -0.1+(0.213 - 0.123|_ls^ ) + & & \{-(52 - 63.6 |_ls^ ) _ b } , + c_1 & = & 0.015 + 0.063(-3500_b^2 ) , + c_2 & = & 610 ^ -6 + 0.137(_b-0.07)^2 , + c_3 & = & 0.8 + 2.3 |_ls^+(70 - 126|_ls^)_b , [ eq : cs ] and j.p .
uzan , , 123510 ( 1999 ) ; t. chiba , 083508 ( 1999 ) ; l. amendola , , 043501 ( 1999 ) ; o. bertolami , p.j .
martins , , 064007 ( 2000 ) ; n. banerjee , d. pavn , , 043504 ( 2001 ) ; 593 ( 2001 ) ; a.a .
sen , s. sen , s. sethi , , 107501 ( 2001 ) ; a.a .
sen , s. sen , , 1303 ( 2001 ) . | we use recent results from the wilkinson microwave anisotropy probe ( wmap ) for the locations of peaks and troughs of the cosmic microwave background ( cmb ) power spectrum , together with constraints from large - scale structure , to study a quintessence model in which the pure exponential potential is modified by a polynomial factor .
we find that the model is compatible with all the recent data for a wide range of cosmological and potential parameters .
moreover , quintessence is favoured compared to @xmath0cdm for @xmath1 and relatively high values of the average fraction of dark energy before last scattering ( `` early quintessence '' ) ; for @xmath2 , quintessence and @xmath0cdm give similar results , except for high values of early quintessence , in which case @xmath0cdm is favoured .
0.5 cm 0.5 cm |
what is the energy source of a macronova ( or kilonova ) ?
macronovae are considered as the emission from the ejecta with mass @xmath2 and the velocity @xmath3 , which are accompanied by the mergers of a ns binary ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
a ns binary merger is one of the most promising sources for the direct detection of the gravitational wave ( gw ) .
the identification of the electromagnetic counterpart to a gw source @xcite would significantly reduce the huge localization error of the gw detectors ( @xmath4 deg@xmath5 ; e.g. , * ? ? ?
* ; * ? ? ?
* ) such as advanced ligo @xcite , advanced virgo @xcite and kagra @xcite .
the strategy of follow - up observations of gw sources should be improved by clarifying the main energy source of macronovae , which determines the behavior of the light curve .
nuclear heating due to the decay of the @xmath0-process elements has been widely discussed as a heating source of macronovae ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
since merger ejecta are neutron - rich , @xmath0-process elements could be synthesized , and the ns binary mergers could be the origin of @xmath0-process elements ( e.g. , * ? ? ? * ) , although iron - rich ejecta are also possible in some cases via the shock and/or neutrino heating ( e.g. , * ? ? ?
the @xmath0-process elements also affect the opacity of the ejecta @xcite , making the long timescale and low temperature of macronovae .
the activities of the central engine are also proposed as alternative energy sources of the macronovae @xcite .
after a ns binary merger , either a bh or a ns is formed as a remnant .
the central remnant releases energy through such as the relativistic jet ( e.g. , * ? ? ?
* ) , disk wind ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and ns wind ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , which may be observed as prompt , extended and plateau emissions in short gamma - ray bursts ( grbs ) ( e.g. , * ? ? ?
* ; * ? ? ?
these outflows can heat the ejecta , which may shine as the macronovae .
similar energy injection is considered in the core - collapse supernovae ( e.g. , * ? ? ?
the energy injection by a highly magnetized ns ( magnetar ) could also produce a brighter transient than the observed macronovae ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , although short grbs do not show any radio signatures of magnetars ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
recently , a macronova candidate is detected as the infrared excess in grb 130603b @xcite . from observations ,
the luminosity , timescale and temperature are @xmath6 erg s@xmath7 , @xmath8 days and @xmath9 k , respectively . in the @xmath0-process model ,
relatively large ejecta mass for the merger of a ns - ns binary , @xmath10 , is required to reproduce the observed infrared excess ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this is important for the equation of state of high - density matter and the progenitor of short grbs ( bh - ns merger or ns - ns merger ) ( e.g. , * ? ? ?
too large ejecta mass in the @xmath0-process model may imply the engine - powered model @xcite . still the previous models using the activities of the extended and plateau emissions also require the large ejecta mass @xmath11 @xcite .
in addition , the previous models also require ejecta with high opacity @xmath12 cm@xmath5 g@xmath7 , which is comparable to the @xmath0-process case , to reproduce the observed long timescale and low temperature .
grb 130603b also shows the long - lasting mysterious x - ray component , which is significantly in excess of the extrapolated power laws based on the optical afterglow @xcite .
the detected luminosity is @xmath14 erg s@xmath7 at the observed time @xmath15 s , which is longer than the timescale of the plateau emission @xcite , and could originate from the central engine such as the accretion disk emission due to the fallback material ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
since the central engine is surrounded by the ejecta , we expect the interaction between the emitted x - ray and the surrounding ejecta .
then the ejecta heated by the irradiation of x - ray may emit infrared photons and reproduce the observed infrared excess according to the ejecta properties .
this is without need for any additional energy source such as the radioactive decay of @xmath0-process elements .
such an x - ray - powered model has a significant advantage that the model uses a single energy source to explain two mysterious signals - x - ray and infrared - which are observed at the same time and with similar luminosities .
this is in contrast to the @xmath0-process and previous engine - powered models which require different unrelated sources for the infrared and x - ray excesses . in this paper
, we consider the x - ray - powered macronovae to explain the observed macronova candidate , grb 130603b . in section [ sec : model ]
we describe our model and constraints on the ejecta properties . in section [ sec : results ] , we present the results for the macronova in grb 130603b .
the other macronova candidates in grb 060614 and grb 080503 are discussed in section [ sec : others ] .
we present the discussion and summary in section [ sec : discussion ] .
, on which the x - rays give the same contribution to the macronova emission . after the jet break time @xmath16 s , the x - ray and infrared fluxes are in comparable excesses of the extrapolated power laws based on the optical emission . , width=264 ]
figure [ figure : lightcurve ] shows the observed x - ray ( 1 kev ; black ) , optical ( r - band ; blue ) and infrared ( h - band ; red ) light curves of grb 130603b @xcite .
the optical light curve shows the jet break at @xmath17 s @xcite , which is supposed to be achromatic .
however , x - ray and infrared fluxes are significantly in excess of the extrapolated power laws after @xmath18 s. remarkably , the x - ray luminosity is similar or slightly larger than the infrared excess known as the macronova emission at the time @xmath19
s. the similar luminosities suggest the same origin for the x - ray and infrared excesses : the infrared macronova could be reproduced if a part of the x - ray excess is converted to the infrared emission , whatever the x - ray excess is .
this x - ray - powered model is naturally expected because the ns binary mergers would eject matter of mass @xmath20 , as shown by general relativistic simulations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and provide a screen to absorb the x - ray and re - emit the infrared emission .
this model explains the observed infrared macronova without the nuclear heating due to the @xmath21-decay and fission of @xmath0-process elements ( e.g. , * ? ? ?
note that the activities of the extended emission ( timescale @xmath22 s ) and plateau emission ( timescale @xmath23 s ) may also contribute energy to the macronova @xcite . in these cases ,
the energy is injected with shorter timescale than that of the macronovae ( @xmath19 s ) and remains trapped with adiabatic cooling ( @xmath24 ) before being released at @xmath19 s. as a result , the required luminosity for explaining the macronova scales as @xmath25 . in figure
[ figure : lightcurve ] , we plot @xmath26 by the green dot - dashed line .
because the x - ray excess component follows @xmath27 @xcite , the later energy injection is more effective for the macronova emission . for the x - ray - powered model to produce the x - ray and infrared excesses detected by observers simultaneously , we consider the following physical setups in figure [ figure : schematic ] .
first , x - rays are generated near the central source and are emitted in nearly isotropic direction .
second , the ejecta lie at radius larger than the x - ray source and cover a fraction of solid angle .
third , the line of sight to the x - ray source is clean for the observers who detect the grb emission .
although our model does not depend on the specific mechanism of x - ray emission , we assume that the x - ray excess at @xmath28 days in grb 130603b originates from the activity near the central engine , such as the accretion disk with super - eddington accretion rate ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) like ultra - luminous x - ray sources . at the early stage of the merger , a relativistic jet is launched from the central engine to penetrate the ejecta @xcite .
because the jet makes a hole in the ejecta , the observers toward the jet axis can directly see inside the ejecta ( see figure [ figure : schematic ] ) .
let us first consider nearly isotropic ejecta with a constant velocity @xmath29 based on the results of the numerical simulations ( e.g. , * ? ? ?
the anisotropy of the ejecta will be discussed in section [ sec : anisotropy ] .
the radius of the ejecta is @xmath30 where @xmath31 is the time after the merger .
since the velocity structure becomes homologous and the ejecta spreads to the radial direction , the typical mass density of the ejecta is given by @xmath32 for the composition of the ejecta , we consider both the iron - rich ejecta and the heavy @xmath0-process ejecta .
since the tidally ejected matter is neutron rich , the heavy @xmath0-process elements may be synthesized ( e.g. , * ? ? ?
on the other hand , the iron - rich ejecta may be dominant if the shocks and/or neutrino irradiation make the electron fraction high ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) .
we consider four conditions to reproduce the infrared excess in our x - ray - powered model .
first , the x - ray photons should be absorbed by the ejecta , i.e. , the optical depth for the x - ray absorption @xmath33 should be larger than unity , @xmath34 second , the absorbed x - ray photons should be thermalized in the ejecta to produce infrared photons .
since the opacities for the iron and @xmath0-process elements are decreasing function of wavelength @xcite , we require the optical depth for infrared photons @xmath35 to be larger than unity , @xmath36 third , the infrared photons should escape from the ejecta to be detected by the observers .
since the optical depth @xmath35 satisfies condition ( [ thermalization_condition ] ) , we consider random walk for the propagation of infrared photons in the ejecta .
the infrared photons can escape from the ejecta if the diffusion timescale @xmath37 is smaller than the dynamical timescale @xmath31 , @xmath38 in the case that the central engine is active longer than the photon diffusion timescale , the light curve of the thermal emission follows the evolution of the central engine , i.e. , the peak time is not necessarily when @xmath39 in the x - ray - powered model .
this is the reason why the x - ray - powered model requires smaller amount of ejecta than the previous models ( see section 3 ) . finally , the temperature of the ejecta @xmath40 should be lower than the observed upper limit @xmath41 k @xcite , @xmath42 which is obtained from the detected infrared flux and the optical upper limit ( blue arrow in figure [ figure : lightcurve ] ) . in following sections [ sec : absorption ] -
[ sec : temperature ] , we consider these four conditions to give constraints on the ejecta properties .
we also consider the luminosity ratio of x - ray and infrared excesses in section [ sec : anisotropy ] . in this subsection ,
we consider condition ( [ absorption_condition ] ) for the x - ray photons to be absorbed in the ejecta .
the absorption optical depth is given by @xmath43 where @xmath44 is the bound - free opacity .
the opacity @xmath44 depends on the ionization state of the ejecta @xcite , @xmath45 where @xmath46 is the average mass number of the composed elements of the ejecta , @xmath47 is the proton mass , @xmath48 is the photoionization cross section and @xmath49 is the neutral fraction of the ejecta .
the ionization state is determined by the balance of the photoionization due to the x - ray emission from the central engine and the recombination .
the neutral fraction @xmath49 is described by the ratio of the absorption rate of the ionizing photons @xmath50 and the recombination rate @xmath51 as @xmath52 the recombination rate is approximately described by @xmath53 where @xmath54 is the number density of the electrons and @xmath55 is the recombination coefficient .
the ionization rate is @xmath56 where the number density of photons whose energy is larger than the ionization threshold energy is , @xmath57 where @xmath58 is the ionization threshold energy and @xmath59 is the fraction of the luminosity above the ionization energy . in the case of iron - rich ejecta ,
the number density of the electrons in the fully - ionized state is @xmath60 , the recombination coefficient for the hydrogen - like iron is @xmath61 @xmath62 s@xmath7 @xcite , the ionization threshold energy of the innermost electronic state is @xmath63 kev , and the photo - ionization cross section at @xmath64 is @xmath65 cm@xmath5 @xcite . using these values and the irradiated x - ray luminosity @xmath66 erg s@xmath7 , kev and regard the excess component as non - thermal emission because the photon index is @xmath67 from observational data at @xmath68 s ( http://www.swift.ac.uk/index.php ) when the excess component dominates the afterglow emission . since the number of photons is smaller in higher energy in the non - thermal emission , we consider that the the photons with higher energy than 9.2 kev do not significantly change the ionization fraction . ]
the ratio @xmath69 is much smaller than unity for the reasonable parameter ranges of the merger ejecta , @xmath70 , @xmath71 and @xmath72 k , at the time @xmath73 days .
then , we use @xmath74 . substituting equations ( [ ejecta_radius ] ) , ( [ ejecta_mass_density ] ) , ( [ absorption_optical_depth ] ) and ( [ bound - free_opacity ] ) with @xmath74 into condition ( [ absorption_condition ] ) , we obtain the constraint on the ejecta velocity @xmath75 as long as the neutral fraction @xmath74 is a good approximation , inequality ( [ absorption_condition_for_velocity ] ) does not depend on the temperature . note that since the photo - ionization cross section for electrons bounded at outer shell is larger than that for the hydrogen - like iron @xcite , lower energy ionizing photons ( @xmath76 kev ) are also absorbed by the ejecta as long as condition ( [ absorption_condition_for_velocity ] )
is satisfied . for the @xmath0-process ejecta
, we assume that most x - ray photons are absorbed by the ejecta within @xmath12 days .
if the x - ray emission is beamed or a limited region of the ejecta is irradiated , the total irradiated luminosity to the ejecta is smaller than that estimated from the observed x - ray luminosity .
note that the ratio @xmath77 is decreasing , so that the neutral fraction remains unity , @xmath74 .
we consider condition ( [ thermalization_condition ] ) for the thermalization in the ejecta .
the optical depth for optical - infrared photons is given by @xmath78 the dominant opacity at the infrared wavelength is bound - bound opacity . in the iron ejecta ,
the opacity for the thermal photons is @xmath79 cm@xmath5 g@xmath7 @xcite .
for simplicity , we neglect the dependence of the opacity @xmath80 on the temperature . from condition ( [ thermalization_condition ] ) and
equations ( [ ejecta_radius ] ) , ( [ ejecta_mass_density ] ) and ( [ infrared_optical_depth ] ) , the upper limit on the ejecta velocity is @xmath81 for the @xmath0-process ejecta , ( @xmath82 cm@xmath5 g@xmath7 ; * ? ? ?
* ; * ? ? ?
* ) , the upper limit on the ejecta velocity becomes large .
thermal photons can escape from the ejecta if condition ( [ escape_condition ] ) is satisfied .
the diffusion timescale @xmath37 for the propagation distance @xmath83 is @xmath84 then , condition ( [ escape_condition ] ) and equations ( [ ejecta_radius ] ) , ( [ ejecta_mass_density ] ) , ( [ infrared_optical_depth ] ) and ( [ diffusion_timescale ] ) give the lower limit on the velocity , @xmath85 since the upper limit on the temperature was obtained from the observations at optical and infrared bands @xcite , the temperature of the ejecta has to satisfy condition ( [ temperature_condition ] ) to reproduce the observations . ) .
] the temperature of the ejecta is determined by the relation @xmath86 where @xmath87 is stefan - boltzmann constant .
then , using condition ( [ temperature_condition ] ) and equation ( [ internal_energy ] ) , the lower limit on the ejecta velocity is @xmath88 since the temperature is determined by the internal energy density due to the irradiation , the lower limit on the velocity ( [ temperature_condition_for_velocity ] ) does not depend on the ejecta mass . in figure
[ figure : lightcurve ] , the x - ray luminosity is approximately an order of magnitude larger than the infrared luminosity at @xmath89 s. the difference between the x - ray and infrared luminosities may come from the collimated x - ray emission due to the geometry of the super - eddington accretion disk ( e.g. , * ? ? ?
note that the ejecta velocity is non - relativistic , so that the reprocessed infrared emission is isotropic .
the anisotropy of the ejecta could also reduce the infrared luminosity compared to the x - ray luminosity . in the bh - ns mergers ,
the dynamical ejecta are highly anisotropic ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
if the ejecta cover @xmath9010% of the solid angle of the x - ray emission , the luminosity of the reprocessed emission from the ejecta is roughly expected to be @xmath91 .
the anisotropy of the ejecta composition could also contribute to the difference between x - ray and infrared luminosities .
numerical simulations including the neutrino transport showed that the electron fraction around the orbital axis is relatively higher than that near the equatorial plane @xcite .
then , iron - rich ejecta may only concentrate on the orbital axis . in the other region ,
@xmath0-process elements may be synthesized and hide the reprocessed photons with the high opacity .
this effect effectively reduces the emission area of the ejecta so that the thermal luminosity becomes lower than the x - ray luminosity .
conversely , the infrared luminosity may be larger than the observed x - ray luminosity because of the limited energy band of the observations .
if a significant energy is emitted as unobserved soft x - rays , the infrared reemission has larger luminosity than the observed x - rays .
in the left panel of figure [ figure : parameter ] , we show the allowed parameter space ( red region ) for the iron ejecta in the x - ray - powered model .
we use @xmath74 , @xmath92 , @xmath93 k , @xmath94 days , @xmath95 cm@xmath5 g@xmath7 and @xmath96 erg s@xmath7 .
the luminosity , temperature and timescale are consistent with the observed values in grb 130603b @xcite . conditions ( [ absorption_condition_for_velocity ] ) , ( [ thermalization_condition_for_velocity ] ) , ( [ escape_condition_for_velocity ] ) and ( [ temperature_condition_for_velocity ] ) are plotted as solid lines .
this panel shows that even if the ejecta are composed of iron , the x - ray - powered model can explain the observed infrared excess by the reasonable range of the parameters ( @xmath97 and @xmath98 ) .
the right panel of figure [ figure : parameter ] shows the case that the ejecta is dominated by the @xmath0-process elements .
the x - ray absorption condition requires such as the recombination rate and photoionization cross section of the @xmath0-process elements . here
, we assume that @xmath0-process elements are partially ionized ( not all f - shell electrons are ionized ) and most x - ray photons are absorbed by the ejecta within @xmath12 days .
then , the bound - bound opacity is significant for the optical - infrared photons .
we use @xmath82 cm@xmath5 g@xmath7 @xcite .
other parameters are the same as in the iron ejecta case in the left panel of figure [ figure : parameter ]
. the right panel of figure [ figure : parameter ] shows that even if the ejecta mass is @xmath99 , the observed infrared excess is reproduced in our x - ray - powered model with the ejecta velocity @xmath100 . in the right panel of figure
[ figure : parameter ] , we also plot the allowed region for the @xmath0-process nuclear heating model in @xcite ( blue region ) .
this panel shows that the allowed region of the parameter space ( red region ) is much larger than that of the @xmath0-process heating model - process heating model . ] . since the @xmath0-process heating model gives the peak luminosity at the time @xmath39 ( e.g. , * ? ? ?
* ) , the allowed region of parameters resides near the line derived from condition ( [ escape_condition ] ) .
the x - ray - powered model ( red region ) has lower ejecta mass than the @xmath0-process model , so that the contribution of the @xmath0-process heating is negligible in the x - ray - powered model . in the both panels of figure [ figure : parameter ] , we plot dashed lines where the kinetic energy of the ejecta @xmath101 equals to the internal energy of the ejecta injected by the irradiation @xmath102 at the time @xmath94 days .
if the kinetic energy of the ejecta before the irradiation is smaller than the internal energy provided by the irradiation ( the grey region in figure [ figure : parameter ] ) , the ejecta would be accelerated and the both energies reach the equipartition , @xmath103 .
this means that the ejecta velocity after the irradiation is determined by the internal energy deposited by the irradiation @xmath102 and the ejecta mass @xmath104 , @xmath105 as long as the velocity is non - relativistic .
we should use this velocity for deriving four conditions in the case @xmath106 .
, most of the regions @xmath103 ( dashed line ) are excluded by the temperature condition ( [ temperature_condition_for_velocity ] ) . ] in the @xmath0-process ejecta , the allowed region with the ejecta mass @xmath107 satisfies @xmath106 . if we consider the ejecta with the mass @xmath108 and the initial velocity @xmath29 which satisfies conditions @xmath109 ( [ thermalization_condition_for_velocity ] ) and @xmath110 ( [ temperature_condition_for_velocity ] ) , the ejecta velocity after the irradiation , equation ( [ ejecta_velocity ] ) , exceeds the upper limit from condition @xmath109 ( [ thermalization_condition_for_velocity ] ) .
therefore , the lower limit on the ejecta mass is @xmath111 in the case of the @xmath0-process ejecta . on the other hand , for the mass range @xmath112 ,
even if the initial ejecta velocity @xmath29 is smaller than the limit from @xmath110 , the irradiation accelerates the velocity to the allowed region .
note that @xmath102 is the lower limit on the injected internal energy because in the early stage the larger energy may be injected , which gives negligible emission at the time @xmath12 days due to the adiabatic cooling @xcite .
the model of x - ray - powered macronovae could also applicable to the macronova candidate following grb 060614 .
as is the case in grb 130603b , grb 060614 has the long - lasting ( @xmath19 s ) x - ray components whose timescale and flux are comparable to those of infrared excess components @xcite .
the luminosity , temperature and timescale are @xmath6 erg s@xmath7 , @xmath113 k and @xmath114 day in the source rest frame , respectively @xcite .
because of relatively long timescale ( @xmath901.7 times longer than that of the macronova in grb 130603b ) , large ejecta mass @xmath115 is required in the @xmath0-process heating model @xcite .
such ejecta mass may be too large for ns - ns mergers as suggested by the simulations ( e.g. , * ? ? ?
thus , @xcite and @xcite suggested a bh - ns merger @xcite for the origin of grb 060614 . on the other hand , if we apply our x - ray - powered model to the observed excesses in grb 060614 , the required minimum mass is @xmath116 for the iron ejecta and @xmath117 for the @xmath0-process ejecta , where the thermalization and photon escape conditions ( [ thermalization_condition_for_velocity ] ) and ( [ escape_condition_for_velocity ] ) give the scaling @xmath118 compared to grb 130603b . the required ejecta velocity @xmath119 )
, the uncertainties of the temperature which come from the assumed afterglow model may reduce the difference . ]
is the same as grb 130603b ( from condition ( [ temperature_condition_for_velocity ] ) ) .
therefore , a ns - ns merger can explain the observed macronova in grb 060614 more naturally in our x - ray - powered model .
the x - ray - powered model may also explain the optical rebrightening at @xmath120 day after the grb 080503 .
the long - lasting x - ray ( @xmath121 day ) has a similar flux to the optical one @xcite .
unfortunately , the redshift of grb 080503 is only limited to @xmath122 @xcite . from the observations @xcite ,
the optical flux and temperature are @xmath123 erg s@xmath7 and @xmath124 k , respectively , which are roughly constant between @xmath125 days in the observer frame .
to explain this event based on the x - ray - powered model , the thermalization condition ( [ thermalization_condition_for_velocity ] ) at @xmath126 days and the escape condition ( [ escape_condition_for_velocity ] ) at @xmath127 day have to be satisfied .
we find that the small mass @xmath128 and the velocity @xmath129 are required in the case of the @xmath0-process dominated ejecta .
the redshift is also restricted to @xmath130 , otherwise the kinetic energy of the ejecta is smaller than the observed radiated energy , which is unphysical . on the other hand , the iron ejecta with the mass @xmath131 and the velocity @xmath129 can also explain the observed optical rebrightening in grb 080503 within the redshift range @xmath132 .
the allowed parameter region for grb 080503 with the iron ejecta slightly overlaps with that for grb 130603b with the @xmath0-process ejecta , but not with the iron ejecta .
if grb 080503 and grb 130603b have the same ejecta mass and velocity , our results imply that the ejecta composition is not isotropic .
namely , some parts of the ejecta are dominated by iron while other parts are dominated by @xmath0-process elements .
this picture is supported by the numerical simulations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
we propose the x - ray - powered macronovae as an alternative interpretation of the observed infrared excess in the short grb 130603b .
our model has important advantages over the @xmath0-process heating model ( e.g. , * ? ? ?
* ) and the previous engine - powered models @xcite . first , the x - ray - powered model explains both the x - ray and infrared excess components with similar fluxes using a single energy source . on the other hand
, the previous models must consider different unrelated sources to explain the excess components .
second , the x - ray - powered model allows for a much broader parameter region of the ejecta properties than that of the @xmath0-process model ( in the right panel of figure [ figure : parameter ] ) .
in particular , the x - ray - powered model explains the macronova with smaller ejecta mass .
the @xmath0-process model must satisfy all the conditions which the x - ray - powered model satisfies , in addition to having the correct amount of @xmath0-process material to generate the observed infrared luminosity by radioactive decay , as well as to having the correct diffusion time at the observed peak time @xmath133 .
then , the allowed parameter range of the ejecta mass has to be much narrow in the @xmath0-process model . in the previous engine - powered models
@xcite , the energy injection occurs at @xmath134 s , earlier than the macronova emission . to confine the injected internal energy in the ejecta up to the macronova timescale @xmath19 s ,
the ejecta satisfy the condition @xmath135 until @xmath19 s. this condition always requires the larger ejecta mass than that of the x - ray - powered model .
we also apply the x - ray - powered model to the other macronova candidates , grb 060614 and grb 080503 .
the @xmath0-process ejecta with the mass @xmath136 and the velocity @xmath119 can explain the optical - infrared excess in grb 060614 . as oppose to the @xmath0-process heating model @xcite , the x - ray - powered model allows relatively small ejecta mass expected for the ns - ns mergers ( e.g. , * ? ? ?
* ) . in grb 080503
, the allowed parameter range does not overlap with that in grb 130603b expect for the case that grb 080503 has the iron ejecta while grb 130603b has the @xmath0-process ejecta .
this may imply ejecta that are in part made of iron and in part made of @xmath0-process elements , as suggested by numerical simulations that find anisotropic electron fraction in the ejecta of ns - ns mergers . in the case that the temperature in the ejecta becomes low ( @xmath137 k ) , dust may form in the ejecta depending on the ejecta composition @xcite .
the dust enhances the opacity of the ejecta .
then , the thermalization occurs for lower mass density of the ejecta than figure [ figure : parameter ] ( condition ( [ thermalization_condition_for_velocity ] ) ) . on the other hand
, photons can escape in the ejecta only with lower mass density ( condition ( [ escape_condition_for_velocity ] ) ) . as a result
, the allowed region moves toward lower ejecta mass than figure [ figure : parameter ] .
@xcite suggested that ejecta is fully ionized and x - ray photons freely escape after @xmath90 1 day for their magnetar model with the x - ray luminosity @xmath138 erg s@xmath7 . in our model ,
the irradiated x - ray luminosity is much smaller than their magnetar model ( @xmath138 erg s@xmath7 ) , so that the ejecta are not fully ionized . as the x - ray excess component
, we consider the activity near the central engine , such as the accretion disk .
using the fallback model with the temporal evolution of the mass accretion rate @xmath139 at @xmath140 and the radiative efficiency @xmath141 , the luminosity from the disk is @xmath6 erg s@xmath142 , where @xmath143 is the total fallback mass ( e.g. , * ? ? ?
note that the ratio of the fallback mass @xmath143 to the ejecta mass @xmath104 is @xmath28 in our model , which is consistent with the numerical simulations @xcite .
if the central engine activity comes from the magnetar , the luminosity is @xmath6 erg s@xmath144 where @xmath145 is the surface dipole magnetic field and the @xmath146 is the rotation period of the ns at 7 days ( e.g. , * ? ? ? * ) .
since the x - ray excess component roughly follows the temporal index @xmath67 , the period @xmath146 has been slowed down at 7 days .
there are important implications of the x - ray - powered macronovae for the search of the electromagnetic counterparts to gw sources .
if we can first observe the x - ray emission originated from the central engine @xcite , the thermal emission from the ejecta could be estimated from the x - ray emission as the luminosity @xmath147 and the temperature @xmath148 as long as the ejecta satisfy the conditions ( [ absorption_condition])@xmath149([escape_condition ] ) .
even if the ejecta initially prevent the detection of x - ray photons from the central engine , the ejecta become optically thin to the x - ray tens of days after the merger .
the luminosity is about @xmath150 erg s@xmath7 , which could be detectable by _
swift_/xrt with integration time @xmath151 s within 100 mpc @xcite .
if the x - ray from the central engine is significantly faint , other heating sources would become dominant such as the @xmath0-process and the shorter timescale activities of the central engine ( extended and plateau emissions ) .
thus , the x - ray observations are important to distinguish the main heating source of the macronovae
. we are grateful to the anonymous referee for useful suggestions .
we would also like to thank k. asano , k. hotokezaka , t. piran , r. sari for fruitful discussions .
this work is supported by sokendai ( the graduate university for advanced studies ) , kakenhi 24103006 ( s.k .
, k.i . ) , 24000004 , 26247042 , 26287051 ( k.i . ) .
was partially supported by an erc starting grant ( grb / sn ) , an isf grant ( 1277/13 ) , an isa grant and an i - core program ( 1829/12 ) .
aasi , j. , abbott , b. p. , abbott , r. , et al . , 2015 , cqgra , 32 , 074001 acernese , f. , agathos , m. , agatsuma , k. , et al .
2015 , cqgra , 32 , 024001 aso , y. , michimura , y. , somiya , k. , ando , m. , miyakawa , o. , sekiguchi , t. , tatsumi , d. , & yamamoto , h. 2013 , phrvd , 88 , 043007 berger , e. 2014 , ara&a , 52 , 43 berger , e. , fong , w. , & chornock , r. 2013 , apjl , 774 , l23 berry , c. p. l. , mandel , i. , middleton , h. , et al .
2015 , apj , 804 , 114 cucchiara , a. , prochaska , j. x. , perley , d. , cenko , s. b. , werk , j. , cardwell , a. , turner , j. , cao , y. , bloom , j. s. , & cobb , b. e. , 2013 , apj , 777 , 94 dai , z. g. , wang , x. y. , wu , x. f. , & zhang , b. 2006 , science , 311 , 1127 de ugarte postigo , a. , thne , c. c. , rowlinson , a. , et al .
2014 , a&a , 563 , 62 essick , r. , vitale , s. , katsavounidis , e. , vedovato , g. , & klimenko , s. 2015 , apj , 800 , 81 fan , y .- z .
, yu , y .- w . , xu , d. , jin , z .- p . , wu , x .- f . , wei , d .- m .
, & zhang , b. 2013 , apjl , 779 , l25 fernndez , r. , kasen , d. , metzger , b. d. , & quataert , e. 2015a , mnras , 446 , 750 fernndez , r. , & metzger , b. d. 2013 , mnras , 435 , 502 fernndez , r. , quataert , e. , schwab , j. , kasen , d. , & rosswog , s. 2015b , mnras , 449 , 390 fong , w. , berger , e. , metzger , b. d. , et al . 2014 , apj , 780 , 118 gao , h. , ding , x. , wu , x .- f , dai , z .- g , & zhang , b. 2015 , apj , 807 , 163 gompertz , b. p. , obrien , p. t. , & wynn , g. a. 2014 , mnras , 438 , 240 gompertz , b. p. , obrien , p. t. , wynn , g. a. , & rowlinson , a. 2013 , mnras , 431 , 1745 grossman , d. , korobkin , o. , rosswog , s. , & piran , t. 2014 , mnras , 439 , 757 hotokezaka , k. , kiuchi , k. , kyutoku , k. , muranushi , t. , sekiguchi , y. , shibata , m. , & taniguchi , k. 2013a , phrvd , 88 , 044026 hotokezaka , k. , kiuchi , k. , kyutoku , k. , okawa , h. , sekiguchi , y. , shibata , m. , & taniguchi , k. 2013b , phrvd , 87 , 024001 hotokezaka , k. , kyutoku , k. , tanaka , m. , kiuchi , k. , sekiguchi , y. , shibata , m. , & wanajo , s. 2013c , apjl , 778 , l16 jin , z .- p . , li , x. , cano , z. , covino , s. , fan , y .- z . , &
wei , d .- m .
2015 , apjl , 811 , l22 kanner , j. , camp , j. , racusin , j. , gehrels , n. , & white , d. 2012 , apj , 759 , 22 kasen , d. , badnell , n. r. , & barnes , j. 2013 , apj , 774 , 25 kasen , d. , & bildsten , l. 2010 , apj , 717 , 245 kawaguchi , k. , kyutoku , k. , nakano , h. , okawa , h. , shibata , m. , & taniguchi , k. 2015 , phrvd , 92 , 024014 kisaka , s. , & ioka , k. 2015 , apjl , 804 , l16 kisaka , s. , ioka , k. , & nakamura , t. 2015 , apjl , 809 , l8 kisaka , s. , ioka , k. , & takami , h. 2015 , apj , 802 , 119 kiuchi , k. , kyutoku , k. , sekiguchi , y. , shibata , m. , & wada , t. 2014 , phrvd , 90 , 041502(r ) kiuchi , k. , kyutoku , k. , & shibata , m. 2012 , phrvd , 86 , 064008 kiuchi , k. , sekiguchi , y. , kyutoku , k. , shibata , m. , taniguchi , k. , & wada , t. 2015 , phrvd , 92 , 064034 kulkarni , s. r. 2005 , astro - ph/0510256 kyutoku , k. , ioka , k. , okawa , h. , shibata , m. , & taniguchi , k. 2015 , phrvd , 92 , 044028 kyutoku , k. , ioka , k. , & shibata , m. 2013 , phrvd , 88 , 041503 lattimer , j. m. , & schramm , d. n. 1974 , apjl , 192 , l145 li , l .- x . , &
paczyski , b. 1998 , apjl , 507 , l59 lippuner , j. , & roberts , l. f. 2015 , arxiv:1508.03133 l , h .- j . , zhang , b. , lei , w .- h .
, li , y. , & lasky , p. d. 2015 , apj , 805 , 89 mangano , v. , holland , s. t. , malesani , d. , et al . 2007 , a&a , 470 , 105 martin , d. , perego , a. , arcones , a. , thielemann , f .- k . , korobkin , o. , & rosswog , s. 2015 , apj , 813 , 2 metzger , b. d. , & berger , e. 2012 , apj , 746 , 48 metzger , b. d. , & bower , g. c. 2014 , mnras , 437 , 1821 metzger , b. d. , martnez - pinedo , g. , darbha , s. , et al .
2010 , mnras , 406 , 2650 metzger , b. d. , & piro , a. l. 2014 , mnras , 439 , 3916 metzger , b. d. , piro , a. l. , & quataert , e. 2009 , mnras , 396 , 304 metzger , b. d. , quataert , e. & thompson , t. a. 2008 , mnras , 385 , 1455 moretti , a. , perri , m. , capalbi , m. , et al .
2007 , proc .
spie , 6688 , 66880 g murguia - berthier , a. , montes , g. , ramirez - ruiz , e. , de colle , f. , & lee , w. h. 2014 , apjl , 788 , l8 nagakura , h. , hotokezaka , k. , sekiguchi , y. , shibata , m. , & ioka , k. 2014 , apjl , 784 , l28 nakar , e. 2007 , phr , 442 , 166 nakar , e. , & piran , t. 2011 , natur .
, 478 , 82 osterbrock , d. e. , & ferland , g. j. 2006 , astrophysics of gaseous nebulae and active galactic nuclei ( 2nd ed . ;
sausalito , ca : university science books ) perley , d. a. , metzger , b. d. , granot , j. , et al .
2009 , apj , 696 , 1871 piran , t. , korobkin , o. , & rosswog , s. 2014 , arxiv:1401.2166 piran , t. , nakar , e. , & rosswog , s. 2013 , mnras , 430 , 2121 proga , d. , & zhang , b. 2006 , mnras , 370 , l61 richers , s. , kasen , d. , oconnor , e. , fernndez , r. , & ott , c. d. 2015 , apj , 813 , 38 rossi , e. m. , & begelman , m. c. 2009 , mnras , 392 , 1451 rosswog , s. 2007 , mnras , 376 , l48 rosswog , s. , korobkin , o. , arcones , a. , thielemann , f .- k . , &
piran , t. 2014 , mnras , 439 , 744 rosswog , s. 2015 , ijmpd , 24 , 1530012 rowlinson , a. , obrien , p. t. , metzger , b. d. , tanvir , n. r. , & levan , a. j. 2013 , mnras , 430 , 1061 sekiguchi , y. , kiuchi , k. , kyutoku , k. , & shibata , m. 2015 , phrvd , 91 , 064059 takahashi , h. r. , & ohsuga , k. 2015 , pasj , 67 , 60 takami , h. , kyutoku , k. , & ioka , k. 2014 , phrvd , 89 , 063006 takami , h. , nozawa , t. , & ioka , k. 2014 , apjl , 789 , l6 tanaka , m. , & hotokezaka , k. 2013 , apj , 775 , 113 tanaka , m. , hotokezaka , k. , kyutoku , k. , wanajo , s. , kiuchi , k. , sekiguchi , y. , & shibata , m. 2014 , apj , 780 , 31 tanvir , n. r. , levan , a. j. , fruchter , a. s. , hjorth , j. , hounsell , r. a. , wiersema , k. , & tunnicliffe , r. l. 2013 , natur .
, 500 , 547 usov , v. v. 1992 , natur . , 357 , 472 wanajo , s. , sekiguchi , y. , nishimura , n. , kiuchi , k. , kyutoku , k. , & shibata , m. 2014 , apjl , 789 , l39 wang , l .- j , dai , z .- g . ,
& yu , y .- w .
2015 , apj , 800 , 79 yang , b. , jin .
z .- p . , li , x. , covino , s. , zheng , x .- z . ,
hotokezaka , k. , fan , y .- z . ,
piran , t. , & wei , d .- m .
2015 , natco , 6 , 7323 yu , y .- w . , zhang , b. , & gao , h. 2013 , apjl , 776 , l40 | a macronova ( or kilonova ) was observed as an infrared excess several days after short gamma - ray burst , grb 130603b .
although the @xmath0-process radioactivity is widely discussed as an energy source , it requires huge mass of ejecta from a neutron star ( ns ) binary merger .
we propose a new model that the x - ray excess gives rise to the simultaneously observed infrared excess via thermal re - emission and explore what constraints this would place on the mass and velocity of the ejecta .
this x - ray - powered model explains both the x - ray and infrared excesses with a single energy source by the central engine like a black hole , and allows for broader parameter region , in particular smaller ejecta mass @xmath1 with iron mixed as suggested by general relativistic simulations for typical ns - ns mergers , than the previous models .
we also discuss the other macronova candidates in grb 060614 and grb 080503 , and implications for the search of electromagnetic counterparts to gravitational waves . |
in many wireless sensor applications , temperature increase caused by sensor operation has to be carefully managed .
for example , wireless sensors implanted in the human body have to be designed such that the temperature due to their operation does not cause any threat for the metabolism .
a line of medical research started by pennes in 1948 @xcite explores the temperature dynamics due to electromagnetic radiation in conjunction with heat losses to the environment and dissipation of heat in the tissue . in the context of sensors that communicate data ,
temperature sensitivity varies depending on the type of tissue . for a given specific tissue
, it is recommended that the temperature does not exceed a critical level , in order to prevent damage to the tissue .
this necessitates careful scheduling of data transmission @xcite .
this problem arises in various types of body area sensor networks , see e.g. , @xcite and references therein .
finally , temperature increase in a sensor is a threat for the proper operation of the hardware itself @xcite . in this context , the electric power that feeds the amplifier circuitry has to be carefully scheduled so as to avoid permanent damage in the circuit . in order to obtain design principles with regard to temperature sensitivity of such systems , determining transmission schemes under a safe temperature threshold @xmath0
is a useful objective . in this paper
, we consider data transmission with energy harvesting sensors under such temperature constraints .
data transmission with energy harvesting transmitters has been the topic of recent research @xcite . in particular , throughput maximization under offline and online knowledge of the energy arrivals
is considered in these references for single - user and multi - user energy harvesting communication systems . in @xcite
, this problem is investigated under imperfections such as battery energy leakage , charge / discharge inefficiency , and presence of processing costs . in the current paper , we aim to bridge physical heat dissipation with data transmission in energy harvesting communication systems . when the sole purpose is to maximize the throughput , the transmitter may generate excessive heat while utilizing the energy resource . in a temperature sensitive
application , the heat accumulation caused by the transmission power policy has to be explicitly taken into account .
in such a case , heat generated in the transmitter circuitry causes a form of information - friction " @xcite .
we study the effect of this friction " in a deadline constrained communication of an energy harvesting transmitter over an awgn channel . for simplicity
, we use transmit power as a proxy for hardware power .
that is , we assume that the energy dissipated by the power amplifier dominates other energy sinks in the circuitry . more work is needed to understand full implications of communication circuitry s energy in this context .
our formulation also relates to @xcite in that the cumulative effect of heat generated in the hardware affects the communication performance .
we determine the throughput optimal offline power scheduling policy under energy harvesting and temperature constraints .
our thermal model is based on a view of the transmitter s circuitry as a linear heat system where transmit power is an input as in @xcite , @xcite , @xcite .
we impose that the temperature does not exceed a critical level @xmath0 .
consequently , we obtain a convex optimization problem .
we solve this problem using a lagrangian framework and kkt optimality conditions .
we first derive the structural properties of the solution for the general case of multiple energy arrivals .
then , we obtain closed form solutions under a single energy arrival . for the general case ,
we observe that the optimal power policy may make jumps at the energy arrival instants , generalizing the optimal policies in @xcite . between energy harvests ,
the optimal power is monotonically decreasing .
we establish for the case of a single energy arrival that the optimal power policy monotonically decreases , corresponding temperature monotonically increases , and both remain constant when the critical temperature is reached .
then , we consider the case of multiple energy arrivals .
we observe that the properties of the solution for the single energy arrival case are guaranteed to hold only in the last epoch of the multiple energy arrival case . in the remaining epochs
, the temperature may not be monotone and the transmitter may need to cool down to create a temperature margin for the future , if the energy harvested in the future is large .
we illustrate possible cases and obtain insights regarding the optimal temperature pattern in the multiple energy arrival case .
we consider an energy harvesting transmitter node placed in an environment as depicted in fig .
[ model ] . the node harvests energy to run its circuitry and wirelessly
send data to a receiver .
the received signal @xmath1 , the input @xmath2 , fading level @xmath3 and noise @xmath4 are related as @xmath5 where @xmath4 is additive white gaussian noise with zero - mean and unit - variance . in this paper ,
the channel is non - fading , i.e. , @xmath6 .
we use a continuous time model : a scheduling interval has a short duration with respect to the duration of transmission and we approximate it as @xmath7 $ ] where @xmath8 denotes infinitesimal time . in @xmath7
$ ] , the transmitter decides a feasible transmit power level @xmath9 and @xmath10 bits are sent to the receiver , where the base of @xmath11 is @xmath12 . to be precise
, the underlying physical signaling is in discrete time and the scalings in snr and rate due to bandwidth and the base of the logarithm are inconsequential for the analysis . . ] as shown in fig .
[ model2 ] , the initial energy available in the battery at time zero is @xmath13 .
energy arrivals occur at times @xmath14 in amounts @xmath15 with @xmath16 .
we call the time interval between two consecutive energy arrivals an _ epoch_. @xmath17 is the deadline . @xmath18 and @xmath19 are known offline and are not affected by the heat due to transmission .
let @xmath20 and @xmath21 be the number of energy arrivals in the interval @xmath22 and by convention we let @xmath23 .
power scheduling policy @xmath9 is subject to energy causality constraints as : @xmath24\end{aligned}\ ] ] in our thermal model , we use the transmit power as a measure of heat dissipated to the environment . in particular , we model the temperature dynamics of the system as follows : @xmath25 where @xmath9 is the transmit power policy and @xmath26 is the temperature at time @xmath27 .
@xmath28 is the constant temperature of the environment that is not affected by the heating effect due to the transmit power level @xmath9 .
@xmath29 and @xmath30 are non - negative constants .
@xmath31 represents the cumulative effect of additional heat sources and sinks and it can take both positive and negative values . in the following ,
we consider the case of no extra heat source or sink , i.e. , @xmath32 . our thermal model in ( [ thermal ] )
is intimately related to the thermal model in @xcite where hardware heating is modeled as a first order @xmath33 heat circuit . in particular , thermal dynamics of a power controlled transmitter due to its amplifier power consumption ( see e.g. , @xcite ) could be modeled as in ( [ thermal ] ) .
we also refer the reader to @xcite for a related heating model . our thermal model is also related to the well - known pennes bioheat equation @xcite .
we assume , for simplicity , that the spatial variation in temperature is not significant and leave the general case of spatial temperature variations as future work
. becomes available for data transmission at time @xmath19 .
@xmath17 is the deadline . ] from ( [ thermal ] ) , the solution of @xmath26 for any given @xmath9 with the initial condition @xmath34 at time @xmath35 is : @xmath36 by inserting @xmath37 in ( [ four ] ) , we get ( c.f .
* eq . ( 3 ) ) ) : @xmath38
the temperature should remain below a critical temperature @xmath0 , i.e. , @xmath39 , where we assume that @xmath40 .
let us define @xmath41 , which is the largest allowed temperature deviation from the environment temperature .
typically , initial temperature is @xmath28 , i.e. , initially the temperature is stabilized at the constant environment temperature @xmath28 . from ( [ ths ] ) , using @xmath42 and @xmath43 , we get the following equivalent condition for the temperature constraint : @xmath44\end{aligned}\ ] ] note that the temperature constraints in ( [ gg ] ) and the energy causality constraints in ( [ causality ] ) do not interact . due to the heat generation dynamics governed by ( [ thermal ] )
, we observe in ( [ gg ] ) that the cost of power increases exponentially in time ( i.e. , the multiplier in front of @xmath45 is exponential in @xmath46 ) while the heat budget also increases exponentially in time ( i.e. , the upper bound on the right hand side of ( [ gg ] ) is exponential in @xmath27 ) .
offline throughput maximization problem over the interval @xmath47 $ ] under energy causality and temperature constraints with initial temperature @xmath43 is : @xmath48 } \quad & \int_0^d \frac{1}{2}\log\left(1 + p(\tau)\right)d\tau \\ \nonumber \mbox{s.t . }
\quad & \int_0^{t } a e^{b \tau } p(\tau ) d \tau \leq t_{\delta}e^{bt } , \quad \forall t \\ \qquad & \int_{0}^{t } p(\tau ) d\tau \leq \sum_{i=0}^{h(t ) } e_i ,
\quad \forall t \label{opt_prob1}\end{aligned}\ ] ] where the space of actions is the set of measurable functions @xmath9 defined over the interval @xmath47 $ ] . note that ( [ opt_prob1 ] ) is a convex functional optimization problem .
the lagrangian for ( [ opt_prob1 ] ) is : @xmath49 taking the derivative of the lagrangian with respect to @xmath9 and equating to zero : @xmath50 which gives @xmath51^+\end{aligned}\ ] ] in addition , the complementary slackness conditions are : @xmath52 in ( [ rf ] ) and ( [ slck1])-([slck2 ] ) , @xmath53 and @xmath54 are distributions that are allowed to have impulses and their total measure over @xmath47 $ ] interval are not both zero , i.e. , @xmath55 or @xmath56 , in order to prohibit @xmath9 from being unbounded .
we note that ( [ rf ] ) and ( [ slck1])-([slck2 ] ) are necessary and sufficient conditions since the problem is convex .
the solution is unique almost everywhere as the objective function is strictly concave .
we note that the problem in ( [ opt_prob1 ] ) could be solved by using calculus of variations .
see @xcite for application of calculus of variations for a similar problem to ( [ opt_prob1 ] ) . as another alternative
, we note that ( [ opt_prob1 ] ) could equivalently be solved by using a hamiltonian approach from optimal control theory .
in particular , we can cast the problem in ( [ opt_prob1 ] ) as an optimal control problem with pure state constraints @xcite . in this case , the state of the system is the tuple @xmath57 $ ] where @xmath58 is the total energy expenditure by the time @xmath27 .
the input is @xmath9 for @xmath59 .
this problem is in the following form : @xmath48 } \quad & \int_0^d \frac{1}{2}\log\left(1 + p(\tau)\right)d\tau \\ \nonumber \mbox{s.t . }
\quad & \frac{d}{dt}t(t ) = f_1(t , b , p ) , \
\frac{d}{dt}b(t)= f_2(t , b , p ) \\ \qquad & g_1(t , b , t ) \leq 0 , \
g_2(t , b , t ) \leq 0 \label{opt_prob2}\end{aligned}\ ] ] where @xmath60 and @xmath61 while @xmath62 and @xmath63 .
note that @xmath64 and @xmath65 do not depend on the input @xmath66 . with these selections ,
optimization problem ( [ opt_prob2 ] ) is in the same form as that stated in ( * ? ? ?
( 2.1)-(2.6 ) ) . in this case
, hamiltonian is @xmath67 and the corresponding lagrangian is @xmath68 where @xmath69 and @xmath70 are the co - state trajectories ; @xmath71 and @xmath72 are multiplier functions .
we note that pontryagin s maximum principle is necessary and sufficient in this case since ( [ opt_prob2 ] ) is a concave maximization problem .
one can derive the equivalence of necessary and sufficient conditions for this optimal control problem to those in ( [ rf ] ) and ( [ slck1])-([slck2 ] ) . in the following ,
we proceed with the lagrangian formulation in ( [ lgrng ] ) and the corresponding optimality conditions in ( [ rf ] ) and ( [ slck1])-([slck2 ] ) .
in this section , we obtain the structural properties of the optimal power scheduling policy using the optimality conditions . in the following lemmas , @xmath9 refers to the optimal policy and @xmath26 is the resulting temperature unless otherwise stated .
we first note that the temperature level never drops below @xmath28 .
in particular , if the initial temperature is between @xmath28 and @xmath0 , the temperature at all times will remain between @xmath28 and @xmath0 .
[ temp ] @xmath73 whenever the initial temperature is @xmath74 . from ( [ thermal ] ) , since @xmath75 we have @xmath76 whenever @xmath77 . the constraint
@xmath39 is satisfied by any feasible policy in ( [ opt_prob1 ] ) .
the following lemma states that if the temperature @xmath26 is constant , then the power @xmath9 is constant also ( while it is not true the other way around , see lemma [ piecewise ] ) , and that if the temperature hits the maximum allowed level @xmath0 , then the power must be below a threshold . [ constant
] whenever @xmath26 is constant over an interval @xmath78 $ ] , @xmath9 is also constant over that interval .
if the temperature hits the level @xmath0 at @xmath79 , then @xmath80 for all sufficiently small @xmath81 .
if @xmath26 is constant in @xmath82 , @xmath83 and from ( [ thermal ] ) , @xmath9 is also constant in the same interval . if @xmath84 for some @xmath85 , then @xmath86 and from ( [ thermal ] ) , @xmath87 .
the following lemma shows that if the power @xmath9 is a monotone increasing function , then so is the temperature @xmath26 .
we first prove this result for piecewise constant functions and then generalize it to arbitrary functions .
we note that a particular instance of a monotone increasing piecewise constant power is observed in the solution of the throughput maximization problem without temperature constraints @xcite .
[ piecewise ] if @xmath9 is a monotone increasing piecewise constant function , then @xmath26 is monotone increasing .
more generally , if @xmath9 is a monotone increasing function , so is @xmath26 .
we first prove the first statement of the lemma which is concerned with piecewise constant functions .
let us start with the case of a single constant power value for the entire duration of communication , i.e. , @xmath88 for @xmath89 $ ] . from ( [ ths ] ) , we have : @xmath90 for @xmath43 , ( [ fin ] ) is a monotone increasing function of @xmath27 . in particular , @xmath91 .
now , let us consider the case of @xmath92 constant power levels for the duration of communication , i.e. , @xmath93 over the interval @xmath94 where @xmath95 for all @xmath96 and @xmath97 where @xmath98 is the number of intervals .
in this case , we have for @xmath99 : @xmath100 where @xmath101 .
hence , the coefficient of @xmath102 in ( [ fin2 ] ) has a negative sign as @xmath103 .
this proves that @xmath26 is monotone increasing . to generalize this result for any monotone increasing function @xmath9
, we obtain any monotone increasing simple approximation @xcite of @xmath9 , denoted as @xmath104 , such that @xmath105 for all @xmath89 $ ] and @xmath106 pointwise .
for example , one can select @xmath107 for @xmath108 and @xmath109 for @xmath110 .
let us call the resulting temperature @xmath111 .
hence , @xmath112 for all @xmath89 $ ] and @xmath113 pointwise . by monotone convergence theorem @xcite
, we have @xmath114\end{aligned}\ ] ] accordingly , @xmath115 pointwise and we have @xmath116\end{aligned}\ ] ] since @xmath104 is a monotone increasing piecewise constant function , from the first part of the proof , @xmath111 is monotone increasing , i.e. , @xmath117 . since @xmath118 pointwise , this implies @xmath119 , i.e. , @xmath26 is monotone increasing as well .
the next lemma shows that if the temperature remains constant over an interval , then that level could only be @xmath28 or @xmath0 , i.e. , any other temperature can not be a stable temperature .
[ const_temp ] if @xmath26 is constant over an interval @xmath78 $ ] , then that constant level could only be @xmath28 or @xmath0 .
assume @xmath26 is constant over @xmath82 . without loss of generality , assume that there is no energy arrival in the interval @xmath82 , and otherwise let @xmath82 be the portion of the interval without any energy arrivals . by lemma [ constant ] , @xmath9 is constant over @xmath82 .
if @xmath120 over @xmath82 , then @xmath77 from ( [ thermal ] ) .
if @xmath121 , we have from ( [ nn ] ) @xmath122 where @xmath123 over the interval @xmath82 by ( [ slck2 ] ) since @xmath124 implies energy constraint is tight and @xmath120 .
therefore , @xmath125 is constant over @xmath82 . if @xmath126 , then by ( [ slck1 ] ) , @xmath127 over @xmath82 and hence @xmath128 is constant over @xmath82 .
however , this makes ( [ nn2 ] ) a time varying function of @xmath27 because of the @xmath129 term in the denominator , and this contradicts the fact that @xmath9 is constant . finally , if @xmath130 , this means that the temperature constraint is never tight .
in this case , the piecewise constant power policy in @xcite is optimum , and the temperature is monotonically increasing from lemma [ piecewise ] , and therefore , can not be a constant over an interval .
the following lemma states that at the end of the communication session either the harvested energy is exhausted or the critical temperature is reached .
[ lm12 ] at @xmath131 , either the temperature constraint or the energy causality constraint or both are tight .
if neither of the constraints are tight , then the power policy @xmath9 could be increased over a set of non - zero lebesgue measure in the last epoch .
this strictly increases the throughput , contradicting the optimality .
the following lemma shows that the optimal power should be monotonically decreasing between energy harvests .
[ new_lm ] @xmath9 is piecewise monotone decreasing except possibly at the energy arrival instants .
in particular , it is monotone decreasing between consecutive energy harvests .
we prove the statement by contradiction .
assume that for some interval @xmath132 $ ] , @xmath9 is strictly monotone increasing , and that the interval @xmath132 $ ] does not contain an energy arrival instant .
define a new power policy as @xmath133 over @xmath134 $ ] and @xmath135 otherwise .
@xmath136 satisfies the energy causality constraint in ( [ opt_prob1 ] ) since @xmath136 uses the same amount of energy as @xmath9 over @xmath132 $ ] and the energy constraint for @xmath9 is not tight in this interval .
@xmath136 also satisfies the temperature constraint . to see this
, we first note that @xmath136 satisfies the following inequality ( see ( * ? ? ?
* theorem on p. 207 ) ) : @xmath137 as both @xmath9 and @xmath129 are monotone increasing .
in addition , since @xmath9 is temperature feasible : @xmath138 combining ( [ cnt1 ] ) and ( [ cnt3 ] ) , we conclude that @xmath136 satisfies the temperature constraint at @xmath139 : @xmath140 additionally , the temperature constraint is satisfied for @xmath141 since @xmath136 and @xmath9 are identical for @xmath141 and @xmath9 is temperature feasible .
hence , we need to show that @xmath136 satisfies the temperature constraint for all @xmath142 to establish the temperature feasibility of @xmath136 .
that is , we need to show : @xmath143 since @xmath144 is constant over @xmath132 $ ] , we have : @xmath145\end{aligned}\ ] ] using ( [ corin ] ) in ( [ gggg ] ) and since @xmath146 , ( [ gggg ] ) takes the following equivalent form : @xmath147 note that the left hand side of ( [ ffgg ] ) is either monotone increasing or monotone decreasing in @xmath27 as it is a linear function of @xmath148 .
since the inequality ( [ ffgg ] ) holds at @xmath149 and @xmath139 as @xmath136 satisfies the temperature constraint at those points , we conclude that @xmath136 satisfies the temperature constraint for all @xmath134 $ ] .
in addition , @xmath136 yields higher throughput than @xmath9 due to the concavity of logarithm .
this contradicts the optimality of @xmath9 .
the proof holds even when @xmath132 $ ] includes an energy arrival instant provided that the energy causality constraint is not tight at that instant .
next , we show that discontinuities in the power level could only occur in the form of positive jumps , and only at the instances of energy harvests .
[ lm1 ] if there is a discontinuity in @xmath9 , it is a positive jump and it occurs only at the energy arrival instants .
the temperature @xmath26 is continuous throughout the @xmath47 $ ] interval .
since @xmath129 is a continuous function of @xmath27 , @xmath53 and @xmath150 , any jump in @xmath9 has to be positive due to ( [ nn ] ) .
any positive jump at instants other than @xmath151 violates monotonicity of @xmath9 within each epoch due to lemma [ new_lm ] . due to
( [ ths ] ) , the resulting temperature @xmath26 is continuous throughout the @xmath47 $ ] interval . by lemma [ lm1 ] , we can take @xmath152 in the form @xmath153 without loss of optimality , where @xmath154 , @xmath155 , are finitely many lagrange multipliers corresponding to the energy causality constraints at the energy harvesting instants @xmath156 and the deadline , @xmath23
. the next lemma shows , for an arbitrary feasible policy @xmath9 , that if the temperature reaches the critical level @xmath0 at some @xmath157 , then the power just before @xmath157 must be larger than a threshold .
[ lm2 ] if @xmath84 for some @xmath85 , then @xmath158 for all sufficiently small @xmath81 .
since @xmath84 , we have : @xmath159 we combine ( [ gg ] ) with ( [ kk ] ) to get @xmath160\end{aligned}\ ] ] which implies in view of the continuity of @xmath9 ( except for the finitely many energy arrival instants ) proved in lemma [ lm1 ] that @xmath158 for all sufficiently small @xmath81 .
we next state the continuity of the optimal power policy @xmath9 at points when it hits the critical temperature @xmath0 .
[ kkk ] if @xmath84 for some @xmath85 then @xmath9 is continuous at @xmath157 and @xmath161 .
the proof follows from lemma [ constant ] and lemma [ lm2 ] and the fact that negative jumps in @xmath9 are not allowed due to lemma [ lm1 ] .
next , we show that when the temperature hits the boundary @xmath0 , it has to return to @xmath0 .
[ extend ] whenever @xmath84 for some @xmath162 , there exists @xmath163 such that @xmath164 .
assume that @xmath84 for some @xmath162 and @xmath126 for all @xmath165 .
by lemma [ kkk ] , @xmath166 . from ( [ four ] ) with @xmath84 , the constraint @xmath39 becomes : @xmath167 since @xmath126 in @xmath165 , only energy causality constraint is active and thus @xmath9 for @xmath165 is the piecewise constant monotone power allocation in @xcite .
on the other hand , @xmath168 satisfies ( [ reef_r2 ] ) with equality for all @xmath27 . therefore , we must have @xmath169 for all @xmath170 for some @xmath171 . however , this contradicts @xmath166 since there can not be a negative jump in @xmath9 by lemma [ lm1 ] .
the following lemma identifies the exact conditions where the power @xmath9 makes a jump .
if there is a jump in @xmath9 , it occurs only at an energy arrival instant , when the battery is empty and the temperature is strictly below @xmath0 . due to the slackness conditions in ( [ slck1])-([slck2 ] ) , a jump occurs
if either the battery is empty or the temperature constraint is tight , i.e. , @xmath164 . by lemma [ kkk ] , @xmath9 is continuous whenever @xmath164 .
therefore , a jump in @xmath9 occurs at an energy arrival instant , when the battery is empty and @xmath126 .
we finally remark that energy may have to be wasted as aggressive use of energy may cause temperature to rise above the critical level .
in this section , we consider a single epoch where @xmath172 units of energy is available at the transmitter at the beginning .
we first develop further structural properties for the optimal power control policy in this specific case and then obtain the solution .
the next lemma shows that , if the power falls below a certain threshold at an intermediate point and remains under that threshold until the deadline , then it should remain constant throughout .
[ lmn ] if @xmath173 for @xmath174 $ ] , then @xmath9 is constant over @xmath175 $ ] .
assume @xmath9 is not constant over @xmath175 $ ] .
let @xmath176 .
define a new policy @xmath177 for @xmath178 $ ] and @xmath135 otherwise .
@xmath136 is both energy and temperature feasible .
energy feasibility holds by construction as @xmath179 and @xmath66 have the same energy over @xmath180 $ ] .
temperature feasibility also holds : @xmath181 since @xmath9 is temperature feasible and as @xmath182 , we have @xmath42 for all @xmath183 from ( [ gg ] ) .
now , by jensen s inequality @xmath136 achieves strictly larger throughput since @xmath11 is strictly concave .
this contradicts the optimality of @xmath9 .
hence , @xmath184 for @xmath178 $ ] .
the following lemma states that the power has to remain constant at the level @xmath185 when the temperature reaches the critical level @xmath0 .
[ multiple_r ] let @xmath186 $ ] denote @xmath187 : t(t)=t_c\}$ ] .
if @xmath188 , then @xmath189 for all @xmath190 $ ] . by lemma [ kkk ] , @xmath191 . by lemma [ new_lm ] ,
@xmath9 is monotone decreasing , and thus @xmath192 for @xmath193 . by lemma [ lmn ] , @xmath194 for all @xmath190 $ ] .
by lemma [ lm1 ] , @xmath9 is continuous and therefore , @xmath189 for all @xmath190 $ ] .
the following lemma states that the optimal power is always larger than a constant value determined by the fixed system parameters .
[ corl ] the optimal policy @xmath9 satisfies : @xmath195\end{aligned}\ ] ] if the temperature constraint is not tight , then the problem reduces to the energy constrained problem in which case @xmath196 . if the temperature constraint is tight , @xmath9 is monotone decreasing by lemma [ new_lm ] and when the temperature level reaches @xmath0 , @xmath9 remains at @xmath185 by lemma [ multiple_r ] .
hence , @xmath197 .
the following lemma shows that , since the power is always larger than a constant value , battery energy level is never zero , except possibly at the deadline .
[ cor1 ] in an optimal policy , energy in the battery is non - zero except possibly at @xmath131 . by lemma [ corl ] ,
the optimal power is always larger than a positive constant .
thus , the battery energy does not drop to zero .
the following lemma shows that the temperature is monotone increasing throughout the transmission duration , and also is a concave function of time .
[ temperature ] the temperature with the optimal power policy is monotone increasing and concave . if the temperature constraint is never tight , then the optimal power level is @xmath198 , and from lemma [ piecewise ] , the temperature is monotone increasing .
concavity in this case follows from the concavity of the explicit expression in ( [ fin ] ) with @xmath43 .
now , assume that the temperature constraint is tight at @xmath131 .
by lemma [ corl ] , @xmath197 . from ( [ thermal ] ) , we have : @xmath199 as @xmath39 by the temperature constraint .
since @xmath9 is monotone decreasing by lemma [ new_lm ] and @xmath26 is monotone increasing , from ( [ sxtn ] ) , @xmath200 is monotone decreasing , proving the concavity of @xmath26 in this case . in view of lemma [ cor1 ] , the energy constraint can be tight only at @xmath131 .
therefore , the corresponding lagrange multiplier is a single variable @xmath201 . from lemma [ temperature ] , @xmath26 is monotone increasing . due to lemma [ multiple_r ] , when @xmath26 reaches @xmath0 , power level has to remain at @xmath185 .
accordingly , we denote the instant when the temperature reaches @xmath0 as @xmath202 . in this case
, the energy constraint is never tight , and @xmath203 . in view of lemma [ lm12 ] , the temperature constraint is tight at @xmath131 .
first , consider the case that @xmath17 is sufficiently large so that there exists @xmath204 such that @xmath205 .
for @xmath206 , @xmath126 and from ( [ slck1 ] ) , @xmath127 . from ( [ nn ] ) , when @xmath206 we have @xmath207 where @xmath208 .
since at @xmath209 the temperature reaches @xmath0 , from lemma [ multiple_r ] , we have @xmath189 for @xmath210 $ ] .
then , the optimal power has the form : @xmath211 where @xmath212 is the unit step function .
now , from lemma [ kkk ] , @xmath9 is continuous at @xmath202 and @xmath213 should be chosen accordingly .
in particular , @xmath214 .
the following lagrange multiplier @xmath215 verifies ( [ kkl ] ) : @xmath216 the corresponding optimal temperature pattern for @xmath217 is : @xmath218 and @xmath164 for @xmath219 .
we note that @xmath202 satisfies : @xmath220 so that @xmath205 .
hence , @xmath26 monotonically increases till it reaches @xmath0 , which is consistent with lemma [ temperature ] .
next , consider the case that @xmath221 .
in this case , @xmath222 where @xmath223 and @xmath224 .
therefore , the optimal @xmath9 in this case is @xmath225 we also remark that @xmath202 level that satisfies ( [ sol ] ) monotonically increases with @xmath226 . to see this ,
we rearrange ( [ sol ] ) as follows : @xmath227 let us define a multi - variable real function @xmath228 as the left hand side of ( [ sol2 ] ) and denote a specific solution as @xmath229 for fixed @xmath226 .
it is easy to see that ( [ sol2 ] ) always has a solution @xmath202 for fixed @xmath226 . to see this
, we evaluate the derivative with respect to @xmath202 as : @xmath230 that is , @xmath228 is monotone decreasing with @xmath202 . at @xmath231 , @xmath232 while @xmath233 as @xmath202 grows . in view of the continuity of @xmath228
, there exists a @xmath202 such that @xmath234 .
additionally , we observe in ( [ sol2 ] ) that for fixed @xmath202 , @xmath228 monotonically increases with @xmath226 .
therefore , if @xmath235 , then , due to monotone increasing property with respect to @xmath226 , @xmath236 for @xmath237 . hence , for @xmath238 such that @xmath239
, we have @xmath240 due to monotone decreasing property with respect to @xmath202 .
note that the optimal power policies in the energy unconstrained cases in ( [ kkl ] ) and ( [ halfway ] ) have finite energies .
if the available energy @xmath172 is larger than the corresponding energy level in ( [ kkl ] ) and ( [ halfway ] ) , then the solution is as in ( [ kkl ] ) and ( [ halfway ] ) .
otherwise , the energy constraint is active and the lagrange multiplier is @xmath241 . from ( [ nn ] )
, we have : @xmath242 we first note that there is a critical energy level @xmath243 such that if @xmath244 , then constant power policy @xmath196 is optimal .
this critical level is : @xmath245 this is the critical level below which the temperature constraint is not tight by the constant power allocation @xmath196 .
the expression in ( [ thrtn ] ) is evaluated from ( [ fin ] ) by inserting @xmath43 , and requiring @xmath246 . when @xmath247 , @xmath127 since temperature constraint is never tight . in this case ,
@xmath243 is the maximum energy level for which a constant power level is optimal .
if @xmath249 , @xmath26 is monotone increasing over @xmath47 $ ] and reaches @xmath0 at @xmath131 . if @xmath250 , the constant power level @xmath251 does not satisfy the temperature constraint .
we note from ( [ thrtn ] ) that @xmath243 increases with the deadline @xmath17 .
therefore , there exists a deadline level @xmath252 for which @xmath253 implies @xmath254 and hence constant power policy is optimal . an alternative way of observing the behavior of the optimal policy is to fix the available energy @xmath172 and @xmath28 and vary the critical temperature @xmath0 . in this case , there is a critical temperature limit @xmath255 for which @xmath196 is optimal whenever @xmath256 : @xmath257 which again is evaluated from ( [ fin ] ) with @xmath43 . in the following ,
we consider @xmath250 or @xmath258 so that both energy and temperature constraints are tight at the end of the communication session .
again , we consider two possibilities : temperature constraint becomes tight at a @xmath204 , and temperature constraint becomes tight at @xmath131 . in both cases , the energy constraint becomes tight at @xmath131 .
first , consider the case that @xmath204 : due to ( [ slck1 ] ) , @xmath127 for @xmath206 and from ( [ nn ] ) , we get : @xmath259 where @xmath208 .
additionally , @xmath189 for the remaining portion of the epoch in view of lemma [ multiple_r ] .
@xmath202 is such that for @xmath260 , @xmath189 and @xmath205 .
since @xmath261 we have : @xmath262 similarly , for @xmath205 , we have from ( [ ths ] ) with @xmath43 : @xmath263 finally , the energy constraint has to be satisfied at @xmath131 : @xmath264 if there exists @xmath265 for ( [ bir])-([uc ] ) , then @xmath9 is : @xmath266 in this case , the corresponding lagrange multiplier is : @xmath267 otherwise , when no such @xmath204 exists , the temperature constraint is tight only at @xmath268 . in this case , @xmath9 is as in ( [ yyy ] ) for @xmath89 $ ] where @xmath269 and @xmath213 have to satisfy : @xmath270 the corresponding lagrange multiplier is @xmath224 .
depending on the energy @xmath172 and the critical temperature @xmath0 , the optimal power scheduling policy @xmath9 varies according to the plots in fig .
[ reg_o ] . for small @xmath172 and fixed @xmath0 or for large @xmath0 and fixed @xmath172
, a constant power policy is optimal . for moderate and large @xmath172
, the optimal power policy is exponentially decreasing and may hit the power level @xmath271 .
note that @xmath202 level at which temperature touches the critical level decreases as @xmath0 is decreased and as @xmath172 is increased . in particular , for fixed @xmath0 , the level of @xmath202 is bounded below by the solution for @xmath272 whereas for fixed @xmath172 , @xmath202 goes to @xmath273 as @xmath0 approaches @xmath28 .
in this section , we extend the solution to the case of multiple energy arrivals .
we start with extending the properties observed for the single energy arrival case when initial temperature @xmath274 is different from @xmath28 .
the following lemma generalizes lemmas [ new_lm ] , [ multiple_r ] and [ temperature ] for the case of an arbitrary @xmath274 .
[ extension ] assume that the initial temperature @xmath274 is in the range @xmath275 instead of @xmath43 and consider the single energy arrival case : @xmath9 is monotone decreasing .
let @xmath276 $ ] denote @xmath187 : t(t)=t_c\}$ ] .
if @xmath277 , then @xmath189 for all @xmath278 $ ] and the temperature is monotone increasing and concave . if @xmath279 , then @xmath280 .
if @xmath274 is in the range @xmath281 then , instead of ( [ gg ] ) , we have the following temperature constraint : @xmath282\end{aligned}\ ] ] where @xmath283 .
note that @xmath284 for all @xmath89 $ ] , i.e. , the right hand side of ( [ gg2 ] ) is always non - negative . the argument in lemma [ new_lm ] is valid in the presence of the additional term @xmath285 in ( [ gg2 ] ) , and therefore @xmath9 is monotone decreasing .
the second claim follows from the argument in lemma [ multiple_r ] .
in particular , in addition to lemma [ new_lm ] , lemma [ lmn ] directly extends with the constraint in ( [ gg2 ] ) .
hence , the result follows by applying the argument in lemma [ multiple_r ] .
finally , @xmath26 is monotone increasing and concave due to the steps followed in lemma [ temperature ] .
in particular , if the temperature constraint is tight at @xmath131 , @xmath197 . hence , ( [ sxtn])-([svnt ] ) hold and the temperature
is monotone increasing and concave . if @xmath279 , then @xmath280 due to the energy constraint .
note that the temperature decreases in case @xmath286 . as in the single epoch case
, we will investigate the solution under special cases .
in particular , we will investigate the solution according to the time when the temperature hits the critical level .
to this end , we specialize in an interval @xmath132 $ ] such that @xmath126 for all @xmath287 and @xmath288 where @xmath289 .
note that the temperature @xmath26 is a continuous function of @xmath27 and hence there exist such intervals .
we assume that the solution is known in @xmath290 $ ] and we let @xmath291 . in this case
, the solution of ( [ opt_prob1 ] ) over the interval @xmath132 $ ] is equal to the solution of the following problem obtained by restricting the temperature constraint to be satisfied at @xmath139 only : @xmath292 } \quad & \int_{t_1}^{t_2 } \frac{1}{2}\log\left(1 + p(\tau)\right)d\tau \\ \nonumber \mbox{s.t . }
\quad & \int_{t_1}^{t_2 } a e^{b \tau } p(\tau ) d
\tau = t_{\delta}e^{bt_2 } - t_g \\ \qquad & \int_{0}^{t } p(\tau ) d\tau \leq \sum_{i=0}^{\tilde{h}(t ) } \tilde{e}_i , \qquad \forall t \in [ t_1,t_2 ] \label{opt_prob_relax}\end{aligned}\ ] ] where @xmath293 . in ( [ opt_prob_relax ] )
, @xmath294 is determined as follows : @xmath295 is the available energy in the battery at time @xmath149 .
@xmath294 for @xmath296 are the energy arrivals at instants @xmath297 .
@xmath298 is defined accordingly .
while the times @xmath299 are exactly those in the original problem , the amounts @xmath294 may be different from the original amounts as some energy may be left for use in the @xmath300 $ ] interval .
for the following argument , whether @xmath294 equals the original energy arrival amount is not relevant and we leave @xmath294 as arbitrary amounts . to obtain the solution of ( [ opt_prob_relax ] ) using this lagrangian framework ,
it is necessary and sufficient to find @xmath301 variables @xmath302 , @xmath303 and @xmath304 such that @xmath305^+ , \quad t \in [ \tilde{s}_{i-1},\tilde{s}_{i } ) , \
i=1,\ldots , \tilde{n}+1\end{aligned}\ ] ] with the corresponding slackness conditions
. therefore , for the @xmath132 $ ] interval , the solution has the structure in ( [ pow ] ) , which is parameterized by finitely many lagrange multipliers . in particular , throughout an epoch over which @xmath126 , power level satisfies @xmath306^{+}$ ] for some @xmath307 and @xmath304 not both equal to zero .
this also holds in a subinterval of an epoch over which @xmath126 . in the following lemma
, we show that in such an epoch , the temperature @xmath26 is unimodal .
[ unimodal ] if @xmath308^{+}$ ] for @xmath134 $ ] for some @xmath309 and @xmath310 , the resulting @xmath26 is unimodal over @xmath132 $ ] . from ( [ four ] )
, we have for @xmath134 $ ] , @xmath311^+ + bt_e\right ) d
\tau + t(t_1 ) \right)\end{aligned}\ ] ] first , we note that when @xmath120 , @xmath312 from ( [ thermal ] ) .
hence , it suffices to show that @xmath26 is unimodal when @xmath313 . by evaluating the integral ,
we get @xmath314 we claim that @xmath26 in ( [ lbbb ] ) is unimodal for @xmath315 .
note that the derivative of @xmath26 is : @xmath316 we let @xmath317 , @xmath318 and concentrate on @xmath319 for @xmath320 .
we note that @xmath319 is a strictly monotone decreasing function of @xmath321 for @xmath322 .
in particular , we have : @xmath323 thus , @xmath324 is strictly monotone decreasing in @xmath27 . as @xmath325 at @xmath149 , we conclude that the factor in ( [ dfsa ] ) that multiplies @xmath148 can take value @xmath273 at most once . in particular , @xmath326 can take positive or negative values at @xmath149 .
if it is positive at @xmath149 , it hits value @xmath273 at most once for @xmath315 .
if it is negative at @xmath149 , it stays negative throughout @xmath315 .
this proves that @xmath26 is unimodal over @xmath132 $ ] . in the following lemma , we show that , in an epoch @xmath327 $ ] , the temperature can not return to @xmath0 if it hits and falls below @xmath0 .
[ new ] if @xmath84 and @xmath328 for some @xmath329 where both @xmath157 and @xmath330 are in @xmath327 $ ] , then @xmath126 for all @xmath331 $ ] . by lemma [ kkk ] , @xmath161 .
by lemma [ new_lm ] , power is monotone decreasing in an epoch .
therefore , if @xmath328 , then @xmath332 and hence @xmath333 for all @xmath331 $ ] .
this , in turn , means that @xmath126 for all @xmath331 $ ] .
next , we complete the unimodal structure of the temperature by showing that it has to be monotone decreasing if it hits and falls below @xmath0 .
[ new2 ] in an epoch @xmath327 $ ] , if the temperature touches @xmath0 at @xmath157 and falls below it , then the temperature is monotone decreasing in @xmath334 $ ] . by lemma
[ new ] , if @xmath328 , then @xmath126 for all @xmath335 $ ] .
therefore , we have @xmath336^{+ } , \quad t \in [ t_h+\delta , s_{i+1}]\end{aligned}\ ] ] for some @xmath337 and @xmath310 . by lemma [ unimodal ] , @xmath26 is unimodal over @xmath335 $ ] . therefore , @xmath26 is monotone decreasing .
we next consider epochs @xmath327 $ ] and its subintervals over which @xmath126 and @xmath164 . by lemma [ unimodal ] and in view of the discussion around ( [ opt_prob_relax ] ) ,
whenever @xmath126 over an epoch , @xmath26 reaches its peak level over that epoch at only one instance .
consequently , if @xmath126 for all @xmath338 $ ] , there are three possible cases .
the first two possibilities are that @xmath26 is monotone increasing or monotone decreasing throughout the epoch .
the third possible case is that @xmath26 is monotone increasing in @xmath339 $ ] and monotone decreasing in @xmath340 $ ] for some @xmath341 .
otherwise , @xmath26 hits @xmath0 and @xmath26 does not return to @xmath0 if it falls below it due to lemma [ new ] .
therefore , if @xmath26 hits @xmath0 in an epoch @xmath342 $ ] , then that epoch is divided into three successive subintervals @xmath343 with @xmath344 , @xmath345 and @xmath346 $ ] for some @xmath347 .
@xmath26 is monotone increasing over @xmath348 , remains at @xmath0 over @xmath349 and is monotone decreasing over @xmath350 .
we finally note that if @xmath126 at @xmath131 , then @xmath126 for all @xmath89 $ ] .
this follows from lemma [ extend ] . in this case , the temperature constraint is never tight and the optimal power policy is identical to the one in @xcite .
is decreased , the energy is spent faster subject to energy causality . ] in fig .
[ exp2 ] , we plot the optimal energy expenditure for different values of critical temperature level @xmath0 . we observe that as @xmath0 is decreased , the temperature budget shrinks and the temperature constraint becomes more likely to be tight . in this case , energy is spent faster not to create high amounts of heat in the system . in general , there is a tension between causing unnecessary heat in the system and maximizing the throughput .
while we have fully characterized this tension in the single energy arrival case , it needs to be further explored in the multiple energy arrivals case . in particular , when a high amount of energy arrives into the system during the progression of communication , the transmitter has to accommodate it by cooling down and creating a temperature margin for future use .
while maximizing the throughput generally requires using the energy in the system to the fullest extent , the transmitter may have to waste energy due to the temperature limit .
we investigate this tension in numerical examples in the next section .
in this section , we provide numerical examples to illustrate the optimal power policy and the resulting temperature profile . for plots in figs .
[ model4 ] , [ model5 ] , [ model6 ] and [ model7 ] , we set @xmath351 , @xmath352 , @xmath353 and @xmath354 . therefore , the critical power level is @xmath355 . in figs .
[ model4 ] and [ model5 ] , we consider the energy unlimited scenario . in this case , the solution of ( [ sol ] ) is found as @xmath356 . in fig .
[ model4 ] , we set @xmath357 and we observe that the optimal power policy is always above the level @xmath185 . in this case ,
power strictly monotonically decreases while temperature strictly monotonically increases with temperature touching the critical level @xmath0 at the deadline . in fig .
[ model5 ] , we set the deadline as @xmath358 .
we calculate that the energy needed to have the power policy in fig .
[ model5 ] is @xmath359 . in other words ,
if the initial energy is @xmath360 then the power policy in fig .
[ model5 ] is optimal .
we observe that the optimal power level monotonically decreases to the level @xmath185 and remains at that level afterwards .
similarly , the temperature level rises to @xmath0 and remains at that level afterwards .
note that the throughput and the energy consumption in fig .
[ model5 ] are higher with respect to those in fig .
[ model4 ] .
parallel to this observation , the monotone decrease is sharper in the power policy in fig .
[ model5 ] compared to that in fig .
[ model4 ] .
since the power level has to be stabilized at @xmath185 , the temperature increase cost paid for achieving certain throughput is minimized if energy consumption starts faster and drops later . in fig .
[ model6 ] , we set the deadline to @xmath361 and the energy limit to @xmath362 .
note that this energy level is slightly less than the energy of the power policy in fig .
[ model5 ] , which translates into a right shift of the point @xmath202 .
in particular , we calculate @xmath363 as the solution of ( [ bir])-([uc ] ) in this case .
similar to the effect of decreasing the deadline observed in the comparison of figs .
[ model4 ] and [ model5 ] , we observe that decreasing the energy level yields a _ smoother _ power policy .
power level drops to @xmath185 and the temperature hits @xmath0 at a later time @xmath202 and both remain constant afterwards . for the single epoch case . ] in fig .
[ model7 ] , we consider the same system as in previous figures with two energy arrivals instead of one and with @xmath364 . in particular , @xmath365 is available initially and @xmath366 arrives at time @xmath367 . in this case
, we calculate @xmath368 .
the energy causality constraint is tight and the power level makes a jump at the energy arrival instant . note that the temperature is continuous at the energy arrival instant even though its first derivative is not .
while the power level has a smooth start , a sharper decrease is observed towards the end since the harvested energy has to be fully utilized . in particular ,
the temperature increase before the energy arrival is kept to a minimum level so as to have a higher heat budget for the larger energy that arrives later .
the temperature hits @xmath0 at @xmath369 after which the power and temperature both remain constant . for the single epoch case . ] finally in fig .
[ model8 ] , we illustrate a curious behavior in the optimal policy . for this example , we set @xmath351 , @xmath370 , @xmath371 and @xmath372 .
initial energy is @xmath373 and energy arrives at @xmath374 with amount @xmath375 and the deadline is @xmath361 .
we observe that energy causality constraint is tight at @xmath374 whereas it is not tight at @xmath131 meaning that some energy is wasted in order not to cause excessive heat .
the temperature generated in this throughput optimal power policy first monotonically increases , hits @xmath0 at @xmath376 , remains there till @xmath377 and drops below @xmath0 .
we interpret the drop in the temperature in the first epoch as an effort to create temperature margin for the high energy arrival in the next epoch .
we calculate @xmath378 as the time after which power level remains at @xmath379 and the temperature remains at @xmath0 .
note that under unlimited energy , temperature would hit @xmath0 at @xmath380 .
due to the energy scarcity in the first epoch , temperature hits @xmath0 later and drops below @xmath0 .
a common behavior we observe in each numerical example is that temperature ultimately increases between two epochs where energy causality constraint is tight .
further research is needed to quantify the relations between the amount of temperature generated while performing optimally in terms of throughput .
while monotonicity of the temperature is lost when multiple energy harvests exist , we note that monotonicity of the temperature is guaranteed in the last epoch due to lemma [ extension ] . and @xmath381 for the single epoch case .
] and @xmath366 at @xmath382 and @xmath364 . ] at @xmath383 and @xmath375 at @xmath374 and @xmath361 . ]
we considered throughput maximization for an energy harvesting transmitter over an awgn channel under temperature constraints .
we used a linear system model for the heat dynamics and determined the throughput optimal power scheduling policy under a maximum temperature constraint by using a lagrangian framework and the kkt optimality conditions .
we determined for the single energy arrival case that the optimal power policy is monotone decreasing whereas the temperature is monotone increasing and both remain constant after the temperature hits the critical level .
we then generalized the solution for the case of multiple energy arrivals .
while monotonicity of the temperature is lost when multiple energy harvests exist , we observed that the temperature ultimately increases while maximizing the throughput .
we also observed that the main impact of the temperature constraints is to facilitate faster energy expenditure subject to energy causality constraints . additionally ,
even though using all of the available energy is optimal for throughput maximization only , with temperature constraints , energy may have to be wasted in order not to exceed the critical temperature .
q. tang , n. tummala , s. gupta , and l. schwiebert , `` communication scheduling to minimize thermal effects of implanted biosensor networks in homogeneous tissue , '' _ ieee trans . on biomed .
_ , vol .
52 , pp .
12851294 , july 2005 .
s. ullah , h. higgins , b. braem , b. latre , c. blondia , i. moerman , s. saleem , z. rahman , and k. kwak , `` a comprehensive survey of wireless body area networks , '' _ jour . medical .
_ , vol .
36 , pp . 10651094 , june 2012 .
a. mutapcic , s. boyd , s. murali , d. atienza , g. de micheli , and r. gupta , `` processor speed control with thermal constraints , '' _ ieee trans .
on circ . and sys .-
i _ , vol .
56 , pp .
19942008 , september 2009 .
o. ozel , k. tutuncuoglu , j. yang , s. ulukus , and a. yener , `` transmission with energy harvesting nodes in fading wireless channels : optimal policies , '' _ ieee jour . on selected areas in commun .
_ , vol . 29 , pp .
17321743 , september 2011 .
o. ozel , j. yang , and s. ulukus , `` optimal broadcast scheduling for an energy harvesting rechargeable transmitter with a finite capacity battery , '' _ ieee trans . on wireless comm .
_ , vol . 11 , pp .
21932203 , june 2012 .
b. devillers and d. gunduz , `` a general framework for the optimization of energy harvesting communication systems with battery imperfections , '' _ jour . of comm . and netw .
_ , vol . 14 , pp . 130 139 , april 2012 .
k. tutuncuoglu , a. yener , and s. ulukus , `` optimum policies for an energy harvesting transmitter under energy storage losses , '' _ ieee jour . on
selected areas in commun .
_ , vol .
33 , pp .
467481 , march 2015 .
j. xu and r. zhang , `` throughput optimal policies for energy harvesting wireless transmitters with non - ideal circuit power , '' _ ieee jour . on selected areas in commun .
_ , vol .
32 , pp . 322332 , february 2014 .
o. ozel , k. shahzad , and s. ulukus , `` optimal energy allocation for energy harvesting transmitters with hybrid energy storage and processing cost , '' _ ieee trans . on signal proc .
_ , vol .
62 , pp .
32323245 , june 2014 . | motivated by damage due to heating in sensor operation , we consider the throughput optimal offline data scheduling problem in an energy harvesting transmitter such that the resulting temperature increase remains below a critical level .
we model the temperature dynamics of the transmitter as a linear system and determine the optimal transmit power policy under such temperature constraints as well as energy harvesting constraints over an awgn channel .
we first derive the structural properties of the solution for the general case with multiple energy arrivals .
we show that the optimal power policy is piecewise monotone decreasing with possible jumps at the energy harvesting instants .
we derive analytical expressions for the optimal solution in the single energy arrival case .
we show that , in the single energy arrival case , the optimal power is monotone decreasing , the resulting temperature is monotone increasing , and both remain constant after the temperature hits the critical level .
we then generalize the solution for the multiple energy arrival case . |
over the recent past there has been considerable experimental activity in the direct detection of dark matter@xcite and further progress is expected in the ongoing experiments@xcite and new experiments that may come online in the future@xcite . at the same time
there have been several theoretical developments which have shed light on the ambiguities and possible corrections that might be associated with the predictions on supersymmetric dark matter .
these consist of the effects on the dark matter analyses of wimp velocity@xcite and of the rotation of the galaxy@xcite , the effects of the uncertainties of quark densities@xcite and the uncertainties of the susy parameters@xcite , effects of large cp violating phases@xcite , effects of scalar nonuniversalities@xcite , effects of nonuniversalities of gaugino masses@xcite and effects of coannihilation@xcite . in this paper
we will discuss some of these briefly but mainly focus on the effects of nonuniversalities of the gaugino masses on dark matter . in the minimal supersymmetric standard model ( mssm )
there are 32 supersymmetric particles and with r parity conservation the lowest mass supersymmetric particle ( lsp ) is absolutely stable . in many unified models , such as in the sugra models@xcite
, one finds that the lightest neutralino is the lsp over most of the parameter space of the model .
thus the lightest neutralino is a candidate for cold dark matter .
the quantity that constrains supersymmetric models is @xmath0 where @xmath1 where @xmath2 is the density of relic neutralinos at the current temperatures , and @xmath3 is the critical matter density , and h is the hubble parameter @xmath4 in units of 100 km / smpc .
the most recent measurements of h from the hubble space telescope give@xcite @xmath5 similarly the most recent analyses of @xmath6 give@xcite @xmath7 if we assume that the component of @xmath8 in @xmath6 is @xmath9 which appears reasonable , then this leads to the result @xmath10 .
perhaps a more cautious choice of the range would be a @xmath11 range which gives @xmath12 the quantity of interest theoretically is @xmath13 here @xmath14 is the freeze - out temperature , @xmath15 where k is the boltzman constant , @xmath16 is the number of degrees of freedom at the time of the freeze - out , @xmath17 is the reheating factor , and @xmath18 is given by @xmath19 where @xmath20 is the thermal average with @xmath21 the neutralino annihilation cross - section and v the neutralino relative velocity .
both direct and indirect methods are desirable and complementary for the detection of milky way wimps .
we shall focus here on the direct detection . in this case
the fundamental detector is the quark and the relevant interactions are the supergravity neutralino - quark - squark interactions .
the scattering of neutralinos from quarks contains squark poles in the s channel and the z boson and the higgs boson ( @xmath22 ) poles in the t channel .
since the wimp scattering from quarks is occuring at rather low energies one may , to a good approximation , integrate on the intermediate squark , z and higgs poles to obtain a low energy effective lagrangian which gives a four - fermi interaction of the following form @xmath23 the contribution of d is generally small and thus the scattering is effectively governed by the terms a , b and c. analysis of dark matter is affected by several factors .
we discuss these briefly below .
two of the quantities that control the detection of dark matter are the wimp mass density and the wimp velocity .
estimates of milky way wimp density lie in the range@xcite @xmath24 and the event rates in the direct detection depend directly on this density .
a second important factor regarding wimps that enters in the dark matter analyses is the wimp velocity .
one typically assumes a maxwellian velocity distribution for the wimps and the current estimates for the rms wimp velocity give @xmath25 with , however , a significant uncertainty .
estimates for the uncertainty lie in the range of @xmath26 km / s to @xmath27 km / s @xcite .
a reasonable estimate then is that the rms wimp velocity lies in the range@xcite @xmath28 analyses including the wimp velocity variations show that the detection rates can have a significant variation , i.e. , a factor of 2 - 3 on either side of the central values@xcite . the scattering of neutralinos from quarks are dominated by the scalar interaction which is controlled by the term c in eq.(6 ) .
the dominant part of the scattering thus arises from the scalar part of the @xmath29 cross - section which is given by @xmath30 here @xmath31 is the reduced mass in the @xmath29 system and @xmath32 ( i = u , d , s quarks ) are quark densities inside the proton defined by @xmath33 there are significant uncertainties in the determination of @xmath32 . to see the range of these uncertainties it is useful to parametrize the quark densities
so that@xcite @xmath34 where we have defined @xmath35 the current range of determinations of @xmath36 give@xcite @xmath37 with the above range of errors one finds that @xmath32 lie in the range @xmath38 of these the errors in @xmath39 generates the largest variations .
a detailed analysis shows that the scalar cross - section can vary by a factor of 5 in either direction due to errors in the quark densities@xcite .
the soft susy breaking parameters that arise in supersymmetric theories after spontaneous breaking are in general complex with phases o(1 ) and can lead to large electric dipole moment of the electron and of the neutron in conflict with current experiment .
recently a cancellation mechanism was proposed as a possible solution to this problem@xcite . with the cancellation mechanism the total edm of the electron and of the neutron can be in conformity with data even with phases o(1 ) and sparticle masses which are relatively light .
the presence of large cp phases can affect dark matter and other low energy phenomena .
effects of cp violating phases on dark matter have been investigated for some time@xcite and more recently such analyses have been extended to determine the effects of large cp phases under the cancellation mechanism@xcite .
one finds that the edm constraints play a crucial role in these analyses .
thus in the absence of cp violating phases one finds that the @xmath29 cross - section can change by orders of magnitude when plotted as a function of @xmath40 ( the phase of @xmath41 ) .
these effects are , however , significantly reduced when the constraints arising from the current experimental limits on the electron and on the neutron edms are imposed . after the imposition of the constraints the effects of cp violating phases are still quite significant in that the @xmath29 cross - section can vary by a factor of @xmath42 .
thus precision predictions of the @xmath29 cross - section should take account of the cp phases if indeed such phases do exist in a given model .
indeed many string and d brane models do indeed possess such phases and thus inclusion of such phases is imperative in making predictions for direct detection in such models .
the effects of coannihilation may become important when the next to the lowest supersymmetric particle ( nlsp ) has a mass which lies close to the lsp mass@xcite .
the size of the effects is exponentially damped by the factor @xmath43 where @xmath44 , @xmath45 and where @xmath46 is the lsp mass .
because of this damping the coannihilation effects are typically important only for regions of the parameter space where the constraint @xmath47 is satisfied .
some of the possible candidates for nlsp are the light stau @xmath48 , @xmath49 , the next to the lightest neutralino @xmath50 , and the light chargino @xmath51 .
an interesting result one finds is that in msugra the upper limit on the neutralino mass consistent with the current experimental constraints on the relic density is extended from 200 gev to 600 gev@xcite when the effects of @xmath52 coannihilation are included . in secs.(2.6 ) and ( 2.7 ) we will show that the allowed range of the neutralino mass can also be extended by inclusion of nonuniversalities in the gaugino masses . the minimal sugra model is based on the universality at the gut scale .
this includes the universality of the scalar masses , of the gaugino masses and of the trilinear couplings at the gut scale . in supergravity
unified models the universality of the soft susy breaking parameters arises from the assumption of a flat kahler potential .
however , the nature of physics at the gut / planck scale is not fully understood and a more general analysis of the soft susy breaking sector requires that one work with a curved kahler potential@xcite .
such an analysis in general leads to nonuniversalities in the scalar sector of the theory .
however , the nonuniversalities in the scalar sector can not be completely arbitrary as there are very stringent constraints on the system from the limits on the flavor changing neutral currents ( fcnc ) .
a satisfaction of the constraints requires essentially a degeneracy in the scalar masses in the first two generations at the gut scale .
however , the constraints on the scalar masses in the higgs sector and on masses in the third generation are far less severe and one could introduce significant amounts of nonuniversalities in these sectors without violating the fcnc constraints . it is found convenient to parametrize the nonuniversalities in the higgs sector by @xmath53 so that at the gut scale ( @xmath54 ) one has @xmath55 , @xmath56 .
similarly one may parametrize the nonuniversalities in the third generation squark sector by @xmath57 so that at the scale @xmath54 one has @xmath58 , @xmath59 .
these nonuniversalities have a significant effect on the low energy physics .
one of the main effects that occurs is through the effect on @xmath41 which is determined via the constraint of the radiative breaking of the electro - weak symmetry and is modified in the presence of the nonuniversalities in the higgs sector and in the third generation sector . to one loop order it is given by@xcite @xmath60 here
@xmath61 is the value of @xmath41 in the absence of nonuniversalities , @xmath62 depends on the top yukawa coupling and defines the position of the landau pole , @xmath63 , and @xmath64 is the loop correction .
we note that the entire effect of nonuniversalities is now explicity exhibited .
one finds that the universalities can significantly affect the event rates .
the effect on the event rates occurs specifically because of the effect on @xmath41 .
thus one finds that for certain regions of the parameter space the nonuniversalities in the higgs and in the third generation sector make a negative contribution to @xmath65 which leads to larger higgsino components for the neutralino . since in the direct detection
the scattering is dominated by the scalar @xmath29 cross - section which in turn depends on the product of the gaugino and the higgsino components one finds that a smaller @xmath41 leads to larger event rates in the direct detection .
a detailed analysis of the effects of nonuniversalities of the scalar masses has been given in refs.@xcite .
we will discuss further this phenomena in the context of the nonuniversalities in the gaugino sector in the next section .
nonuniversality of gaugino masses arises in grand unified models via corrections to the gauge kinetic energy functions@xcite .
thus in grand unified models a non - trivial gauge kinetic energy function leads to a gaugino mass matrix which has the form@xcite @xmath66 as an example if we consider the gut group to be su(5 ) then the gauge kinetic energy function @xmath67 transforms as follows @xmath68 where @xmath69 stands for the symmetric product .
the term that transforms like the singlet of su(5 ) in the gauge kinetic energy function leads to universality of the gaugino masses , while the @xmath70 plet , the @xmath71 plet and the @xmath72 plet will generate corrections to universality . in general one could have an admixture of the various representations and this will lead to gaugino masses of the form @xmath73 where @xmath74 depend on r and for the representations @xmath75 they are given in table 1@xcite .
the nonuniversality of the gaugino masses also leads to corrections of the gauge coupling constants at the gut scale and in general one has @xmath76 .
we note , however , that the coefficients @xmath77 that enter in @xmath78 are different than those that enter in @xmath79 .
this is so because the corrections to @xmath78 involve only the gauge kinetic energy function while the corrections to @xmath79 involve the gauge kinetic energy function as well as the nature of gut physics .
su(5 ) rep & @xmath80 & @xmath81 & @xmath82 + & 1 & 1&1 + & -1 & @xmath83&@xmath84 + & -5&3 & @xmath85 + & 10 & 2 & 1 + + the gaugino sector nonuniversalities affect @xmath41 . to exhibit this effect
we can expand @xmath41 determined via the constraint of the radiative breaking of the electro - weak symmetry in terms of the parameter @xmath86 .
one finds the following expansion @xmath87 and for @xmath88 one has@xcite @xmath89 thus in these cases the nonuniversalities lead to a smaller value of @xmath90 .
now as already mentioned in the previous section the higgsino components become more dominant as @xmath41 becomes smaller .
we can exhibit this analytically for the case when @xmath41 is small but we are still in the scaling region@xcite where @xmath91 . in this case
it is possible to analytically investigate the size of the gaugino - higgsino components @xmath92 of the lsp defined by @xmath93 where @xmath94 is the bino , @xmath95 is the wino , and @xmath96 , and @xmath97 are the two higgsinos . in this case one
finds that the gaugino components of the lsp are given by @xmath98 , and @xmath99 while the higgsino components are given by@xcite @xmath100 , @xmath101 . from
the above one finds that the higgsino components have a dependence on the inverse power of @xmath41 and thus a smaller @xmath41 will lead to a larger scalar @xmath102 cross - section .
the literature on the analyses of dark matter relic density@xcite and direct detection in mssm@xcite and in sugra models@xcite is quite extensive@xcite .
we discuss here the quantitative effects of the gaugino mass nonuniversality on dark matter .
some features of the effects of gaugino mass nonuniversalities have already been discussed in the literature@xcite and we review here the more recent developments@xcite .
the techniques used in the analysis are as discussed in ref.@xcite and in the analysis we impose the @xmath103 constraint@xcite . in fig.[exfig1 ]
we plot the scalar @xmath29 cross - section as a function of the neutralino mass for the case of gut scale nonuniversalities with values of @xmath104 in the range -0.1 to 0.08 .
one finds that the scalar cross - section is enhanced for negative values of @xmath104 just as one would expect from the general discussion above because it is for the case of @xmath104 negative that @xmath41 becomes small .
one finds that in general the scalar cross - section increases systematically as @xmath105 increases for negative values of @xmath104 and an enhancement of the scalar cross - section by as much as a factor of 10 can be gotten relative to the universal case of @xmath106 .
one also finds an enhancement of the allowed range of the neutralino mass consistent with the constraints . in fig.[exfig2 ] we plot the maximum and the minimum of the scalar cross - section as a function of the neutralino mass for the case of gut scale nonuniversalities where the nonuniversalities arise from the 200 plet representation with @xmath107 when the other parameters are varied over their assumed naturalness range .
the current experimental limits from dama@xcite and from cdms@xcite are also plotted .
further , the currents limits would certainly be significantly improved in other dark matter detectors in the future@xcite and in fig.[exfig2 ] we also plot the expected limits from future cdms , and from genius@xcite .
one finds that the current experiment does constrain the theory in a small region of the parameter space .
further , the expected sensitivity in future experiment , i.e. , in cdms and in genius will explore a major part of the parameter space of this model .
we also note that the inclusion of nonuniversality significantly increases the allowed range of the neutralino parameter space .
nonuniversality of gaugino masses is rather generic in string theory .
however , the specific nature of the nonuniversality will depend on the details of the compactification .
we discuss here the effects of gaugino mass nonuniversality on dark matter in the context of d brane models .
the possibility of nonuniversal gaugino phases in brane models arises from the choice of embedding of the different gauge groups of the standard model on different branes .
one may consider , from example , models that arise from type iib string compactified on a six - torus @xmath108 which contains 9 branes , @xmath109 and @xmath110 ( i=1,2,3 ) branes and 3 branes .
not all the branes can be present simultaneuosly due to the constraint of n=1 supersymmetry which requires that one has either 9 branes and @xmath110 ( i=1,2,3 ) branes or @xmath109 ( i=1,2,3 ) branes and 3 branes . in the following
we will make the choice of embedding on @xmath111 branes and 5 branes@xcite .
one of the major problems in developing a sensible string phenomenology is that the mechanism of supersymmetry breaking in string theory is still lacking .
however , some progress can be made by use of an efficient parametrization of supersymmetry breaking .
here we use the parametrization where the breaking of supersymmetry arises from the breaking generated by the dilaton and the moduli vev s of the following form@xcite @xmath112@xmath113@xmath114 , @xmath115@xmath113@xmath116 .
we consider now a specific 9 - 5 brane model .
here one embeds the @xmath117 gauge group on 9 branes and @xmath118 gauge group on a @xmath119 brane . the alternative possibility of embedding the standard model gauge group on five branes
is discussed in the last two papers of ref.@xcite . for the @xmath120 brane model
the soft susy breaking sector of the theory is given by@xcite @xmath121 here @xmath122 is the goldstino direction in the dilaton s ( moduli @xmath123 ) vev space .
we discuss now dark matter on d branes . in fig.[exfig3 ] we give a plot of the scalar @xmath29 cross section as a function of the neutralino mass for the @xmath120 d brane model .
one of the interesting feature of the d brane model is that the scalar masses are in general not universal .
however for @xmath124 one has @xmath125 and the scalar masses are universal although the gaugino masses are still nonuniversal . since we are mostly interested here in investigating the effects of nonuniversalities of the gaugino masses in this analysis , we impose universality of the scalar masses and set @xmath124 . in fig.[exfig3 ] we give a plot of the minimum and the maximum of the scalar @xmath29 cross section under this constraint .
one finds that under the assumed constraints the allowed domain of the parameter space has the general features which are similar to the gut scale nonuniversalities .
one common feature is that the allowed domain of the parameter space is extended close to 500 gev .
one may note that if in addition to the constraint @xmath124 one also sets @xmath126 one finds also universality of the gaugino masses .
this situation is exhibited by the vertical dark line in the enclosed region on the left hand side in fig.[exfig3 ] .
in this paper we have given a brief review of the recent theoretical developments in the analyses of supersymmeteric dark matter .
our emphasis has been in exploring the effects of uncertainties of the input data and the effects of nonuniversalities of the gaugino masses on dark matter analyses .
it is found that the uncertainties of the wimp velocities can change detection rates by up to factors of 2 - 3 while the uncertainties in quark masses and densities can change the @xmath29 cross - section by up to factors of 5 in either direction .
the effects of gaugino mass nonuniversalities on dark matter analyses is found to be quite dramatic .
it is seen that gaugino mass nonuniversalities can increase the higgsino components of the lsp and significantly increase the @xmath29 cross - section from scalar interactions and also increase the allowed range of the lsp consistent with relic density constraints .
thus an increase in the scalar @xmath29 cross - sections by up to a factor of 10 can occur while the allowed range of the neutralino masses can move up to 500 gev consistent with the relic density constraints .
data from current dark matter experiments is beginning to put constraints on models with nonuniversalities .
these constraints will become more severe as the sensitivity of dark matter experiments increase in the future . + * acknowledgements * + this research was supported in part by nsf grant phy-9901057 .
99 r. belli et.al .
, `` search for wimp annual modulation signature : results from dama / nai-3 and dama / nai-4 and the global combined analysis '' , dama collaboration preprint infn / ae-00/01 , 1 february , 2000 .
u. chattopadhyay , t. ibrahim and p. nath , phys . rev . *
d60*,063505(1999 ) ; t. falk , a. ferstl and k. olive , hep - ph/9908311 ; s. khalil and q. shafi , nucl . phys . *
b564 * , 19(1999 ) ; k. freese and p. gondolo , hep - ph/9908390 ; s.y .
choi , hep - ph/9908397 ; s. khalil , hep - ph/9910408 .
chamseddine , r. arnowitt , and p. nath , phys .
lett . * 49 * , 970(1982 ) ; for a review see , p. nath , r. arnowitt and a.h .
chamseddine , applied n=1 supergravity , trieste lectures , 1983(world scientific , singapore,1984 ) ; h. p. nilles , phys .
rep . * 110 * , 1(1984 ) ; s. abel et.al . , hep - ph/0003154 .
t. ibrahim and p. nath , phys .
* b 418 * , 98(1998 ) ; phys . rev .
* d57 * , 478(1998 ) ; phys . rev .
* d58 * , 111301(1998 ) ; t. falk and k olive , phys . lett .
* b 439 * , 71(1998 ) ; m. brhlik , g.j .
good , and g.l .
kane , phys .
* d59 * , 115004 ( 1999 ) ; a. bartl , t. gajdosik , w. porod , p. stockinger , and h. stremnitzer , phys . rev . * 60 * , 073003(1999 ) ; t. falk , k.a .
olive , m. prospelov , and r. roiban , nucl . phys . *
b560 * , 3(1999 ) ; s. pokorski , j. rosiek and c.a .
savoy , nucl .
b570 * , 81(2000 ) ; e. accomando , r. arnowitt and b. datta , hep - ph/9907446 ; m. brhlik , l. everett , g. kane and j.lykken , phys .
lett . * 83 * , 2124 , 1999 ; hep - ph/9908326 ; e. accomando , r. arnowitt and b. datta , phys . rev . *
d61 * , 075010(2000 ) . s. k. soni and h. a. weldon , phys
b126 * , 215(1983 ) ; v. s. kaplunovsky and j. louis , phys . lett . *
b306 * , 268(1993 ) ; d. matalliotakis and h.p .
nilles , nucl.phys.*b435 * , 115(1995);m .
olechowski and s. pokorski , phys.lett . *
b344 * , 201(1995 ) ; n. polonski and a. pomarol , phys.rev.*d51*,6532(1995 ) .
j. ellis , k. enqvist , d .
v. nanopoulos , and k. tamvakis , phys . lett . *
155b * , 381(1985 ) ; m. drees , phys . lett . *
b158 * , 409(1985 ) ; t. dasgupta , p. mamales , p. nath , phys.rev . * d52 * , 5366(1995 ) ; d. ring , s. urano and r. arnowitt , phys . rev . *
d52*,6623(1995 ) .
g. anderson , c.h .
chen , j.f .
gunion , j. lykken , t. moroi , and y. yamada , hep - ph/9609457 ; g. anderson , h. baer , c - h chen and x. tata , hep - ph/9903370 ; k. huitu , y. kawamura , t. kobayashi and k. puolamaki , hep - ph/9903528 .
a small sample of papers in direct detection in mssm consists of : w.a .
goodman and e. witten , phys . rev . * d31 * , 3059(1983 ) ; g. greist , phys .
lett . * 61 * , 666(1988 ) ; j. ellis and r. flores , phys . * b263*,259(1991 ) ; a. bottino et.al .
astro . part
* 1 * , 61(1992 ) ; m. drees and m. nojiri , phys .
rev.*d48 * , 3483(1993 ) ; v.a .
bednyakov , h.v .
klapdor - kleingrothaus and s.g .
kovalenko , phys . rev . * d50 * , 7128(1994 ) .
r. arnowitt and p. nath , phys .. rev .
lett.*70 * , 3696(1993 ) ; ibid , 74 , 4592(1995 ) ; g. kane , c. kolda , l. roskowski and j. wells , phys .
rev.*d49 * , 6173(1994 ) ; h. baer and m. brhlik , phys .
rev.*d53 * , 597(1996 ) .
p. nath and r. arnowitt , phys .
b336 * , 395(1994 ) ; f. borzumati , m. drees , and m.m .
nojiri , phys .
rev.*d51 * , 341(1995 ) ; v. barger and c. kao , phys .
rev . * d57 * , 3131(1998 ) ; h. baer , m. brhlik , d. castano and x. tata , phys . rev .
* d58 * , 015007(1998 ) . | several new effects have been investigated in recent analyses of supersymmetric dark matter .
these include the effects of the uncertainties of wimp velocity distributions , of the uncertainties of quark densities , of large cp violating phases , of nonuniversalities of the soft susy breaking parameters at the unification scale and of coannihilation on supersymmetric dark matter .
we review here some of these with emphasis on the effects of nonuniversalities of the gaugino masses at the unification scale on the neutralino - proton cross - section from scalar interactions .
the review encompasses several models where gaugino mass nonuniversalities occur including sugra models and d brane models .
one finds that gaugino mass nonuniversalities can increase the scalar cross - sections by as much as a factor of 10 and also significantly extend the allowed range of the neutralino mass consistent with constraints up to about 500 gev .
these results have important implications for the search for supersymmetric dark matter .
= 14.5pt |
the problem of understanding the large gluon density regime in high energy scattering has always been one of the challenges of perturbative qcd ( pqcd ) .
unitarity of the total cross section and saturation of the gluon distribution are among the most important issues related to the problem .
the bfkl equation @xcite is the only well established tool of pqcd which allows us to explore this high density region by resumming the leading longitudinal logarithmic contribution to the scattering process . in bfkl evolution
the small@xmath0 partons are produced overlapping each other in the transverse coordinate space @xcite , therefore creating high density regions in the hadron s wave function ( hot spots ) .
the next to leading order correction to bfkl equation has been calculated recently @xcite .
although the final conclusion one should draw from the calculation of @xcite is still to be understood , there are some serious problems associated with the interpretation of the result @xcite .
however , we are not going to address these issues in this paper for the following reason . as was shown in @xcite the effects of the second order bfkl kernel become important in hadron
hadron scattering at the rapidities of the order of @xmath2 , with @xmath3 the strong coupling constant . at the same time the unitarity constraints , associated with the multiple ( leading order ) hard pomeron exchanges
are expected to be reached at @xmath4 @xcite , which is parametrically smaller than @xmath5 for small coupling constant .
therefore multiple pomeron exchanges become important at lower center of mass energies than the effects of subleading corrections , possibly leading to unitarization of the total hadron
hadron cross section .
hence the problem of resummation of the multiple pomeron exchanges seems to be more important for describing the recent experimental results , such as zeus 1995 data @xcite , which probably shows evidence of saturation of the @xmath1 structure function at low @xmath6 . in this paper
we are going to consider deep inelastic scattering ( dis ) of a virtual photon on a large nucleus and will resum all multiple pomeron exchanges contributing to the @xmath1 structure function of the nucleus in the leading longitudinal logarithmic approximation in the large @xmath7 limit .
the first step in that direction in pqcd was the derivation by gribov , levin and ryskin ( glr ) of an equation describing the fusion of two pomeron ladders into one in the double logarithmic approximation @xcite .
the resulting equation with the coefficient in front of the quadratic term calculated by mueller and qiu @xcite for a low density picture of a spherical proton of radius @xmath8 reads @xmath9 ^ 2.\ ] ] this equation sums up all multiple hard pomeron exchanges in the gluon distribution function in the double logarithmic limit . since then there have been several attempts to generalize the glr equation .
recently an equation has been proposed by ayala , gay ducati , and levin in @xcite , which tries to incorporate the glauber type multiple rescatterings of a probe on the nucleons in a nucleus ( see fig .
[ glauber ] ) .
using the results of mueller in @xcite for a pair of gluons multiply rescattering inside a nucleus , the authors of @xcite obtained the following equation for the gluon distribution of the nucleus in the double logarithmic approximation @xmath10 \right\}.\end{aligned}\ ] ] if one expands the right hand side of eq .
( [ glamu1 ] ) to the second order in @xmath11 one recovers the glr equation [ eq .
( [ glamu1 ] ) is written here for a cylindrical nucleus .
therefore the coefficients in the obtained glr equation will not match those of eq .
( [ glreq ] ) ] .
an extensive work on resumming the multiple pomeron exchanges in the gluon distribution function in the leading @xmath12 approximation ( i.e. without taking the double logarithmic limit ) has been pursued by jalilian - marian , kovner , leonidov , mclerran , venugopalan , and weigert @xcite . starting with a model of a large nucleus @xcite , which provides some effective action @xcite , they develop a renormalization group procedure which integrates out harder longitudinal gluonic degrees of freedom in the nucleus and allows one to resum leading @xmath12 contribution to the gluon distribution function .
the resulting equation , written in a functional form in @xcite is supposed to resum these leading logarithms including also all powers of color charge density of the nucleus , which in a more traditional language corresponds to resummation of multiple pomeron exchanges .
however , even though that equation at the lowest ( one pomeron ) level reduces to the expected bfkl equation @xcite , in general it is a very complicated functional differential equation which can not be solved even numerically .
recently the double logarithmic limit of that equation was obtained @xcite , providing us with another equation for @xmath13 : @xmath14 @xmath15.\end{aligned}\ ] ] eq .
( [ alik ] ) is written here for a cylindrical nucleon with transverse area @xmath16 . in the limit of small gluon density eq .
( [ alik ] ) reduces to glr equation @xcite .
our approach in this paper is pursuing the same goals as authors of @xcite .
we will also write an equation which resums all multiple pomeron exchanges on a nucleus in the leading logarithmic approximation .
however our strategy is a bit different from @xcite .
we will consider the scattering of a virtual photon on a nucleus at rest , therefore putting all the qcd evolution in the wave function of the virtual photon .
this is different from what was done by the authors of @xcite , since they were developing the qcd evolution inside the nucleus .
the virtual photon s wave function including the leading logarithmic evolution was constructed in the large @xmath7 limit by mueller in @xcite . this so called dipole wave function in fact contains all multiple pomeron exchanges , which in the large @xmath7 language can be pictured as multiple color `` cylinders '' .
a numerical analysis of the unitarization of the total onium onium cross section through multiple pomeron exchanges was carried out in the framework of the dipole model by salam in @xcite .
considering the scattering of a virtual photon ( quarkonium state ) on a nucleus simplifies the problem , allowing us to treat it analytically . in sect .
ii we will use the dipole wave function to write down an equation which governs the evolution of the @xmath1 structure function of the nucleus ( formula ( [ eqn ] ) ) .
our equation is directly related to a physical observable ( @xmath1 ) and , therefore , is free from all the problems and ambiguities associated with dealing with the gluon distribution function @xmath13 .
it is a non linear integral equation , not a functional differential equation like in @xcite
. therefore one should be able to solve eq .
( [ eqn ] ) at least numerically .
we will dedicate sect .
iii to exploring the equation resulting from taking double logarithmic ( large @xmath6 ) limit of the equation derived in sect .
we see that our equation reduces to glr equation , failing to reproduce eqs .
( [ glamu1 ] ) , ( [ alik ] ) . finally , in sect .
iv we will conclude by discussing the limitations of the large @xmath7 approximation , as well as some advantages of our approach .
we start by considering a deep inelastic scattering process on a nucleus . as shown in fig.[glauber ] , the incoming virtual photon with a large @xmath17 component of the momentum splits into a quark
antiquark pair which then interacts with the nucleus at rest .
we model the interaction by no more than two gluon exchanges between each of the nucleons and the quark
antiquark pair .
this is done in the spirit of the quasi
classical approximation used previously in @xcite .
the interactions are taken in the eikonal approximation . then
, as could be shown in general , i.e. , including the leading logarithmic qcd evolution , the total cross section , and , therefore , the @xmath1 structure function of the nucleus can be rewritten as a product of the square of the virtual photon s wave function and the propagator of the quark
antiquark pair through the nucleus @xcite.the expression reads @xcite @xmath18 \
n({\bf x}_{01},{\bf b}_0 , y ) , \ ] ] where the incoming photon with virtuality @xmath19 splits into a quark antiquark pair with the transverse coordinates of the quark and antiquark being @xmath20 and @xmath21 correspondingly , such that @xmath22 .
the coordinate of the center of the pair is given by @xmath23 .
@xmath24 is the rapidity variable @xmath25 .
the square of the light cone wave function of @xmath26 fluctuations of a virtual photon is denoted by @xmath27 and @xmath28 for transverse and longitudinal photons correspondingly , with @xmath29 being the fraction of the photon s longitudinal momentum carried by the quark . at the lowest order in electromagnetic coupling ( @xmath30 )
@xmath27 and @xmath31 are given by [ @xcite and references therein ] [ wfn ] @xmath32 \right\},\ ] ] @xmath33 with @xmath34 .
we consider massless quarks having only one flavor . the quantity @xmath35 has the meaning of the forward scattering amplitude of the quark
antiquark pair on a nucleus @xcite . at the lowest ( classical ) order
not including the qcd evolution in rapidity it is given by [ gamma ] @xmath36 \right\},\ ] ] with @xmath37 as defined in @xcite and @xmath38 being the mean free path of a gluon in a nuclear medium , as defined in @xcite . in the logarithmic approximation for large @xmath6 (
small @xmath39 ) eq . ( [ gamma1 ] ) can be rewritten as @xmath40 \right\}.\ ] ] @xmath41 is the propagator of the @xmath42 pair through the nucleus .
the propagator could be easily calculated , similarly to @xcite , giving the glauber multiple rescattering formula ( [ gamma ] ) . here and throughout the paper we assume for simplicity that the nucleus is a cylinder , which appears as a circle of radius @xmath8 in the transverse direction and has a constant length @xmath43 along the longitudinal @xmath29 direction .
therefore its transverse cross sectional area is @xmath44 . in formula ( [ gamma ] )
@xmath45 is the atomic number of the nucleus , @xmath3 is the strong coupling constant and @xmath46 is the gluon distribution in a nucleon in the nucleus , taken at the lowest order in @xmath3 , similarly to @xcite .
( [ gamma ] ) resums all glauber type multiple rescatterings of a @xmath26 pair on a nucleus . as was mentioned before , since each interaction of the pair with a nucleon in the nucleus is restricted to the two gluon exchange , the formula ( [ gamma ] ) effectively sums up all the powers of the parameter @xmath47 . or , looking at the power of the exponent in ( [ gamma ] ) we conclude that since @xmath48 , it resums all the powers of @xmath49 .
this is the definition of quasi
classical limit , a more detailed discussion of which could be found in @xcite .
since the nucleus is at rest in order to include the qcd evolution of @xmath1 structure function , we have to develop the soft gluon wave function of the incoming virtual photon . in the leading longitudinal logarithmic approximation ( @xmath50 ) the evolution of the wave function is realized through successive emissions of small@xmath0 gluons .
the @xmath26 pair develops a cascade of gluons , which then scatter on the nucleus . in order to describe the soft gluon cascade
we will take the limit of a large number of colors , @xmath51 .
then , this leading logarithmic soft gluon wavefunction will become equivalent to the dipole wave function , introduced by mueller in @xcite .
the physical picture becomes straightforward .
the @xmath26 pair develops a system of dipoles ( dipole wave function ) , and each of the dipoles independently scatters on the nucleus , as shown in fig .
[ dipole ] . since the nucleus is large , we may approximate the interaction of a dipole ( quark antiquark pair ) with the nucleus by @xmath52 given by eq .
( [ gamma ] ) , with @xmath53 and @xmath54 being the dipole s transverse separation and impact parameter .
that means that each of the dipoles interacts with several nucleons ( glauber rescattering ) in the nucleus independent of other dipoles .
the interaction of each of the dipoles with the nucleus is the same as was shown in fig .
[ glauber ] for the initial @xmath42 pair . to construct the dipole wave function
we will heavily rely on the techniques developed in @xcite .
following @xcite we define the generating functional for dipoles @xmath55 ( see formulae ( 16 ) and ( 17 ) in @xcite ) .
the generating functional then obeys the equation ( see eq . ( 12 ) in @xcite ) @xmath56 + \end{aligned}\ ] ] @xmath57 \end{aligned}\ ] ] @xmath58 where @xmath59 , @xmath60 and the integration over @xmath61 is performed over the region where @xmath62 and @xmath63 .
this @xmath64 serves as an ultraviolet cutoff in the equation and disappears in the physical quantities .
@xmath65 is the position of the center of the initial dipole in the transverse plane @xcite .
@xmath66 in the large @xmath7 limit .
the generating functional is defined such that @xmath67 [ see @xcite ] .
analogous to @xcite we now define the dipole number density by @xmath68 @xmath69 convoluted with the virtual photon s wave function gives the number of dipoles of transverse size @xmath0 at the impact parameter @xmath70 with the smallest light cone momentum in the pair greater or equal to @xmath71 .
similarly to the dipole number density we can introduce dipole pair density @xcite for a pair of dipoles of sizes @xmath72 and @xmath73 at the impact parameters @xmath74 and@xmath75 by @xmath76 our notation is different from the conventional approach of @xcite by the factor of a factorial , for reasons which will become obvious later . generalizing the definition ( [ dippair ] ) to @xmath77 dipoles of sizes @xmath78 situated at the impact parameters @xmath79
we easily obtain : @xmath80 one can now see that in order to include all the multiple pomeron exchanges one has to sum up the contributions of different numbers of dipoles interacting with the nucleus .
namely we should take the dipole number density @xmath81 and convolute it with the propagator of this one dipole in the nucleus @xmath82 .
then we should take the dipole pair density @xmath83 and convolute it with two propagators @xmath84 and @xmath85 , etc . that way we obtain an expression for @xmath86 : @xmath87 @xmath88 @xmath89 where we put the minus sign in front of @xmath90 to make it positive , since @xmath91 is negative .
( [ nn ] ) clarifies the physical meaning of @xmath90 as a total cross section of a @xmath26 pair interacting with a nucleus .
one can understand now the factorials in the definitions of the dipole number densities ( [ n1 ] ) , ( [ dippair ] ) and ( [ kdip ] ) : once the convolutions with the propagators @xmath91 are done then the dipoles become `` identical '' and we have to include the symmetry factors . in order to write down an equation for @xmath35 we have to find the equations for @xmath92 s first .
following the techniques introduced in @xcite we have to differentiate the equation for the generating functional ( [ zz ] ) with respect to @xmath93 putting @xmath94 at the end , keeping in mind that @xmath95 . differentiating formula ( [ zz ] ) once we obtain an equation for @xmath96 : @xmath97 + \end{aligned}\ ] ] @xmath98 \int_\rho d^2 \tilde{x}_2 \frac{x_{01}^2}{x_{02}^2 x_{12}^2 } \ , 2 \ , n_1 ( x_{02 } , y , { \overline { \bf b}}_1 , { \bf x}_1 ) , \end{aligned}\ ] ] where , following @xcite , we have defined @xmath99 .
differentiating eq .
( [ zz ] ) twice we obtain an equation for @xmath100 @xmath101 \int_\rho d^2 \tilde{x}_2 \frac{x_{01}^2}{x_{02}^2 x_{12}^2}\end{aligned}\ ] ] @xmath102,\end{aligned}\ ] ] where @xmath103
. now higher order differentiation of eq .
( [ zz ] ) becomes apparent , and could be easily done yielding the following equation for the number density of @xmath104 dipoles : @xmath105 \int_\rho d^2 \tilde{x}_2 \frac{x_{01}^2}{x_{02}^2 x_{12}^2}\end{aligned}\ ] ] @xmath106 , \end{aligned}\ ] ] where we anticipate the integration over the dipole sizes and treat the dipoles as identical objects . in principle eq .
( [ eqni ] ) should contain the permutations of the arguments of the gluon densities in the product on the right hand side , but for the abovementioned reason we do not write this terms explicitly .
multiplying formula ( [ eqni ] ) by @xmath107 integrating over the dipole sizes and impact parameters , and summing all such equations , i.e. summing over @xmath104 from 1 to @xmath108 in ( [ eqni ] ) one obtains the equation for @xmath109 @xmath110 + \frac{\alpha c_f}{\pi^2 } \int_0^y d y \ , \exp \left [ - \frac{4 \alpha c_f}{\pi } \ln \left ( \frac{x_{01}}{\rho } \right ) ( y - y ) \right]\end{aligned}\ ] ] @xmath111 .\end{aligned}\ ] ] equation ( [ eqn ] ) , together with equations ( [ defn ] ) and ( [ wfn ] ) , provide us with the leading logarithmic evolution of the @xmath1 structure function of a nucleus including all multiple pomeron exchanges in the large@xmath7 limit . throughout the preceding calculations we never made an assumption that @xmath6 is large .
of course it should be large enough for the perturbation theory to be applicable .
the only assumption about the incoming photon s momentum that we made was that its light
cone component @xmath17 is large , therefore we could neglect the inverse powers of @xmath17 .
this is eikonal approximation , which is natural for leading @xmath12 calculation .
however if the inverse power of @xmath17 comes with an inverse power of @xmath112 , forming something like @xmath113 we do not neglect these terms , therefore resumming all the inverse powers of @xmath6 ( `` higher twist terms '' ) . that way we proceed to conclude that equation ( [ eqn ] ) sums up in the leading logarithmic approximation all diagrams that include the effects of multiple pomeron exchanges , with pomeron ladders together with pomeron splitting vertices being incorporated in the dipole wave function . in terms of conventional ( not `` wave functional '' ) language eq .
( [ eqn ] ) resums the so called `` fan '' diagrams ( see fig .
[ fan ] ) which were summed up by conventional glr equation @xcite .
the difference between our equation and glr is that eq .
( [ eqn ] ) does not assume leading transverse logarithmic ( large @xmath6 ) approximation .
in order to reconcile our approach with traditional results in this section we will take the large @xmath6 limit of eq .
( [ eqn ] ) and show that in this double logarithmic approximation eq .
( [ eqn ] ) reduces to the glr equation @xcite .
we consider a scattering of a virtual photon , characterized by large momentum scale @xmath19 , on a nucleus at rest characterized by the scale @xmath114 .
the @xmath115 limit implies that the dipoles produced at each step of the evolution in the dipole wave function must be of much greater transverse dimensions than the dipoles off which they were produced .
basically , since in the double logarithmic approximation the transverse momentum of the gluons in the dipole wave function should evolve from the large scale @xmath19 to the small scale @xmath114 , than the transverse sizes of the dipoles should evolve from the small scale @xmath116 to the large scale @xmath117 . in the limit when the produced dipoles are much larger than the dipole by which they were produced ( large @xmath6 limit ) , the kernel of eq .
( [ eqn ] ) becomes @xmath118 where @xmath119 , and the upper cutoff of the @xmath120 integration is given by the inverse momentum scale characterizing the nucleus , @xmath121 . since this integration is done in the region of large transverse sizes the ultraviolet cutoff @xmath64 is no longer needed .
one can easily see that including virtual corrections would bring in the exponential factor @xmath122 in the eqs .
( [ zz ] ) and ( [ eqn ] ) instead of @xmath123 $ ] . in the double logarithmic approximation @xmath124 and @xmath125 , therefore @xmath126 . that way the factor of @xmath122 can be neglected . the resulting limit of eq .
( [ eqn ] ) is @xmath127 @xmath128,\end{aligned}\ ] ] which after differentiation with respect to @xmath24 yields @xmath129 , \end{aligned}\ ] ] where , for simplicity , we suppressed the difference in the impact parameter dependence of @xmath90 on the left and right hand sides of eq .
( [ dlla ] ) .
this is done in the spirit of the large cylindrical nucleus approximation .
also one should keep in mind that for this double logarithmic limit in the definition of @xmath130 given by eq .
( [ nn ] ) the integration over the dipole s transverse sizes should be also done from @xmath131 to @xmath121 .
now we have to make a connection between @xmath109 and the gluon distribution function @xmath132 of a nucleus .
@xmath35 is a forward scattering amplitude of a @xmath26 pair on a nucleus and is a well defined physical quantity .
however there is some freedom in the definition of the gluon distribution .
if one makes us of the general definition of the gluon distribution as a matrix element of leading twist operator , then an attempt to take into account higher twist operators would lead only to renormalization of their matrix elements ( see @xcite and references therein ) .
the evolution equation for @xmath13 would be linear , with all the non - linear saturation effects included in the initial conditions .
the goal of the glr type of approach is to put these non - linear effects in the evolution equation .
therefore in the double logarithmic approach one usually defines the gluon distribution function through a cutoff operator product expansion , i.e. , as a matrix element of the @xmath133 operator , with @xmath6 an ultraviolet cutoff imposed on the operator ( see the discussion on pp .
442 - 443 of @xcite ) . in the spirit of this approach
we define the gluon distribution by @xmath134 with the coefficient fixed by the two gluon exchange between the quark antiquark pair and the nucleus ( in the large @xmath7 limit ) . substituting eq .
( [ xg ] ) into eq .
( [ dlla ] ) one obtains @xmath135 ^ 2 ] .\end{aligned}\ ] ] differentiating the resulting equation with respect to @xmath136 and using @xmath137 , which is valid in the double logarithmic limit , we end up with @xmath138 ^ 2,\end{aligned}\ ] ] which exactly corresponds to the glr equation @xcite , with the factors matching those corresponding to cylindrical nucleus case in references @xcite .
one has to note that the problems with the definition of the gluon distribution function outlined above bear no consequence on eq .
( [ eqn ] ) .
this equation describes the evolution of @xmath86 in the leading @xmath12 and does not assume collinear factorization or impose transverse momentum cutoffs , therefore posing no problems like the mixing of operators of different twists @xcite .
one would also like to rederive the equation for @xmath11 derived earlier in @xcite , given by eq .
( [ glamu1 ] ) here . in the double logarithmic limit
we did not reproduce those results , given by formula ( 48 ) in @xcite , formula ( 3.19 ) in @xcite , and , most explicitly by formula ( 41 ) in @xcite .
nevertheless this equation could be obtained from eq .
( [ eqn ] ) in the following peculiar limit .
consider the scattering of a virtual photon with moderately large virtuality @xmath19 on a nucleus composed of very heavy quarks , so that the typical momentum scale characterizing the nucleus given by the mass of the heavy quarks @xmath139 is much larger than @xmath19 . in that case
the double logarithmic limit would correspond to production of small dipoles in the dipole wave function .
now , using certain freedom we have in the definition of the gluon distribution function , we can treat eq .
( [ gamma]b ) as a definition of @xmath11 , which should be substituted in ( [ gamma]b ) instead of @xmath140 . plugging the generalized eq .
( [ gamma]b ) into eq .
( [ eqn ] ) taken in the small dipole limit we obtain eq.([glamu1 ] ) . that way we will rederive the results of @xcite , however not quite for the same process as the one for which they were derived originally .
finally we note that we failed to find a limit in which eq .
( [ eqn ] ) reproduces formula ( 39 ) in @xcite , that was given by eq .
( [ alik ] ) above , which is another candidate for the double logarithmic evolution of @xmath13 including all multiple pomeron exchanges .
one of the diagrams that were not included in our analysis above is shown in fig .
[ loop ] ( a ) .
there a hard pomeron ladder splits into two and then the two ladders again fuse into one which connects to the nucleus . in general
this diagram is of the same order in coupling and rapidity as the usual two pomeron exchange diagram of fig .
[ loop ] ( b ) , and , therefore should be also considered .
the effects of that type can play an important role in unitarization and saturation of the onium
onium scattering amplitude @xcite . in the dipole wave function language
the graph in fig .
[ loop ] ( a ) can be interpreted in two ways . in one case
it could be a part of the dipole wave function , which should be redefined to include this graph in the form of color quadrupoles .
on the other hand the graph in fig .
[ loop ] ( a ) could correspond to the usual `` two dipole '' part of wave function ( @xmath141 ) with both dipoles interacting with the same nucleon in the nucleus .
however , for the given problem of scattering on a large nucleus , the graph in fig .
[ loop ] ( a ) is suppressed by powers of atomic number @xmath45 compared to the graph in fig .
[ loop ] ( b ) for both cases .
the inclusion of the pomeron fusion effects in the dipole wave function is a difficult task @xcite .
( [ zz ] ) does not take them into account .
construction of dipole wave function which includes the diagrams shown in fig .
[ loop ] ( a ) is an interesting and important problem , which is still to be solved .
one should note that inclusion of the graphs of the type shown in fig .
[ loop ] ( a ) in the dipole wave function would result in appearance of color quadrupoles @xcite . in the case of onium onium center of mass scattering this wave function can be either considered as suppressed by factors of @xmath142 compared to the one dipole wave function , or , more correctly , one may note that its contribution to the scattering amplitude will suppressed by the factor of @xmath143 @xcite , with @xmath144 the intercept of the bfkl pomeron .
however , this argument does not apply for the case when one of the onia is at rest @xcite , or to our case when the nucleus is at rest .
the graph in fig .
[ loop ] ( a ) in our case of nucleus at rest brings in suppression by powers of @xmath45 .
the reason for that is very simple : in the first diagram ( fig .
[ loop ] ( a ) ) there is only one pomeron ladder ( dipole ) interacting with the nucleus below , whereas in the second diagram ( fig .
[ loop ] ( b ) ) there are two ladders .
since each dipole is convoluted with its propagator through the nucleus , each ladder brings in a factor of @xmath145 . for large transverse size dipoles
this factor is proportional to @xmath146 and for small dipoles it scales as @xmath147 . in any case
one can see that the graph in fig .
[ loop ] ( a ) , having the same parametrical dependence on @xmath3 and @xmath24 as the graph in fig .
[ loop ] ( b ) , is suppressed by some power of @xmath45 compared to this second graph .
therefore , one should note that by neglecting the effects of the graph in fig .
[ loop ] ( a ) we assume that we are doing the leading calculation in the powers of the atomic number of the nucleus @xmath45 . that allows us to avoid complications associated with the incorporation of the diagram in fig .
[ eqn ] ( a ) in the evolution equation ( [ eqn ] ) .
( [ eqn ] ) can also be derived directly from eq .
( [ zz ] ) by putting @xmath148 in it and defining @xmath149 .
that way we have a method of resumming all multiple pomeron exchanges for any higher order corrections to the dipole kernel .
if one calculates the dipole kernel , say , at the next - to - lowest order , then we can write down an equation for generating functional @xmath150 , similar to eq .
( [ zz ] ) .
though the next - to - lowest order equation will in addition have cubic terms in @xmath150 on the right hand side . then , putting @xmath151 and @xmath152 one would easily obtain an equation resumming multiple pomeron exchanges in the subleading logarithmic approximation .
therefore dipole model provides us with a relatively straightforward way of taking into account the multiple pomeron exchanges once the one pomeron exchange contribution has been calculated . in other words
if the dipole kernel is known at any order in the coupling constant one can easily generalize the resulting equation for generating functional to include the multiple pomeron exchanges on a nucleus .
finally we note that it would be interesting to try fitting the recent hera data @xcite using the evolution of the @xmath1 structure function given by eq .
( [ eqn ] ) .
i would like to thank alex kovner , genya levin , larry mclerran and alfred mueller for a number of informative and very helpful discussions .
this work is supported by doe grant de - fg02 - 87er40328 . | we derive an equation determining the small-@xmath0 evolution of the @xmath1 structure function of a large nucleus which includes all multiple pomeron exchanges in the leading logarithmic approximation using mueller s dipole model @xcite .
we show that in the double leading logarithmic limit this evolution equation reduces to the glr equation @xcite .
16.5 cm -1 cm 22.4 cm nuc - mn-99/1-t + tpi - minn-99/05 * small@xmath0 @xmath1 structure function of a nucleus including multiple pomeron exchanges * + yuri v. kovchegov + + _ school of physics and astronomy , university of minnesota , + minneapolis , mn 55455 _ + + + |
the interchain , interplane transport of electrons in low dimensional ( quasi 1d , 2d ) materials attracts much attention @xcite in view of striking differences between longitudinal and transverse transport mechanisms revealing a general problematics of strongly correlated electronic systems . beyond the low field ( linear ) conduction , the tunnelling current - voltage j - u characteristics @xmath0 , @xmath1 are of particular importance .
the interest has been renewed thanks to recently developed @xcite design of intrinsic tunnelling devices where electronic transitions between weakly coupled chains or planes take place in the bulk of the unperturbed material .
the first feature one expects to see at any tunnelling experiment in gapful conductors is the regime of free electrons when the current onset corresponds to the voltage @xmath2 of the gap in the spectrum of electrons .
but contrarily to usual systems , like semi- or even superconductors , there is also a possibility for tunnelling within the subgap region @xmath3 .
it is related to the pseudogap ( pg ) phenomenon known for strongly correlated electrons in general , well pronounces in quasi 1d systems and particularly in cases where the gap is opened by a spontaneous symmetry breaking ( see @xcite and refs . therein ) .
the pg is originated by a difference , sometimes qualitative , between three forms of electronic states : a ) short living excitations which are close to free electrons , b ) dressed stationary excitations of the correlated systems , and c ) added particles which modify the ground state itself @xcite . for our typical examples of electrons on a flexible lattice , the modification results in self - trapped states ( b ) like single particle @xmath4 polarons with energies @xmath5 below the single electron ( a ) activation energy @xmath6 ; then the new gap @xmath7 will be observed as a true threshold with the pg in between .
there may be also contributions of two - particle @xmath8 states ( c ) - bipolarons , which energy gain per electron is larger than for polarons @xmath9 .
while the cases ( a , b ) are common for low symmetry and discrete symmetry cases , for ( c ) there is a further drastic effect of a spontaneous symmetry breaking like the case of the polyacethylene @xmath10 or of some doubly commensurate cdws .
now the bipolarons are decoupled into particles with a nontrivial topology , solitons or kinks , changing the sign of the order parameter of the dimerization .
the situation is further intricate in systems with a continuous gs degeneracy like incommensurate charge density waves ( icdw ) or wigner crystals .
here even the self - trapping of a single electron is allowed to lead to topologically nontrivial states , the amplitude solitons ass . in the same class
we find a more common case of acoustic polarons in a 1d semiconductor @xcite .
properties of systems with different types of the gs degeneracy , and required theoretical approaches , are quite different . here
we shall concentrate on systems with a discrete , precisely double , degeneracy which also include most basic elements of non - degenerate systems .
theoretically , the tunnelling in cdws was studied in details for regimes of free electrons @xcite when the current onset corresponds to the voltage @xmath2 of the gap in the spectrum of electrons .
we shall consider the tunnelling in the pg regime .
we shall follow the method @xcite developed for studies of single particle spectral density @xmath11 in applications to pes and arpes intensities .
we refer to this publication for details in techniques and literature . a word of notations . in the following we shall invoke many quantities with the dimension of energy ( or frequency , since we shall keep @xmath12 ) which will be classified according to different characters ( with indices ) .
@xmath13 and @xmath14 will be the external voltage difference for tunnelling and the external frequency for pes or optics .
@xmath15 will always stay for electronic eigenvalue in a given potential , negative values will be addressed explicitly as @xmath16 .
@xmath17 will be branches of a total energy ( of deformations together with electronic energies ) supporting eigenstates @xmath18 which may be filled with occupation numbers @xmath19 .
@xmath20 will be total energies of stationary states , that is @xmath21 .
@xmath22 will be the frequency of a collective mode ( phonons specifically to cdws ) which interaction with electrons is responsible for their self - trapping .
the collective deformation @xmath23 will also be measured as the potential energy experienced by electrons .
we shall keep the electron charge @xmath24 hence potentials will be measured as energies and the interchain current @xmath25 will have the dimension _ number - of - particles / unit - time / unit - length_. the indices @xmath26 will number coupled chains ; indices @xmath27 will number moments @xmath28 of time for virtual processes .
a possibility of tunnelling or of other excitations within the gap in spectra of free electrons @xmath29 is related to a more general phenomenon of the pseudogap pg . for electrons ,
the pg signifies the remnants of the spectral density @xmath30 , or the integrated one @xmath31 , at @xmath32 where @xmath33 is the absolute boundary of the spectrum .
@xmath33 is the energy of a fully dressed state of one electron interacting with other degrees of freedom .
( there may be totally external modes like deformations or polarizations for usual polarons , external modes essentially modified by the bath of electrons like in cdws , internal collective modes of electronic system itself like in sdws . ) most commonly , the self - trapped state of one electron is known as the `` polaron '' while more complex objects , solitons , can appear for systems with continuously degenerate gss ( see @xcite for a review ) .
the functions @xmath34 and @xmath30 are measured directly in pes and arpes experiments ( these abbreviations stay for the integrated photo - emission spectroscopy and for the angle ( that is momentum ) resolved one ) . as such
they have been studied theoretically for the pg region by the present authors @xcite and we refer to these publications for a more comprehensive discussion and for the literature review .
the one electron spectra can be accessed also in traditional external tunnelling experiments : junctions or stm . for the last case , and
practically for macroscopic point junctions , only the integrated @xmath34 is measured .
elements of the full dependence @xmath30 become necessary to describe strongly anisotropic materials ( layered quasi-2d or chain quasi-1d ones ) where the coherent tunnelling is realized in internal junctions of `` mesa '' type devices @xcite . here
the tunnelling goes between adjacent layers within the single crystal of the same material , hence the momentum is preserved . in a simpler version ,
the internal subgap tunnelling takes place from free electrons of some metallic bands or pockets to polaronic states within gapful spectra which probably takes place in @xmath35 @xcite .
otherwise it measures actually the joint spectral density for creation a particle - hole excitation at adjacent chains ( the interchain exciton ) . in this respect
it will be instructive to compare the coherent tunnelling and the subgap optical absorption oa ( see a short excursion and references in @xcite , iii.e ) .
a less expected version of the internal tunnelling is a possibility for bi - electronic transfers ( tunnelling of bipolarons or of kink - antikink pairs ) which usually is attributed only to superconductors .
we shall see that these processes extend the pg further down to even lower voltages . in any case , the tunnelling current @xmath0 is given by the transition rate of electrons between two subsystems @xmath36 kept at the potential difference @xmath37 . for a weak coupling @xmath38 , the electron tunnelling from @xmath39 to @xmath40 is given by the convolution of spectral densities @xmath41 if the momentum is conserved , or of their integrals @xmath42 for the incoherent tunnelling .
( everywhere we assume @xmath43 . )
recall that for free electrons with a spectrum @xmath44 we have @xmath45 while @xmath34 becomes the dos @xmath46 , e.g. for @xmath47 @xmath48 near the bottom of the free band where the electron effective mass is @xmath49 .
consider briefly the case where one of reservoirs , say @xmath50 , is composed by free electrons with a known dos @xmath51 .
one of applications of a tunnelling between the free spectrum and the pg may be the case of several families of conjugated polymers ( polypyrolle , polythiophene ) where origins of filled , @xmath52 or empty , @xmath53 bands below and above the gap are essentially different .
then the polaronic effect , hence the pg , may be pronounced only for one type of particles : electrons or holes .
the same concerns 1d systems made of semiconducting wires where both effective masses and deformation potentials for electrons and holes are usually very different .
then , for the incoherent tunnelling , @xmath54 gives directly either the tunnelling current @xmath55 ( if @xmath56 has a sharp peak at the fermi level , which is typical for using junctions with superconductors ) or the tunnelling differential conductance @xmath57 ( if @xmath58 at the fermi surface ) . below we shall be mostly interested in systems with the charge conjugated symmetry ( or qualitatively equivalent ones ) ; the examples are carbon nanotubes , symmetric conjugated polymers like polyphenylenes , polyanilines and polymers where the gap is formed ( partly at least ) by the spontaneous symmetry breaking : the polyacethylenes ( @xcite ) .
numerical details will be presented for the last rich case . in all these cases
the pg will exist near both rimes @xmath59 of the free excitation gap @xmath60 .
recall now some known results for @xmath34 within the pg @xcite .
it has the form @xmath61 where the action @xmath62 is proportional to the big parameter of our adiabatic approximation : @xmath63 .
@xmath64 is determined by an optimal fluctuation localized in space and time ( an instanton ) which supports the necessary split - off local level @xmath65 . in principle
, the prefactor @xmath66 also depends on @xmath67 and may show power law dependencies near extremals @xmath68 .
but within constraints of the adiabatic approximation @xmath69 the dependence @xmath70 is negligible in comparison with the one of @xmath64 .
the characteristic value of @xmath71 may be important for estimates of the overall magnitude of observable effects .
thus for the single particle integrated intensity @xmath72 and @xmath73 for the momentum resolved intensity ; here @xmath49 is the effective electron mass @xmath74 .
appendix [ app:1 ] contains derivation of the prefactor specifically for the tunnelling processes . in limiting cases we have @xcite \1 .
near the entry to the pg , just below the free edge @xmath75 : @xmath76 . \label{nh}\ ] ] \2 .
near the low end of the pg , just above the true spectral boundary @xmath33 : @xmath77 .
\label{nl}\]]the total dependence @xmath34 and the values of numerical constants in the above limiting laws , can be determined approximately @xcite with the help of the instanton techniques simplified by the zero dimensional reduction ( the anzats of an effective particle which we shall recall and extend below ) .
the resulting curve is plotted at the figure [ fig : s4pol ] .
moreover , the regime ( 1 . ) can be mapped exactly @xcite upon the problem of a particle in a quenched random uncorrelated potential which here is created by instantaneous quantum fluctuations of the media .
the known exact solution @xcite provides the reference value of the coefficient in the exponent of ( [ nh ] ) , from which our approximate value differs only by @xmath78 @xcite . recall for comparison the usual regime @xmath79 of the allowed tunnelling which is dominated by free electronic states .
the current of the coherent tunnelling between chains @xmath36 is given as @xmath80with @xmath81 and @xmath82 being the bloch functions .
it is instructive to compare the interchain tunnelling probability with the on chain interband optical absorption oa when the matrix element of density @xmath83 changes to the one of the momentum : @xmath84 . in both cases
the @xmath85 pair is created and the same spectral densities are involved .
the difference is in matrix elements : the oa takes place between states of opposite parity while the tunnelling requires for the same parity .
the on - chain oa between edges @xmath86 of the free gap is known to be allowed since the parity of states near @xmath86 is opposite , hence @xmath87 is finite and the oa intensity as a function of frequency @xmath88 rises as @xmath89 .
but for the same reason , the tunnelling matrix element between identical chains is prohibited at @xmath90 and the tunnelling will show only a weak edge onset @xmath91 .
nevertheless , in many cases of gaps opened due to spontaneous dimerization , the neighboring chains tend to order in antiphase .
now the shift by half a period permute states with @xmath92 then the parity of states near opposite rims @xmath86 at neighboring chains is equal , the tunnelling becomes allowed and the usual singularity is restored : @xmath93 .
going down into the pg @xmath94 , the above analysis applies to the on - chain optics but changes drastically for the interchain tunnelling .
the tunnelling will be studied in details below , here we shall only mention in advance an effect of spatial incoherence of optimal quantum fluctuations at different chains which removes completely the constraints of orthogonality .
the case of the on - chain oa can be analyzed briefly already here .
the oa is given by the convolution of two fast decaying functions of the energy @xmath95
\left\vert \lambda ( \omega_{1},\omega_{1}-\omega)\right\vert^{2}d\omega_{1}\sim \nonumber\\ \lambda^{2}(\omega /2,-\omega /2)\left(s^{\prime \prime}\right)^{-1/2 } \exp \left [ -2s(\omega /2)\right]= \nonumber\\ \left(s^{\prime \prime}(\omega /2)\right)^{-1/2 } i^{2}(\omega /2);\
s^{\prime \prime}=\frac{d^{2}s}{d\omega^{2}}. \label{ioa}\end{aligned}\]]here we have used that for the convex function @xmath64 , as given by ( [ nh],[nl ] ) , the minimum of the expression @xmath96 lies at the middle @xmath97 . at this point the electron levels @xmath65 and @xmath98 are placed symmetrically , wave functions have opposite parity , hence @xmath99 is finite .
this is the case of typical peierls insulators .
but for systems where the basis wave functions of valent and conductive bands have the same parity ( the dipole oa is not allowed ) , @xmath100 and we have to consider in ( [ ioa ] ) the deviations from the symmetry condition .
now @xmath101 and the saddle point integration in ( [ ioa ] ) gives another factor of @xmath102 which is small as @xmath103 .
we arrive at the answer similar to ( [ ioa ] ) but with the small prefactor @xmath104 .
until now we did not consider the dependencies on the momentum @xmath105 .
in the full range of @xmath67 and @xmath105 , the spectral function @xmath106 has a rich structure which can be tested in the arpes experiments . in observable quantities
, the momentum dependence appears twice : via the matrix element @xmath107 and via the action @xmath108 .
the analysis is simplified for the regime 2 : the low polaron boundary @xmath109 . here
the action dependence on @xmath88 and @xmath105 comes through the single variable @xmath110 where @xmath111 , is a heavy mass of the polaron center motion .
this kinetic energy contribution can be neglected in compare to the matrix element dependence on @xmath105 which confines @xmath112 within the characteristic momenta distribution @xmath113 of the wave function @xmath114 of the self - trapped electronic state localized over the scale @xmath115 : beyond @xmath116 , the function @xmath107 falls off exponentially .
( at this scale , the recoil kinetic energy @xmath117 is small in compare to the energy width @xmath118 .
then the final integration over @xmath105 affects only @xmath107 and gives a constant factor @xmath119 . )
altogether we find for tunnelling just the law ( [ nl ] ) with @xmath120 . in the regime 1 .
, near the free edge , the states are shallow @xmath121 and extended @xmath122 .
the effective mass @xmath123 for the center of motion becomes light , energy dependent @xmath124 but the characteristic energy scale of the form factor @xmath125 is still small in comparison with the characteristic energy width @xmath126 of ( [ nh ] ) .
so again we integrate separately the factor @xmath127 to obtain an additional prefactor @xmath128 for the tunnelling law ( [ nh ] ) with @xmath120 .
recall that for the arpes with independent variations of @xmath88 and @xmath105 , their interference may lead to rather unexpected and potentially observable phenomena ( @xcite , section iii.d ) .
one of them is the `` quasi spectrum '' : the intensity maximum over the line @xmath129 within the pg @xmath130 ( @xcite , section iii.d , case b1 , eq.49 ) .
another effect is the emergence of instantons at high @xmath105 within the domain of free electron region @xmath131 leading to the enhanced intensity within the band @xmath132 ( @xcite , section iii.d , case b3 , eq.51 ) .
we shall follow the adiabatic method of earlier publications @xcite assuming a smallness of collective frequencies @xmath133 in compare with the electronic gap : @xmath134 .
now , electrons are moving in a slowly varying potential @xmath135 , so that at any instance @xmath136 their energies @xmath137 and wave functions @xmath138 are defined from a stationary schrodinger equation @xmath139 ( eq . ( [ dirac ] ) below will give an example ) .
the hamiltonian @xmath140 depends on the instantaneous configuration @xmath135 so that @xmath141 and @xmath142 depend on time only parametrically .
exponentially small probabilities which we are studying here are determined by steepest descent paths in the joint space @xmath143 $ ] of configurations and the time , that is by a proximity of the saddle point of the action @xmath144 .
it is commonly believed , in analogy with the usual wkb , that the saddle point , the extremum of @xmath144 over @xmath145 and @xmath136 , lie at the imaginary axis of @xmath136 so that , as usual , we shall assume @xmath146 and correspondingly @xmath147 since now on .
consider the system of two weakly coupled chains @xmath26 which are put at the electric potential difference @xmath37 .
the system is described by the total action @xmath148where @xmath149 $ ] are single chain actions and the term @xmath150 describes the interchain hybridization of electronic sates .
@xmath151 are operators of electronic states.the average transverse current is given by the functional integral @xmath152~ it_{\perp } ( \hat{\psi}_{a}^{\dag}(x , t)\hat{\psi}_{b}(x , t)- \hat{\psi}_{b}^{\dag}(x , t)\hat{\psi}_{a}(x , t))\exp [ -s_{ab } ] } { \int dx d[\delta_{j}(x , t)]\exp [ -s_{ab}]}. \label{j}\end{aligned}\ ] ] we consider first the processes originated by the transfer of one electron between the chains .
they appear already in the first order of expansion of the exponent in ( [ s - total ] ) in powers of @xmath153 , which contribution to the current ( [ j ] ) can be written as @xmath154 \int d(x - y)\int d(\tau_{1}-\tau_{2 } ) \\ \nonumber [ \psi_{a}^{\ast}(x,\tau_{1})\psi_{b}(x,\tau_{1 } ) \psi_{a}(y,\tau_{2})\psi_{b}^{\ast}(y,\tau_{2 } ) \exp(-s(\tau_{1}-\tau_{2},\delta_{j}(x , t ) ) ) , \label{j1}\end{aligned}\]]where the normalizing factor @xmath155 is the denominator in ( [ j ] ) taken at @xmath156 . here the time dependent action @xmath157 describes ( in imaginary time ) the process of transferring one particle from the doubly occupied level @xmath158 of the chains @xmath39 to the unoccupied level @xmath159 of the chain @xmath40 at the time @xmath160 and the inverse process at the time @xmath161 .
we have @xmath162+\int_{\tau_{1}}^{\tau_{2}}dt[(l_{a}(-1)+l_{b}(1)-u ] \nonumber\\ & = & \int_{-\infty}^{\infty}dt[l_{a}(0)+l_{b}(0)]+ \int_{\tau_{1}}^{\tau_{2}}dt[(e_{b}+e_{a}-u ] , \label{s(1,2)}\end{aligned}\]]where @xmath163,\nu)$ ] are lagrangians of the @xmath164-th chain with the number of electrons changed by @xmath165 . they are given as a sum of the kinetic term and the potential @xmath166 : @xmath167~;\ v_{\nu}=v_{0}+|\nu |e . \label{l(nu)}\ ] ] here the potential term @xmath166 contains the energy of deformations and the sum over electron energies in filled states @xmath168 which include both the vacuum states and the split off ones : @xmath169= \int dx\frac{\delta^{2}}{2g^{2}}+\sum_{e_{\alpha}<e_{f}}e_{\alpha}[\delta ( x , t)]-w_{gs } \label{v(nu)}\ ] ] ( here @xmath170 is the coupling constant ) .
@xmath166 is counted with respect to the gs energy @xmath171 so that in the non perturbed @xmath172 state @xmath173 ( the particle , electron for @xmath174 or hole for @xmath175 , added instantaneously to the non deformed gs is placed at the lowest allowed energy , the gap rim @xmath75 ) .
the exact extremal ( saddle point ) trajectory is defined by equations @xmath176 actually the explicit calculation of the action requires for approximations .
we shall follow a way @xcite of the zero dimensional reduction which reduces the whole manyfold of functions @xmath177 to a particular class @xmath178\rightarrow s[e_{j}(t),x_j(t ) ] \label{reduction}\ ] ] of a given function @xmath179 of @xmath180 ( relative to a time dependent center of mass coordinate @xmath181 ) .
@xmath182 is parameterized by a conveniently chosen ( see @xcite for examples ) parameter for which a universal and economic choice is the eigenvalue @xmath137 .
the requirement for the manyfold @xmath183 is that it supports a pair of eigenvalues @xmath18 split off inside the gap @xmath184 which span the whole necessary interval .
the last simplification is to assume , in the spirit of all approaches of optimal fluctuations @xcite , that the potential supports one and only one pair of localized eigenstates @xmath185 .
explicit formulas for the peierls case are given in the appendix [ app:2 ] .
recall that for the oa problem we deal with one chain characterized by one pair of functions @xmath141 and @xmath186 . but for the interchain tunnelling , the functions @xmath137 at chains @xmath26 are not obliged to be identical and also the wells may be centered around different points @xmath181 .
within such a parametrization the variational equation in ( [ dd , d1,d2 ] ) yields the equation of motion for @xmath141 @xmath187 where @xmath188 are the hamiltonians which must be constants within each interval of integration in ( [ s(1,2 ) ] ) . apparently , at the outer intervals @xmath189,@xmath190 @xmath191 to provide the return to the gs with @xmath192 at @xmath193 .
at the inner interval @xmath194 @xmath195 to preserve the continuity of velocities @xmath196 at @xmath197 . since the values @xmath198 are determined uniquely by the equation of motion at the outer intervals , then @xmath198 coincide for both @xmath26 , hence @xmath199 and the functions @xmath137 become identical at any time @xmath200 .
( still , the shapes are allowed to be shifted by different centers @xmath181 : @xmath201 ) .
finally the extremal conditions ( [ dd , d1,d2 ] ) with respect to impact times @xmath202 in ( [ s(1,2 ) ] ) yield @xmath203 the action is finite @xmath204 , hence the transition probability is not zero , only for a closed trajectory , that is at presence of a turning point ( as examples , see figures [ fig : u12],[fig : u14],[fig : u18],[fig : u16+c ] in the appendix [ app:2 ] ) .
there must be a minimal value of @xmath205 where @xmath206 hence @xmath207 and @xmath208 .
the last condition requires for @xmath209 that is for @xmath210 which determines the threshold voltage at twice the polaron energy .
we arrive at the effective one chain problem with the doubled effective action .
the extremal tunnelling action is @xmath211 which is twice the exponent appearing in the spectral density @xmath212 with limiting laws ( [ nh],[nl ] ) .
the full expression is @xmath213 we obtain a final expression for the current after integration over @xmath177 around the extremal taking into account the zero modes related with translations of the instanton centers positions @xmath181 .
( details of calculations are given in the appendix [ app:1 ] ) @xmath214 , \label{pre0}\]]where @xmath215 is the fourier transforms of the wave functions @xmath114 , the time @xmath216 is defined as @xmath217 .
the mean fluctuational displacement @xmath218 of the center of mass between the impact moments is given as @xmath219 where @xmath220 is the translational mass : @xmath221 note that the prefactor in eq .
( [ pre0 ] ) , which is the matrix element between orthogonal states @xmath222 and @xmath223 , is always nonzero due to the integration over zero modes @xmath181 ( in contrast to results for the rigid lattice where it obeys the selection rules ) ; see more in the appendix [ app:1 ]
. comparing with the pes intensity @xmath34 calculated in @xcite we see that , up to pre - exponential factors , the tunnelling current is proportional to the square of the pes intensity @xmath212 : @xmath224 .
e.g. near the threshold @xmath225 we can write @xmath226 \exp \left [ c_{2}\frac{(u-2w_{1})}{g\omega_{0}}\log \frac{2c_{3}\delta_{0}}{(u-2w_{1})}\right ] .
\label{j1low}\ ] ] the coefficients @xmath227 can be found numerically from ( [ s(u ) ] ) as ( for the peierls model ) @xmath228 , @xmath229 , @xmath230 .
( these values differ from the corresponding ones in @xcite because of different normalizations of frequency @xmath133 in compare to @xmath231 ) .
it is known that the joint self - trapping of two electrons allows to further gain the energy resulting in stable states different from independent polarons . in general nondegenerate systems
this is the bipolaron , confined within the length scale twice smaller than that of the polaron , the energy gain of the bound state @xmath232 is four times that of the polaron and the total energy gain of the bipolaron @xmath233 is also four times that of two polarons .
( certainly these results neglect the energy loss due to the coulomb repulsion which may become critical for the stability of a shallow bipolaron . ) the same time , the total energy of one bipolaron @xmath234 is larger than the energy of one polaron @xmath235 and even than the free electron energy @xmath75 .
this is why bipolarons can not be seen as thermal excitations while they are favored in case of doping .
the information on their existence comes from the ground state of doped systems where bipolarons are recognized by their spinless character and special optical features ( see @xcite for experimental examples on conducting polymers and @xcite for relevant theoretical models ) .
an important advantage of tunnelling experiments is a possibility to see bipolarons directly , at voltages @xmath37 below the two - polaron threshold @xmath236 that is within the true single particle gap .
this possibility comes from the fact that , for bipolarons as particles with the double charge @xmath237 , the voltage gain by transferring from one chain to another is @xmath238 , hence the threshold will be at @xmath239 .
the probability of the bi - electron tunnelling is small as it appears only in the higher order @xmath240 in interchain coupling .
but it can be seen as extending below the one - electron threshold where no other excitations can contribute to the tunnelling current .
the bi - electronic contribution to the current can be written , by expanding ( [ s - total ] ) and ( [ j ] ) , as @xmath241d[\delta_b]\int \prod_{i=1}^{3}dy_{i}d\tau_{i}\exp \left(-s(\tau_{i})\right ) \nonumber\\ & & \left [ \psi_{a}^{\ast}(x,\tau)\psi_{b}(x,\tau)\psi_{a}^{\ast}(y_{1},\tau_{1 } ) \psi_{b}(y_{1},\tau_{1})\psi_{b}^{\ast}(y_{b},\tau_{2 } ) \psi_{a}(y_{2},\tau_{2})\psi_{b}^{\ast}(y_{3},\tau_{3})\psi_{a}(y_{3},\tau_{3 } ) -\{2\leftrightarrow 3\}\right ] , \nonumber\end{aligned}\]]which generalizes expressions ( 7 ) and ( 8) for the one electron tunnelling . here
+ \nonumber\\ & & \left\{\int_{\tau}^{\tau_{1}}+\int_{\tau_{2}}^{\tau_{3}}\right\ } dt[(l_{a}(1)+l_{b}(1)-u]+\int_{\tau_{1}}^{\tau_{2}}dt[l_{a}(2)+l_{b}(2)-2u ] .
\label{2e}\end{aligned}\ ] ] within our model ( [ l(nu)],[v(nu ) ] ) the potentials @xmath243 are additive in energy @xmath65 , then the action can be simplified as @xmath244 the extremal solution is defined , as above , by equations of the type ( [ dd , d1,d2 ] ) but with four impact times @xmath28 instead of two .
( actually , in view of the time reversion symmetry , the number of boundary conditions is twice smaller . )
a similar analysis of the extremal solution shows that optimal fluctuations @xmath177 are identical in shape , up to shifts of their centra : @xmath181 , @xmath245 .
hence the energies are identical @xmath246 , and also the resonance conditions @xmath247 take place at the impact moments @xmath28 .
moreover , the simple hierarchy of our model @xmath248 shows that all branches @xmath17 cross at the same point @xmath249 ( see figures [ fig : u12],[fig : u14],[fig : u18 ] below ) .
then the evolution @xmath141 switches directly from the branch @xmath250 to the branch @xmath8 and back , without following the intermediate branch @xmath4 .
it means that the intervals @xmath251 and @xmath252 of one - electron transfers @xmath4 are confined to zero : @xmath253 .
in other words , only processes of simultaneous tunnelling of pairs of particles are left .
notice that this picture changes in more general models , particularly taking into account important coulomb interactions .
they add , to the energy branch of a shallow bipolaron , the energy @xmath254 where @xmath255 is the dielectric susceptibility of the media in the interchain direction , @xmath256 is the localization length of @xmath257 , such that @xmath258 . now the intermediate intervals @xmath259,@xmath260 appear where the evolution follows the @xmath4 branches , see figure [ fig : u16+c ] . with increasing coulomb interactions this single particle interval becomes more pronounced and the bipolaronic threshold is shifted towards the one of two independent polarons . in any case , the extremum solution for the action ( [ 2e ] ) is achieved on the instanton trajectory given be the equation @xmath261 the extremal action is @xmath262 this action is finite if the turning point @xmath263 does exist , that is if @xmath264 .
notice that , neglecting coulomb interactions , the energy @xmath166 is determined only by the total number @xmath265 of electrons and holes .
then the energy of the bipolaron ( both @xmath266 and @xmath267 are either empty or doubly occupied ) and the energy of the exciton ( both @xmath268 and @xmath269 are singly occupied ) are the same .
then the trajectory of the bi - electronic tunnelling becomes the same as the one for the case of optical absorption @xcite , only the action is doubled @xmath270 . up to the pre - exponential factor
we have @xmath271^{2},\]]where @xmath272 is the optical absorption probability for one chain . for common systems with a nondegenerate ground state ,
the dependence @xmath273 resembles qualitatively the law [ j1low ] for the one - electron contribution , with a similar behavior near the threshold @xmath274 .
the situation changes for a doubly degenerate ground state where the bipolaron dissolves into a diverging pair of solitons ( dimerization kinks ) .
thus for the peierls model the evaluation of ( [ s2(u ) ] ) gives , similar to the oa law of @xcite , near the two particle threshold @xmath275 with @xmath276 .
the overall dependence for the @xmath277 is shown at the figure [ fig : s4sol ] . here
we see explicitly that in the order @xmath240 the threshold voltage @xmath278 is smaller than @xmath279 obtained in the order @xmath280 .
therefore this is the main contribution to the current in the region @xmath281 .
figure [ fig : s4sol ] shows that the dependence of @xmath282 near the low @xmath37 onset is much sharper than that of @xmath0 at the figure [ fig : s4pol ] near the polaronic onset which corresponds to the higher singularity in the limiting formula ( [ j2min ] ) in compare to ( [ j1low ] ) .
in quasi 1d systems with a gapful electronic spectrum , the interchain tunnelling ( as well as pes or oa ) can be used to test virtual electronic states within the pseudogap . due to the interaction of electrons with a low frequency mode , phonons in our examples ,
the tunnelling is allowed in the subgap region @xmath283 which forms the pseudogap .
the one electron processes lead to universal results similar both for systems with the build - in gap and for those where the gap is due to the spontaneous breaking of a discrete symmetry .
the pg is entered with the law ( [ nh ] ) and continues down to the threshold @xmath284 , approached with the law ( [ nl ] ) .
this threshold corresponds to the interchain transfer of fully dressed particles : polarons with the energies @xmath33 .
but in tunnelling the pg is stretched even further down thanks to processes of a simultaneous tunnelling of two electrons .
it terminates at the lower threshold @xmath285 or @xmath286 , @xmath287 . here
@xmath288 is the energy of the bipolaron - a bound state of two electrons selftrapped together . in degenerate systems the bipolaron dissolves into unbound solitons , hence the threshold at @xmath289 with a more pronounced dependence of the tunnelling rate ( [ j2min ] ) as well of the oa .
numerical results are presented at figures [ fig : s4pol],[fig : s4sol ] .
there is an important difference between subgap processes and the usual overgap transitions at @xmath290 of free electrons in a rigid system .
it comes , beyond intensities , from different character of matrix elements . actually within the pg region
there are no particular selection rules since the wave functions of virtual electronic states split off within the gap are localized having a broad distribution of momenta . then the pg absorption is allowed independent on the interchain ordering .
contrarily , the regular tunnelling across the free gap shows an expected dos singularity @xmath291 for the out of phase interchain order while for the in - phase order the threshold is smooth @xmath292 .
this difference may be important to choose an experimental system adequate for studies of pgs .
the smearing of the free edge singularity is a natural criterium for existence of the pg below it @xcite .
but the total absence of this strong feature in systems with forbidden overgap transitions can allow for a better resolution of the whole pg region , down to the absolute threshold .
probably a very smooth manifestation of gaps in usual tunnelling experiments on cdws @xcite , while the gaps show up clearly through activation laws , is related to this smooth crossover from the overgap to the subgap region .
( notice that the existing experiments refer mostly to icdws which , with their continuous degeneracy of the gs , must be studied specially which is beyond the scope of this article . )
finally we shall discuss relations with other theoretical approaches .
most theories of tunnelling , see @xcite , keep the following assumptions : i. they refer to the overgap region where interactions or fluctuations are not important and usually are not taken into account .
they refer to the incoherent tunnelling , local in space , which is a usual circumstance of traditional experiments .
the pg in tunnelling was considered by monz et al in @xcite in the framework of the approach @xcite .
this method became popular recently in theories of the pg thanks to its easy implementation : it is sufficient to average results for a rigid system over a certain distribution of the gap values .
apparently this is the way to describe an average over a set of measurements performed on similar systems with various values of the gap , e.g. manipulating with the temperature , the pressure or a composition .
but actually , as we could see above , the pg is formed by fluctuations localized both in space and time , the instantons , with localization parameters depend on the energy deficit being tested .
there is an intermediate approach applied @xcite to a complex of the pg phenomenon from optics to conductivity and susceptibility .
it treats fluctuations as an instantaneous disorder due to quantum zero point fluctuations of the gap .
indeed , this picture can be well applied , as it was done already in @xcite , but only to dynamical processes and only in the upper pg region , just below the free gap @xmath293 , which leads to the law ( [ nh ] ) .
but deeper within the pg , the fluctuations are not instantaneous : they require for an increasingly longer time and become self - consistent with the measured electronic state leading to another law and to appearance of the lower threshold .
generalizations and deeper analysis of the model of the instantaneous disorder lead to interesting theoretical studies @xcite , but their applicability is very limited unless the variable time scale is realized as we have demonstrated in this and preceding articles . our approach can be compared to the work @xcite on the fluctuational creation of pairs of phase solitons in a 1d commensurate cdw under the longitudinal electric field .
but in our case me deal , in effect , with the interchain tunnelling of pairs of solitons under the transverse field ; also the solitons have a more complex character of a multielectronic origin . in conclusion ,
the presented and earlier @xcite studies recall for the necessity of realizing the variable time scale of subgap processes both in theory and in diverse interpretations of different groups of experiments ( dynamic , kinetic , thermodynamic ) which address excitations with very different life times .
s. m. acknowledges the hospitality of the laboratoire de physique thorique et des modle statistiques , orsay and the support of the cnrs via the ens - landau foundation .
we consider the system of weakly coupled dimerised chains .
each chain is described by a usual electron - phonon hamiltonian ( peierls , ssh models ) .
electron levels @xmath65 and wave functions @xmath294 are determined by equations @xmath295\mathbf{\psi}=e\mathbf{\psi } , \label{dirac}\ ] ] where @xmath296 are the pauli matrices , @xmath297 , @xmath298 are the components of electron wave functions near fermi points @xmath299 , and the real function @xmath182 is the amplitude of the alternating dimerization potential .
the ground state of each chain is the peierls dielectric with the gap @xmath300 .
the electron spectrum has the form @xmath301 ( in the following we shall put the fermi velocity @xmath302 , @xmath303 and , as everywhere , the plank constant @xmath304 ) . the excited states are solitons ( kinks ) , polarons and bi - solitons ( kink - antikink pairs ) which are characterized by electron levels localized deeply within the gap ( see the review @xcite ) . the one parametric family of configurations @xmath305 supporting the single split - off pair of levels @xmath18 can be written as @xmath306 evolving from a shallow potential well at @xmath307 through the stationary configuration for a polaron @xmath4 to the pair of diverging kinks at @xmath308 as shown at the figure [ fig : shapes ] .
the potentials @xmath166 ( for the @xmath65 level filling @xmath165 ) as functions of @xmath65 are given as @xmath309 the translational mass can be found as @xmath310.\ ] ] consider the matrix element between levels @xmath18 in the peierls state .
the wave function has two components @xmath311 according to @xmath312 .
explicit expressions for split - off states are @xmath313 .
the equation for the bound eigenstate ( [ dirac ] ) shows the following symmetry : @xmath314 , @xmath315 , @xmath316 .
then @xmath317 , @xmath318 , with @xmath319 , which demonstrates explicitly the orthogonality of @xmath257 and @xmath320 .
the matrix element in eq .
( [ pre0 ] ) becomes @xmath321 . at @xmath90 , @xmath322 ,
hence for identical chains the transition at the free gap @xmath323 is forbidden which removes the singularity at the gap threshold in a rigid system .
but the true threshold at @xmath236 for the subgap absorption or tunnelling are not subjected to this selection rule since the wave functions of localized states associated to the optimal fluctuation are distributed over the momentum region @xmath324 .
figure [ fig : shapes ] shows exact shapes @xmath183 of the equilibrium polaron ( upper thick line ) and of a well formed @xmath325 pair of solitons ( lower thick line ) .
thin lines show exact shapes of optimal fluctuations necessary to create these states by tunnelling .
notice the much less pronounced shapes for optimal fluctuations in compare to the final states which facilitates the tunnelling . of the equilibrium polaron , @xmath326 ( upper thick line ) and of a
nearly formed ( @xmath327 ) pare of solitons ( lower thick line ) .
thin lines show exact shapes of optimal fluctuations necessary to create these states by tunnelling.,height=207 ] figure [ fig : v1 ] plots the total energy @xmath328 of the single particle branch as a function of the associated energy of the bound state .
@xmath329 , @xmath330 corresponds to the particle added to the unperturbed ground state , at the bottom of the continuous spectrum .
@xmath331 is the mid - gap state reached for the limit of two divergent solitons when the total energy approaches the maximal value @xmath332 . in between , at @xmath333 , @xmath334
, the minimum corresponds to the stationary polaronic state . the short thin vertical line between plots @xmath65 and @xmath328 points to the configuration ( upper thin curve at the figure [ fig : shapes ] ) of the fluctuation necessary for tunnelling to the polaron ( the minimum of @xmath328 , upper thick curve at the figure [ fig : shapes ] ) .
as a function of the energy @xmath65 of the associated bound state.,height=207 ] next three figures plot the total energies @xmath335 for different branches as a function of the energy @xmath65 of the associated bound state ( all in units of @xmath75 ) .
branches are distinguished by their ordering at @xmath329 .
figure [ fig : u12 ] corresponds to the potential @xmath336 which is below the bi - electronic threshold ; no branch is crossing @xmath337 axis , hence no final action is allowed and the current is zero .
for different tunnelling branches @xmath335 as a function of the energy @xmath65 of the associated bound state .
this figure corresponds to @xmath336 which is below the bi - electronic threshold.,height=207 ] figure [ fig : u14 ] corresponds to the potential @xmath338 which is between the bi - electronic threshold @xmath339 and the polaronic one @xmath340 ; the bi - electronic branch crosses the axis @xmath337 at the point @xmath263 , the action is finite , hence a nonzero tunnelling of two electrons is allowed . as a function of the energy @xmath65 of the associated bound state
this figure corresponds to @xmath338 which is between the thresholds for tunnelling of bipolarons and polarons.,height=207 ] figure [ fig : u18 ] corresponds to the potential @xmath341 , above the bi - electronic threshold @xmath342 , exactly at the polaronic one @xmath340 .
now two parallel processes of one- and two- electron tunnelling are allowed . as a function of the energy @xmath65 of the associated bound state ( over a selected interval ) .
this figure corresponds to @xmath341 which is just at the polaronic threshold , above the bipolaronic one.,height=207 ] the figure [ fig : u16+c ] corresponds to the potential @xmath343 between the bi - electronic threshold @xmath342 , and the polaronic one @xmath340 .
contrary to the figure [ fig : u14 ] , the coulomb interaction is taken into account which lifts the degeneracy of the earlier crossing point of three branches .
the one electron term @xmath4 does not cross @xmath337 axis yet , but it passes below two other terms in a vicinity of their crossing .
now the optimal bi - electronic tunnelling takes place via a sequence of two single electronic processes confined in time . as a function of the energy @xmath65 of the associated bound state .
this figure corresponds to @xmath343 which is between the thresholds for tunnelling of bipolarons and polarons .
contrarily to the figure [ fig : u14 ] , the coulomb interaction is taken into account which lifts the crossing degeneracy.,height=207 ]
we need to perform the integration over @xmath177 around the extremal taking into account the zero modes related to translations of positions @xmath181 of the instanton centers .
the path integration over the gapless mode @xmath186 is important , particularly for the matrix element : the overlap of wave functions evolves following @xmath344 while their localization follows the evolution of @xmath141 .
we shall work within the zero dimensional reduction of eq.([reduction ] ) .
@xmath346d[x_j]j_{x_j}j_{e_j}\exp(-s)\nonumber\\ \psi_{a}^{\ast}(x - x_{a}(\tau_1),e(\tau_1))\psi_{a}(y - x_{a}(\tau_2),e(\tau_2 ) ) \psi_{b}(x - x_{b}(\tau_1),-e(\tau_1))\psi_{b}^{\ast}(y - x_{b}(\tau_2),-e(\tau_2 ) ) , \label{jjj}\end{aligned}\ ] ] where @xmath347 , @xmath348 are the jacobians of the transformation ( [ reduction ] ) .
( @xmath349 is the number of points for the intermediate discretization of the time axis . )
we integrate over the zero mode @xmath186 and take into account fluctuations of the instanton shape due to variations of the parameter @xmath350 .
the action in ( [ reduction ] ) has the form @xmath351=\sum_{j = a , b}dt\left(m(e_{j}(t))\dot{x_{j}}^{2}/2+f(e_{j})\dot{e}_{j}^{2}/2 + { v}_u(e_j)\right)\ ] ] with @xmath352 from ( [ s2(u ) ] ) .
the integration over @xmath353 is carried out exactly after the transformation @xmath354 using the known expression here @xmath363 , and the same for @xmath364 , are functions of energies in these points which finally become @xmath365 . using fourier transforms , we rewrite the product of wave functions as @xmath366 integration over @xmath367 gives @xmath368 , @xmath369 and integration over @xmath370 gives @xmath371 .
after integration over @xmath372 we arrive at the result ( [ pre0 ] ) .
the factor @xmath373 in ( [ pre0 ] ) after integration over @xmath374 $ ] which was performed using again the equation ( [ dashen ] ) .
i. latyshev , a. a. sinchenko , l. n. bulaevskii , v.n .
pavlenko , p. monceau , jetp letters , * 75 , * 93 ( 2002 ) ; yu.i .
latyshev , l. n. bulaevskii , t. kawae , a. ayari , and p. monceau , j. phys .
iv france , * 12 * , pr9 , 109 ( 2002 ) .
we view the pg as a partial filling of an expected clear spectral gap , which is adequate to cdw physics .
another view , typical in the field of hihg - tc superconductors , is that the pg is a suppressiion of the expected metalic dos .
these too approaches might be convergent .
artemenko , a.f.volkov , sov .
: jetp * 60 * , 395 ( 1984 ) ; k.m .
munz , w.wonneberger , z.phys . b**79 * * , 15 ( 1990 ) ; a.m. gabovich , a.i .
voitenko , phys .
b * 52 * , 7437 ( 1995 ) ; k. sano eur . phys
. j. b * 25 * , 417 ( 2002 ) _ and rfs .
theirin_. s. brazovskii , n. kirova , `` electron selflocalization and superstructures in quasi one - dimensional dielectrics '' in _
soviet scientific reviews _
, i. m. khalatnikov ed .
( harwood ac . publ . , ny , 1984 ) , vol . * 5 * , p. 99 .
latyshev , p.monceau , a. sinchenko , l. bulaevskii , s. brazovskii , t. kawae , t. yamashita , _ in _ proceedings of the international workshop on strongly correlated electrons in new materials ( scenm02 ) , j. physics a : * 36 * , 9323 ( 2003 ) .
b. halperin , m. lax , phys .
rev . * 148 * , a722 ( 1966 ) ; j. zittarts , j.s .
langer , phys . rev . * 148 * , a741 ( 1966 ) ; i.m .
lifshits , s.a .
gredeskul and l.a .
`` introduction to the theory of disordered systems '' , _ ( wiley , new york , 1988 ) .
lifshitz , yu . m. kagan , sov .
jetp * 35*,206 ( 1972 ) ; s.v .
iordanskii , a. m. finkelshtein , sov .
jetp * 35 * , 215 ( 1972 ) ; s. v. iordanskii and e. i. rashba , sov .
: jetp * 47 * , 975 ( 1978 ) .
epstein , a.g .
macdiarmid , journal of molecular electronics * 4 * , 161(1988 ) ; z. vardeny , e. ehrenfreund , and o. brafman , m. nowak , h. schaffer , a. j. heeger , and f. wudl , phys .
lett . * 56 * , 671 ( 1986 ) .
s. brazovskii , n. kirova , jetp letters * 33 * , 4 ( 1981 ) _ and _ chemica scripta * 17 * , 171 ( 1981 ) ; d. k. campbell and a. r. bishop , phys .
b 24 , r4859 ( 1981 ) ; k. fesser , a.r . bishop , d.k .
campbell , phys .
b , * 27 * , 4804 ( 1983 ) ; s. brazovskii , n. kirova , s. matveenko , sov .
phys . : jetp * 59 * , 434 ( 1984 ) ; s. matveenko , sov .
: jetp * 60 * , 1026 ( 1984 ) . | we suggest a theory of internal coherent tunnelling in the pseudogap region where the applied voltage is below the free electron gap . we consider quasi 1d systems where the gap is originated by a lattice dimerization ( peierls or ssh effect ) like in polyacethylene , as well as low symmetry 1d semiconductors .
results may be applied to several types of conjugated polymers , to semiconducting nanotubes and to quantum wires of semiconductors .
the approach may be generalized to tunnelling in strongly correlated systems showing the pseudogap effect , like the family of high tc materials in the undoped limit .
we demonstrate the evolution of tunnelling current - voltage characteristics from smearing the free electron gap down to threshold for tunnelling of polarons and further down to the region of bi - electronic tunnelling via bipolarons or kink pairs . the interchain tunnelling is described in a parallel comparison with the on chain optical absorption , also within the subgap region . |
the properties of magnetic impurities in a metal is one of the most intensively studied problems in solid state physics .
the work of friedel @xcite and anderson @xcite laid the foundation to understand why some transition - metal impurities form a local magnetic moment while others do nt .
kondo @xcite showed that multiple scattering of conduction electrons by a magnetic impurity yields a divergent contribution to the resistance in perturbation theory .
yoshida @xcite introduced the concept that the ( spin 1/2 ) magnetic impurity forms a singlet state with the conduction electrons and is non - magnetic at zero temperature .
these new insights stimulated a large body of theoretical and experimental work ( see for example @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite ) .
the majority of experimental and theoretical work has focussed on the singlet kondo ground state .
however , the `` magnetic state '' of the impurity is of equal or even greater importance because magnetic impurities are always present , including in micro - chips and nanostructures , and influence the thermodynamic and transport properties of the hosts .
since many experiments and almost all technical applications are * not * performed at low temperatures the magnetic impurities are generally far above their kondo temperature @xmath0 and show their full magnetic behavior .
the theoretical investigation of the magnetic state has been explored in much less detail than the kondo ground state for spin @xmath1 impurities . in many cases
the kondo temperature is very low , in the range of liquid helium temperature . in this case
the impurity is in the magnetic state at relatively low temperature .
( the word impurity is in this paper reserved to impurities which possess - at sufficiently high temperature - a magnetic moment ) .
when the temperature is several times the kondo temperature one is sufficiently above @xmath0 to destroy the kondo ground state . on the other hand one may expect that the properties of the magnetic state are not yet influenced by the thermal excitations due to the finite temperature .
therefore a number of theoretical investigations treat the magnetic state at zero temperature , i.e. as a magnetic ground state .
this approach is probably justified but it leaves the work always vulnerable to the criticism that there is no magnetic moment at zero temperature . therefore in this paper
i prefer to use the effect of a magnetic field on the kondo state .
a magnetic field which is an order of magnitude larger than @xmath2 ( @xmath3=bohr magneton ) destroys the kondo singlet state as well and yields the magnetic state .
its side effects are that it changes the energy of the d - states by @xmath4 and shifts the conduction bands by @xmath4 .
the latter yields the pauli susceptibility but has otherwise only a negligible effect on the interaction between the impurity and the conduction electrons because the fermi level for spin - up and down electrons readjusts to the same height ( as before ) .
friedel @xcite and anderson @xcite derived a criterion for the instability of the paramagnetic state , i.e. the formation of a magnetic moment : take the density of states @xmath5 of the d - resonance at the fermi energy ( in the paramagnetic state ) and multiply it by the coulomb repulsion energy @xmath6 . if the product @xmath7 then a magnetic moment is formed . within mean field theory the d - density of states
is given by a lorentz function@xmath8 where @xmath9 is an effective energy of the d - electrons in the spin - up or down state , @xmath10 while @xmath11 is the average occupation of the d - electron with the opposite spin ) and the resonance width @xmath12 is given in mean field theory by@xmath13 here @xmath14 is the s - d - hopping matrix element between a conduction electron and the d - state at the impurity and @xmath15 is the density of states of the conduction electrons . in the mean field theory
an occupied @xmath16 electron state can only make transitions into @xmath17-states .
( throughout this paper i express electron states by their creation operators ) .
it is well known that the mean field theory has a number of shortcomings . during the last few years
the group of the author has developed a new approach to the impurity problem , in particular the friedel - anderson and the kondo impurity . in this * fair * method a * * f**riedel state is * * a**rtificially built from each conduction band and * * i**nserted as a * * r**esonance state into the conduction or s - band of spin - up and spin - down electrons . in the appendix a short review of the fair solution for the friedel impurity
is sketched .
the fair solution for the magnetic state yields a considerably lower energy for the `` magnetic ground state '' and requires a much larger critical coulomb interaction to form a magnetic state .
this is of some practical importance since the mean field approximation is used in a number of numerical spin - density functional theory calculations for the magnetic moment of impurities in an ( s , p ) metal host @xcite , @xcite , @xcite , @xcite , @xcite .
in addition to the size of the magnetic moment one would like to know the density of states in the magnetic state . the answer of the mean field theory has been discussed above .
but there have been a number of suggestions that the d - resonance is broader than the mean field suggests ( see for example logan @xcite ) .
the mean field theory decouples the spin - up d - electron from the spin - down d - electron , but in reality
the d - electrons are coupled through the coulomb energy . a transition in the @xmath16 electron state changes the energy and the state of the @xmath18 electron as well . therefore it has been suggested in the past that the d - resonances in the friedel - anderson impurity are larger than the mean field theory predicts .
a wider d - resonance in the friedel - anderson impurity together with the condition @xmath7 would require a larger coulomb energy to form a magnetic moment .
in this connection the previous result of the author that the fair solution requires a ( two times ) larger coulomb energy to form a magnetic moment would find a simple physical interpretation .
it is the goal of this paper to calculate the density of states of the `` magnetic ground state '' in the fair solution and compare it with the mean field density of states . in section ii the theoretical background of the magnetic state of the friedel - anderson impurity is sketched . in section
iii electrons and holes are introduced into the magnetic ground state .
their interactions and excitation energies are derived . in section
iv the results of the numerical calculations are presented . finally in section
v and vi the results are discussed together with the conclusion . in the appendix a the basic idea of the fair method is sketched .
the simplified hamiltonian for a magnetic impurity is generally described by the friedel - anderson ( fa ) hamiltonian@xmath20\right\ } + un_{d\uparrow } n_{d\downarrow } \label{h_fa}\ ] ] here the operators @xmath21 represent s - electrons , i.e. the conduction band . in the friedel - anderson hamiltonian in equ .
( [ h_fa ] ) the d - state for each spin interacts with every electron in the conduction band .
imagine how much easier the task would be if the d - electron would interact only with a single electron state ( in each spin band ) . all other conduction band states would represent just a background or quasi - vacuum .
this is the fair approach . during the last few years the author introduced such a solution to the friedel - anderson impurity problem
in which only four electron states , the spin - up and spin - down d - states @xmath16 and @xmath18 and two fair states , @xmath22 and @xmath23 interact through the coulomb and s - d - hopping potential .
these states @xmath22 and @xmath24 are composed of the spin - up and spin - down conduction band states .
they are the friedel artificially inserted resonance states or fair states .
the interaction of the remaining conduction electron states with the d - states is insignificant ; they just yield a background .
this yields very good ground - state properties .
the fair states are composed of the corresponding conduction bands@xmath25{ccc}a_{0\uparrow}={\textstyle\sum_{\nu=0}^{n-1 } } \alpha_{0}^{\nu}c_{\nu\uparrow } & & b_{0\downarrow}={\textstyle\sum_{\nu=0}^{n-1 } } \beta_{0}^{\nu}c_{\nu\downarrow}\end{array}\ ] ] the remaining @xmath26 states in each spin band are constructed orthogonal to the corresponding fair state , orthonormal to each other and sub - diagonal with respect to the band energy hamiltonian @xmath27 this yields new bases for the conduction bands @xmath28 and @xmath29 with @xmath30 .
these new bases are uniquely determined by the two fair states . within the new bases the fa - hamiltonian ( [ h_fa ] )
can be expressed as @xmath31 with@xmath32 where @xmath33 \label{h_0'}\]]@xmath34 + \sum_{i=1}^{n-1}v_{i}^{\left ( a\right ) sd}\left [ d_{\uparrow } ^{\dag}a_{i,\uparrow}+a_{i,\uparrow}^{\dag}d_{\uparrow}\right ] \label{h_1'}\ ] ] and the spin - down hamiltonians are obtained by replacing @xmath35 by @xmath36 and the @xmath37-states by @xmath38-states .
the hamiltonian @xmath39 is diagonal in the band states @xmath40 and @xmath41 for @xmath42 .
the only interaction takes place between the states @xmath43 , @xmath44 , @xmath45 and @xmath46 .
i call these states the * nest states*. the ground state of the hamiltonian @xmath39 is straight forward .
it consists of the coupled state between the nest states and a partially occupied spin - up and down band .
i occupy each spin component with @xmath47 electrons , putting @xmath48 electrons into each conduction band states and one spin - up and one spin - down electron into the nest .
this yields the magnetic ground state as described in equ .
( [ psi_ms ] ) .
@xmath49 \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow } \right\rangle \label{psi_ms}\ ] ] where @xmath50 represents a kind of quasi - vacuum ( @xmath51 ) .
the calculation of the coefficients @xmath52 yields a secular hamiltonian @xmath53 which i call the nest - hamiltonian and which has the form @xmath54{cccc}e_{0}^{\left ( a\right ) } + e_{0}^{\left ( b\right ) } & v_{b}^{sd } & v_{a}^{sd } & 0\\ v_{b}^{sd } & e_{0}^{\left ( a\right ) } + e_{d } & 0 & v_{a}^{sd}\\ v_{a}^{sd } & 0 & e_{d}+e_{0}^{\left ( b\right ) } & v_{b}^{sd}\\ 0 & v_{a}^{sd } & v_{b}^{sd } & 2e_{d}+u \end{array } \right)\ ] ] here the abbreviations are used : @xmath55 and @xmath56 ( see equ.([h_0 ] ) ) .
the superscript @xmath57 stands for nest and the subscript @xmath58 gives the number of nest electrons in the spin - up and spin - down state .
( the energy of the occupied band states is not included .
it yields the same contribution to each component ) .
@xmath53 has four eigenvalues and eigenstates . the lowest eigenvalue yields the ground state .
the components of the ground state , nest plus the band states , are shown in fig.1 .
the first order correction to the energy , i.e. the expectation value of @xmath59 is zero .
but in addition the second order perturbation of @xmath59 is extremely small .
this is demonstrated in appendix b. it may appear remarkable that the neglect of the interactions between the d - electron and all the band states @xmath60 and @xmath61 yields a realistic ground state .
but it is not unheard off that one can obtain an excellent ground state while neglecting a major part of the interaction in the system .
the bcs theory is a good example because it only includes the electron - phonon interaction between cooper pairs of time - reversed electrons .
the interaction between all the other electrons is neglected although their number is much larger .
one major part of the numerical calculation is , of course , the optimization of the two fair states @xmath43 and @xmath62 so that the expectation value of the energy @xmath63 of the hamiltonian @xmath39 has a minimum .
the optimization procedure is described at length in previous papers @xcite , @xcite , @xcite and is taken for granted in this paper and will not be described here .
( the fair states are rotated in hilbert space ) .
since we do nt count the fair states any more as band states the number of band states is reduced by one and their energy is slightly shifted ( by less than the original energy spacing ) .
the band states enter in the energy @xmath64 only through the kinetic ( band ) energy of the occupied band states . in a way they just prepare the nest for the states @xmath65 .$ ] @xmath66{fig1.eps}}\\ & \begin{tabular } [ c]{l}fig.1 : the composition of the magnetic state $ \psi_{ms}$ in the nest is shown.\\ it consists of four slater states .
each slater state has a half - full spin - up and\\ down band , two fair states ( circles in within the bands ) and two d - states\\ ( circles on the left and right of the band ) .
full black circles represent\\ occupied states and light grey represent empty states .
the band\ at the right\\ with the half - filled circles symbolizes the magnetic solution with four\\ slater states .
\end{tabular}\end{aligned}\ ] ] in the numerical calculation we will present the results for two examples with the parameters @xmath67 , @xmath68 and @xmath69 and @xmath70 .
the smaller value of the s - d - matrix element permits a better fit of the resulting resonance curve with a lorentz curve since the effect of the finite width of the band is smaller . in the construction of the magnetic ground state
@xmath71 the hamiltonian @xmath59 has been completely neglected .
below we will derive the excitation energies by introducing an additional electron ( hole ) into an empty ( occupied ) states .
for this calculation it is important to know whether the empty state is really empty or whether transitions from the ground state into the state due to @xmath59 have partially occupied this state .
( this problem is well known from the calculation of the electron - phonon mass enhancement . in the calculation of
the electron - phonon self - energy one injects an electron into an `` empty state '' @xmath72 above the fermi energy .
the transitions of this electron via the electron - phonon interaction into other empty states @xmath73 contribute to the self - energy @xmath74 .
however , the state @xmath72 was not really empty because transitions from the ground state into @xmath72 already created a finite occupation of @xmath72 .
one has to correct the self - energy due to these processes ) . in the appendix
i show that transitions from the ground state into empty band states ( due to @xmath59 ) are practically zero .
the interference between transitions from the d - state and from the corresponding fair state almost perfectly cancel each other .
the total weight in all perturbation states is only of the order of @xmath75 and can be completely neglected . therefore the band states are either completely empty or fully occupied .
in fig.2 a spin - up electron is injected into one of the empty states @xmath76 of the @xmath77-band .
this yields the slater states ( @xmath78 ) .
the slater state ( @xmath79 ) is obtained by injecting an electron into the nest , either into the state @xmath22 or @xmath16 . @xmath80{fig2.eps}}\\ & \begin{tabular } [ c]{l}fig.2 : an electron has been injected into the spin - up band and the spin up nest.\\ this induces electron or hole transitions in the spin - down band .
the resulting\\ slater states are shown as ( $ \gamma$ ) and ( $ \delta$ ) .
( each band with half circles consists\\ of two slater states . ) \end{tabular}\end{aligned}\ ] ] the injection into @xmath22 or @xmath16 yields@xmath81{lll}$a_{0\uparrow}^{\dag}\psi_{ms}=\left ( a_{d , b}a_{0}^{\dag}d_{\uparrow}^{\dag } b_{0\downarrow}^{\dag}+a_{d , d}a_{0}^{\dag}d_{\uparrow}^{\dag}d_{\downarrow } ^{\dag}\right ) \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow } \right\rangle $ & for & $ a_{0}^{\dag}$\\ $ d_{\uparrow}^{\dag}\psi_{ms}=-\left ( a_{a , b}a_{0\uparrow}^{\dag}d_{\uparrow } ^{\dag}b_{0\downarrow}^{\dag}+a_{a , d}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag } d_{\downarrow}^{\dag}\right ) \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & for & $ d_{\uparrow}^{\dag}$\end{tabular } \ \ \ \ \label{a_psi_ms}\ ] ] both final states yield a double occupancy of the spin up nest states .
furthermore these states are not eigenstates of the nest . with respect to the ( basis )
states @xmath82 and @xmath83 the nest hamiltonian takes the form@xmath84{cc}\left ( e_{d}+e_{0}^{\left ( a\right ) } \right ) + e_{0}^{\left ( b\right ) } & v_{b}^{sd}\\ v_{b}^{sd } & \left ( e_{d}+e_{0}^{\left ( a\right ) } \right ) + e_{d}+u \end{array } \right ) \label{hnst_3}\ ] ] by diagonalization one obtains the eigenstates @xmath85 and @xmath86 with @xmath87 the states ( @xmath78 ) and ( @xmath79 ) in fig.2 are the initial states which one obtains through injection of a spin - up electron into the ground state . due to the perturbation hamiltonian @xmath59
these states interact with each other and ( @xmath79 ) interacts with the states ( @xmath88 ) and ( @xmath89 ) . in table
i the possible states which can be obtained through the injection of a spin - up electron plus linear coupling through @xmath59 are collected .
these states are ( @xmath78 ) the two - electron nest ground state plus one electron , ( @xmath79 ) a nest with two spin - up and one spin - down electron , ( @xmath88 ) a full spin - up and empty spin - down nest plus one spin down electron and ( @xmath89 ) a full spin - up and full spin - down nest and one spin down hole . in table i these states , their number and their energies are listed .
@xmath90{|l|l|l|l|}\hline $ \mathbf{\psi}_{f}$ & \textbf{number } & \textbf{energy } & \\\hline $ a_{j\uparrow}^{\dag}\psi_{ms}$ & $ n/2 $ & $ e_{j}^{\left ( a\right ) } $ & \\\hline $ \psi_{2/1}^{\left ( \alpha\right ) } $ & $ 2,\alpha=1,2 $ & $ e_{d}+e_{0}^{\left ( a\right ) } + e_{2/1}^{\alpha}-e_{00}$ & \\\hline $ a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag}b_{j\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ n/2 $ & $ ed+e_{0}^{\left ( a\right ) } + e_{j}^{\left ( b\right ) } -e_{00}$ & \\\hline $ b_{k\downarrow}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag}b_{0\downarrow}^{\dag } d_{\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ n/2 - 1 $ & $ \begin{array } [ c]{c}2e_{d}+u+e_{0}^{\left ( a\right ) } + e_{0}^{\left ( b\right ) } \\
-e_{k}^{\left ( b\right ) } -e_{00}\end{array } $ & \\\hline \end{tabular } \\ & \begin{tabular } [ c]{l}table i : this table describes the states of fig.2 which are generated\\ by the injection of one spin - up electron into the magnetic ground state\\ and transition from the resulting states through $ h_{1}^{\prime}$. the state $ \psi_{ms}$ is\\ given by equ .
( \ref{psi_ms } ) and $ \psi_{2/1}^{\left ( \alpha\right ) } $ is given by ( \ref{psi2/1 } ) .
the\\ energy is measured from the ground - state energy $ e_{00}$. \end{tabular}\end{aligned}\ ] ]
fig.2 and table i show all the spin - up electron excitations which interact linearly in @xmath59 .
this is a total of @xmath91 states .
it is straight forward to construct the secular matrix ( i.e. the excitation hamiltonian @xmath92 ) between the excitations in table i. one may put the two nest states @xmath93 and @xmath86 at the positions one and two , followed by the @xmath94 additional single particle excitations .
we denote these @xmath91 states as @xmath95 .
the diagonal of the hamiltonian is given by the energies in table i. the off - diagonal elements of @xmath92 are the matrix elements of @xmath59 between the states ( @xmath96 ) in fig.2 .
the single particle excitations interact only with the first two nest states through @xmath59 but not among each other .
( in appendix b the corresponding matrix elements are shown in table iii for similar transitions from the ground state . )
this hamiltonian @xmath92 is diagonalized and yields a set of @xmath97 new eigenstates @xmath98 with eigenenergies @xmath99 .
the components of the eigenstates @xmath98 in terms of @xmath95 are given as columns @xmath100 .
the @xmath101th row of the matrix @xmath100 yields the amplitude of our @xmath101th original state @xmath95 in terms of the new eigenstates @xmath98 .
if this @xmath101th original state is , for example , @xmath102 then it can be expressed in the new eigenstates as @xmath103 its density of states is then @xmath104 since electron injection creates only the states ( @xmath78 ) and ( @xmath79 ) one obtains the full ( spin - up ) excitation spectrum by summing over these @xmath105 states .
the weight of states @xmath106 is one , however , the weight of @xmath107 is only @xmath108 since @xmath109 \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow } \right\rangle \\ & = \left [ a_{d , b}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag}b_{0\downarrow } ^{\dag}+a_{d , d}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag}d_{\downarrow}^{\dag } \right ] \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow } \right\rangle\end{aligned}\ ] ] a similar result is found for the weight of @xmath110 which is @xmath111 .
since @xmath107 and @xmath110 are not eigenstatet of @xmath39 they represent a combination of the two eigenstates @xmath112 . in fig.3
is sketched what happens when an electron is injected into either the state @xmath113 or @xmath16 .
the electron injection yields a superposition of the two eigenstates @xmath112 . from these states
the electron can make a transition into any of the ( @xmath114 ) states via @xmath59 .
the two transition amplitudes interfere in this transition .
this interference has to be included in the calculation of the spectral weight density of @xmath107 and @xmath110 ( which requires just the scalar products between @xmath107 and @xmath115 ( or @xmath110 and @xmath115 ) .
this is discussed in more detail in appendix c. @xmath116{fig3.eps}}\\ & \begin{tabular } [ c]{l}fig.3 : an electron has been injected into the $ a_{0\uparrow}^{\dag}$ state ( or $ d_{\uparrow}^{\dag}$ state,\\ dashed arrow ) .
the resulting state is a superposition of two nest states.\\ from these nest states the electron makes ( as one possibility ) a transition\\ into the state $ a_{j\uparrow}^{\dag}$ where the two amplitudes interfere .
the energies of the\\ two nest states are shown .
the different thickness of the arrows shows\\ different probabilities for the two paths . \end{tabular}\end{aligned}\ ] ] for the full spectrum of excitations one has to include the injection of holes into the occupied states .
this is shown in fig.4 .
the hole can be injected into the occupied states @xmath76 yielding @xmath117 or into the nest . in the latter case the spin - up part of the nest
is emptied .
this yields for the secular matrix of the nest in analogy to equ .
( [ hnst_3])@xmath118{cc}e_{0}^{\left ( b\right ) } & v_{b}^{sd}\\ v_{b}^{sd } & e_{d}\end{array } \right ) \label{hnst_1}\]]@xmath119{fig4.eps}}\\ & \begin{tabular } [ c]{l}fig.4 : an hole has been injected into the spin - up band and the spin up nest.\\ this induces electron or hole transitions in the spin - down band .
the resulting\\ slater states are shown as ( $ \gamma$ ) and ( $ \delta$ ) .
\end{tabular}\end{aligned}\ ] ] @xmath90{|l|l|l|}\hline \textbf{state } & \textbf{number } & \textbf{energy}\\\hline $ a_{j\uparrow}\psi_{ms}$ & $ n/2 - 1 $ & $ -e_{j}^{\left ( a\right ) } = \left\vert e_{j}^{\left ( a\right ) } \right\vert $ \\\hline $ \psi_{\downarrow}^{\left ( \alpha\right ) } = \left ( b_{b}b_{0\downarrow } ^{\dag}+b_{d}d_{\downarrow}^{\dag}\right ) \left\vert \mathbf{0}_{a,\uparrow } \mathbf{0}_{b,\downarrow}\right\rangle $ & $ 2,\alpha=1,2 $ & $ e_{\downarrow } ^{\left ( \alpha\right ) } -e_{00}$\\\hline $ b_{j\downarrow}b_{0\downarrow}^{\dag}d_{\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ n/2 - 1 $ & $ e_{d}+e_{0}^{\left ( b\right ) } -e_{j}^{\left ( b\right ) } -e_{00}$\\\hline $ b_{j\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ n/2 $ & $ e_{j}^{\left ( b\right ) } -e_{00}$\\\hline \end{tabular } \\ &
\begin{tabular } [ c]{l}table ii : this table describes the states of fig.4 which are generated\\ by the injection of one spin - up hole into the magnetic ground state\\ and transition from the resulting states through $ h_{1}^{\prime}$. the energy is\\ measured from the ground - state energy $ e_{00}$. \end{tabular}\end{aligned}\ ] ] the construction of the excitation or secular hamiltonian @xmath92 is in complete analogy to the electron injection .
this time the number of excitations is @xmath120 .
the spectrum is obtained in the same way as before .
for the numerical calculation a conduction band with a finite number of states is used .
we follow here wilson @xcite by using an s - electron band with constant density of states and the fermi level in the center , which we divide into energy cells @xmath121 . in each energy cell
( which may contain @xmath122 @xmath72-states @xmath123 we rearrange the states ( by an orthogonal transformation ) so that one state @xmath124 accumulates all the interaction with the d - states while the other @xmath125 states have zero interaction with the d - states .
wilson normalized the energy in terms of the fermi energy so that his band extended from @xmath126 @xmath127to @xmath128 .
wilson s logarithmic scale for the energy cells is not opportune for the present investigation because it is not fine enough at the energy of the d - resonance .
therefore i use a linear sub - division of the energy band @xmath129 . for the majority of calculations
the energy band is sub - divided into @xmath130 and @xmath131 energy cells .
the state @xmath132 represents all the s - electron states in the cell @xmath133 and possesses the average energy @xmath134 ( corresponding to @xmath135 for @xmath136 ) . for @xmath137 energy cells with constant width of @xmath138
the s - d - matrix elements @xmath139 is given by @xmath140 so that @xmath141 .
in the following i show the results for the friedel - anderson hamiltonian with the d - level energy @xmath142 the coulomb energy @xmath67 and an s - d - hopping matrix element of @xmath70 .
the magnetic moment of this impurity is @xmath143 ( for @xmath144 ) .
the calculations are performed with @xmath130 or @xmath131 energy levels of constant spacing .
a second set of results is derived for the parameters @xmath68 , @xmath145 and @xmath146 . to assure that the magnetic state @xmath71 is indeed the ground state ( i.e. to prevent the formation of a singlet state ) a magnetic field @xmath147
is applied yielding a magnetic energy @xmath148 .
this energy is chosen so that @xmath149 .
i estimate the kondo energy from the difference between the energies of the singlet and the triplet state .
this energy difference is about @xmath150 for @xmath146 and of the order of @xmath151 for @xmath70 .
it turned out that the required magnetic energy is in both cases so small that it yields no noticeable changes .
this is partly due to the fact that the absolute smallest band energies which are @xmath152 act as a finite temperature as wilson pointed out @xcite . for @xmath136
this corresponds to a temperature of @xmath153 which is a very large temperature compared with most kondo temperatures .
but magnetic field is of academic importance to assure the magnetic state @xmath71 is the appropriate ground state .
the use of equidistant energy levels is important to identify the resonance state within the electron bands .
but it has the drawback that it does not describe well the behavior of the wave function at low energies . at low energies the logarithmic energy scale which wilson introduced would be more appropriate .
but the evaluation of the density of states is much more difficult for a non - linear energy scale
. for each spin band one obtains a spectrum with a total weight of @xmath154 , corresponding to @xmath137 s - electron states and one d - electron state .
however , the number of energy levels is @xmath155 .
( this is the number of eigenstates of the excitation hamiltonians for electrons and holes together ) .
this means that the weight at the individual energies is at least for @xmath156 energies much less than one . in fig.5 the spectral weight at different energies
is shown in the energy range from @xmath126 to @xmath157 for the minority band . for negative energies
the weight is either very close to one or very small . here
one can calculate the density of states by the separation of the levels with weight close to one ( by dividing the weight by the level separation ) .
the evaluation is more complicated for positive energies .
however , here the sum of neighboring energy levels is close to one
. then one can calculated the `` center of weight '' for two neighboring levels which have a total weight close to one and then proceed as before .
it turns out that the best approach is to start from the lower and upper ends of the band in the evaluation .
@xmath158{fig5.eps}}\\ & \begin{tabular } [ c]{l}fig.5 : the spectral weight for the different excitation energies .
\end{tabular}\end{aligned}\ ] ] in fig.6a the density of states of the excitation spectrum for the majority spin is shown .
the full circles are obtained with @xmath136 states and the stars use @xmath144 equally spaced wilson states .
the full curve represents a lorentz curve with the parameters @xmath159 the resonance energy is is @xmath160 and the resonance ( half ) width is @xmath161 .
the corresponding mean field resonance width is @xmath162 .
@xmath163{fig6a.eps}}\\ & \begin{tabular } [ c]{l}fig.6a : the density of states for the majority spins .
the full circles\\ are calculated with $ n=40 $ states and the stars use $ n=80 $ equally\\ spaced wilson states .
the full curve represents a lorentz curve\\ with the resonance energy $ e_{r}=-0.53 $ and the width $ \gamma_{r}=0.08$.\\
this width is twice the mean - field value of $ \gamma_{mf}=0.039$. \end{tabular}\end{aligned}\ ] ] in fig.6b the density of states of the minority spin is drawn .
again the full curve represents a lorentz curve with the resonance at @xmath164 and a resonance half - width of @xmath165 .
@xmath166{fig6b.eps}}\\ & \begin{tabular } [ c]{l}fig.6b : the density of states for the minority spins .
the full circles\\ are calculated with $ n=40 $ states and the stars use $ n=80 $ equally\\ spaced wilson states .
the full curve represents a lorentz curve\\ with the resonance energy $ e_{r}=0.52 $ and the width $ \gamma_{r}=0.08$.\\ again this width is twice the mean - field value of $ \gamma_{mf}=0.039$. \end{tabular}\end{aligned}\ ] ] for comparison fig.7 shows the result of a similar calculation and evaluation for a simple friedel resonance , where the coulomb energy @xmath6 is set equal to zero .
the evaluation yields a lorentz curve with @xmath167 and @xmath168 the s - d - matrix element is still given by @xmath70 .
@xmath169{fig7.eps}}\\ & \begin{tabular } [ c]{l}fig.7 : the density of states for an impurity with $ u=0$. the full circles\\ and the stars are calculated with $ n=40 $ and $ n=80 $ states.\\ the full curve represents a lorentz curve with the resonance energy\\ $ e_{r}=-0.51 $ and the width $ \gamma_{r}=0.044$. \end{tabular}\end{aligned}\ ] ] alternatively we tried to obtain the density of states by broadening the @xmath89-shaped energy spectrum in fig.5 with a gaussian curve @xmath170 $ ] .
this method worked quite well .
the optimal density curve was obtained when @xmath171 was equal to the level distance @xmath138 .
the width of the two resonances was essentially the same as in fig.6a , b and fig.8a , b .
only the heights were slightly reduced .
however , i prefer to use the other evaluation method in this paper so that there is no doubt that the broadening of the resonance width is a real physical effect and not due to an artificial broadening with a gaussian curve . in a second series
the majority and minority density of states have been calculated for the parameters @xmath68 , @xmath145 , and @xmath146 .
this impurity possesses a magnetic moment of @xmath172 . in this case
the mean - field theory yields a resonance width of @xmath12=0.0@xmath173 .
the best fit to the numerical results yields @xmath174 .
it is again about twice the value of the mean field .
( the resonance curves are no longer perfectly symmetrical because of the finite width of the conduction band ) .
the corresponding friedel density of states ( which is not shown here ) has a resonance with of 0.08 which is quite close to the theoretical value of @xmath175 .
@xmath176{fig8a.eps}}\\ & \begin{tabular } [ c]{l}fig.8a : the density of states for the majority spins for $ \left\vert v_{sd}^{0}\right\vert ^{2}=0.05$\end{tabular}\end{aligned}\]]@xmath166{fig8b.eps}}\\ & \begin{tabular } [ c]{l}fig.8b : the density of states for the minority spins for $ \left\vert v_{sd}^{0}\right\vert ^{2}=0.05$\end{tabular}\end{aligned}\]]@xmath19
in the mean field approximation the magnetic state has two d - resonances at the energies @xmath177 . since in the symmetric case one
has @xmath178 one finds @xmath179 since @xmath180 . so generally the resonances are closer to fermi energy than @xmath181 in our case for the parameters @xmath142 @xmath145 and @xmath70 the magnetic moment is within 1% @xmath182 ( in units of @xmath183 and one expects the resonance almost at @xmath184 .
the resonance width in mean field is given by @xmath185 where @xmath186 is the density of states of the s - electrons at the fermi level .
this yields for the above parameters @xmath187 .
( the matrix element @xmath188 and the density of states are normalized to the atomic volume as the sample volume ) .
this is the same width that one expects for a friedel resonance with @xmath70 . indeed in fig.7
the friedel resonance has a @xmath189-value of @xmath190 .
this agrees within 10% or @xmath191 with the numerical result .
the important result is that the resonance width of the minority and majority spins is larger than the friedel resonance width by a factor of two .
therefore the obtained resonance width is also twice the mean field resonance width .
this suggests that the any calculation which uses mean field yields an incorrect density of states . it will be interesting to check how the spin - density functional theory is affected by this result because the latter uses the mean field approximation . since the coulomb interaction broadens the d - resonance by a factor of two it also reduces the height of the resonance by the same factor of two .
therefore it is very plausible that the mean field theory overestimates the tendency to form a magnetic moment .
if the criterion for the formation of a magnetic moment , @xmath192 , is accepted then one expects that the critical coulomb energy for the formation of a magnetic moment is increased by a factor of two .
this was the previous result by the author @xcite . in the density of states of the majority and minority spins in fig.6ab and fig.8a ,
b one observes a scattering and a small maximum at zero energy .
this is probably due to the fact that i used a constant cell width for the wilson states .
this means that i average over all states within an energy cell of the width @xmath193 .
this is definitely a poor approximation for the two energy cells @xmath194 and @xmath195 ( which touch the fermi level ) .
wilson avoids this problem by using a logarithmic energy scale .
however , the present method to evaluate the density of states does not work for an energy - dependent cell width .
details of this question will have to be clarified in the future .
in this paper the density of states of the friedel - anderson impurity is calculated in the magnetic ground state .
the magnetic ground state is enforced by the application of a magnetic field whose zeeman energy is an order of magnitude larger than the kondo energy .
( for the parameters chosen in the numerical calculation the effect of the magnetic field is so small that it can be neglected ) .
the fair ground state is the eigenstate of a hamiltonian @xmath39 . to construct this hamiltonian two artificial friedel resonance states @xmath43 and @xmath62 are reverse engineered out of the original spin - up and spin - down conduction bands . there is a numerical procedure to optimize these two fair states .
when this is done one has an extremely simple hamiltonian @xmath39 . in @xmath39 only the two s - states @xmath43 and @xmath62 interact with the d - state .
these states @xmath196 @xmath62 , @xmath16 and @xmath18 are called the nest states .
@xmath39 is diagonal in the ( modified ) conduction band electrons @xmath197 and @xmath61 .
a perturbation hamiltonian @xmath198 which is the difference between the original fa - hamiltonian and @xmath39 , has zero energy expectation value in the fair ground state .
in addition it is shown in second order self - consistent perturbation theory ( see appendix c ) that the total occupation of all perturbation states is only of the order of @xmath75 , i.e. the fair ground state has still an amplitude of @xmath199 .
this is important when an electron or hole is injected into the ground state .
the excitation spectrum is obtained by injecting an electron or a hole into the ground state .
the resulting excited states interact via the perturbation hamiltonian @xmath59 and yield a spectrum of energy resonances .
it turns out that the injection of an electron into the spin - up conduction band creates also transition between the spin - down conduction band and the nest states .
the resulting density of states possesses the shape of a resonance curve .
however , the resonance width is about twice the value of the mean - field theory . as a consequence the height of the resonance density of states
is reduced by a factor of two .
since the formation of a magnetic moment depends on the product of the coulomb interaction and the density of d - states , one would expect that the mean field overestimates the tendency towards a magnetic moment .
indeed i observed in the first paper about the magnetic ground state that the formation of a magnetic moment requires about twice the coulomb energy that the mean field theory predicts .
this consorts well with the present finding of the reduced resonance density of states .
the next step in the future investigation is the calculation of the density of states of the kondo resonance within the fair model . for this calculation
one has to use a wilson spectrum with a logarithmic energy scale . @xmath19
the basic idea of the fair method can be best explained for a friedel resonance with the hamiltonian @xmath200\ ] ] this is done in the following steps : 1 . from the free ( or s- ) electron basis an artificial friedel resonance ( * fair ) * state
is constructed with @xmath201 together with a full orthonormal basis @xmath202 so that the free electron - electron hamiltonian @xmath203 takes the form @xmath204\ ] ] the requirement that the matrix elements between different @xmath205 and @xmath206 @xmath207 vanish has the consequence that a given fair state @xmath208 determines uniquely the full basis @xmath209 . in the new basis
the total friedel hamiltonian takes the form@xmath210 where @xmath211\ ] ] the perturbation hamiltonian has the form@xmath212 + \sum_{i=1}^{n-1}v_{i}^{\left ( a\right ) sd}\left [ d^{\dag}a_{i}+a_{i}^{\dag}d\right ] \right\}\ ] ] here the new matrix elements are given as @xmath213 and @xmath214 .
a trial state @xmath215 is defined as@xmath216 \left\vert \mathbf{0}_{a}\right\rangle \label{psi_f}\ ] ] where @xmath217 @xmath51 .
the right side is abbreviated as@xmath218 where , for example , for @xmath219 one has @xmath220 3 .
the energy expectation value @xmath221 of the @xmath39 with respect to the trial state @xmath215 is calculated .
the contribution of @xmath59 is not included .
this results in a @xmath222 secular matrix @xmath223{cc}{\textstyle\sum_{i=1}^{n-1 } } e_{i}+e_{0}^{\left ( a\right ) } & v_{0}^{\left ( a\right ) sd}\\ v_{0}^{\left ( a\right ) sd } & { \textstyle\sum_{i=1}^{n-1 } } e_{i}+e_{d}\end{array } \right)\ ] ] whose lowest eigenvalue yields @xmath64 and the corresponding eigenvector yields the coefficients @xmath224 .
4 . the fair state is rotated ( by variation ) in the @xmath137-dimensional hilbert space until the lowest eigenvalue of the secular matrix reaches a minimum .
it has been shown by the author @xcite , @xcite that this procedure results in the * exact * @xmath225-particle ground state of a friedel hamiltonian ( given by ( [ psi_f ] ) ) .
the matrix elements of the perturbation hamiltonian @xmath59 between this ground state and any excited state vanish .
it is interesting to look at the result in some more detail . * the states
@xmath37 @xmath226 enter the secular matrix only through the total energy of the occupied states @xmath227 and contribute only the the background energy . * the coefficients and the relative weight of the states @xmath208 and @xmath228
are only determined by the energies of the fair and the d - state @xmath229 and their coupling @xmath230 . in a way one can say that the states @xmath37 prepare just the background - a kind of nest - for @xmath208 and @xmath228 .
the secular matrix represents an effective hamiltonian for these two states in the nest . in the following
i will call the secular matrix without the kinetic energy @xmath227 the * nest hamiltonian*.@xmath231{cc}e_{0}^{\left ( a\right ) } & v_{0}^{\left ( a\right ) sd}\\ v_{0}^{\left ( a\right ) sd } & e_{d}\end{array } \right)\ ] ] the state @xmath208 represents an artificially inserted friedel resonance state .
therefore i call @xmath208 a `` * * f**riedel * * a**rtificially * * i**nserted * * r**esonance '' state or * fair*-state .
the use of the fair - states is at the heart of my approach to the fa- and kondo impurity problem .
therefore i call this approach the * fair * method .
the hamiltonian of the friedel - anderson impurity is given in equ .
( [ h_fa ] ) .
one obtains the mean - field hamiltonian from equ.([hfa0 ] ) by replacing @xmath232 = @xmath233@xmath234 @xmath235 @xmath236 . after adjusting @xmath237 and @xmath238 self -
consistently one obtains two friedel resonance hamiltonians with a spin - dependent energy of the @xmath239-state : @xmath9 @xmath240 .
@xmath241\right\}\ ] ] the mean - field wave function is a product of two friedel ground states for spin up and down @xmath242 .
now we express each friedel ground state @xmath243 by the fair solution , for example@xmath244 for the two friedel states in the mean - field wave function
i use the form of equ .
( [ psi_f ] ) and obtain for the mean - field solution @xmath245 \left [ \left ( a_{b,\downarrow}a_{0-\downarrow}^{\dag } + a_{s,\downarrow}d_{\downarrow}^{\dag}\right ) \prod_{i=1}^{n-1}b_{i\downarrow}^{\dag}\right ] \phi_{0}\label{psi_mf}\\ & = \left [ a_{a , b}a_{0\uparrow}^{\dag}b_{0\downarrow}^{\dag}+a_{a , d}a_{0\uparrow}^{\dag}d_{\downarrow}^{\dag}+a_{d , b}d_{\uparrow}^{\dag } b_{0\downarrow}^{\dag}+a_{d , d}d_{\uparrow}^{\dag}d_{\downarrow}^{\dag } \right ] \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow } \right\rangle \nonumber\end{aligned}\ ] ] where @xmath246 and @xmath247 are two ( different ) bases of the @xmath137-dimensional hilbert space .
this solution can be rewritten as equation ( [ psi_ms ] ) . in the mean - field solution
@xmath248 the coefficients @xmath249 are restricted by two conditions @xmath250 ( @xmath251 ) .
therefore this state does not describe well the correlation effects .
in contrast the state ( [ psi_ms ] ) opens a wide playing field for improving the solution : ( i ) the fair states @xmath208 and @xmath252 can be individually optimized , each one defining a whole basis @xmath253 and @xmath254 .
this yields a much better treatment of the correlation effects .
the resulting state is denoted as the ( potentially ) magnetic state @xmath71 .
the magnetic state @xmath71 has the same structure as the mean field solution @xmath248 ; the only difference is that its components are optimized for the friedel - anderson hamiltonian .
in the construction of the magnetic ground state @xmath71 only the hamiltonian @xmath39 ( equ . [ h_0 ] ) has been used .
the expectation value of the `` perturbation '' hamiltonian @xmath59 is zero , @xmath255 where @xmath256 and @xmath257 + \sum_{j=1}^{n-1}v_{j}^{\left ( a\right ) sd}\left [ d_{\sigma}^{\dag } a_{j,\sigma}+a_{j,\sigma}^{\dag}d_{\sigma}\right ] \label{h1p}\ ] ] but @xmath59 yields transitions from the ground state @xmath71 into excited states . from equ .
( [ h1p ] ) one recognizes that @xmath258 only permits transitions between the nest @xmath259 and a band state @xmath260 but no transition among band states .
if one considers only final states through this transitions which are linear in @xmath59 then fig.9 shows the possible final states . for the spin - up band
this are the transitions from the nest state into an electron excitation @xmath205 , leaving the spin - up part of the nest empty or a transition from an occupied state by creating a hole @xmath261 and filling both state @xmath22 and @xmath16 of the spin - up part of the nest .
the corresponding states are generated for the spin - down band .
@xmath80{fig9.eps}}\\ & \begin{tabular } [ c]{l}fig.9 : final states which can be obtained by transitions from the\\ ground state $ \psi_{ms}$ through the perturbation hamiltonian $ h_{1}^{\prime } .$\end{tabular}\end{aligned}\ ] ] as an example we consider the transitions in the spin - up band .
applying @xmath262 to the ground state @xmath71 yields@xmath263{c}\left [ v_{j}^{\left ( a\right ) fr}a_{a , b}+v_{j}^{\left ( a\right ) sd}a_{d , b}\right ] a_{j,\sigma}^{\dag}b_{0\downarrow}^{\dag}\\ + \left [ v_{j}^{\left ( a\right ) fr}a_{a , d}+v_{j}^{\left ( a\right ) sd}a_{d , d}\right ] a_{j,\sigma}^{\dag}d_{\downarrow}^{\dag}\\ + \left ( -v_{j}^{\left ( a\right ) fr}a_{d , b}+v_{j}^{\left ( a\right ) sd}a_{a , b}\right ) a_{j,\uparrow}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag } b_{0\downarrow}^{\dag}\\ + \left ( -v_{j}^{\left ( a\right ) fr}a_{d , d}+v_{j}^{\left ( a\right ) sd}a_{a , d}\right ) a_{j,\uparrow}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag } d_{\downarrow}^{\dag}\end{array } \right\ } \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow } \right\rangle\ ] ] the top two lines represent an electron excitation , shown as the left state in fig.9 and the bottom two lines a hole excitation shown as the second state in fig.9 .
the final states and the corresponding matrix elements are collected in table iii .
@xmath90{|l|l|l|l|}\hline $ \mathbf{\psi}_{i}$ & $ \mathbf{\psi}_{f}$ & $ \left\langle \psi_{i}\left\vert h_{1}^{\prime}\right\vert \psi_{f}\right\rangle $ & \textbf{excitation}\\\hline $ \psi_{ms}$ & $ a_{j\uparrow}^{\dag}b_{0\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ \left ( v_{j}^{\left ( a\right ) fr}a_{a , b}+v_{j}^{\left ( a\right ) sd}a_{d , b}\right ) $ & electron , $ j\geq n$\\\hline $ \psi_{ms}$ & $ a_{j,\uparrow}^{\dag}d_{\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ \left ( v_{j}^{\left ( a\right )
fr}a_{a , d}+v_{j}^{\left ( a\right ) sd}a_{d , d}\right ) $ & electron , $ j\geq n$\\\hline $ \psi_{ms}$ & $ a_{j,\uparrow}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag } b_{0\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ \left ( -v_{j}^{\left ( a\right ) fr}a_{d , b}+v_{j}^{\left ( a\right ) sd}a_{a , b}\right ) $ & hole , $ 0<j < n$\\\hline $ \psi_{ms}$ & $ a_{j,\uparrow}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag } d_{\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & $ \left ( -v_{j}^{\left ( a\right ) fr}a_{d , d}+v_{j}^{\left ( a\right ) sd}a_{a , d}\right ) $ & hole , $ 0<j < n$\\\hline \end{tabular } \ \ \ \ \\ & \begin{tabular } [ c]{l}table iii : for the spin - up band the final states $ \psi_{f}$ , and the matrix elements\\ of $ h_{1}^{\prime}$ between the ground state $ \psi_{ms}$ and the final states $ \psi_{f}$ are listed.\\ the final states still have to be expanded into new eigenstates of the nest\\ plus band . for the spin - down band one
obtains equivalent transitions .
\end{tabular } \
\\end{aligned}\ ] ] there are two important aspects of this result , ( i ) there are always two transitions into each final state , for example @xmath264 and @xmath265 .
these two transitions interfere and cancel each other almost completely ( as will be shown below ) .
( ii ) the first two final states in table iii ( as well as the third and forth final state ) are not eigenstates of @xmath39 .
one has to expand these final states in terms of the eigenstates of the nest , for example@xmath81{|l|}\hline $ a_{j,\uparrow}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag}b_{0\downarrow}^{\dag } \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ \\\hline $ a_{j,\uparrow}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag}d_{\downarrow}^{\dag } \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ \\\hline \end{tabular } \ \ \ \ \ \longleftrightarrow\begin{array } [ c]{c}a_{j,\uparrow}\psi_{2/1}^{\left ( 1\right ) } \\
a_{j,\uparrow}\psi_{2/1}^{\left ( 2\right ) } \end{array}\ ] ] this is in complete analogy to the calculation of the excitations in section iii . in fig.10 the logarithm of the effective matrix element is plotted for the first transition in table iii .
this represents an electron excitation which is restricted to positive energies .
as one recognizes the effective matrix elements are strongly reduced in the energy range in which transitions are possible .
the value of @xmath266 lies in the range between @xmath267 and @xmath75 while the original matrix elements for @xmath14 are @xmath268 .
this applies for all eight possible excitations . in the energy range where an excitation is permitted the matrix elements
are strongly reduced . among the eight possible transitions there is only one transition whose matrix elements exceed @xmath267
this is the second one in table iii where the matrix element reaches values of @xmath269 , still much smaller than the original matrix elements . @xmath270{fig10.eps}}\\
& \begin{tabular } [ c]{l}fig.10 : the logarithm of the effective matrix elements for a transition\\ from the ground state $ \psi_{ms}$ into the excited state $ a_{j\uparrow}^{\dag } b_{0\downarrow}^{\dag}\left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ . this\\ transition is only possible for positive energies , and there the\\ interference between the the d - state $ d_{\uparrow}^{\dag}$ and the fair state $ a_{0\uparrow}^{\dag}$\\ cancels the transition almost completely .
\end{tabular}\end{aligned}\ ] ] the strong reduction of the matrix elements is due to the introduction of the fair states which compensate the transitions involving the d - states .
it is also the reason why the magnetic ground state is so well represented by @xmath71 . with the eigenenergies of the states in fig.9 and the matrix elements for
the transition from the ground state @xmath71 into these states one can now perform a self - consistent perturbation calculation .
the number of excited states is @xmath271 for each spin - band .
one can build the the full secular matrix for these states which consists of @xmath272 states , where the additional one is the ground state .
the result for our standard example ( @xmath142 @xmath145 , @xmath273 , @xmath274 is rather dramatic .
the weight of all @xmath275 excitations together is only about @xmath75 .
the amplitude of @xmath71 after diagonalization is @xmath276 .
this demonstrates the fair solution is an excellent approximation to the ground state .
with the secular hamiltonian @xmath92 in section 3.1 one can construct the green s functions ( gf)of the excitations @xmath95@xmath277 or @xmath278 the resulting diagonal elements of @xmath279 are the green s function of the excitations , for example @xmath280 is the green s function of @xmath93 . since the @xmath281 are a function of the energy the above relation is not very practical for a numerical nor analytical calculation of @xmath282 . for a numerical evaluation
it is much easier to calculate the eigenvalues @xmath99 and eigenvectors @xmath98 of @xmath92 where @xmath283 .
then the green s function @xmath284 of the excitation can be expressed as @xmath285 each energy eigenvalue contributes to the spectrum of the state @xmath286 for the particle density of states we need the green s function ( or the spectrum ) of the @xmath287 states @xmath102 and @xmath107 and @xmath110 .
the @xmath47 electron green s functions @xmath288 follow directly from the above calculation .
the @xmath289 and @xmath290 for spin up have to be derived from the green s functions of @xmath291 and @xmath86 . in fig.3
a spin - up electron is injected into the nest .
this electron can be injected into the @xmath22 or the @xmath16 state .
these processes yield complementary amplitudes of @xmath81{lll}$a_{0\uparrow}^{\dag}\psi_{ms}=\left ( a_{d , b}a_{0}^{\dag}d_{\uparrow}^{\dag } b_{0\downarrow}^{\dag}+a_{d , d}a_{0}^{\dag}d_{\uparrow}^{\dag}d_{\downarrow } ^{\dag}\right ) \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow } \right\rangle $ & for & $ a_{0}^{\dag}$\\ $ d_{\uparrow}^{\dag}\psi_{ms}=-\left ( a_{a , b}a_{0\uparrow}^{\dag}d_{\uparrow } ^{\dag}b_{0\downarrow}^{\dag}+a_{a , d}a_{0\uparrow}^{\dag}d_{\uparrow}^{\dag } d_{\downarrow}^{\dag}\right ) \left\vert \mathbf{0}_{a,\uparrow}\mathbf{0}_{b,\downarrow}\right\rangle $ & for & $ d_{\uparrow}^{\dag}$\end{tabular } \ \ \ ] ] each of the two resulting states has to be expanded in the two eigenstates @xmath93 and @xmath86 of the nest .
for example , the amplitudes for @xmath107 are obtained through the scalar product between @xmath107 and @xmath112 .
then they can make a transition into any of the other states with a single excitation , for example the state @xmath102 . again the resulting amplitudes interfere .
so the particle components of the gfs @xmath292 and @xmath290 are composed of the gfs of @xmath112 with a weight which is the square of the scalar product between the amplitude in equ .
( [ a_psi_ms ] ) and @xmath112 . these gfs @xmath289 and @xmath290
have also a hole component .
j. friedel , philos . mag .
43 , 153 ( 1952 ) ; adv .
phys . 3 , 446 ( 1954 ) ; philos .
suppl . 7 , 446 ( 1954 ) ; can .
34 , 1190 ( 1956 ) ; nuovo cimento suppl . 7 , 287 ( 1958 ) ; j. phys .
radium 19 , 573 ( 1958 ) , the electronic structure of the transition metals and their alloys and of the heavy metals | by applying a magnetic field whose zeeman energy exceeds the kondo energy by an order of magnitude the ground state of the friedel - anderson impurity is a magnetic state . in recent years the author introduced
the friedel artificially inserted resonance ( fair ) method to investigate impurity properties . within this fair approach
the magnetic ground state is derived .
its full excitation spectrum and the composition of the excitations is calculated and numerically evaluated . from the excitation spectrum the electron density of states
is calculated .
majority and minority d - resonances are obtained .
the width of the resonances is about twice as wide as the mean field theory predicts
. this broadening is due to the fact that any change of the occupation of the d - state in one spin band changes the eigenstates in the opposite spin band and causes transitions in both spin bands .
this broadening reduces the height of the resonance curve and therefore the density of states by a factor of two .
this yields an intuitive understanding for a previous result of the fair approach that the critical value of the coulomb interaction for the formation of a magnetic moment is twice as large as the mean field theory predicts .
pacs : 75.20.hr , 71.23.an , 71.27.+a |
the rapid experimental progress in the manipulation of ultracold quantum gases has enabled the creation of strongly correlated many - body systems , which are challenging to describe theoretically @xcite . the unambiguous identification of these phases requires precise measurements to characterize its properties .
a powerful tool represents the response function , which provides static properties of the system as well as reveals information about the excitations of the examined state . here ,
we propose a method for an _ in situ _ measurement of the response function to a external and weak probing field for a trapped gas of ultracold atoms . in seminal experimental measurements
the bogoliubov excitation spectrum in a bose - einstein condensate has been measured by probing the dynamic structure factor @xcite .
nowadays , this method has been applied to access the excitation spectrum within the bec - bcs crossover @xcite , as well as the smooth transition of the excitation gap from the superfluid to the mott insulating phase @xcite .
a major drawback of the present approaches is that the setup measures the number of excitations created in the systems : the atomic system is destroyed during the measurement process , and in addition , it requires the application of strong fields violating in most situations the condition of a weak probe . as a consequence
, it becomes difficult to distinguish , whether the measurements probes the dynamical evolution of a strongly driven system or characterizes the ground state properties .
several alternative methods to access information on ground state properties by measuring the light field passing through an atomic system have been discussed , e.g. , the detection of magnetic order @xcite and particle fluctuations @xcite . within a remarkable experiment , the possibility to access the dynamical structure factor by a measurement of the probe field has recently been demonstrated @xcite , which opens a way for weakly probing and analyzing cold atomic gases . in this article , we propose a setup for realizing an _ in - situ _ measurement of the linear response function for a cloud of trapped ultracold atomic atoms .
the setup uses two different cavity fields , see fig [ fig : setup ] , and is motived by recent progress of bose - einstein condensates coupled to high finesse cavities @xcite .
note that the experiment can also be performed in a ring cavity setup to avoid standing waves created by the probe lasers ; similar setups have been proposed for cavity state preparation @xcite .
the main idea is that the interaction of the atomic system with the cavity modes transfers information on the response of the atomic system onto the cavity fields @xcite . by measuring the fields that leave the cavities one can reconstruct the linear response function as well as the dynamic structure factor .
our method puts only modest requirements on the cavity finesse and allows for fast measurements with sufficiently high photon count rates without destroying the atomic system . for simplicity , we focus in the following on the density response function ; the extension to alternative setups accessing e.g. , the spin structure factor is straight forward . for weak probe fields @xmath0 ,
the linear response function is defined by the deformation of the particle density @xmath1 with the fourier transformation relation @xmath2 . while the real part of the response function @xmath3 accounts for the dispersive properties of the media on the probe field , the imaginary part @xmath4 describes the creation of excitations within the media .
consequently , the response function contains important information about the two - particles excitations as well as collective excitations , which is obvious from its relation to the dynamical structure factor @xmath5s({{\bf q}},\omega).\ ] ] in previous experiments the dynamic structure factor @xmath6 has been studied extensively .
however , it is important to point out , that the real part and the imaginary part of the response function are not independent of each other , but rather are related via the kramers - kronig relation @xmath7 as a consequence , it is possible to access information about the excitation spectrum by probing the real part of the response function off - resonantly and without creating any excitations in the media .
such a method is in strong contrast to the current approach , where the created excitations strongly distort the probed state .
the method present in this manuscript allows one to access both the real part of the response function as well as the imaginary part , and opens a way to study _ in situ _ and for very weak probes the response of an quantum many - body system with cold atomic and molecular gases .
and the excited state @xmath8 are driven by cavity fields @xmath9 , @xmath10 with a large detuning @xmath11 from the atomic resonance .
the momentum difference @xmath12 of the cavity fields is transfered to the atomic system . ]
we consider a system of @xmath13 atoms with two internal states @xmath14 and @xmath8 ( see fig . [
fig : excite ] ) , coupled to two cavity fields @xmath15 and @xmath16 with frequencies @xmath17 and @xmath18 , respectively .
the dynamics of the cavity modes is governed by the hamiltonian @xmath19 for a large detuning @xmath11 of the cavity fields from the atomic resonance , we can adiabatically eliminate the state @xmath8 . ignoring effects from spontaneous emission ,
the interaction between the two cavity fields and the particle density operator @xmath20 of the atomic system takes the form @xmath21,\ ] ] with @xmath22 being the two - photon coupling constant , while @xmath23 denotes the difference in momenta between the cavity fields .
the first term in @xmath24 accounts for the stark shift of the atoms and only leads to a renormalization of the cavity resonance frequencies , i.e. , @xmath25 . note that here we assume @xmath26 denotes the total number of atoms in the cavity and is a conserved quantity .
in addition we would also like to point out that the cavity lasers can also be focused and probe only parts of the atomic system . then
, @xmath27 accounts only for the atoms within the mode volume of the lasers , and the stark shift becomes a local shift in the chemical potential .
we consider the terms involving the density operator @xmath28 as small perturbations and solve the system first in absence of these coupling terms .
then , the quantum langevin equation for the cavity fields reduce to @xmath29 where @xmath30 are the input fields of the cavities @xcite .
we assume that the cavity is driven by two lasers with frequencies @xmath9 and @xmath10 , i.e. , @xmath31 .
then , the steady state solution for the averaged fields @xmath32 is given by @xmath33 to simplify the analysis , we set the cavity detunings to zero , @xmath34 and assume equal driving fields @xmath35 , while the generalization to unequal driving fields is straightforward .
furthermore , we can absorb a phase factor in the definition of @xmath36 such that @xmath37 and @xmath38 with @xmath39 . to study the effect of the atom - field interaction , we consider small fluctuations around the free field solution .
for this , we define a new set of operators as @xmath40 with the inverse transformation @xmath41 these operators have the expectation values @xmath42 and @xmath43 , and we expand them into their mean values and small fluctuations @xmath44 , @xmath45 .
then , the interaction hamiltonian @xmath46 reduces to @xmath47 with @xmath48 , the potential strength @xmath49 , and the effective coupling @xmath50 .
the first term describes a classical driving field @xmath51 for the atomic system , while the last terms account for the coupling between the atomic system and the cavity fields . for @xmath52 ,
this set of equations reduces to the case previously studied in the context of cavity cooling @xcite .
we now introduce the quadrature operators @xmath53\hspace{0.5 cm } & p_b & = -{\textup{i}}\left[{\delta b}-{\delta b^\dagger}\right]\\ x_a & = \frac{1}{2}\left[{\delta a}+ { \delta a^\dagger}\right ] & p_a & = -{\textup{i}}\left[{\delta a}-{\delta a^\dagger}\right]\end{aligned}\ ] ] with the commutation relations @xmath54 & = [ x_b , p_b ] = & { \textup{i}}\\ { } [ x_b , p_a ] & = [ x_a , p_b ] = & 0.\end{aligned}\ ] ] the equation of motions for the cavity field then reduce to @xmath55-\frac{\gamma}{2}p_a\\ \dot{x}_b & = & { \textup{i}}\frac{g_{\mathrm{eff}}}{\hbar}[\rho_{{\bf q}}{\textup{e}}^{-{\textup{i}}\omega t}-\rho_{-{{\bf q}}}{\textup{e}}^{{\textup{i}}\omega t}]-\frac{\gamma}{2}x_b\\ \dot{p}_b & = & -\frac{\gamma}{2}p_b.\end{aligned}\ ] ] these equations describe the back action of the atomic system with the particle density @xmath20 onto the cavity fields , i.e. , a fluctuation in the particle density operator of the atomic system @xmath56 influences the cavity fields within characteristic quantum - non - demolition setup .
the leading term @xmath57 drives a perturbation onto density of the atomic system with the field @xmath58 .
the condition of a weak probe field reduces to @xmath59 with @xmath60 the characteristic energy scale of the atomic system @xcite .
then , the response of the atomic system to this external probe is well described within linear response theory @xmath61 which implies for the present drive with momentum transfer @xmath62 and frequency @xmath63 the response @xmath64 . then
, it is possible to replace the operators in the langevin equations by their expectation values , and we obtain the coupled differential equations @xmath65 we can immediately identify the steady state solution @xmath66 consequently , the linear response of the atomic system is imprinted onto the cavity fields : the imaginary part of the response function describing the creation of two - particle excitations and collective excitations in the atomic system is encoded by a shift in the amplitude operator @xmath67 accounting for the scattered of photons from one cavity mode onto the other . in turn , the real part of the response function characterizing the dispersive part of the media leads to a shift in the phase @xmath68 of the fields .
finally , we provide the experimental parameters for the measurement of the response function .
we will focus on the measurement of the phase quadrature @xmath69 , as it exhibits two advantages over the measurement of @xmath70 : first , we note that @xmath71 is directly given by the sum of the expectation values @xmath72 and @xmath73 of the physical cavity fields @xmath74 and @xmath75 , which can be efficiently measured within a homodyne detection . for appropriate phases of the cavity fields
this means that the measurements are done against zero background , which is not the case when measuring @xmath76 .
second , the kramers - kronig relation eq .
( [ kramerskronig ] ) allows us to compute @xmath77 _ on resonance _ by using measurement data taken _ off resonance_. therefore , we are able to obtain the dynamic structure factor of the atomic system without transferring energy onto the atomic system .
as the sum of the physical observables @xmath78 , @xmath79 .
a signal can be detected only when the ratio @xmath80 is large compared to phase fluctuations . ]
the validity of the expansion and ignoring the quadratic terms in the fluctuating fields requires a large photon number in the cavity , i.e. , @xmath81 . in turn
, the weak driving field @xmath82 implies a weak coupling @xmath83 .
such weak coupling is easily achieved by controlling the detuning to the excited level , see fig .
[ fig : excite ] . note , that his weak coupling is in high contrast to the recent experimental works on high finess cavities @xcite , and significantly simplifies the requirements for the experimental realization of such a setup .
the phase quadratures can be efficiently been measured within a homodyne detection with the sensitivity , see fig .
[ fig : paxa ] , @xmath84 the range of interesting values for the response function are @xmath85 with @xmath13 the total number of particles and @xmath86 the characteristic energy scale of the atomic system .
this requires a sensitivity to measure the phase angle @xmath87 for the last equation , we have used the relation between the driving field @xmath88 and the number of photons in the cavity .
the number of atoms within the cavity are typically in the range @xmath89 , and therefore , with a cavity decay rate @xmath90 corresponding to a quality factor @xmath91 with up to 100 photons within the cavity , and a characteristic energy @xmath92 , the required sensitivity is in the range @xmath93 , which can be achieved with present techniques @xcite .
in summary we have demonstrated the experimental feasibility of an _ in situ _ measurement of the dynamic structure factor of arbitrary atomic systems , providing a novel tool for the coherent manipulation of ultracold quantum gases .
the proposed method employs two cavity fields which are driven by external lasers .
we have shown that the required resolution can be achieved using a cavity finesse that has already been reached in present experiments .
furthermore , the method allows for sufficiently fast measurements .
the current setup is in close analogy to the recently experimentally realized system @xcite .
the main difference is , that in the experimental setup a strong imbalance between the two photon modes is applied . as a consequence ,
the validity of the expansion up to linear order in the fluctuating fields is satisfied by the strong coupling laser , while for the weak probe beam a very low photon number can be used .
then , the present calculations can be performed again in a straightforward manner with the only modification , that the real part of the response function is written on the phase quadrature @xmath94 , while the imaginary part is encoded onto the amplitude quadrature @xmath95 . in the experiment
the amplitude quadrature was accessed with a heterodyne detection @xcite , while the real part of the response function again could be probed by a simpler homodyne detection .
we would like to point out that the contrast of the signal can be strongly enhanced by using a cavity for the weak probe field .
finally , we would like to point out , that the present setup can also be used to study the relaxation of excitations in the atomic system : first , the lasers imprint several excitations on resonance into the atomic system , while at a later stage the probe of the real part allows one to analyze whether these excitations are still present or have relaxed .
we thank r. lw for fruitful discussions on the experimental parameters .
the work was supported by a grant from the army research office with funding from the darpa ole program , by the deutsche forschungsgemeinschaft ( dfg ) within sfb / trr 21 , by the national science foundation through a grant for the institute for theoretical atomic , molecular and optical physics at harvard university and smithsonian astrophysical observatory , and by a fellowship within the postdoc program of the german academic exchange service ( daad ) . | we propose an experimental setup to efficiently measure the dynamic structure factor of ultracold quantum gases .
our method uses the interaction of the trapped atomic system with two different cavity modes , which are driven by external laser fields . by measuring the output fields of the cavity the dynamic structure factor of the atomic system
can be determined .
contrary to previous approaches the atomic system is not destroyed during the measurement process . |
a superfluid flows without friction but becomes dissipative above a certain critical velocity @xmath1 via generating its elementary excitations such as phonons and vortices .
the landau criterion provides a conventional energetic consideration to determine the critical velocity , stating @xmath10 $ ] @xcite , where @xmath11 is the energy of an elementary excitation of momentum @xmath12 . for a homogeneous system ,
the landau critical velocity is equal to the speed of sound @xmath5 .
however , the dynamic response of a superfluid flow is significantly sensitive to the boundary condition of the system and hence a quantitative understanding of the critical velocity has been a challenging task in the study of superfluidity .
one of the paradigmatic situations considered in fluid mechanics is a two - dimensional ( 2d ) flow past a circular cylinder . for an incompressible flow ,
the local velocity is increased by a factor of 2 at the lateral sides of the cylinder @xcite and the local landau supersonic criterion suggests a critical velocity of @xmath13 that is independent of the radius @xmath14 of the cylinder .
theoretical studies showed that the onset of dissipation involves generating a counter - rotating vortex pair @xcite .
more rigorous calculations , taking into account the compressibility of the superfluid and quantum pressure near the boundary of the cylinder , predicted that the critical velocity converges to @xmath15 in the large cylinder limit @xmath16 @xcite , where @xmath8 is the superfluid healing length .
experimental verification of the predictions on @xmath9 is highly desirable . in previous ultracold atom experiments ,
a similar situation was investigated by stirring superfluid samples with a repulsive laser beam @xcite .
the existence of finite critical velocities @xcite and generation of vortex dipoles @xcite were successfully demonstrated . however , the measured values of @xmath9 ranged widely from 0.1 to 0.45 , which did not allow a quantitative study of the homogeneous 2d problem .
theoretical investigations showed that the inhomogeneous density distribution of a trapped sample @xcite , 3d vortex dynamics @xcite , or the manner of stirring @xcite should be critical in the measurements .
recently , the vortex shedding dynamics was also investigated experimentally with polariton superfluids flowing past static defects @xcite . in this paper
, we systematically study the critical velocity for vortex shedding in a bose - einstein condensate with a repulsive gaussian potential .
we measure the critical velocity as a function of the barrier height @xmath0 of the potential over a wide range of the beam width , @xmath17 . in particular , in order to address the 2d homogeneous regime , we employ spatially large and highly oblate condensates , ensuring vortex dynamics in two dimensions .
the key difference of a gaussian potential from a hard cylinder is its soft boundary .
a gaussian potential , @xmath18 , produces a density - depleted hole in the condensate when @xmath19 , where @xmath3 is the chemical potential of the condensate .
the radius of the hole and the potential slope at the hole boundary are given as @xmath20 respectively . in comparison to the case with a hard cylinder
, the soft boundary reduces the density in the proximity of the obstacle and consequently lowers the local speed of sound .
then , from the local landau criterion it is naturally suggested that the critical velocity of the hard cylinder defines an upper bound for that of the obstacle formed by the gaussian potential . when the hole radius @xmath14 becomes larger with higher @xmath0 [ fig .
1(b ) ] and/or the beam width @xmath7 decreases for fixed @xmath14 [ fig .
1(c ) ] , the obstacle would converge to the hard cylinder with stiffening its boundary .
the main result of our measurements is that in the deep non - penetrable regime ( i.e. , @xmath21 ) , the critical velocity @xmath1 increases with decreasing @xmath6 for fixed @xmath22 and approaches about @xmath23 .
this observation is consistent with the expectation from the aforementioned discussion based on the local landau criterion .
furthermore , the measured value of @xmath9 with our stiffest obstacle is in good agreement with theoretical predictions for a 2d superfluid flow past a hard cylinder . .
( a ) normalized potential slope at the obstacle boundary @xmath24 as a function of @xmath6 and @xmath25 , where @xmath3 and @xmath8 is the chemical potential and the healing length of the condensate , respectively , and @xmath14 is the obstacle radius such that @xmath26 . the boundary becomes stiffer as ( b ) @xmath25 becomes larger or ( c ) @xmath6 decreases .
the corresponding trajectories are indicated by the arrows in ( a).,width=328 ]
our experiment starts with a bose - einstein condensate of @xmath27na atoms in a harmonic trap formed by combining optical and magnetic potentials @xcite .
the condensate fraction of the sample is over 90% . in a typical sample condition , where the trapping frequencies are @xmath28 = @xmath29 hz and the atom number of the condensate is @xmath30 , the condensate healing length is @xmath31 m and the speed of sound is @xmath32mm / s at the trap center , where @xmath33 is the planck constant divided by 2@xmath34 and @xmath35 is the atomic mass . by adjusting the trapping frequencies or the atom number of the condensate , @xmath8
is varied up to @xmath36 m . the thomas - fermi radius and thickness of the condensate
are @xmath37 and @xmath38 , respectively . in this highly oblate condensate , vortex line excitations
are strongly suppressed @xcite and the vortex dynamics is expected to be two dimensional .
we adiabatically ramp up the power of a repulsive gaussian laser beam in 1 s and hold it for 0.2 s to ensure that the condensate is stationary .
then , we translate the laser beam horizontally by @xmath39 m by using a piezo - driven mirror [ fig . 2(a ) ] .
the velocity @xmath40 of the laser beam is kept constant during the translation and controlled by adjusting the traveling time .
the sweeping region of the laser beam is centered in the condensate [ fig 2(b ) ] .
the density variation over the sweeping region is less than 10% and the speed of sound can be well approximated to be spatially constant . after completing the sweeping , we slowly ramp down the laser beam power for 0.5 s , and take an absorption image of the condensate after expansion by releasing the trapping potential to detect vortices . of having vortex dipoles ( blue circles ) and the number of vortices ( red diamonds ) as a function of the velocity @xmath40 of the optical obstacle .
the barrier height @xmath41 and the gaussian width @xmath42 .
the solid line is a sigmoidal function fit to @xmath43 .
the inset shows an image of a condensate containing a vortex dipole .
each data point was obtained from 15 realizations of the same experiment and the error bars indicate the standard deviations of the measurements.,width=294 ] the width @xmath7 of the laser beam is calibrated from the in - situ images of very dilute samples penetrated through by the laser beam , taking into account the imaging resolution of our system ( @xmath44 m ) .
the beam waist of the laser beam is @xmath45 m at the focal plane and the gaussian width @xmath7 of the optical obstacle is controlled by defocusing the laser beam at the sample plane .
the sample thickness ( @xmath46 m ) is much shorter than the rayleigh length of the laser beam and we ignore the beam divergence .
the beam width is much smaller than the condensate radius and the chemical potential of the condensate is negligibly affected by the presence of the laser beam .
a vortex dipole is identified with two density - depleted holes that are symmetrically located in the condensate with respect to the sweeping line of the laser beam [ fig .
2(c ) inset ] .
after beinggenerated in the center region of the condensate , the vortex dipole moves toward the edge of the condensate and splits into two individual vortices that subsequently travel along the boundary of the condensate in the opposite direction .
this peculiar obrit motion of the vortices was studied in ref .
a single vortex was observed occasionally at low @xmath40 with probability @xmath47 . because the vortex lifetime is over 10 s @xcite
, we attribute the single vortex to uncontrolled perturbations in sample preparation and we do not count it as a vortex dipole .
the critical velocity @xmath1 for vortex shedding is determined from the probability distribution @xmath43 for having vortex dipoles after sweeping with the laser beam . here
@xmath43 is obtained from 15 realizations of the same experiment with a given sweeping velocity @xmath40 , i.e. , @xmath43 is the ratio of the number of images showing vortex dipoles to the total number of measurements .
the critical velocity @xmath1 is determined by fitting a sigmoidal function to the probability distribution as @xmath48 [ fig .
we use the value of @xmath49 as the measurement uncertainty of @xmath50 , corresponding to the range of @xmath51 .
figure 3(a ) displays the results of the critical velocity as a function of the barrier height @xmath0 for various beam widths .
the critical velocity shows a dip structure having a minimum at @xmath52 , clearly distinguishing the two regimes : a penetrable regime with @xmath53 and a non - penetrable regime with @xmath19 .
the dip structure of the critical velocity can be accounted for by a consideration based on the local landau criterion . for @xmath53 ,
the density minimum is located at the top of the gaussian potential and decreases for higher @xmath0 , and thus , lowering the local speed of sound . on the other hand , when @xmath19 , as described before , the potential slope becomes steeper with higher @xmath0 and the density in the proximity of the obstacle boundary is gradually restored back to the bulk density , leading to a higher critical velocity .
the dip structure around @xmath54 becomes more pronounced with larger @xmath6 by lowering the minimum value [ fig .
3(a ) inset ] . versus the relative barrier height @xmath55 for various beam widths @xmath6 : @xmath56 $ ] ( red diamonds ) , @xmath57 $ ] ( black squares ) , @xmath58 $ ] ( blue circles ) , and @xmath59~\mu$]m ( open squares ) .
the inset displays @xmath60 at @xmath61 as a function of @xmath6 and the solid line is a line to guide the eye .
( b ) same data in the non - penetrable regime ( @xmath19 ) as a function of the hole radius @xmath25 , together with additional data points for @xmath62.,width=313 ] in terms of the vortex shedding mechanism , the penetrable regime is different because there is no density - depleted region in the fluid and it has been anticipated that vortex nucleation would be initiated by generation of rarefaction pulses @xcite .
recently , the critical velocity of penetrable obstacles for dissipation was investigated in effective 1d systems @xcite . in one dimension ,
the critical velocity is predicted to vanish as @xmath0 approaches @xmath3 @xcite but it is not the case in our 2d situation .
moreover , the critical velocity shows a quite intriguing dependence on @xmath6 [ fig .
3(a ) inset ] .
further investigation of the functional form of @xmath63 in the penetrable regime is warranted .
however , in this work , we focus on the non - penetrable regime to address the 2d hard cylinder situation . in fig .
3(b ) , we recast the data for the non - penetrable regime ( @xmath19 ) as a function of the hole radius @xmath14 , together with additional data obtained for @xmath64 with @xmath65 and @xmath66 .
note that the hole radius is weakly dependent on the barrier height as @xmath67 . from the previous discussion of the boundary stiffness effect and @xmath68 [ fig .
1(b ) ] , one may expect that @xmath9 would be saturated to a certain value with increasing hole radius and also that the saturation behavior would be faster with smaller beam width .
we see that the experiment data are roughly fit to the expectation in a small @xmath14 region .
however , when @xmath14 is increased to be larger than @xmath7 , the growth rate of the critical velocity slows down and becomes negative .
note that at @xmath69 , the critical velocity with small @xmath70 is even lower than that with large @xmath71 .
one possible explanation for such climbing - over behavior of the critical velocity is the imperfection of the laser beam profile .
if the beam profile is not perfectly gaussian , for example , the intensity profile of the outer part of the laser beam decays slower than exponential and then the potential slope @xmath72 at the obstacle boundary would decrease with increasing @xmath14 .
we see that the climbing - over of @xmath1 occurs at @xmath73 in both of the data sets with @xmath70 and 35 .
this seems to support the beam profile effect because in our experiment @xmath7 is varied by defocusing the same laser beam .
the @xmath74 factor of the laser beam is measured to be 1.2 . to further investigate the soft boundary effect on the critical velocity , we take a different scanning trajectory in the parameter space of the gaussian obstacle : decreasing @xmath6 for fixed @xmath22 . in this
setting , the hole radius @xmath14 is also varied proportionally with @xmath7 [ eq .
( 1 ) ] and the trajectory corresponds to a diagonal line in fig .
1(a ) . a scanning with fixed @xmath25 , as depicted in fig .
1(c ) , might be more ideal in terms of isolating the finite-@xmath25 effect @xcite , but this would require exponentially high @xmath22 for small @xmath6 as @xmath75 , necessarily recalling the outer part of the laser beam .
we set @xmath76 , where @xmath77 and the normalized boundary slope @xmath78 . in the previous measurements ( fig .
3 ) , @xmath79 shows a maximum around this potential height . in the new set of measurements , we observe that the critical velocity monotonically increases as @xmath6 decreases ( fig
. 4 ) , which is consistent with our expectation from the boundary stiffness .
it is worth noting that although the data are obtained from various samples with different healing lengths , they agree with each other in the plane of the dimensionless parameters @xmath6 and @xmath9 .
this demonstrates the 2d character of the vortex dynamics in our system because the healing length is the only relevant length scale in 2d superfluid hydrodynamics . versus @xmath6 for fixed @xmath80 . here
@xmath81 m for black closed squares and @xmath82 m for black closed circles .
the blue open circles show the results of theoretical calculation for a 2d hard cylinder of radius @xmath83 from ref .
the red open diamonds indicate the numerical results for a gaussian potential with @xmath84 from ref .
@xcite , where the potential slope @xmath85 would be about two times higher than ours .
the dashed line denotes the theoretically predicted value , @xmath86 for a 2d homogeneous case with a hard cylinder in the large obstacle limit @xcite.,width=302 ] for our stiffest obstacle , @xmath9 is measured to be about 0.40 , which is slightly higher than the predicted value of 0.37 for a 2d hard cylinder in the large-@xmath25 limit . in our measurement
, @xmath87 marginally satisfies the large obstacle condition , and the deviation of the measured value might be attributed to the finite-@xmath25 effect .
the dependence of the critical velocity on @xmath25 was investigated theoretically @xcite and it was shown that @xmath9 gradually increases from the value of 0.37 as @xmath88 . in fig . 4 , for comparison , we display the theoretical results of ref .
@xcite for a 2d hard cylinder of radius @xmath89 ( blue open circles ) and the result of numerical simulations performed with a gaussian potential with @xmath84 in ref .
@xcite ( red open diamonds ) .
the experimental results converge to the theoretical predictions when @xmath6 decreases , i.e. , the optical obstacle becomes similar to a hard cylinder by stiffening its boundary .
our results are inconsistent with the @xmath90 dependence of @xmath1 that was predicted from analytic analyses on the stability of a superfluid flow @xcite .
it is clearly seen that when @xmath91 increases by a factor of 5 , the @xmath9 decreases less than a factor of 2 . even without including the additional reduction effect due to the soft boundary of the obstacle , the critical velocity decreases much smaller than what would be expected from the @xmath90 dependence .
finally , we want to recall a few aspects of the experimental condition that should be considered for the quantitative comparison of the measurement results to theoretical predictions for a 2d homogeneous case .
first , the critical velocity was not measured for a steady flow condition but by sweeping a finite section of a trapped condensate .
the measurement can be affected by the sweeping manner .
for example , if vortex nucleation requires a finite time , which might be longer than the sweeping time near the critical velocity @xcite , it would result in a systematic , upward shift of the measured value of @xmath1 .
second , although we expect suppression of 3d vortex dynamics in a highly oblate condensate , it is an inevitable fact that the condensate has an inhomogeneous density distribution along the axial direction .
thus , one can not completely ignore 3d responses of the condensate , in particular , at the moment of vortex nucleation .
it might be necessary or sufficient to introduce an effective speed of sound that would be lower than the peak value @xmath5 of the condensate @xcite .
we have presented the measurements of the critical velocity for vortex shedding in highly oblate bose - einstein condensates and investigated the soft boundary effect of the moving obstacle formed by a gaussian potential .
our results are consistent with a picture based on the local landau criterion and the measure value of @xmath9 with the stiffest obstacle is in good agreement with the theoretical predictions for a homogeneous 2d superflow past a cylindrical object .
this work has established a reliable experimental method to measure the critical velocity of a trapped condensate and its intriguing extension is to investigate the temperature dependence of the critical velocity , which might provide a new setting to study the role of thermal atoms in vortex nucleation @xcite .
l. d. landau , j. phys .
ussr * 5 * , 71 ( 1941 ) .
l. d. landau and e. m. lifshitz , _ fluid mechanics _ ( pergamon press , oxford , 1987 ) .
t. frisch , y. pomeau , and s. rica , phys .
lett . * 69 * , 1644 ( 1992 ) .
b. jackson , j. f. mccann , and c. s. adams , phys .
lett . * 80 * , 3903 ( 1998 ) . c. huepe and m. e. brachet , physica d * 140 * 126 ( 2000 ) .
n. g. berloff and p. h. roberts , j. phys .
a : math . gen . * 33 * , 4025 ( 2000 ) .
s. rica , physica d * 148 * 221 ( 2001 ) .
pham , c. nore , and m. e. brachet , physica d * 210 * , 203 ( 2005 ) .
f. pinsker and n. g. berloff , phys .
rev . a * 89 * , 053605 ( 2014 ) . c. raman , m. khl , r. onofrio ,
d. s durfee , c. e. kuklewicz , z. hadzibabic , and w. ketterle , phys .
lett . * 83 * , 2502 ( 1999 ) .
r. onofrio , c. raman , j. m. vogels , j. r. abo - shaeer , a. p. chikkatur , and w. ketterle , phys .
lett . * 85 * , 2228 ( 2000 ) .
s. inouye , s. gupta , t. rosenband , a. p. chikkatur , a. grlitz , t. l. gustavson , a. e leanhardt , d. e. pritchard , and w. ketterle , phys .
87 * , 080402 ( 2001 ) .
t. w. neely , e. c. samson , a. s. bradley , m. j. davis , and b. p. anderson , phys .
lett . * 104 * , 160401 ( 2010 ) .
r. desbuquois , l. chomaz , t. yefsah , j. lonard , j. beugnon , c. weitenberg , and j. dalibard , nat . phys .
* 8 * , 645 ( 2012 ) .
w. j. kwon , g. moon , j. choi , s. w. seo , and y. shin , phys .
a * 90 * , 063627 ( 2014 ) .
t. winiecki , b. jackson , j. f. mccann , and c. s. adams , j. phys .
b : at . mol .
. phys . * 33 * , 4069 ( 2000 ) .
p. o. fedichev and g. v. shlyapnikov , phys .
a * 63 * , 045601 ( 2001 ) .
m. crescimanno , c. g. koay , r. peterson , and r. walsworth , phys .
a * 62 * , 063612 ( 2000 ) . c. nore , c. huepe , and m. e. brachet , phys .
lett . * 84 * , 2191 ( 2000 ) .
b. jackson , j. f. mccann , and c. s. adams , phys .
rev . a * 61 * , 051603(r ) ( 2000 ) .
a. aftalion , q. du , and y. pomeau , phys .
lett . * 91 * , 090407 ( 2003 ) .
k. fujimoto and m. tsubota , phys .
a * 83 * , 053609 ( 2011 ) .
g. w. stagg , a. j. allen , n. g. parker , and c. f. barenghi , phys .
rev . a * 91 * , 013612 ( 2015 ) .
g. nardin , g. grosso , y. lger , b. pietka , f. morier - genoud , and b. deveaud - pldran , nat . phys .
* 7 * , 635 ( 2011 ) .
a. amo , s. pegeon , d. sanvitto , v. g. sala , r. hivet , i. carusotto , f. pisanello , g. lemnager , r. houdr , e. giacobino , c. ciuti , and a. bramati , science * 332 * , 1167 ( 2011 ) .
b. jackson , n. p. proukakis , c. f. barenghi , and e. zaremba , phys .
a * 79 * , 053615 ( 2009 ) .
s. j. rooney , p. b. blakie , b. p. anderson , and a. s. bradley , phys .
a * 84 * , 023637 ( 2011 ) .
t. aioi , t. kadokura , t. kishimoto , and h. saito , phys .
x * 1 * , 021003 ( 2011 ) .
b. jackson , ph.d .
thesis , durham university , 2000 .
p. engels and c. atherton , phys .
lett . * 99 * , 160405 ( 2007 ) .
m. albert , t. paul , n. pavloff , and p. leboeuf , phys .
* 100 * , 250405 ( 2008 ) .
d. dries , s. e. pollack , j. m. hitchcock , and r. g. hulet , phys . rev .
a * 82 * , 033603 ( 2010 ) .
k. c. wright , r. b blakestad , c. j. lobb , w. d. phillips , and g. k. campbell , phys .
a * 88 * , 063633 ( 2013 ) .
g. w. stagg , n. g. parker , and c. f. barenghi , j. phys .
b : at . mol .
phys . * 47 * , 095304 ( 2014 ) .
j. s. stieberger and w. zwerger , phys .
a * 62 * , 061601(r ) ( 2000 ) .
g. huang , v. a. makarov , and m. g. velarde , phys .
a * 67 * , 023604 ( 2003 ) . j. e. williams , e. zaremba , b. jackson , t. nikuni , and a. griffin , phys .
88 * , 070401 ( 2002 ) .
a. a. penckwitt , r. j. ballagh , and c. w. gardiner , phys .
* 89 * , 260402 ( 2002 ) .
k. kasamatsu , m. tsubota , and m. ueda , phys .
a * 67 * , 033610 ( 2003 ) . | we present measurements of the critical velocity for vortex shedding in a highly oblate bose - einstein condensate with a moving repulsive gaussian laser beam . as a function of the barrier height @xmath0
, the critical velocity @xmath1 shows a dip structure having a minimum at @xmath2 , where @xmath3 is the chemical potential of the condensate . at fixed @xmath4 , we observe that the ratio of @xmath1 to the speed of sound @xmath5 monotonically increases for decreasing @xmath6 , where @xmath7 is the beam width and @xmath8 is the condensate healing length .
we explain our results with the density reduction effect of the soft boundary of the gaussian obstacle , based on the local landau criterion for superfluidity .
the measured value of @xmath9 with our stiffest obstacle is about 0.4 , which is in good agreement with theoretical predictions for a two - dimensional superflow past a circular cylinder . |
the seesaw mechanism is implemented by the inclusion of a majorana mass @xmath6 for the right handed neutrinos ( allowed due to their singlet character under all the symmetries of the standard model ( sm ) ) and by considering a large separation between this mass and the electroweak ( ew ) scale @xcite .
after ew symmetry breaking , the full @xmath7 neutrino mass matrix is given in terms of the @xmath8 dirac mass matrix , @xmath9 , and the @xmath8 majorana mass matrix @xmath6 . here
@xmath10 is the @xmath8 neutrino yukawa coupling and @xmath11 with @xmath12 gev .
the ratio of the two higgs doublets vacuum expectations values is @xmath13 .
the assumption of @xmath14 leads to the usual seesaw equation , @xmath15 , which guaranties the smallness of the light neutrino masses .
after the diagonalisation of the full @xmath7 neutrino mass matrix one obtains six physical majorana neutrinos : three light @xmath16 , with masses @xmath17 , and three heavy @xmath18 , with masses @xmath19 .
notice that we work in a lepton basis where both the right handed mass matrix and the charged lepton mass matrix are diagonal in flavour space .
the flavour mixing in the light neutrino sector is given by the maki - nakagawa - sakata matrix @xmath20 @xcite for which we use the standard parameterization , which is written in terms of three mixing angles @xmath21 , @xmath5 and @xmath22 and three cp violating phases @xmath23 , @xmath24 and @xmath25 .
we use here the parameterisation proposed in ref .
@xcite , where the solution to the seesaw equation is written as @xmath26 , with @xmath27 being a @xmath28 orthogonal complex matrix , defined by three complex angles @xmath29 ( @xmath2 ) .
the attractiveness of this parameterisation is that it allows to easily implement the requirement of compatibility with low energy neutrino data .
it also clearly shows that in the singlet seesaw scenario one can have large neutrino yukawa couplings , @xmath30 , by simply choosing large entries in @xmath31 .
the main implication of these large yukawa coupling is that they can induce large lepton flavour violating ( lfv ) rates @xcite .
the total number of parameters of the neutrino sector in this scenario is 18 , which in this particular parameterisation are summarised by @xmath32 , @xmath23 , @xmath24 , @xmath25 , @xmath33 , @xmath34 and @xmath29 . by adjusting the light neutrino parameters to the low energy neutrino data , one
is left with 9 input parameters given by @xmath34 and @xmath29 . regarding the numerical estimates we consider two scenarios .
the first one with quasi - degenerate light neutrinos , with masses @xmath35 ev , @xmath36 and @xmath37 , and degenerate heavy neutrinos with mass @xmath38 .
the second one is with hierarchical light and heavy neutrinos , with masses @xmath39 and @xmath40 .
here we use @xmath41 ev , @xmath42 ev , @xmath43 , @xmath44 , @xmath45 , and for simplicity we fix @xmath46 .
in addition to the previous seesaw parameters , there are the susy sector parameters which , within the assumed constrained minimal supersymmetric standard model ( cmssm ) framework , are given by @xmath47 and @xmath48 . the universality of the soft - susy breaking terms is imposed at a high scale @xmath49 which we fix here to the @xmath50-@xmath51 gauge couplings unification scale @xmath52 gev . in particular ,
in the numerical analysis we consider specific choices of these parameters , given by the msugra - like `` snowmass points and slopes '' ( sps ) @xcite listed in table [ sps : def:15 ] , which represent different examples of possible susy spectra . [ cols="^,^,^,^,^,^",options="header " , ] there are several important conclusions to be drawn from fig .
[ fig : modt2:argt2:1214 ] .
let us first discuss the case @xmath53 ev .
we note that one can obtain a baryon asymmetry in the range @xmath54 to @xmath55 for a considerable region of the analysed @xmath56 range .
notice also that there is a clear separation between the predictions of @xmath57 and @xmath58 , with the latter well above the present experimental bound .
this would imply an experimental impact of @xmath5 , in the sense that the br predictions become potentially detectable for this non - vanishing @xmath5 value . with the planned meg sensitivity @xcite
, both cases would be within experimental reach . however , this statement is strongly dependent on the assumed parameters , in particular @xmath59 .
for instance , a larger value of @xmath60 ev , illustrated on the right panel of fig .
[ fig : modt2:argt2:1214 ] , leads to a very distinct situation regarding the sensitivity to @xmath5 .
while for smaller values of @xmath56 the branching ratio displays a clear sensitivity to having @xmath5 equal or different from zero ( a separation larger than two orders of magnitude for @xmath61 ) , the effect of @xmath5 is diluted for increasing values of @xmath56 .
let us now address the question of whether a joint measurement of the brs and @xmath5 can shed some light on experimentally unreachable parameters , like @xmath62 .
the expected improvement in the experimental sensitivity to the lfv ratios supports the possibility that a br could be measured in the future , thus providing the first experimental evidence for new physics , even before its discovery at the lhc .
the prospects are especially encouraging regarding @xmath63 , where the experimental sensitivity will improve by at least two orders of magnitude . moreover , and given the impressive effort on experimental neutrino physics , a measurement of @xmath5 will likely also occur in the future @xcite . given that , as previously emphasised , @xmath64 is very sensitive to @xmath5 , whereas this is not the case for br(@xmath65 ) , and that both brs display the same approximate behaviour with @xmath62 and @xmath66 , we now propose to study the correlation between these two observables .
this optimises the impact of a @xmath5 measurement , since it allows to minimise the uncertainty introduced from not knowing @xmath66 and @xmath62 , and at the same time offers a better illustration of the uncertainty associated with the @xmath27-matrix angles . in this case
, the correlation of the brs with respect to @xmath62 means that , for a fixed set of parameters , varying @xmath62 implies that the predicted point ( br(@xmath65 ) , br(@xmath67 ) ) moves along a line with approximately constant slope in the br(@xmath65)-br(@xmath67 ) plane . on the other hand ,
varying @xmath5 leads to a displacement of the point along the vertical axis . in fig .
[ fig : doublebr ] , we illustrate this correlation for sps 1a , choosing distinct values of the heaviest neutrino mass , and we scan over the bau - enabling @xmath27-matrix angles ( setting @xmath68 to zero ) as @xmath69 we consider the following values , @xmath70 , @xmath71 , @xmath72 and @xmath73 , and only include in the plot the br predictions which allow for a favourable bau .
other sps points have also been considered but they are not shown here for brevity ( see @xcite ) .
we clearly observe in fig .
[ fig : doublebr ] that for a fixed value of @xmath62 , and for a given value of @xmath5 , the dispersion arising from a @xmath74 and @xmath75 variation produces a small area rather than a point in the br(@xmath76)-br(@xmath67 ) plane . the dispersion along the br(@xmath65 ) axis is of approximately one order of magnitude for all @xmath5 .
in contrast , the dispersion along the br(@xmath64 ) axis increases with decreasing @xmath5 , ranging from an order of magnitude for @xmath77 , to over three orders of magnitude for the case of small @xmath5 ( @xmath78 ) . from fig .
[ fig : doublebr ] we can also infer that other choices of @xmath62 ( for @xmath79 $ ] ) would lead to br predictions which would roughly lie within the diagonal lines depicted in the plot . comparing these predictions for the shaded areas along the expected diagonal `` corridor '' , with the allowed experimental region ,
allows to conclude about the impact of a @xmath5 measurement on the allowed / excluded @xmath62 values .
the most important conclusion from fig .
[ fig : doublebr ] is that for sps 1a , and for the parameter space defined in eq .
( [ doublebr : input ] ) , an hypothetical @xmath5 measurement larger than @xmath78 , together with the present experimental bound on the br(@xmath64 ) , will have the impact of excluding values of @xmath80 gev .
moreover , with the planned meg sensitivity , the same @xmath5 measurement can further constrain @xmath81 gev .
the impact of any other @xmath5 measurement can be analogously extracted from fig .
[ fig : doublebr ] . as a final comment
let us add that , remarkably , within a particular susy scenario and scanning over specific @xmath74 and @xmath75 bau - enabling ranges for various values of @xmath5 , the comparison of the theoretical predictions for br(@xmath64 ) and br(@xmath65 ) with the present experimental bounds allows to set @xmath5-dependent upper bounds on @xmath62 .
together with the indirect lower bound arising from leptogenesis considerations , this clearly provides interesting hints on the value of the seesaw parameter @xmath62 . with the planned future sensitivities
, these bounds would further improve by approximately one order of magnitude .
ultimately , a joint measurement of the lfv branching ratios , @xmath5 and the sparticle spectrum would be a powerful tool for shedding some light on otherwise unreachable susy seesaw parameters .
it is clear from this study that the connection between lfv and neutrino physics will play a relevant role for the searches of new physics .
m.j . herrero would like to thank alberto lusiani for the invitation to participate in this interesting and fruitful conference .
she also acknowledges project fpa2003 - 04597 of spanish mec for finantial support .
e. arganda and m. j. herrero , phys . rev .
d * 73 * ( 2006 ) 055003 [ arxiv : hep - ph/0510405 ] .
s. antusch , e. arganda , m. j. herrero and a. teixeira , arxiv : hep - ph/0607263 , to appear in jhep .
p. minkowski , phys .
b * 67 * ( 1977 ) 421 ; m. gell - mann , p. ramond and r. slansky , in _ complex spinors and unified theories _ eds .
p. van .
nieuwenhuizen and d. z. freedman , _ supergravity _ ( north - holland , amsterdam , 1979 ) , p.315 [ print-80 - 0576 ( cern ) ] ; t. yanagida , in _ proceedings of the workshop on the unified theory and the baryon number in the universe _ , eds . o. sawada and a. sugamoto ( kek , tsukuba , 1979 ) , p.95 ; s. l. glashow , in _ quarks and leptons _ , eds . m. lvy _ et al . _ ( plenum press , new york , 1980 ) , p.687 ; r. n. mohapatra and g. senjanovi , phys .
* 44 * ( 1980 ) 912 .
z. maki , m. nakagawa and s. sakata , prog .
* 28 * ( 1962 ) 870 ; b. pontecorvo , sov .
jetp * 6 * ( 1957 ) 429 [ zh .
fiz . * 33 * ( 1957 ) 549 ] ; sov .
jetp * 7 * ( 1958 ) 172 [ zh .
* 34 * ( 1957 ) 247 ] .
j. a. casas and a. ibarra , nucl .
b * 618 * ( 2001 ) 171 [ arxiv : hep - ph/0103065 ] .
f. borzumati and a. masiero , phys .
* 57 * ( 1986 ) 961 .
b. c. allanach _ et al .
_ , in _ proc .
of the aps / dpf / dpb summer study on the future of particle physics ( snowmass 2001 ) _ ed .
n. graf , eur .
j. c * 25 * ( 2002 ) 113 [ econf * c010630 * ( 2001 ) p125 ] [ arxiv : hep - ph/0202233 ] .
j. hisano , t. moroi , k. tobe and m. yamaguchi , phys . rev .
d * 53 * ( 1996 ) 2442 [ arxiv : hep - ph/9510309 ] .
w. porod , comput .
commun . * 153 * ( 2003 ) 275 [ arxiv : hep - ph/0301101 ] .
m. l. brooks _ et al . _
[ mega collaboration ] , phys .
* 83 * ( 1999 ) 1521 [ arxiv : hep - ex/9905013 ] . b. aubert _ et al .
_ [ babar collaboration ] , phys .
lett . * 96 * ( 2006 ) 041801 [ arxiv : hep - ex/0508012 ] .
b. aubert _ et al . _ [ babar collaboration ] , phys .
* 95 * ( 2005 ) 041802 [ arxiv : hep - ex/0502032 ] .
u. bellgardt _ et al . _
[ sindrum collaboration ] , nucl .
b * 299 * ( 1988 ) 1 .
b. aubert _ et al . _ [ babar collaboration ] , phys .
. lett .
* 92 * ( 2004 ) 121801 [ arxiv : hep - ex/0312027 ] .
s. ritt [ mega collaboration ] , on the web page http://meg.web.psi.ch/docs/talks/ + s@xmath82ritt / mar06@xmath82novosibirsk / ritt.ppt .
a. g. akeroyd _ et al .
_ [ superkekb physics working group ] , arxiv : hep - ex/0406071 . t. iijima , overview of physics at super b - factory " , talk given at the 6th workshop on a higher luminosity b factory , kek , tsukuba , japan , november 2004 .
j. aysto _ et al .
_ , arxiv : hep - ph/0109217 .
the prime working group ,
search for the @xmath83 conversion process at an ultimate sensitivity of the order of @xmath84 with prism " , unpublished ; loi to j - parc 50-gev ps , loi-25 , http://psux1.kek.jp/jhf-np/loilist/ + loilist.html y. kuno , nucl .
* 149 * ( 2005 ) 376 .
s. pascoli , s. t. petcov and c. e. yaguna , phys .
b * 564 * ( 2003 ) 241 [ arxiv : hep - ph/0301095 ] .
s. t. petcov , s. profumo , y. takanishi and c. e. yaguna , nucl .
b * 676 * ( 2004 ) 453 [ arxiv : hep - ph/0306195 ] .
h. chankowski , j. r. ellis , s. pokorski , m. raidal and k. turzynski , nucl .
b * 690 * ( 2004 ) 279 [ arxiv : hep - ph/0403180 ] .
a. masiero , s. k. vempati and o. vives , new j. phys .
* 6 * ( 2004 ) 202 [ arxiv : hep - ph/0407325 ] .
m. fukugita and t. yanagida , phys .
b * 174 * ( 1986 ) 45 .
d. n. spergel _ et al .
_ , arxiv : astro - ph/0603449 . s. t. petcov , w. rodejohann , t. shindou and y. takanishi , nucl . phys .
b * 739 * ( 2006 ) 208 [ arxiv : hep - ph/0510404 ] .
e. ables et al .
[ minos collaboration ] , fermilab - proposal-0875 ; g. s. tzanakos [ minos collaboration ] , aip conf .
( 2004 ) 179 ; m. komatsu , p. migliozzi and f. terranova , j. phys .
g * 29 * ( 2003 ) 443 [ arxiv : hep - ph/0210043 ] ; p. migliozzi and f. terranova , phys .
b * 563 * ( 2003 ) 73 [ arxiv : hep - ph/0302274 ] ; p. huber , j. kopp , m. lindner , m. rolinec and w. winter , jhep * 0605 * ( 2006 ) 072 [ arxiv : hep - ph/0601266 ] ; y. itow _ et al .
_ , arxiv : hep - ex/0106019 ; a. blondel , a. cervera - villanueva , a. donini , p. huber , m. mezzetto and p. strolin , arxiv : hep - ph/0606111 ; p. huber , m. lindner , m. rolinec and w. winter , arxiv : hep - ph/0606119 ; j. burguet - castell , d. casper , e. couce , j. j. gomez - cadenas and p. hernandez , nucl .
b * 725 * ( 2005 ) 306 [ arxiv : hep - ph/0503021 ] ; j. e. campagne , m. maltoni , m. mezzetto and t. schwetz , arxiv : hep - ph/0603172 . | in these proceedings we present the results for lepton flavour violating tau and muon decays within the susy seesaw scenario .
specifically , we consider the constrained minimal supersymmetric standard model extended by three right handed neutrinos , @xmath0 and their corresponding susy partners , @xmath1 , ( @xmath2 ) , and use the seesaw mechanism for neutrino mass generation .
we include the predictions for the branching ratios of two types of lepton flavour violating channels , @xmath3 and @xmath4 , and compare them with the present bounds and future experimental sensitivities .
we first analyse the dependence of the branching ratios with the most relevant susy seesaw parameters , and we then focus on the particular sensitivity to @xmath5 , which we find specially interesting on the light of its potential future measurement .
we further study the constraints from the requirement of successfully producing the baryon asymmetry of the universe via thermal leptogenesis , which is another appealing feature of the susy seesaw scenario .
we conclude with the impact that a potential measurement of @xmath5 can have on lepton flavour violating physics .
this is a very short summary of the works in refs .
@xcite and @xcite to which we refer the reader for more details . |
a cell is a system that transforms nutrients into substrates for growth and division . by assuming that the nutrient flow from the outside of a cell is an energy and material source
, the cell can be regarded as a system to transform energy and matter into cellular reproduction .
it is important to thermodynamically study the efficiency of this transformation@xcite . regarding material transformation ,
the yield is defined as the molar concentration of nutrients ( carbon sources ) needed to synthesize a molar unit of biomass ( cell content ) and has been measured in several microbes @xcite . as the conversion of nutrients to cell content
is not perfect and material loss to the outside of a cell occurs as waste , the yield is generally lower than unity .
the yield also changes with nutrient conditions , and measurements in several microbes show that the yield is maximized at a certain finite nutrient flow rate .
the basic logic underlying the optimization of yield at a finite nutrient flow rate rather than at a quasi - static limit is not fully understood .
a cell can also be regarded as a type of thermodynamic engine to transform nutrient energy into cell contents . in this case
, it is necessary to study the thermodynamic efficiency or entropy production during the process of cell reproduction .
the thermodynamic efficiency of metabolism has been measured in several microbes under several nutrient conditions @xcite , and westerhoff and others computed it by applying the phenomenological flow - force relationship of the linear thermodynamics to catabolism and anabolism @xcite to show that the efficiency is optimal at a finite nutrient flow .
although such a phenomenological approach is important for technological application , a physiochemical approach is also necessary to highlight difference between cellular machinery and the carnot engine by characterizing the basic thermodynamic properties in a simple protocell model . indeed ,
when viewed as a thermodynamic engine , a cell has remarkable differences from the standard carnot - cycle engine .
the cell sits in a single reservoir , without a need to switch contacts between different baths .
the cell grows autonomously to reproduce . to consider the nature of such a system , it is necessary to establish the following three points distinguishing the cell from the standard carnot engine@xcite .
first , cells contain catalysts ( enzymes ) .
the enzyme exists only within a compartmentalized cell encapsulated by a membrane and thus enables reactions to convert resources to intracellular components to occur within a reasonable time scale within a cell but not outside the cell . without the catalyst , extensive time
is required for the reaction .
thus , the reaction is regarded to occur only in the presence of the catalyst .
this leads to an intriguing non - equilibrium situation : let us consider the reaction @xmath0 with @xmath1 as the resource , @xmath2 as the product , and @xmath3 as the catalyst .
then , under the existence of @xmath3 , the system approaches an equilibrium concentration ratio with @xmath4/[p ] = \exp(-\beta(\mu_r-\mu_p))$ ] and @xmath5 and @xmath6 as the standard chemical potential of the resource and product , respectively , and with @xmath7 as the inverse temperature .
in contrast , outside the cell , @xmath1 and @xmath2 are disconnected by reactions within the normal time scale ; therefore , their concentration ratio can take on any value . in this sense ,
the external environment is non - equilibrium in nature , in contrast to the intracellular environment .
this leads to a remarkable difference from the standard carnot engine .
second , while considering the dynamical process , it is important to note that the catalysts are synthesized within the cell as a result of catalytic reactions .
the time scale to approach equilibrium can depend on the abundance of the catalyst , which depends on the reaction dynamics themselves .
based on the first and second points mentioned above , the approach to equilibrium in the intracellular environment depends on catalyst abundance , which also depends on the flow rate of nutrients from outside the cell .
hence , the thermodynamic efficiency could show non - trivial dependence upon the nutrient flow .
third , cell volume growth results from membrane synthesis from nutrient components , facilitated by the catalyst , whereas the concentrations of catalyst and nutrient are diluted by cell growth , which results in a non - standard factor for thermodynamic characteristics .
these three issues , which are fundamental to cell reproduction , are mutually connected and thus inherent to a self - reproducing , or autopoietic , system .
in contrast to dynamical systems studies for self - reproduction in catalytic reaction networks @xcite , however , the thermodynamic characteristics for such systems have not been fully explored .
+ in the present study , we determine these characteristics using simple reaction dynamics consisting of the nutrient , catalyst , and membrane . in sec .
ii , we consider a simple protocell model consisting of a membrane precursor and catalyst under a given nutrient flow . the entropy production by chemical production per unit cell volume growth is shown to be minimized at a certain finite nutrient flow .
the mechanism underlying this optimization is discussed in relation to the abovementioned three characteristics of a cell .
the entropy production by material flow is discussed in sec .
ii.a and basically does not change the conclusion described above .
a protocell model consisting of a variety of catalysts that form a network , together with nutrients and membrane precursors , has been investigated to confirm that the conclusion described above is not altered .
the biological relevance of our results is discussed in sec . iii .
first , we study the entropy production @xmath8 resulting from the intracellular reaction for the minimal protocell model consisting only of the synthesis of the enzyme and membrane precursor from the nutrient , which then leads to cellular growth @xcite(see fig.[fig : scheme ] for schematic representation ) .
the model consists of nutrient , membrane precursor , and enzyme , where the enzyme and membrane precursor are synthesized from the nutrient under catalysis by the enzyme .
moreover , by assuming that the diffusion constant of the nutrient is sufficiently large , the internal nutrient concentration is regarded to be equal to the external nutrient concentration . based on the rate equation for chemical kinetics , our model is given by the following two - component ordinary differential equation @xmath9 where the variables @xmath10 and @xmath11 denote the concentrations of the enzyme and membrane precursor , respectively , whereas @xmath12 denotes the cell volume growth rate to be determined .
+ here , the notation of parameters is as follows : * @xmath13 : nutrient concentration . * @xmath14 , @xmath15 , with @xmath16 and @xmath17 as the standard chemical potential of nutrient , @xmath10 , and @xmath11 , respectively .
* @xmath18 : catalytic capacity of the enzyme for @xmath19 component ( @xmath20 ) . * @xmath21 : consumption rate of the membrane precursor to produce the membrane , such that the volume growth rate @xmath22 is given by @xmath23 , where @xmath24 is the conversion rate from membrane molecules to cell volume . in the stationary state
, @xmath22 takes a positive constant value of @xmath25 for @xmath26 . exists
however this solution is an unstable fixed point of the differential equation for @xmath27 . ]
thus , the protocell volume increases exponentially in time . here , we define the entropy production per unit volume at this steady growth state as @xmath8 . in computing @xmath8 , spatial inhomogeneity
is not considered through the assumption of local homogeneous equilibrium .
thus , the entropy produced during the doubling in the protocell volume is given by @xmath28 where @xmath29 is the initial cell volume and @xmath30 is doubling time of the protocell volume .
+ we denote @xmath31 as the entropy production per unit cell - volume growth . generally , if @xmath32 is smaller , the thermodynamic efficiency for a cell growth is higher . for larger @xmath32
, more energetic loss occurs in the reaction process .
hereafter , we study the dependence of @xmath32 on the nutrient condition and the growth rate @xmath22 .
+ in this subsection , we consider only the entropy production by the chemical reaction ; the entropy production by the flow of chemicals from the outside of the cell will be considered in the next section .
the calculation of entropy production among different components is performed by virtually introducing chemical baths for different components that are mutually in disequilibrium and then applying linear non - equilibrium thermodynamics for calculation .
this may result in stringent requisites ; however , this step is adopted to address the thermodynamic efficiency of a cell with growth , as general steady - state thermodynamics are not established currently .
then , the entropy production by the reactions is given by @xmath33 , where @xmath34 is the chemical flow and @xmath35 is the affinity for each reaction .
here we set @xmath36 without losing generality .
+ for calculation , we assume that @xmath37 and @xmath38 are identical for simplicity , denoted as @xmath39 . then , by rescaling the variables as @xmath40 eq.([eq:2 ] ) is written as @xmath41 where @xmath42 and @xmath43 .
the stationary solution of the equation for @xmath44 is given by @xmath45 following this assumption , the entropy production by chemical reaction @xmath8 at the stationary state is calculated as @xmath46 with @xmath47 for the enzymatic reaction @xmath48 and for the membrane reaction @xmath49 . here
, the flows are given by @xmath50 and @xmath51 , whereas the affinities are given by @xmath52 and @xmath53 .
we omit the tilde for affinities because the affinities are not affected by scale transformation .
therefore , we obtain @xmath54 the dependence of @xmath55 upon @xmath56 and @xmath57 , thus obtained , is plotted in fig.[fig : eff_en_graph ] for @xmath44 . as shown , the entropy production rate per unit growth shows a non - monotonic dependence on the nutrient concentration and is minimized at a non - zero nutrient concentration .
because nutrient uptake rate is a monotonic function of nutrient concentration , this result means that the entropy production rate per unit growth @xmath32 is minimal at a finite nutrient uptake rate .
this result is in strong contrast with the thermal engine , where the entropy production is minimal at a quasi - static limit .
fig.[fig : eff_en](a),(b ) shows the entropy production rate per unit growth @xmath58 for each reaction which produces component @xmath10 and @xmath11 , respectively .
this shows that the non - monotonic dependence on the nutrient in fig.[fig : eff_en_graph ] is attributable to @xmath59 . as mentioned above
, an important characteristic of cells is that intracellular reactions are facilitated by enzymes that are autonomously synthesized .
thus , the equilibrium distribution of chemicals in the presence of enzymes is different from the external chemical distribution .
the decrease in @xmath32 under low nutrient concentrations is explained accordingly : the extracellular concentrations of the nutrient and of the membrane precursor are far from equilibrium in the presence of catalysts .
therefore , their intracellular concentrations under conditions of low nutrient uptake remain far from equilibrium and still similar to the external concentrations because of insufficiency of the enzyme . however , when the amount of nutrient uptake increases , the amount of enzyme increases and the system approaches intracellular equilibrium ; therefore , the entropy production rate per unit growth decreases .
+ in contrast , with further increases in nutrient uptake , the entropy production rate increases as a result of the increase in cellular growth ; entropy production @xmath60 by the reaction increases linearly with the reaction speed @xmath34 . in the steady state , the reaction speed @xmath34 is roughly estimated by @xmath61 , with @xmath10 as the concentration of the product of the reaction .
for example , the dynamics of the enzyme concentration are given by @xmath62 .
at steady state , the enzyme production rate @xmath63 is balanced with @xmath61 according to eq.([eq:2 ] ) .
thus , @xmath64 increases with @xmath61 . in summary , for a cell with a high growth rate , increased enzyme abundance is needed , which , however , leads to higher entropy production . ) , @xmath65 balances with the synthesis of the membrane precursor , but the tendency does not change . ] ( with @xmath66 ) is adopted , in which @xmath34 is thermodynamic flow , and @xmath67 the conjugate thermodynamic force , and @xmath68 s are the transport coefficients . here , the degree of coupling is defined by @xmath69 , while the thermodynamic efficiency is defined by the entropy production by the first process divided by the second process , given by @xmath70 this efficiency is known to take a local maximum at a finite flow , if the coupling is weak , i.e. , @xmath71 .
this optimality , however , is not related with the optimality discussed here .
as discussed , ours is due to the equilibration under the existence of catalysts , whose synthesis speed is essential to give the optimality , in addition to the dilution by the cell - volume increase .
second , through straightforward calculation of thermodynamic variables for each reaction , it is shown that the optimality in our case is achieved under the tight - coupling regime i.e. , @xmath72 , where the optimality at a finite flow is not possible in the standard linear thermodynamics . ] .
in contrast , if the enzyme concentration is fixed externally , the entropy production rate per unit growth @xmath32 is minimized at the zero limit of nutrient concentration . in this case , the reaction dynamics eq.([eq:2 ] ) are reduced to @xmath73 where @xmath74 is a constant representing the concentration of the enzyme . in this case ,
the stationary solution is given by @xmath75 $ ] , and accordingly @xmath76 .
there is no optimal nutrient concentration in this expression because @xmath77 is always positive for any @xmath78 .
this is consistent with the explanation mentioned above for eq.([eq : tilde ] ) .
if the enzyme abundance is fixed to be independent of the nutrient uptake , the speed of approaching equilibrium is not altered by the nutrient condition ; therefore , the entropy production just increases monotonically because of the cell volume growth .
thus far , we considered only entropy production by chemical reactions .
in addition , the material flow also contributes to entropy production , which is taken into account here .
+ to discuss the flow of nutrients , the dynamics of the nutrient concentration can not be neglected . by including the temporal evolution of the nutrient concentration ,
the dynamics of the cellular state are given by @xmath79 where @xmath80 and @xmath81 are the enzyme , membrane precursor , and nutrient concentration , respectively , and @xmath82 .
the rate constants @xmath83 and @xmath84 are determined by the standard chemical potential of each chemical .
additionally , the nutrient is taken up with rate @xmath85 from the extracellular environment with a concentration @xmath86 .
+ entropy production by chemical flow is derived from nutrient uptake and membrane consumption , which ( again by assuming linear nonequilibrium thermodynamics ) are given by @xmath87 and @xmath88 , respectively , where @xmath89 is the material flow of component @xmath19 and @xmath90 is the chemical potential
. integration of the terms over a narrow layer having a spatial gradient results in @xmath91 and @xmath92 .
we neglect the entropy production of the solvent with the assumption that intra- and extracellular solvent concentrations are identical ) , we adopted entropy production of membrane consumption as a diffusion process . ] .
the contribution of dilution of the nutrient resulting from cellular growth is approximated as @xmath93 by using the formula of entropy change resulting from the isothermal expansion of an ideal solution to a terminal volume @xmath94 is @xmath95 per unit mole . because @xmath22 is the volume expansion rate in this context and @xmath96 , the change in entropy density is written as @xmath97 per unit mole .
the approximated formula is obtained by expanding @xmath98 into the taylor series and taking the limit of @xmath99 to zero . ] ; for other species , we use the same formula . +
we choose that @xmath100 and @xmath21 are equal to unity and that @xmath101 , for the sake of simplicity .
indeed , the characteristic behavior of @xmath32 is independent of this choice .
then , the fixed - point solutions of eq.([eq:3 ] ) are obtained against two parameters @xmath83 and @xmath86 . from the solution ,
the entropy production per unit growth is computed , as shown in fig.[fig : eff_en_3nodes_diff](a ) .
we note that here again the minimal @xmath32 is achieved for a finite nutrient uptake , i.e. , under nonequilibrium chemical flow . in fig.[fig : eff_en_3nodes_diff](b ) , we plotted @xmath102 , the entropy production excluding that derived from the chemical reaction .
it increases monotonically with the external nutrient concentration .
entropy production is primarily derived from chemical reactions ; therefore , the conclusion of subsection a is unchanged .
+ note that the so - called thermodynamic efficiency is defined as @xmath103 where @xmath104 and @xmath105 are the rates of catabolism and anabolism , and @xmath106 and @xmath107 are the affinities of catabolism and anabolism @xcite . here
, the optimality with regard to entropy production @xmath32 also leads to the optimal thermodynamic efficiency , which , in the present case , is computed by @xmath108 where @xmath109 and @xmath110 are the absolute values of the uptake ( and consumption ) flow of chemical species @xmath81 ( and @xmath11 ) , and @xmath111 is the chemical potential of the @xmath19th chemical species .
it is computed by using the chemical potential of nutrient @xmath112 with @xmath113 as the standard chemical potential for the nutrient and @xmath114 as its standard concentration ( the chemical potential for @xmath10 and @xmath11 are computed in the same way ) .
this thermodynamic efficiency also takes a local maximum value at a non - zero nutrient uptake rate ( see fig.[fig : thermodynamic_efficiency ] ) .
it is worthwhile to check the generality of our result for a system with a large number of chemical species as in the present cell . for this purpose ,
we introduce a model given by @xmath115 where the variables @xmath116 , @xmath117 , and @xmath118 denote the concentrations of the nutrient , membrane precursor , and enzymes , respectively , and @xmath119 is the external concentration of the nutrient .
each element of the reaction tensor @xmath120 is unity if the reaction of @xmath121 to @xmath19 catalyzed by @xmath83 exists ; otherwise , it is set to zero . here ,
the nutrient and the membrane precursor can not catalyze any reaction , whereas the other components @xmath122 form a catalytic reaction network @xcite .
all chemical reactions are reversible in our model ; therefore @xmath120 is equal to unity if and only if @xmath123 equals unity . for the sake of simplicity
, we assume that catalytic capacity , nutrient uptake rate , membrane precursor consumption rate , and the conversion rate from membrane molecule to cell volume are unity .
the standard chemical potential @xmath124 for each chemical species is assigned by uniform random numbers within @xmath125 $ ] , whereas @xmath126 is given by @xmath127 accordingly @xcite .
+ numerical simulations reveal that there again exists an optimal point of @xmath32 for each randomly generated reaction network of @xmath128 .
the dependence of @xmath32 on the nutrient concentration is plotted in fig.[fig : n](a ) , overlaid for different networks .
although the nutrient concentration to give the optimal value is network - dependent , it always exists at a finite nutrient concentration ; therefore , the entropy production is minimized at a non - zero nutrient concentration . to determine a possible relationship with the optimality of @xmath32 and equilibrium in the presence of a catalyst we also computed the kullback - leibler ( kl ) divergence of the steady state distribution from the equilibrium boltzmann distribution as a function of the external nutrient concentration , expressed as @xmath129 where @xmath130 is the concentration of the @xmath19 th chemical species in the steady state .
the kl divergence for each network shows non - monotonic behavior , as shown in fig.[fig : n](b ) .
although the optimal nutrient concentration does not agree with the optimum for @xmath32 , each kl divergence decreases in the region where @xmath32 is reduced . in this sense
, it is suggested that the reduction of @xmath32 in our model eq.([eq : n_system ] ) is related to the equilibration process of abundant enzymes synthesized as a result of a relatively high rate of nutrient uptake as discussed for eq.([eq:2 ] ) and eq.([eq:3 ] ) .
to discuss the thermodynamic nature of a reproducing cell , we have studied simple protocell models in which nutrients are diffused from the extracellular environment and necessary enzymes for the intracellular reactions are synthesized to facilitate chemical reactions , including the synthesis of membrane components , which leads to the growth of cell volume . in the models ,
cell growth is achieved through nutrient consumption by the reactions described above .
we computed @xmath32 , which is the entropy production per unit cell volume growth and found that the value was minimized at a certain nutrient uptake rate .
this optimization stems from the constraint that cells have to synthesize enzymes to facilitate chemical reactions , i.e. , the autopoietic nature of cells . in general , the concentrations of nutrients and membrane components in extracellular environments are different from those in equilibrium achieved in the presence of enzymes , and the intracellular state moves towards equilibrium by synthesizing enzymes to increase the speed of chemical reactions .
the equilibration reduces the entropy per unit chemical reaction .
however , faster cell volume growth leads to a higher dilution of chemicals ; therefore , faster chemical reactions are required to maintain the steady - state concentration of chemicals . because entropy production by the reaction increases ( roughly linearly ) with the frequency of net chemical reactions
, @xmath32 then increases for a higher growth range .
thus , the existence of an optimal nutrient content is explained by the requirement for reproduction mentioned in the introduction , i.e. , equilibration of non - equilibrium environmental conditions facilitated by the enzyme , autocatalytic processes to synthesize the enzyme , and cell - volume increase resulting from membrane synthesis .
+ in the present model , all chemical components thus synthesized are not decomposed ; they are only diluted .
however , each component generally has a specific decomposition time or deactivation time as a catalyst .
we can include these decomposition rates , which can also be regarded as diffusion to the extracellular environment with a null concentration .
then , the equilibration effect is clearer , although the results regarding optimal nutrient uptake are unchanged .
+ in the present study we focused on the entropy production that corresponds to dissipated energy per unit growth . in microbial biology , however , material loss is discussed as biological yield , as mentioned in the introduction , and it is thus reported that the optimal yield is achieved at a certain finite nutrient flow .
material loss is not directly included in the present model ; therefore , we can not discuss the yield derived directly from entropy production .
however , it may be possible to assume that energy dissipation is correlated with material dissipation .
+ for example , the stoichiometry of metabolism is suggested to depend on dissipated energy @xcite . here , metabolism consists of two distinct parts : catabolism and anabolism . for catabolism , the energy is transported through energy currency molecules such as atp , nadph , and gtp , which are synthesized from the nutrient molecule . in this process
, molecular decomposition also occurs , leading to the loss of nutrient molecules .
in addition , the abundance of energy - currency molecules and the utilized energy are correlated .
hence , for both catabolism and anabolism , the energy dissipation and material loss are expected to be correlated .
indeed , a linear relationship between the yield and the inverse of thermodynamic loss ( i.e. , quantity similar to @xmath131 here ) is suggested from microbial experiments @xcite .
+ considering the correlation between energy and matter , the minimal entropy production at a finite nutrient flow that we have shown here may provide an explanation for the finding of optimal yield at a finite nutrient flow .
future studies should examine the relationship between minimal entropy production and optimal yield in the future by choosing an appropriate model that includes atp synthesis and waste products in a cell .
currently , although our models are too simple to capture such complex biochemistry in a cell , they should initiate discussion regarding the thermodynamics of cellular growth .
the authors would like to thank a. kamimura , n. takeuchi , y. kondo , t. hatakeyama , a. awazu , y. izumida , t. sagawa , and t. yomo for the useful discussion .
the present work was partially supported by the platform for dynamic approaches to living system from the ministry of education , culture , sports , science , and technology of japan and the dynamical micro - scale reaction environment project of the japan science and technology agency .
optimality at a finite flow rate was also discussed at the so - called weak - coupling regime in the linear non - equilibrium thermodynamics , where a linear relationship between the fluxes and forces , @xmath134 ( with @xmath135 ) is adopted , in which @xmath34 is thermodynamic flow , and @xmath67 the conjugate thermodynamic force , and @xmath68 s are the transport coefficients . here , the degree of coupling is defined by @xmath136 , while the thermodynamic efficiency is defined by the entropy production by the first process divided by the second process , given by @xmath137 this efficiency is known to take a local maximum at a finite flow , if the coupling is weak , i.e. , @xmath71 .
this optimality , however , is not related with the optimality discussed here . as discussed , ours is due to the equilibration under the existence of catalysts , whose synthesis speed is essential to give the optimality , in addition to the dilution by the cell - volume increase .
second , through straightforward calculation of thermodynamic variables for each reaction , it is shown that the optimality in our case is achieved under the tight - coupling regime i.e. , @xmath72 , where the optimality at a finite flow is not possible in the standard linear thermodynamics .
entropy production during isothermal expansion of an ideal solution from the initial volume @xmath138 to a terminal volume @xmath94 is @xmath139 per unit mole . because @xmath140 is the volume expansion rate in this context and @xmath96 , the change in entropy density
is written as @xmath141 per unit mole .
the approximated formula is obtained by expanding @xmath142 into the taylor series and taking the limit of @xmath99 to zero . | cells generally convert external nutrient resources to support metabolism and growth . understanding the thermodynamic efficiency of this conversion is essential to determine the general characteristics of cellular growth . using a simple protocell model with catalytic reaction dynamics to synthesize the necessary enzyme and membrane components from nutrients , the entropy production per unit cell - volume growth is calculated analytically and numerically based on the rate equation for chemical kinetics and linear non - equilibrium thermodynamics .
the minimal entropy production per unit cell growth is found to be achieved at a non - zero nutrient uptake rate , rather than at a quasi - static limit as in the standard carnot engine .
this difference appears because the equilibration mediated by the enzyme exists only within cells that grow through enzyme and membrane synthesis .
optimal nutrient uptake is also confirmed by protocell models with many chemical components synthesized through a catalytic reaction network .
the possible relevance of the identified optimal uptake to optimal yield for cellular growth is also discussed . |
electron decoherence is one of the key ingredients of the many - body ground state in the presence of disorder and electron - electron interactions .
the existing non - perturbative theory of this phenomenon at low temperatures in realistic disordered conductors @xcite is rather complicated , to a large extent because of the necessity to properly account for fermi statistics for interacting electrons . at the same time
the main physical reason for electron dephasing appears obvious already without unnecessary technical details : it is the electron interaction with the fluctuating quantum electromagnetic field produced by other electrons moving in a disordered potential . in order to be able to quantitatively
describe and understand the latter effect guinea @xcite suggested a model which mimics all essential features of the `` real '' problem of interacting electrons in a disordered conductor except for the pauli exclusion principle .
this model describes a quantum particle moving on a ring with radius @xmath3 and interacting with quantum dissipative environment . for a system in thermodynamic equilibrium quantum decoherence manifests itself as effective suppression of off - diagonal density matrix elements beyond a certain length @xmath1 .
provided there exists nonzero electron dephasing due to its interaction with quantum environment at @xmath2 , this dephasing length @xmath1 should stay finite down to zero temperature .
hence , all effects sensitive to quantum coherence , such as , e.g. , persistent currents ( pc ) and aharonov - bohm ( ab ) oscillations in mesoscopic rings , should be suppressed by interactions as soon as the ring perimeter @xmath4 exceeds @xmath1 .
a great deal of information can be obtained by modelling the environment by a bath of caldeira - leggett ( cl ) oscillators . in this case
it was demonstrated @xcite that pc is reduced by interactions _ in the ground state _ implying suppression of quantum coherence exactly at @xmath5 .
for the same cl environment guinea @xcite found that ab oscillations for a quantum particle on a ring are suppressed by the factor @xmath6 , where the length @xmath1 is set by interactions and remains finite down to @xmath5 .
a similar result was also obtained earlier from the real - time analysis @xcite .
furthermore , the problem @xcite is exactly equivalent to that of coulomb blockade in the so - called single electron box where exponential reduction of the effective charging energy at large conductances @xcite is presently considered as a well established result .
thus , it is now widely accepted that pc for a quantum particle on a ring is exponentially reduced at large ring perimeters down to @xmath5 due to strong dephasing produced by interaction between the particle and the cl bath .
it appears that presently no such consensus exists for another important model of the environment @xcite formed by a diffusive electron gas .
renormalization group arguments developed for this model @xcite suggest very weak power law @xmath7 suppression of ab oscillations at @xmath8 , where the factor @xmath9 is set by interactions . on the contrary , the combination of semiclassics , instanton technique and quantum monte carlo ( mc ) analysis @xcite yields much stronger suppression of quantum coherence , namely exponential suppression @xmath10 ( with temperature independent @xmath1 ) at not too low @xmath11 and power law suppression @xmath12 with @xmath13 at @xmath8 and for @xmath14 .
more recently this problem was reconsidered by means of variational approach @xcite , perturbation theory @xcite and mc simulations @xcite .
in contrast to ref . , either no @xcite or very weak @xcite @xmath3-dependent suppression of pc was found .
note , however , that the variational calculation @xcite reduces the pc problem to that of mass renormalization in the @xmath15 topological sector while , as we show here , pc is determined by other topological sectors ( @xmath16 ) that are distinct from the @xmath15 sector , unlike the variational result in ref . .
perturbative in the interaction calculations @xcite can also miss the correct behavior of pc at not too small @xmath3 as it was already demonstrated in ref .
for the problem under consideration and was also discussed elsewhere in a broader context @xcite .
more arguments along the same lines will be presented below in this paper .
as far as numerical mc results are concerned , the authors @xcite ascribed the difference between their conclusions and those of the previous work @xcite to insufficient trotter number values employed in the mc analysis @xcite .
while this particular issue definitely requires further analysis , it is worth pointing out that the mc data @xcite cover only the perturbative regime @xmath17 where rather weak suppression of pc was found in ref .
as well , cf .
fig . 3 in that paper .
hence , mc results @xcite for small @xmath3 do not appear conclusive for the most interesting non - perturbative regime @xmath18 where strong ( though weaker than exponential ) interaction - induced suppression of pc was predicted @xcite down to @xmath8 . leaving detailed discussion of mc procedure and results to the future , here we note that despite significant theoretical activity in the field
currently no well - controlled non - perturbative approach is available which would allow to analytically study pc for the `` particle + dirty electron gas '' model @xcite in the low temperature limit and at sufficiently large values of @xmath3 . given both fundamental importance of the problem and remaining controversies in the literature , it is highly desirable to formulate such an approach which would unambiguously resolve the whole issue .
a step in this direction is undertaken in the present work .
the structure of our paper is as follows . in sec .
2 we define our model which is essentially identical to that pioneered by guinea @xcite except we additionally introduce a periodic potential for the particle on a ring . in sec .
3 we formulate our main formalism and evaluate persistent current in the ring without interactions by means of the standard instanton technique .
this approach is conveniently generalized to the interacting regime in sec . 4 where we analyze suppression of both quantum coherence and pc by coulomb interaction at sufficiently low temperatures down to @xmath8 and arbitrarily large ring perimeters
. discussion of our main observations and conclusions is presented in sec .
some technical details of our calculation are specified in appendices a and b.
( color online ) the system under consideration : a particle on a ring in the presence of a periodic potential .
the ring is pierced by the magnetic flux and the particle interacts with an effective environment formed by a dirty electron gas.,width=324 ] we will consider a quantum particle with mass @xmath19 and electric charge @xmath20 on a ring with radius @xmath3 threaded by external magnetic flux @xmath21 , see fig .
[ f1 ] . as before @xcite
, it will be convenient to describe the particle position by a vector @xmath22 and consider the angle @xmath23 as a quantum variable .
in contrast to refs . where the quantum particle on a ring was described only by its kinetic energy ( i.e. no potential energy was included into consideration ) , in this paper we will assume that the particle moves in a periodic potential which just for the sake of definiteness is chosen in the form @xmath24 . here
@xmath25 is the total number of periods of the potential @xmath26 which the particle should pass before it makes one full circle on the ring .
accordingly , a non - interacting particle on a ring is described by the hamiltonian @xmath27 where @xmath28 is the magnetic flux operator , @xmath29 and @xmath30 is the flux quantum ( here and below we set the planck s constant equal to unity @xmath31 ) .
now let us include the interaction between the particle on a ring and an effective dissipative environment .
specifically , we will assume that the ring is embedded in the environment formed by the so - called `` dirty electron gas '' @xcite .
the total hamiltonian for our system reads @xmath32 where @xmath33 is the standard hamiltonian for electrons in a disordered conductor and @xmath34 describes interaction between the particle and the electronic environment
. fluctuating electrons in this environment produce stochastic electromagnetic field @xmath35 described by the equilibrium correlator @xmath36 where @xmath37 is the matsubara frequency , @xmath38 is the dielectric susceptibility of the environment and @xmath39 .
similarly to refs .
we will model the environment by 3d diffusive electron gas with @xmath40 where @xmath41 is the drude conductivity of this gas , @xmath42 is the electron diffusion coefficient and @xmath43 is the electron elastic mean free path .
interaction between the particle on a ring and fluctuating electrons in the environment is described by the standard coulomb term @xmath44 in what follows we will assume that the whole system remains in thermodynamic equilibrium at a temperature @xmath11 .
our first and standard step is to integrate out all environmental degrees of freedom effectively described by the collective variable @xmath35 . in the limit of weakly disordered environment @xmath45 to be analyzed below fluctuations of the field @xmath35
can be considered gaussian . in
this case integration over this field is carried out exactly @xcite .
after that one arrives at the grand partition function of the system expressed as a single path integral over the angle variable @xmath46 : @xmath47 ) . \label{z}\end{aligned}\ ] ] here we defined @xmath48 , @xmath49 and the winding - number - projected partition functions @xmath50 . the first term in the exponent in eq .
( [ z ] ) takes care of the magnetic flux while the second term @xmath51 $ ] describes the effective action for our interacting particle on a ring .
this action consists of two terms , @xmath52=s_0[\theta ] + s_{\rm int}[\theta ] .
\label{act}\ ] ] the term @xmath53=\int_0^{\beta}d\tau
\left[\frac{1}{4e_c}\left(\frac{\partial \theta}{\partial \tau}\right)^2+u_0(1-\cos ( \kappa \theta ) ) \right ] \label{part}\ ] ] defines the action for a particle in the absence of the environment .
this action is identical to one for the josephson junction @xcite where @xmath54 plays the role of the charging energy .
the term @xmath55 describes the effect of interaction between the particle and the environment . for our model
it has the form @xcite @xmath56=\alpha\int_0^{\beta}d\tau \int_0^{\beta}d\tau'\frac{\pi^2t^2k(\theta ( \tau ) -\theta ( \tau ' ) ) } { \sin^2 [ \pi t(\tau -\tau ' ) ] } , \label{actiond}\\ k(z)=1-\frac{1}{\sqrt{4r^2\sin^2(z/2)+1 } } , \label{f}\end{aligned}\ ] ] where the constant @xmath57 effectively controls the interaction strength in our model and @xmath58 .
note that the integral in eq .
( [ actiond ] ) is understood as a principal value .
the formal divergence at @xmath59 is regularized by requiring @xmath60 which explains the origin of the first term in ( [ f ] ) . for the sake of physical consistency of our model below
we will set @xmath61 , where the first inequality just means that interaction should remain weak , @xmath62 , while the second one implies that the distance between the neighboring potential minima should be much larger than @xmath43 .
accordingly , the parameter @xmath58 obeys the inequality @xmath63 note that it can also be convenient to rewrite the function @xmath64 in terms of the fourier series @xmath65 , \label{fourier}\ ] ] where the fourier coefficients are @xmath66 for @xmath67 and @xmath68 otherwise .
let us first evaluate persistent current in our system in the absence of interactions , i.e. we set the interaction constant equal to zero @xmath69 everywhere in this section .
in what follows we will mainly be interested in sufficiently large values of the ring radius @xmath3 .
hence , without loss of generality one can consider the limit @xmath70 .
below we will perform our calculation for a somewhat more stringent condition @xmath71 which simplifies our analysis but is by no means important for any of our key conclusions . in the tight binding limit ( [ rrr ] ) and at sufficiently low temperatures
the particle is located at the bottom of one of the potential wells ( see fig . [ f1 ] ) , i.e. in the vicinity of the points @xmath72 , where @xmath73 is an integer number . accordingly , in eq .
( [ z ] ) one should substitute @xmath74 the particle can move around the ring only by hopping between the neighboring minima @xmath72 and @xmath75 of the periodic potential @xmath26 .
each of these tunneling events is described by the well known instanton ( kink ) trajectory @xmath76 and corresponds to the tunneling rate @xmath77 , where @xmath78 and @xmath79 . in order to evaluate the grand partition function
@xmath80 it is necessary to sum over all possible tunneling events of the particle between all potential minima to all orders in @xmath81 . the minimum number of such hops should be equal to @xmath82 for any trajectory corresponding to the winding number @xmath83 .
taking into account that effective duration of each tunneling event is @xmath84 and that the total imaginary time span equals to @xmath85 we can distinguish two different limits . in the limit
@xmath86 the average distance between instantons is large as compared to their typical size , i.e. in this case we are dealing with dilute instanton gas . in the opposite limit @xmath87 instantons
are very close to each other and essentially merge forming a single trajectory .
below we will also demonstrate that in the above conditions it suffices to set @xmath88 .
we begin with the low temperature limit @xmath89 .
to proceed let us evaluate a somewhat more general than in eq .
( [ gpf ] ) matrix element @xmath90 . for this purpose
we consider multi - instanton trajectories @xmath91 where @xmath92 and @xmath93 are respectively the topological charges and collective coordinates of instantons and @xmath94 is defined in eq .
( [ inst ] ) .
the trajectory ( [ multiinst ] ) describes tunneling of the particle between the states @xmath95 and @xmath96 after @xmath83 winds around the ring provided we fix @xmath97 i.e. we consider configurations containing totally @xmath98 instantons corresponding to @xmath99 hops clockwise and @xmath100 hops counterclockwise .
taking into account all possible tunneling events restricted by the condition ( [ rest ] ) and summing over all winding numbers @xmath83 , we obtain @xmath101 making use of the integral representation for the kronecker symbol @xmath102 after performing a summation over @xmath83 with the aid of poisson s resummation formula @xmath103 we obtain @xmath104 this formula allows to easily recover the low - lying energy levels of our problem which contain all necessary information in order to evaluate pc .
for instance , the ground state energy of the particle @xmath105 is obtained in a standard way by taking the limit @xmath106 in eq .
( [ sum ] ) which yields @xmath107 for @xmath108 .
( [ gse ] ) should be continued periodically outside this interval .
this expression determines the periodic dependence ( with period equal to the flux quantum @xmath109 ) of the ground state energy on the magnetic flux @xmath21 . in the limit
@xmath110 this dependence reduces to a set of parabolas @xmath111 turning back to the grand partition function ( [ gpf ] ) , from eq .
( [ sum ] ) we find @xmath112 pc can now be easily obtained from the general formula @xmath113 which yields diamagnetic current @xmath114 this expression fully determines pc in the ring at temperatures @xmath115 and in the absence of interactions with dissipative environment .
the dependence of the maximum pc value on temperature is depicted in fig .
[ f2 ] .
( color online ) maximum value of pc as a function of temperature for different values of @xmath25 .
inset : the magnetic flux value @xmath116 at which pc reaches its maximum value as a function of @xmath117.,width=324 ] eq .
( [ genexp ] ) is further simplified at temperatures above and below the interlevel distance @xmath118 . in the limit @xmath119
the leading contribution to the partition function is defined by configurations with minimal number of instantons .
hence , in the sum over winding numbers in eq .
( [ z ] ) it is sufficient to keep only the terms with @xmath120 terms . for the term with @xmath15
it is necessary to sum over all configurations , whereas for the case @xmath121 only configurations with @xmath25 instantons contribute .
after some algebra we get @xmath122 where @xmath123 is the modified bessel function of imaginary argument . for @xmath121
we find @xmath124 as a result we obtain @xmath125 where @xmath126 at low temperatures @xmath127 eq .
( [ genexp ] ) reduces to a simple formula @xmath128 which also trivially follows from eq .
( [ gse ] ) .
this formula demonstrates that at @xmath5 the magnitude of pc in our system is proportional to @xmath81 while its flux dependence deviates from the simple sinusoidal form for all @xmath129 . in particular , in the limit
@xmath110 this dependence approaches a sawtooth one @xmath130 comparing the expressions for pc derived above with those for a free particle on a ring @xcite we observe that the main physical difference between these two models is the presence of two distinct energy scales
@xmath131 and @xmath81 in our case whereas only one energy scale @xmath54 remains in the limit @xmath132 .
otherwise significant features of the effect are essentially the same in both models . indeed , at high temperatures ( @xmath133 here and @xmath134 for @xmath132 ) the dependence of pc on the magnetic flux @xmath21 is sinusoidal with the period @xmath109 and its amplitude decreases with increasing @xmath11 , cf .
( [ hight ] ) with eq .
( 11 ) in ref . .
in the low temperature limit ( @xmath135 here and @xmath136 for @xmath132 ) the dependence @xmath137 strongly deviates from sinusoidal ( except for a special case @xmath138 ) . in the limit @xmath110 the sawtooth dependence ( [ lk ] ) is identical to that for the case @xmath132 where one has @xmath139 .
finally , let us compare the dependence of the pc amplitude @xmath140 on the ring radius @xmath3 obtained in these two cases .
provided the parameter @xmath25 is fixed and does not change with @xmath3 , eqs .
( [ lk ] ) and ( [ rate ] ) yield exponential decay of @xmath140 with increasing @xmath3 since @xmath141 .
this exponential decay is due to the fact that for larger @xmath3 the potential profile changes in a way that the particle should tunnel at a longer distance @xmath142 between the two neighboring states @xmath143 and @xmath144 .
alternatively , one can keep the distance between the adjacent potential minima @xmath145 and @xmath146 unchanged while increasing @xmath3 .
this is achieved by varying the parameter @xmath25 with @xmath3 as @xmath147 . under this condition
the tunneling rate @xmath81 ( [ rate ] ) becomes independent of @xmath3 and the magnitude of pc @xmath140 ( [ lk ] ) decreases with increasing @xmath3 as @xmath148 exactly as in the case @xmath132 .
note , however , that due to the restriction ( [ rrr ] ) this regime can apply only at not too large values of @xmath3 .
now let us turn to the case of higher temperatures @xmath149 which can be realized in the limit @xmath150 . in this case the leading contribution to the partition function originates from one multi - instanton trajectory .
this trajectory of merged instantons @xmath151 can easily be evaluated due to the presence of the integral of motion which is the classical energy @xmath152 corresponding to the winding number @xmath83 .
this energy is fixed by the periodic boundary condition @xmath153 and the trajectory @xmath151 is obtained from the equation @xmath154 making use of the standard quasiclassical technique one arrives at the following expression for the partition function : @xmath155 where @xmath156 is the classical action .
as before , the leading contribution to this partition functions comes from the lowest winding numbers @xmath120 . for @xmath15
one recovers the oscillator - like expression @xmath157 whereas for @xmath121 the instanton trajectory approaches the straight line in which case the integrals can be easily evaluated and yield @xmath158 with the aid of these results one again arrives at the expression for pc in the form ( [ sin ] ) with @xmath159 different regimes considered in this section can also be summarized graphically by means of the diagram depicted in fig .
[ f3 ] .
( color online ) different regimes for pc in the non - interacting case .
the regions i ( lowest @xmath11 ) and ii ( higher @xmath11 ) are described within the dilute instanton gas approximation ( eqs .
( [ lowt ] ) and ( [ sin ] ) , ( [ hight ] ) respectively ) , whereas the high temperature region iii corresponds to the case of merged instantons , eqs .
( [ sin ] ) , ( [ vhight ] ) .
the dashed lines @xmath160 and @xmath161 indicate the crossover between these regimes .
, width=324 ]
now let us turn on interactions and analyze the effect of fluctuations in a dissipative environment . to this end
we again employ the above instanton technique . evaluating the path integral in ( [ z ] ) in the limit ( [ rrr ] ) and for @xmath162 we follow the same scheme and substitute the trajectory ( [ multiinst ] ) describing quantum tunneling of the particle between different potential minima into the full effective action ( [ act ] ) .
as before , let us fix the winding number equal to @xmath83 .
then we again arrive at configurations of totally @xmath163 instantons restricted by the condition ( [ rest ] ) where we now also set @xmath164 . evaluating the interaction term @xmath55 ( [ actiond ] ) , ( [ f ] ) on multi - instanton trajectories ( [ multiinst ] ) after some algebra ( see appendix a ) we obtain @xmath165=4\pi \alpha rk/\kappa\qquad\qquad\qquad\nonumber \\
-2\alpha\sum\limits_{a , b=1,\ a < b}^{k}\nu_a \nu_b g(\varphi_{ab } ) \ln\left[\frac{\sin[\pi t(\tau_b-\tau_a)]}{\pi t\omega^{-1}}\right ] , \label{instint}\end{aligned}\ ] ] where @xmath166 and @xmath167 we observe that the interaction term ( [ instint ] ) consists of two different contributions .
one of them , @xmath168 , describes interaction - induced suppression of quantum tunneling of the angle variable @xmath23 between different potential minima .
this term yields effective @xmath169-dependent renormalization of the tunneling amplitude @xmath170 the remaining contribution in eq .
( [ instint ] ) describes logarithmic interaction between different instantons which occurs for @xmath171 due to the presence of a dissipative environment .
this logarithmic interaction is absent for @xmath172 in which case @xmath173 . at not too low temperatures
@xmath174 inter - instanton interactions just provide further renormalization of the tunneling amplitude @xmath175 .
one can also write down the renormalization group ( rg ) equation @xmath176 which yields both @xmath177 and @xmath178dependent renormalized tunneling amplitude of the form @xmath179 at even lower temperatures @xmath180 interactions again yield renormalization of the tunneling amplitude which now becomes @xmath181 .
the corresponding rg equation ( [ rg ] ) remains the same but relevant energy scale now becomes @xmath182 eq .
( [ renlt ] ) together with eq .
( [ dren ] ) defines the renormalized tunneling amplitude in the limit @xmath8 and for @xmath171 . in the particular case
@xmath183 our results reduce to those established for the so - called spin - boson model with ohmic dissipation @xcite .
we also note that eq .
( [ renlt ] ) is formally applicable for @xmath184 , while for larger values of the interaction strength @xmath185 and at @xmath5 the tunneling amplitude is renormalized to zero , @xmath186 .
this is the consequence of the quantum dissipative phase transition which similarly to other models @xcite occurs at the critical interaction strength @xmath187 and implies localization of a quantum particle in one of potential wells at any @xmath185 exceeding the critical value @xmath188 .
accordingly , no pc can flow at @xmath5 and @xmath189 .
this formal conclusion , however , does not appear to be of substantial physical significance , since the applicability range of our model is restricted to small values of @xmath190 .
finally , we should point out that , employing a regular perturbation theory in @xmath185 , already in the first order one recovers additional terms which can not be captured within the rg equation ( [ rg ] ) .
the corresponding analysis is presented in appendix b. it turns out that the behavior of pc may be quite different depending both on temperature and on the parameter @xmath25 .
therefore , it is appropriate to distinguish several different cases . as we already discussed , in the case @xmath138 logarithmic interaction between instantons is absent , and the only effect of interactions is @xmath177dependent renormalization of the tunneling amplitude ( [ dren ] ) .
accordingly , for pc in this case we obtain @xmath191 this result is valid at all temperatures in the range @xmath192 .
it demonstrates that for @xmath138 coulomb interaction yields exponential suppression of pc down to @xmath5 provided the ring perimeter @xmath4 exceeds an effective dephasing length @xmath193 which is set by the effective interaction strength @xmath185 and does not depend on temperature .
note that exactly the same length scale was found in the absence of the periodic potential @xmath132 in ref . .
in the case @xmath171 instantons interact logarithmically and in addition to ( [ dren ] ) at not too low temperatures the tunneling amplitude is renormalized according to eq .
( [ renht ] ) . combining these two equations and substituting the renormalized tunneling amplitude @xmath194 into eq .
( [ hight ] ) instead of @xmath81 we obtain @xmath195 where @xmath196 and @xmath197 in the corresponding limits is defined respectively in eq .
( [ hight ] ) and ( [ vhight ] ) . in this case
we again recover the same exponential suppression of pc at ring perimeters @xmath198 with _ temperature independent _ dephasing length defined in eq .
( [ dephl ] ) . at the same time , the pre - exponent in the expression for @xmath199 depends on temperature as a power law @xmath200 with @xmath201 . for @xmath171 and small values of the interaction strength
@xmath190 we have @xmath202 , i.e. @xmath203 grows with decreasing temperature .
this growth , though somewhat weaker than in the non - interacting case ( [ hight ] ) ( since @xmath204 ) , implies that eqs .
( [ sin11 ] ) , ( [ hightic ] ) can be trusted only at @xmath205 whereas at even lower temperatures one expects a crossover to a different regime to be discussed below .
we also note that qualitatively similar behavior of pc at not too low temperatures follows from the numerical analysis @xcite of the model with @xmath132 . in order to evaluate pc in the limit @xmath206
we will make use of eq .
( [ b14 ] ) for the free energy . combining this expression with eq .
( [ tok ] ) and observing that the difference @xmath207 in the last term in eq .
( [ b14 ] ) can be safely neglected within the accuracy of our calculation ( since it only produces extra terms @xmath208 ) , we obtain @xmath209 , \label{pct0}\ ] ] where @xmath194 is defined in eqs .
( [ renlt ] ) , ( [ dren ] ) .
this result allows to make the following observations .
firstly , taking into account the dependence of @xmath194 on @xmath169 we conclude that exactly at @xmath5 and at sufficiently large ring perimeters pc is exponentially suppressed as @xmath210 i.e. in this case for any fixed value @xmath25 we can define an effective zero temperature dephasing length @xmath211 this result demonstrates that in the limit of weak interactions @xmath190 the effective length ( [ tildeph ] ) turns out to be approximately @xmath25 times longer than @xmath1 , i.e. @xmath212 .
should , however , we assume that @xmath213 , no finite dephasing length could be defined from eq .
( [ dephatzero ] ) , although also in this case even for small @xmath185 pc can suffer exponentially strong suppression by electron - electron interactions due to the condition ( [ lr ] ) .
secondly , the result ( [ pct0 ] ) demonstrates that at @xmath8 the effect of electron - electron interactions on pc does not just reduce to renormalization of the tunneling amplitude @xmath214 .
we observe that eq .
( [ pct0 ] ) also contains additional terms evaluated here within the first order perturbation theory in @xmath185 .
this @xmath169-dependent first order contribution turns out to be singular at values of @xmath215 close to @xmath216 and at all other half - integer numbers which indicates insufficiency of the first order perturbation theory , at least for such values of @xmath215 .
higher order terms of the perturbation theory in the interaction may contain similar ( or even stronger ) singularities and , on top of that , may grow with increasing @xmath169 .
hence , at @xmath8 no perturbation theory in @xmath185 can in general be trusted , in particular at sufficiently large @xmath169 .
a more detailed analysis of this issue is beyond the scope of the present paper .
here we only conjecture that such analysis might yield an additional dependence of pc on the ring radius @xmath169 not accounted for in the expression for @xmath194 .
it is quite likely that such kind of @xmath169-dependence of pc at @xmath5 was also observed within a numerical treatment @xcite in the case @xmath132 .
additional support for this conjecture is provided by the exact solution presented below .
as we already pointed out , for @xmath183 our problem is exactly mapped onto the well known spin - boson model with ohmic dissipation @xcite . in this case interaction between instantons
is given by eq .
( [ intk2 ] ) and the grand partition function reads ( see also appendices a and b ) @xmath217=\sum\limits_{n=0}^\infty\left(\delta_r\cos z\right)^{2n}\int\limits_0^\beta d\tau_1\int\limits_{\tau_1}^\beta d\tau_2 ... \int\limits_{\tau_{2n-1}}^{\beta}d\tau_{2n}\exp\left(4\alpha\sum\limits_{j < k=1}^{2n}(-1)^{j+k}\ln\left(\frac{\sin(\pi t(\tau_{k}-\tau_{j})}{\pi t\omega^{-1}}\right)\right ) . \label{sb}\ ] ] in order to non - perturbatively evaluate pc at all values of @xmath169 and at all temperatures including @xmath5 we can profit from the exact solution known for the particular value of the interaction strength @xmath218 , the so - called toulouse limit . introducing the parameter @xmath219 one can conveniently write down the exact expression for the partition function ( [ sb ] ) with @xmath218 in the form @xmath217=\exp\left[\int\limits_{\pi t/\omega}^\infty dx\frac{1-e^{-\frac{x}{2\pi\xi(z)}}}{x\sinh x}\right].\ ] ] this result allows to immediately establish pc which reads @xmath220 at not too low temperatures @xmath221 we obtain @xmath222 in low temperature limit @xmath223 one finds @xmath224 evaluating the integral in ( [ 52 ] ) we arrive at the final result @xmath225\sin(2\pi\phi_x ) .
\label{tlt0}\ ] ] we observe that both at non - zero temperatures and exactly at @xmath5 the above exact expressions for pc demonstrate its exponential suppression at ring perimeters exceeding @xmath226 ( these two length scales coincide for @xmath227 and @xmath228 , see eq .
( [ tildeph ] ) ) .
in addition , the result ( [ tlt0 ] ) demonstrates that at @xmath5 ( a ) the dependence of pc on @xmath169 deviates from purely exponential ( which is in agreement with our above conjecture ) and ( b ) the dependence of pc on @xmath215 deviates from purely sinusoidal and contains the logarithmic singularity at half - integer values of @xmath215 , cf .
( [ pct0 ] ) .
in this paper we proposed a model which allows for detailed non - perturbative treatment of the effect of electron - electron interactions on pc in normal nanorings at low temperatures .
our investigation employs a well - controlled instanton technique and yields a rich structure of different regimes .
the main features observed within our analysis can be summarized as follows : ( @xmath229 ) coulomb interaction yields @xmath3-dependent renormalization of the tunneling amplitude ( [ dren ] ) which , in turn , results in exponential suppression of pc at large enough @xmath3 , ( @xmath230 ) logarithmic interaction between instantons yields additional renormalization of the tunneling amplitude described by the rg equation ( [ rg ] ) and ( @xmath231 ) electron - electron interactions generate yet additional contributions not captured by eqs .
( [ dren ] ) and ( [ rg ] ) , see eq .
( [ pct0 ] ) and appendix b. these contributions may become particularly important at @xmath8 indicating the failure of the naive perturbation theory in the interaction at sufficiently large @xmath3 .
although the effect ( @xmath231 ) still requires additional non - perturbative analysis , already ( @xmath229 ) and ( @xmath230 ) result in exponential suppression of pc at any temperature including @xmath232 for any given @xmath25 and at ring perimeters exceeding the dephasing length set by interactions and defined in eqs .
( [ dephl ] ) and ( [ tildeph ] ) .
note that the length scale identical to ( [ dephl ] ) also follows from the earlier non - perturbative analysis @xcite developed for the case @xmath132 .
thus , similarly to ref . the decoherence effect in our model is controlled by the parameter @xmath233 rather than by @xmath185 or @xmath234 as it was sometimes suggested in the literature in the case @xmath132 .
it is worthwhile to stress that exponential dependence of pc on @xmath3 of the form @xmath235 by itself does not yet necessarily imply decoherence .
for instance , even in the absence of interactions pc @xmath236 ( [ lowt ] ) can decrease exponentially with increasing @xmath3 provided the parameter @xmath25 is fixed to be independent of the ring radius .
obviously , quantum coherence is not destroyed in this case .
exponential reduction of pc with increasing @xmath3 at @xmath8 can also occur in superconducting nanorings due to proliferation of quantum phase slips @xcite .
also in that case the dependence @xmath237 can be interpreted just as a non - trivial coordinate - dependent renormalization effect @xcite . an important qualitative difference between nanorings with dissipation considered in sec .
4 and the two last examples is that in our problem dissipation explicitly _ violates time - reversal symmetry _ ( thus causing genuine decoherence of a quantum particle ) , while no such symmetry is violated in superconducting nanorings @xcite or in the absence of dissipation ( sec .
3 ) . hence ,
quantum coherence remains fully preserved in the last two cases despite exponential suppression of pc at large @xmath3 .
one can discriminate between decoherence and pure renormalization in a number of ways .
for instance , one can drive the system out of equilibrium and investigate its relaxation by means of a real - time analysis .
this approach was employed in refs . where a finite dephasing time @xmath238 was found at @xmath5 with @xmath1 defined in eq .
( [ dephl ] ) and @xmath239 being the particle velocity .
this observation allows to unambiguously identify quantum decoherence . another way @xcite amounts to analyzing fluctuations of pc in the ground state of an interacting system . for instance , one can study the correlator @xmath240 , where @xmath241 is the current operator which expectation value @xmath242 defines pc in the ground state . within the model
@xcite it was demonstrated that , while the average pc in the ground state decreases with increasing interaction strength , its fluctuations _ increase _ , thus implying genuine decoherence rather than pure renormalization .
a similar situation occurs within the model studied here .
following @xcite here we intentionally disregarded fermi statistics by suppressing electron exchanges between the ring and the environment .
the question arises if inclusion of the pauli principle into the model could alter our main conclusion about non - vanishing electron decoherence at zero temperature .
golubev and one of the present authors addressed this issue by developing two entirely different non - perturbative in the interaction techniques @xcite .
both these methods yield the same conclusion : although fermi statistics is crucially important for other properties of a dirty interacting electron gas , it practically _ does not affect _ interaction - induced quantum decoherence at @xmath8 .
quantitative agreement was demonstrated @xcite between the corresponding results derived for the models with @xcite and without @xcite the pauli principle . despite all these developments ( including recently obtained _ exact _ solution of the problem @xcite ) it is sometimes argued in the literature that the pauli principle can preclude from electron decoherence at @xmath8 . e.g. this conclusion was reached on the basis of the first order perturbation theory in the interaction @xcite .
insufficiency of such kind of perturbation theory for the problem in question was repeatedly demonstrated elsewhere @xcite and was also observed here within the model considered .
more recently , von delft and co - authors @xcite re - iterated the same incorrect conclusion @xcite . unfortunately the analysis @xcite is fundamentally flawed since it violates causality effectively implying that dynamics of interacting electrons would be affected by photons coming both from the past and from the future .
hence , the results @xcite can not be considered seriously . for more details on this issue
we refer the reader to the paper @xcite . *
acknowledgments * this work was supported in part by rfbr grant 09 - 02 - 00886 .
a.g.s . also acknowledges support from the landau foundation and from the dynasty foundation .
let us analyze the effect of electron - electron interactions on the dilute instanton gas employed in our work . combining eqs .
( [ actiond ] ) and ( [ fourier ] ) we can rewrite the dissipative part of the action describing such interactions within our model in the following form : @xmath243=-\frac{\alpha}{2}\sum\limits_na_n\int\limits_{-\beta/2}^{\beta/2}d\tau \int\limits_{-\beta/2}^{\beta/2 } d\tau'\frac{\pi^2 t^2 ( 1-e^{in\theta(\tau)})(1-e^{-in\theta(\tau')})}{\sin^2(\pi t(\tau-\tau'))}. \label{a1}\ ] ] substituting multi - instanton trajectories @xmath244 ( [ multiinst ] ) into eq .
( [ a1 ] ) and employing the approximation @xmath245 appropriate for well - separated instantons , with the aid of the identity @xmath246 we obtain @xmath247=-\frac{\alpha}{2}\sum\limits_na_n\sum\limits_{j_1,j_2}c_{j_1}\bar c_{j_2 } \int\limits_{-\beta/2}^{\beta/2}d\tau\int\limits_{-\beta/2}^{\beta/2 } d\tau ' f'_{nj_1}(\tau)\bar f'_{nj_2}(\tau ' ) % \times\qquad\nonumber\\ \times \ln|\sin(\pi t(\tau-\tau'))|.\end{aligned}\ ] ] let us first consider the terms in the above sum which correspond to @xmath248 .
they are @xmath249 since an effective instanton width @xmath250 is much smaller than the inverse temperature @xmath251 it is possible to split the contribution ( [ a4 ] ) into regular and singular ( in the limit @xmath252 ) parts , respectively @xmath253 and @xmath254 consider now the remaining terms with @xmath255 . provided the distance between instantons remains much larger than their width @xmath256 we find @xmath257 % \times\qquad\nonumber\\ \times \ln(\sin(\pi t(\tau_{j_2}-\tau_{j_1 } ) ) ) .
\label{a7}\end{aligned}\ ] ] performing summation over @xmath258 in eq .
( [ a7 ] ) and combining the result with eqs .
( [ a5 ] ) and ( [ a6 ] ) we arrive at eq .
( [ instint ] ) .
let us note that in the particular case @xmath259 the logarithmic interaction between instantons vanish , while in the case of @xmath183 and for large enough @xmath169 we get @xmath260=2\pi\alpha r(n_1+n_2)+\tilde s(\alpha ) $ ] where the logarithmic inter - instanton interaction @xmath261 takes the form @xmath262 hence , the partition function for our model with @xmath183 is formally equivalent to that for the well known spin - boson model with ohmic dissipation @xcite .
let us rewrite the partition function of our problem as @xmath263 } \delta_{\sum\limits_i \nu_i , m\kappa}. \label{b1}\ ] ] with the aid of the poisson s resummation formulae eq .
( [ b1 ] ) can be transformed to @xmath264=\sum\limits_{k=1}^{\kappa}z[2\pi(\phi_x - k)/\kappa;\beta],\ ] ] where @xmath217=\sum\limits_{k=0}^{\infty}\sum\limits_{\nu_1=\pm1 } .. \sum\limits_{\nu_k=\pm 1}\left(\frac{\delta_r}{2}\right)^n\int\limits_0^\beta d\tau_1\int\limits_{\tau_1}^\beta d\tau_2 ... \int\limits_{\tau_{n-1}}^{\beta}d\tau_n e^{iz\sum\limits_{j=1}^k\nu_j+2\alpha\sum\limits_{a < b=1}^k\nu_a\nu_bg(\varphi_{ab})w(\tau_a-\tau_b ) } \label{b3}\ ] ] and @xmath265}{\pi t\omega^{-1}}\right]=-\sum\limits_{n=-\infty}^{\infty}w_ne^{2\pi int(\tau_a-\tau_b)}.\ ] ] expanding the exponent in eq .
( [ b3 ] ) one recovers the perturbation series for @xmath266 $ ] . extending the definition of the fourier coefficients @xmath267 to negative @xmath258 in such a way that @xmath268 and @xmath269 we may write @xmath270 with the aid of this equation it is easy to define all orders of the perturbative expansion @xmath271 , where @xmath272=e^{-2\beta\delta_r\sin^2(z/2)}\ ] ] and @xmath273=-2\alpha\delta_r^2\sum\limits_{n =- r}^ra_n\sin^2\left(z+\frac{\pi n}{\kappa}\right)\sin^2\left(\frac{\pi n}{\kappa}\right ) \int\limits_0^\beta d\tau_1\int\limits_{\tau_1}^\beta d\tau_2z_0[z;\tau_1]z_0\left[z+\frac{2\pi n}{\kappa};\tau_2-\tau_1\right ] z_0[z;\beta-\tau_2]w(\tau_2-\tau_1).\ ] ] the last term can also be represented graphically by the diagram in fig .
[ f4 ] .
performing the fourier transformation of the partition function @xmath274=\int\limits_{-\infty}^\infty\frac{d\lambda}{2\pi}e^{-i\lambda \beta}\tilde z[z,\lambda ] \label{b8}\ ] ] and introducing the self - energy @xmath275 as a sum of all irreducible diagrams , we obtain @xmath276=\frac{1}{-i\lambda+2\delta_r\sin^2(z/2)-\tilde \sigma[z;\lambda]}. \label{b9}\ ] ] to the first order in the interaction one finds @xmath277=2\alpha\delta_r^2\sum\limits_{n =- r}^ra_n\sin^2\left(z+\frac{\pi n}{\kappa}\right ) \sin^2\left(\frac{\pi n}{\kappa}\right)\sum\limits_{k=-\infty}^\infty\frac{w_k}{-i\lambda-2\pi i kt+2\delta_r\sin^2(\frac{z}{2 } + \frac{\pi n}{\kappa})},\ ] ] where @xmath278 $ ] and @xmath279 for @xmath280 . performing summation over @xmath163 , we get @xmath281=2\alpha\delta_r^2\sum\limits_{n =- r}^ra_n\sin^2\left(z+\frac{\pi n}{\kappa}\right)\sin^2\left(\frac{\pi n}{\kappa}\right ) \qquad\qquad\qquad\qquad\qquad\qquad\qquad\nonumber\\ \times \frac{\ln\frac{2\omega}{\pi t}+\gamma+\frac12\psi\left(1+\frac{\lambda+2i\delta_r\sin^2(\frac{z}{2}+\frac{\pi n}{\kappa})}{2\pi t}\right)+\frac12\psi\left(1-\frac{\lambda+2i\delta_r\sin^2(\frac{z}{2}+\frac{\pi n}{\kappa})}{2\pi t}\right)}{-i\lambda+2\delta_r\sin^2(\frac{z}{2}+\frac{\pi n}{\kappa})},\end{aligned}\ ] ] where @xmath282 is the digamma function and @xmath283 is the euler constant . substituting this expression into eq .
( [ b9 ] ) , we observe that @xmath284 $ ] has a pole at @xmath285 , where @xmath286.\end{aligned}\ ] ] combining the above results with eq .
( [ b8 ] ) one arrives at the final perturbative in the interaction expression for the partition function .
of particular interest is the low temperature limit @xmath180 . in this case one has @xmath287 and the free energy reduces to @xmath288 . \label{b13}\end{aligned}\ ] ] it is straightforward to observe that the first logarithmic term in the square brackets simply yields the renormalization @xmath181 and does not alter the dependence of pc on the flux @xmath215 .
hence , this perturbative contribution can be absorbed in the first term in ( [ b13 ] ) simply by substituting @xmath194 instead of @xmath289 , where @xmath194 is defined in eq .
( [ renlt ] ) .
the second logarithmic term in the square brackets in ( [ b13 ] ) contains an additional flux dependence which turns singular at @xmath215 close to the half - integer numbers .
clearly , this perturbative contribution can not be just reduced to the renormalization ( [ renlt ] ) . as a result
, we obtain @xmath290 employing this expression we arrive at eq .
( [ pct0 ] ) .
golubev and a.d .
zaikin , phys .
* 81 * , 1074 ( 1998 ) ; phys . rev .
b * 59 * , 9195 ( 1999 ) ; _ ibid . _ * 62 * , 14061 ( 2000 ) ; j. low . temp . phys . * 132 * , 11 ( 2003 ) .
d.s . golubev and a.d .
zaikin , new j. phys .
* 10 * , 063027 ( 2008 ) ; physica e * 40 * , 32 ( 2007 ) .
semenov , d.s .
golubev , and a.d .
zaikin , phys .
b * 79 * , 115302 ( 2009 ) ; a.g .
semenov and a.d .
zaikin , arxiv:0905.2045 .
f. guinea , phys .
b * 65 * , 205317 ( 2002 ) .
p. cedraschi , v.v .
ponomarenko , and m. bttiker , phys .
lett . * 84 * , 346 ( 2000 ) ; ann . phys . * 289 * , 1 ( 2001 ) .
d.s . golubev and a.d .
zaikin , physica b * 255 * , 164 ( 1998 ) .
s.v . panyukov and a.d .
zaikin , phys .
lett . * 67 * , 3168 ( 1991 ) ; j. low temp .
phys . * 73 * , 1 ( 1988 ) .
herrero , g. schn , and a.d .
zaikin , phys .
b * 59 * , 5728 ( 1999 ) and further reference therein .
golubev , c.p .
herrero , and a.d .
zaikin , europhys . lett . * 63 * , 426 ( 2003 ) . b. horovitz and p. le doussal , phys .
b * 74 * , 073104 ( 2006 ) .
d. cohen and b. horovitz , j. phys .
a : math . theor . * 40 * , 12281 ( 2007 ) ; europhys . lett .
* 81 * , 30001 ( 2008 ) .
v. kagalovsky and b. horovitz , phys .
b * 78 * , 125322 ( 2008 ) . g. schn and a.d .
zaikin , phys .
rep . * 198 * , 237 ( 1990 ) .
leggett , s. chakravarty , a.t .
dorsey , m.p.a .
fisher , a. garg , and w. zwerger , rev .
phys . * 59 * , 1 ( 1987 ) .
u. weiss , _ quantum dissipative systems _ ( world scientific , singapore , second edition , 1999 ) .
matveev , a.i .
larkin , and l.i .
glazman , phys .
rev . lett . * 89 * , 096802 ( 2002 ) .
arutyunov , d.s .
golubev , and a.d .
zaikin , phys .
rep . * 464 * , 1 ( 2008 ) .
golubev , g. schn , and a.d .
zaikin , j. phys .
. jap . * 72 * , suppl .
a , 30 ( 2003 ) .
aleiner , b.l .
altshuler , and m.e .
gershenzon , waves random media * 9 * , 201 ( 1999 ) .
golubev , a.d .
zaikin , and g. schn , j. low temp .
phys . * 126 * , 1355 ( 2002 ) .
j. von delft , int . j. mod .
b , * 22 * , 727 ( 2008 ) ; f. marquardt , j. von delft , r.a . smith , and v. ambegaokar , phys . rev .
b * 76 * , 195331 ( 2007 ) .
d.s . golubev and a.d .
zaikin , j. phys .
. ser . * 129 * , 012016 ( 2008 ) ; cond - mat/0512411 . | employing instanton technique we evaluate equilibrium persistent current ( pc ) produced by a quantum particle moving in a periodic potential on a ring and interacting with a dissipative environment formed by diffusive electron gas .
the model allows for detailed non - perturbative analysis of interaction effects and depending on the system parameters yields a rich structure of different regimes .
we demonstrate that at low temperatures pc is exponentially suppressed at sufficiently large ring perimeters @xmath0 where the dephasing length @xmath1 is set by interactions and does not depend on temperature .
this behavior represents a clear example of quantum decoherence by electron - electron interactions at @xmath2 . |
it has long been recognized that a periodic potential , such as a rotating bar , can drive density waves in a differentially rotating disk .
a linear inviscid theory of the excitation and propagation of these waves in such disks was described in the seminal work of goldreich & tremaine ( 1979 , 1980 ) .
subsequently , yuan ( 1984 ) showed that the gravitational potential associated with a rotating bar can effectively excite nonlinear density waves with high radial streaming velocities at the outer lindblad resonance ( olr ) in the limit that self - gravity in the gas disk could be neglected .
the excitation and wave propagation with self - gravity was later considered in the linear and nonlinear approximation by yuan & cheng ( 1989 , 1991 ) .
further progress was achieved by the development of a nonlinear asymptotic theory by yuan & kuo ( 1997 ) , who explicitly included the effects of gas pressure , viscosity , and self - gravitation in the formulation . however , the theory s applicability was somewhat limited as it was derived only in the vicinity of the resonances . building upon this study ,
an improved nonlinear asymptotic global theory of spiral density waves was developed by yuan & yang ( 2006 ) to theoretically examine the nearby grand - design spiral galaxy ngc 5248 .
although these nonlinear asymptotic theories can be useful in addressing the structure of disk galaxies , they do not provide insight on the time - dependent evolution of the structure in gas disks in response to a rotating bar potential . to examine this phase ,
a full hydrodynamic description must be obtained via numerical simulations . to this end ,
early numerical investigations of the evolution of disk galaxies driven by a rotating bar potential were carried out by huntley et al .
( 1978 ) and sanders & tubbs ( 1980 ) .
an excellent description of the general properties of gas flows in galaxies with bars can be found in the investigation by athanassoula ( 1992 ) .
simulations designed for direct comparisons with the observed morphology and velocity fields of specific galaxies have been published by england ( 1989 ) and lindblad & kristen ( 1996 ) for ngc 1300 and by lindblad et al .
( 1996 ) for ngc 1365 .
building on this approach , we report on a quantitative comparison between observations of the barred spiral ngc 6782 and detailed two - dimensional hydrodynamical simulations in order to probe the underlying properties of the rotating bar . in the next section ,
observational data on the structure of ngc 6782 based on ground - based and hst images , and an h@xmath2 fabry - perot velocity field , are summarized . the numerical technique for solving the governing hydrodynamic equations as well as the description of the initial set - up of the calculations
are described in 3 .
procedures adopted for theoretically modeling the observational data are outlined in 4 and the comparisons of the numerical results with the data are presented in 5 . finally , a discussion is presented in 6 and conclusions in 7 .
ngc 6782 is an extremely interesting case , as shown by the images in figures [ images ] and [ hubble ] .
the type of the galaxy is ( r@xmath3r@xmath4)sb(r)a in the system of the de vaucouleurs atlas of galaxies ( buta et al .
2007 ) , implying a multi - ringed early - type barred spiral galaxy .
the upper panel images were obtained with a tek2 ccd attached to the ctio 1.5 m telescope in 1992 august .
these include a @xmath5-band image and a @xmath6 color index map coded such that blue features are dark while red features are light .
a @xmath7-band image was also obtained , and the exposure times were 600s in @xmath5 , and 300s each in @xmath7 and @xmath8 .
the lower panel images are 1200s exposure red continuum and h@xmath2+[nii ] images obtained in 1993 august also with the ctio 1.5 m but with a tek1 ccd ( crocker et al .
1996 ) . for both ccds used , the readout noise was 3 - 5@xmath9 and the gain was 1.6@xmath9/adu .
the pixel scale was 0@xmath10435 in each case .
all of these images were processed using standard routines in iraf .
the redshift of ngc 6782 is [email protected] ( de vaucouleurs et al .
we choose the value of the hubble constant to be 100 km s@xmath1 mpc@xmath1 . as a result
, we have here adopted a distance of 38.95 mpc , yielding a scale of 10@xmath12 = 1.89 kpc .
however , the exact distance is not important in this particular investigation .
the images show that ngc 6782 has a nearly circular bright nuclear ring connected with a pair of almost straight dust lanes , the other ends of which attach to a diamond - shaped ( or pointed - oval ) inner ring .
two faint arms in turn extend from the two tips of the inner ring to the outermost parts of the galaxy , forming a faint double outer ring - pseudoring morphology ( buta & crocker 1991 ) where the r@xmath3 component is stronger .
each ring feature is a zone of enhanced blue colors , consistent with active star formation .
the pointed - oval shape of the inner ring is an unusual feature of the galaxy and is not commonly seen . as noted by crocker et al .
( 1996 ) , the h@xmath2+[nii ] emission is concentrated in arcs around the major axis of the inner ring , which coincides with the bar axis .
this is seen mainly in inner rings having a significant intrinsic elongation along the bar axis .
we address this concentration with our model in this paper . according to grouchy et al .
( 2008 ) , 24% of the total h@xmath2+[nii ] flux comes from the inner ring , while 54% comes from the nuclear ring and central source .
the same paper gives the star formation rate in the inner ring as 0.84 m@xmath13 yr@xmath1 .
it is noteworthy that virtually all of the recent star formation in ngc 6782 is found in its three main ring features .
figure [ hubble ] shows a hubble heritage image of ngc 6782 that reveals finer details of the star formation and dust in the inner regions of the galaxy .
individual star clusters are detectable near the pointy ends of the inner ring .
the weakness of the northwest portion of the nuclear ring in the h@xmath2+[nii ] image is due mainly to a strong dust lane .
the bar itself is very prominent in this image as a feature largely made of old stars .
in addition to the strong leading dust lanes , the image reveals very much weaker dust lanes near the trailing sides of the bar .
figure [ profiles ] shows azimuthally - averaged surface brightness and color index profiles of ngc 6782 , based only on the ctio images .
the surface brightness scales were calibrated from observations of landolt ( 1992 ) standard stars .
the three rings ( whose radii are from buta & crocker 1993 ) cause the @xmath5-band profile to have significant humps and the color index profiles to have significant dips .
the @xmath14 and @xmath15 colors decrease at the location of each main ring as expected from the color index map in figure [ images ] .
these profiles are typical of early - type barred galaxies .
table 1 summarizes some of the basic parameters of ngc 6782 derived from the 1992 broadband images and the profiles in figure [ profiles ] .
the isophotal dimensions , total magnitude , and total colors are rc3-style ( de vaucouleurs et al . 1991 ) global parameters that can improve on the previously published values . for our adopted distance
, the total apparent magnitude implies an absolute blue magnitude brighter than @xmath1620 , indicating that ngc 6782 is a high luminosity early - type system .
various published classifications indicate that the galaxy is an sb0/a - sba type with multiple rings .
the photometric orientation parameters are based on ellipse fits to @xmath17 outer isophotes beyond the main ring features .
these parameters are well - defined and indicate a system inclined only about 27@xmath18 to the line of sight in position angle 35@xmath18 .
these parameters are in agreement with those based on analysis of the observed velocity field ( section 2.2 ) .
table 1 also includes the deprojected diameters , axis ratios , and relative bar - ring position angles for the nuclear , inner , and outer r@xmath3 rings based on these adopted orientation parameters .
the rings were mapped visually from the @xmath5-band image on a tv monitor , and an ellipse was fitted to the mappings by least squares .
the inner ring has an extreme intrinsic axis ratio ( 0.69 ) compared to the average , [email protected] , found for sb inner rings by buta ( 1995 ) . within the uncertainties of the visual mapping
, the inner ring is aligned almost exactly parallel to the bar .
the outer r@xmath3 ring is also found to be intrinsically elongated with an axis ratio of 0.79 , and is aligned almost perpendicular to the bar .
buta ( 1995 ) found that r@xmath3 rings on average have an intrinsic axis ratio of [email protected] .
the ratio of the outer and inner ring diameters is 2.44 , somewhat larger than the median of 2.13 found by buta ( 1995 ) for 532 sb galaxies .
the nuclear ring deprojects into a nearly circular shape , and no bar - ring position angle is given .
ngc 6782 is well - known as a double - barred galaxy ( buta & crocker 1993 ; erwin 2004 ) , with the primary bar crossing the inner ring and the secondary bar crossing the nuclear ring .
the secondary bar is most prominent in the red continuum image in figure [ images ] , and has an orientation tipped 29@xmath18 behind the primary bar ( if the spiral arms are trailing ) .
neither bar looks exceptionally strong , and in fact the primary bar is classified as sab by corwin et al .
the position angle of the primary bar is 177@xmath19 ( crocker et al .
the inner ring of ngc 6782 is so highly elongated that it also must be considered bar - like .
the optical velocity field of ngc 6782 was obtained on 1997 july 27 ( ut ) with the ( now decommissioned ) rutgers fabry - perot ( rfp ) interferometer attached to the ctio 4-m telescope .
the broad etalon with a full width at half maximum ( fwhm ) of 2.3@xmath20 was used for the observations , and h@xmath2 was the only emission line used for the radial velocity measurements .
the instrumental velocity dispersion is 44 km s@xmath1 .
owing to the relatively low inclination , only 16 frames were needed to cover the full velocity range of ngc 6782 .
frames ranging from 6642@xmath20 ( 3620 km s@xmath1 ) to 6656@xmath20 ( 4260 km s@xmath1 ) were taken in 1@xmath20 steps .
each frame was exposed for 600s .
using local position standards from the stsci guide star catalog , the scale of the frames was measured to be 0@xmath10349 @xmath11 0@xmath10001 pix@xmath1 .
the same stars showed that the images had to be rotated counterclockwise by 90@xmath1850 to get north to the top and east to the left .
although the observations were obtained during strongly variable cloudiness , the ctio 4-m control system had a transparency monitoring capability that allowed us to stop and restart the integrations as needed when the transparency improved .
the full width at half maximum of the point spread function on the images was 4.4 pixels ( 1@xmath105 ) .
the main steps in the reduction of the images and the calibration of radial velocities are described already by buta et al .
( 2001 ) , and we refer interested readers to that paper for details .
the process gave line , continuum , radial velocity , and velocity dispersion maps with values for more than 26,000 pixels .
the h@xmath2 line map is similar to the h@xmath2+[nii ] image shown in figure [ images ] , except that it lacks the nuclear source .
this source must be dominated by [ nii ] emission .
the resulting heliocentric velocity field is shown with false colors in figure [ vfield ] .
the dotted partial circle indicates the boundary of the actual field of view , which is 2@xmath218 in angular diameter .
part of the velocity field is cut - off on the northeast side due to this boundary , but the bulk of the galaxy s ionized gas is covered .
the velocity field immediately reveals a line of nodes inconsistent with the major axes of the inner or outer r@xmath3 rings .
these features are intrinsically elongated if in the same plane , as is likely . at the position of the nuclear ring ,
two spots of high velocity relative to the nucleus are seen .
these spots are symmetrically placed around the center and are in a position angle of 31@xmath18 .
analysis of the velocity field in terms of deprojected circular motion ( warner et al .
1973 ) gave a rotation center close ( within 1@xmath10 ) to the position of the nucleus and a systemic velocity of 3921@xmath113 km s@xmath1 , in excellent agreement with the hi parkes all - sky final survey catalog value of 3920@xmath1114 km s@xmath1 ( nasa / ipac extragalactic database ) .
the line of nodes for circular rotation was found to be in position angle 37@xmath185 by the warner et al .
( 1973 ) method , but because the inclination is low , the method did not strongly favor a particular value , giving similar quality fits over a range of at least 15@xmath18 - 30@xmath18 . as a check
, we also used aips routine gal , which fits an exponential rotation curve to the projected velocity field .
this gave a line of nodes position angle of 32@xmath189 and an inclination of 22@xmath185 .
this latter value is close enough to the photometric inclination of 27@xmath18 that we have adopted 27@xmath18@xmath115@xmath18 for our subsequent analysis . for the line of nodes we adopted a position angle of 35@xmath18@xmath112@xmath18 ,
also consistent with the photometric value .
kinematic parameters are collected in table 1 .
figure [ rotc ] shows the receding and approaching halves of the rotation curve based only on points within 30@xmath18 of the line of nodes .
considerable symmetry is seen , including a very rapid rise to a maximum rotation velocity of @xmath22250 km s@xmath1 at the radius of the nuclear ring .
there is also an indication of a slightly falling rotation curve at large radii , again seen on both sides of the major axis .
rubin et al . (
1985 ) showed that the average sa galaxy rotation curve is either flat or slightly rising for the luminosity of ngc 6782 . since the emission from ngc 6782
is confined to rings , two of which are intrinsically significantly elongated , the details of the rotation curve must be viewed with caution . to further highlight the kinematics , we show in figure [ vpadiagrams ] mean radial velocity versus position angle every 5@xmath18 for each ring defined by the indicated ranges of sky - plane radii . if the rotation were perfectly circular , these curves would be symmetric , but subtle asymmetries , especially for the inner ring , indicate the presence of noncircular motions .
fits of circular rotation to these diagrams confirm a declining rotation velocity , from 264 km s@xmath1 in the nuclear ring to 209 km s@xmath1 in the outer ring / pseudoring .
the outputted velocity dispersion map showed a relatively quiescent velocity field with low dispersions throughout most of the disk . within the same ranges of sky - plane radii
as used in figure [ vpadiagrams ] , we find @xmath23=33@xmath116 km s@xmath1 for the nuclear ring and 11@xmath1113 km s@xmath1 for the inner and outer rings , where the errors are standard deviations .
dispersions are significantly higher in the nuclear ring , indicating greater gas turbulence in that region .
the simulations are performed with a high - order godunov code ( antares ) , in which the calculation of hydrodynamic fluxes on zone interfaces is obtained from the exact riemann solution ( yuan & yen 2005 ) . although the physical problem is naturally described using polar coordinates in cylindrical symmetry , there are advantages in using cartesian coordinates in the present computations . in particular , one can avoid the need to impose an artificial inner boundary condition to handle the coordinate singularity at the origin .
moreover , the very short time steps resulting from small grid cells near the origin in the azimuthal direction can be avoided in using cartesian coordinates . at the outer boundary ,
radiation boundary conditions are imposed .
that is , wave characteristic decomposition is performed in each boundary cell , allowing waves to propagate outward , but suppressing incoming waves . in this
treatment , reflection is not permitted at the boundary .
we consider the evolution of the gaseous disk in the isothermal approximation .
the equations governing the flow corresponding to mass conservation and motion are given as : @xmath24 where @xmath25 denotes the surface density of the disk and @xmath26 denotes the velocity vector .
here , @xmath27 is the gas pressure integrated in the @xmath28-direction . for an isothermal gas , @xmath29 where @xmath30 is the sound speed .
the total gravitational potential , @xmath7 , is composed of three components : @xmath31 the first term , @xmath32 , is a central potential supporting a uniform , differentially rotating disk , i.e. , @xmath33 where the angular speed @xmath34 is determined from the observed rotation curve .
the second term , @xmath35 , is the rotating non - axisymmetric potential of interest . for the purpose in this paper ,
we adopt a stellar bar rotating as a rigid body : @xmath36,\ ] ] where @xmath37 is the angular speed of the bar and only the first non - vanishing harmonic term is included .
note that the axisymmetric part of the bar potential is already contained in @xmath32 .
a simplified functional form for the amplitude is adopted as : @xmath38 where @xmath39 , @xmath40 , and @xmath41 = 1.0 kpc . @xmath42 and @xmath43 are both dimensionless . the functional dependence on the radial coordinate , @xmath43 , is chosen to preserve the asymptotic behavior of the potential .
it is proportional to @xmath44 as @xmath45 and approaches @xmath46 for large @xmath43 .
thus , the force of the bar at @xmath47 is zero , while it behaves as @xmath48 when @xmath43 is large .
the parameter @xmath49 denotes the radial distance for which the bar potential is a minimum .
finally , @xmath50 , the last term of @xmath7 , represents the self - gravitational potential of the gaseous disk .
the hydrodynamic code is coupled to a fast fourier transform ( fft ) poisson solver to include the self - gravitation of the disk in the calculation .
the poisson equation for the disk is intrinsically 3-dimensional , however , the complete potential function problem is not solved .
in contrast the force is calculated in the plane of the disk by integration where the force can be written as a double summation of the product of surface density and a kernel , which is in the form of a convolution . the computational demand associated with the convolution of two vectors of length @xmath51
is reduced from @xmath52(@xmath53 ) to @xmath52(@xmath51ln@xmath51 ) with the fft . as a result
, the entire force computation has a demand of @xmath52(@xmath53(ln@xmath51)@xmath54 ) .
since the fft is only used to accelerate the computation , our poisson solver does not require the periodic boundary condition .
the initial surface density distribution of the gas is taken to be an exponential law of the form @xmath55 where @xmath56 is the initial surface density at the center , taken to have a value of 10 @xmath57 .
the value of @xmath58 is specified by requiring @xmath25 = ( 1/100)@xmath56 at @xmath59 = 15 kpc .
the actual values of the densities are unimportant , as long as the self - gravity of the gas is neglected , since they can be normalized without affecting eqs .
( 1 ) and ( 2 ) . for the axisymmetric background force , we use the representation of the rotation curve of ngc 6782 illustrated in figure [ meanrotc ] . here the points from figure [ rotc ] have been folded around the center .
since it is likely that the derived rotation curve is affected by noncircular perturbations resulting from the spiral arms and rings , we use a least - squares approximation to fit the data points , instead of interpolating between them , in the form of a nearly flat rotation curve , which is written as @xmath60 with @xmath61 = 0.01 .
the optimal value of @xmath62 is found to be 224 km s@xmath1 .
the corresponding precession frequency curves are shown in figure [ pfreqs ] .
the horizontal line represents the pattern speed of the bar , @xmath37 .
the intersection of @xmath37 with the @xmath63 and @xmath64 curves , determine the location of the inner lindblad resonance ( ilr ) and outer lindblad resonance ( olr ) as well as the inner and outer 4:1 resonances .
here , @xmath65 is the epicyclic frequency . at the start of the simulation
the gas is placed on circular orbits with rotational velocities given by the rotation curve .
these velocities are adjusted to take account of pressure gradients caused by the gradient in the density distribution to ensure that the gas is initially in a state of equilibrium . in order to reduce numerical noise , the amplitude of the bar potential , @xmath66 ,
is increased gradually , growing linearly from zero to the full value during one rotational period of the bar .
this amount of time was found to be necessary to avoid the formation of transients resulting from the growth of the perturbing potential . to complete the specification of the input parameters ,
we adopt a sound speed , @xmath30 , for the gaseous disk of 10 km s@xmath1 ( lindblad & kristen 1996 ) .
provided that the sound speed is within reasonable limits , 5@xmath6715 km s@xmath1 , the gaseous response is not very sensitive to its value .
all the calculations presented in this paper are performed with a 1024 @xmath68 1024 uniform grid .
the computational domain corresponds to a physical region of 40 kpc @xmath68 40 kpc , resulting in a cell size of 39 @xmath68 39 pc@xmath69 ( or [email protected]@xmath12 ) .
the appearance of ngc 6782 is bisymmetrical , and the selection criteria of a successful model are based on the following requirements which are similar to those used by lindblad & kristen ( 1996 ) : \1 . the nearly circular bright nuclear ring should be reproduced .
the gas lanes in the model should have approximately the same shape and position as the observed dust lanes .
\3 . the diamond - shaped ( or pointed oval ) inner ring should be reproduced .
the faint double outer ring - pseudoring should be reproduced .
the characteristics in the observed velocity field , e.g. the circular rotating central disk and the bent isovelocity curves on the diamond - shaped ring , should be reproduced .
we judge our models by visual inspection of the model density and velocity maps overlaid on the observed ones instead of adopting a quantitative method , e.g. @xmath71 , which can be misleading if the strong density features or steep velocity gradients present in the models do not exactly overlap their observed counterparts
. such comparisons would be sufficient in the context of our simple model .
as pointed out by yuan & kuo ( 1997 ) , the spiral waves excited at the outer inner lindblad resonance ( oilr ) , where the epicyclic frequency , @xmath65 , of the fluid is twice the angular frequency in the rotating frame of the bar , propagate inward . the negative angular momentum carried by these waves
will be deposited over the annular region through which they propagate .
the disk material within these regions , after losing angular momentum , would flow toward the center .
while their theory applies only to the vicinity of the resonance , the results from the global nonlinear asymptotic theory of spiral density waves ( see yuan & yang 2006 ) reveal that the wavelength of the density waves decreases as they propagate inward from the oilr .
thus , the spiral arms become more tightly wound and take the appearance of a ring around the center .
we note however that these results are all based on steady - state solutions and do not describe the morphology after material has been swept by the waves and accreted to the center . by carrying out numerical simulation experiments ,
we find the quasi - steady state of a bar - driven gaseous disk with an oilr is characterized by an `` eye - shape '' morphology which includes a pair of open , strongly shocked spiral arms in the outer depleted region and a pair of tightly wound spiral arms embedded in a more widespread ring in the circumnuclear region . based on these experimental results , an oilr is required in order to simulate the nuclear ring and the gas dust lanes , confirming the results obtained by athanassoula ( 1992 ) .
an additional test used to validate our numerical simulations is to consider the case when the olr and oilr simultaneously exist . for this test
, a proper angular speed for the bar is chosen such that the locations of the olr and the oilr are in the appropriate range .
if the strength of the bar is weak , waves excited at the olr propagate outward and those at the oilr propagate inward and no waves are excited in between , as predicted by linear theory . by increasing the strength of the bar , the evanescent ( wave - forbidden ) region between the olr and oilr narrows in extent and the inner 4:1 higher harmonic spiral density waves between them are excited .
their interaction with the waves excited at the oilr form a shape similar to the diamond - shaped inner ring seen in ngc 6782 . based on this test and the observations , the values for the angular pattern speed , the location of the potential minimum , and the strength of the bar potential can be obtained . the pattern speed @xmath72 controls the positions of the resonances and thus the morphology of the spirals . according to the images of ngc 6782 and the results of our simulation tests
, @xmath37 takes values between 25 @xmath73 38 km s@xmath1 kpc@xmath1 , corresponding to a radial interval for the radius of the olr of 10 kpc @xmath74 r@xmath75 @xmath74 15.3 kpc .
the location of the minimum and the strength of the bar potential can also be determined .
in particular , our test results show that the nuclear ring forms inside the location of the minimum and that the stronger the bar potential , the smaller the ring . considering both factors simultaneously , the location of the minimum of the bar potential , a@xmath3 in eq.7 , is taken in the range 2.3 @xmath74 a@xmath76 3.2 kpc . to parameterize the strength of the bar potential , we denote by b the ratio of the effective radial force exerted on the disk by the rotating bar ( yuan & kuo 1997 ) to the force for the circular motion from the rotation curve at the location of olr .
this choice is motivated by the fact that the waves are excited at the resonances .
it is found that b should lie between 3% and 10% .
in order to quantitatively compare our theoretical results with the observations , the plane of the theoretical disk must be oriented with respect to the line of sight . from the analysis of the 2d fabry - perot velocity field , the p.a . of the line of nodes and
the inclination of ngc 6782 are 35@xmath19 and 27@xmath19 respectively . the p.a . of the major bar of the galaxy
is 177@xmath19 .
the best - fit values of the parameters to the observations are @xmath37 = 25 km s@xmath1
kpc@xmath1 , a@xmath3 = 3.2 kpc , and b = 5% .
the inferred value of @xmath37 implies the positions of the inner / outer lindblad resonances ( ilr / olr ) , the inner 4:1 ultraharmonic resonance ( uhr ) , and the corotation resonance ( cr ) are located at r@xmath77 = 2.61 kpc , r@xmath78 = 5.78 kpc , r@xmath79 = 8.92 kpc , and r@xmath75 = 15.3 kpc .
these locations are shown relative to the galaxy s morphology in figures [ resonances ] and [ resonances1 ] . in figure
[ resonances ] , the circles are superposed on a deprojected @xmath5-band image where the bar axis has been rotated to a horizontal orientation .
the inner ring is confined between ilr and uhr , while the outer r@xmath3 ring lies between cr and olr , close to the _ outer
_ 4:1 uhr at r = 12.1 kpc ( where @xmath80 ) according to figure 8 .
the positioning of the r@xmath3 ring is very similar to what treuthardt et al .
( 2008 ) found for the barred spiral galaxy ngc 1433 using sticky particle simulations .
the nuclear ring is well inside ilr .
figure [ resonances1 ] shows the same circles superposed on a contour plot of six @xmath5-band isophotes . in this plot
, the horizontal line indicates the extent of the bar as estimated by rautiainen et al .
the plot shows that the isophotes of ngc 6782 continue to be oval and nearly aligned with the bar all the way to cr .
the projected surface density distribution for the best - fit is illustrated in figure [ surfdens2]a .
its superimposition onto the optical multicolor image and the @xmath5-band image are shown in figures [ surfdens2]c and [ surfdens2]e , respectively .
specifically , figure [ surfdens2]c shows that the location and shape of the bright nuclear ring , the dust lanes , as well as the pointy oval inner ring are reproduced by the simulations . at the northwestern outer arm in figure [ images ] , upper left panel , there are two faint branches near the tip of the inner ring , which are also reproduced by our simulation ( see figure [ surfdens2]e ) .
we note , however , that the fit is not without deviations since the simulation can not fit the outer part of the outer arms .
observationally , the arms extend almost one half of a revolution , but those in our simulation only extend one fourth revolution and are much broader .
this is because only the gas response is considered in the simulation and the stellar response has not been taken into account .
although it is possible to produce longer spiral arms in the simulation by adding a weak spiral potential ( lindblad et al 1996 , lindblad & kristen 1996 ) , we have chosen not to do so in this paper .
the imposition of a spiral potential can generate waves by the resonant excitation as well as by forced oscillation ( wang 2006 ) .
these two types of waves may interact , thereby complicating the morphology .
since the main features in the images of ngc 6782 are well delineated and smooth , we seek to use a simple model to reproduce these features as much as possible in order to gain a fundamental understanding of their formation and evolution .
the evolution of the density features has been followed in the simulation to gain insight into their formation .
initially , spiral waves excited at the ilr appear first .
as the strength of the bar potential is increased , the spiral waves intensify eventually leading to the formation of a shock .
subsequently , the inner parts of the spirals become straighter along the major axis of the bar and waves propagate inward . simultaneously , the waves excited at the inner 4:1 resonance appear , growing stronger and propagating inward .
their propagation leads to their interaction with the waves excited at the ilr to connect with the straight gas lanes to form the pointy oval inner ring .
this evolution is consistent with schwarz s ( 1984 ) suggestion that inner rings are connected to the inner 4:1 resonance located just inside the cr .
the straight gas lanes propagating inward become more tightly wound and gradually form a nuclear ring .
a quasi - steady state is reached after one and three fourths revolutions of the bar , or 430 myr .
the sequential images are shown in figure [ gasdisk ] .
the general result concerning the formation of a nuclear ring is similar to that found by piner , stone & teuben ( 1995 ) in their numerical simulations of barred galaxies .
the nuclear ring has a rich structure and resembles a tightly wrapped spiral .
this is consistent with the result presented here , but the requirements necessary for its formation differ in detail .
specifically , piner et al . (
1995 ) require that their model galaxy must have two inner lindblad resonances and the extrema of the @xmath81 curves along the major and minor axes between the resonances must be at the same radial position .
in contrast , only one inner lindblad resonance exists in our model galaxy . as the density waves excited at the ilr propagate inward , their wavelength decreases , evolving to a more tightly wound structure about the center ( yuan & yang 2006 ) .
as the galaxy evolves , mass is continuously accreted into the central region .
the tightly wound spirals become more dense and a ring is formed .
the existence of strong inner rings , offset dust lanes , and circumnuclear rings has also been found in the n - body simulation study by byrd et al .
some of their numerical results resemble ngc 6782 provided that the pattern speed of ngc 6782 is slow such that both an inner second harmonic resonance and an inner lindblad resonance are allowed .
the isovelocity contours from the simulation velocity field are plotted on top of both the simulation density map and the observed fabry - perot velocity field in figures [ isovels]a and b respectively . for these comparisons
, we have smoothed the simulation velocity field to the same spatial resolution as that of the observation .
strong velocity gradients across the dust lanes are seen in figure [ isovels]a . across the dust lanes
, matter collides with the arm from the concave ( or inner ) side and is deflected by the shock , resulting in mass inflow to fuel the nuclear star formation ring .
it is noteworthy in figure [ isovels]b that at the southeastern part of the pointy inner ring , the velocity contours bend outwards along the spiral arms in both the simulation and observation .
this is one of the distinct features for the density waves excited at the ilr and is an important diagnostic .
based upon this comparison , the inner ring is directly related to the ilr .
the outward bending of the curves is caused by the same mechanism that produces the sharp velocity gradients across the dust lanes . from the amount of bending of the isovelocity curves at the respective positions
, it is inferred that the shock in the dust lanes is stronger than that in the pointy inner ring .
since the deflection of the isovelocity curves across the dust lanes is very large , the amount of shear is so high that it is likely that a cloud at the shock loci will be sheared out before collapsing .
therefore , the absence of hii regions or signs of recent star formation in these dust lanes can be naturally understood . on the other hand ,
clouds located at the position of the density enhancements in the inner ring , which undergo compression upon passage through the shock , will have sufficient time to collapse before being sheared out . as a result ,
the inner ring shows clear signs of recent star formation .
this is consistent with athanassoula s ( 1992 ) result , according to which these features should exist in strongly barred galaxies .
although the ratio of the effective radial force to the force for the circular motion is 5% at olr , it is 21% at ilr . since almost all the important features of our simulation lie inside the cr ,
we conclude that the force associated with the rotating bar in our simulation for ngc 6782 is strong . for comparison , laurikainen et al .
( 2004 ) used a near - infrared image from the ohio state university bright galaxy survey ( eskridge et al .
2002 ) to estimate the maximum relative torque in ngc 6782 .
their value , @xmath82 = [email protected] , located at r = 24.4 arcsecs ( equal to 4.61 kpc ) , indicates only a moderate bar strength characteristic of sab galaxies ( see also buta & block 2001 ) .
the value of @xmath82 in our model for the best fit case , which is located at a different distance , r = 3.2 kpc , is found to be 0.135 . as noted in the paper by crocker et al .
( 1996 ) and shown again in figure [ images ] , the hii regions in the pointy inner ring of ngc 6782 are bright around the cusps and mostly concentrated in the vicinity of the ring major axis . in their sample of 32 ringed galaxies , not only ngc 6782 but also ic 1438 and ugc 12646 exhibit most of their emission near the ring major axis with very little emission near the ring minor axis .
this is similar to ngc 1433 ( buta et al . 2001 ) and ngc 3081 ( buta et al .
2004 ) , but there is more emission near the minor axes in these two cases .
an explanation for this phenomenon has been suggested by contopoulos ( 1979 ; see also byrd et al .
2006 ) involving the relative particle motions within the cr region .
in particular , particles would remain longer at the apocentra of their orbits than at the pericentra , which would enhance the density along the major axis of all orbits within the cr .
thus , near the true major axis of an elongated inner ring which lies inside cr , hii regions are expected to be concentrated . in our fully hydrodynamical
simulation , the high density region around the two tips of the pointy inner ring is explained in terms of the interaction of the gas with the curved shocks . to see this
, we present the flow pattern in a reference frame rotating with the bar in figure [ gasflows ] . here , the arrows represent the velocity vectors at a given point overlaid on the face - on surface density map of the gas . since we seek to understand the existence and spatial distribution of the hii regions in the inner ring , the area in figure [ gasflows ] lies inside the cr , and we only plot the flow lines related to the inner ring . note that since the gas flow field in our simulation is not steady , but only quasi - steady we use flow lines rather than streamlines to demonstrate their instantaneous directions . in order to see clearly how flow lines bunch up , we only plot them for less than one revolution .
flow lines start from a number of equidistant points along the direction of the bar semimajor axis along the x - axis . focusing on the left side of figure [ gasflows ]
, it can be seen that they hit the shock on the upper left and are deflected and bunched up to compress the gas there .
the compressed gas may trigger star formation and then move along the flow lines .
a moment later , the gas hits the shock on the lower left and the same process repeats itself . from the velocity vectors
, it can be seen that the gas around the tips of the pointy inner ring is moving more slowly than that near the ring minor axis , implying that the gas remains for a longer time near the ring major axis , consistent with contopoulos result ( 1979 ) . as a result
, the hii regions in the pointy inner ring would be expected to be present around the cusps .
simulations carried out with and without self - gravitation produce similar morphologies with the same overall shapes .
the primary quantitative differences are the radii of the nuclear rings .
in particular , the nuclear ring in the simulation with self - gravitation is larger in extent than in the case without the inclusion of self - gravity effects .
this result is consistent with the asymptotic solutions in yuan & cheng ( 1989 ) and yuan & kuo ( 1997 ) who showed that the self - gravity will affect the radial location and orientation of the spiral pattern , but not its overall shape .
this result follows from the fact that self - gravitational effects will shift the q - barrier to move toward the corotation region . as a result
, the entire spiral pattern is shifted toward corotation , inward for the olr and outward for the ilr .
since the spiral pattern in ngc 6782 lies inside the corotation point , the effect of self - gravity should move the spiral pattern outward and increase its extent . for stronger self - gravity ,
the effect is greater and it would be manifested most clearly on the dense nuclear ring in the simulation result .
our findings on ngc 6782 may be compared with other recent studies .
patsis et al .
( 2003 ) analyzed the shapes of inner sb rings from the point of view of two- and three - dimensional periodic orbits in a potential consisting of a miyamoto disk , a plummer bulge , and a 3d ferrers bar , and considered the case of ngc 6782 as an example .
they concluded that the pointed - oval inner ring of ngc 6782 can be interpreted in terms of 2d 6:1 orbits of the @xmath83 family " and two associated groups of stable 3d families .
no orbits of the bar - supporting @xmath84 family were needed to explain the shape and location of the ring .
rautiainen et al .
( 2005 ) used near - infrared images to infer gravitational potentials in 38 galaxies , including ngc 6782 , and then to evolve a cloud - particle disk in these potentials until the simulated morphology matches the @xmath5-band morphology .
the goal was to measure the ratio of the corotation radius ( @xmath85 ) to the bar radius ( @xmath86 ) and compare it with model expectations for fast and slow bars . for ngc 6782 ,
the best - fitting simulation gave a corotation radius of 37@xmath101 , compared to 47@xmath102 in our model .
with a measured bar radius of 31@xmath103 , rautiainen et al .
derived @xmath87=1.2 , in the fast bar " domain of debattista & sellwood ( 2000 ) . for our value of @xmath85 ,
the ratio is 1.5 , which would be in the slow bar " domain .
as we have shown , ngc 6782 has an oval zone that almost fills the minor axis dimension of the outer r@xmath3 ring .
our cr radius fully encompasses this feature which extends well beyond the ends of the main bar .
quillen & frogel ( 1997 ) focussed on the shape of the outer r@xmath3 ring in ngc 6782 as a means of deducing the mass - to - light ratio of the bar . for our adopted orientation parameters
, we have shown that the deprojected position angle of the r@xmath3 ring relative to the bar is 74@xmath18@xmath113@xmath18 , while buta ( 1995 ) derived a mean orientation of 90@xmath18@xmath119@xmath18 for r@xmath3 rings .
on the assumption that the ring is actually aligned perpendicular to the bar , quillen & frogel ( 1997 ) concluded that the inclination of the galaxy must lie in the range 35@xmath18 @xmath88 @xmath89 @xmath88 45@xmath18 . to reproduce the morphology of the ring for an inclination of 41@xmath18 , quillen & frogel deduced that the bar mass - to - light ratio must be greater than 70% of that required for a maximal disk .
this study lacked the rotation information that we have available in our paper .
romero - gomez et al . ( 2006 ) have considered an alternative interpretation of the rings in a galaxy like ngc 6782 .
they focussed on what they referred to as `` rr@xmath3 '' rings , a combination of an inner ring and an r@xmath3 outer ring like that seen in ngc 6782 . instead of specific resonances ,
this pattern is attributed to what the authors call invariant manifolds " of periodic orbits associated with the l@xmath3 and l@xmath90 lagrangian points in the potential of a rotating bar .
the manifolds can account for a highly elongated , pointed - oval inner ring aligned with the bar and an r@xmath3 outer ring which dips in at its minor axis .
romero - gomez et al . ( 2007 ) further show that the variety of ring types , such as r@xmath3r@xmath91 and r@xmath91 , as well as open spirals such as is seen in ngc 1365 , can also be explained by these manifolds .
we have performed two dimensional modeling of the barred spiral galaxy ngc 6782 using the antares hydrodynamic code in the isothermal approximation within the framework of a simple model based on a nearly flat rotation curve ( eq .
( 9 ) ) , which is fit from the observations . for a perturbing bar potential ( eqs .
( 6 ) and ( 7 ) ) described by the angular speed , the location of the potential minimum , and the strength of the bar , we show that the primary density and velocity features of the galaxy can be reproduced provided that the parameters are in the range , @xmath37 = 25 km s@xmath1 kpc@xmath1 , a@xmath3 = 3.2 kpc , and b = 5% . specifically , the observed shapes and positions of the bright nuclear ring , offset gas lanes , and the pointy oval inner ring are well reproduced in our model .
in addition , our simulation results also reproduce the two faint branches near the tip of the inner ring at the northwestern outer arm in the @xmath5-band image .
based on the time - dependent evolution of the gas disk , the formation of the primary features has been examined . in particular , it has been found that the pointy inner ring is formed via the interaction between the density waves excited at the ilr and the 4:1 higher harmonic density waves under the gravitational influence of a slowly rotating strong bar .
the nuclear ring , on the other hand , is initially a pair of tightly wound spirals which gradually evolve into a ring after matter has continuously accreted into the central region .
upon comparison with the observed velocity field , the numerical simulations provide an understanding for the origin of the outward bending of the isovelocity contours along the spiral arms at the southeastern part of the pointy inner ring .
this feature is caused by gas colliding with the spiral arm from the interior ( concave side ) .
finally , the existence of cuspy hii regions in the pointy oval inner ring can also be naturally explained by the concentration of gas flow lines around the tips of the inner ring . in the future ,
high resolution observations are desirable in order to discern the strong velocity gradients across the dust lanes seen in our simulation to provide further confirmation of the model presented in this paper .
we thank ted williams for assistance with the fabry - perot observations of ngc 6782 , and we also thank guy b. purcell and ted williams for the programs used for reducing the data .
we are grateful to ronald e. taam and paul t.p . ho for many valuable comments .
r.b . acknowledges the support of nsf grant ast-0507140 to the university of alabama .
l .- h.l . and c.y .
acknowledge the support of national science council , taiwan , nsc96 - 2752-m-001 - 002-pae .
athanassoula , e. 1992 , mnras , 259 , 345 buta , r. 1995 , apjs , 96 , 39 buta , r. & block , d. l. 2001 , apj , 550 , 243 buta , r. & crocker , d. a. 1991 , aj , 102 , 1715 buta , r. & crocker , d. a. 1993 , aj , 105 , 1344 buta , r. , ryder , s. d. , madsen , g. j. , wesson , k. , crocker , d. a. , & combes , f. 2001 , aj , 121 , 225 buta , r. , byrd , g. , & freeman , t. 2004 , aj , 127 , 1982 buta , r. , corwin , h. g. , & odewahn , s. c. 2007 , the de vaucouleurs atlas of galaxies , cambridge , cambridge university press byrd , g. , freeman , t. , & buta , r. 2006 , aj , 131 , 1377 byrd , g. , rautiainen , p. , salo , h. , buta , r. , & crocker , d. a. 1994 , aj , 108 , 476 contopoulos , g. 1979 , in photometry , kinematics , and dynamics of galaxies , ed .
d. s. evans ( austin : univ . of texas dept . of astronomy ) , 425 corwin , h. g. , de vaucouleurs , a. , & de vaucouleurs , g. 1985 , southern galaxy catalogue , univ .
texas mono . in astr .
4 crocker , d. a. , baugus , p. d. , & buta , r. 1996 , apjs , 105 , 353 debattista , v. & sellwood , j. 2000 , apj , 543 , 704 de vaucouleurs , g. , de vaucouleurs , a. , corwin , h. g. , buta , r. j. , paturel , g. , & fouqu , p. 1991
, third reference catalogue of bright galaxies , new york , springer england , m.n .
1989 , apj , 344 , 669 eskridge , p. et al .
2002 , apjs , 143 , 73 goldreich , p. , & tremaine , s. 1979 , apj , 233 , 857 erwin , p. 2004
, a&a , 415 , 941 grouchy , r. , buta , r. , salo , h. , & laurikainen , e. 2008 , in preparation huntley , j. m. , sanders , r. h. , & roberts , w. w. 1978 , apj , 221 , 521 landolt , a. u. 1992 , aj , 104 , 340 laurikainen , e. , salo , h. , buta , r. , & vasylyev , s. 2004 , mnras , 355 , 1251 lindblad , p. a. b. , lindblad , p. o. , & athanassoula , e. 1996 , a&a , 313 , 65 lindblad , p. a. b. , & kristen , h. 1996 , a&a , 313 , 733 patsis , p. a. , skokos , ch . , & athanassoula , e. mnras , 346 , 1031 piner , b. g. , stone , j. m. , & teuben , p. j. 1995 , apj , 449 , 508 quillen , a. c. & frogel , j. a. 1997 , apj , 487 , 603 rautiainen , p. , salo , h. , & laurikainen , e. 2005 , apj , 631 , l129 romero - gomez , m. , masdemont , j. j. , athanassoula , e. , & garcia - gomez , c. 2006 , a&a , 453 , 39 romero - gomez , m. , athanassoula , e. , masdemont , j. j. , & garcia - gomez , c. 2006 , a&a , 472 , 63 rubin , v. , burstein , d. , ford , w. k. , & thonnard , n. 1985 , apj , 289 , 81 sanders , r. h. , & tubbs , a. d. 1980 , apj , 235 , 803 schwarz , m. p. 1984 , mnras , 209 , 93 treuthardt , p. , salo , h. , rautiainen , p. , & buta , r. 2008 , arxiv:0804.3356 ( astro - ph ) wang , h .- h .
2006 , ms .
thesis , national taiwan university warner , p. j. , wright , m. c. , & baldwin , j. e. 1973 , mnras , 163 , 163 yuan , c. 1984 , apj , 281 , 600 yuan , c. , & cheng , y. 1989 , apj , 340 , 216 yuan , c. , & cheng , y. 1991 , apj , 376 , 104 yuan , c. , & kuo , c .-
1997 , apj , 486 , 750 yuan , c. , & yen , d. c .- c .
2005 , jkas , 38 , 197 yuan , c. , & yang , c .- c . 2006 , apj , 644 , 180 ll @xmath92 & 19@xmath93 19@xmath94 35@xmath951 + @xmath96 & @xmath1660@xmath18 01@xmath21 12@xmath10 + sgc type & ( r)sab(r)a + csrg type & ( r@xmath97)sb(r)0/a + de vaucouleurs atlas type & ( r@xmath3r@xmath91)sb(r)a + logd@xmath98 ( 0@xmath211 ) & 1.355 + logr@xmath98 & 0.050 + @xmath99 & 52@xmath185 ( 1950 ) + @xmath100 & 12.85 + @xmath101 & 0.91 + @xmath102 & 1.22 + photometric axis ratio @xmath103 ( 70@xmath10 @xmath74 @xmath59 @xmath74 89@xmath10 ) & 0.894 @xmath11 0.002 + photometric major axis @xmath104 ( 70@xmath10 @xmath74 @xmath59 @xmath74 89@xmath10 ) & 34@xmath189 @xmath11 0@xmath185 ( 1950 ) + photometric inclination @xmath105 & 27@xmath182@xmath110@xmath182 + kinematic inclination @xmath106 & 27@xmath18@xmath115@xmath18 + kinematic line of nodes position angle @xmath107 & 35@xmath18@xmath112@xmath18 + heliocentric systemic radial velocity @xmath108 & 3921@xmath113 km s@xmath1 + deprojected nuclear ring major axis diameter & 9@xmath109@xmath110@xmath102 + deprojected nuclear ring axis ratio & [email protected] + deprojected inner ring major axis diameter & 53@xmath102@xmath110@xmath102 + deprojected inner ring axis ratio & [email protected] + deprojected bar - inner ring position angle & 1@xmath188@xmath111@xmath181 + deprojected
outer r@xmath3 ring major axis diameter & 130@xmath10@xmath111@xmath10 + deprojected outer r@xmath3 ring axis ratio & [email protected] + deprojected bar - outer r@xmath3 ring position angle & 74@xmath18@xmath113@xmath18 + | ngc 6782 is an early - type barred spiral galaxy exhibiting a rich and complex morphology with multiple ring patterns . to provide a physical understanding of its structure and kinematical properties ,
two - dimensional hydrodynamical simulations have been carried out .
numerical calculations reveal that the striking features in ngc 6782 can be reproduced provided that the gas flow is governed by the gravitational potential associated with a slowly rotating strong bar .
in particular , the response of the gaseous disk to the bar potential leads to the excitation of spiral density waves at the inner lindblad resonance giving rise to the appearance of a nearly circular nuclear ring with a pair of dust lanes . for a sufficiently strong bar potential ,
the inner 4:1 spiral density waves are also excited .
the interaction of the higher harmonic waves with the waves excited at the inner lindblad resonance and confined by the outer lindblad resonance results in the observed diamond - shaped ( or pointy oval ) inner ring structure . the overall gas morphology and kinematical features
are both well reproduced by the model provided that the pattern speed of the bar is @xmath0 km s@xmath1 kpc@xmath1 . |
the inner @xmath9au of protoplanetary disks is thought to be the region in which most exoplanets form @xcite .
initially , the inner disk likely has a smooth radial distribution of gas and dust that is truncated by the stellar magnetic field at the co - rotation radius , within 0.1au .
from this point , disk material accretes onto the central star through magnetospheric accretion ( e.g. * ? ? ?
however , over time , inner disks evolve , and gaps and inner holes frequently develop within a few myr @xcite . some of these gaps may be carved by photo - evaporation @xcite , and others by dynamical clearing by planets coupled with pressure - driven dust migration @xcite .
the evolutionary window available for planet formation is therefore limited by , and linked to , the time scale of inner disk dispersal of a few myr @xcite .
when observing the inner regions of protoplanetary disks and the process of planet formation , a central challenge is that the angular size of these regions is small ( @xmath10 ) , even for the closest star - forming clouds .
further , the gas temperatures in the planet - forming region typically spans @xmath11k , pushing the main observational line tracers into the near- to far - infrared wavelength range .
this leads to the common use of infrared molecular spectroscopy to investigate the properties of inner disk gas .
the co molecule is a ubiquitous and sensitive tracer of gas in protoplanetary disks .
while pure rotational co lines observed by millimeter interferometers trace cold gas extending to the outermost disk regions , co lines from rovibrational transitions at infrared wavelengths are typically excited closer to the central star , where the disk is hotter and exposed to strong infrared ( ir ) and ultraviolet ( uv ) radiation fields .
surveys of rovibrational co emission from the fundamental ( @xmath12 ) band near 4.7@xmath0 m have demonstrated that the lines probe the properties of inner disks between the magnetospheric accretion radius and the inner 10 au ( e.g. * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
co line profiles , when spatially- and velocity - resolved , have shown a wealth of co isotopologues and complex line profiles , as indicative of a range in excitation conditions and of physically distinct emission components @xcite .
emission from @xmath13co @xmath14=1 - 0 and @xmath14=2 - 1 lines is strong enough to be detected in the majority of gas - rich disks , demonstrating that co is often vibrationally excited . in the innermost disk
, co can be efficiently excited by ir photons from the stellar radiation and the local @xmath0m - size hot dust through ir pumping ( e.g. * ? ? ?
detection of higher vibrational lines in some disks implies vibrational temperatures that are several times larger than the rotational temperatures , providing evidence for the radiative cascade from pumping of directly uv - irradiated co gas @xcite .
detection of rarer isotopologues , including @xmath15co that is commonly seen in its @xmath14=1 - 0 transitions , has been usually attributed to large column densities of co gas found in young disks .
recent analyses of velocity - resolved co rovibrational emission in one protoplanetary disk found that the @xmath14=1 - 0 lines are a superposition of a broad ( fwhm @xmath16kms@xmath17 ) and a narrow ( fwhm @xmath18kms@xmath17 ) components @xcite .
@xcite found evidence for similarly broad and narrow co components in @xmath19 more embedded disks , and found that , while the two components are blended in @xmath13co @xmath14=1 - 0 lines , only the broad component contributes to the @xmath13co @xmath14=2 - 1 lines , while the @xmath15co lines are dominated by the narrow component . in this paper , we systematically apply a spectral decomposition analysis to high resolution ( @xmath20 ) spectra of a sample of @xmath21 protoplanetary disks , to test the hypothesis that co rovibrational emission in protoplanetary disks may generally be described by two distinct components in velocity and excitation : specifically , a _ narrow component _ , or nc , and a distinctly _ broad component _ , or bc .
the decomposition allows us to estimate both the location and excitation temperature of the emitting co gas .
while most disks in the sample do show two distinct co components that get colder at larger disk radii , some show only narrow emission lines and signatures of having cleared - out inner disk regions .
together , this dataset reveals a sequence of inner disk dispersal across a wide range of stellar masses , from 0.3 to 3@xmath22 .
the disks sample includes 37 disks with 4.7@xmath0 m spectra obtained with the cryogenic infrared echelle spectrometer ( crires , * ? ? ?
* ) on the very large telescope ( vlt ) of the european southern observatory ( eso ) , as part of the eso large program 179.c-0151 @xcite .
further , one spectrum was added from program 093.c-0432 , providing an additional epoch of data for a disk around the variable star exlup @xcite .
the original survey included disks mostly around young solar - mass stars ( m@xmath23 m@xmath24 , t tauri stars ) as well as some around intermediate - mass stars ( @xmath25 m@xmath26 m@xmath24 , mostly herbig ae / be stars ) .
the disks were also selected to cover a range of evolutionary phases , from the early stages when the disks are still embedded in diffuse envelopes to the clearing stages of transitional disks .
for this paper , we selected those disks with spectra that show co lines in emission , indicative of the presence of warm co gas in the inner part of the disk .
the sample includes 30 t tauri disks and 7 herbig disks ; 2 disks have been observed to have diffuse envelopes , and 7 disks are known to be transitional ( see table [ tab : incl ] ) .
all spectra were reduced using procedures described in @xcite , and were already presented in several papers , providing general overviews as well as more detailed studies of various subsamples @xcite .
the observations were carried out between 2007 and 2010 , but consistent instrumental parameters were maintained to ensure homogeneity of the data .
the slit width was @xmath27 , providing a spectral resolution of @xmath28kms@xmath17 , which is generally confirmed by measured widths of narrow atmospheric absorption lines .
early - type telluric standard stars were observed at similar airmasses to the science targets ( typically within 0.1 difference ) to allow an accurate telluric correction .
the observations were generally obtained using the crires adaptive optics correction .
the four crires detectors provide a combined instantaneous spectral coverage of @xmath29@xmath0 m .
the observing strategy targeted the @xmath13co rovibrational p - branch lines in several non - contiguous settings , with at least one centered between 4.65 and 4.77@xmath0 m ( see figure [ fig : 5um spectrum ] ) .
additionally , a spectral setting covering wavelengths up to 5.0@xmath0 m was obtained for most targets .
the crires spectra do not provide continuous coverage due to gaps from detector chips and the telluric absorption from atmospheric co lines doppler - shifted from the targeted lines depending on the month of observation . in some cases ,
the spectra were repeated by applying a small shift to the central wavelength or taken in different months of the year to compensate for those effects and provide a more complete spectral sampling ( see * ? ? ?
this strategy was adopted to obtain complete velocity - resolved co line profiles .
in this section we describe the methodology applied to the analysis of crires spectra , while comparison of our results to previous work and discussion are provided in section [ sec : disc ] .
we apply a spectral decomposition algorithm , which essentially consists of the following three basic steps .
_ step 1 _ : : high signal - to - noise profiles of @xmath13co @xmath14=1 - 0 and @xmath14=2 - 1 emission are built by stacking multiple lines from different rotational levels .
_ step 2 _ : : the stacked line profiles are split into two ( broad ; bc and narrow ; nc ) components by scaling the @xmath13co @xmath14=2 - 1 profile to match the wings of the @xmath13co @xmath14=1 - 0 profile , and then subtracting it .
this step reveals if the emission is made of a single ( sc ) rather than of two components ( bc and nc , figure [ fig : multi - comp analysis ] ) .
_ step 3 _ : : the line widths and the vibrational @xmath3 ratios are then measured for each detected component . this procedure is applied to each co spectrum in our sample .
the first two steps are described in sections [ sec : ana1 ] and [ sec : ana2 ] .
the third step provides the data used to find two empirical relations that probe the properties of inner disks ( sections [ sec : ana3 ] and [ sec : ana4 ] ) .
we build high signal - to - noise co line profiles for each science target by stacking several transitions observed in the 4.7@xmath0 m spectra .
nominally , the maximum spectral coverage obtained in this survey includes ro - vibrational co transitions from p(1)p(32 ) for @xmath13co @xmath14=1 - 0 , p(1)p(26 ) and r(0)r(6 ) for @xmath13co @xmath14=2 - 1 , and p(1)p(23 ) and r(0)r(15 ) for @xmath15co @xmath14=1 - 0 . in practice
, the number of transitions available for each disk depends on the spectral coverage obtained for individual targets and on the gaps caused by detector chips and telluric absorption , which in turn depends on the time of year of a given observation . for stacking and averaging line profiles
, we use the lines that are most commonly available , which have similar integrated flux , and which are not significantly blended with other transitions .
these criteria preserve both the line flux and shape in the stacked profiles , and enforce homogeneity in the analysis method for the entire disk sample .
note that , within the relatively narrow spectral range @xmath30 m , lines out of a given vibrational level have very similar fluxes and excitation energies ( within 10 - 30% ) and line profiles depend only very weakly on the rotational quantum number ( e.g. * ? ? ? * ; * ? ?
* ; * ? ? ?
* ) . for high rotational quantum numbers
, this may no longer be true ( @xmath31 ) .
given these constraints , the @xmath13co @xmath14=1 - 0 lines that are most suitable for stacking are those between p(7 ) and p(12 ) , while the @xmath13co @xmath14=2 - 1 lines are those between p(3 ) and p(6 ) . as for the @xmath15co lines , several
can typically be used over the entire spectrum , where not blended with @xmath13co .
each line is normalized to its local continuum , re - sampled on a common velocity grid , and the stacked line profiles are built by weighted average of the pixel values from the individual transitions . while the stacked @xmath14=2 - 1 line profiles typically have continuous spectral sampling , @xmath14=1 - 0 lines often have spectral gaps on one side due to the counterpart telluric absorption lines at the doppler velocity of the target at the time of observation ( see for instance the gap on the red side of the @xmath14=1 - 0 line in figure [ fig : multi - comp analysis ] ) . in some disks ( cw tau , doar 24e s , ht lup , vssg 1 ) , an absorption component is present on top of @xmath13co emission in the @xmath14=1 - 0 lines .
this is typically due to absorption from unrelated foreground gas @xcite . in these targets , we exclude pixels affected by the absorption component from the stacked profiles and further analysis .
l l l l c l c l c l aatau & s & tt & 0.85 & _ m1 _ & - & - & 71(1 ) & _ i1 _ + as205 n & d & tt & 1.1 & _ m2 _ & 20(5 ) & _ i2 _ & 25 & _ i3 _
+ as209 & d & tt & 1.4 & _ m3 _ & - & - & 3840 & _ i3 _ + cra - irs2 & d & tt , em & 1.0 & _ m4 _ & 17(11 ) & _ this work _ & - & - + cvcha & d & tt & 2.1 & _ m3 _ & 30(12 ) & _ this work _ & - & - + cwtau & d & tt & 1.2 & _ m2 _ & 28(10 ) & _ this work _ & - & - & @xmath32 agrees with @xmath33 from _ i4 _ + dftau & d & tt & 0.5 & _ m5 _ & 65(15 ) & _ this work _ & - & - & @xmath32 agrees with @xmath34 from _ i5 _ + doar24e s & d & tt & 0.7 & _ m2 _ & 20(5 ) & _ i2 _ & - & - + doar44 & s & tt , tr & 1.3 & _ m3 _ & - & - & 40(20 ) & _ i3 _ & + drtau & d & tt & 1.0 & _ m2 _ & 9(5 ) & _ i2 _ & 37(3 ) & _ i6 _ & + ec82 & s & tt & 1.0 & _ m4 _ & 51(18 ) & _ this work _ & - & - & + exlup 08 & d & tt & 0.6 & _ m6 _ & 44(8 ) & _ this work _
& - & - & ex lup observation from august 2008 + exlup 14 & d & tt & 0.6 & _ m6 _ & 48(17 ) & _ this work _
& - & - & ex lup observation from april 2014 + fntau & s & tt & 0.35 & _ m7 _ & - & - & 20(10 ) & _ i7 _ + fztau & s & tt & 0.7 & _ m3 _ & 38(15 ) & _ this work _ & - & - & + gqlup & d & tt & 0.8 & _ m2 _ & 65(10 ) & _ i2 _ & - & + hd135344b & s & h , tr & 1.6 & _ m2 _ & 14(4 ) & _ i2 _ & 23(0.3 ) & _ i8 _ + hd142527 & s & h , tr & 3.5 & _ m2 _ & 20(2 ) & _ i2 _ & 23(8 ) & _ i9 _ + hd144432s & s & h & 1.7 & _ m2 _ & 27(3 ) & _ i2 _ & - & - + hh100 & d & tt , em & 0.4 & _ m3 _ & 26(4 ) & _ this work _
& - & - & + htlup & d & tt & 2.5 & _ m3 _ & 28(8 ) & _ this work _ & - & - & + imlup & d & tt & 1.0 & _
m4 _ & - & - & 49(4 ) & _ i10 _ + irs48 & s & h , tr & 2.0 & _ m3 _ & 42(6 ) & _ i11 _ & 48(8 ) & _ i12 _ + lkha330 & s & h , tr & 2.5 & _ m2 _ & 12(2 ) & _ i2 _ & 42 & _ i13 _
+ rno90 & d & tt & 1.5 & _ m2 _ & 37(4 ) & _ i2 _ & - & - + rulup & d & tt & 0.7 & _ m2 _ & 35(5 ) & _ i2 _ & - & - + scra s & d & tt & 0.6 & _ m3 _ & 26(4 ) & _ this work _
& - & - + scra n & d & tt & 1.5 & _ m2 _ & 10(5 ) & _ i2 _ & - & - + sr21 & s & h , tr & 2.2 & _ m2 _ & 15(1 ) & _ i2 _ & 15(0.4 ) & _ i8 _ & + ttau n & d & tt & 2.1 & _ m3 _ & 20(13 ) & _ this work _
& - & - & @xmath35 agrees with @xmath36 from _ i14 _ + twcha & s & tt & 1.0 & _ m8 _ & 67(23 ) & _ this work _
& - & - + twhya & s & tt , tr & 0.7 & _ m2 _ & 4(1 ) & _ i2 _ & 12 & _ i6 _ & + vssg1 & d & tt & 0.5 & _ m3 _ & - & - & 53 & _ i3 _ & + vvser & s & h & 3.0 & _ m2 _ & 65(5 ) & _ i2 _ & 70(5 ) & _ i15 _ & + vwcha & d & tt & 0.6 & _ m3 _ & 44(17 ) & _ this work _ & - & - + vzcha & d & tt & 0.8 & _ m3 _ & 25(10 ) & _ this work _ & - & - + waoph6 & d & tt & 0.9 & _ m9 _ & - & & 39 & _ i3 _ + wxcha & d & tt & 0.6 & _ m8 _ & 87(31 ) & _ this work _ & - & - l c c c c c c[ht ] as205 n & 52 & 0.38 & 0.17 & 970 & 13 & 2.3 + as209 & 160 & 0.91 & 0.07 & 1054 & 33 & 1.7 + cra - irs2 & 53 & 0.37 & 0.11 & 960 & 17 & 1.1 + cvcha & 102 & 0.25 & 0.18 & 790 & 47 & 0.9 + cwtau & 54 & 0.32 & 0.32 & 890 & 45 & 0.5 + dftau & 89 & 0.43 & 0.20 & 1060 & 53 & 0.6 + doar24e s & 81 & 0.30 & 0.04 & 860 & 29 & 0.4 + drtau & 37 & 0.39 & 0.06 & 990 & 13 & 0.5 + exlup 08 & 155 & 0.61 & 0.06 & 1420 & 32 & 1.3 + exlup 14 & 150 & 0.51 & 0.07 & 1220 & 40 & 1.0 + gqlup & 107 & 0.28 & 0.20 & 830 & 62 & 0.6 + hh100 & 80 & 1.02 & 0.04 & 1200 & 12 & 1.9 + htlup & 190 & 0.65 & 0.05 & 1510 & 35 & 1.6 + imlup & 200 & 1.16 & 0.05 & 1400 & 40 & 1.3 + rno90 & 92 & 0.21 & 0.23 & 730 & 52 & 0.7 + rulup & 111 & 0.50 & 0.07 & 1200 & 18 & 2.5 + scra s & 101 & 0.40 & 0.04 & 1000 & 15 & 1.8 + scra n & 57 & 0.30 & 0.05 & 860 & 12 & 1.1 + ttau n & 110 & 0.57 & 0.07 & 1340 & 28 & 1.1 + vssg1 & 66 & 0.36 & 0.27 & 940 & 35 & 1.0 + vwcha & 80 & 0.32 & 0.16 & 890 & 33 & 0.9 + vzcha & 53 & 0.44 & 0.18 & 1080 & 24 & 0.9 + waoph6 & 133 & 0.33 & 0.07 & 900 & 32 & 1.2 + wxcha & 140 & 0.45 & 0.11 & 1100 & 67 & 0.5 l c c c c[ht ] aatau & 115 & 0.14 & 0.2 & 630 + doar44 & 72 & 0.11 & 0.4 & 590 + ec82 & 49 & 0.03 & 0.9 & 420 + fntau & 9 & 0.01 & 1.8 & 330
+ fztau & 31 & 0.20 & 1.0 & 720 + hd135344b & 14 & 0.05 & 1.7 & 420 + hd142527 & 25 & 0.04 & 2.3 & 460 + hd144432s & 40 & 0.09 & 0.8 & 560 + lkha330 & 9 & 0.09 & 4.7 & 420 + twcha & 65 & 0.16 & 0.7 & 660 + twhya & 6 & 0.09 & 0.3 & 500 + vvser & 55 & 0.16 & 2.9 & 660 + + irs48 & 14 & 0.38 & 16 . & 970 + sr21 & 9 & 0.64 & 6.5 & 1480 + hd100546 @xmath37 & 16 & 0.27 & 15 . & 810 + hd97048 @xmath37 & 17 & 0.33 & 14 . & 900 + hd179218 @xmath37 & 18 & 0.37 & 21 . & 950 + hd141569 @xmath37 & 16 & 0.56 & 17 . & 1300 + hd190073 @xmath37 & 16 & 0.27 & 6.0 & 810 + hd98922 @xmath37 & 21 & 0.29 & 9.3 & 840 the presence of separate broad and narrow components is apparent by visual inspection of many disk spectra ( e.g. figure [ fig : 5um spectrum ] ) . in practice ,
the difference is formalized by comparison of the @xmath14=1 - 0 and @xmath14=2 - 1 stacked line profiles .
we scale the @xmath14=1 - 0 to match the broad wings of the @xmath14=2 - 1 profile , and then subtract one from the other .
the best scaling factor is found by @xmath38 minimization between the scaled @xmath14=1 - 0 line profile and the @xmath14=2 - 1 line profile , excluding the central region of narrow emission , if present .
the nc is typically much narrower than bc , such that most of the line profile can be used for the @xmath38 minimization .
the choice of the wavelength exclusion region is determined iteratively , starting by using the full line profile .
if the residuals after subtraction of the two line profiles are consistent with zero over the entire spectral range of the @xmath14=2 - 1 line , the nc is not detected and the procedure stops there . in the contrary case , the fit is iterated after exclusion of increasingly large portions at the center of the line profile , until the residuals are composed of a flat continuum and of the nc emission line .
figure [ fig : multi - comp analysis ] shows a demonstrative sketch of the outcome , while spectral decomposition plots for all disks are shown in appendix [ app : co_decomp ] .
two basic parameters are measured from the decomposition procedure : the line width and the vibrational flux ratio @xmath39 .
line widths are measured at 50% of the line peak ( the full width at half maximum ; fwhm ) from fits to the line profiles , allowing up to two gaussian functions to account for double - peaked profiles .
the vibrational flux ratio is measured from the scaling factor that matches the @xmath14=1 - 0 to the @xmath14=2 - 1 lines as explained above , and is exactly equivalent to the integrated line intensity ratio between the @xmath14=2 - 1 and the @xmath14=1 - 0 lines in each spectrum ( further details are provided in the appendix ) .
it is possible to measure vibrational ratios in all single - component disks and for the bc in all double - component disks . for the nc , @xmath3
can be measured in only three disks : as 205 n , dr tau , and s cra n. in these disks , the nc can also be identified in the @xmath14=2 - 1 lines through fits of multiple gaussians to the line profile , one for each of the broad and narrow components ( plots are shown in the appendix ) .
the non - detection of the nc in the @xmath14=2 - 1 lines of all other double - component disks is generally consistent with an upper limit on the vibrational flux ratio of @xmath40 .
tables [ tab:2co comp ] and [ tab:1co comp ] report line widths and vibrational ratios for the entire disk sample . by implementing this decomposition procedure , we find that all disks fall into two phenomenological classes : the single - component " disks ( 14 of 37 ) , where the @xmath14=1 - 0 line profile , once scaled down , is consistent with that of the @xmath14=2 - 1 line , and the double - component " disks ( 23 of 37 ) , where the @xmath14=2 - 1 profile matches only the wings of the @xmath14=1 - 0 profile , revealing an additional narrower component of the @xmath14=1 - 0 profile . in the double - component disks ,
the @xmath15co lines are typically significantly narrower than the bc , but match the profile of the nc ( figure [ fig : multi - comp analysis ] ) .
only as 209 and i m lup have @xmath15co lines detected in the bc only , rather than in the nc , while a few disks tentatively show both the bc and nc in @xmath15co ( cra - irs 1 , as 205 n , and ru lup ) . in double - component disks
, the bc profile shows properties distinct from those of the single - component disks : broader lines ( @xmath41 50200kms@xmath17 ) , a clear separation in terms of line flux ratios with a boundary at @xmath42 , and weak or undetected @xmath15co lines .
conversely , the nc shares the same line widths ( @xmath41 1070kms@xmath17 ) , same vibrational ratios @xmath43 , and same @xmath15co detection rate as for co in 12/14 single - component disks ( sc ) of our sample . when using line widths to infer radial properties of keplerian disks , it is necessary to correct for inclination effects .
consequently , we collect values of disk inclinations from the literature , where available , for the disks in our sample .
we carefully searched for the most reliable estimates obtained using spatially - resolved disk images , either from millimeter interferometry , optical / near - infrared ( nir ) scattered - light imaging , or spectro - astrometry from crires .
note that for spectro - astrometry the derived disk inclination is degenerate with an assumed stellar mass . for consistency ,
we therefore adopt both the stellar masses and the disk inclinations reported by @xcite .
table [ tab : incl ] shows the adopted disk inclinations , with notes on the technique used to measure the inclination for each disk . using ten double - component disks with known inclinations ( seven from spectro - astrometry , and three from interferometry )
, we find strong correlation between the co line width and disk inclination , with pearson coefficients of 0.7 for the bc and of 0.9 for the nc .
strong correlation between nir co line widths and disk inclination has been reported previously by @xcite from a sample of five herbig disks .
correlation between these parameters is expected when the gas is in keplerian rotation around the star , a condition that is suggested by the high detection rate of double - peaked line profiles in our disk sample ( @xmath44% ) and that has been proposed since the first surveys of nir co emission in disks ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
note that the two - component decomposition significantly decreases the scatter in this relation relative to what is seen when the line profiles are not fully velocity - resolved as with crires , or if the line profiles were all modeled with a single gaussian profile to derive average line widths ( i.e. without separating bc from nc ) .
gas in keplerian rotation at a disk radius @xmath45 produces emission at a maximal velocity @xmath46 from the rest frequency of a given co line following kepler s law as @xmath47 , where @xmath48 is the disk inclination , @xmath49 is the gravitational constant and @xmath50 is the stellar mass . using this relation , it is possible to derive an emitting radius for the gas in a rotating disk from the observed line widths . in general , emission from higher velocities
originate from smaller orbital radii .
these velocities , usually taken at the 50% or 10% of the line peak value , therefore provide estimates of the disk radii where co is emitting .
we refer to the innermost emitting radius of co , @xmath51 , when derived from @xmath52hwhm as proposed by @xcite . in comparison ,
the velocity at @xmath53hwhm provides an estimate of the disk radius @xmath54 that emits the peak line flux ( in agreement with the emitting radii derived from spectro - astrometry by * ? ? ?
* ; * ? ? ?
figure [ fig : fwhm_incl ] shows the measured co line widths divided by @xmath55 . removing the stellar mass dependence on the inclination relation leaves only the dependence on the keplerian radius @xmath45 , which can be compared to a simple model of gas emission in keplerian rotation around a @xmath56 star ( shown in figure [ fig : fwhm_incl ] at three different disk radii ) . in these terms
, the bc has @xmath57 between 0.01 and 0.1 au .
this is consistent with the disk region of the corotation and magnetospheric accretion radii for solar - mass stars ( e.g. * ? ? ?
* ) , as found by previous studies of rovibrational co emission from disks ( e.g. * ? ? ?
* ; * ? ? ?
the nc , instead , emits from larger disk radii , with @xmath57 between 0.1 and 1 au .
the single - component disks show co emission at radii similar to , or larger than , those of the nc . in figure [
fig : fwhm_incl ] we show linear fits to the bc and nc in disks with known inclinations , obtained using the bayesian method by @xcite .
the regression takes into account the uncertainties on both variables , as well as the intrinsic scatter in the regression relation due to physical quantities not explicitly accounted for .
typical errors on the line widths are @xmath58 for the bc and @xmath59 for the nc , the stellar masses are assumed to be known to 20% and the disk inclinations are typically known to accuracies better than 10% .
the best - fit parameters and their errors are taken as the median and standard deviation of the posterior distributions of the regression parameters .
the median intercept and slope are used to plot the lines in figure [ fig : fwhm_incl ] , while the posterior distributions are plotted in the appendix .
the single linear fits shown in figure [ fig : fwhm_incl ] assume that differences in @xmath57 due to the dependence on stellar and disk luminosity for either the bc and the nc is weak relative to the inclination dependence .
this is supported by the bayesian regression of line widths versus disk inclinations , which confirms strong correlation ( @xmath60 ) for both the bc and nc , although the uncertainties in the regression parameters for bc are 34 times larger than for nc .
this is also consistent with previous results for disks around solar - mass stars where @xmath57 was found to vary only within a factor of a few @xcite , and with the theoretical expectation that the inner radius increases only with @xmath61 .
the residual intrinsic scatter on the linear relation for the bc is four times larger than for the nc , which may point to a remaining dependence on the stellar and ( variable ) accretion luminosity . given the empirical line width - inclination relation , it is possible to estimate inclinations for disks with high resolution co spectroscopy even when resolved imaging is not yet available .
in doing so , it is assumed that the two relations found for the bc and nc represent a general property of rovibrational co emission in the double - component disks .
the advantage of having two relations , one for each velocity component , is that for a given double - component disk there are two constraints to derive one parameter .
we take the measured values of the bc and nc line widths in each disk and use them to derive the disk inclination by using the regression coefficients @xmath62 from the empirical linear relations : @xmath63 where @xmath64 and @xmath65 for the bc , and @xmath66 and @xmath67 for the nc .
we derive two inclination values , one for the bc and one for the nc , and adopt the mean of the 1-@xmath68 range in common between the two . among the single - component disks , only three have unknown inclination : ec 82 , fz tau , tw cha . for these disks ,
we adopt the nc relation , since single - component disks with known inclinations span disk radii similar to those of the nc of double - component disks ( figure [ fig : fwhm_incl ] ) . following this procedure
, we estimate an inclination for a total of fifteen new disks .
these values are included in table [ tab : incl ] . using the inclinations described in the previous section
, we derive the co emitting radii for all disks in the sample . in figure
[ fig : vibr_rin ] we show a diagram made with the vibrational line flux ratios plotted against the co emitting radii .
the diagram at the left of the figure shows a strong logarithmic anti - correlation between @xmath69 and @xmath3 , with a pearson coefficient of -0.9 .
in addition to spanning spatially separated @xmath54 ranges , as already shown in figure [ fig : fwhm_incl ] , the bc and nc clearly span distinct values in @xmath3 . here too , the single - component disks have similar values to those of the nc .
two obvious outliers are sr 21 and irs 48 , both transitional disks with large inner gaps , which have vibrational excitation temperatures as high as the bc at much smaller disk radii ( they will be discussed in section [ sec : disc ] ) . excluding them
, figure [ fig : vibr_rin ] demonstrates that the vibrational flux ratio decreases with disk radius as : @xmath70 the vibrational line flux ratios measure how much the vibrational level @xmath14=2 is populated as compared to @xmath14=1 , which in turn is a measure of the vibrational excitation temperature @xmath71 ( e.g. * ? ? ?
the relation between @xmath3 and @xmath54 can thus be converted into a radial profile for the co vibrational excitation temperature in inner disks .
the conversion of the vibrational flux ratio to a temperature depends on the co column density , so the model - independent relation is that using the vibrational flux ratio ( eqn .
2 ) . to estimate the corresponding temperature relation , we use a model of a slab of gas in thermal equilibrium , where the level populations are defined by a single temperature following the boltzmann distribution . other analyses of co rovibrational emission from disks consistently find a relatively narrow range of column densities .
consequently , we adopt a single @xmath13co column density of @xmath72@xmath73 , as an order - of - magnitude value representative of the range found in previous studies ( @xmath74@xmath75@xmath73 , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the details of the conversion are described in the appendix [ app : vibr_convers ] .
the resulting temperature radial profile is : @xmath76 if disks without previously known inclinations are excluded , the exponent becomes @xmath77 , within the error , indicating that the result is not biased by our empirical derivation of disk inclinations .
the excitation temperatures estimated for each disk are reported in tables [ tab:2co comp ] and [ tab:1co comp ] .
the nominal uncertainty on the temperatures , once the column density is fixed , is less than 50 k as due to the uncertainty on the vibrational ratios ( which is typically @xmath78 ) .
@xcite proposed that the non - detection of @xmath15co lines in the bc , combined with the curvature seen in rotation diagrams , constrained the bc column density to within @xmath79@xmath72@xmath73 . on the other hand ,
rotation diagrams of @xmath15co lines were found to require column densities of @xmath80@xmath75@xmath73 for the nc .
if we relax our assumption and allow the bc to have columns as low as @xmath74@xmath73 , and nc as large as @xmath75@xmath73 , then the temperatures will decrease / increase for the nc / bc , respectively , and the overall temperature profile may exhibit a steeper index up to @xmath81 ( see appendix [ app : vibr_convers ] ) .
the vibrational excitation temperatures of the bc match those previously found from rotational analyses of @xmath13co lines ( 7001700k , e.g. * ? ? ?
* ) , while the co in the nc and sc disks typically have lower temperatures , corresponding to those determined for @xmath15co line fluxes ( 200600k , e.g. * ? ? ?
this is consistent with our finding that the @xmath15co line profiles match those of the nc , as indicative of probing a similar disk region . in turn
, this suggests that the nc is formed in an optically thick layer , leading to high @xmath82 ratios , while the bc must be more optically thin .
a similar conclusion was reached for disks in embedded young stars @xcite .
the radii derived from the bc span the range found previously for disks around solar - mass stars ( 0.010.1au ) while the nc and sc span the range found for herbig and transitional disks ( 0.110 au ) .
this is consistent with the finding that double - component disks ( where bc is detected ) are primarily disks around low - mass stars , while the single - component disks are primarily herbig and transitional disks . the nc and part of sc share similar properties in terms of line widths , orbital radii , and vibrational ratios ( section [ sec : ana ] and table [ tab : co summary ] ) , and they may trace the same gas component , regardless of the presence of a bc .
l c c c c c broad ( bc ) & 50200 & 0.20.6 & 0.040.3 & 8001500 & 3/23 + narrow ( nc ) & 1070 & @xmath83 & 0.23 & @xmath84 & 18/23 + single i ( sc ) & 670 & 0.010.2 & 0.25 & 300 - 700 & 9/12 + single ii ( sc ) & 920 & 0.30.6 & 620 & 800 - 1500 & 8/8 while a disk origin for the bc is supported by the common double - peak profiles , as well as by a keplerian signal seen in spectro - astrometry @xcite , the origin of the nc is less clear .
a disk wind has been proposed for single - peak @xmath13co @xmath14=1 - 0 line profiles ( those dominated by the nc , e.g. as205 n ) to reproduce the sub - keplerian motions of the line peak , observed with spectro - astrometry @xcite . in one case ,
a corresponding large - scale disk wind has possibly been seen with alma @xcite .
an observational characteristic of a disk wind launched at small radii , and preserving angular momentum as it expands , is that it the gas contributing to the line becomes increasingly sub - keplerian closer to the line center .
specifically , the line wings are exactly keplerian , while the inner few kms@xmath17 are strongly sub - keplerian , tracing radii significantly smaller than what their velocity would suggest .
this could explain why observables constructed from the line wings of both the broad and narrow components , as we do here , yield results consistent with a nearly keplerian disk .
broad and narrow components that may be similar to those of the rovibrational co lines have been found in the 6300[oi ] line by @xcite .
these authors suggest a disk origin for the broad component , and an unbound cool disk wind as the origin of the narrow component .
it is not clear whether the narrow components of [ oi ] and co are generally tracing the same disk wind , but they share close resemblance in at least a few disks . for the disks in common between this work and @xcite , the fwhm of nc in co and [ oi ] agrees to within @xmath85kms@xmath17 in dr tau , vz cha , ru lup , tw hya , and cw tau ; in dr tau , the same is true also for the bc of co and [ oi ] . in section
[ sec : ana4 ] we found a strong logarithmic anti - correlation between the co vibrational excitation temperature and emitting radius .
this relation corresponds to an empirical temperature profile of a typical inner disk surface .
however , it was noted that two disks from the sample with large co emitting radii , the transitional disks sr 21 and irs 48 , clearly depart from the relation , showing significantly higher vibrational ratios at their respective co emitting radii than the relation predicts . to explore this further
, we add 6 disks from the crires survey of 12 disks around intermediate - mass ( @xmath86 ) stars from @xcite .
this survey presents 4.7@xmath0 m spectroscopy with a similar experimental design and wavelength coverage as our survey .
specifically , we include the full subset of disks for which lines are detected from both @xmath13co @xmath14=1 - 0 and @xmath14=2 - 1 transitions in a wavelength range similar to that of our survey .
following the same procedure as for other disks , we derive @xmath54 and @xmath71 for the disks around hd100546 , hd179218 , hd190073 , hd97048 , hd141569 , hd98922 ( included in table [ tab:1co comp ] ) .
their locations in the temperature - radius diagram are included in figure [ fig : tempprofs_compar ] , which shows that disks around intermediate - mass stars have vibrational temperatures that are typically higher at a given radius than the relation defined by the lower - mass stars at smaller disk radii .
it is clear that sr 21 and irs 48 belong to this vibrationally hot class of disks .
the combination of crires surveys of disks spanning a wide range of stellar masses therefore demonstrates the existence of two trends : a vibrational temperature decrease out to radii of @xmath87au followed by a temperature increase at radii @xmath88au .
we note that the remaining six disks from van der plas et al .
( 2015 ) lack @xmath14=2 - 1 measurements due either to poor telluric corrections and low signal - to - noise , or the line widths and disk inclinations put them in the ir regime of our diagram ( @xmath85au ) , where low @xmath14=2 - 1 line fluxes are expected .
.,scaledwidth=50.0% ] analyses by @xcite , @xcite , and @xcite included two possible excitation regimes for co rovibrational emission : in the inner disk , ir pumping due to the local dust continuum radiation , and at larger disk radii , where the dust temperature is too low to generate sufficient amounts of 4.7@xmath0 m photons , pumping by uv photons from the star and/or accretion shock .
the specificity of uv pumping is to populate very high vibrational levels ( @xmath89 ) by decay of an excited electronic state @xcite , while ir pumping drives the co level population to equilibrate to the local color temperature of the dust . at dust temperatures of 500 - 1000k
, this will only populate the @xmath14=1 and @xmath14=2 levels .
indeed , figure [ fig : tempprofs_compar ] can be interpreted as evidence for a transition between the ir and uv pumping regimes near 3au .
the ir regime " , at small disk radii , and the uv regime " , at larger disk radii , are correspondingly marked in the figure .
further evidence is provided by comparison to rotational temperatures t@xmath90 derived in the same disks from previous studies @xcite .
figure [ fig : tempprofs_compar ] shows that @xmath91 in the ir regime , indicating that the co rovibrational population is thermalized to the local dust temperature in the inner disk @xcite .
conversely , in the uv regime , @xmath92 @xcite .
this is consistent with the finding of others , for instance @xcite found @xmath93k and @xmath94k in the disk of hd141569 .
note that , while the vibrational temperatures of the @xmath14=2 - 1 transitions are higher than the rotational temperature in the uv regime , they are typically much lower than the vibrational temperature of even higher - lying levels ( @xmath14=6 - 5 can be as high as 6000k ) .
therefore , while we use the @xmath14=2 - 1 temperature since it is easy to measure in most disks , our results are consistent with previous work citing very high temperatures of high - lying vibrational levels .
+ in figure [ fig : final ] we illustrate how the temperature - radius diagram of rovibrational co emission resembles key aspects of the current theoretical picture of protoplanetary disk dispersal . in the phenomenological class of the double - component disks , the bc
is observed from a hot disk region extending to the magnetospheric accretion radius at @xmath5au @xcite , demonstrating the presence of abundant gas at the smallest radii dynamically permissible for keplerian orbits .
these disks are likely to be in a primordial phase , before giant planets have fully formed ( sketch 1 ) .
dr tau is an example of a primordial disk as based on its co emission .
the next stage in the co sequence leads to the suppression and eventual disappearence of bc ( sketch 2 ) , strongly suggesting the opening of partly devoid inner gaps or an inner region depleted in gas compared to the primordial disks .
in fact , it has been observed by monitoring co spectra of a strong episodic accretor , ex lup , that the bc gets significantly weaker relative to nc after an accretion outburst that depleted the gas mass in the inner disk by an order of magnitude @xcite . in this stage ,
double- and single - component disks overlap in the diagram .
nc is detected from fractions of an au up to @xmath87au .
at similar radii , and even larger , co is observed in single - component disks , where the bc is not detected .
tw hya is an example of this second class of disks , which probably has significant overlap with the `` pre - transitional '' disks as defined from dust emission @xcite .
indeed , for tw hya , an inner gap in the optically thick dust disk has been detected @xcite .
note that a non - detection of the bc does not directly imply the complete absence of dust or gas at @xmath950.1 au .
residual gas and dust in the innermost region , which absorbs most of the direct uv radiation from the star , may likely explain why the temperature profile in some single - component disks continues along an ir - pumped power law .
photo - evaporation probably plays a role in opening disk gaps , and it is interesting to note that uv photo - evaporation is expected to take over at a few au @xcite , close to where the transition between the ir and uv regimes is found in the co diagram .
other clearing mechanisms may dominate the disk dissipation at smaller disk radii , including disk accretion and dust evolution to growing planetesimals and protoplanets ( e.g. * ? ? ?
the transition to a uv - dominated regime of co rovibrational excitation defines a third class of disks , as discussed in section [ sec : disc2 ] . in order for uv radiation to excite the gas to high vibrational temperatures beyond a few au
, the inner disk region must be largely devoid of shielding dust .
we identify these disks as having large inner gaps , with very little amounts of residual material in the inner disk ( sketch 3 ) .
this interpretation is consistent with that proposed by @xcite , who also suggested that uv - pumped co emission may be a distinct signature of devoid inner gaps , and @xcite , who imaged large inner dust gaps in two herbig disks in the uv branch of our diagram , one of which was previously unknown ( in hd97048 ) . in total , large inner gaps in the dusty disks have been detected in up to 6 out of 8 disks in the uv branch so far , and for the remaining two ( hd98922 and hd190073 ) the current lack of resolved imaging may simply be due to their large distance ( @xmath96 pc ) .
potentially , our analysis suggests that all disks that show a single rovibrational co component are developing or have developed gaps in their inner 10 au region , making them prime targets for future imaging detections .
a stellar - mass dependence of @xmath54 may be implied by figure [ fig : final ] , where higher - mass stars tend to have larger @xmath54 relative to lower - mass stars .
co gas from @xmath97au is rarely detected in disks around stars of @xmath98 .
this can not be explained by stellar magnetospheric truncation of the disk , which is expected to happen at radii @xmath99au across the stellar mass spectrum @xcite , and an inner disk radius set by dust sublimation may be marginally consistent only in a few disks , and only at @xmath100au given their bolometric luminosities @xcite .
alternatively , a stellar - mass dependent inner disk lifetime would be consistent with the co temperature - radius diagram , with disks around intermediate - mass stars dispersing faster than those around low- and solar - mass stars ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
that is , we detect fewer broad co components in disks with short dispersal time scales . in this scenario , a mass - dependence in figure [ fig : final ] would be directly linked to the timescale for gas dispersal in inner disks , and therefore to planet - formation processes and timescales . in summary , we suggest that the co temperature - radius diagram shows a sequence of gap opening in protoplanetary disks , where formation of smaller gaps ( interpreted as an early transition , in evolutionary terms ) is marked by the disappearance of the co broad component , while opening of larger gaps ( or a more evolved phase ) is marked by an inversion in the trend of the temperature profile
. the uv pumping regime may represent the very last step before the primordial gas in the disk is dispersed , setting the beginning of the debris disks phase ( see the beta pic point in figure [ fig : final ] , from rovibrational co @xmath14=1 - 0 lines detected by * ? ? ? * ) .
a high - resolution spectrum of co emission at 4.7@xmath0 m may therefore be a powerful discriminator between primordial , pre - transitional and transitional disk stages at radii between 0.110au .
observations of the relative radial profiles of co gas and of dust grains of different sizes in inner disks may provide a powerful tool to distinguish between competing gap formation mechanisms .
the lifetime of the planet - forming region of protoplanetary disks is linked to the observed orbital distribution of exoplanets . in particular , the semi - major axis distribution of giant exoplanets is known to contain structure indicating the presence of multiple populations , including the `` hot jupiters '' , orbiting very close to the star at a small fraction of an au @xcite , and a deficit of giant planets out to 0.5au @xcite .
it is generally accepted that hot jupiters migrated inwards from a natal orbit at a few au , either due to angular momentum exchange with protoplanetary gas @xcite during the lifetime of the disk , or due to a later dynamical process , such as the kozai mechanism @xcite .
( from the database of exoplanets.org ) .
the exoplanet histogram is arbitrarily scaled for comparison to the co histogram .
_ bottom _ : same as plot above , but separated into two stellar mass bins .
here we show the innermost co gas in each disk , therefore excluding the nc in double - component disks.,scaledwidth=50.0% ] to explore this in the context of our observed inner disk radii , we show the distribution of @xmath51 compared to the semi - major axes of known exoplanets with @xmath101 in figure [ fig : corad_compar ] .
the bc radius distribution matches the orbital distribution of hot jupiters ( 0.010.1 au ) , as expected in the type ii migration scenario .
that is , we observe that gas is indeed often available at the right disk radii within 12myr to allow hot jupiters to migrate inwards to the inner gas disk truncation radius . however , hot jupiters probably comprise a small fraction of giant exoplanets @xcite . if the inner disk dissipates rapidly , most planets may not have time to migrate , and the orbital distribution of `` cold jupiters '' will be determined by the inner gas radius of transitional disks , as traced by the narrow co component . indeed ,
figure [ fig : corad_compar ] shows that the nc distribution peaks slightly inward of the bulk of the exo - jupiter distribution ( 110 au ) .
a stellar - mass dependence of the inner disk gas radius may also be reflected in the exoplanet population .
it is known that intermediate - mass stars have very few hot jupiters compared to lower - mass stars @xcite , although a possible overestimation of the intermediate - mass stellar masses is currently under debate @xcite .
@xcite proposed that the paucity of hot jupiters around intermediate - mass stars may be due to a stellar - mass dependent disk lifetime , where inner disks around intermediate - mass stars dissipate before any giant planets have time to migrate to the disk corotation radius . in the lower panel of figure [ fig : corad_compar ]
, we compare the distributions of innermost gas disk radii separated into two stellar mass bins at @xmath102 .
this distribution suggests that the bc component is rarely found in disks around intermediate - mass stars , relative to disks around low- and solar - mass disks .
that is , the paucity of the bc within 0.1au in disks around intermediate - mass stars is reflected by a similar paucity of hot jupiters in the same stellar mass bin .
a likely interpretation of our co observations is that the gaseous disk at 0.01 - 0.5au around stars with @xmath103 indeed dissipates faster than the planet migration time scale .
larger samples of co rovibrational spectra of disks and exoplanets around intermediate - mass stars are needed to confirm this link between the natal disk properties and planetary architectures .
we presented a new analysis of vlt - crires high - resolution ( r@xmath196,000 ) 4.7@xmath0 m spectra of rovibrational co emission from protoplanetary disks .
we applied an empirical decomposition of co line profiles to the sample of co - emitting disks , finding that rovibrational line profiles consist of two distinct velocity components , a broad ( bc ) and a narrow ( nc ) component , primarily in disks around solar - mass stars , and of a single co component in transitional and herbig disks .
when splitting the observed line profiles into two components , we find an empirical relation between the co line width and the disk inclination as expected when the gas is in keplerian rotation in a disk .
this allowed us to estimate the inclination for 15 disks in the sample for which it was previously unknown . using the profile decomposition to separate blended information from different disk radii
, we define an empirical diagram of the keplerian co emitting radius and the vibrational excitation temperature , leading to the definition of a universal excitation temperature profile . between 0.05 and 3au , we find a power law @xmath104 . beyond 3au , the vibrational temperature turns around and begins to increase with radius . we interpret this as the presence of two different excitation regimes , with pumping by infrared photons dominating in the inner disk , and fluorescence excitation by uv photons beyond 3au .
we discuss the co temperature - radius diagram in the context of the theoretical picture of inside - out protoplanetary disk evolution , where the innermost disk is dissipated first .
we find evidence that the co rovibrational line profiles are tracers of the evolutionary stage of inner protoplanetary disks , potentially providing a gaseous equivalent to the traditional definition of classical , pre - transitional , and transitional disks as based on the dust continuum emission .
specifically , we propose that disks exhibiting only a single component in rovibrational co should be in more advanced stages of gap opening and inner disk dissipation than the double - component disks , where the disappearance of bc marks an earlier stage and the uv pumping regime a final stage before the debris disks phase .
rovibrational co emission is sensitive to the innermost location of abundant gas , including for transitional disks , which may have more than one inner region due to the presence of gaps .
consequently , we compared the observed distribution of inner disk molecular gas to the semi - major axis distribution of giant exoplanets .
we find that the distribution of co emitting radii for the broad component matches that of the semi - major axis distribution of hot jupiters .
this is consistent with the majority of hot jupiters migrating inward due to type ii migration , and being halted by the inner truncation of the gaseous disk during an early evolutionary phase .
the radial distribution of the narrow co component peaks at larger radii , close to the `` period deficit '' between the hot jupiter population and the bulk of the giant exoplanet distribution . finally , we find that disks around intermediate - mass stars rarely have gas in the hot jupiter region .
this matches an observed deficit of hot jupiters around intermediate - mass stars , lending support to the idea that their inner disks may evolve too fast for giant planets to migrate inwards to the disk corotation radius .
this analysis ultimately supports the potential of future spatially - resolved imaging of rovibrational co to move significant steps forward in our understanding of disk evolution and planet formation in protoplanetary disks .
herbig disks can already be resolved with adaptive - optics assisted observations , for instance with crires , but disks around solar - mass stars require either spectro - astrometric techniques , infrared interferometric imaging ( e.g. , vlt - matisse ) , or full aperture spectro - imaging at 4.7@xmath0 m with the future generation of extremely large telescopes ( e.g. , e - elt - metis ) .
the authors acknowledge helpful discussions with c. salyk , e. van dishoeck , and g. blake , and constructive comments from an anonymous referee . this research has made use of the exoplanet orbit database and the exoplanet data explorer at exoplanets.org .
acknowledges financial support by a nasa origins of the solar system grant no .
oss 11-oss11 - 0120 , a nasa planetary geology and geophysics program under grant nag 5 - 10201 .
this work is based on observations made with eso telescopes at the paranal observatory under programs 179.c-0151 and 093.c-0432 .
the vibrational ratio @xmath105 is the flux ratio @xmath3 between the @xmath106 lines and the @xmath107 lines with same rotational quantum numbers in the two vibrational levels . in this work ,
@xmath105 values are measured from stacked lines built by using multiple rotational lines and thus provide an average estimate of the vibrational ratio between @xmath108 and @xmath109 .
previously , vibrational ratios of co emission were measured by @xcite for 25 disks that are included in our sample , using the same crires spectra we use here .
they published the median of five values measured in each disk , as obtained from ratios of the corresponding rotational lines in the two vibrational levels .
figure [ fig : vibr_plev ] ( panel @xmath110 ) compares the values from @xcite to those measured in this work .
the values measured in the single - component disks agree to a 1:1 relation , showing that our estimates are consistent .
however , the vibrational ratios in previous work have been underestimated in the double - component disks , because the different contributions from bc and nc in the @xmath14=1 - 0 lines were not separated .
the vibrational ratio @xmath105 depends on the temperature and column density of the gas , as well as on the excitation process . in this work
we use a model of a slab of gas in thermal equilibrium that accounts for the line opacity , where the level populations are determined by the boltzmann distribution ( the model is explicitly described in * ? ? ?
the molecular parameters for co rovibrational lines are taken from the 2012 hitran and the hitemp databases @xcite .
we produced a grid of models to convert the measured @xmath105 into an estimate of the vibrational excitation temperature @xmath71 of the gas .
we used three values for the co column density n@xmath111 to span the range of values proposed by previous studies of nir co emission from protoplanetary disks ( @xmath74@xmath75@xmath73 , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
similarly , for the temperature we use eleven values between 300 and 5000 k. for each set of @xmath71 and @xmath112 a co model spectrum is produced , assuming a line fwhm of @xmath113kms@xmath17 ( the median value for the bc ) and using a pixel sampling equivalent to the crires spectra .
we then re - sample the model spectrum using a typical value for the rms noise as measured in the observed spectra , and then measure vibrational ratios following three methods : 1 ) using directly the individual line fluxes before convolution with the observed fwhm ( the true " value ) , 2 ) taking the median of seven ratios of the line fluxes measured in the convolved spectrum ( equivalent to the method used by * ? ? ?
* ) , and 3 ) using the method adopted in this work .
figure [ fig : vibr_plev ] ( panel @xmath114 ) shows one model as an example , for illustration of the outcomes of this procedure .
it is important to highlight that the true " ratio is not a single value . in the example , the vibrational ratio varies between 0.2 and 0.6 depending on the line used to measure it .
moreover , vibrational ratios measured from the individual line fluxes as observed in the convolved spectrum tend to underestimate the true value .
this is due to blending of the @xmath14=1 - 0 lines with higher vibrational lines , which contribute increasingly to the line flux as @xmath71 , @xmath112 , and the line fwhm increase .
the averaged ratios estimated in this work in general agree well with the range of true values for each model ( panel @xmath115 in figure [ fig : vibr_plev ] ) .
this is true in particular in the case of a co column density of @xmath72@xmath73 , which is a good approximation for most disks in our sample ( e.g. * ? ? ?
* ) . for higher values of @xmath116@xmath73
, the vibrational ratios measured here may overestimate the true values by up to 50% .
we convert the vibrational ratios measured in this work into temperatures using the curves shown in panel @xmath115 of figure [ fig : vibr_plev ] . in this regard , it is not important that our measured @xmath105 may overestimate the true value , as long as we are self - consistent in the conversion , i.e. as long as we use the same method to measure @xmath105 in the data and in the model . in tables
[ tab:2co comp ] and [ tab:1co comp ]
we put the temperature values estimated by using a co column of @xmath72@xmath73 . in the case of as 209 ,
hh 100 , and i m lup , the measured vibrational ratios are too high to be modeled with a co column of @xmath72@xmath73 , which never produces values higher than @xmath10.8 .
therefore , for these disks we use a co column of @xmath75@xmath73 , which can reproduce @xmath105 as high as 1.2 and would be appropriate if the emission is more optically thick ( our analysis indeed finds that the bc in as 209 and i m lup is more optically thick than in other double - component disks , as the bc is detected also in @xmath15co , see figure [ fig : multi - comp analysis ] ) .
for each one of the three values adopted for the co column density , we obtain a different empirical relation between the temperature and the radius of the co emission .
this is shown in panel @xmath117 of figure [ fig : vibr_plev ] , where we plot the linear fits obtained as in figure [ fig : vibr_rin ] .
the index @xmath118 of the temperature profile @xmath119 varies between 0.20 ( for @xmath112 = @xmath75@xmath73 ) to 0.33 ( for @xmath120@xmath73 ) . if we allow a radial gradient in @xmath112 such to have higher values of @xmath121@xmath73 as representative of the more optically - thick nc , and lower values of @xmath122@xmath73 as representative of the bc ( as found by * ? ? ?
* ) , the index may become as steep as @xmath123 .
in figure [ fig : decomp figure ] we show the outcome of the co decomposition procedure as applied to all disk spectra in our sample .
we implement the bayesian method by @xcite in the linear fits to the empirical relations presented in sections [ sec : ana3 ] and [ sec : ana4 ] .
@xcite provided a bayesian method for linear regression that accounts for errors on both the dependent and independent variables as well as an intrinsic scatter due to physical properties that are not explicitly included .
a weighted sum of gaussian distributions is used to obtain an accurate approximation of the true probability distribution of the independent variable . to sample the posterior distribution we use the markov chain monte carlo method from the metropolis - hastings algorithm , which is more efficient than the gibbs sampler in case of small sample sizes .
a thorough discussion of the method and of its comparison to other statistical methods is provided in @xcite .
the method has been made publicly available through the idl routine ` linmix_err.pro ` , that we use in our analysis .
figure [ fig : post_distr ] shows the posterior distributions we obtained for the linear regression parameters presented in section [ sec : ana ] . | the fundamental rovibrational band of co near 4.7@xmath0 m is a sensitive tracer of the presence and location of molecular gas in the planet - forming region of protoplanetary disks at 0.0110au .
we present a new analysis of a high - resolution spectral survey ( r@xmath196,000 , or @xmath2 ) of co rovibrational lines from protoplanetary disks spanning a wide range of stellar masses and of evolutionary properties .
we find that the co emission originates in two distinct velocity components .
line widths of both components correlate strongly with disk inclination , as expected for gas in keplerian rotation . by measuring the line flux ratios between vibrational transitions
@xmath3 , we find that the two velocity components are clearly distinct in excitation .
the broad component ( @xmath4 ) probes the disk region near the magnetospheric accretion radius at @xmath5au , where the gas is hot ( @xmath6k ) .
the narrow component ( @xmath7 ) probes the disk at larger radii of 0.110au , where the gas is typically colder ( 200700k ) .
co excitation temperatures and orbital radii define an empirical temperature - radius relation as a power law with index @xmath8 between 0.053au . the broad co component , co - spatial with the observed orbital distribution of hot jupiters , is rarely detected in transitional and herbig ae disks , providing evidence for an early dissipation of the innermost disk .
an inversion in the temperature profile beyond 3au is interpreted as a tracer of a regime dominated by uv pumping in largely devoid inner disks , and may be a signature of the last stage before the disk enters the gas - poor debris phase .
_ accepted by the astrophysical journal _ |
clusters of galaxies are the largest self - gravitating systems in the universe , and x - ray observations have revealed characteristics of intracluster medium ( icm ) which carries @xmath19% mass of known baryons , such as temperature , icm mass , and metals therein
. however , the icm emission has been detectable only to @xmath20 times the virial radius @xmath21 ( e.g. * ? ? ?
* ) due to limited sensitivities of detectors .
this means @xmath22% of an entire cluster volume has been unexplored in x - rays .
according to the hierarchical structure formation scenario based on cold dark matter ( cdm ) paradigm , mass accretion flows onto clusters , in particular , along filamentary structures , are still on - going .
the icm in the cluster outskirts , therefore , is expected to be sensitive to the structure formation and cluster evolution . however , since the x - ray data around the virial radius of most clusters were unavailable , we have not yet known the icm characteristics in detail , such as density , temperature , pressure and entropy .
the icm around the virial radius is a fascinating frontier for x - ray observations . in the era of _ suzaku _
@xcite , detection of the icm beyond 0.6 @xmath21 has become possible thanks to low and stable particle background of the xis detectors @xcite .
in fact , @xcite measured temperature profile of abell 2044 out to @xmath23 , a radius within which the mean cluster mass density is 200 times the cosmic critical density , and found that the profile is in good agreement with hydrodynamic simulations .
@xcite determined radial profiles of density , temperature , entropy , gas fraction , and mass up to @xmath24 mpc of pks 0745 - 191 , and found that the temperature profile deviates from that expected from the icm in hydrostatic equilibrium with a universal mass density profile as found by navarro , frenk & white ( 1996 , 1997 , hereafter nfw profile ) .
@xcite also detected x - ray emission to @xmath23 and found evidence for departure from hydrostatic equilibrium at radii as small as @xmath25 mpc ( @xmath26 ) .
@xcite obtained metallicity of the icm up to the cluster virial radii for the first time from a link region between the galaxy clusters abell 399 and abell 401 .
a gravitational lensing study is complementary to x - ray measurements , because lensing observables do not require any assumptions on the cluster dynamical states .
the huge cluster mass distorts shapes of background galaxy images due to differential deflection of light paths , in a coherent pattern . as a result , multiple images , arcs and einstein rings caused by the strong gravitational field of a cluster , appear around the cluster center , which is so - called strong gravitational lensing effect .
it is capable of measuring mass distributions of a cluster central region very well . outside the core ,
weak gravitational lensing analysis , which utilizes the coherent distortions in a statistical way , is a powerful tool to constrain the mass distribution out to the virial radius . in particular ,
the _ subaru / suprime - cam _ is the best instrument to conduct the weak lensing analysis ( e.g. * ? ? ?
* ; * ? ? ?
? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , thanks to its high photon - collecting power and high imaging quality , combined with a wide field - of - view .
since the strong lensing analysis is complementary to weak lensing one , a joint analysis enables us to determine accurately the cluster mass profile from the center to the virial radius .
pioneer joint strong and weak lensing studies on abell 1689 @xcite have shown that observed lensing profile is well fitted by an nfw mass profile with a high concentration parameter which is defined as a ratio of the virial radius to the scaled radius .
therefore , one of the best targets for the study of the icm out to cluster outskirts is abell 1689 .
comparisons of x - ray observables with the well - determined lensing mass would allow us to conduct a powerful diagnostic of the icm states , including a stringent test for hydrostatic equilibrium , for the first time .
in addition , two mass measurements using x - ray and lensing analysis are of vital importance to understand the systematic measurement bias and construct well - calibrated cluster mass , which will improve the accuracy of determining cosmological parameters using a mass function of galaxy clusters ( e.g. * ? ? ? * ; * ? ? ? * ; * ? ? ?
* ) . in this paper
, we describe the _ suzaku _ observations and data reduction in [ section : obs ] , and x - ray background estimation in [ section : xbkg ] . in
[ section : result ] , we show results of spatial and spectral analyses and obtain radial profiles of temperature , electron density , and entropy . in [ section : dis ] , we discuss the hydrostatic equilibrium around the virial radius , and compare x - ray observables , such as hydrostatic mass , pressure and entropy , with lensing mass derived from a joint strong and weak lensing analysis .
we also discuss the icm characteristics in the outskirts in light of its cosmological environment , such as the large - scale structure and low - density voids .
we summarize the conclusions in [ section : summary ] . in the present work ,
unless otherwise stated , we use @xmath27 ( @xmath28 ) , assuming a flat universe with @xmath29 .
this gives physical scale @xmath30 kpc at the cluster redshift @xmath31 @xcite .
the definition of solar abundance is taken from anders and grevesse ( 1988 ) .
errors are given at the 90 % confidence level .
as shown in figure [ fig : mosaic ] , we performed four - pointing _ suzaku _ observations of abell 1689 , named offset1 ( northeast direction ) , offset2 ( southeast ) , offset3 ( southwest ) , and offset4 ( northwest ) , in the end of july 2008 with exposure of @xmath32 ks for each pointing .
the pointings were coordinated so that the x - ray emission centroid of abell 1689 was located at one corner of each pointing and the entire field - of - view ( fov ) of the xis covered the icm emission up to the virial radius ( @xmath33 mpc , * ? ? ? * ) .
the observation log is summarized in table [ table : observation ] .
all the xis sensors ( xis0 , xis1 and xis3 ) were in normal clocking mode without window or burst options throughout the observations .
in the present paper , we used the xis data created through version 2.2 pipe - line data processing , which is available from the _ suzaku _ ftp site .
we re - processed unscreened xis event files using _ suzaku _ software version 11 in heasoft 6.6 , together with the calibration database ( caldb ) released on 2009 january 9 .
we applied ` xiscoord ` , ` xisputpixelquality ` , ` xispi ` , and ` xistime ` in this order , and performed the standard event screenings .
bad pixels were rejected with ` cleansis ` , employing the option of chipcol = segment .
we selected events with grade 0,2,3,4 , and 6 .
good time intervals ( gti ) were determined by criteria as ` " saa_hxd==0 ` ` & & ` ` t_saa_hxd``>436 & & cor>6 & & elv>5 ` ` & & dye_elv>20 & & aocu_hk_cnt3_nml_p==1 & & ` ` sn_dtrate<3 ` ` & & ang_dist<1.5 " ` .
detailed procedures of the data processing and screening are the same as those described in _ the suzaku data reduction guide_. redistribution matrix files ( rmfs ) of the xis were produced by ` xisrmfgen ` , and auxillary response files ( arfs ) by ` xissimarfgen ` @xcite . as an input to the arf generator
, we prepared an x - ray surface brightness profile of abell 1689 using a double-@xmath34 model ( sum of two @xmath34 models , * ? ? ?
the parameters of the model were determined through a least chi - square fit to a background - subtracted and vignetting - corrected 0.510 kev surface brightness profile , created from the _ xmm - newton _ mos1 image of abell 1689 .
point sources were detected and removed from the image , using ` ewavelet ` task of the sas software version 8.0.0 with a detection threshold set at @xmath35 .
as the background , we used a public mos1 blank - sky dataset which is available from the _ xmm - newton _ website . among the blank - sky files
, we chose data with hydrogen column densities between @xmath36 @xmath37 and @xmath38 @xmath37 , to match the hydrogen column density of abell 1689 , @xmath39 @xmath37 ( * ? ? ?
* weighted average over @xmath40 cone radius ) .
the total exposure of the selected mos1 blank - sky data is 125 ks .
the resulting parameters of double-@xmath34 model are ( @xmath41 , @xmath34)@xmath42(@xmath43 kpc , @xmath44 ) for the narrower component and ( @xmath45 kpc , @xmath46 ) for the wider component , where @xmath41 is the core radius . absorption below 2 kev , caused by a carbon - dominated contamination material on the xis optical blocking filters , is included in the arfs . in the calculation of contamination by ` xissimarfgen ` , differences of the contaminant thickness among the xis sensors are taken into account , along with its radial dependences and secular changes @xcite . non x - ray background ( nxb ) of each xis sensor was created using ` xisnxbgen ` , which sorts spectra of night - earth observations according to the geomagnetic cut - off - rigidity ( cor ) and makes an averaged spectrum weighted by residence times for which suzaku stayed in each cor interval @xcite .
integrated period for the nxb creation is @xmath47 days of the abell 1689 observations .
we assumed systematic errors ( 90% confidence limit ) of 6.0% and 12.5% for the nxbs @xcite from xis - fi ( front illuminated ccd , see * ? ? ?
* xis0 and xis3 in this paper ) and xis - bi ( backside illuminated ccd , see * ? ? ?
* xis1 in this paper ) , respectively , and added them in quadrature to the corresponding statistical errors . in the next section
, we estimate the x - ray background ( xrb ) around abell 1689 which is crucial to detect weak signals from the icm around the virial radius .
in order to estimate the x - ray background ( xrb ) , we selected the nearest two blank - sky _ suzaku _ observations to abell 1689 ; one is q1334 - 0033 ( observation i d = 702067010 , hereafter q1334 ) which is @xmath48 offset from abell 1689 , and the other is ngc4636_galactic_1 ( observation i d = 802039010 , hereafter n4636_gal ) which is @xmath49 offset from abell 1689 .
the exposures are 12 ks and 35 ks for q1334 and ngc4636_gal , respectively , after the event processing and screening which are the same as those described in [ subsection : obs : reduction ] . the hydrogen column densities ( * ? ? ?
* weighted average over @xmath40 cone radius ) of abell 1689 , q1334 , and n4636_gal , are @xmath39 @xmath37 , @xmath50 @xmath37 , and @xmath51 @xmath37 , respectively .
they are in good agreement with one another within 11% , suggesting the same level of x - ray background in soft x - ray band .
next we checked the count rates obtained from rosat all sky survey ( rass ) which are available via the nasa s heasarc website .
since the rass count rates in the direction of abell 1689 is contaminated by its own icm emission , we took an annular region around abell 1689 of which inner and outer radii are @xmath52 and @xmath53 , respectively .
the rass count rates of the two blank - sky fields were defined as circular regions of @xmath40 radius . as shown in table [ table : rass - cnt ] , the count rates of q1334 and n4636_gal match those of the annular abell 1689 field within errors in all the energy bands , except for count rates of 1/4 kev and 3/4 kev bands in the n4636_gal field , which are significantly higher than the corresponding rates in the abell 1689 field .
therefore , the cosmic x - ray background ( cxb ) of abell 1689 is considered to be represented well by both q1334 and n4636_gal observations , while galactic foreground emission ( gfe ) should be represented only by the q1334 observation .
thus , in the following spectral analyses , we ignored spectra of n4636_gal below @xmath54 kev . using xspec12 version 12.5.0ac , we fitted nxb - subtracted xis - fi ( averaged over xis0 and xis3 ) and xis - bi spectra of q1334 and n4636_gal simultaneously with a model which represents the cxb and gfe emissions .
rmfs and arfs were created for each observations by ` xisrmfgen ` and ` xissimarfgen ` , respectively .
the input image for the arfs is uniform emission over a circular region of @xmath55 radius . as the cxb model
, we used a power law with a fixed photon index of 1.41 @xcite and a free normalization . to represent the gfe , we employed a model of @xcite , which consists of three _ apec _
@xcite components ; a non - absorbed 0.08 kev component , an absorbed 0.11 kev one , and an absorbed 0.27 kev one , representing the emission from the local bubble ( lb ) , cool halo , and hot halo , respectively .
the three _ apec _ temperatures were all fixed .
the hydrogen column density was fixed to the galatic values of @xmath50 @xmath37 and @xmath51 @xmath37 for q1334 and n4636_gal , respectively .
the metal abundance and redshift were also fixed to one solar and zero , respectively .
relative normalizations of the two halo components were tied as cool halo : hot halo @xmath56 @xcite .
thus , the gfe model has only two free parameters ; normalizations of the lb and halo .
as shown in figure [ fig : bgd_fit ] , this cxb+gfe model reproduces the blank - sky spectra successfully .
the fitting result is summarized in table [ table : fit - bgd ] .
the 2.010.0 kev surface brightness of the cxb component is @xmath57 erg s@xmath58 @xmath37 sr@xmath58 ( 90% statistical and systematic errors ) , where the systematic error , 10% , refers to appendix 1 of @xcite .
this is consistent with the average asca result of absolute cxb intensity @xcite , @xmath59 erg s@xmath58 @xmath37 sr@xmath58 ( 90% statistical and systematic errors ) .
in order to see how far the icm emission of abell 1689 is detected , we made a surface brightness profile from the 0.510 kev xis - fi ( xis0 @xmath60 xis3 ) mosaic image ( figure [ fig : mosaic ] ) .
the center was chosen to be ( @xmath61 , @xmath62)=(@xmath63 ) in j2000 coordinates as a position of x - ray emission centroid in a 0.510 kev _ xmm - newton _ mos1 image .
this position is close to the brightest - cluster galaxy , maxbcg j197.87292 - 01.34110 bcg , @xmath64 ( @xmath65 kpc ) offset from the center .
we identified point sources from the mos1 image using ` ewavelet ` task of the sas software version 8.0.0 with a detection threshold set at 7@xmath66 .
as to point sources outside the mos1 image , we selected visible sources in a full band image of rosat pspc of which flux limit corresponds to @xmath67 erg s@xmath58 in 2.010.0 kev .
as shown in figure [ fig : mosaic ] ( right ) , circular regions ( @xmath68 in radius ) around 49 point sources were masked in the xis - fi mosaic image ( their positions and flux are summarized in table [ table : pntsrc ] in appendix b ) . in figure
[ fig : prof_all ] , we show 0.52 kev and 210 kev raw ( background inclusive , but point - source excluded ) surface brightness profiles as black crosses .
we created corresponding 0.52 kev and 210 kev nxb profiles from an nxb mosaic image with ` xisnxbgen ` .
the 0.52 kev and 210 kev nxb profiles are shown in magenta in figure [ fig : prof_all ] .
then , we obtained an xrb mosaic image .
based on the cxb and gfe models of abell 1689 described in
[ subsection : bgd : ana ] , we simulated xrb ( cxb+gfe ) images of the four offset observations using ` xissim ` with exposures 10 times longer than those of actual observations . the 0.52 kev and 210 kev xrb profiles are shown in green in figure [ fig : prof_all ] . the cxb level of abell 1689 is not the same as that of the blank - sky fields because point sources are excluded in abell 1689 .
therefore , intensity of the cxb model when the flux is integrated to our flux limit , @xmath69 erg s@xmath58 in 2.010.0 kev , was calculated from equation 6 of @xcite , and used in the simulation .
the resulting cxb intensity of abell 1689 is 65% of the absolute cxb intensity .
finally , we obtained a background - subtracted profile ( black @xmath70 magenta @xmath70 green in figure [ fig : prof_all ] ) , as shown in red in figure [ fig : prof_all ] . in the error bars in this background - subtracted profile ( red crosses ) , 1@xmath66 uncertainties of the xrb and nxb are added in quadrature to the corresponding statistical 1@xmath66 errors .
we adopted 3.6% as the @xmath71 uncertainty of nxb @xcite , while that of xrb was calculated from @xmath71 statistical errors of normalizations in the cxb and gfe models , determined from the spectral fitting of blank sky observations ( [ subsection : bgd : ana ] ) .
these are 2.1% and 15.8% for the cxb and the gfe , respectively , corresponding to 8.5% of 0.510 kev xrb flux . as a result ,
the background - subtracted profiles both in 0.52 kev and 210 kev bands extend up to the virial radius ( @xmath33 mpc ) .
however , as we see in
[ subsection : ana : syserr ] , more careful treatment of uncertainties related to a rather broad _ suzaku _ point spread function ( psf ) , @xmath72 in half power diameter ( hereafter hpd ) , is needed to evaluate the significance of the signal around the virial radius .
the signals around the virial radius found in the x - ray surface brightness profile ( red crosses in figure [ fig : prof_all ] ) have to be tested for systematic uncertainties caused by ( 1 ) contaminated signals from inner icm emission , and ( 2 ) residual signals from the excluded point sources . as for the item ( 1 ) , significant amount of signals from inner regions of the icm is expected to contaminate the signal around the virial radius , since the psf of _ suzaku _ is @xmath72 in hpd @xcite . however , as described in detail in appendix a , the simulated double - beta profile determined from the 0.510 kev _ xmm - newton _ mos1 image ( [ subsection : obs : reduction ] ) is consistent with the observed profile , and the contaminated signals which originate inside @xmath73 contributes only 22% of signals outside @xmath73 .
therefore , the signal observed around the virial radius can not be explained by the contaminated signals from the inner icm emission .
the item ( 2 ) is also due to the @xmath74 hpd of _
suzaku_. although we excluded circular regions around the point sources , half of signals from point sources is expected to escape from the excluded circular regions of @xmath68 radius .
as shown in figure [ fig : pntsrc_prof ] , we simulated this residual signals from point sources .
details of the simulation are described in appendix b. in 0.52 kev band , the contribution of the point sources in the @xmath75@xmath76 region is only 26% of the observed profile . meanwhile
, the observed 210 kev profile can be explained well by the point sources . by combining the contributions of ( 1 ) and ( 2 ) ,
we conclude that in 0.52 kev band , these effects explain at most 48% of the signal around the virial radius ( @xmath75@xmath76 ) , and the detection is significant at 4.0@xmath66 level . on the other hand , in 210 kev range ,
the signal is not significant .
this indicates that the detected icm temperature around the virial radius might be low , compared with the central temperature of @xmath77 kev @xcite . in table
[ table : significance ] , we show signal significances when x - ray surface brightness profiles are evaluated separately for the four pointing directions . in the offset1 direction ( northeast ) , the signal in 210 kev band is also detected with @xmath78 level , suggesting harder spectrum in this direction .
since we surely detected the icm emission out to the virial radius from the analysis of x - ray surface brightness , we proceeded to spectral analysis .
for each pointing ( offset1 through offset4 ) , we divided the xis image ( figure [ fig : mosaic ] ) to five concentric annular regions centered on the x - ray emission centroid ( [ subsection : ana : sb ] ) ; their inner and outer radii are @xmath79@xmath80 , @xmath80@xmath81 , @xmath81@xmath82 , @xmath82@xmath75 , and @xmath75@xmath76 .
we have four azimuthal regions at the same distance from the center .
the 49 point sources were masked by circles of @xmath68 radius as we did in obtaining the x - ray surface brightness profile ( [ subsection : ana : sb ] ) . after extracting spectra of the xis0 , xis1 , and xis3 from each region
, we averaged xis - fi ( xis0 and xis3 ) spectra . in the @xmath79@xmath80 and @xmath80@xmath81 regions ,
all spectra of offset1 , an xis - fi spectrum of offset2 , and an xis - bi spectrum of offset4 are unavailable because regions irradiated by the calibration source are removed from the analysis .
as shown in figure [ fig : spec ] , we used xspec12 version 12.5.0ac and included the xis - fi and xis - bi spectra from the regions at the same distance , and corresponding nxb and response ( rmf and arf ) files created through the procedure described in [ subsection : obs : reduction ] . in @xmath75@xmath76 analysis , for example , we included four xis - fi source spectra and four xis - bi source spectra . in order to determine the xrb
, we also included the spectra of blank - sky fields ( figure [ fig : bgd_fit ] ) with their nxb and response files ( [ subsection : bgd : ana ] ) .
then , the nxb - subtracted source spectra were fitted with an absorbed thin - thermal emission model represented by _ phabs _
@xmath83 _ apec _ , added to the xrb model ( [ subsection : bgd : ana ] ) , while the blank - sky spectra were fitted with the xrb model simultaneously with the source spectra .
relative cxb normalization of abell 1689 to the blank - fields was fixed to 65% ( [ subsection : ana : sb ] ) .
free parameters in the _ apec _ model are temperature , metal abundance , and normalization .
the hydrogen column density was fixed to the galactic value of @xmath39 @xmath37 and redshift was also fixed to 0.1832 .
these free parameters for different pointings ( offset1 through offset4 ) were determined separately when we saw azimuthal differences , or tied when we obtained azimuthal averages .
relative normalization between the xis - fi and xis - bi was left free and its deviations from unity were within @xmath84% in all the regions .
the fitting result is summarized in table [ table : fit - result ] .
although the x - ray background model is determined separately for the individual annular regions , the normalizations of cxb , lb , and halo are consistent among the regions within errors . in this spectral analysis , we did not explicitly include the two uncertainties described in
[ subsection : ana : syserr ] , namely , contaminated signals from inner icm emission and residual signals from the excluded point sources .
we estimate their effect on the temperature and electron number density of abell 1689 in the following section , obtaining radial profiles of these icm parameters .
the top panel of figure [ fig : prof ] shows radial profiles of gas temperature , obtained from the spectral analysis ( [ subsection : ana : spectral ] ) . in figure
[ fig : prof ] ( and following figures of any radial profiles ) , the center of each radial bin is defined as emission - weighted center derived from the x - ray surface brightness obtained in [ subsection : ana : sb ] ; @xmath85 , @xmath86 , @xmath87 , @xmath88 , and @xmath89 for the five annular regions .
although the temperature profile is a projected one obtained by the two - dimensional spectral analysis , it coincides well with a three - dimensional temperature profile obtained with _
data @xcite in the overlapping region .
the temperature is @xmath90 kev within @xmath81 ( @xmath91 kpc ) , and it gradually decreases toward the outskirts down to @xmath92 kev around the virial radius ( @xmath93 mpc ) .
the decrease of temperature is independent of the azimuthal directions up to @xmath94 , at which the mean interior density is @xmath95 times the critical mass density of the universe .
note that the overdensity radius @xmath14 is determined by lensing analysis , as is the case for @xmath21 ( see [ sec : joint ] and eq .
( [ eq : viral ] ) ) .
the logarithmic slopes of temperature between the third and forth bins , corresponding to @xmath7 are @xmath96 for all , offset1 and offset234 directions , respectively , which are consistent within errors with each other .
the temperature variations in direction becomes significant in the outskirts of @xmath3 : the temperatures for offset1 ( @xmath4kev ) and offset234 ( @xmath97kev ) are about half and one - sixth of a mean temperature @xmath98 kev , which is an emission - weighted temperature of @xmath80@xmath81 and @xmath81@xmath82 regions , respectively .
this temperature variations can be seen directly in spectra . in figure
[ fig : spec - compbgd ] , we show the spectra of offset1 and offset2 in @xmath75@xmath76 region . the signal of offset1 is clearly detected from both xis - fi and xis - bi in @xmath99@xmath100 kev range , whereas the signal of offset2 is comparable to the background in the same energy range .
this means that the xis spectra of offset1 is harder than those of offset2 , resulting in the higher temperature in the offset1 direction .
the logarithmic slopes of temperature in the outskirts corresponding to @xmath3 are @xmath101 for all , offset1 and offset234 directions , respectively .
the slope in offset234 and all directions are surprisingly steep . contaminated signals from the inner icm emission due to the _ suzaku _ psf may affect the temperature determination .
however , in pks 0745 - 191 case @xcite , the effect is at most @xmath102 kev even in a central region where @xmath103% of photons come from the other regions . in the other regions where photon mixing is @xmath104% , the temperature uncertainties are @xmath105 kev .
therefore , in our case of abell 1689 , we expect that the psf effect is at most @xmath106 kev in the cluster outskirts outside @xmath107 , which is much smaller than the statistical errors
. the effect inside @xmath107 would be also negligible because our temperature profile is consistent with the _ chandra _ result @xcite as described above . in order to see the other effect , the residual signals from the excluded point sources in @xmath75@xmath76 , we added a power - law of photon index 1.4 into the fit with its normalization fixed to the summed escaped flux of the 49 excluded point sources .
this corresponds to increasing the cxb level by @xmath108 . as a result ,
the effect was as small as @xmath109 kev in all the azimuthal directions , which is also smaller than the statistical errors .
we compared our temperature profiles with that based on the scaled temperatures derived from a sample of 15 clusters by @xcite .
we used the mean temperature @xmath98 kev to calculate the scaled profile .
the temperature profile in the offset1 direction is in good agreement with the scaled temperature one ( dashed line in figure [ fig : prof](a ) ) , while the temperatures of the other regions ( offset2 - offset4 ) around the virial radius are lower than the scaled value by a factor of @xmath92 .
the electron number density profile was calculated from the normalization parameter of _ apec _ model , defined as @xmath110 where @xmath111 is the angular size distance to the source in units of cm , @xmath112 is the electron density in units of @xmath113 , and @xmath114 is the hydrogen density in units of @xmath113 .
since metal abundance was poorly constrained from our fitting result ( table [ table : fit - result ] ) , we used @xmath115 assuming metal abundance to be 0.3 solar .
we calculated a geometrical volume which contributes to each two - dimensional region assuming that abell 1689 is spherically symmetric , and obtained projected volume - averaged density for the region .
then , the projected density profile was deprojected by the following equation @xmath116 where @xmath117 is the deprojected electron density , @xmath118 is the projected electron density , and @xmath119 is the volume contributing to @xmath120-th annular region .
we assumed that no emission exists outside the @xmath75@xmath76 region and hence @xmath121 in the @xmath75@xmath76 region .
the result is summarized in table [ table : density ] .
the electron number density profile ( middle panel of figure [ fig : prof ] ) decreases down to @xmath122 around the virial radius .
the density profiles are in good agreement with a generalized @xmath34 model @xcite obtained by fitting _
data within @xmath123 ( * ? ? ?
* model 3 in table 9 ) .
in contrast to the temperature profile , the electron densities of the four offset directions are consistent with one another within errors in any circular annulus .
the slope of the number density is @xmath6 in the range of @xmath7 .
our result coincides with previous studies around @xmath14 : @xmath124 for the reflex - dxl clusters @xcite , @xmath125 for the locuss sample @xcite , @xmath126 for the rexcess sample @xcite , and @xmath127 for 13 clusters observed with _
a consistent result , @xmath128 , has been also obtained with _
suzaku _ in abell 1795 around @xmath14 @xcite . in the outer annulus of @xmath3 ,
the density slope becomes flatter to @xmath129 .
the same tendency of flattening is seen in a electron density profile of pks 0745 - 191 around @xmath130 @xcite .
excess signals in a x - ray surface brightness profile above a @xmath34 model outside @xmath131 seen in abell 1795 @xcite also indicates the flattening of slope in the cluster outskirts .
the electron density profile outside @xmath132 is significantly shallower than asymptotic matter density slope @xmath9 of an nfw profile @xcite , defined by @xmath133 where @xmath134 is the central density parameter and @xmath135 is the scale radius . since the electron density profile is consistent with the _ chandra _ result within @xmath123 @xcite , we can say that the systematic effects of the _ suzaku _ psf are safely within the statistical errors in this region . in our @xmath75@xmath76
, however , 48% of the signal can leak from the other regions or point sources as discussed in
[ subsection : ana : syserr ] .
this fraction corresponds to 28% in terms of the electon density , which is comparable to the statistical error in this region ( see table [ table : density ] ) . the steepest slope including this systematic effect is @xmath136 , and is still much shallower than the nfw profile .
@xcite analytically constructed a physical model of the icm in hydrostatic equilibrium with the dark matter potential , in the context of dark - matter jeans equilibria @xcite constrained by the _ dark matter entropy _ in a single power - law form @xmath137 , as motivated by n - body simulations of cold dark matter halos @xcite .
their _ equilibrium _ model predicts a shallow gas density slope of ( @xmath138 ) at the entropy slope of @xmath139 in a slow and constant accretion phase , in good agreement with our observed gas density slope of @xmath6 in the range of @xmath7 . beyond @xmath14 of abell 1689
, we calculated the ratio of the dark matter circular velocity to the gas sound velocity from the observed gas density and entropy slopes , by following their model ( equation 7 in * ? ? ? * ) .
the resulting ratio is @xmath140 , which conflicts with our observations that the dark matter velocity is much higher than the sound velocity . as we discuss in
[ section : dis ] , the icm in most regions at the outermost annulus is likely _ not _ to be in hydrostatic equilibrium .
therefore , some modifications to the model would be needed to describe our results outside @xmath14 .
the bottom panel in figure [ fig : prof ] shows the entropy profile calculated as @xmath141 using the temperature and electron density profiles obtained above .
the entropies in all directions increase from the center to the radius of @xmath75 .
the entropy distribution in the last radial bin is anisotropic , similarly to the temperature profiles .
the radial dependence of the offset1 entropy obeys roughly @xmath142 ( dashed line in figure [ fig : prof](c ) ) , which is predicted by models of accretion shock heating @xcite , while the entropies of the other regions ( offset2-offset4 ) in the last radial bin are lower than those in their @xmath82@xmath75 annuli .
in this section we discuss the distributions of gas temperature and entropy shown in figure [ fig : prof ] , particularly focusing on their anisotropic distributions in the last bin ( @xmath3 ) . in the framework of the hierarchical clustering scenario based on cdm paradigm , mass aggregation processes onto clusters , in particular along the large - scale filamentary structures , are still on - going .
continuous mass accretion flows along filamentary structures , within which clusters are embedded , sometimes trigger _ internal shocks _ in gas within clusters .
it converts most of the tremendous amounts of kinetic energy into the thermal energy to heat the icm .
therefore , the accretion flows are considered to play an important role in cluster evolution as well as in the icm thermodynamics .
if the shock heating dominates in the cluster outskirts , the entropy generated in the shock process becomes higher than those in the pre - shock regions . on the other hand , if the gas is adiabatically compressed during the infall with a sub - sonic velocity , the gas temperature increases but the entropy stays constant . according to analytical models and simulations considering both the continuous accretion shock and adiabatic compression processes ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , the entropy profile is predicted to increase with @xmath143 by the accretion shock , as long as the initial entropy is low , assuming that the icm is in hydrostatic equilibrium and the kinetic energy of the infalling gas is instantly thermalized via shocks .
this radial dependence has been generally confirmed in observed profiles ( e.g. * ? ? ?
* ; * ? ? ?
* ) within @xmath14 .
interestingly , our entropy profile in the offset1 direction coincides very well with the model prediction up to the virial radius .
it indicates a possibility that the icm in the offset1 @xmath144 region is thermalized and close to the hydrostatic equilibrium .
indeed , the outskirts temperature in the offset1 direction agrees with the predictions of the analytical models ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) in which , under the hydrostatic equilibrium assumption , the temperature around the virial radius declines at most @xmath145 percent of the inner values .
contrary to the offset1 , the sharp decline in both temperature and entropy in the offset2-offset4 regions indicates that the thermal pressure is not sufficient to balance the total gravity of the cluster .
if the gas is not completely thermalized , the gas temperature should be lower than the value needed to maintain hydrostatic equilibrium . in this case , the accreted gas retains some fraction of the total energy in bulk motion , and the bulk pressure partly supports the gravity .
in fact , recent numerical simulations ( e.g. * ? ? ?
* ) have shown that the kinetic energy of bulk motion carries @xmath146 of the total energy around the virial radius .
in addition to the bulk pressure , pressure of turbulent motions in the icm may also contribute to balance the total gravity ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in order to study the validity for the assumption of hydrostatic equilibrium , we measured the hydrostatic equilibrium masses ( h.e .
mass ) for the all , offset1 and offset234 regions .
we here assume that the mass distribution is spherically symmetric and the icm is in hydrostatic equilibrium at any radii .
the hydrostatic mass , @xmath147 , is calculated from the parametric density and temperature profiles obtained by fitting data points in each azimuthal region ( all , offset1 , & offset234 ) with models ( appendix [ appen : n+tpara ] and table [ table : nfit ] ) , as follows @xmath148 where @xmath149 is the gas pressure and @xmath150 is the gas mass density .
the errors ( 68% cl uncertainty ) are calculated with monte carlo simulations because model parameters are correlated with each other .
the resulting mass is shown in figure [ fig : mhe ] .
the cumulative mass in the offset1 direction increases with radius , while the ones in all and offset234 regions _ unphysically _ decrease outside @xmath151@xmath152 . this _
unphysical _ decrease in the cumulative mass is due to the sharp temperature drops in the offset234 region . here
, in the parametric density model ( equation [ eq : ne_pr ] ) , we adopted a modified @xmath34 model to reproduce the density flatness in outskirts ( @xmath153 ) .
the ratio of core radii @xmath154 in equation [ eq : ne_pr ] was fixed to be 3 , since the flattening factor in which @xmath155 is included was not constraind well owing to few data points . however , even by choosing the ratio @xmath156 , the radii , at which the masses have maximum values , are changed only by @xmath157 for offset234 and @xmath158 for all regions , respectivly .
this is because the curvature of the parametric density profile is insensitive to a choice of the ratio of core radii .
thus , the hydrostatic equilibrium we assumed is inadequate to describe the icm in the outskirts except the offset1 direction , that is , most of the icm in the cluster outskirts is far from hydrostatic equilibrium .
one alternative proposal to the bulk motion or turbulent motion as a cause of the non - hydrostatic equilibrium would be convective instability in the cluster outskirts .
we investigated the possibility of convective instability in the radial direction .
the convective motion in offset234 is unstable outside @xmath159 but the time scale of the growing mode is comparable to the age of the universe at cluster redshift @xmath160 . therefore , the convective instability is negligible for the _ unphysical _ decrease in the cumulative mass .
another possible proposal is that ions have a higher temperature than that of electrons @xcite , and that the thermal pressure of ions supports the cluster mass .
we computed the thermal equilibration time between electrons and ions through the coulomb interaction .
in the cluster outskirts of abell 1689 , the time scale is given by @xmath161 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
. the coulomb interaction would provide us with the maximum time scale of thermal equilibrium , because there might be other processes , such as plasma instability , to facilitate the interaction between electrons and ions .
since the resulting time is much shorter than the dynamical time scale , it is difficult to explain the _ unphysical _ decrease .
therefore , the bulk and/or turbulent motions would be the main source(s ) of additional pressure needed in the offset234 directions . in
[ subsection : pwl ] , we shall discuss on this point again based on a joint x - ray and lensing analysis .
what makes the anisotropic distributions of gas temperature and entropy in the outskirts ?
the internal shocks associated with mass accretion flows would heat the icm .
numerical simulations ( e.g. * ? ? ?
* ; * ? ? ?
* ) have shown that , the low - density gas in low - density regions , so - called voids , accretes onto the vicinity of clusters by the gravitational pull , with an order of thousands of kilometers per second . as a result , the internal shocks form outside virial radius at which gas density sharply changes .
it is also referred to as _ virial shocks _ @xcite .
meanwhile , the accretion flow along filamentary structures forms internal shocks inside the virial radius .
these shocks migrate relatively deep into a cluster potential well , because a large amount of matter , such as gas , galaxies and groups , accretes from filaments @xcite .
therefore , the thermalization in the outskirts along the direction of filamentary structure takes place faster .
based on this accretion shock scenario , the observed anisotropic distributions of gas temperature and entropy in the outskirts inside virial radius would be associated with large - scale structures . another possibility for
the anisotropy is that infalling galaxies , especially massive galaxies , might be reservoirs of cold gas .
the cold and dense gas , which is bound by the dark matter potential of galaxies , would decrease the emission - weighted temperature in the outskirts .
although we excluded point sources in the spectral study , we can not rule out this possibility only from the x - ray results .
if this is the case , member galaxies around the virial radius ( @xmath75@xmath76 ) in the offset234 directions would be more numerous than those in the offset1 direction . in order to search the large - scale structures or the anisotropic galaxy distribution around the virial radius , we made maps of galaxy number density using the _ sdss _ dr7 catalogue ( abazajian et al .
2009 ) from _ sdss _ casjobs site .
we first looked into bright red - sequence galaxies which are expected to host cold gas .
we selected red - sequence galaxies by criteria of @xmath162 and @xmath163 , where we used _
psfmag _ for magnitude and _ modelcolor _ for color .
we made projected galaxy distribution , smoothed with a gaussian kernel , @xmath164 , of fwhm@xmath165 .
the resulting map is shown in the left panel of figure [ fig : mapsdss ] .
we also looked into the same map derived from _
subaru _ data .
it is consistent with the _ sdss _ map , although a part of the outer region is not available in the _ subaru _ data due to the limitation of its field - of - view . in figure [ fig : mapsdss ] ( left ) , the central part of density map is clearly elongated in the north - south direction .
@xcite have found the same elongation in a two - dimensional mass distribution .
however , we could find neither an apparent feature nor a significant difference between the projected distributions in the offset1 and the other directions in the cluster outskirts of @xmath166 .
therefore , the temperature and entropy anisotropies in the outskirts are not due to the cold gas in galaxies .
next , we made a larger map to investigate the presence of large - scale structures associated with abell 1689 .
@xcite has shown that the maximum line - of - sight velocity of infall galaxies is @xmath167 .
therefore , taking into account the uncertainty of redshift due to line - of - sight velocities , we selected galaxies in the range of @xmath168 , where @xmath169 is the cluster redshift , @xmath170 is a photometric redshift , @xmath171 , and @xmath172 is the light velocity .
the mean photometric redshift for our sample is @xmath173 , where the first error is the statistical error and the second one is the systematic error due to photometric uncertainty of each galaxy .
our sample of galaxies thus statistically represents a slice around abell 1689 in redshift space .
the resulting map of galaxies smoothed with a gaussian kernel of fwhm=200 is shown in the right panel of figure [ fig : mapsdss ] . in the offset1 direction , a filamentary overdensity region outside the virial radius
is found , where the galaxy number density is more than 1.5 times higher than those in the other directions . in the offset4 direction outside the fov of xis
, a narrow sheet of galaxies is also found , which is connected to a northwest overdensity region .
we note that the projected elongated direction of cluster mass and galaxies ( left panel of figure [ fig : mapsdss ] ) does not coincide with the filamentary direction ( right panel ) , suggesting that a mass structure seen in the central region of a cluster is not necessarily connected directly to a filament .
our result does not change even when we choose twice or half @xmath174 in the galaxy selection .
we also tried to investigate the line - of - sight filamentary structure by a monte -carlo simulation computing the redshift distribution of galaxies with photometric errors .
however , we could not obtain reliable results within a redshift resolution smaller than @xmath175 .
we could not see the line - of - sight structure from the spectral data of _ sdss _ either , because available number of galaxies is too small .
the high temperature and entropy region in the outskirts is clearly correlated with the galaxy overdensity region associated with the large - scale structure outside abell 1689 , as shown in figure [ fig : map_kt_sdss ] .
this indicates that the icm in the outskirts is significantly affected by surrounding environments of galaxy clusters , such as the filamentary structures and the low density void regions .
the large - scale structure would play an important role in the thermalization process of the icm in the outskirts .
in particular , our result suggests that the thermalization in the outskirts along the filamentary structure takes place faster than that in the void region . in this subsection
, we carry out a joint x - ray and lensing analysis , incorporating _
subaru / suprime - cam _ and _ hst / acs _ data .
abell 1689 has been the focus of intensive lensing studies in recent years ( e.g , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
it has been shown by @xcite and @xcite that joint lensing profiles of abell 1689 obtained from their acs and subaru data with sufficient quality are consistent with a continuously steepening density profile over a wide range of radii , @xmath176kpc@xmath177 , well described by the general nfw profile @xcite , whereas the singular isothermal sphere model is strongly disfavored .
they also revealed that the concentration parameter of the nfw profile , namely the ratio of the virial radius to the scale radius , @xmath178 , is much higher than predicted by cosmological @xmath179-body simulations ( e.g. , * ? ? ?
* ; * ? ? ?
we note that the virial radius , @xmath21 , within which the mean interior density is @xmath180 @xcite times the critical mass density , @xmath181 , at a cluster redshift is given by @xmath182 in the present paper , the full lensing constraints derived from the joint acs and subaru data allow us to compare , for the first time , x - ray observations with the total mass profile for the entire cluster , from the cluster center to the virial radius .
the lensing analysis is free from any assumptions about the dynamical state of the cluster , so that a joint x - ray and lensing analysis provides a powerful diagnostic of the icm state and any systematic offsets between the two mass determination methods .
here we first summarize our lensing work on abell 1689 before presenting the results from our joint analysis .
@xcite combined hst / acs strong lensing data with subaru weak lensing distortion and magnification data in a two - dimensional analysis to reconstruct the projected mass profile . their full lensing method , assuming the spherical symmetry , yields best - fit nfw model parameters of @xmath183 and @xmath184 ( including both statistical and systematic uncertainties ; see also @xcite ) , which properly reproduce the observed einstein radius of @xmath185 for @xmath186 @xcite .
here we deproject the two - dimensional mass profile of @xcite and obtain a non - parametric @xmath187 profile simply assuming spherical symmetry @xcite . this method is based on the fact that the surface - mass density @xmath188 is related to the three - dimensional mass density @xmath189 by an abel integral transform ; or equivalently , one finds that the three - dimensional mass @xmath190 out to spherical radius @xmath191 is written in terms of @xmath188 as @xmath192 where @xmath193 @xcite .
we propagate errors on @xmath188 using monte - carlo techniques taking into account the error covariance matrix of the full lensing constraints .
this deprojection method allows us to derive in a non - parametric way three - dimensional virial quantities ( @xmath194 ) and values of @xmath195 within a sphere of a fixed mean interior overdensity @xmath196 with respect to the critical density of the universe at the cluster redshift .
the resulting parameters of the non - parametric spherical , as well as nfw , models are summarized in table [ table : lensmass ] .
here we compare our x - ray hydrostatic mass measurements ( in all regions ; [ subsec : icm_outskirts ] ) with the spherical - lensing and nfw mass profiles in figure [ fig : mhe_vs_mwl ] . as discussed in
[ subsec : icm_outskirts ] , the hydrostatic equilibrium assumption is invalid in the cluster outskirts and hence , a mass comparison is limited within the radius @xmath197 , inside which the hydrostatic mass increases with radius . the hydrostatic mass is systematically lower than the lensing mass at all radii .
in particular , the hydrostatic mass in the central region ( @xmath198 ) is significantly lower than the lensing one ( @xmath199 ) . since our hydrostatic mass estimates agree with the latest _
chandra _ results @xcite in the overlapping region , this would not be due to the low angular resolution of the _ suzaku _ satellite . at intermediate radii of @xmath200 , the hydrostatic to lensing mass ratio , @xmath201 ,
is in a range of @xmath202 .
some numerical simulations ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) have shown that x - ray hydrostatic - equilibrium mass estimates would underestimate the actual cluster mass , because the turbulent pressure and the bulk / rotational kinetic pressure partially support the gravity of the cluster .
for instance , @xcite found that the hydrostatic mass is biased low , and is underestimated on average by @xmath203 and @xmath204 at overdensities @xmath205 and @xmath95 , respectively , for a sample of 12 simulated clusters at redshift @xmath206 .
@xcite found that the hydrostatic mass derived from the @xmath34-model gas density profile and spectroscopic - like temperature model , similarly to our modeling , is biased low by @xmath207% and @xmath208% at overdensities of @xmath205 and @xmath95 , respectively .
they found that , on average , the systematic underestimate bias of @xmath147 ( with respect to the actual cluster mass ) increases with decreasing @xmath196 , or increasing the pivot radius @xmath209 .
we find that the degree of systematic bias in the x - ray mass estimate , @xmath210 , is in rough agreement with the results from numerical simulations , except for the central region .
however , such a comparison with simulated clusters in the central region would be practically difficult owing to _ over - cooling _ problems in cosmological hydrodynamical simulations , failing to reproduce the observed icm features . if the icm is not in thermal balance with the gravity and an additional pressure support is provided by the gas bulk / turbulent motions , as suggested by simulations , the x - ray micro - calorimeter instrument ( sxs ) on the next japanese x - ray satellite _ astro - h _
@xcite will directly detect them and measure their contributions to the pressure balance .
another possible cause of the mass discrepancy is the triaxial structure of clusters .
if the major axis of a triaxial halo is pointed to the observer , then this orientation boosts the projected surface mass density @xmath211 and hence the lensing signal . as a result
, halo triaxiality and orientation bias can affect the lensing - based mass and concentration measurements ( e.g , * ? ? ?
@xcite demonstrated in the context of the @xmath212cdm model that clusters with large einstein radii ( @xmath213 ) , or the so - called _ superlenses _ , are likely to be highly biased , with major axes preferentially aligned with the line - of - sight .
the level of correction for the orientation bias in lensing mass estimates ( obtained a priori assuming the spherical nfw model ) is derived from a semi - analytical representation of simulated cdm triaxial halos by @xcite .
@xcite fitted the two - dimensional mass map of umetsu & broadhurst ( 2008 ) with a triaxial halo model and obtained @xmath214 and @xmath215 ( table [ table : lensmass ] ) where the quoted errors here are statistical only ( 68.3% cl ) .
this triaxial description of the full lensing constraints presented in umetsu & broadhurst ( 2008 ) is in better agreement with our x - ray results .
on the other hand , more recently , @xcite demonstrated that a simple triaxial modeling of _ chandra _ x - ray data can relax the mass discrepancy with respect to strong and weak lensing .
they found that the total spherical mass and the spherically - averaged mass density are essentially unchanged under different triaxiality assumptions , as previously reported by @xcite and @xcite . by assuming a prolate symmetry of the gas structure with axis ratio @xmath216
, they found that their x - ray mass estimate agrees with those of strong and weak lensing within 1% ( @xmath217 ) and 25% ( @xmath218 ) , respectively . using the triaxial halo model of @xcite , the x - ray to lensing
mass ratio becomes consistent with unity at larger radii within the errors ; however , it is still much lower than unity in the central region due to the inferred high concentration parameter . based on independent weak lensing data
, @xcite also obtained high concentrations from their triaxial halo modeling in conjunction with the strong lensing prior on the einstein radius . due to the tight strong lensing constraints that prefer a highly - concentrated mass distribution , it is difficult to explain the mass discrepancy in the central region . from our combined x - ray and lensing data set
, we measure the cumulative gas mass fraction , @xmath219 where @xmath220 and @xmath221 are the gas and lensing masses contained within a sphere of radius @xmath191 .
the gas mass is calculated by fitting a modified @xmath34 model to the electron number density profile ( appendix [ appen : n+tpara ] ) .
the errors are obtained by 1000 monte - carlo simulations taking into account the error covariance matrix .
the resulting gas fraction , shown in the right panel of figure [ fig : mhe_vs_mwl ] , increases toward the outskirts , but does not reach the cosmic mean baryon fraction ( wmap5 ; * ? ? ?
* ) within the virial radius . in figure
[ fig : mhe_vs_mwl ] ( right ) , we also compare our result with those from previous statistical studies @xcite : the mean gas fractions for 12 clusters from the local substructure survey ( locuss ) based on a joint _ xmm - newton _ x - ray and _ subaru / suprime - cam _ weak lensing analysis @xcite and those for 4 high - mass clusters ( @xmath222 ) from a joint sunyaev- zeldovich effect ( sze ) _ amiba _ and _ subaru _ lensing analysis @xcite .
we note that these gas fraction measurements , based on the lensing - derived total mass , are independent of the cluster dynamical state .
the mean gas fractions from the two statistical studies are in close agreement with each other . on the other hand , our gas fraction measurements of abell 1689 based on the joint x - ray / lensing analysis , assuming spherical symmetry , are about half of the mean cluster values derived from @xcite and @xcite . for the case of the triaxial mass model @xcite
, the upper bounds of the gas fraction come closer to , and are marginally consistent with , the mean values derived by statistical work .
recently @xcite proposed a method of modeling the icm distributions in gravitational potential wells of clusters , including star formation , energy input , and nonthermal pressure support calibrated by numerical simulations and x - ray observations . in their standard model ,
the gas initially contained inside the virial radius expands outward .
consequently , the cluster baryon fraction reaches the cosmic mean at @xmath223 , which is in rough agreement with our result .
we compare gas thermal pressure obtained from the two x - ray observables , temperature and electron number density , with a pressure calculated from the lensing mass profile assuming the hydrostatic equilibrium ( hereafter lensing pressure ) .
the lensing pressure is a total pressure to balance fully the gravity of the cluster lensing mass ( see also * ? ? ?
we calculated the lensing pressure profile , @xmath224 , by integrating eq .
( [ eq : he ] ) with a boundary condition that the external pressure is zero at @xmath225 ( @xmath226 ) .
the inner pressure is little affected by choices of the outer boundary condition as long as the pressure outside the virial radius is significantly lower than the values at inner radii .
we extrapolated the observed lensing mass and density profile outside the field - of - view .
the errors were computed by 1000 monte carlo simulations .
the lensing pressure is systematically higher than the gas thermal pressure at any radii ( the left panel of figure [ fig : pts ] ) . within the radius of @xmath227 ,
the gas thermal pressure is at most @xmath145 @xmath16 of the lensing pressure . in the outskirts around the virial radius ,
the gas pressure in offset1 is comparable to lensing one , while ones in the offset234 and all regions is @xmath17 @xmath18 .
this is inconsistent with a result from sze analysis of clusters ; a stacked sze signal profile for galaxy clusters using wmap three - year data @xcite has shown that the gas closely follows the dark matter distribution out to the virial radius .
if we assume that the mass discrepancy is fully explained by the gradient of additional pressure supports such as turbulent and/or bulk pressures , they contribute up to @xmath228
@xmath229 of the lensing pressures within @xmath14 and @xmath230
@xmath231 around the virial radius , respectively .
hydrodynamic numerical simulation @xcite has shown the fraction of turbulent pressure to total pressure on average increases up to @xmath232 at @xmath233 , roughly corresponding to @xmath21 .
they also found that the ratio at @xmath14 spans in a range from @xmath234 to @xmath235 for simulated relaxed clusters and from @xmath236 to @xmath237 for unrelaxed clusters , respectively .
@xcite has found that the turbulent energy accounts for @xmath238 @xmath239 of the thermal energy content .
@xcite has shown that the bulk energy changes from @xmath240 of the total energy at the center to @xmath232 at the virial radius and the turbulent energy ( @xmath241 ) is less than the kinetic one .
the fractions of additional non - thermal pressure we derived above are slightly higher than those in simulated clusters .
they should be considered as upper limits , taking into consideration the possibility of triaxial halo model , that decreases the amount of non - thermal pressure relative to total lensing one .
we next computed entropy profiles from the lensing mass profile .
since the fraction of thermal pressure to the total one has not yet been revealed in reality , we can not calculate the entropy of the thermal gas using lensing mass .
we therefore compared the normalized gas entropy with lensing entropy ( @xmath242 ) , as shown in the right panel of figure [ fig : pts ] .
the normalization is applied at the third radial bin .
the lensing entropy surprisingly agrees with the normalized gas entropy in the offset1 direction out to the virial radius .
the normalized entropies in the offset234 and all regions also trace the lensing entropy up to @xmath243 , while , in the outskirts , observed entropies deviate downward . although we can not constrain from figure [ fig : pts ] how much thermal energy is converted from the shock energy , when we remember that the slope of gas entropy in offset1 also coincides with the accretion shock models ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* [ subsection : ana : prof ] ) , our result indicates that the entropy generation by accretion shocks would occur , tracing the cluster mass profile .
we observed abell 1689 with _ suzaku _ and detected the icm emission out to the virial radius .
then , we derived radial profiles of temperature , electron density , and entropy , and conducted a joint x - ray and lensing analysis . our conclusions in this work can be summarized as below . * the icm emission out to the virial radius
was detected with @xmath2 significance .
* the temperature gradually decreases toward the virial radius from @xmath90 kev to @xmath92 kev .
* the temperature variations in azimuthal direction becomes significant in @xmath3 : the temperatures of offset1 ( @xmath4kev ) and offset234 ( @xmath97kev ) are about half and one - sixth of @xmath98 kev which is an emission - weighted temperature of @xmath80@xmath81 and @xmath81@xmath82 regions , respectively .
* the temperature profile in the offset1 direction is in good agreement with the scaled temperature one by @xcite , while the temperatures in the other directions ( offset2 - offset4 ) around the virial radius are lower than the scaled value by a factor of @xmath92 .
* the electron density profile decreases down to @xmath122 around the virial radius .
* the electron densities in the four offset directions are consistent with one another within errors in any circular annulus . *
the slope of the electron number density is @xmath6 in intermediate range of @xmath7 . in the outer annulus of @xmath3 , the density slope , @xmath244 where the lower limit is the systematic error due to
the contamination of x - ray emissions from point sources and inner region , becomes shallower .
this outskirts value is significantly shallower than the universal matter density profile @xmath9 by navarro , frenk , & white ( 1996 ; 1997 ) .
* the entropies in all directions increase from the center to the radius of @xmath75 .
the entropy distribution in the last radial bin ( @xmath75@xmath76 ) is anisotropic , similarly to the temperature profiles . *
the entropy profile of the offset1 direction increases as @xmath142 , as predicted by models of accretion shock heating @xcite , while the entropies of offset2-offset4 in the outskirts are lower than those in their annuli of @xmath82@xmath75 .
the slopes of the entropy profiles in the offset1 direction and the other directions match those predicted by lensing spherical masses with the hydrostatic equilibrium assumption , from the core to the virial radius and to @xmath243 , respectively . * the cumulative hydrostatic mass in the offset1 direction increases with radius ,
while the ones in the other ( offset234 ) regions _ unphysically _ decrease outside @xmath159 .
the icm in the cluster outskirts , except the offset1 direction , is far from hydrostatic equilibrium . * the anisotropic distributions of gas temperature and entropy in the outskirts
are clearly correlated with the large - scale structure associated with abell 1689 . in the offset1 ( northeast )
direction , the hot and high entropy region around the virial radius is connected to the overdensity region of galaxies in the _ sdss _ map , whereas the low - density region ( void ) outside the virial radius is found in the other directions .
it indicates that the thermalization in the outskirts along the filamentary structure takes place faster than that in the void region .
* the hydrostatic to spherical lensing mass ratio , @xmath201 , at intermediate radii of @xmath200 , is in a range of @xmath245 @xmath246 .
using the triaxial halo model , the x - ray to lensing mass ratio in the same annulus is consistent with unity within errors . to the contrary
, the ratio within @xmath80 , @xmath247 , is significantly biased low , and the discrepancy can not be compromised by either the spherical or triaxial models * the gas fraction computed from combined gas density and spherical lensing mass does not reach the cosmic mean baryon fraction within the virial radius .
although our fraction is lower than the mean value of previous statistical studies based on lensing mass , the upper bounds of the gas fraction derived from the triaxial mass model @xcite come closer to , and are marginally consistent with , the statistical works . *
the thermal gas pressure within @xmath243 is at most @xmath248 @xmath16 of the total pressure predicted by spherical lensing mass model . in the cluster outskirts outside @xmath14 ,
the thermal pressure in the offset1 direction is comparable to total pressure , whereas it is @xmath249
@xmath18 of the total pressure in the other directions .
taking into account the triaxial model , these values give the lower limits . with such values
the thermal pressure would be insufficient to balance the total gravity of the cluster , requiring additional pressure support(s ) .
we are very grateful to molnar , s. , m. for helpful comments .
mk acknowledges support from a grant based on the special postdoctoral researchers program of riken .
the present work is supported in part by the grant - in - aid for scientific research ( s ) , no
. 18104004 .
no and mt are in part supported by a grant - in - aid from the ministry of education , culture , sports , science , and technology of japan ( no : 20740099 ; mt : 19740096 ) .
ku acknowledges support by the national science council of taiwan under the grant nsc95 - 2112-m-001 - 074-my2 .
this work is supported by a grant - in - aid for the coe program `` exploring new science by bridging particle - matter hierarchy '' and g - coe program `` weaving science web beyond particle - matter hierarchy '' in tohoku university , funded by the ministry of education , science , sports and culture of japan .
this work is in part supported by a grant - in - aid for science research in a priority area `` probing the dark energy through an extremely wide and deep survey with _
telescope '' ( 18072001 ) from the ministry of education , culture , sports , science , and technology of japan .
we simulated icm emissions from abell 1689 on xis - fi ( xis0 + xis3 ) with ` xissim ` , which spatially follow the double - beta model determined from the 0.510 kev _ xmm - newton _ mos1 image ( [ subsection : obs : reduction ] ) .
as a spectrum model , we used 6.2 kev _
apec _ model ( metal abundance is 0.3 solar ) with a flux of @xmath250 ergs @xmath251 s@xmath58 ( integrated within @xmath252 in radius ) , which was determined from simultaneous fit to xis - fi ( averaged over xis0 and xis3 ) and xis1 spectra of the offset1 observation , extracted from whole xis regions .
since offset1 observation does not cover the central @xmath253 region , the spectrum model reproduces the emission from cluster outskirts .
as shown in figure [ fig : prof_sim ] , the simulated profile is in good agreement with the observation .
this indicates that the observed signals around the virial radius can be described by extrapolating the double - beta profile . from another simulation which used the same double - beta model but truncated at @xmath73
, we found that the stray signals which originate from @xmath254 regions contribute only 22% of signals in @xmath255 regions . in order to study an effect of temperature gradient expected in the icm ,
we also simulated xis - fi ( xis0 + xis3 ) profiles of 9 kev and 2 kev thermal emission with _
apec _ model ( metal abundance is 0.3 solar ) . when normalization of these two simulated profiles are scaled to match at the center , the 9 kev profile is higher than the 2 kev one by 10% outside @xmath73
. therefore , the temperature gradient does not affect the simulated profile significantly .
out of the 49 point sources in table [ table : pntsrc ] , 35 sources have counterparts in the incremental second xmm - newton serendipitous survey catalog ( 2xmmi , * ? ? ?
* ) , on which flux for 5 energy bands , 0.20.5 kev , 0.51.0 kev , 1.02.0 kev , 2.04.5 kev , and 4.512.0 kev , are available from a heasarc website . for sources which are not on the 2xmmia catalog , we estimated their 0.22.0 kev flux by comparing their background - subtracted counts in the circular @xmath68 regions in the pspc total band ( 0.12.0 kev ) image and that of source 7 ( table [ table : pntsrc ] ) of which flux is known . from the estimated 0.22.0 kev flux ,
we further estimated flux for the 5 energy bands , assuming a power law spectrum with photon index of 1.4 . using the derived crude 5-band spectra of the 49 point sources , we simulated xis - fi ( xis0 + xis3 ) images for the 4 observations using ` xissim ` . in order to improve statistics , we set exposures which are 10 times longer than the actual observations .
as shown in figure [ fig : pntsrc_sim ] , the whole area of xis is contaminated with signals from point sources . figure [ fig : pntsrc_prof_azimuth ] shows the simulated 0.52 kev and 210 kev profiles of residual point - source signals in the four pointing directions .
we parametrized the electron density and temperature profiles shown in figure [ fig : prof ] , which is necessary to measure the hydrostatic equilibrium mass ( [ subsec : icm_outskirts ] ) as well as to conduct a joint analysis of _ suzaku _ x - ray satellite and _ subaru / suprime - cam_+_hst / acs _ lensing analysis ( [ sec : joint ] ) .
the @xmath34-model ( e.g. * ? ? ? * ) or its generalized version @xcite are often used to express electron density profiles in clusters . since the low angular resolution of the _ suzaku
_ satellite does not require the complicated modeling @xcite , the analytic expression we use for deprojected electron density profile is given by @xmath256 where @xmath257 is the central electron number density , @xmath34 is the slope parameter , and @xmath258 and @xmath155 are the core radii . in this modified expression of @xmath34 model ,
the number density of the outskirts @xmath153 are much better fitted with the outer radius @xmath155 , than a single @xmath34 model .
unfortunately , since the number of the data points is small , the core radius , @xmath155 , is not constrained well .
we therefore fixed @xmath259 .
our results ( [ section : dis ] ) are not significantly changed by choosing a constant factor of @xmath154 .
the best - fit parameters for all , offset1 and offset234 regions are summarized in table [ table : nfit ] .
the observed projected temperature profile is described using a simple parametrization @xmath260^{-\alpha } , \label{eq : t_pr}\end{aligned}\ ] ] with the temperature normalization @xmath261 , the scale radius @xmath262 and the slope parameter @xmath61 .
we note that the assumed inner slope @xmath263 is insignificant for our main purpose to describe the temperature profile in the outskirts .
we fit the density and temperature profiles in the whole region of the cluster , offset1 and offset234 regions with the above two models , respectively . in the fitting of offset1 region ,
we utilize the data of the other region instead of lack of the first two bins in the profiles .
the best - fit parameters for all , offset1 and offset234 regions are summarized in table [ table : nfit ] .
abazajian , k. n. , adelman - mccarthy , j. k. , ageros , m. a. , allam , s. s. , allende p. c. , an , d. , anderson , k. s. j. , anderson , s. f. , et al .
2009 , apjs , 182 , 543 afshordi , n. , lin , y .- t . , nagai , d. , & sanderson , a. j. r. 2007 , mnras , 378 , 293 akahori , t. & yoshikawa , k. 2009 , pasj , submitted .
preprint : arxiv0909.5025 anders , e. , & grevesse , n. 1989 , gecoa , 53 , 197 atrio - barandela , f. , kashlinsky , a. , kocevski , d. & ebeling , h. 2008 , apj , 675 , l57 bautz , m. w. , miller , e. d. , sanders , j. s. , arnaud , k. a. , mushotzky , r. f. , porter , f. s. , hayashida , k. , henry , j. p. , et al .
2009 , pasj , 61 , 1117 bode , p. , ostriker , j. p. & vikhlinin , a. 2009 , apj , 700 , 989 borgani , s. , finoguenov , a. , kay , s. t. , ponman , t. j. , springel , v. , tozzi , p. , & voit , g. m. 2005 , mnras , 361 , 233 broadhurst , t. , bentez , n. , coe , d. , sharon , k. , zekser , k. , white , r. , ford , h. , bouwens , r. , et al .
2005 , apj , 621 , 53 broadhurst , t. , takada , m. , umetsu , k. , kong , x. , arimoto , n. , chiba , m. , & futamase , t. 2005 , apj , 619 , l143 broadhurst , t. & barkana , r. 2008 , mnras , 390 , 1647 bullock , j. s. , kolatt , t. s. , sigad , y. , somerville , r. s. , kravtsov , a. v. , klypin , a. a. , primack , j. r. & dekel , a. 2001 , mnras , 321 , 559 cavaliere , a. , & fusco - femiano , r. 1976 , a&a , 49 , 137 cavaliere , a. , lapi , a. , & fusco - femiano , r. 2009 , apj , 698 , 580 corless , v. l. , king , l. j. , & clowe , d. 2009 , mnras , 393 , 1235 croston , j. h. , pratt , g. w. , bhringer , h. , arnaud , m. , pointecouteau , e. , ponman , t. j. , sanderson , a. j. r. , temple , r. f. , et .
al 2008 , a&a , 487 , 431 dickey , j. m. , & lockman , f. j. 1990 , ara&a , 28 , 215 dolag , k. , vazza , f. , brunetti , g. & tormen , g. 2005 , mnras , 364 , 753 ettori , s. & fabian , a. c. 1998 , mnras , 293 , l33 faltenbacher , a. , hoffman , y. , gottlber , s. , & yepes , g. 2007 , mnras , 376 , 1327 fang , t. , humphrey , p. , & buote , d. 2009 , apj , 691 , 1648 fox , d. c. , & loeb , a. 1997 , 491 , 459 fujita , y. , tawa , n. , hayashida , k. , takizawa , m. , matsumoto , h. , okabe , n. , & reiprich , t. h. 2008 , pasj , 60 , 343 gavazzi , r. 2005,a&a , 443 , 793 george , m. r. , fabian , a. c. , sanders , j. s. , young , a. j. , & russell , h. r. 2009 , mnras , 395 , 657 hamana , t. , miyazaki , s. , shimasaku , k. , furusawa , h. , doi , m. , hamabe , m. , imi , k. , kimura , m. , et al .
2003 , apj , 597 , 98 hamana , t. , miyazaki , s. , kashikawa , n. , ellis , r. s. , massey , r. j. , refregier , a. , taylor , j. e. 2009 , pasj , 61 , 833 henley , d. b. , & shelton , r. l. 2008 , apj , 676 , 335 ishisaki , y. , maeda , y. , fujimoto , r. , ozaki , m. , ebisawa , k. , takahashi , t. , ueda , y. , ogasaka , y. , et .
al 2007 , pasj , 59 , 113 jeltema , t. e. , hallman , e. j. , burns , j. o. , & motl , p. m. 2008 , apj , 681 , 167 king , i. r. 1962 , aj , 67 , 471 komatsu , e. & seljak , u. 2001 , mnras , 327 , 1353 komatsu , e. , dunkley , j. , nolta , m. r. , bennett , c. l. , gold , b. , hinshaw , g. , jarosik , n. , larson , d. , et al .
2009 , apjs , 180 , 330 koyama , k. , tsunemi , h. , ; dotani , t. , bautz , m. w. , hayashida , k. , tsuru , t. g. , matsumoto , h. , ogawara , y. , et al .
2007 , pasj , 59 , 23 kushino , a. , ishisaki , y. , morita , u. , yamasaki , n. y. , ishida , m. , ohashi , t. , & ueda , y. 2002 , pasj , 54 , 327 lau , e. t. , kravtsov , a. v. , & nagai , d. 2009 , apj , 705 , 1129 lapi , a. & cavaliere , a. 2009 , apj , 692 , 174 lemze , d. , broadhurst , t. , rephaeli , y. , barkana , r. , & umetsu , k. 2009 , apj , 701 , 1336 limousin , m. 2007 , apj , 668 , 643 mitsuda , k. , bautz , m. , inoue , h. , kelley , r. l. , koyama , k. , kunieda , h. , makishima , k. ogawara , y. , et al .
2007 , 59 , s1 miyazaki , s. , komiyama , y. , sekiguchi , m. , okamura , s. , doi , m. , furusawa , h. , hamabe , m. , imi , k. , et al . 2002 , pasj , 54 , 833 molnar , s. m. , hearn , n. , haiman , z. , bryan , g. , evrard , a. e. , & lake , g. 2009 , apj , 696 , 1640 molnar , s. et al .
2010 , submitted to apjl nagai , d. & kravtsov , a. v. 2003 , apj , 587 , 514 nagai , d. , vikhlinin , a. , & kravtsov , a. v. 2007 , apj , 655 , 98 nakamura , t. t. , & suto , y. 1997 , prog.theor.phys .
, 97 , 49 nakazawa , k. , sarazin , c. l. , kawaharada , m. , kitaguchi , t. , okuyama , s. , makishima , k. , kawano , n. , fukazawa , y. , et al . 2009 , pasj , 61 , 339 navarro , j. f. , frenk , c. s. , & white , s. d. m. 1996 , apj , 462 , 563 navarro , j. f. , frenk , c. s. , & white , s. d. m. 1997 , apj , 490 , 493 neto , a. f. , gao , l. , bett , p. , cole , s. , navarro , j. f. , frenk , c. s. , white , s. d. m. , springel , v. , et al .
2007 , mnras , 381 , 1450 oguri , m. , takada , m. , umetsu , k. , & broadhurst , t. 2005 , apj , 632 , 841 oguri , m. , hennawi , j. f. , gladders , m. d. , dahle , h. , natarajan , p. , dalal , n. , koester , b. p. , sharon , k. & bayliss , m. 2009 , apj , 699 , 1038 okabe , n. , & umetsu , k. 2008 , pasj , 60 , 345 okabe , n. , takada , m. , umetsu , k. , futamase , t. , & smith , g. p. 2009 , pasj submitted .
preprint : arxiv:0903.1103 okabe , n. , zhang , y .,- y .
, finoguenov , a. , takada , m. , umetsu , k. , smith , g. p. , & futamase , t. 2010 , apj submitted .
ostriker , j. p. , bode , p. , & babul , a. 2005 , apj , 634 , 964 piffaretti , r. , jetzer , p. , & schindler , s. 2003 , a&a , 398 , 41 piffaretti , r. , & valdarnini , r. 2008 , a&a , 491 , 71 peng , e .- h . , andersson , k. , bautz , m. w. , & garmire , g. p. 2009 ,
apj , 701 , 1283 ponman , t. j. , sanderson , a. j. r. , & finoguenov , a. 2003 , mnras , 343 , 331 pratt , g. w. , arnaud , m. , & pointecouteau , e. 2006 , a&a , 446 , 429 pratt , g. w. , bhringer , h. , croston , j. h. , arnaud , m. , borgani , s. , finoguenov , a. , & temple , r. f. 2007 , a&a , 461 , 71 reiprich , t. h. , hudson , d. s. , zhang , y. -y . , sato , k. , ishisaki , y. , hoshino , a. , ohashi , t. , ota , n. , et al .
2009 , a&a , 501 , 899 ryu , d. , kang , h. , hallman , e. , & jones , t. w. 2003 , apj , 593 , 599 serlemitsos , p. j. , soong , y. , chan . k. , okajima , t. , lehan , j. p. , maeda , y. , itoh , k. , mori , h. , et al .
2007 , pasj , 59 , 9 smith , r. k. , brickhouse , n. s. , liedahl , d. a. , & raymond , j. c. 2001 , apj , 556 , l91 snowden , s. l. , mushotzky , r. f. , kuntz , k. d. , & davis , d. s. 2008 , a&a , 478 , 615 spitzer , l. jr . 1962 , physics of fully ionized gases ( new york : wiley ) struble , m. f. , & rood , h. j. 1999 , apjs , 125 , 35 sunyaev , r. a. , & zeldovich , y. b. 1972 , comments astrophys .
space phys . , 4 , 173 takahashi , t. , kelley , r. , mitsuda , k. , kunieda , h. , petre , r. , white , n. , dotani , t. , fujimoto , r. , et al .
2008 , proc .
spie , 7011 , 70110o takizawa , m. 1998 , apj , 509 , 579 takizawa , m. 1999 , apj , 520 , 514 tawa , n. , hayashida , k. , nagai , m. , nakamoto , h. , tsunemi , h. , yamaguchi , h. , ishisaki , y. , miller , e. d. , et al .
2008 , pasj , 60 , 11 taylor , j. e. , & navarro , j. f. 2001 , apj , 563 , 483 tozzi p. , & norman , c. 2001 , apj , 546 , 63 umetsu , k. , & broadhurst t. 2008 , apj , 684 , 177 umetsu , k. , birkinshaw , m. , liu , g. , wu , j. p. , medezinski , e. , broadhurst , t. , lemze , d. , zitrin , a. , et al .
2009 , apj , 694 , 1643 umetsu , k. , medezinski , e. , broadhurst , t. , zitrin , a. , okabe , n. , hsieh , b. c. , & molnar , s. m. 2009 , apj submitted .
preprint : arxiv0908.0069 vazza , f. , brunetti , g. , kritsuk , a. , wagner , r. , gheller , c. , & norman , m. 2009 , a&a , 504 , 33 vikhlinin , a. , forman , w. , & jones , c. 1999 , apj , 525 , 47 vikhlinin , a. , kravtsov , a. , forman , w. , jones , c. , markevitch , m. , murray , s. s. , & van speybroeck , l. 2006 , apj , 640 , 691 vikhlinin , a. , burenin , r. a. , ebeling , h. , forman , w. r. , hornstrup , a. , jones , c. , kravtsov , a. v. , murray , s. s. , et al .
2009 , apj , 692 , 1033 voit , g. m. , bryan , g. l. , balogh , m. l. , & bower , r. g. 2002 , apj , 601 watson , m. g. , schrder , a. c. , fyfe , d. , page , c. g. , lamer , g. , mateos , s. , pye , j. , sakano , m. , et al . 2009 ,
a&a , 493 , 339 zhang , y .- y . ,
bhringer , h. , finoguenov , a. , ikebe , y. , matsushita , k. , schuecker , p. , guzzo , l. , & collins , c. a. 2006 , a&a , 456 , 55 .
zhang , y .- y . , finoguenov , a. , bhringer , h. , kneib , j .-
p . , smith , g. p. , czoske , o. & soucail , g. 2007 , a&a , 467 , 437 zhang , y .- y . , finoguenov , a. , bhringer , h. , kneib , j .- p . ,
smith , g. p. , kneissl , r. , okabe , n. , & dahle , h. 2008 , a&a , 482 , 451 zhang .
, y- . , y. , okabe , n. , finoguenov , a. , graham , p. s. , piffaretti . , r. , valdarnini , r. , babul , a. , evrard , a. e. , et al
. 2010 apj submitted .
preprint : arxiv:1001.0780 kev xis - fi ( xis0 @xmath60 xis3 ) images smoothed with a gaussian kernel of @xmath264 .
background is not subtracted and vignetting is not corrected .
ccd regions where @xmath265fe calibration sources are irradiated ( @xcite ) are excluded .
( right ) the same as the left panel , but point - source regions are specified as circles of @xmath68 radius .
see text for details of the point - source detection .
, title="fig:",scaledwidth=30.0% ] kev xis - fi ( xis0 @xmath60 xis3 ) images smoothed with a gaussian kernel of @xmath264 .
background is not subtracted and vignetting is not corrected .
ccd regions where @xmath265fe calibration sources are irradiated ( @xcite ) are excluded .
( right ) the same as the left panel , but point - source regions are specified as circles of @xmath68 radius .
see text for details of the point - source detection .
, title="fig:",scaledwidth=31.5% ] ( right ) are removed , but vignetting is not corrected .
raw ( background inclusive ) profile ( black ) is shown with xrb ( green ) , nxb ( magenta ) , and xrb@xmath60nxb ( blue ) profiles . a resulting background - subtracted profile ( black @xmath70 blue )
is shown in red .
the error bars in these profiles are 1@xmath66 . in the error
bars in the red profile , 1@xmath66 uncertainties of the xrb and nxb are added in quadrature to the corresponding statistical 1@xmath66 errors .
( right ) the same as the left panel , but for 210 kev profiles .
, title="fig:",scaledwidth=40.0% ] ( right ) are removed , but vignetting is not corrected .
raw ( background inclusive ) profile ( black ) is shown with xrb ( green ) , nxb ( magenta ) , and xrb@xmath60nxb ( blue ) profiles . a resulting background - subtracted profile ( black @xmath70 blue )
is shown in red .
the error bars in these profiles are 1@xmath66 . in the error
bars in the red profile , 1@xmath66 uncertainties of the xrb and nxb are added in quadrature to the corresponding statistical 1@xmath66 errors .
( right ) the same as the left panel , but for 210 kev profiles .
, title="fig:",scaledwidth=40.0% ] uncertainties of the xrb and nxb are added in quadrature to the corresponding statistical 1@xmath66 errors .
( right ) the same as the left panel , but for 210 kev profiles .
, title="fig:",scaledwidth=40.0% ] uncertainties of the xrb and nxb are added in quadrature to the corresponding statistical 1@xmath66 errors .
( right ) the same as the left panel , but for 210 kev profiles .
, title="fig:",scaledwidth=40.0% ] . for comparison ,
background - subtracted 0.510 kev surface brightness profile of xis - fi averaged over xis0 and xis3 ( black ) is also shown . in this black profile , the residual contribution of point sources estimated in [ subsection : ana : syserr ] is also subtracted . in the error bars of the red profile ,
1@xmath66 uncertainties of the xrb and nxb are added in quadrature to the corresponding statistical 1@xmath66 errors .
, scaledwidth=37.0% ] xis3 ) mosaic image of point sources , smoothed with a gaussian kernel of @xmath264 . @xmath68
circular regions around these sources are excluded .
simulated exposures of the 4 observations are set 10 times longer than the actual ones . , scaledwidth=37.0% ] lccc offset1 ( 803024010 ) & 2008/july/23 04:42:40 & 2008/july/24 08:22:23 & 39.2 + offset2 ( 803025010 ) & 2008/july/24 08:23:04 & 2008/july/25 10:09:18 & 38.2 + offset3 ( 803026010 ) & 2008/july/25 10:09:59 & 2008/july/26 11:23:19 & 39.3 + offset4 ( 803027010 ) & 2008/july/26 11:24:00 & 2008/july/27 12:13:14 & 37.9 + ccc & fraction & significance + offset1 0.52 kev & 0.21 & 5.4@xmath66 + offset2 0.52 kev & 0.25 & 2.9@xmath66 + offset3 0.52 kev & 0.45 & 2.0@xmath66 + offset4 0.52 kev & 0.09 & 2.6@xmath66 + offset1 210 kev & 0.47 & 2.1@xmath66 + offset2 210 kev & 0.80 & 0.0@xmath66 + offset3 210 kev & @xmath282 & 0.0@xmath66 + offset4
210 kev & 0.64 & 0.2@xmath66 + cccccccccc @xmath283@xmath74 & & & @xmath284 & @xmath285 & @xmath286 & @xmath287 & @xmath288 & @xmath289 & @xmath290 + @xmath74@xmath291 & & & @xmath292 & @xmath293 & @xmath294 & @xmath295 & @xmath296 & @xmath297 & @xmath298 + @xmath291@xmath107 & @xmath299 & @xmath295 & @xmath300 & @xmath301 & @xmath302 & @xmath303 & @xmath304 & @xmath305 & @xmath306 + @xmath107@xmath73 & @xmath307 & @xmath308 & @xmath309 & @xmath310 & @xmath311 & @xmath312 & @xmath313 & @xmath314 & @xmath315 + @xmath73@xmath316 & @xmath317 & @xmath318 & @xmath319 & @xmath320 & @xmath321 & @xmath322 & @xmath323 & @xmath324 & @xmath325 + region & & & @xmath326/d.o.f + @xmath283@xmath74 & & & @xmath327 + @xmath74@xmath291 & & & @xmath328 + @xmath291@xmath107 & & & @xmath329 + @xmath107@xmath73 & & & @xmath330 + @xmath73@xmath316 & & & @xmath331 cccccc @xmath283@xmath74 & & & @xmath332 & @xmath333 & @xmath334 + @xmath74@xmath291 & & & @xmath335 & @xmath336 & @xmath337 + @xmath291@xmath107 & @xmath338 & @xmath339 & @xmath340 & @xmath341 & @xmath342 + @xmath107@xmath73 & @xmath343 & @xmath344 & @xmath345 & @xmath346 & @xmath347 + @xmath73@xmath316 & @xmath348 & @xmath349 & @xmath350 & @xmath351 & @xmath352 + ccccccc all & @xmath353 & @xmath354 & @xmath355 & @xmath356 & @xmath357 & @xmath358 + offset1 & @xmath359 & @xmath360 & @xmath361 & @xmath362 & @xmath363 & @xmath364 + offset234 & @xmath365 & @xmath366 & @xmath367 & @xmath368 & @xmath369 & @xmath370 + | we present results of new , deep _
suzaku _ x - ray observations ( @xmath0ks ) of the intracluster medium ( icm ) in abell 1689 out to its virial radius , combined with complementary data sets of the projected galaxy distribution obtained from the _ sdss _ catalog and the projected mass distribution from our recent comprehensive weak and strong lensing analysis of _ subaru / suprime - cam _ and _ hst / acs _ observations .
faint x - ray emission from the icm around the virial radius ( @xmath1 ) is detected at @xmath2 significance , thanks to low and stable x - ray background of _ suzaku_. the _ suzaku _ observations reveal anisotropic gas temperature and entropy distributions in cluster outskirts of @xmath3 correlated with large - scale structure of galaxies in a photometric redshift slice around the cluster .
the high temperature ( @xmath4kev ) and entropy region in the northeastern ( ne ) outskirts is apparently connected to an overdense filamentary structure of galaxies outside the cluster .
the gas temperature and entropy profiles in the ne direction are in good agreement , out to the virial radius , with that expected from a recent _ xmm - newton _ statistical study and with an accretion shock heating model of the icm , respectively . to the contrary , the other outskirt regions in contact with low density void environments have low gas temperatures ( @xmath5kev ) and entropies , deviating from hydrostatic equilibrium .
these anisotropic icm features associated with large - scale structure environments suggest that the thermalization of the icm occurs faster along overdense filamentary structures than along low - density void regions .
we find that the icm density distribution is fairly isotropic , with a three - dimentional density slope of @xmath6 in the radial range of @xmath7 , and with @xmath8 in @xmath3 , which however is significantly shallower than the navarro , frenk , & white universal matter density profile in the outskirts , @xmath9 .
a joint x - ray and lensing analysis shows that the hydrostatic mass is lower than spherical lensing one ( @xmath10 ) but comparable to a triaxial halo mass within errors , at intermediate radii of @xmath11 . on the other hand , the hydrostatic mass within @xmath12 is significantly biased as low as @xmath13 , irrespective of mass models .
the thermal gas pressure within @xmath14 is , at most , @xmath15@xmath16 of the total pressure to balance fully the gravity of the spherical lensing mass , and @xmath17@xmath18 around the virial radius .
although these constitute lower limits when one considers the possible halo triaxiality , these small relative contributions of thermal pressure would require additional sources of pressure , such as bulk and/or turbulent motions . |
lattice calculations of bulk thermodynamic observables , e.g. the energy density and pressure , clearly show that the transition to the high temperature phase of qcd is accompanied by the liberation of partonic degrees of freedom .
asymptotically , at infinite temperature , energy density and pressure approach the limit of an ideal gas of quarks and gluons . at @xmath0
the deviations from this asymptotic behaviour are still about 15% which is too large to be accounted for by ordinary high temperature perturbation theory .
this shows the relevance of non - perturbative effects leading , for instance , to thermal quark and gluon masses . for temperatures close to @xmath1 , _
i.e. _ @xmath2 , these non - perturbative features dominate bulk thermodynamic behaviour , leading to large deviations from the ideal gas relation @xmath3 , a strong reduction of the velocity of sound and large screening lengths @xcite .
the observation of large elliptic flow in heavy ion collisions at rhic , its successful description in terms of ideal hydrodynamics as well as the observed strong modification of jets also led to the conclusion that the medium produced in heavy ion collisions at rhic , which is expected to be generated and thermalized at temperatures @xmath2 , is opaque and still strongly interacting @xcite .
it thus became of considerable interest to understand in more detail the structure of the interacting medium in the vicinity of the transition temperature . in this context
it has been suggested that a large set of colored bound states of light quarks could exist above @xmath1 and dominate the bulk thermodynamics for @xmath4 @xcite .
indeed , lattice calculations of in - medium properties of heavy quark bound states have led to the conclusions that correlation functions for some of these states , in particular @xmath5 and @xmath6 , are not significantly modified in the vicinity of @xmath1 and that these states thus can persist to exist as bound states even at @xmath7 @xcite . on the other hand , there is ample evidence that correlation functions of light quarks are strongly modified immediately above @xmath1 .
for instance , scalar and pseudo - scalar correlation functions become degenerate as a consequence of chiral symmetry restoration ; in the chiral limit the pseudo - scalar particle no longer is a goldstone particle which is reflected in the structure of spectral functions where the temperature independent , @xmath8-function like peak present at low temperature gets replaced by a broad temperature dependent bump at energies @xmath9 @xcite .
moreover , the analysis of spatial correlation functions in the scalar and vector channels showed a strong sensitivity to modifications of temporal boundary conditions for the fermion ( quark ) fields above @xmath1 while they are insensitive to this below @xmath1 @xcite .
this suggests that in these quantum number channels a bosonic bound state does not give the dominant contribution to the correlation functions above @xmath1 ; the fermionic substructure of independently propagating quarks and anti - quarks becomes visible .
if light quark bound states exist above @xmath1 they thus must have quite peculiar quasi - particle properties . in recent lattice studies of 2-flavour qcd at non - zero quark ( @xmath10 ) chemical potential higher order derivatives of the qcd partition function ( generalized susceptibilities ) with respect to quark chemical potentials
have been used to also analyze higher moments of fluctuations of quark number and electric charge @xcite .
in particular , the second order derivatives with respect to @xmath11 or @xmath12 are monotonously rising across @xmath1 while the fourth order derivatives have pronounced peaks at the transition temperature .
we will show in the following that this reflects the transition from hadrons to quarks as the dominant degrees of freedom that carry baryon number and electric charge in the hot medium @xcite .
with quark masses corresponding to a pseudo - scalar ( pion ) mass of about 770 mev.,width=188 ] 0.6 cm with quark masses corresponding to a pseudo - scalar ( pion ) mass of about 770 mev.,width=188 ] 0.2 cm with quark masses corresponding to a pseudo - scalar ( pion ) mass of about 770 mev.,width=188 ] 0.6 cm with quark masses corresponding to a pseudo - scalar ( pion ) mass of about 770 mev.,width=188 ] -1.0 cm -0.4 cm fluctuations of quark number and charge at vanishing net baryon number density ( vanishing chemical potential ) can be obtained from the qcd partition function through appropriate combinations of derivatives with respect to @xmath13-quark chemical potentials , @xmath14 where @xmath15 and @xmath16 .
results for these cumulants obtained from a calculation within 2-flavour qcd are shown in fig .
[ fig : data ] . the ratios of the quark number and electric charge cumulants , @xmath17 , @xmath18 , are sensitive observables that allow to identify the carriers of quark number and electric charge in a thermal medium .
this is quite apparent in the low temperature phase where a hadron resonance gas ( in boltzmann approximation ) is known to give a good description of the bulk thermodynamics @xcite .
this immediately leads to the expectation that @xmath19 for @xmath20 .
the analysis of electric charge fluctuations is a bit more subtle as doubly charged baryons start contributing to the thermodynamics significantly for @xmath21 . taking into account
separately the contribution from the mesonic isospin triplet sector ( @xmath22 ) and the baryonic isospin doublet ( @xmath23 ) and quartet ( @xmath24 ) sectors one finds for the charge fluctuations @xcite , @xmath25 at low temperature this ratio is dominated by charge fluctuations in the pion sector , which contributes to @xmath22 .
the ratio will thus approach unity at low temperatures and is monotonically increasing with temperature .
this indicates that in the low temperature limit all charged degrees of freedom carry one unit of charge ; @xmath26 increases only due to contributions arising from isospin quartet baryons which can carry charge @xmath27 .
the generic features expected for quark number and charge fluctuations in a hadron gas are reproduced by the lattice results for @xmath28 shown in fig .
[ fig : r42 ] . in the high temperature phase
these ratios apparently change drastically .
in fact , in the infinite temperature , ideal gas limit the ratios will approach @xmath29 and @xmath30 , respectively . as can be seen from fig .
[ fig : r42 ] these asymptotic values are almost reached for @xmath31 . -1.0 cm
-0.4 cm like in the hadronic phase one would expect also for the high temperature phase that the presence of doubly charged states or states carrying more than one unit of quark number would lead to an increase of the ratios @xmath28 .
the analysis of light quark bound states @xcite indeed suggests that in addition to colored @xmath32 and @xmath33 states also @xmath34-states could exist above @xmath1 .
the latter , however , would be only weakly bound and presumably would disappear already at @xmath35 .
if such states would contribute to the thermodynamics above @xmath1 they also would lead to an increase in charge and quark number fluctuations .
the dashed lines shown in fig . [ fig : r42 ] indicate the increase in @xmath28 at the presumed melting temperature of @xmath34-states , @xmath35 @xcite , assuming that at this temperature the @xmath34-states contribute only with half their statistical weight to the qcd partition function .
this clearly overestimates the fluctuations observed at high temperature in lattice calculations .
we have discussed some generic features of quark number and charge fluctuations in 2-flavour qcd .
we argue that the ratio of quartic and quadratic fluctuations are sensitive observables that can directly provide information on the constituents of the thermal medium that carry net quark number and electric charge , respectively .
we have shown that below the qcd transition temperature these ratios are in reasonable agreement with a hadronic resonance gas .
above @xmath1 the ratios rapidly drop and approach the high temperature ideal gas values .
this suggests that already for @xmath36 quark number and charge are predominantly carried by states with the quantum numbers of quarks .
9 f. karsch , lect .
notes phys .
* 583 * ( 2002 ) 209 .
first three years of operation of rhic , nucl .
a * 757 * ( 2005 ) 1 - 283 .
shuryak and i. zahed , phys .
d * 70 * ( 2004 ) 054507 ; + j. liao and e. v. shuryak , hep - ph/0508035 .
f. karsch , eur .
j. c * 43 * ( 2005 ) 35 .
f. karsch et al . , phys .
b * 530 * ( 2002 ) 147 .
g. boyd , s. gupta , f. karsch and e. laermann , z. phys .
c * 64 * ( 1994 ) 331 . c. r. allton et al .
d * 71 * ( 2005 ) 054508 .
s. ejiri , f. karsch and k. redlich , hep - ph/0509051 .
f. karsch , a. tawfik and k. redlich , phys .
b * 571 * ( 2003 ) 67 and eur .
j. c * 29 * ( 2003 ) 549 . | we analyze fluctuations of quark number and electric charge , in 2-flavour qcd at finite temperature and vanishing net baryon number density . in the hadronic phase we find that an enhancement of charge fluctuations arises from contributions of doubly charged hadrons to the thermodynamics .
the rapid suppression of fluctuations seen in the high temperature phase suggests that in the qgp quark number and electric charge are predominantly carried by quasi - particles with the quantum numbers of quarks . |
the so(10 ) grand unified theory has several interesting features @xcite . it can accommodate left - right symmetry as one of the intermediate symmetry and hence provides an explanation of parity violation @xcite .
@xmath4 is a generator of the group so(10 ) and hence lepton number violation takes place spontaneously .
this explains the origin of lepton number violation and neutrino majorana mass naturally .
the smallness of neutrino masses is assured by the see - saw mechanism , so that by keeping the scale of @xmath4 violation high the smallness of the neutrino mass is guaranteed .
quarks and leptons are treated equally in so(10 ) gut .
gauge coupling unification is consistent with all low energy results . on the one hand
there are many attractive features of the so(10 ) gut , on the other the predictability becomes low .
depending on the symmetry breaking pattern and higgs scalar contents , the model can have widely differing predictions .
attempts have been made to construct a minimal model . in one approach
the minimal model is constructed with minimum numbers of parameters , while in the other approach minimum numbers of higgs scalars are included in the model .
there are also models without any intention of minimality or simplicity , where the main aim is to explain all experiments and have maximum predictability . in the present article
we shall study an so(10 ) gut , which has the minimum dimensions of the higgs scalars . in any so(10 ) gut the minimal number of higgs scalar includes an @xmath5 symmetry breaking higgs scalar , which can give masses to the fermions and some scalars that can break the @xmath4 symmetry along with the @xmath6 symmetry and can give majorana masses to the neutrinos .
in addition , there is one higgs scalar which breaks the group so(10 ) gut .
one conventional model includes a higgs bi - doublet ( a * 10 * -plet of so(10 ) higgs scalar , which is doublet under both the @xmath5 and @xmath7 groups ) and both @xmath5 and @xmath6 higgs triplets . in such models with triplet higgs scalars
neutrinos acquire masses at the tree level .
the so(10 ) representation that contains this higgs scalar is of * 126 * dimensions . in another version of the model ,
one breaks the left - right symmetry and the @xmath4 symmetry by doublets of @xmath5 and @xmath6 groups .
this higgs field belongs to a * 16 * -plet representation of so(10 ) . for symmetry breaking and giving fermion masses another * 10
* -plet higgs scalar is introduced , which contains the bi - doublet higgs and hence gives tree level masses to the fermions and the neutrinos acquire only dirac masses of the order of other fermion masses .
there is no majorana mass term for the neutrinos and hence see - saw mechanism is not possible .
however , there exist effective higher - dimensional operators , which can give correct majorana masses to both the left - handed and right - handed neutrinos .
recently it has been pointed out that it is possible to consider an so(10 ) gut , which does not have any higgs bi - doublet scalar belonging to a * 10 * -plet of so(10 ) @xcite .
the higgs scalar that breaks left - right symmetry and the @xmath4 symmetry belongs to a * 16 * -plet of higgs scalar .
since tree level fermion masses are not allowed without the bi - doublet scalar , all fermion masses come from higher - dimensional operators in the see - saw form . in supersymmetric theories
the non - renormalizablity theorem does not allow radiative generation of such higher dimensional operators .
we shall then restrict ourselves to only non - supersymmetric so(10 ) gut .
the source of see - saw suppression for the fermion number violating majorana mass terms are different from the source of see - saw suppression for the fermion number conserving dirac mass terms , which maintain the large hierarchy between the charged fermion masses and the neutrino masses . in this article
we shall study some aspects of this model in detail . in the next section
we shall discuss the model . in sec .
[ hpot ] we shall present details of the scalar potential minimization and the allowed symmetry breaking pattern and in sec . [ fmass ] the generation of fermion masses is discussed .
we shall then study the renormalization group equation for this model with the specific choice of the higgs scalars . in sec .
[ gcu ] we shall study the gauge coupling unification and in sec . [ yukawa ] the yukawa coupling evolution for the different fermions .
since all fermion masses have the same see - saw origin , the perturbative unification becomes an important question in these models .
some of the yukawa couplings could become large , although the effective fermion masses still remains small . in sec .
[ lepto ] leptogenesis in our model is discussed and in the last section we summarize our results .
the starting point for any so(10 ) gut is the choice of the symmetry breaking pattern .
there exists many chains of symmetry breaking pattern , which are all consistent with our present knowledge .
so the particular choice of a symmetry breaking pattern defines a specific model . we shall consider a symmetry breaking pattern , which requires a minimum number of higgs scalars , given by : @xmath8 \nonumber \\ & { \stackrel{\phi_4 } \longrightarrow } & { \rm su(3)}_c\times{\rm su(2)}_l\times{\rm su(2)}_r\times{\rm u(1)}_{b - l } \quad [ { \rm g}_{3221 } ] \nonumber\\ & { \stackrel{\chi_r}\longrightarrow } & { \rm su(3)}_c\times{\rm su(2)}_l\times{\rm u(1)}_y \quad[{\rm g}_{321 } ] \nonumber \\ & { \stackrel{\chi_l}\longrightarrow}&{\rm su(3)}_c\times { \rm u(1)}_{\rm em}\ , , \label{sbp}\end{aligned}\ ] ] where the higgs fields responsible for the symmetry breakings , @xmath9 , @xmath10 , @xmath11 and @xmath12 are explicitly shown in the above equation . both @xmath9 and @xmath10
are contained in @xmath13 , which transforms as the * 210 * dimensional representation of so(10 ) . the * 210
* -plet decomposes under the pati - salam subgroup @xmath14 of so(10 ) as , @xmath15 in the above decomposition @xmath9 corresponds to @xmath16 and @xmath10 corresponds to ( * 15,1,1 * ) .
the higgs fields @xmath12 and @xmath17 belong to the * 16 * dimensional spinor representation @xmath18 of so(10 ) and the other fields @xmath11 and @xmath19 belong to the conjugate representation @xmath20 , called @xmath21 , of so(10 ) .
the group transformation properties of the @xmath22 fields under @xmath23 and @xmath14 are as follows : @xmath24 in the above equations @xmath25 , and @xmath26 denote the group transformation properties of the higgs fields under @xmath23 and @xmath14 . at this stage
we shall digress to discuss one important feature of the left - right symmetric models , namely the question of parity @xmath27 .
the discrete @xmath28 symmetry , that interchanges the two su(2 ) subgroups of the lorentz group o(3,1 ) , is called the parity .
this parity can be identified as the discrete @xmath28 symmetry operator that interchanges the groups @xmath5 and @xmath6 of the left - right symmetric model , which implies that under parity @xmath29 .
this definition extends to scalars also .
that is , an @xmath5 doublet scalar field @xmath12 will transform to an @xmath6 doublet scalar field @xmath11 under the operation of parity @xmath30 . in the conventional left - right symmetric models , the parity is spontaneously broken along with the group @xmath6 . in other words , when the left - right symmetric group @xmath31 is spontaneously broken , parity is also spontaneously broken .
there is another possibility of breaking parity spontaneously without breaking the left - right symmetric group .
since the scalar fields transform trivially under the lorentz group , the vev of a parity odd field can break parity spontaneously without breaking the left - right symmetry .
unlike the conventional case , now the parity acting on the fermions and vector bosons is not spontaneously broken . to distinguish these two cases , this second type of parity
is called a @xmath1-parity .
thus when @xmath1-parity is broken , the left - handed and right - handed scalars can have different mass and vev and hence the gauge coupling constants of @xmath5 and @xmath6 can also be different . in the present model @xmath1-parity
plays a very crucial role , both for symmetry breaking as well as for fermion masses .
it also plays some role in gauge coupling unification .
the @xmath32 representation of so(10 ) is a totally antisymmetric tensor of rank four @xmath33 and the singlet @xmath9 is the component @xmath34 in the notation , in which , @xmath35 are @xmath36 indices and @xmath37 are @xmath38 indices . thus under @xmath1-parity @xmath9 is odd ( @xmath39 ) and consequently when it gets its vacuum expectation value ( vev ) at the gut scale , @xmath40 , it breaks the left - right parity of the theory . due to this spontaneous breaking of the left - right parity at the @xmath40 scale
we will have @xmath41 at a lower energy scale .
this @xmath1-parity odd field is also required to give masses to the light neutrinos .
next we write down the fermions in our model and their group transformation properties .
the left - handed quarks , leptons , anti - quarks and anti - leptons belong to a * 16 * -plet representation of so(10 ) , which transform under @xmath14 as : @xmath42 @xmath43 is the generation index . the right - handed fermions and anti - fermions belong to the conjugate representation , @xmath44 in addition to the above mentioned conventional particles our model consists of an extra so(10 ) gauge singlet fermion per generation : @xmath45 @xmath46 .
under @xmath23 the states @xmath47 and @xmath48 transform as : @xmath49 and as a result the fermions can be labelled as : @xmath50 and @xmath51 the generators of the left - right symmetry group @xmath23 are related to the electric charge of the particles by , @xmath52 where @xmath53 in the conventional left - right symmetric models there is one bi - doublet higgs scalar @xmath54 , which gives masses to quarks and charged leptons and a dirac mass to the neutrinos through its couplings of the form @xmath55 .
in addition , there are triplet higgs scalars @xmath56 and @xmath57 , which can give majorana masses to the left - handed and right - handed neutrinos through the couplings @xmath58 . in our present model
all these higgs scalars @xmath59 , @xmath60 and @xmath61 are absent and hence there are no tree level fermion masses for the quarks and the leptons . after discussing the structure of the higgs vacuum expectation values in this model , we shall come back to the question of fermion masses .
in the conventional left - right symmetric models , the combinations of the higgs fields , @xmath59 , @xmath60 and @xmath61 ensures that for certain choices of parameters , @xmath61 can acquire a very large vev compared to other fields breaking left - right symmetry at a large scale .
it is clear that in the absence of the field @xmath59 , both the fields @xmath62 would acquire equal vevs .
it has also been shown that in the absence of the field @xmath59 in a left - right symmetric model with only the doublet higgs scalars @xmath12 and @xmath11 , the minimization of the potential would result in equal vevs for both @xmath12 and @xmath11 , which would lead to inconsistency and parity will be conserved at low energy .
this problem could be solved if parity is broken in these theories either explicitly or spontaneously . in the present model
this problem does not occur .
it was mentioned in the original version of the model that the @xmath1-parity odd singlet field @xmath63 under g@xmath64 contained in the @xmath32 representation would allow @xmath65 and break left - right symmetry at some high scale . in this section
we shall minimize the scalar potential and discuss the various possible solutions , which allows left - right symmetry breaking at some high scale .
let us consider the higgs potential @xcite : @xmath66 + m_d \phi ( \gamma^\dagger \gamma ) + \lambda_{\phi \gamma } \phi^2 ( \gamma^\dagger \gamma)\ , .
\label{hp}\end{aligned}\ ] ] the coupling @xmath67 is the most important term that is required for the left - right breaking to take place at a higher scale compared to standard model symmetry breaking .
@xmath1-parity is broken when @xmath9 acquires a non - vanishing vev , @xmath68 , at the @xmath40 scale , since @xmath9 is odd under @xmath1-parity .
@xmath10 will get a non - vanishing vev at the @xmath69 scale .
there will be many terms in eq .
( [ hp ] ) including @xmath70 , but as we are analyzing the structure of eq .
( [ hp ] ) in the @xmath23 phase and mainly interested on the vevs of the @xmath22 fields , we do not explicitly write down the terms including @xmath71 . @xmath70 has no important contribution in the expressions of the vevs of the @xmath22 fields .
we now discuss the masses of the components of @xmath18 and the vevs .
the scalar potential responsible for the masses of the fields @xmath12 and @xmath11 is given by , @xmath72 the masses of these fields are then given by , @xmath73 if @xmath1-parity is conserved , @xmath74 and the masses of both @xmath12 and @xmath11 become equal . since the vev of @xmath9 breaks @xmath1-parity , it will be possible to fine tune parameters to obtain the mass of @xmath12 to be orders of magnitude smaller than the mass of @xmath11 . from phenomenological consideration
we also require @xmath75 we shall next check if this widely different vevs for @xmath12 and @xmath11 is possible .
@xmath76 breaks the electroweak symmetry , while @xmath77 breaks the left - right symmetry at a very high scale , close to the gut scale .
we denote the vevs of the fields @xmath12 and @xmath11 as : @xmath78 instead of minimizing the potential , we shall first write down the potential in terms of the vevs of the fields and then find the conditions satisfied by the vevs . with the above vevs we can write the higgs potential in the @xmath23 phase as : @xmath79 ( v^*_r v_r + v^*_l v_l ) - \frac{\lambda_\gamma}{4}(v^*_r v_r + v^*_l v_l)^2 \nonumber\\ & & -\,\frac{\lambda'_\gamma}{4}(v^4_l + v^4_r + v^{*\,4}_l + v^{*\,4}_r ) + m_d \eta ( v^*_r v_r - v^*_l v_l)\ , .
\label{pot1}\end{aligned}\ ] ] the @xmath80 symbols in the above equation stands for terms containing @xmath70 . setting @xmath81 and @xmath82 , which amounts to saying that there is no cp violation and all vevs
are considered to be real , the extremum conditions of @xmath83 comes out to be : @xmath84 = 0\,,\nonumber \\
\label{vl}\\ \frac{\partial v}{\partial v_r}&= & 2v_r \left [ m^2_\gamma + ( \lambda_{\phi \gamma } \eta^2 + \cdot \cdot \cdot ) - \frac{\lambda_\gamma}{2 } ( v^{2}_l + v^{2}_r ) -\lambda'_\gamma v^2_r + m_d \eta \right ] = 0\,.\nonumber \\ \label{vr}\end{aligned}\ ] ] the above equations imply , @xmath85=0\ , . \label{vlvr}\end{aligned}\ ] ] neglecting the trivial solution @xmath86 , the other interesting relation between @xmath76 and @xmath77 that comes out from the above equation is , @xmath87 two things can be noted from the above equation .
first as it was stated previously , in understanding the relation between @xmath76 and @xmath77 we do not require the vev of @xmath10 .
secondly if @xmath88 has some value comparable to @xmath40 and @xmath89 is not too high , then it is apparent from eq .
( [ vlvrsq ] ) that @xmath90 .
if the energy scale where @xmath11 and @xmath17 gets a non vanishing vev be @xmath91 then we can say that @xmath92 where @xmath93 gev .
thus this model allows left - right symmetry breaking at a much higher scale compared to the standard model symmetry breaking scale .
it is clear from the above discussions that this model works only if @xmath1-parity is broken spontaneously .
in addition , severe fine tuning is required to obtain and maintain this solution . to make the masses of @xmath12 and @xmath11 different @xmath94 , a fine tuning is required .
then the next fine tuning is required to keep the vev @xmath76 to be orders of magnitude smaller than @xmath77 .
this is the usual fine tuning required in all non - supersymmetric theories .
we can write eq .
( [ vl ] ) as @xmath95 = 0\,.\end{aligned}\ ] ] since the vev @xmath76 will be proportional to @xmath96 , a fine tuning is performed to keep @xmath97 gev .
the second fine tuning makes sure that the vev @xmath77 does not destabilize the vev of @xmath76 through radiative corrections .
in the left - right symmetric theories the left - handed fermions are doublets under @xmath5 and the right - handed fermions are doublets under @xmath6 . hence the fermion masses would require a bi - doublet higgs scalar .
following our discussions at the end of sec .
[ sbpat ] , it is clear that in the present model there are no yukawa couplings giving dirac or majorana masses to the quarks and leptons . in this model both the majorana and the dirac masses
originate from dimension-5 effective operators , given by : @xmath98 where @xmath99 , @xmath100 and @xmath101 are some heavy mass scales in the theory . in general , the mass scales appearing in the operators which contribute to the dirac masses ( @xmath99 , @xmath100 ) and the mass scales that appear in the operators contributing to the majorana masses ( @xmath101 ) will be different , since in one of them total fermion number is violated by 2 units .
when the higgs scalars @xmath12 and @xmath11 acquire vevs , the first two operators @xmath102 and @xmath103 give the quark masses : @xmath104 similarly the third and the fourth operators @xmath105 and @xmath106 contribute to the charged lepton and neutrino dirac masses : @xmath107 while the last two operators @xmath108 and @xmath109 contribute to the majorana masses for the left - handed and right - handed neutrinos respectively : @xmath110 we shall now discuss some of the possible origin of these operators and their consequences .
the see - saw masses of the neutrinos in theories with only doublet higgs may arise from various cases as , some higher dimensional effective operators in a non supersymmetric theory , from non - renormalizable gravitational interactions or from supersymmetric extensions of models with doublet higgs @xcite .
in the present case the see - saw masses of the neutrinos can be obtained in three different ways .
they may be mediated by exchange of scalar fields or fermion fields or may be induced radiatively .
as we shall argue now , the first two possibilities are not very attractive and hence we shall study the radiative mechanism in details . when the intermediate field is a scalar , it has to be a field which transforms as @xmath111 and hence the field could be either a @xmath112 or a @xmath113 or a @xmath114 . if the scalar field transform as @xmath113 , the fermion mass matrix will be totally antisymmetric and hence phenomenologically unacceptable .
if the scalar field @xmath115 transform as a @xmath116 or a @xmath114 , its components will receive induced vevs through its couplings @xmath117 and @xmath118 .
then we can eliminate the @xmath12 and @xmath11 in the resulting theory and revert to the conventional theories with bi - doublet higgs @xmath59 and triplet higgs scalars @xmath62 .
so , we shall not discuss this possibility any further in the rest of the article .
we shall now consider the possibility of intermediate heavy fermions generating the effective operators for the quark and lepton masses .
for each of the operators we require two fermions , one left - handed and the other right - handed , both having same gauge transformation properties . for the majorana mass terms generated by the last two operators a self - conjugate singlet fermion is sufficient .
the singlet fermion @xmath119 , we already included in the present model , can give the majorana masses to the left - handed and right - handed neutrinos .
to generate the operator @xmath102 , we need two fermions @xmath120 and @xmath121 coupling to @xmath122 and @xmath123 respectively .
both these fields should then transform similarly @xmath124 or * 210 * and the lagrangian must contain the couplings @xmath125 to give masses to the up - quarks by the operator @xmath102 .
the down quark masses are obtained by an effective operator @xmath126 , which may be generated by adding the field @xmath127 or * 126 * or @xmath128 and introducing the couplings in the lagrangian : @xmath129 the operators @xmath130 and @xmath106 may be obtained by introducing the fields @xmath131 or * 45 * and @xmath132 with the couplings @xmath133 respectively .
then we may give masses to the up and the down quarks as well as to the charged leptons and the neutrinos if there are heave fermions transforming as * 120 * and * 45*. the singlet field @xmath119 per generation is required to give majorana masses to the neutrinos with its couplings , which we shall discuss later .
we shall now come back to the present model , where the quark and lepton masses are generated radiatively .
the fermion content of the model has been discussed in sec .
[ sbpat ] .
the most general yukawa couplings are then given by , @xmath134 in this expression generation indices have been suppressed .
one loop diagram of fig .
[ fmass1.f ] then generates effective operators @xmath135 which are of the form @xmath102 and @xmath130 and contributes to the down - quark and charged - lepton masses . on the other hand the one loop diagram of fig .
[ fmass2.f ] generates effective terms : @xmath136 which are of the form of the operators @xmath103 and @xmath137 and contributes to the masses of the up - quarks and the dirac masses of the neutrinos .
the up - quark , down - quark and charged - lepton masses can now be estimated from fig .
[ fmass1.f ] and fig .
[ fmass2.f ] to be : @xmath138 here @xmath139 or @xmath140 , depending on whether @xmath141 or @xmath140 is larger and @xmath142 , and @xmath143 we thus obtain different up and down quark masses and on the other hand @xmath144 unification .
the other mass relations in the down - quark sector and the charged - lepton mass relations could come from higher order terms , since the remaining matrix elements are of the order of @xmath145 to @xmath146 compared to the 33-element @xcite .
for example , operators of the form @xmath147 contribute differently to the down - quark and charged - lepton masses , since the effective vev transform as @xmath148 and @xmath149 , which behaves as the field @xmath150 and hence can solve the fermion mass problem in guts , a la georgi - jarlskog mechanism .
the neutrino masses come from the couplings of the neutrinos with the singlet fermions @xmath119 , given by eq .
( [ yukawac ] ) . in the basis
@xmath151 the tree level neutrino mass matrix becomes : @xmath152 which gives two heavy states , which are mostly @xmath153 and @xmath154 . in the limit @xmath155 , the two heavy mass eigenvalues are @xmath140 and @xmath156 . on the other hand ,
when @xmath157 , the two heavy states are almost degenerate with eigenvalues @xmath158 with a mass splitting of about @xmath140 .
the latter case may be more interesting for leptogenesis , which we shall discuss at the end .
the lightest state @xmath159 remains massless at the tree level .
however , if we include the effect of @xmath1-parity violation , this problem could be solved .
we thus continue our discussion taking @xmath1-parity violation into consideration . the effective operator : @xmath160 and a similar @xmath1-parity violating effective operator @xmath161 which could come from the fig . [ xyfig3]a and fig .
[ xyfig3]b , together give a neutrino mass matrix : @xmath162 where , @xmath163 .
this mass matrix is obtained by integrating out the heavy modes @xmath164 . in the absence of @xmath1-parity violation , this mass matrix remains symmetrical and one of the eigenvalues vanishes , leading to a massless left - handed neutrino .
when @xmath1-parity violating effect is included , the symmetry between the left and the right handed neutrinos is lost and the left - handed neutrinos become light and massless . in the limit @xmath165 and @xmath166 , diagonalization of this matrix gives a light neutrino with mass @xmath167 this gives the correct order of magnitude for neutrino mass for @xmath168 gev and @xmath169 gev .
this tiny neutrino mass is of the see - saw type and in fact all fermion masses are of the see - saw type in this model .
in this section we shall study the renormalization group equations for the evolution of the coupling constants in our model .
we start with the one - loop renormalization group equation for the gauge coupling constants @xmath170 where @xmath171 where @xmath172 stands for the energy - scale of our theory .
@xmath173 is the gauge coupling constant of the group @xmath174 which is a subgroup of the semi - simple gauge group @xmath175 and the beta functions @xmath176 contain contributions from gauge bosons , fermions and scalars as : @xmath177.\end{aligned}\ ] ] to two - loop the @xmath178 functions of any semi - simple gauge group is given as @xcite : @xmath179 where the @xmath180s and @xmath181s are the one - loop and two - loop @xmath178 function coefficients respectively . here
@xmath182 is the number of groups whose direct product is the semi - simple gauge group of the theory , @xmath183 takes on values from @xmath184 .
first we concentrate on the one - loop effect and later we will see the effects of the two - loop coefficients on the gauge coupling evolutions . the @xmath180s calculated for the various phases are supplied below @xcite . @xmath185 in the above table @xmath186 and the superscripts @xmath187 , @xmath188 , @xmath189 indicates the phase in which the numbers are calculated .
the renormalization group ( rg ) equations can now be used to write down the the standard model gauge couplings in terms of the so(10 ) coupling .
writing @xmath190 , the gauge coupling constant matching conditions at the @xmath91 scale are : @xmath191_{{\rm g}_{\rm sm}}&= & \left[\frac35\frac{1}{\alpha_{2r}(m_r ) } + \frac23 \frac{1}{\alpha_{b - l}(m_r)}\right]_{{\rm g}_{3221 } } , \\
\left[\frac{1}{\alpha_{2l}(m_r)}\right]_{{\rm g}_{\rm sm}}&= & \left[\frac{1}{\alpha_{2l}(m_r)}\right]_{{\rm g}_{3221 } } , \\ \left[\frac{1}{\alpha_{2l}(m_r)}\right]_{{\rm g}_{\rm
sm}}&= & \left[\frac{1}{\alpha_{2r}(m_r)}\right]_{{\rm g}_{3221 } } , \\ \left[\frac{1}{\alpha_{3c}(m_r)}\right]_{{\rm g}_{\rm sm}}&= & \left[\frac{1}{\alpha_{3c}(m_r)}\right]_{{\rm g}_{3221 } } .
\label{mmr}\end{aligned}\ ] ] the matching conditions at the @xmath69 scale are : @xmath192_{{\rm g}_{3221}}&= & \left[\frac{1}{\alpha_{4c}(m_r)}\right]_{{\rm g}_{422 } } , \\
\left[\frac{1}{\alpha_{2l}(m_r)}\right]_{{\rm g}_{3221}}&= & \left[\frac{1}{\alpha_{2l}(m_r)}\right]_{{\rm g}_{422 } } , \\
\left[\frac{1}{\alpha_{2r}(m_r)}\right]_{{\rm g}_{3221}}&= & \left[\frac{1}{\alpha_{2r}(m_r)}\right]_{{\rm g}_{422 } } , \\
\left[\frac{1}{\alpha_{3c}(m_r)}\right]_{{\rm g}_{3221}}&= & \left[\frac{1}{\alpha_{4c}(m_r)}\right]_{{\rm g}_{422 } } .
\label{mmx}\end{aligned}\ ] ] finally at the @xmath40 scale , @xmath193_{{\rm g}_{422}}&= & \left[\frac{1}{\alpha_{10}(m_u)}\right]_{{\rm so}_{10 } } , \\
\left[\frac{1}{\alpha_{2l}(m_u)}\right]_{{\rm g}_{422}}&= & \left[\frac{1}{\alpha_{10}(m_u)}\right]_{{\rm so}_{10 } } , \\
\left[\frac{1}{\alpha_{2r}(m_u)}\right]_{{\rm g}_{422}}&= & \left[\frac{1}{\alpha_{10}(m_u)}\right]_{{\rm so}_{10 } } .
\label{mmu}\end{aligned}\ ] ] with the help of the above matching conditions and the rg equation we can write to one - loop , @xmath194\ , , \label{aly}\\ \frac{1}{\alpha_{2l}(m_z)}&=&\frac{1}{\alpha_{10}(m_u ) } + 8\pi\left [ a^{(sm)}_{2l}\ln\left(\frac{m_r}{m_z}\right ) + a^{(lr)}_{2l } \ln\left(\frac{m_x}{m_r}\right)\right.\nonumber\\ & & + \left.a^{(x)}_{2l}\ln\left(\frac{m_u}{m_x}\right)\right ] , \label{al2l}\\ \frac{1}{\alpha_{3c}(m_z)}&=&\frac{1}{\alpha_{10}(m_u ) } + 8\pi\left [ a^{(sm)}_{3c}\ln\left(\frac{m_r}{m_z}\right ) + a^{(lr)}_{3c } \ln\left(\frac{m_x}{m_r}\right)\right.\nonumber\\ & & + \left.a^{(x)}_{4c}\ln\left(\frac{m_u}{m_x}\right)\right ] .
\label{al3}\end{aligned}\ ] ] the linear combinations of the gauge couplings that yields @xmath195 and @xmath196 are the following : @xmath197 where @xmath198 and @xmath196 are related to the electromagnetic and strong interaction coupling constants in the present symmetry broken phase . using the experimental numbers @xcite , @xmath199 eq .
( [ wang ] ) and eq .
( [ as ] ) reduces to the following : @xmath200 the above equations can be utilized for calculating the intermediate scales like @xmath91 and @xmath69 in our theory . here
we discuss two cases . in this case from eq .
( [ aly ] ) , eq .
( [ al2l ] ) and eq .
( [ cons1 ] ) and using the @xmath178 function coefficients given in the last table we get , @xmath202=29.66\,.\nonumber\\ \label{mumx1}\end{aligned}\ ] ] similarly from eq .
( [ aly ] ) , eq .
( [ al2l ] ) , eq .
( [ al3 ] ) and eq .
( [ cons2 ] ) and the @xmath178 function coefficients we get , @xmath203=106.444\ , .
\label{mumx2}\end{aligned}\ ] ] eliminating @xmath40 from the above two equations we get @xmath204 and if we take @xmath205 gev then @xmath206 gev . the above value of @xmath91
can be taken as the lowest possible value of it in our model and all the predictions in our model will be made assuming @xmath207 gev . in this case
the two equations corresponding to eq .
( [ mumx1 ] ) and eq .
( [ mumx2 ] ) are : @xmath209=29.66\ , , \label{munmx1}\end{aligned}\ ] ] and @xmath210=106.444\ , .
\label{munmx2}\end{aligned}\ ] ] eliminating @xmath40 from the above two equations gives us a relation between @xmath69 and @xmath91 as , @xmath211\ , .
\label{mrmxp}\end{aligned}\ ] ] from the above equation it can be verified that if we take @xmath205 gev and impose @xmath212 then @xmath213 gev . in the next subsection
we include the two loop results and the above results are re - derived computationally . from the computational results we see that the above value of @xmath214 gev is two orders of magnitude smaller than the actual one .
after the discussion on gauge coupling unification to one - loop we discuss about the two - loop effects of the rg equations . to two - loop
the @xmath178 functions are given in eq .
( [ gbeta ] ) .
the @xmath180s for the various phases has been supplied in the table appearing in the beginning of this section and the @xmath215s for the various phases are as follows : @xmath216 here @xmath217 , @xmath218 and here @xmath219 , @xmath220 where @xmath221 .
s in the sm phase , left - right phase upto @xmath40 .
the abscissa is @xmath222 , where @xmath172 initial is @xmath223 gev and @xmath224 gev in our case .
the figure in the right shows explicitly the gauge coupling unification at @xmath225 gev.,height=623 ] if we start from @xmath226 gev ( where @xmath227 ) and fix @xmath224 gev then the evolution of the gauge coupling constants are as given in fig .
[ gcsm : f ] . in the next phases the coupling constant evolution shows that at @xmath69 both @xmath228 and @xmath229 unite to produce @xmath230 . fig .
[ gcsm : f ] shows that from @xmath69 on wards the development of @xmath231 and @xmath232 are identical . at around @xmath225 gev the gauge coupling constants unite .
our computational results show that the highest value of @xmath91 is just slightly above @xmath3 gev .
if @xmath91 is much above the above mentioned value then @xmath69 comes down and @xmath233 . in models with triplet higgs scalars ,
it is possible to consider @xmath234 . at the scale @xmath91 , the group @xmath6 is broken into @xmath235 and stays orthogonal to @xmath236 .
later when @xmath236 is broken to @xmath237 .
subsequently at a much lower scale the symmetry breaking , @xmath238 , takes place .
however , this is not possible in scenarios with only doublet higgs scalars , since the right - handed higgs scalar doublet does not have any component with @xmath239 quantum number to be zero and hence it breaks @xmath6 and @xmath240 simultaneously . as a result , the highest value of @xmath91 could be @xmath3 gev .
as we discussed earlier @xcite , the gauge coupling unification also requires a lower bound on @xmath241 gev . from this
lower ( higher ) bound for @xmath91 , we estimated computationally higher ( lower ) bound for @xmath69 and @xmath40 as shown in figs .
[ bound1 ] , [ bound2 ] , [ bound3 ] .
the lower and higher bounds for @xmath69 are @xmath242 gev and @xmath243 gev respectively . also lower and higher bounds for @xmath40
are estimated as @xmath244 gev and @xmath245 gev respectively . in this regard
it is important to note that the stability of the proton offers further constraints on the parameter space of their model , since the gut scale is lower than most of the conventional models and is smaller than @xmath246gev .
the difference between @xmath69 and @xmath91 can go up from 0 ( i.e , @xmath212 ; no intermediate @xmath247 symmetry ) to @xmath248 gev as shown in fig .
[ bound4 ] .
this range of @xmath91 plays an important role for yukawa coupling unification and leptogenesis in this model .
in sec . [ fmass ] , eq .
( [ yukawac ] ) gives the only yukawa couplings of our theory .
( [ yukawac ] ) is valid in the so(10 ) level . in the @xmath14 level the yukawa couplings will become :
@xmath249 the gauge transformation properties of the various fields , except @xmath153 which is a gauge singlet , are shown below : @xmath250 in the above equations @xmath251 designates the transformation properties under @xmath14 .
the yukawa couplings in the @xmath23 phase is as : @xmath252 the gauge transformation properties of @xmath253 , @xmath254 , @xmath255 , @xmath256 , @xmath19 and @xmath11 are specified in eq .
( [ lql ] ) , eq .
( [ rql ] ) and eq .
( [ higgsconj ] ) . here
we specify the gauge transformation properties of the other two fields present in the above equation .
@xmath257 in the above equations @xmath258 designates the @xmath259 transformation properties and @xmath260 designates the @xmath14 transformation properties . in the next stage , that is in the @xmath261 phase
, the yukawa couplings look like : @xmath262 here the various standard model fermion fields transform under @xmath261 as : @xmath263 similarly the various higgs fields transform as : @xmath264 in the above expressions the first triplet @xmath265 designates the transformation properties under @xmath261 , the next four numbers @xmath258 designates the @xmath23 transformation properties and the triplet @xmath260 stands for the @xmath14 transformation properties .
now we give an order of magnitude estimation about the running of the effective top yukawa coupling in our theory . from eq .
( [ mup ] ) it is seen that the effective top quark yukawa coupling looks like @xmath266 . in the @xmath267 phase it looks like @xmath268 in the convention adopted to name the yukawa couplings in eq .
( [ 321 ] ) .
calling this effective coupling as @xmath269 it will evolve simply like : @xmath270 as in the standard - model up to the @xmath91 scale .
the gauge and quartic coupling contributions will be negligible compared to @xmath269 .
starting from the top - quark mass at the electroweak scale , the evolution equation gives the effective top - quark yukawa coupling at the left - right symmetry breaking scale to be of the order of @xmath271 for @xmath272 gev .
since the effective coupling constant is a product of three couplings @xmath268 and if @xmath273 then each of these couplings can individually take values large enough as @xmath274 . as a result the yukawa sector becomes non - perturbative in the @xmath259 phase .
but on the other hand if we have @xmath275 then the situation changes . in this case
the individual coupling becomes of the order of @xmath276 which implies the theory is still perturbative .
@xmath69 can be the heavy gauge boson masses or the singlet fermion @xmath153 mass .
the previous condition means that the heavy gauge bosons or the singlet fermion can not be heavier than @xmath277 gev in our theory if we take @xmath272 gev for a perturbative scenario in the yukawa sector . above the left - right symmetry breaking scale @xmath91 up to the unification scale @xmath40 , the coupling constants @xmath278 and
@xmath279 will evolve separately .
the separate yukawa couplings remains finite up to the unification scale .
since the neutrino masses now depend on the couplings with the singlets , there is no stringent restriction coming from the up quark masses . as a result
, it may be possible to get large neutrino mixing angles .
the right - handed neutrinos and the new singlet fermions can now decay into light leptons .
the majorana masses of the left - handed and right - handed singlets violate lepton numbers , which in turn can generate enough lepton asymmetry . before the electroweak phase transition
this asymmetry can then generate a baryon asymmetry of the universe @xcite . since there is no supersymmetry , the gravitino bounds are not present .
the out - of - equilibrium condition can be satisfied near the gut scale since the couplings are large to get the required neutrino mass with large see - saw scale . in this model
there is another interesting feature that the singlets combine with the right - handed neutrinos to form pseudo - dirac particles and hence resonant leptogenesis may also be possible @xcite . for leptogenesis ,
consider the interactions of eq .
( [ yukawac ] ) . unlike usual see - saw models with triplet higgs scalars @xcite , in this model
the right - handed neutrinos can not decay into left - handed neutrinos and higgs bi - doublets dominantly .
the simplest lepton number violating interactions come from the decays of @xmath164 : @xmath280 the majorana masses of @xmath164 allow the singlet to decay into both leptons and antileptons violating lepton number by two units .
in the present model both @xmath254 and @xmath12 are very light and hence these decays are allowed . for cp violation
there are two types of one - loop diagrams which interferes with the tree - level diagrams for the decays of @xmath164 .
these are the vertex type diagrams of fig .
[ vertex ] and fig .
[ selfenergy ] .
the right - handed neutrinos do not take part in leptogenesis directly , but due to mixing of the right - handed neutrinos with the heavy singlets @xmath164 , the right - handed neutrino decays also enter in the picture of leptogenesis . in the limit of @xmath155 , the right - handed neutrinos and the @xmath164 are both heavy and distinct . in this case
the amount lepton asymmetry due to cp violation is given by , @xmath281 } { \sum_\alpha @xmath164 are diagonal and the eigenvalues are hierarchical @xmath282 where @xmath283 are masses of @xmath164 .
we can write the effective lepton asymmetry as : @xmath284 where @xmath285 is a suppression factor which depends on the amount of departure from equilibrium . the exact value of @xmath285 can be obtained by solving the boltzmann equation taking all the interactions into consideration .
however , it is also possible to make an order of magnitude estimate for the amount of asymmetry , which will be very close to the actual value .
since @xmath286 is the lightest of the singlets , the decay of this singlet will be able to generate the lepton asymmetry . the asymmetry generated or washed out by the heavier ones
@xmath287 or @xmath288 will be smeared out by the interactions of @xmath286 after @xmath287 and @xmath288 had decayed away .
so , for an estimate we shall only consider the decays of @xmath286 .
this assumption is well justified when the singlets @xmath164 have a hierarchical mass structure .
the out - of - equilibrium condition is parametrized by : @xmath289 when @xmath290 , there is no boltzmann suppression of the generated asymmetry and the out - of - equilibrium condition is satisfied . in this case
the generated asymmetry is given by @xmath291 and it is not washed out after it is created and one gets @xmath292 . however ,
if @xmath293 , then the interaction strength is so slow that the generated asymmetry can never reach the value @xmath291 .
although the interactions can not wash out the asymmetry after it is generated , the amount of asymmetry is less than @xmath291 . in the case of @xmath294 ,
the generated asymmetry is same as the cp asymmetry of @xmath291 , but even after the asymmetry is created , the interaction remains strong enough to deplete the asymmetry .
although the depletion is exponentially fast , it can not compete with the expansion of the universe for long and the final amount of asymmetry is not exponentially depleted .
it was shown that @xcite the suppression factor @xmath285 is almost linearly proportional to @xmath295 . in the present model
we come across this last scenario . in the present model @xmath296 gev . while a lower value of @xmath91 is preferable for out - of - equilibrium condition , since the yukawa couplings grow very fast above the scale @xmath91 we have to consider the highest value of @xmath91 . taking the hierarchical structure of @xmath164
, we consider the mass of the lightest singlet to be around @xmath297 gev . taking @xmath298 gev , @xmath299 and @xmath300
we find that @xmath295 is much lower than 1 , which gives a strong suppression factor of @xmath301 . on the other hand
the yukawa couplings in this model comes out to be of the order of 1 and hence we get a large enhancement in the cp asymmetry and @xmath291 in our case can be as large as @xmath302 and so the lepton asymmetry parameter @xmath303 . at this stage ,
@xmath304 and @xmath305 is given by @xmath291 .
thus @xmath306 the final baryon asymmetry after the electroweak phase transition is thus given by , @xmath307 in the case of @xmath157 , the right - handed neutrinos and the @xmath164 singlets of every generation are almost degenerate .
the mass splitting between the states @xmath308 and @xmath164 with mass @xmath309 is of the order of @xmath140 .
although @xmath164 decays will now generate a lepton asymmetry , both the heavy mass eigenstates contain the states @xmath164 . as a result ,
when these two almost degenerate states decay , there may be resonant leptogenesis ( which will however require new interactions and fine - tuning ) and hence the scale of leptogenesis could be very low . for the present scenario since @xmath91 and hence @xmath310 can not be much lower , this is not important and hence we shall not discuss it in any further detail .
in conclusion , we constructed an so(10 ) gut without any higgs bi - doublets .
all the symmetry breaking could be achieved by only two higgs scalars , a @xmath32 and a @xmath311 . by including a massive singlet fermion per generation we break chiral symmetry which can then give masses to all the fermions radiatively without introducing any new scalar fields .
all fermion masses have the same see - saw form . the spontaneous parity breaking plays a crucial role in breaking the left - handed and right - handed su(2 ) groups at two widely different scales and also giving masses to the left - handed neutrinos in this scenario . the spontaneous breaking of an ungauged discrete symmetry , the d - parity , which is a special feature of this model may cause formation of very heavy domain walls of gut - scale mass . since the gut scale in the model is rather low @xmath312 gev )
, some of the parameters of the model may have to be constrained further to prevent fast proton decay .
these features require further study and will be taken up in future .
the model allows large neutrino mixing and required neutrino masses .
the baryon asymmetry of the universe can be explained through leptogenesis .
gr acknowledges the support of the raja ramanna fellowship of the department of atomic energy , government of india and the hospitality of the physics department , university of california , riverside .
brd acknowledges the support in part by the u.s .
department of energy under grant no .
de - fg03 - 94er40837 .
b. bajc , g. senjanovi and f. vissani , phys.rev.lett .
* 90 * , 051802 ( 2003 ) ; + h.s .
goh , r.n .
mohapatra and s.p .
ng , phys.lett . * b570 * , 215 ( 2003 ) and + phys.rev . * d68 * , 115008 ( 2003 ) ; + s. bertolini , m. frigerio and m. malinsky , phys.rev . *
d70 * , 095002 ( 2004 ) .
t. clark , t. kuo and n. nakagawa , phys.lett .
* b115 * , 26 ( 1982 ) ; + c.s .
aulakh and r.n .
mohapatra , phys.rev . * d28 * , 217 ( 1983 ) ; + c.s .
aulakh , b. bajc , a. melfo , g. senjanovi and f. vissani , phys.lett .
* b588 * , 196 ( 2004 ) ; + t. fukuyama , a. ilakovic , t. kikuchi , s. meljanac and n. okada , eur.phys.j . * c42 * , 191 ( 2005 ) and + j.math.phys . *
46 * , 033505 ( 2005 ) ; + c.s .
aulakh and a. giridhar , nucl.phys . *
b711 * , 275 ( 2005 ) .
s. rajpoot , phys.rev . *
d22 * , 2244 ( 1980 ) ; + r.n .
mohapatra , phys.rev . *
d34 * , 3457 ( 1986 ) ; + a. font , l.e .
ibez and f. quevedo , phys.lett . *
b228 * , 79 ( 1989 ) ; + s.p .
martin , phys.rev . *
d46 * , 2769 ( 1992 ) ; + c.s .
aulakh , k. benakli and g. senjanovi , phys.rev.lett . * 79 * , 2188 ( 1997 ) ; c.s .
aulakh , a. melfo , a. rain and g. senjanovi , phys.lett . *
b459 * , 557 ( 1999 ) and + nucl.phys . *
b597 * , 89 ( 2001 ) ; + t. hambye and g. senjanovi , phys.lett . *
b582 * , 73 ( 2004 ) ; + g. dambrosio , t. hambye , a. hektor , m. raidal and a. rossi , phys.lett . *
b604 * , 199 ( 2004 ) .
d. chang and r.n .
mohapatra , phys.rev.lett . * 58 * , 1600 ( 1987 ) ; + b.s .
balakrishna , phys.rev.lett .
* 60 * , 1602 ( 1988 ) ; + k.s . babu and r.n .
mohapatra , phys.rev.lett . * 62 * , 1079 ( 1989 ) and + phys.rev . * d41 * , 1286 ( 1990 ) . m. flanz , e.a
. paschos , and u. sarkar , phys.lett
. * b345 * , 248 ( 1995 ) ; + l. covi , e. roulet and f. vissani , phys.lett . *
b384 * , 169 ( 1996 ) ; + m. flanz , e.a .
paschos , u. sarkar and j. weiss , phys.lett . *
b389 * , 693 ( 1996 ) ; + a. pilaftsis , nucl.phys . *
b504 * , 61 ( 1997 ) and + phys.rev . *
d56 * , 5431 ( 1997 ) ; + w. buchmller and m. plmacher , phys.lett . *
b431 * , 354 ( 1998 ) ; + a. pilaftsis , int.j.mod.phys .
* a14 * , 1811 ( 1999 ) ; + a. pilaftsis and t.e.j .
underwood , nucl.phys .
* b692 * , 303 ( 2004 ) ; + t. hambye , j. march - russell and s.m . west , jhep * 0407 * , 070 ( 2004 ) .
e. ma and u. sarkar , phys.rev.lett .
* 80 * , 5716 ( 1998 ) ; + g. lazarides and q. shafi , phys.rev . * d58 * , 071702 ( 1998 ) ; + w. grimus , r. pfeiffer and t. schwetz , eur.phys.j . * c13 * , 125 ( 2000 ) ; + e. ma , m. raidal and u. sarkar , phys.rev.lett .
* 85 * , 3769 ( 2000 ) and + nucl.phys .
* b615 * , 313 ( 2001 ) ; + t. hambye , e. ma and u. sarkar , nucl.phys . *
b602 * , 23 ( 2001 ) ; + t. hambye , m. raidal and a. strumia , phys.lett . *
b632 * , 667 ( 2006 ) .
a. acker , h. kikuchi , e. ma and u. sarkar , phys.rev .
* d48 * , 5006 ( 1993 ) ; + p.j .
odonnell and u. sarkar , phys.rev . *
d49 * , 2118 ( 1994 ) ; + m. plumacher , z.phys .
* c74 * , 549 ( 1997 ) ; + t. hambye and g. senjanovi , phys.lett . * b582 * , 73 ( 2004 ) ; + s. antusch and s.f . king , phys.lett . * b597 * , 199 ( 2004 ) ; + narendra sahu and s. uma sankar , nucl.phys . *
b724 * , 329 ( 2005 ) ; + k.s . babu , a. bachri and h. aissaoui , nucl.phys . *
b738 * , 76 ( 2006 ) . | in this work we study an so(10 ) gut model with minimum higgs representations belonging only to the * 210 * and * 16 * dimensional representations of so(10 ) .
we add a singlet fermion @xmath0 in addition to the usual * 16 * dimensional representation containing quarks and leptons
. there are no higgs bi - doublets and so charged fermion masses come from one - loop corrections .
consequently all the fermion masses , dirac and majorana , are of the see - saw type .
we minimize the higgs potential and show how the left - right symmetry is broken in our model where it is assumed that a @xmath1-parity odd higgs field gets a vacuum expectation value at the grand unification scale . from the renormalization group equations we infer that in our model unification happens at @xmath2 gev and left - right symmetry can be extended up to some values just above @xmath3 gev .
the yukawa sector of our model is completely different from most of the standard grand unified theories and we explicitly show how the yukawa sector will look like in the different phases and briefly comment on the running of the top quark mass .
we end with a brief analysis of lepton number asymmetry generated from the interactions in our model . |
dihadron azimuthal correlation has been a successful tool in understanding the interactions between jet and medium , and in extracting the properties of the sqgp . over the years
, the correlation analyses have been carried out in various regions of transverse momentum ( @xmath0 ) for the triggers and partners .
many interesting features have been discovered . in the high @xmath0 region ,
the correlation distributions show narrow peaks around @xmath1 ( near - side ) and @xmath2 ( away - side ) @xcite , consistent with fragmentation of jets escaping the dense medium with small energy loss . in the low @xmath0 region ,
the correlation distributions are dominated by a double hump structure around @xmath3 at the away - side ( the cone ) @xcite and a structure elongated along the @xmath4 but centered around @xmath1 ( the ridge ) @xcite , characteristic of a complicated response of the medium to energy deposited by the quenched jets . in the meanwhile , many theoretical models @xcite have been proposed to interpret the data .
but to date , a complete and consistent picture accommodating the vast amount data is still missing .
our goal is to provide a brief overview of the dihadron correlation results , with an eye towards the reciprocal relation between jet quenching and medium response , and discuss several insights distilled from the data .
in general the dihadron correlations depend on the @xmath0 of both hadrons in the pair , and the full characterization of their modification patterns have to be studied differentially as function of trigger @xmath0 ( @xmath5 ) and
partner @xmath0 ( @xmath6 ) .
such a survey study has been carried out recently by the phenix @xcite and star collaboration @xcite .
[ fig:1a ] summarizes dihadron @xmath7 distribution in a broad transverse momentum space , which shows many distinctive features appearing at different @xmath0 regions ( indicated by the circles and lines ) .
these features fits well into a simple two - component picture as illustrated in fig .
[ fig:1b ] separately for both the near- and away - side : a jet fragmentation component that dominates for @xmath8 gev/@xmath9 , and a medium response component that dominates at @xmath10 gev/@xmath9 .
the rich @xmath0 dependent correlation patterns simply reflect the competition between fragmentation of survived jets and medium response to quenched jets on both the near- and away - side .
the observed patterns are rather complicated in fig .
[ fig:1a ] since 1 ) the medium response and jet fragmentation have very different angular distribution and very different spectral slope , 2 ) the shapes of the medium response are also quite different between the near- and away - side .
a new variable @xmath11 was introduced recently to describe the medium response at low @xmath0 @xcite .
@xmath11 quantify the medium modification of hadron pair yield from the expected yield , in a way similar to @xmath12 for describing the modification of single hadron yield .
the hadron pair yield is proportional to the dijet yield , and in the absence of nuclear effects , it should scale with @xmath13 , and @xmath14 .
[ fig:2 ] shows @xmath11 as a function of pair proxy energy ( @xmath15 ) for the near- ( left panel ) and away - side ( right panel ) .
the star autocorrelation result @xcite is shown as a single point at @xmath16 gev/@xmath9 .
in contrast to a constant suppression at large @xmath17 , the pair yields are not suppressed or even enhanced at @xmath18 gev/@xmath9 .
this enhancement directly reflects the energy transport that redistribute energy of the quenched jets to low @xmath0 hadrons ( medium response ) .
we would like to point out that @xmath17 is a natural variable for the near - side correlation since it approximates the jet energy , and data show an approximate scaling in @xmath17 .
however , even the data points for the away - side tend to group together , probably because the medium response is a function of jet energy , which increase monotonously with @xmath19 .
the transition from jet fragmentation dominated to medium dominated region in dihadron correlation happens around @xmath20 gev/@xmath9 , a region similar to that for the single particle from soft physics ( hydrodynamics+ recombination ) dominated region to hard physics ( jet ) dominated region .
naturally , we expect the physics important for single particle production should play an important role for the dihadron correlation .
[ fig:3 ] shows schematically the @xmath0 dependence of the modifications of the single particle yield ( via @xmath12 ) and hadron pair yield ( via @xmath11 ) .
their @xmath0 dependence trend are quite different , especially at low @xmath0 , which can be explained qualitatively as follows .
even though jet production dominates single particle yield at @xmath21 gev/@xmath9 in @xmath22 collisions , the strong energy loss and collective flow modify the @xmath0 distribution by shifting hard hadrons to lower @xmath0 and pushing soft hadrons to higher @xmath0 .
this reshuffling changes single - hadron and correlated hadron - pair yield , hence the @xmath12 and @xmath11 shape .
indeed , several theoretical models suggest that collective flow and recombination play a significant role in modifying the angular shape , spectra slope and particle composition of the correlated pairs @xcite .
@xmath11 provides a mean to quantify the contribution of jet fragmentation hadrons or jet induced hadrons in this @xmath0 region .
previously , the modification of dihadron yield is characterized with @xmath23 ( ratio of per - trigger yield between au+au and @xmath22 ) @xcite .
@xmath23 is a good variable at high @xmath0 , since most triggers come from jets and most jets fragment into at most one trigger , such that per - trigger yield ( pty ) is a good representation of per - jet yield .
however at lower @xmath0 region , non - fragmentation triggers from soft production mechanisms or medium response mechanisms become important .
these triggers tend to dilutes the @xmath23 , since they either has no correlation or non - jet like correlation ( such as ridge ) .
[ fig:4 ] illustrate the dilution effects for near - side @xmath4 correlation .
we estimate dilution factor ( @xmath24 ) for 3 - 4 gev/@xmath9 triggers based on their correlations with 5 - 10 gev/@xmath9 hadrons as shown by the inserted panel : requiring 5 - 10 gev/@xmath9 hadrons ensures the pairs are dominated by the jet fragmentation ( left panel of the insert ) , thus deviation of @xmath23 from one for soft triggers reflects the level of dilution ( the red arrow ) .
once the dilution factor is corrected , we subtract out the jet fragmentation contribution and obtain the ridge distribution ( black circles ) .
the estimated ridge contribution is approximately flat , consistent experimental data at large @xmath4 .
however , this dilution effect was not observed in some star analysis @xcite , which shows that the pty@xmath25 subtracted by the estimated ridge before any dilution correction already equals pty@xmath26 .
in most correlation analyses and model calculations , it was normally assumed that one hadron ( `` trigger '' ) comes from the jet , and the second hadron ( `` partner '' ) comes from either fragmentation or feedback , i.e. only jet - jet and jet - medium pairs are considered . in this picture , the trigger comes from a jet that is biased to the surface , which losses some energy and fragments outside the medium .
the fragments contribute to the near - side jet peak , and the feedback of the lost energy gives rise to the near - side ridge . in parallel
, the away - side jet is quenched as it traverses a longer medium , contributing to the away - side cone .
this picture does not include the medium - medium pairs ( both hadrons come from medium feedback of quenched jets ) .
these pairs could be important at intermediate and low @xmath0 , since each jet can induce more correlated hadrons via jet quenching than via fragmentation .
for example , most medium response models induce correlation by local heating of medium by the jet , such as momentum kick , jet deflection , mach - cone , etc @xcite , which are very effective in generating large yield of correlated hadron pairs . in addition
, the whole overlap volume contributes to the observed medium - medium pairs , while both jet - jet and jet - medium pairs suffer a strong suppression .
this point is illustrated by fig .
[ fig:5 ] , which shows the typical geometrical origin for the three types of correlated pairs .
the jet fragmentation contribution is proportional to the number of survived jet ( @xmath27 , i.e. the constant suppression level at large @xmath0 , @xmath28 in most central bin . ) , while the medium response is proportional to the number of quenched jets ( @xmath29 ) . for jet - jet pairs , both hadron
are emitted from the surface ( tangential emission ) ; for jet - medium pairs , the jet hadron is emitted from the surface ( surface emission ) and the other from the whole volume ; for medium - medium pairs , both hadrons are emitted from the whole volume .
the production rate for jet - jet , jet - medium and medium - medium pairs scale approximately with @xmath30 , @xmath31 and @xmath32 .
clearly , if @xmath33 , the medium - medium pairs becomes dominant .
recently , several analyses have been carried out to quantify the properties of the near - side ridge and away - side cone structures @xcite .
the data show very similar properties between the ridge and the cone , i.e. both have similar slope and bulk like particle compositions , and both are important up to 4 gev/@xmath9 .
these similarities suggests that their production mechanisms are connected .
the medium - medium pairs from quenched jets are natural candidates for creating these similarities . because medium - medium pairs come from quenched jets originated deep inside the medium , they contribute to both the near - side and away - side on a equal footing .
the near - side pairs could contain correlations among mach cone particles , and away - side pairs could also contain correlation between the ridge and mach cone particles ( see ref .
@xcite for a possible realization ) .
a strong modification of the away - side correlation was also observed at the top sps energy ( @xmath34 gev ) @xcite .
the strong away - side broadening has been used to argue for a similar interpretation ( such as mach cone ) as for the rhic results .
however a quantitative analysis of the energy dependence of the modification patterns ( see fig.[fig:6 ] ) shows that the yield of medium response are quite different between rhic and sps energies .
in fact the near - side yield drop by almost factor of 8 going from 200 gev to 17 gev while the away - side shoulder yield drops by factor of 2 in the same energy range .
but there are little dependence of the yields on @xmath35 in the away - side head region , where the jet fragmentation is important . to quantify the energy dependence of away - side shape
, we calculate the ratio of the yield density in the head region to that in shoulder region , @xmath36 @xcite , in fig.[fig:7]a . the @xmath36 increases with decreasing collision energy , with a ratio slightly above one in sps energy .
this value is comparable with that obtained for rather peripheral ( @xmath37 ) in au+au collisions at 200 gev ( fig.[fig:7]b ) .
these results suggest a much weaker medium response in sps energy ( the ridge almost disappeared and cone strongly suppressed ) than that at rhic , but a similar jet fragmentation contribution , probably related to smaller energy loss and stronger cronin effects at lower energy @xcite . cl & lastly , it was shown that the mach - cone angle found from the three particle ( 3-p ) correlation ( 1.4 rad ) is significantly larger than the two particle ( 2-p ) correlation analysis ( 1.1 rad ) .
sps also seems to see a 3-p correlation signal @xcite .
however one should realized that , it is possible that the kinematics of jets contributing to 3-p signal is different from those contributing to the 2-p signal .
it is likely that most jet have multiplicity @xmath38 that the 3-p only samples a small fraction of all jets that contribute to the dihadron correlations .
due to the surface bias and steeply falling parton spectra , the observed high @xmath0 single hadrons and dihadron pairs mainly come from those jets that suffer minimal interaction with the medium .
this energy loss bias limits their usefulness as tomography tools . on the other hand ,
medium responses are directly sensitive to the energy loss and energy dissipation processes used to model the high @xmath0 production .
for example the collisional energy loss would imply that momentum kick dominates the low @xmath0 pairs , the radiative energy loss would favor for the gluon feedback mechanism .
finally , the jet quenching and medium responses are modeled separately in most theoretical calculations . a unified framework , including both jet quenching and medium response , which can describe the full @xmath39 evolution of the jet shape and yield at both near- and away - side is required to understand the details of the parton - medium interactions .
99 j. adams _ et al . _ [ star collaboration ] , phys . rev .
lett . * 95 * , 152301 ( 2005 ) j. adams _ et al . _ [ star collaboration ] , phys .
lett . * 97 * , 162301 ( 2006 ) a. adare _ et al . _
[ phenix collaboration ] , phys .
c * 77 * , 011901(r ) ( 2008 ) a. adare _ et al .
_ [ phenix collaboration ] , arxiv:0801.4545 [ nucl - ex ] .
m. j. horner [ star collaboration ] , j. phys .
g * 34 * , s995 ( 2007 ) s. s. adler _ et al .
_ [ phenix collaboration ] , phys .
lett . * 97 * , 052301 ( 2006 ) j. adams _ et al . _ [ star collaboration ] , phys .
c * 75 * , 034901 ( 2007 ) n. armesto , c. a. salgado and u. a. wiedemann , phys . rev .
lett . * 93 * , 242301 ( 2004 ) , phys . rev .
c * 72 * , 064910 ( 2005 ) ; i. vitev , phys .
b * 630 * , 78 ( 2005 ) ; i. m. dremin , jetp lett . *
30 * , 140 ( 1979 ) v. koch , a. majumder and x. n. wang , phys .
* 96 * , 172302 ( 2006 ) j. casalderrey - solana , e. v. shuryak and d. teaney , hep - ph/0602183 .
t. renk and j. ruppert , phys . rev .
c * 73 * , 011901 ( 2006 ) ; c. b. chiu and r. c. hwa , phys .
c * 74 * , 064909 ( 2006 ) ; a. d. polosa and c. a. salgado , phys .
c * 75 * , 041901(r)(2007 ) ; p. romatschke , phys .
c * 75 * , 014901 ( 2007 ) ; a. majumder , b. muller and s. a. bass , phys . rev .
lett . * 99 * , 042301 ( 2007 ) ; c. y. wong , phys .
c * 76 * , 054908 ( 2007 ) ; e. v. shuryak , phys .
c * 76 * , 047901 ( 2007 ) ; v. s. pantuev , arxiv:0710.1882 [ hep - ph ] ; j. putschke , eur .
j. c * 49 * , 57 ( 2007 ) .
s. afanasiev _ et al . _
[ phenix collaboration ] , arxiv:0712.3033 [ nucl - ex ] .
a. frantz , [ phenix collaboration ] these proceedings .
j. jia and r. lacey , arxiv:0806.1225 [ nucl - th ] .
m. ploskon [ ceres collaboration ] , nucl .
a * 783 * , 527 ( 2007 ) a. adare _ et al . _
[ phenix collaboration ] , arxiv:0801.4555 [ nucl - ex ] .
j. g. ulery , these proceedings . | a brief review of the @xmath0 dependence of the dihadron correlations from rhic is presented .
we attempt to construct a consistent picture that can describe the data as a whole , focusing on the following important aspects , 1 ) the relation between jet fragmentation of survived jet and medium response to quenched jets , 2 ) the possible origin of the medium response and its relation to intermediate @xmath0 physics for single hadron production , 3 ) the connection between the near - side ridge and away - side cone , 4 ) and their relations to low energy results . |
anisotropic flow is an important observable in ultra - relativistic heavy - ion collisions as it signals the presence of multiple interactions between the constituents of the created matter .
anisotropic flow has been observed in nucleus
nucleus collisions from low energies up to @xmath0 tev at the large hadron collider ( lhc ) @xcite .
the azimuthal anisotropic flow is usually characterized by the fourier coefficients @xcite : @xmath1\rangle , \label{eq : harmonics}\ ] ] where @xmath2 is the azimuthal angle of the particle , @xmath3 is the azimuthal angle of the initial state spatial plane of symmetry , and @xmath4 is the order of the harmonic .
the second fourier coefficient @xmath5 is called elliptic flow @xcite . because the magnitude of the anisotropic flow depends strongly on the friction in the created matter , characterized by the shear viscosity over entropy density ratio @xmath6 , the large elliptic flow observed at rhic provided compelling evidence for strongly interacting matter which , in addition , appears to behave like an almost perfect fluid @xcite .
however , a precise determination of @xmath6 in the partonic fluid is complicated by uncertainties in the initial conditions of the collision , the relative contributions from the hadronic and partonic phase , and the unknown temperature dependence of @xmath6 .
because of these uncertainties it was not even clear if the elliptic flow would increase or decrease when going from rhic to lhc energies ; a measurement of elliptic flow at the lhc was therefore one of the most anticipated results .
in the left panel of fig . [ fig : figure1 ] the alice measurement at 2.76 tev @xcite shows that the integrated elliptic flow of charged particles increases by about 30% compared to flow measured at the highest rhic energy of 0.2 tev .
this result indicates that the hot and dense matter created in these collisions at the lhc still behaves like a fluid with almost zero friction , providing strong constraints on the temperature dependence of @xmath6 . experimentally , because the planes of symmetry @xmath3 in eq .
[ eq : harmonics ] are not known , the anisotropic flow coefficients are estimated from measured correlations between the observed particles . the right panel shows the integrated elliptic flow ( @xmath7 and @xmath8 gev/@xmath9 ) obtained from two- and multi - particle correlations as a function of collision centrality , compared to star measurements at rhic . here the elliptic flow estimated from two - particle correlations is denoted by @xmath5\{2 } , while those estimated from multi - particle correlations are denoted by @xmath5\{4 } , @xmath5\{*q*-dist } and @xmath5\{lyz } , for the four particle cumulant , fit of the flow * q*-vector distribution and the lee - yang zeros method , respectively ( see @xcite and references therein ) . there is a significant difference between flow estimates from two- and multi - particle correlations both at lhc and at rhic energies .
this difference is caused by nonflow contributions ( these are other sources of azimuthal correlations due to jets and resonance decays , for instance ) and by event - by - event fluctuations in the elliptic flow . in the left panel of fig .
[ fig : figure2 ] , the collision centrality dependence of @xmath5 is plotted in narrow bins ( 12% ) , to reduce trivial event - by - event fluctuations within a centrality bin . in this figure
the effect of the nonflow on two - particle estimates is apparent from the difference in @xmath5 calculated from particles with @xmath10 and @xmath11 .
also shown are results from four- , six- , and eight - particle cumulant estimates which are consistent within uncertainties , indicating that the genuine 4-particle , and higher order , nonflow is negligible .
the flow estimates from the cumulants can therefore be written as @xcite @xmath12 where @xmath13 is the event averaged elliptic flow , @xmath14 the standard deviation of the event - by - event fluctuations and @xmath15 the residual nonflow contribution . assuming that @xmath16 , eq .
[ eq : cumulants ] is valid up to order @xmath17 .
we use the hijing event generator ( which does not include flow ) to estimate @xmath15 .
the right panel of fig .
[ fig : figure2 ] shows @xmath18 for @xmath10 and @xmath11 before and after the estimated nonflow correction . after this correction
both estimates are in good agreement , indicating that hijing seems to correctly describe the two - particle nonflow contribution .
we currently assign the entire hijing based nonflow correction for @xmath19 as a conservative estimate of the systematic uncertainty .
using the nonflow corrected @xmath18 and @xmath20 we obtain from eq .
[ eq : cumulants ] @xmath21 the results are plotted in fig . [
fig : figure3 ] ( left ) , together with the ratio @xmath14/@xmath13 ( right ) .
the ratio @xmath14/@xmath13 is found to be large @xmath22% ; a similar result has been obtained at rhic @xcite . because the magnitude of the elliptic flow is proportional to the eccentricity @xmath23 of the initial nuclear overlap region in the transverse plane
, we expect that the event - by - event fluctuation in @xmath5 will be proportional to that in @xmath23 .
the measured ratio @xmath24 is compared in fig .
[ fig : figure3 ] to the ratio @xmath25 , calculated with eq .
[ eq : flucts ] from a mc - kln @xcite and a mc - glauber model @xcite ( full curves ) .
the mc - kln under - predicts the data whereas the mc - glauber over - predicts the data for more central collisions . to investigate to which extent the ratio obtained from eq .
[ eq : flucts ] represents @xmath25 ( and by implication @xmath26 ) we have calculated this ratio directly from the distributions generated by the two models , that is , without the assumption that @xmath27 . the result is plotted in the right panel of fig .
[ fig : figure3 ] for mc - kln ( dotted curve ) and mc - glauber ( dot - dashed curve ) and indicates , in comparison to the full curves , that eq .
[ eq : flucts ] breaks down for @xmath28 . .
the bands indicate rhic results measured by star .
figure taken from @xcite.,title="fig:",width=302 ] .
the bands indicate rhic results measured by star .
figure taken from @xcite.,title="fig:",width=302 ] to eliminate the trivial dependence of @xmath23 and @xmath5 on centrality and to increase the sensitivity to the properties of the medium it is illustrative to plot the ratio @xmath29 . in the left panel of fig .
[ fig : figure4 ] we show the ratios @xmath30 and @xmath31 with @xmath23 calculated from mc - kln or mc - glauber .
these two ratios should be identical , provided that the fluctuations are correctly described in the models , which is seen not to be the case for the more central collisions in mc - glauber . to reduce the sensitivity to fluctuations we plot
@xmath32 as defined by eq . [ eq : flucts ] ( curves in fig . [
fig : figure4 ] ) .
the mismatch between the two curves indicates a sensitivity to the initial conditions which still obscures the direct relation between @xmath29 and the properties of the medium .
elliptic flow as a function of transverse momentum @xmath33 is sensitive to the evolution and freeze - out conditions of the created system .
the right panel of fig .
[ fig : figure4 ] shows that the charged particle @xmath33-differential elliptic flow , compared to rhic , does not change within uncertainties at low @xmath33 @xcite which is remarkable because the beam energies differ by more than one order of magnitude . the 30% increase in the integrated flow , shown in fig .
[ fig : figure1 ] , must therefore be due to an increase in average transverse momentum . in hydrodynamical model calculations
this increase in mean @xmath33 is due to a larger transverse flow at higher energies .
this leads to a more pronounced mass dependence of the elliptic flow . in fig .
[ fig : figure5 ] we show identified particle @xmath33-differential elliptic flow compared to hydrodynamic model predictions . -differential elliptic flow for pions , kaons and antiprotons for 40%50% ( left ) and 10%20% ( right ) collision centrality
. the curves are hydrodynamical model calculations . figures taken from @xcite.,title="fig:",width=302 ] -differential elliptic flow for pions , kaons and antiprotons for 40%50% ( left ) and 10%20% ( right ) collision centrality .
the curves are hydrodynamical model calculations
. figures taken from @xcite.,title="fig:",width=302 ] the hydrodynamic model predictions from @xcite ( curves in fig .
[ fig : figure5 ] ) describe for mid - central collisions very well the measured @xmath5(@xmath33 ) for pions , kaons and antiprotons at low @xmath33 ( left panel of fig .
[ fig : figure5 ] ) . for more central collisions ( right panel )
the hydrodynamical model predictions well describe the flow of pions and kaons but not that of the antiprotons .
this mismatch is also observed for the spectra @xcite and may indicate a larger radial flow in the data . at rhic
this was also observed and there a better description of the antiproton flow was obtained by introducing a hadronic cascade afterburner in the calculations @xcite .
-differential elliptic flow for pions , kaons and antiprotons compared to results from rhic ( shaded bands ) and hydrodynamical calculations at rhic ( dashed curves ) and lhc ( full curves ) .
right : elliptic flow versus transverse kinetic energy both divided by the number of constituent quarks .
figures taken from @xcite . , title="fig:",width=302 ] -differential elliptic flow for pions , kaons and antiprotons compared to results from rhic ( shaded bands ) and hydrodynamical calculations at rhic ( dashed curves ) and lhc ( full curves ) .
right : elliptic flow versus transverse kinetic energy both divided by the number of constituent quarks .
figures taken from @xcite . , title="fig:",width=302 ] the energy dependence of the mass splitting at low @xmath34 gev/@xmath9 is shown in more detail in the left panel of fig .
[ fig : figure6 ] . to enable a direct comparison to the star pion and
antiproton data ( bands in the figure ) , the elliptic flow measured by alice is plotted for the centrality bin of 30%40% .
we observe a small but significant increase with energy of the mass splitting between pions and antiprotons .
an increased mass splitting is also observed in hydrodynamical calculations between rhic ( dashed curves ) and lhc energies ( full curves ) although the antiproton elliptic flow is overestimated in both these calculations .
hydrodynamics predicts that the mass splitting pattern persists at large values of @xmath33 in contrast to what is observed in the data , as is apparent from fig .
[ fig : figure5 ] above @xmath35 2 gev/@xmath9 .
an elegant explanation of the particle type dependence and magnitude of @xmath5 at larger @xmath33 is provided by the coalescence picture . at rhic
it was observed that @xmath36 showed a universal scaling when plotted versus @xmath37 .
here @xmath38 is the number of constituent quarks and @xmath39 is the transverse mass , with @xmath40 the mass of the particle .
the right panel of fig .
[ fig : figure6 ] shows this scaling at 2.76 tev for 40%-50% collision centrality measured by alice .
the data show a clear scaling for pions and kaons but not for antiprotons . -differential elliptic flow for different collision centralities @xcite .
right : identified particle elliptic flow at high-@xmath33 .
for comparison the @xmath41 data measured by phenix are also shown .
figures taken from @xcite .
, title="fig:",width=302 ] -differential elliptic flow for different collision centralities @xcite .
right : identified particle elliptic flow at high-@xmath33 .
for comparison the @xmath41 data measured by phenix are also shown .
figures taken from @xcite . , title="fig:",width=302 ] at sufficiently high transverse momentum , hadron yields are thought to contain a significant contribution from the fragmentation of high energy partons produced at initial hard scatterings .
these high energy partons are predicted to lose energy when traversing nuclear matter .
this energy loss is expected to depend strongly on the color charge density of the medium and on the path length traversed by the propagating parton . because this path length depends on the azimuthal emission angle with respect to the reaction plane ,
an azimuthal anisotropy of particle emission is introduced at large @xmath33 .
indeed , significant values of @xmath5 are found between 8 and 20 gev/@xmath9 as is shown in the left panel of fig .
[ fig : figure7 ] .
this @xmath5 increases from central to more peripheral collisions as is expected from the path length dependence of parton energy loss . to investigate where the coalescence regime stops and where the parton energy loss mechanism might become dominant ,
we show in the right panel of fig .
[ fig : figure7 ] the identified particle @xmath5 at large @xmath33 .
we see that up to about 8 gev/@xmath9 the proton @xmath5 is larger than the pion @xmath5 , as one would expect from coalescence .
due to fluctuations in the initial matter distribution the initial spatial geometry has not a smooth almond shape but , instead , a more complex spatial geometry which may possess also odd harmonic symmetry planes .
these are predicted to give rise to odd harmonics like triangular flow @xmath42 .
recently it was realized that these odd harmonics are particularly sensitive to both @xmath43/s and the initial conditions , which generated strong theoretical and experimental interest @xcite . ,
@xmath42 and @xmath44 , full and open squares show @xmath45 and @xmath46 respectively .
in addition we show @xmath47 and @xmath48 , which represent the triangular flow measured relative to the second order event plane and the reaction plane , respectively .
the dashed curves are hydrodynamical predictions @xcite described in the text .
figure adapted from @xcite right : the @xmath33-differential triangular flow for pions , kaons and antiprotons .
figure taken from @xcite .
, title="fig:",width=302 ] , @xmath42 and @xmath44 , full and open squares show @xmath45 and @xmath46 respectively . in addition
we show @xmath47 and @xmath48 , which represent the triangular flow measured relative to the second order event plane and the reaction plane , respectively .
the dashed curves are hydrodynamical predictions @xcite described in the text .
figure adapted from @xcite right : the @xmath33-differential triangular flow for pions , kaons and antiprotons . figure taken from @xcite .
, title="fig:",width=302 ] the left panel of fig .
[ fig : figure8 ] shows that @xmath42 is indeed significant and does not depend strongly on centrality .
the magnitude and centrality dependence of @xmath42 is reasonably well described by predictions from a hydrodynamic model calculation with glauber initial conditions and @xmath49 ( dotted curve ) , in contrast to a calculation based on mc - kln cgc initial conditions with @xmath50 , which under - predicts the data ( dashed dotted curve ) .
this suggests that the value of @xmath6 for the matter created in these collisions is small .
however , currently there is no model calculation which can describe both the integrated and @xmath33-differential @xmath5 and @xmath42 at lhc energies with one set of parameters . to investigate further its hydrodynamic origin
, the @xmath33-differential @xmath42 has been measured for pions , kaons and antiproton , as is shown in the right panel of fig .
[ fig : figure8 ] .
it is seen that the mass splitting pattern in elliptic flow is clearly present in triangular flow as well .
in addition to providing constraints on @xmath6 and the initial conditions , it is shown in @xcite that the measured @xmath5 and @xmath42 flow coefficients also provide a natural explanation for the two - particle correlations structures the so - called mach cone and soft ridge first seen at rhic and later at the lhc @xcite .
in this overview we have presented first results on anisotropic flow measured in pb
pb collisions at @xmath0 tev by alice at the lhc . details on these measurements can be found elsewhere in these proceedings @xcite .
we find that for transverse momenta of up to about 2 gev/@xmath9 hydrodynamical model calculations can give a good description of the measurements , even though currently there is no simultaneous description of all the anisotropic flow data with a single choice of initial conditions and @xmath6 .
the stronger elliptic flow , compared to rhic , shows that the system created at lhc collision energies still behaves like an almost perfect fluid .
the first measurements of the higher harmonic anisotropic flow coefficients , in particular @xmath42 , already provide new strong constraints on @xmath6 and the initial conditions .
barely six months after the first lead on lead collisions , the alice anisotropic flow measurements show that the properties of the created matter at the lhc can be studied with unprecedented precision .
99 s. a. voloshin , a. m. poskanzer and r. snellings , in landolt - boernstein , _ relativistic heavy ion physics _ , vol . 1/23 , p 5 - 54 ( springer - verlag , 2010 ) k. aamodt _ et al . _
[ alice collaboration ] , phys .
lett . * 105 * , 252302 ( 2010 ) s. voloshin and y. zhang , z. phys .
c * 70 * 665 ( 1996 ) | the alice detector at the lhc recorded first pb pb collisions at @xmath0 tev in november and december of 2010 .
we report on the measurements of anisotropic flow for charged and identified particles . from the comparison with measurements at lower energies and with model predictions we find that the system created at these collision energies is described well by hydrodynamical model calculations and behaves like an almost perfect fluid . |
since the detection of the first extra - solar planet orbiting a solar - like star in 1995 , 51 peg b @xcite , several new discoveries have been made .
the most used planetary detection methods are radial velocity and transit .
together , both techniques led to the discovery of the vast majority of the over 570 planets observed so far ( according to the extra - solar planets encyclopaedia , http://exoplanet.eu ) . when the same planet is detected by both methods ,
the complete set of planetary and orbital parameters is determined .
two space missions designed to detect planetary transits are operational right now : corot , from the cnes with a consortium of many countries , and kepler , from nasa .
one of the motivations for the search of extra - solar planets is the quest for places suitable for the emergence of life . for this to be possible
, the planet needs to be located in a place known as the habitable zone @xcite .
this is the region around the host star , where the radiation received by the planet is such that allows for water to exist in its liquid state .
furthermore , the planet needs to be rocky .
due to the observations bias , the majority of the planets found up to date are gaseous giants , what excludes them from being habitable even if they are in the habitable stellar zone . through observation of the solar system
, we can infer that the existence of moons and planetary rings around gas exoplanets may be common .
also , we can deduce that moons are rocky by nature .
planetary rings may exist in very different configurations , from the rarefied rings of uranus to the giant rings of saturn .
the detection of both moons and/or rings may uncover important clues about the planet formation and migration process .
moreover , if a gas planet located in the stellar habitable zone has moons , these moons may also be habitable , making them a good place for life to evolve .
@xcite suggest that exomoons could be detected using photometry .
the presence of moons around an exoplanet would cause distortions in the transit light curve , in the form of steps or asymmetries .
the detection of these distortions would provide the moon s orbital parameters . using this technique and photometric data from the hubble space telescope ,
@xcite searched for moons and rings around the exoplanet hd 209458b .
however , nothing was found .
@xcite proposes an algorithm capable of generating transit light curves of a planet with a moon analytically , and shows how this can be used on the search for moons using photometry . on the other hand , @xcite studied the detection possibility using gravitational micro - lensing events .
@xcite showed that it is possible to detect moons in orbit of gaseous giant planets in the habitable zone using spectroscopy in the 1.5 - 5 @xmath1 m band .
@xcite suggests that moon detection is possible from the effects on the planet motion , that may be detectable through precise measurements of the timing and duration of the transits .
these are known as _ transit time variation _
( ttv ) and _ transit duration variation _ ( tdv ) .
@xcite suggest that rings around a saturn - like exoplanet could be detected with photometric precision of ( 1 - 3 ) @xmath2 10@xmath3 and time resolution of 15 minutes , as long as the ring is not viewed close to edge - on .
this resolution is within the accuracy that the kepler space telescope can achieve for sun - like or brighter stars .
@xcite study ring detection through photometry and spectroscopy , and conclude that rings with significantly great inclination can be detected by today radial velocity precision of 1 m / s .
@xcite study the nature of rings that can exist around exoplanets , and conclude that rings around close - in planets can show non - trivial laplacian planes .
this is caused by the increased effects of the planet s quadrupole moment .
these so called warped rings furnish important clues about the planet interior and spin period . despite all these proposed methods , no exo - moon or planetary rings
have been observed so far . here
, we propose a model of planetary transit simulation that may be used in the search for exo - moons through their signature in the light curve of their host planets . beside the moons
, the model may also simulate the transit of a planet with a ring around it .
the model we propose here is capable of generating transit light curves numerically , unlike @xcite .
albeit the larger processing time needed to fit light curves using this method , it has the advantage of being easily adaptable to include new features , as additional moons , changes in the ring plane or starspots . the goal is to apply this model to the observations of the corot and kepler space telescopes , looking for the photometric signals that may indicate the presence of moons or rings around exoplanets .
our model does not consider the detection of the warped rings shown by @xcite .
such rings are predicted around close - in exoplanets , with orbital period of a few days .
thus , our model is limited to the detection of rings located in the planet s equatorial plane , with larger semi - major axis . as the occurance of such warped rings depends on the quadrupole moment j@xmath4 and on the density of the planet , it is difficult to stablish a general criterion to the validity of our model .
for example , for the case of an exoplanet with density 3 g @xmath5 and j@xmath4 = 10@xmath6 , warped rings occur for semi - major axis below 0.05 au ( see figure 6 on @xcite ) .
planets with larger semi - major axis will have rings located in the planet equatorial plane and could be detected by our model .
the next section describes the model used in this work , whereas section 3 presents the application of this model to a few study cases .
section 4 discusses the detectability threshold of moon and ring systems .
finally , the last section lists our main conclusions .
the model used here is based on that of @xcite , where the star is considered a disc with limb darkening and the planet a completely dark disc .
both stellar and planetary parameters can be fit by the model . to minimize the number of free parameters , we consider a simple model where the moon s orbit is coplanar with the planetary one , and is also circular .
looking at the solar system s moons , we see that these assumptions are a little limiting .
however , as this is our first study on the detection of moons , we choose to consider the simplest model . in a future work
we can consider a more realistic model , including orbital inclinations and eccentricity for the moon .
regarding to the planet eccentricity , following the strategy proposed by @xcite , the model simulates eliptical orbits only for the cases were eccentricity was measured by other methods , such as radial velocities .
stellar input parameters to the model are : rotation period , radius , and mass ( the latter two in solar units ) .
the rotation period is used in the calculation of the longitude position of starspots in successive transits .
we also have two limb darkening parameters .
the planetary input parameters are : orbital period , radius , mass , orbital inclination angle , eccentricity and position of the periastron . from the orbital period ,
the orbit s semi - major axis is calculated by kepler s third law .
lastly , the moon input parameters are orbital period , radius , mass , and the angular position of the moon in its orbit at the starting time of the first transit . for the rings ,
the input parameters consists of inner and outer radii , two inclination angles , one with respect to the plane of the sky and the other with the planet s orbital plane .
the transparency of the rings may also be fit .
the temporal resolution of the resulting light curve is chosen within the model , by adjusting the time interval for each simulated photometric data point .
here we used a temporal resolution of 32 s for future comparison with corot data , and 1 min for kepler data . using stellar and planetary parameters ,
the planetary orbit is calculated .
the line of transit is the projection of this orbit on the stellar surface . at each time interval ,
the program calculates the position of the planet and its moon .
the total luminosity of the star planet
moon system is then determined by summing the intensities of all the pixels in the image , where the pixels occupied by the planet and its moon have zero value . in the case of rings ,
the occupied pixels intensity is multiplied by the ring transparency factor . at the end of the transit ,
the light curve is normalized to one from the outside transit data points .
figure [ fig : img_transitos ] shows two examples of transits from the model .
figure [ fig : img_transitos_a ] shows the transit of a planet and a moon in front of the star , and figure [ fig : img_transitos_b ] shows the transit of a planet with rings .
it is also possible to add spots on the stellar surface , where each spot is modeled by three extra parameters : radius , intensity ( with respect to stellar central intensity ) , and longitude position at the start of the first transit .
the spot latitude is considered to be that of the projected transit line . from the stellar rotation period ,
the spot longitude position on consecutive transits is calculated .
a further assumption is that the spot position does not change within a single transit ( usually lasting a few hours ) , which is reasonable when the stellar rotation period is much larger than the transit duration . in this work
, no attempt has been made for a dynamical analysis of a planet with moons or rings .
we used for this matter , regarding to planet - moon systems , the results of @xcite , where the authors through numerical simulations obtain semi empirical expressions for the semi - major axis maximum and minimum that a satellite in orbit of a giant extrasolar planet may have as a function of the stellar and planetary parameters .
@xcite also estimates the maximum mass of a satellite such that its orbit is stable .
an interesting result is that planets with very close in orbits to their host star can only hold moons with very small masses , but more distant planets may have stable moons with masses even larger than that of the earth .
for example , _ rho crb _ at a distance of 0.224 au from its host star , with an orbital period of 39.8 d , may harbor a satellite of up to 1.5 m@xmath7 .
further away planets , such as hd 92788 , at approximately 1 au from its star , allows for a satellite with a maximum mass of @xmath8 m@xmath7 .
the transiting extrasolar planets with the longest periods observed so far are corot-9b ( 95.27 d ) , hd 80606 ( 114.34 d ) and kepler-11 g ( 118.37 d ) .
thus , these planets may harbor a moon with mass similar or even larger than earth , that may be detectable from the application of the present method to corot or kepler lightcurves .
figure [ fig : planetas_sem_lua ] shows two examples of lightcurves obtained from our model .
the figure on the left , [ fig : planetas_sem_lua_a ] , is for the transit of a planet like hd 209458b ( @xcite ) , whereas the figure on the right , [ fig : planetas_sem_lua_b ] , is for a planet similar to corot-2b ( @xcite ) .
figure [ fig : planetas_com_lua ] shows two examples of lightcurves of a planet with an orbiting moon .
both panels are for a jupiter size planet transiting a star identical to the sun with an orbital period of 45 days . in the model ,
the added moon around this planet has 2 earth radii . in [ fig : planetas_com_lua_a ] , the moon orbital period is 5 days , while in figure [ fig : planetas_com_lua_b ] it is 2 days . also , the position angle of the moon with respect to the planet is @xmath9 in the first case , and @xmath10 in the second one .
when adding a moon around a planet in the model , two characteristics signatures may occur .
the first one is a step , that may appear at the beginning or the at the end of a transit , while the second effect is a discontinuity , a jump , seen in the bottom of the transit .
the step at the beginning ( see figure [ fig : planetas_com_lua_a ] ) occurs when the moon ingress happens before that of the planet , while the step near the end of the transit ( figure [ fig : planetas_com_lua_b ] ) is caused by the moon egress after the planet .
both effects are marked with an upward arrow .
furthermore , when the moon orbital period is small , an asymmetry arises in the bottom of the transit , shown by the downward arrows in figure [ fig : planetas_com_lua ] .
this asymmetry is generated when the moon is eclipsed by the planet .
it may also happen if at the start of the transit the moon is already behind ( or in front ) of the planet , but falls out of eclipse during the transit .
the width and intensity of this signal depends on the moon s relative orbital period , radius , and also on the position angle @xmath11 of the moon with respect to the planet at the initial transit time .
figure [ fig : planetas_com_aneis ] presents the modeled light curve produced by a transiting saturn - like planet around a star with solar parameters ( left panel ) . to enhance the signal produced by the ring presence ,
their size were augmented by 50% ( right panel ) .
the main effect of the planetary ring system is to increase the covered area of the stellar surface .
the result of which is to deepen the transit light curve .
moreover , the transit shape becomes rounder at the ingress and egress instants , than if no rings are present .
both these factors depend strongly on the ring transparency , being larger in the case of more opaque rings , as expected .
an interesting effect appears when the ring size is increased . in this case , the light curve becomes more triangular , similar to the light curves produced during grazing eclipses of binary star systems .
if the star has one or more spots on its surface , similar to sunspots , @xcite has shown that the photometric light curve during a planetary transit will display a small bump " .
this occurs because when the planet occults a darker region of the stellar surface , occupied by the spot , there is a relative increase in the total intensity .
moreover , the detection of the same starspots on different transits may be used to determine the rotation period of the star ( @xcite ) .
figure [ fig : planetas_com_mancha ] shows the same light curves from figure [ fig : planetas_sem_lua ] , except now a spot was added to the stellar surface .
this spot , located at 30@xmath12 longitude ( measured from the center of the star ) , has a size of half each planet radius and a relative intensity of 0.6 times the stellar central intensity . [ !
htb ] in some cases , the photometric signal produced by the presence of moons may be mistaken by that caused by starspots .
distinguishing between the two cases is possible when successive transit are observed , since the spots signal will have a periodicity equal to the stellar rotation period , while those of the moons will have a different period , following its own orbital period .
examples of how confusing things can get , especially how the presence of starspots may hinder the detection of moons , are shown in figure [ fig : planetas_com_lua_e_manchas ] .
the figure presents the transit of a planet with an orbiting moon in front of a star , in which surface there is a spot , for two situations : only one spot ( [ fig : planetas_com_lua_e_manchas_a ] ) and star with 4 spots ( [ fig : planetas_com_lua_e_manchas_b ] ) . in both panels ,
the planet is identical to jupiter with a period of 45 d and the moon has 3 earth radius and an orbital period of 2 d around the planet . in the case of figure
[ fig : planetas_com_lua_e_manchas_b ] , it is clearly seen that the presence of starspots can significantly difficult the detection of any moon effect on the light curve .
to verify the detectability of moon and ring systems around a planet , it is possible to add gaussian noise to the modeled light curve and then try and recover the input parameters from a fit . for corot data simulation
, we used a noise level of @xmath13 , such as that observed for corot-2 @xcite , and in the case of the kepler data , @xmath14 .
once the noise has been added to the model light curve , we fit this light curve with the following sequence of algorithms .
first the pikaia @xcite algorithm is used to obtain a first guess that is then inputed into the amoeba @xcite routine .
pikaia is a genetic algorithm the main advantage of which is to find the global maximum of the function to be maximized , using only the allowed interval value for each free parameter .
these results are then used as input to the amoeba routine that refines the result using the simplex method _ for convergence minimization_. the error estimates for the fitted parameters were obtained using the mpfit algorithm @xcite .
after the fitting procedure , we compute the _ goodness - of - fit _ using the parameter q , as defined in @xcite as _ the probability that the chi - square should exceed a particular value @xmath15 by chance _ and calculated by : @xmath16 where @xmath17 is the incomplete gamma function @xmath18 .
generally , if q is larger than 0.01 , then the fit is acceptable . in this case
, it can be said that such event ( moon or ring ) is detectable .
if q is between 0.001 and 0.01 , then the fit may or not be acceptable , and more studies about the noise are required .
this can be caused , for example , if the measurement errors are underestimated .
q below 0.001 indicates a bad fit . on the other hand ,
q very close to 1 is not expected also .
this is , literally , too good to be true .
this can be caused for nonnormal error distribution , or overestimated measurement errors .
next , we show some examples of the fit to light curves with noise added , and the results they yield .
the first test to the model is to apply it to the two transiting planets described in section [ secao_transito_planeta ] , where gaussian noise has been added similar to that of corot and kepler observations . in this fit , all the stellar parameters and the orbital period ( or semi - major axis ) of the planet
were fixed , and the only free parameters were the planetary radius and the orbital inclination angle .
the comparison between the original parameters and those obtained from the fit of the noisy modeled light curve are seen in tables [ tab : ajuste_planetas_corot ] and [ tab : ajuste_planetas_kepler ] , and and in figures [ fig : fit_planetas_sem_lua_a ] and [ fig : fit_planetas_sem_lua_b ] for hd 209458 and corot-2 . both light curves had noise level similar to the corot satellite .
the two panels show the simulated data points as dots and the fit as a solid line .
the residuals are shown in the lower part of each panel .
.fit results for planetary transits of hd 209458b and corot-2b for corot simulations . [
cols="^,^,^,^,^,^",options="header " , ] [ tab : ajuste_saturno_kepler ] the q value in table [ tab : ajuste_saturno_kepler ] and the residuals in figure [ fig : fit_saturno_kepler ] show that the fit was satisfactory .
to determine the radius of the smallest detectable moon , several simulations were performed decreasing the radius at each run .
every simulation had its light curve fitted yielding the moon parameters . whenever the q value indicates a good fit , we consider the moon detectable .
thus , in the case of corot , the smallest detectable moon in orbit around a jupiter size planet has a radius of 1.3 r@xmath7 , while in the case of kepler , this minimum radius decreases to 0.3 r@xmath7 .
the latter moon would be smaller than ganimedes , the larger moon of our solar system , which size is 0.41 r@xmath7 . in all the simulations , the star was considered sun - like with no spots . for the case of planetary rings ,
the process is slightly different , because , in this case , the biggest problem is distinguishing between single planets and those with rings . in this case , then , we adopt a criterion that takes into consideration the residuals .
when fitting the light curve of a planet with rings , first only the planet parameters are fit . if the q value indicates a bad fit , then we may conclude that the rings can be detectable . to establish a detectability criterion
, first we define the effective area of the rings as the ring projection on the plane of the sky multiplied by its opacity , according to : @xmath19 where @xmath20 and @xmath21 are the internal and external radii , respectively , @xmath22 is the inclination angle of the rings relative to the plane of the sky , and @xmath23 is the transparency .
totally transparent rings have @xmath24 , whereas totally opaque ones have a zero value of transparency .
the results of our simulations showed that in the case of the kepler data , planetary rings are detectable when the effective area corresponds to approximately 3% of the area covered by the planet .
for example , rings with @xmath25 and inclination angle of 78@xmath26 with the plane of the sky in orbit a saturn - like planet , will be detectable with rings about 51% the size of those of saturn . for the corot data
, the effective area corresponds to approximately the area of the planet .
therefore , for the same planet as in the previous case , the rings should have sizes 45% larger than those of saturn to be detectable above the corot noise level .
here we reported the results of a model capable of simulating planetary transits , including the effect of existing moons and rings around that planet .
light curves were presented for each case .
estimates of the threshold detection for such events were made .
it was shown how these events may be detected using the pikaia and amoeba routines for curve fitting .
as expected , the main limiting factor for the possible detection is the noise level of the telescopes .
this can be seen by comparing the results from simulations of photometric light curves of corot and kepler .
the corot data allows for the detection of moons with at least 1.3 r@xmath7 in size around a planet identical to jupiter . in the case of kepler data ,
this detection limit on the moon radius drops to 0.3 r@xmath7 . for planetary rings
, a saturn - like planet would be detectable with kepler , however , this same planet - ring system would be mistakenly inferred as a single planet with slightly larger radius by corot .
detection limits for both telescopes were estimated above .
next , this model will be applied to real light curves to search for the types of signatures described in this work .
data from both corot and kepler will be used .
an improvement that can be made to the model is the implementation of analysis of the transit time and duration variations caused by the presence of the moon that makes the planet orbit the common center of mass @xcite .
these timing variations would confirm the presence of moons . finally , by detecting the moon by both methods , photometry and timing variation , it is possible to establish a complete set of physical parameters for the satellite .
the authors acknowledge financial support from the brazilian agencies capes and fapesp ( grant number 2006/50654 - 3 ) .
r. alonso et al .
2008 , , 482 , l21 barnes , j. w. , & fortney , j. j. 2004 , , 616 , 1193 brown , t. m. , charbonneau , d. , gilliland , r. l. , noyes , r. w. , & burrows , a. 2001 , , 552 , 699 charbonneau , p. 1995
, , 101 , 309 domingos , r. c. & winter , o. c. & yokoyama , t. 2006 , , 373 , 1227 han , c. , & han , w. 2002 , , 580 , 490 huang , s. s. 1959 , american scientist , 47 , 397 .
kasting , j. f. , whitmire , d. p. , & reynolds , r. t. 1993 , , 101 , 108 kipping , d. m. 2008 , , 389 , 1383 kipping , d. m. 2009 , , 392 , 181 kipping , d. m. 2009 , , 396 , 1797 kipping , d. m. 2011 , , 416 , 689 markwardt , c. b. 2009 , astronomical data analysis software and systems xviii asp conference series , vol .
411 , proceedings of the conference held 2 - 5 november 2008 at hotel loews le concorde , qubec city , qc , canada . edited by david a. bohlender , daniel durand , and patrick dowler .
san francisco : astronomical society of the pacific , 2009 .
, p.251 mayor , m. , & queloz , d. 1995 , , 378 , 355 nelder , j. a. & mead , r. 1965 , comput .
j. 7 , 308 .
ohta , y. , taruya , a. , & suto , y. 2009 , , 690 , 1 press , w.h . , & flannery , b.p . , & teukolsky , s.a .
& vetterling , w.t .
numerical recipes in c : the art of scientific computing .
new york : cambridge university press , 1992 .
sartoretti , p. , & schneider , j. 1999 , , 134 , 553 schlichting , h. e. , & chang , p. 2011
, , 734 , 117 silva , a. v. r. 2003 , , 585 , l147 silva - valio , a. 2008 , , 683 , l179 silva , a. v. r. , & lanza , a. f. 2010 , , 529 , a36 southworth , john .
2010 , , 408 , 1689 williams , d. m. , & knacke , r. f. 2004 , astrobiology , 4 , 400 | since the discovery of the first exoplanet , those most adequate for life to begin and evolve have been sought . due to observational bias , however , most of the discovered planets so far are gas giants , precluding their habitability .
if these hot jupiters are located in the habitable zone of their host stars , and if rocky moons orbit them , these moons may be habitable . in this work
we present a model for planetary transit simulation considering the presence of moons and planetary rings around the planet .
the moon s orbit is considered to be circular and coplanar with the planetary orbit .
the other physical and orbital parameters of the star , planet , moon and rings can be adjusted in each simulation .
it is possible to simulate as many successive transits as desired .
since the presence of spots on the surface of the star may produce a signal similar to that of the presence of a moon , our model also allows for the inclusion of starspots .
the result of the simulation is the light curve with the planetary transit .
white noise may also be added to the light curves , to produce curves similar to those obtained by the corot and kepler space telescopes .
the goal is to determine the criteria for detectability of moons and/or ring systems using photometry .
the results show that it is possible to detect moons with radii as little as 1.3 r@xmath0 with corot and 0.3 r@xmath0 with kepler . |
there has been a lot of interest in neutrino interactions in the few gev energy region , coming from the oscillation experiments and the demand to better constrain the systematic errors . for the neutrino energies around @xmath3 gev the most abundant reaction
is charged current quasi - elastic ( ccqe ) scattering : @xmath4 and it is also the most important process in the investigation of the oscillation phenomenon , e.g. in the t2k experiment @xcite . due to the standard
conserved vector current ( cvc ) and partially conserved axial current ( pcac ) hypotheses , using the electron scattering data and assuming a dipole form of the axial form factor , the weak transition matrix element contains only one unknown parameter , the axial mass @xmath5 .
recent neutrino ccqe cross section measurements , notably the high statistics muon double differential cross section results from the miniboone ( mb ) experiment @xcite , suggest @xmath5 values significantly larger than estimations from the older deuteron target neutrino measurements and from the pion electro - production data @xcite .
it is becoming clear that in order to understand correctly the mb data , a two body current contribution to the cross section must be considered @xcite . in the experimental event identification
this contribution can easily be confused with the genuine ccqe events .
the older and recent ccqe @xmath5 estimates can be consistent , because in the case of the neutrino - deuteron scattering the two body current contribution is small @xcite .
theoretical models of the two body current contribution ( called in this paper also n - particles n - holes ( @xmath6 ) ) give quite different estimates of the size of the effect both for neutrino and antineutrino scattering .
recently the mb collaboration published the first large statistics antineutrino ccqe - like ( a sum of ccqe and @xmath6 contributions ) cross section results @xcite .
the data has been already analysed by the theoretical groups @xcite . in ref .
@xcite a ratio of the ccqe - like cross sections for neutrinos and antineutrinos was discussed as a function of neutrino energy .
if the errors become smaller , this kind of data can allow to discriminate the models .
it is important to look also for alternative ways to evaluate @xmath5 and/or investigate the size of the two body current contribution .
the minerva collaboration analysed the energy deposit near ccqe - like interaction vertex looking for an evidence for multinucleon knock - out events @xcite .
another option is to look at the neutral current elastic ( ncel ) reaction : @xmath7 , where @xmath8 denotes proton or neutron . as it will be explained in detail in sect .
[ ncel ] the basic theoretical framework to investigate ncel scattering is similar to the one used for ccqe .
the neutral current nc hadronic current is expressed in terms of the vector and axial form factors , that are linear combinations of the form factors present in the ccqe reaction with an addition of new terms sensitive to the strange quark content of nucleons .
thus , the ncel scattering data allows for extraction of both @xmath5 and the strange quark contribution to the form factors .
in fact , most of the interest in the ncel reaction comes from its potential to measure the strangeness of the nucleon .
the axial strange form factor can be determined from the ncel neutrino-@xmath9 scattering data because it enters proton and neutron matrix elements with opposite signs .
typically , in experiments , one tries to extract @xmath10 , a value of the axial strange form factor at @xmath11 .
it corresponds to the fraction of strange quarks and antiquarks that contribute to the total proton spin , commonly denoted as @xmath12 .
the first estimation based on @xmath13 data ( bnl e734 experiment ) was done in ref .
@xcite with the result @xmath14 . for the later discussions see refs .
@xcite .
about three years ago the mb collaboration measured the flux averaged ncel differential cross section on the @xmath9 target @xcite .
two observables were considered .
the first one is the distribution of events in the total reconstructed kinetic energy of the final state nucleons . the measurement is based on the mb detector ability to analyse the scintillation light in the absence of cherenkov light from the final state muon . in the second observable a proton enriched sample of events
is analysed with the kinetic energies above the cherenkov radiation threshold . the observable
is defined as the ratio of the cross section for the proton enriched sample to the total ncel - like ( events with no pions in the final state ) cross section .
the mb collaboration made two separate parameter extractions . assuming @xmath15 the value of @xmath16 gev was obtained from the first observable .
taking the value @xmath17 gev from the ccqe analysis @xcite , mb found @xmath18 from the proton enriched sample of events .
the description of both data samples in terms of @xmath19 gev and @xmath20 is consistent .
the mb collaboration did not attempt to make a simultaneous fit to both theoretical parameters ( such fits were only discussed in ref .
@xcite ) . in the mb analysis
the values of @xmath5 used in modelling scattering off carbon and off free protons , were different . for protons
the @xmath5 was fixed to be @xmath21 gev .
for carbon the axial mass values was treated as a free parameter . a large _ effective _
@xmath5 value is expected to account for the @xmath6 contribution present in neutrino - carbon scattering but absent in the neutrino - proton scattering .
mb collaboration used nuance neutrino event generator with nuclear effects described by fermi gas model and final state interactions ( fsi ) @xcite .
nuance does not include the @xmath6 contribution . the data from ref .
@xcite has been already discussed in some phenomenological papers .
butkevich and perevalov @xcite investigated an impact of more realistic nuclear model in the mb data analysis . a relativistic distorted wave impulse approximation ( rdwia ) @xcite model leads to the values @xmath22 gev and @xmath23 , consistent with those reported in ref .
the spectral function formalism was used by ankowski in ref .
@xcite , with the conclusion that the shape of the mb measured distribution of events in @xmath24 is reproduced with @xmath25 gev . however , there is a @xmath26 discrepancy in the overall normalisation with the mb results ( the measured cross section is larger ) .
meucci , gusti and pacati analysed the predictions of four nuclear models @xcite .
it turned out that relativistic green function ( rgf ) model is able to reproduce the mb ncel data with the value of @xmath5 close to those obtained in old deuteron - target experiments .
the purpose of this paper is to perform the first analysis of the mb ncel data using a model that includes the @xmath6 contribution .
the model is provided by the nuwro monte carlo event generator @xcite developed over last 9 years at the wrocaw university .
nuwro describes the neutral current @xmath6 contribution with the effective transverse enhancement ( te ) model @xcite .
the te model provides the contribution to the neutrino inclusive cross section and predictions for the final state nucleons are obtained with a procedure described in ref .
the advantage of the te model is that it can be safely used for neutrinos of energies larger than @xmath27 gev . in this paper
only nucleons resulting from the @xmath6 events will be analysed and differences between microscopic ( see refs .
@xcite ) and effective @xmath6 models are made smaller by fermi motion and final state interactions effects .
the recent minerva ccqe - like data analysis demonstrates that with @xmath28 gev and the te model both neutrino and antineutrino @xmath24 distributions are well reproduced @xcite .
another important difference with respect to the previous studies of the mb ncel data is in the treatment of nuclear effects .
we will compare nuwro predictions to the quantities ( visible energy ) that are directly observable and we will not rely on the nuance fsi model . our main result is a simultaneous fit to the @xmath5 and @xmath29 done using the mb data for the distribution of events in the total reconstructed kinetic energy of the final state nucleons , with the outcome : @xmath30 gev and @xmath2 .
we will argue that the second mb observable is very sensitive to details of fsi and the @xmath6 kinematics model and it is very difficult to use it as a reliable source of an information about theoretical model parameters . on the contrary ,
the first observable is quite robust to such details and can be successfully used to extract values of the interesting quantities .
our paper is organised as follows : in sec .
[ theory ] a general description of the ncel reaction is given ; in sec .
[ nuwromc ] the main features of the nuwro generator are summarised ; sec . [ dataanal ] contains a details of the data analysis and the mb energy unfolding procedure ; in sec [ results ] we present our main results , and conclusions can be found in sec .
[ conclusions ] .
[ ncel ] we consider neutral current neutrino - nucleon scattering : @xmath31 where @xmath8 , @xmath32 , @xmath33 denote the initial and final state nucleons and the outgoing lepton with the four momenta : @xmath34 , @xmath35 and @xmath36 , respectively .
the four momentum transfer is given by @xmath37 , @xmath38 .
in the born approximation the scattering matrix element factorises into the product of leptonic and hadronic contributions : @xmath39 where @xmath40 is cabibbo angle , @xmath41 is the fermi constant , while @xmath42 and @xmath43 are the expectation values of the leptonic and hadronic currents .
the leptonic part is : @xmath44 while getting the hadronic contribution requires an extra phenomenological input in the general formula @xmath45 with the effective hadronic vertex @xmath46 . in order to construct @xmath46 for the ncel scattering
, one has to follow the pattern given by the standard model .
the cvc theory , the pcac hypothesis and the @xmath47 isospin symmetry relate the neutral current form factors to those present in the electromagnetic and charged current hadronic matrix elements @xcite .
the nc hadronic vertex reads : @xmath48 where indices @xmath34 and @xmath49 refer to proton and neutron .
the nc form factors can be expressed as : @xmath50 @xmath51 signs refer to proton / neutron , @xmath52 is the weinberg angle , @xmath53 .
@xmath54 are the proton ( neutron ) electromagnetic form factors , @xmath55 is the axial nucleon form factor : @xmath56 @xmath57 and @xmath58 are the vector and the axial strange form factors .
the electromagnetic form factors @xmath59 , @xmath60 are obtained from the analysis of the elastic @xmath61 scattering data ( for a review see ref .
@xcite ) and can be expressed in terms of the electric and magnetic form factors : @xmath62 , \nonumber \\ \\
f_2^{p(n ) } ( q^2 ) & = & \frac{4m^2}{q^2 + 4m^2 } \left [ g_m^{p(n)}(q^2 ) - g_e^{p(n)}(q^2)\right ] , \end{aligned}\ ] ] where @xmath63 is the average nucleon mass . in the breit frame
the electric ( @xmath64 ) and the magnetic ( @xmath65 ) nucleon form factors are related to the nucleon electric charge and current distributions .
for instance , for the electric proton form factor @xmath66 , where @xmath67 is the mean - square radius of the charge distribution . similarly , one can introduce the electric ( @xmath68 ) and magnetic ( @xmath69 ) isoscalar strange form factors . in the first approximation
one can assume that `` the effective sizes '' of the proton and neutron strange sea are the same , therefore we use the same strange form factors for both nucleons .
the latest global analysis of the elastic parity violating ( pv ) @xmath70 and bnl e734 neutrino scattering data indicate that the values of vector strange form factors are consistent with zero @xcite , and in our analysis we set @xmath71 . in order to estimate the axial strange form factor
we assume that the radius of the strange sea is comparable with the axial `` charge '' radius of the proton so that @xmath72 with the free parameter @xmath29 that has to be extracted from the data .
in our numerical analysis we use the bbba05 vector form factors @xcite .
we also investigated a possible impact on the results from the form factors corrected by the two - photon exchange effect @xcite .
nuwro is a monte carlo event generator developed at the wrocaw university @xcite .
it simulates neutrino - nucleon and neutrino - nucleus interactions including : ( quasi- ) elastic scattering , pion production through @xmath73 resonance ( with a contribution from the non - resonant background ) , more inelastic processes and coherent pion production .
nuwro covers neutrino energy range from @xmath74 mev to tev .
there are three basic nucleus models implemented in nuwro : global / local relativistic fermi gas ( fg ) model , spectral function and the effective momentum and density dependent potential . in the analysis described in this paper we use the local fermi gas model . in the impulse approximation picture ,
all the hadrons arising at a primary vertex are propagated through the nuclear matter using the nuwro cascade model .
nuwro is the open - source project and the code is freely available @xcite .
there are three charged current ( cc ) two - body current models implemented in nuwro : the ific model @xcite , the martini et al .
model @xcite and the te model @xcite . in all of them
double differential cross section contribution for the final state muons is the external input and the hadronic part is modelled using the scheme ( proposed in @xcite ) : 1 . set randomly the four - momenta ( @xmath75 and @xmath76 ) of the initial nucleons from the fermi sphere with a radius determined by the local nuclear density ; 2 .
calculate the four - momentum of the hadronic system @xmath77 where @xmath78 is the four - momentum transferred to the hadronic system ; 3 .
repeat steps 1 . and 2 .
until the invariant hadron mass is larger than mass of two nucleons ( @xmath79 ) ; 4 .
make the lorentz boost to the hadronic center of mass system ; 5 .
select isotropically momenta of two final state nucleons ; 6 .
boost back to the laboratory frame ; 7 .
apply final state interactions .
results from three models depend on the distribution of events in the energy transfer .
the differences are smeared out by the fermi motion and the fsi effects . in our analysis of the ncel data we use the te model @xcite , because to date this is the only nc @xmath6 model available in nuwro . in the te model the @xmath6 contribution to the nc scattering cross section on carbon
is introduced by the modification of the vector magnetic form factors : @xmath80 where @xmath81 and @xmath82 .
using @xmath83 in the ncel differential cross section formula one obtains a cross sections for a sum of the ncel and @xmath6 reactions .
the @xmath6 cross section is obtained by subtracting the ncel part described with the standard magnetic form factors : @xmath84 the value of the axial mass in the te model is set to be @xmath85 mev , as assumed in ref .
the total cross section for the nc @xmath6 scattering is shown in fig .
[ fig : mec_nc ] . at the typical mb flux
neutrino energies , @xmath86 mev , the @xmath6 contribution in the te model amounts to about @xmath87 of the ncel cross section . for comparison
we show also the predictions from the martini et al . model taken from ref .
the martini et al .
model predicts much larger @xmath6 cross section .
also for the cc @xmath6 reaction the predictions from this model are larger than those from the ific and te models . in the cc @xmath6 reactions
there can be either neutron - neutron ( @xmath49-@xmath49 ) or proton - neutron ( @xmath49-@xmath34 ) pair in the initial state .
the probability of the mixed isospin pair is defined in nuwro by a parameter @xmath88 , with the default value @xmath89 . for the nc @xmath6 interactions
every isospin initial state pair is possible ( @xmath49-@xmath34 , @xmath49-@xmath49 , @xmath34-@xmath34 ) . to keep the same proportion between @xmath49-@xmath49 and @xmath34-@xmath49 pairs and assuming the same probability to have @xmath49-@xmath49 and @xmath34-@xmath34 pairs
, we introduce the parameter @xmath90 giving the likelihood of a @xmath91 pair to be selected in a nc two body current reaction .
hadrons resulting in the primary vertex propagate through the nuclear matter with the nuwro cascade model : 1 .
nucleons are assumed to be in the potential well of depth @xmath92 , where @xmath93 is the fermi energy and @xmath94 mev ; 2 .
formation zone ( fz ) can be applied ( fz is set to be zero for the @xmath6 events ) and the following steps are repeated : 1 .
a nucleon free path ( @xmath95 ) is drawn from the exponential distribution taking into account the nucleon - nucleon cross section and the local nuclear density ; 2 . if @xmath96 fm the nucleon is propagated by @xmath95 , the interaction kinematics is generated and a check for pauli blocking is done to decide if the interaction happened ; 3 . if @xmath97 the nucleon is propagated by @xmath98 . in the cascade the fermi motion of target nucleons
is taken into account , and every new nucleon , which participate in the cascade , brings in an extra kinetic energy .
when a nucleon leaves the nucleus , its kinetic energy is reduced by @xmath99 setting it on - shell .
if the kinetic energy is smaller than @xmath99 , the nucleon is assumed to be stuck inside the nucleus . for more detailed description of the nuwro cascade model
we are going to discuss two data samples provided by the mb collaboration .
the first one ( the ncel sample ) contains the distribution of the total reconstructed kinetic energy of all nucleons in the final state , normalised to the number of events seen in the detector . the second data sample ( the ncel high energy sample )
is provided in a form of the ratio : @xmath100 where @xmath101 denotes a contribution from a special class of events , called single proton or proton enriched .
those are the events with visible cherenkov light and proton angle @xmath102 . in mc simulations the largest contribution to those events comes from the ncel scattering on protons , which then do not undergo reinteractions . in the case of multiple proton events ,
the energy of individual proton is in general too low to produce the cherenkov light .
even if a high energy proton appears in the multiple proton event , it has typically larger scattering angle than protons unaffected by fsi . the denominator ( @xmath103 )
denotes the contribution from all the ncel - like interactions .
both data samples are presented as a function of reconstructed energy ( @xmath104 ) , measured in the detector . to compare those data with the theoretical predictions given in terms of the true kinetic energy ( @xmath105 )
, one needs the unfolding procedure , allowing a passage from @xmath105 to @xmath104 . in the next subsections we describe the original mb unfolding procedure .
we had to propose a treatment of @xmath6 events , not considered in the mb analysis . for all but @xmath6 events
we follow closely the approach proposed by perevalov in ref .
a similar unfolding procedure is used in the both data samples .
five types of events giving contribution to the final distributions are considered : 1 .
\(a ) ncel on hydrogen ; 2 .
\(b ) ncel on a proton from carbon unaffected by fsi effects ; 3 .
\(c ) ncel on a proton from carbon with fsi effects ; 4 .
\(d ) ncel on a neutron from carbon ; 5 .
\(e ) irreducible background ( pions produced in a primary vertex and absorbed due to fsi effects ) .
the probabilities of scenarios ( b - e ) depend on the details of the cascade models implemented in nuance or in nuwro . for each type of the signal events , @xmath106 ,
there is a response matrix ( @xmath107 ) provided in ref .
@xcite , which simulates the energy smearing , the detector efficiency and defines a relation between true and reconstructed energy distributions : @xmath108 @xmath107 are either @xmath109 or @xmath110 matrices for the two data samples , respectively .
the columns of matrices label the true kinetic energy bins and rows label the reconstructed energy .
there are @xmath111 bins starting from @xmath112 mev up to @xmath113 mev plus an extra overflow bin for the ncel sample in the true kinetic energy
. for ncel high energy sample there are @xmath114 bins , starting from @xmath115 mev up to @xmath113 mev plus underflow and overflow bins . to obtain the reconstructed kinetic energy distribution and compare with the data from ref .
@xcite one goes through the following steps : 1 .
use a theoretical model and calculate the flux - averaged distributions for five different types of signal events using the same bins as in the response matrices ; 2 . use the proper response matrices to translate each histogram to the reconstructed kinetic energy distribution ; 3 .
sum all the histograms and add the background events ( @xmath116 contains dirt , beam - unrelated , and other backgrounds provided by the mb collaboration in ref .
@xcite ) to get the total reconstructed energy spectrum : + @xmath117 4 . use the provided error matrices ( @xmath118 ) to calculate @xmath119 : + @xmath120 + unlike in the ccqe mb data published in ref .
@xcite , the flux normalisation error is already included in the error matrix @xmath118 .
an alternative way to convert the true kinetic energy to the reconstructed one is to translate it on the event by event basis . for each value of the true kinetic energy
the corresponding column in the response matrix gives a probability distribution with the information how the given true energy value is smeared out in the detector , normalised to the efficiency . to obtain the reconstructed kinetic energy distribution one proceeds as follows : 1 . for each event
calculate the total true kinetic energy of all nucleons in the final state ( @xmath105 ) and get a bin number @xmath121 ; 2 .
find the type of signal ( @xmath122 ) , see sect .
[ anal1 ] ; 3 .
choose @xmath121-th column of the @xmath107 response matrix as the probability distribution ; 4 . use the mc method to decide if the event is accepted ( according to the efficiency ) and what energy is visible in the detector . in the @xmath6 events
there are typically two nucleons after a primary interaction and both of them propagate independently through nucleus . in the mb analysis
there are no @xmath6 events included and no response matrices were prepared for them . to take two body current events into account
, we must express them in terms of the five signals defined by the mb .
a naive interpretation may suggest a treatment of each nucleon from the @xmath6 events separately .
however , it would be incorrect .
the signal is recorded by photomultiplier tubes ( pmt ) , which absorb the light emitted in the scintillator ( and also the cherenkov radiation ) .
the event is accepted if there is a sufficient number of pmt hits .
any of two individual nucleons may have too low energy to generate enough pmt hits , but together they can make it @xcite .
in our analysis we treated both nucleons from a two body current event together and summed up the kinetic energies of all the nucleons in the final state as if they were coming from only one nucleon . in the detector @xmath6 events
are seen as multiple protons events , so as signals ( c ) - ncel on proton from carbon affected by fsi effects and ( d ) - ncel on neutron from carbon .
one expects events without a proton in the final state to be more smeared out in the detector , and we apply response matrix for the signal ( d ) if there were two neutrons in the primary vertex or for the signal ( c ) in other cases @xcite . in @xmath6 events
the energy transferred to the hadronic system is shared by two nucleons , and the probability that there will be a proton with energy large enough to produce the cherenkov light is low .
thus , the @xmath6 events make the ratio @xmath123 ( eq . [ eq : ratio ] ) smaller .
the ratio @xmath123 ( eq . [ eq : ratio ] ) as a function of the total reconstructed kinetic energy of all nucleons in the final state , broken down to individual contributions from : elastic scattering on carbon , np - nh , elastic scattering on hydrogen , irreducible background and other backgrounds .
the nuwro result is obtained with the @xmath124 gev and @xmath125 values . ]
we first repeat the analysis without the @xmath6 contribution to check if our numerical procedures reproduce the mb results .
we assumed a fixed value of the axial mass for hydrogen @xmath126 gev and we minimised the @xmath119 function for the effective axial mass for carbon ( @xmath127 ) using the data for the reconstructed energy distribution . following the mb procedure we fixed the value @xmath128 .
the value @xmath129 was obtained with @xmath130 ( a confidence level ( cl ) @xmath131 ) .
this value is larger than the one reported by the mb collaboration ( @xmath132 gev ) @xcite but consistent within the 1@xmath133 error bars .
a discrepancy is probably caused by the presence of the pionless @xmath73 resonance decays in the nuance but not in the nuwro generator .
moreover , we use a lower value of @xmath5 for hydrogen than the mb collaboration . to investigate a possible impact of the choice of the electromagnetic form factors parameterisation on the final results
, we repeated the computations using the form factors corrected by a two - photon exchange @xcite and obtained almost identical results . using the data for the ratio @xmath123 ( eq . [ eq : ratio ] ) we examined the strange quark contribution to the ncel cross section .
we assumed a fixed value of the axial mass : @xmath134 gev for hydrogen and effective value @xmath135 gev for carbon .
we found the strange quark contribution to be @xmath136 with @xmath137 ( cl @xmath138 ) .
this results is consistent with values published by the mb collaboration @xcite and the bnl e734 experiment @xcite .
following the same steps the analysis was repeated including the @xmath6 contribution . assuming @xmath128 the minimum of the @xmath119 for the distributions of the total reconstructed kinetic energy of the final state nucleons was found for the axial mass value @xmath139 gev with @xmath140 ( cl @xmath131 ) .
the extraction of the strangeness from the ratio @xmath123 , assuming @xmath141 gev , leads to the value @xmath142 with @xmath143 ( cl @xmath144 ) , which is inconsistent with zero ( as assumed in the first fit ) . as it was discussed before the two body current events contribute mostly to the denominator of the ratio @xmath123 making its value smaller .
also , a lower @xmath5 makes the ratio @xmath123 smaller . to compensate for both effects
a lower value of @xmath29 can be expected .
the inconsistency described in the previous subsection encouraged us to try to make a simultaneous fit of both theoretical model parameters . in the case of the first observable we obtained the following results : * without @xmath6 events : @xmath145 with @xmath146 ; * with @xmath6 events : @xmath147 with @xmath148 . in the case of second observable
we discovered that the best fit values are very sensitive to many details of the theoretical model : * from the fig .
[ fig : ds_bf ] it is clearly seen that the ratio @xmath123 depends strongly on other backgrounds ; * above @xmath149 mev of the kinetic energy a significant contribution comes from irreducible background ( pion production and absorption ) known with a precision not better than 20 - 30% ; * we constructed monte carlo @xmath6 toy models based on the te model with modified distribution of energy transfer and the obtained best fit values depend strongly on such modifications ; on the other hand the results from the first observable are affected in the much weaker way .
[ fig : kontur ] shows our results for the simultaneous two - dimensional fits without and with the @xmath6 contribution included in the nuwro simulations together with @xmath150 confidence regions .
the inclusion of the @xmath6 events makes the best fit result for @xmath5 consistent with the world average .
it confirms that the difference between recent and older axial mass measurements can be explained by taking into account two body current contribution .
the value of the strange quark contribution is found to be consistent with zero .
we calculated the value of @xmath119 for the second mb observable with the values @xmath151 and @xmath152 and obtained @xmath153 .
it means that the reported values are consistent also with the proton enriched sample observable .
our best fit for the distribution of the total reconstructed kinetic energy of the final state nucleons and the overall nuwro prediction ( broken down to individual contributions from elastic scattering on carbon , elastic scattering on hydrogen , two body current , irreducible background and other backgrounds ) is demonstrated in fig .
[ fig : ma_bf ] .
the contribution coming from the two body current amounts to approximately @xmath154 of the overall distribution affecting both its shape and the normalisation .
as we focus on the sum of the kinetic energies of all the nucleons in the final state , the result is not much sensitive to the assumptions made on the @xmath6 kinematics described in sec [ sec : mec ] . our predictions for the ratio @xmath123 obtained with the values @xmath151 and @xmath152 together with the contributions to the numerator coming from various signal events
are presented in fig .
[ fig : ds_bf ] . as mentioned in sec .
[ sec : nucmodel ] the formation zone for the @xmath6 events is assumed to be zero , because there is no clear physical motivation to introduce it . in order to estimate how important the fz effect can be for the ncel analysis we repeated the computations assuming the fz for the @xmath6 events to be @xmath3 fm .
it turned out that this assumption does not affect the final results in a statistically relevant way .
we investigated the impact of the uncertainty of the value of the parameter @xmath155 defining a relative abundance of @xmath91 and @xmath156 pairs on which two body current scattering occur .
the default value of @xmath155 is @xmath157 and we repeated the computations with the value @xmath158 .
no influence on the final results was found .
the impact of the two body current events on the analysis of the mb data for the neutrino ncel scattering on @xmath0 was investigated in detail .
this is the first analysis of this kind yet .
we performed a simultaneous fit to two theoretical model parameters and obtained the values @xmath30 gev and @xmath2 .
our results provide a new evidence that large axial mass measurements can be explained by the two - body current contribution to the cross section neglected in the experimental data analysis
. it would be interesting to repeat this analysis using one of the microscopic models of the nc two body current contribution . also , it is desirable to include other nuclear effects like random phase approximation ( rpa ) correlations .
as mentioned above , nuwro contains implementations of ific and martini et al .
models for @xmath6 in the cc channel only . recently using the theoretical computations from @xcite nuwro was upgraded with the implementation of rpa , but again only for the cc reactions .
we are planning to include @xmath6 dynamics and the rpa effects for nc channels but it requires a significant amount of theoretical and programming work . for the rpa an appropriate recalculation of the components of the polarisation tensor is needed as in the relativistic formalism adopted in ref .
@xcite the response functions are calculated in the analytical form .
recently the miniboone collaboration made public the preliminary results from the antineutrino ncel analysis @xcite .
they are supplementary to those discussed in this paper .
when the data become available a combined analysis will have more potential to investigate the @xmath6 contribution and put more constraints on the @xmath29 value . as discussed in sect .
[ sec : npnh ] the mb proton enriched distribution of events contains an important information about hadrons resulting from @xmath6 events not explored in this study . in order to make use of this information a very good control over the fsi effects is required .
finally , we would like to notice that there is an interesting idea for an alternative measurement of the ncel cross section described in ref .
the authors investigate the relation between the rate of the observed @xmath159 rays coming from nuclear deexcitation in water - cherenkov detectors and the ncel cross section .
we would like to thank denis perevalov , arie bodek and eric christy for useful information .
m. martini , m. ericson , g. chanfray , and j. marteau , phys .
c * 81 * ( 2010 ) 045502 ; j. nieves , i. ruiz simo , and m.j .
vicente vacas , phys .
b * 707 * ( 2012 ) 72 ; a. meucci , c. giusti , phys .
rev . d*85 * ( 2012 ) 093002 ; j. e. amaro , m. b. barbaro , j. a. caballero , t. w. donnelly , and j. m. udias , phys .
d * 84 * ( 2011 ) 033004 .
fiorentini et al .
[ minerva collaboration ] , _ measurement of muon neutrino quasi - elastic scattering on a hydrocarbon target at @xmath161 gev _ , arxiv:1305.2243 [ hep - ex ] ; l. fields et al .
[ minerva collaboration ] , _ measurement of muon antineutrino quasi - elastic scattering on a hydrocarbon target at @xmath161 gev _ , arxiv:1305.2234 [ hep - ex ] . | results of the analysis of the miniboone experiment data for the neutral current elastic neutrino scattering off the @xmath0 target with the nuwro monte carlo generator are presented .
inclusion in the nuwro the two body current contribution leads to the axial mass value @xmath1 gev consistent with the older evaluations based on the neutrino - deuteron scattering data . the strange quark contribution to the nucleon spin
is estimated with the value @xmath2 . |
there are more than 100 giant molecular clouds in the galaxy with mass of 10@xmath11m@xmath12 or greater .
the internal velocity dispersion of these clouds , measured using @xmath3co emission , is in most cases approximately equal to their virial velocity of about 15 ( dame 1986 ) .
however , there are at least two very unusual molecular clouds , located at @xmath13(@xmath14 ) and ( @xmath15 ) , which have extremely large velocity dispersions of about 100 ( fwhm ) as seen in several molecular lines including @xmath3co
. hereafter we shall refer to these objects as wide line clouds ( wlcs ) .
oort drew attention to the large velocity dispersion of the wlc at @xmath16 longitude in his influential review article almost two decades ago ( 1977 ) .
however , these clouds remained largely unexplored until the mid - eighties when stark and bania ( 1986 ) and bitran ( 1987 ) mapped them at high spatial resolution in @xmath3co , @xmath17co , and cs , and confirmed their unusually large velocity dispersion .
the observed velocity dispersion even in cs , which is associated with high density cores of clouds , is about @xmath18 compared to @xmath19@xmath20 for giant molecular clouds .
both clouds are also visible in the large - scale oh absorption - line survey of boyce and cohen ( 1994a ) where the properties of seven wlcs are discussed in the longitude range @xmath21 . for
an overall @xmath10@xmath22 map of this galactic centre region see boyce and cohen ( 1994b ) .
in addition , the same authors have mapped the cloud at @xmath23 using @xmath24 measurements and give a mass estimate of @xmath25 ( boyce _ et al .
_ 1989 ) .
according to bitran ( 1987 ) and stark and bania ( 1986 ) the mass of the clouds , estimated from the @xmath3co emission , is about @xmath26 , however , this is very uncertain because the ratio of h@xmath27 mass to co emission for clouds in the inner few kpc is likely to be different from the solar neighborhood ( stacy 1988 ) .
we have mapped the two wlcs in @xmath28-cm in order to provide a lower limit to their mass , to determine their velocity dispersion , and to ascertain if they are different from giant molecular clouds . in the next section
we describe the observational details , sect .
3 contains the results and a comparison with the co - observations .
the main results are summarized in sect . 4 .
we have observed the two wide - line clouds in @xmath28-cm with the effelsberg @xmath29 m telescope using a cooled hemt receiver with a system temperature less than about @xmath30 k on cold sky but increased to about @xmath31 k due to radio emission from the galactic centre and spillover from the ground .
the total bandwith of the receiver was @xmath32 mhz around the neutral hydrogen hyperfine transition frequency of @xmath33 mhz ; the corresponding velocity range was @xmath34 to @xmath35 . using the @xmath36 channel autocorrelation spectrometer a resolution of about @xmath37
was obtained after hanning filtering of the spectra .
the clouds have been mapped in steps of @xmath38 which is half the beam width of the telescope . the total angular coverage in @xmath10 and @xmath22 for the two clouds were ( @xmath39)(@xmath40 ) and ( @xmath41)(@xmath42 ) , respectively .
thus each cloud was scanned at @xmath43 observation points with an integration time of @xmath44 s per position .
the temperature was calibrated by periodically switching to a stabilized noise tube , and the calibration of the noise tube was done by observing the standard area s7 before and after mapping each cloud .
the resulting temperature calibration factors are used to convert the measured voltage to brightness temperature including atmospheric extinction . for both clouds the rms noise level in the raw data was about @xmath45 mk .
we did not however correct the maps for effects from the antenna side lobes ; the resulting error is expected to be small ( the relative sensitivity of the first side lobes , separated from the main beam by @xmath46 , is about @xmath47 ) . at low velocities the observed spectra
are dominated by strong emission from foreground material .
the wlcs can be seen as broad emission structures extending up to @xmath48 . at both ends of the observed velocity interval
no neutral hydrogen emission was detected , and we have used these emission free regions to subtract a linear baseline from each spectrum with a baseline error of about @xmath49 k.
in fig . 1 we show average @xmath28-cm spectra of the clouds , i.e. the brightness temperature averaged over the observed area .
for the wlc at @xmath50 this average contains three distinct peaks .
the first one at @xmath51 is caused by the emission from the wlc itself , the second one at @xmath52 is partly from the wlc and also contains a strong contribution from an arm - like structure at @xmath53 which extends over the entire observed @xmath10-range ( compare fig .
the high velocity peak at @xmath54 is entirely due to hi near the galactic center unrelated to the wlc . in the composite hi spectra of the wlc at @xmath55 ( fig .
1 ) two peaks can be seen .
the one at @xmath56 is probably unrelated to the wlc .
the strong peak at @xmath57 is again due to a extended arm - like structure which is however more uniformly distributed over the observed area with a strong contribution from emission at @xmath58 .
this wlc is weaker than the wlc at @xmath50 and is harder to recognize in the average spectrum .
2 shows @xmath10-@xmath59-maps for both of the observed regions .
the wlcs are clearly seen and appear to cover a large velocity range of more than @xmath29 .
it is a remarkable fact that in both cases the clouds end at high velocity ridges extending over the entire observed longitude range .
these ridges can also be identified in the corresponding co - maps , although the ratio of intensities between the clouds and the ridges are much larger in co compared to hi , i.e. the ridges appear much weaker in co. contour maps of the temperature integrated over the velocity range from @xmath60 to @xmath61 as a function of @xmath10 and @xmath22 are shown in fig
. 3 for both wlcs in 21-cm as well as @xmath3co emission . both wlcs clearly stand out against the background , and their positions , sizes , and orientations obtained in hi coincide with the @xmath62 emission ; it is evident that we are looking at the same objects in both of these lines .
there is a small offset of about @xmath63 arcmin between the centers of the hi- and the @xmath62-map for the wlc at @xmath64 .
this is however insignificant since the shift is much less than the beam width for both observations .
it can also be seen that the wlc at @xmath7 is partly obscured by the hi - emission from the galactic plane but stands out more strongly in @xmath62 .
the wlcs are almost certainly not nearby objects .
the reason for this is that the observed velocity range of the wlcs seen in @xmath28-cm , as well as @xmath3co , is approximately 0200 km s@xmath9 , with a mean velocity of about 100 km s@xmath9 .
thus if these clouds were nearby objects on roughly circular orbit , then their orbital velocity projected along the line of sight would be small and therefore their observed internal random velocity should have comparable negative and positive values .
this is contrary to the observations .
the observed velocity of an object , on a circular orbit , in the galactic plane at longitude @xmath10 is @xmath65\sin\ell\ ] ] where @xmath66 and @xmath67 are the distances of the observer and the object from the galactic center , respectively , and @xmath68 is the angular rotation speed . taking @xmath69 to be approximately constant
we find the distance of the wlc at @xmath64 from the galactic center to be @xmath70 , and the one at @xmath71 to be @xmath72 .
thus the distance @xmath73 of the wlcs from us is approximately @xmath66 .
the total neutral hydrogen mass of the wlcs is estimated from the observed flux in @xmath28-cm using the formula @xmath74 where @xmath75 is brightness temperature in k , @xmath59 is the velocity in , and @xmath10 and @xmath22 are measured in degrees .
this expression follows from the number of hydrogen atoms along the line of sight as given by burton ( 1988 ) multiplied with the atomic hydrogen mass and the suface element @xmath76 . substituting the observed quantities in the above equation we find that the hi mass in the cloud regions , i.e. without subtracting any background emission , are about @xmath77 for the @xmath50 cloud , and @xmath78 for the cloud at @xmath7 .
these values were obtained by integrating the brightness temperature over the velocity range of 50200 @xmath79 in order to eliminate most of emission from foreground hi .
however , there is a substantial @xmath28-cm emission in this velocity range associated with the galactic plane at the angular location of both of the wlcs that must also be subtracted in order to determine the atomic mass of these clouds .
we have tried several different ways of estimating this _ background _ emission .
one method is to calculate the total hi emission outside the region occupied by each cloud , and multiply it by the fractional area covered by the cloud to yield the contribution of the background emission .
this procedure gives the _ corrected _ hi mass of the wlc at @xmath16 to be @xmath80 , and the wlc at @xmath81 to be about @xmath82 .
however , this method overestimates the _ background _ contribution , or underestimates cloud mass , since the emission from outside the cloud region is very nonuniform and is dominated by the strong @xmath28-cm emission from the galactic plane .
we have also subtracted average spectra from the cloud region before calculating the mass from the integral of equation ( 1 ) ; various areas outside the clouds were used to determine the average spectra .
when the galactic plane is not included in the area that is used in estimating the average spectra , then we obtain cloud mass that are approximately the averages of the two extreme values mentioned above , and is likely to be closer to the actual atomic mass of these clouds which lie outside of the galactic plane .
our best estimate of the hi mass of the wlc at @xmath16 is @xmath83 , and the wlc at @xmath81 is @xmath84 with an uncertainty of about a factor of 2 .
the hi mass is obviously the lower limit to the total cloud mass , most of which is molecular hydrogen .
neutral hydrogen in a typical giant molecular clouds is observed in a halo surrounding the cloud , with a ratio of hi to molecular mass of roughly 10% ( e.g. anderson , wannier & morris 1991 , elmegreen & elmegreen 1987 ) .
the neutral hydrogen mass of the wlcs is about 1% of their total mass as determined from the @xmath3co emission ( bitran 1987 ) .
it has been suggested that the ratio of h@xmath27 mass to @xmath3co emission for clouds near the galactic center is perhaps smaller by a factor of about 10 compared to the value in the solar neighborhood ( see stacy 1988 ) .
we also expect to see a larger fraction of the @xmath3co molecules in these clouds because of their large velocity dispersion . taking these factors into consideration
we conclude that the ratio of the hi mass to the total mass for the wlcs is not very different from giant molecular clouds .
we have mapped two molecular clouds at @xmath50 and @xmath7 in the @xmath28-cm line emission of neutral hydrogen .
these clouds show unusually large velocity dispersions of more than @xmath29 ( fwhm ) which has also been seen in lines of @xmath62 , @xmath4 , and @xmath85 .
the position , size , and orientation of both of the wide line clouds ( wlcs ) are the same in hi and @xmath3co maps .
the wlcs are very prominent features in the @xmath10-@xmath59-maps and show up as highly extended vertical bands .
these clouds lie about 0.3@xmath86 from the galactic plane and are almost certainly within about 1 kpc of the galactic center .
the mass of the clouds calculated from their @xmath62 emission , using the standard ratio of h@xmath27 to @xmath3co mass , is about @xmath87 ( bitran 1987 ) ; however for clouds in the galactic center region the ratio of h@xmath27 mass to @xmath3co luminosity is perhaps smaller by a factor of about 10 compared to the value in the solar neighborhood . from our hi observation
we estimate the atomic mass of the wlcs at @xmath50 and @xmath81 to be @xmath6 , and @xmath8 respectively , with an uncertainty of about a factor of @xmath63 due to the error in the determination of the background emission .
the ratio of hi to molecular mass of the wlcs does not appear to be abnormal for clouds near the galactic center , provided we use the ratio of h@xmath27 to @xmath3co preferred for this region .
it is striking that the wlcs extend in velocity from roughly 0 km s@xmath9 ( lsr ) to about 200 km s@xmath9 .
moreover , both of these clouds appear to end at high velocity ridges , in the @xmath10v plane , that are very bright in 21-cm and extend over the entire observed longitude range ( fig . 2 ) . these ridges are also visible in @xmath3co maps extending to several degrees in longitude ; however the ridges are much less prominent in co compared to hi .
boyce and cohen ( 1994a ) have found similar structures ( which they call filaments ) in their oh absorption - line measurements , and these filaments seem to connect several of the seven wlcs described in this survey .
we believe that these features provide important constraints to the physical origin of the enormous velocity dispersion of the wlcs .
* acknowledgment : * pk is grateful to pat thaddeus for suggesting this problem , numerous discussions and the use of his co data .
we are indebted to tom dame for many interesting discussions about molecular clouds and for suggesting several improvements to the presentation of our work .
this work was carried out while pk was visiting the institute for advanced study , princeton .
he is very grateful to john bahcall for his comments , support and hospitality .
anderson , b.g . , wannier , p.g .
, morris , m. 1991 , _ astrophys .
j. _ , * 366 * , 464 bitran , m.e .
1987 , _ ph.d . thesis _ , university of florida boyce , p.j . ,
cohen , r.j . , dent , w.r.f .
1989 , _ mon . not .
_ , * 239 * , 1013 boyce , p.j . ,
cohen , r.j .
1994a , _ astron
. astrophys .
_ , * 107 * , 563 boyce , p.j . ,
cohen , r.j .
1994b , _ iau symp .
_ * 169 * , 311 , eds .
l. blitz & p. teuben , kluwer press burton , w.b 1988 , in : _ galactic and extragalactic radio astronomy _ ( @xmath88 ed . ) , eds .
verschuur & k.i .
kellermann , p. 295
, springer - verlag , berlin .
dame , t.m . ,
elmegreen , b.g . ,
cohen , r.s . , thaddeus , p. , 1986
, _ astrophys .
j. _ , * 305 * , 892 dame , t.m . , ungerechts , h. , cohen , r.s .
, de geus , e.j . , grenier , i.a . , may , j. , murphy , d.c . ,
nyman , l .- . , thaddeus , p. 1987
, _ astrophys .
, * 322 * , 706 elmegreen , b.g . , & elmegreen d.m . , 1987 , apj 320 , 182 oort , j. 1977 , _ ann .
_ , * 15 * , 295 stacy , j.g . , bitran m.e . ,
dame , t.m . ,
thaddeus , p. 1988 , _ the galactic center ( iau symp . ) _ * 136 * ) , 157 , ed .
m. morris , reidel , dordrecht stark a.a , bania t.m .
1986 , _ astrophys .
j. _ , * 306 * , l17
* fig . 1 . * composite hi spectra of the wlcs as a function of velocity .
the solid lines show the brightness temperature averaged over the entire observed area .
the dashed lines are the corresponding variances .
the left / right panels are for clouds at @xmath89 * fig
longitude - velocity contour maps of both wlcs .
the maps show the brightness temperature integrated over the @xmath22-range ( @xmath90 ) for the cloud at @xmath50 , and over ( @xmath91 ) for the @xmath71 cloud .
contour levels are separated by @xmath92 k deg . *
longitude - latitude contour maps of both wlcs .
the solid lines represent the hi data ( brightness temperature ) , and the dashed lines show the @xmath3co data ( antenna temperature ) .
velocity integration ranges are ( @xmath93 ) and ( @xmath94 ) for the hi and @xmath62 maps , respectively .
no background subtraction has been performed for these maps .
the corresponding contour levels are separated by @xmath95 k ( hi ) and @xmath60 k ( @xmath62 ) for the wlc at @xmath50 , and @xmath60 k ( hi ) and @xmath96 k ( @xmath62 ) for the @xmath7 cloud .
the @xmath3co data is courtesy of p. thaddeus & t. dame , and was obtained using a 1.2 meter telescope ; for details please see dame et al .
( 1987 ) & bitran ( 1987 ) . | we have mapped two molecular clouds at ( @xmath0 ) = ( @xmath1 ) and ( @xmath0 ) = ( @xmath2 ) in 21-cm line and continuum emission .
these clouds show unusually large velocity dispersions of more than 100 ( fwhm ) which has also been seen in @xmath3co , @xmath4 , and cs emissions .
this dispersion is roughly an order of magnitude larger compared to giant molecular clouds . from our hi observation
we estimate that the atomic mass of the cloud at @xmath5 is about @xmath6 and the mass of the cloud at @xmath7 is @xmath8 .
the ratio of total molecular mass to atomic mass for these clouds appears to be normal for clouds near the galactic center .
the main peculiar feature of these clouds is their abnormally large velocity dispersion ; the extent in velocity is from about 0 km s@xmath9 ( lsr ) to 200 km s@xmath9 .
these clouds are connected , in the @xmath10v plane , to high velocity ridges that extend over several degrees of the longitude . these properties , we believe ,
provide important clues to the physical process responsible for their large velocity dispersion . |
an intense gamma - ray burst , grb030329 , allowed us to carry out the very detailed observations of optical , x - ray , and radio counterparts to the burst .
its photometry and spectroscopy led to the separation between the afterglow and sn contributions @xcite .
the sn component is similar to sn 1998bw ; an unusual type ic sn @xcite .
it makes clear that at least some grbs arise from core - collapse sne .
grb030329 was seen as a double - pulsed burst with the total duration time of @xmath1630 sec @xcite .
its total isotropic energy is @xmath17 erg .
the grbs with known redshifts distribute the emission energy @xmath18 over the wide range of energy such as @xmath19 erg @xmath20erg@xcite .
the duration times of grbs extend from @xmath1610ms to @xmath21 sec@xcite .
the light curves of grbs show really various profiles . however , to fully understand the emission mechanisms of the observed @xmath22-ray bursts and of the afterglows one must have a firm handle on the generation process of the @xmath22-ray bust . the fireball model of the @xmath22-ray bust has succeeded in understanding the characteristics of some light profiles of grbs @xcite .
its model was proposed to solve the compactness problem , that is , to understand both the short time variation of @xmath22-ray radiation and the optically thin process in the emitting region of the @xmath22-ray photons @xcite .
the solution of the fireball model requires the beaming jets with the emitting region relativistically moving to our direction .
the two processes are proposed for the formation of the jets , one is the magnetohydrodynamical process @xcite , and the other is the neutrino process @xcite .
further more the two different roles are expected for the neutrinos to form the jets .
the frequent collisions between neutrinos and antineutrinos produce the fireball with light elements which expands relativistically in a funnel around a rotation axis .
the other role is the acceleration of plasmas by the collisions of neutrinos with nucleons or electrons , which produces the neutrino driven jets .
the sufficient collisions between @xmath23 and @xmath24 at the central region of the accretion disk were proposed by @xcite .
they assumed the accretion disk is optically thin for the transfer of neutrinos .
but , @xcite showed that the accretion disk becomes optically thick for neutrino transfer and then the neutrinos trapped in the matter fall into a black hole .
however , the composition of accreting matter should be carefully considered .
the production rates of neutrinos and their opacities directly depend on the density of free nucleons .
@xcite were thought in question whether the neutrinos can cool the accretion disk .
they treated the accreting material in @xmath25 equilibrium and showed that there exist just a small number of free nucleons and then the neutrino processes are no longer effective in cooling of the disk .
this result is in a striking contrast to the previous works on the accretion disk in which the material almost consists of free nucleons produced by the photodisintegration process and the dominant cooling was carried out by neutrinos .
material at high temperature @xmath26k becomes in nuclear statistical equilibrium .
its composition of material depends on the number ratio of photon to baryon , @xmath27 , where @xmath28 is the mean molecular weight , @xmath29 and @xmath30 are the number densities of photons and baryons @xcite .
if @xmath31 is large , @xmath32 , the baryons form a dissociated gas of nucleons and alpha particles . however ,
if @xmath31 is of order unity or less , iron - group nuclei are typically favored .
the ratio @xmath31 is expressed as @xmath33 in terms of the temperature @xmath34 in units of @xmath35k and the density @xmath36 in units of @xmath37g/@xmath9 . in the case of a typical accretion with @xmath38/sec and @xmath39 , the temperature and
the density become are @xmath40(k ) and @xmath41(g/@xmath9 ) at the inner region of the accretion disk .
then the values of @xmath31 become as @xmath42 .
one should treat the accreting matter with a large abundance of heavy nuclei .
we determine the densities of free neutron and proton , alpha particle , and heavy nuclei by using the equation of state for hot , dense matter given by @xcite .
the changes of lepton number per a baryon , @xmath43 , are calculated through the reactions of neutrinos with these baryons .
the thermal equilibrium structure of the accretion disk should be determined in the frame of a general relativity .
@xcite showed that the core - collapse in the evolved star with the main sequence mass , @xmath44 , forms a black hole at the center with the rotating disk balanced with gravity when the evolved star has the some large specific angular momentum .
then the specific angular momentum parameter @xmath45 of the falling matter into a black hole becomes larger to be @xmath46 .
the falling velocity of the rotating disk is required to be very small in comparison with light velocity , since the time duration of some grbs is a few seconds , i.e. , @xmath47 , where @xmath48 is the gravitational radius of a black hole and @xmath49 is the light velocity .
the accretion disk with slow falling velocity should be investigated in the frame of rotating spacetime .
the heating rate due to viscosity is proportional to the shear stress @xmath50 . in the static spacetime
its stress increases with the differential rotation of the accreting flow @xmath51 , where @xmath52 is the angular velocity of the falling flow . in the rotating spacetime the differential rotation of the fluid observed in the frame co - rotating with spacetime
is physically meaningful @xcite . in the extreme kerr
the relative angular velocity @xmath53 has the maximum at the radius @xmath54 when @xmath52 is the angular velocity of the keplerian orbit and @xmath55 is the angular velocity of spacetime .
it is convenient for the analysis of energy balance and of the transfer of angular momentum to introduce the orbiting frame with orthogonal tetrad which rotates with keplarian velocity . in this paper
the shear stress @xmath50 and the transfer of angular momentum are solved by using the orbiting frame .
the power by advection contributes seriously to the energy balance in the accretion disk .
the previous works on the accretion disk with neutrino loss had been studied in the frame of an advection - dominated accretion flow @xcite .
the role of the advection in the energy balance depends on the velocity profile since the power by the advection is expressed as @xmath56 , where @xmath57 and @xmath58 are the pressure and velocity of the flow .
when @xmath59 , the accreting matter is cooled by the expansion of fluid . on the other hand when @xmath60 , it is heated by the compression like as in the core - collapse in a supernova explosion .
the radial velocity profile of the accreting flow is determined by the energy balance of the disk and by the transfer of angular momentum .
the scale height at the inner portion of the disk with the temperature @xmath61k is very thin .
the angular velocity of the accretion disk is then approximated to be keplerian . we determine the radial profile of the velocity by solving the transfer of angular momentum with the thermal energy balance of the disk and evaluate the efficiency of advection in the energy balance .
the thermal stability of the accretion disk might depend on the scattering optical depth of @xmath5 .
the thermal energy in the disk is carried out mainly by @xmath5 .
the collision frequency of @xmath5 with the nucleons or electrons reaches to @xmath62 within the disk for the typical case of the accretion .
the escaping time from the disk is then @xmath63(ms ) , where @xmath64 is the height of the disk .
the thermal energy produced by viscous heating stays within the disk in the time @xmath65 and might induce the thermal instability .
we shall briefly analyze the thermal instability .
if the luminosity of each flavor of neutrino and its mean energy are observed the physics of grbs will be cleared .
we shall calculate the luminosity @xmath66 and the mean energy @xmath67 for each type of neutrino @xmath68 emitted from the accretion disk .
the diagram determining the mass of a central black hole , @xmath69 , and its specific angular momentum , @xmath45 , from the observations of neutrinos is presented . in this paper
we study the accretion of hot dense matter onto a black hole . in sect .
2 we describe the equation of state for the accreting hot , dense matter . the reactions of neutrinos and the changes of neutrino fractions are presented in section 2.2 .
the relativistic model of accretion disk with neutrino loss is given in sect .
3 . in sect .
4 the results of thermally equilibrium disk are given .
the profiles of compositions along the accretion flow are shown in section 4.1 .
the flow structure and chemical potentials of nucleons are depicted in section 4.2 .
the emissivities and opacities of neutrinos are presented in section 4.3 .
the luminosity and mean energy of neutrinos versus mass and angular momentum of a black hole are discussed in section 4.4 . in sect .
5 the thermal stability of the accretion disk is investigate .
the neutrino transfer in homogeneous disk , changing rates of positrons and the dynamical properties of accretion disk are presented in appendix .
the cross sections of neutrinos , opacities , emission rates and the reaction rates of neutrinos are summarized in tables , which are shown in appendix .
it is important for the study of thermal and dynamical properties of massive accretion disks precisely to determine the fraction of free nucleons in the accreting matter .
its fraction provides the emissivities and absorptions of neutrinos .
the mass fraction of free nucleons depends on the photon - to - baryon ratio @xmath70@xcite , @xmath71 where @xmath72 is avogadro s number .
we compute @xmath70 in terms of the temperature @xmath73 and density @xmath74 as @xmath75 where @xmath76 is the helicity for photons .
we evaluate the ratio @xmath70 for the neutrino dominated accreting flow ( ndaf ) given by @xcite .
the loci of ndaf with @xmath77sec@xmath2 and the contours of @xmath78 in the @xmath74 @xmath73 plane are shown in fig 1 .
the ratio @xmath70 on the flow is @xmath79 , showing little number density of photons . on the other hand
the previous works on ndaf without @xcite adopted the approximate formula for the mass fraction of free nucleons , @xmath80 , @xcite : @xmath81 the contour of @xmath82 is also plotted in fig .
1 . at the inner side of ndaf its fraction , @xmath80 , reaches to unity .
we also calculated the flow profile of the standard accretion disk with the keplerian angular momentum in which the fraction of free nucleons is provided through this formula , @xmath80 .
the accreting velocity of the standard disk is much less than that of ndaf , which produces the denser profile of the accretion flow for the same accretion rate , @xmath83sec@xmath2 .
then its ratio @xmath70 is very minute , @xmath84 .
thus it is necessary to handle the accreting matter with an appreciable abundance of heavy nuclei .
( g @xmath85 ) @xmath73(k ) plane .
ndaf is the locus of the neutrino - dominant accretion flow with @xmath86 sec@xmath2 and @xmath39 .
sad1 is the locus of the standard accretion disk with keplarian angular momentum where heavy nuclei are ignored .
sad2 is the locus of the standard accretion disk with heavy nuclei .
the lines with the constant photon - to - baryon ratio @xmath87 and those with constant mass fraction of free nucleons @xmath88 are plotted .
the phase boundary between heavy nuclei and uniform evaporated matter for the case @xmath89 is described by the dot - dashed line with the label @xmath90.,width=302,height=226 ] we adopt a simplified , analytical equation of state for hot , dense matter given by @xcite , in which the heavy nuclei are treated as being in a b.c.c . lattice . in accordance with the wigner - seitz approximation each heavy ion is considered to be surrounded by a charge - neutral spherical cell consisting of a less dense vapor of neutrons , protons and alpha particles as well as electrons .
the volume of this cell is given by @xmath91 , where @xmath92 is the number density of heavy nuclei .
we construct a three - dimensional table of the relevant thermodynamic quantities , i.e. , @xmath93 , as a function of the inputs ( @xmath94 ) , where @xmath95 are the number densities of free neutron , free proton , @xmath96 particle , @xmath97 is the difference of chemical potential between the neutron and the proton , @xmath98 and @xmath99 are the mean mass number and proton number of heavy nuclei .
the contours of chemical potential , @xmath100 , are depicted in fig.9 .
[ cols="<,<,<,<",options="header " , ] is the unit step function , @xmath101 if @xmath102 , and @xmath103 if @xmath104 . ]
the analytic formulas of neutrino s transfers are derived by using the simplified treatment of radiative transfer given by @xcite . introducing the two streams , with intensities
@xmath105 moving at an angle @xmath106 relative to @xmath107 and @xmath108 , define the net flux density @xmath109 and the total intensity @xmath110 such as @xmath111 and @xmath112 @xcite .
the transfers of neutrinos are then described by the production rate of neutrinos , @xmath113 , and the absorption opacity , scattering opacity and the total opacity , @xmath114 and @xmath115 , as @xmath116 the boundary condition at the center of the disk is @xmath117 . to derive the simplest solution ,
let us assume that @xmath118 are independent of @xmath119 @xcite
. equations ( [ i1 ] ) and ( [ i2 ] ) can then be integrated to give @xmath120 where @xmath121 and @xmath122 are the values of @xmath113 and @xmath110 at @xmath123
. the boundary condition at the surface of the disk is @xmath124 , i.e. , @xmath125 . applying this boundary condition
we find @xmath126 where @xmath127 is the absorbing optical depth , @xmath128 , and @xmath129 is the total optical depth , @xmath130 .
the escaping flux density of any sort of neutrino , @xmath68 , is then given by @xmath131 where @xmath132 and @xmath133 are the total optical depth and the absorbing one of @xmath68 , and @xmath134 is the production rate of @xmath68 per unit volume .
the number density of positrons which are equilibrium with thermal photons is described as @xmath135 ( landau and lifshitz 1964 ) , where the brackets indicate the mean values of the phase space blocking factors .
these factors are evaluated with the average energy of electrons and positrons produced by @xmath136-pair creation@xcite : @xmath137 } \bigr\}^{-1 } .\ ] ] the average energies , @xmath138 and @xmath139 , are approximately given by @xmath140 where @xmath141 and @xmath142 are the energy and number densities of positrons .
when the fraction of positrons becomes large , @xmath143 , the changing rate of positron number , @xmath144 , due to the reactions , @xmath145 , @xmath146 , can not be negligible .
we calculate the changing rate of the positrons along the flow by using the following equation : @xmath147 where @xmath148 is the net production rate of positrons , @xmath149 .
dynamically equilibrium structure of the disk is given by the euler equation .
the equation of motion of the fluid , @xmath150 , for the steady , axially symmetric and rotating ideal fluid can be expressed in the total differential form : @xmath151 the equipressure surfaces are given by the equation @xmath152constant .
the quantity @xmath153 is equal in the newtonian limit to the total potential ( gravitational plus centrifugal ones ) expressed in the units of @xmath154 @xcite .
we solve the disk structure in terms of cylindrical coordinates @xmath155 and @xmath156 instead of radial coordinates which are combined through the relations , @xmath157 if no pressure force exerts in the @xmath158 direction , the equation @xmath159 determines the velocity of the geodesic circular orbit , @xmath160 .
let s introduce the new variables , @xmath161 defined by @xmath162 , @xmath163 .
the following functions are defined by : @xmath164 the circular velocity is then given by @xmath165 this velocity at the equatorial plane is expresses as @xmath166 the vertical force is described by the gradient of the equipotential , @xmath167 .
the first order of the expansion of @xmath168 in @xmath169 gives the approximate expression such as @xmath170 \nonumber \\ & - & ( 1 + \frac { \omega}{\omega } + \frac{1}{v_{(\varphi ) } v_\omega})\frac{\acute{a}}{\psi } + \frac{1-m / r}{v_{(\varphi ) } v_\omega d } \biggr\ } , \nonumber \\ & \theta(r)&=1 + \frac{am^{1/2}}{r^{3/2 } } , \quad d=1-\frac{2m}{r } + \big(\frac{a}{r}\big)^2,\nonumber \\
& \psi & = 1 + \big(\frac{a}{r}\big)^2\big(1 + 2\frac{m}{r}\big ) , \nonumber \\ & \acute{a } & = 2 + ( 4- 3\frac{m}{r})\big(\frac{a}{r}\big)^2 + \big(\frac{a}{r}\big)^4 , \nonumber \\ & v_\omega & = \omega e^{\psi-\nu}=v_{(\varphi ) } + \omega e^{\psi-\nu } \label{psi}\end{aligned}\ ] ] | the accretion of hot , dense matter which consists of heavy nuclei , free nucleons , degenerated electrons , photons and neutrinos is studied .
the composition of free nucleons and their chemical potentials are provided through the equation of state for hot , dense matter proposed by @xcite .
the numbers of leptons are calculated through the reactions of neutrinos and through the simplified transfers of neutrinos in the accreting matter .
the thermally equilibrium fluid structure of the accretion is solved .
the angular momentum transfer in the disk is analyzed with the shear stress which is driven in the frame of the general relativity .
when the mass of a central black hole and the accretion rate are selected as @xmath0 and @xmath1sec@xmath2 , which provide the typical luminosity of gamma ray bursts(grbs ) , the fraction of free protons in the accreting matter becomes very small , @xmath3 , while that of free neutrons is closer to unity , @xmath4 .
then the antielectron neutrinos @xmath5 can freely escape through the disk but the electron neutrinos @xmath6 are almost absorbed in the disk . thus the frequent collisions of @xmath5 with @xmath6 over the disk could nt be occurred .
the accretion disk is cooled mainly by @xmath5 , which suppresses the increase of temperature and increases the density in the accreting matter such as @xmath7k and @xmath8g/@xmath9 at the inner side of the disk .
the scattering optical depths of @xmath5 and @xmath6 then reach to be very large , @xmath10 .
thus the accretion disk could be thermally unstable within the duration of diffusion time of neutrinos , @xmath11(ms ) . the ram pressure produced by many scattering of @xmath5
are very strong , which might produce the neutrino driven wind or jets .
the luminosity and mean energy of neutrinos , @xmath12 , @xmath13 , ejected from the disk increases with the specific angular momentum of a black hole .
the mean energy is inversely proportional to the central mass , @xmath14 , while the luminosity @xmath12 is independent of its mass , @xmath15 .
the diagram determining the physical properties of a central black hole from the observation of each flavor of neutrino is proposed .
[ firstpage ] accretion , accretion disks black hole physics gamma rays : bursts neutrinos . |
galactic studies have been one of the main topics of the tartu observatory over the whole period of its existence .
these studies began with the work by f.g.w .
struve on double stars and measurements of stellar parallaxes in early 19th century , when tartu observatory was founded in the kaiserliche universitt zu dorpat .
modern era of galactic studies began about 95 years ago when ernst pik determined the dynamical density of matter in the disk of the galaxy in 1915 , the distance to the andromeda nebula in 1922 , and found main principles of stellar structure and evolution in 1938 .
his student grigori kuzmin developed principles of galactic modeling and calculated the local density of matter near the sun , suggesting the absence of local dark matter in large quantities .
the present generation of astronomers follows these traditions . in the following
i give an overview of the development of the ideas of the structure of galaxies and the universe over the whole period of activity of the tartu observatory .
the period of last 50 years is described in more detail .
it was a tradition in classical universities to have an astronomical observatory .
tartu university observatory was built in 1810 , its astronomical instruments were installed by astronomy and mathematics student friedrich georg wilhelm struve ( 1793 - 1864 ) , who after defending his phd was nominated to professor of astronomy and mathematics and director of the observatory .
struve understood well the needs of contemporary astronomy and geodesy .
the main goal at this time was to understand the nature of stars and stellar systems .
struve started to measure stellar positions , which was basis for better description of the universe .
he soon realised that conventional astronomical instruments , available in tartu , were not sufficient to solve most interesting problems .
thus he applied to get support to buy a new larger telescope .
his efforts succeeded and in 1825 a 9 inch fraunhofer refractor was installed in observatory . for about 15 years
this was the largest and best telescope of this type in the world .
struve made excellent use of the new telescope .
first he made a survey of the whole northern sky and published a catalogue of double stars detected .
a catalogue followed , which contained exact measurements of positions of double stars@xcite .
after repeated observations have been made it is possible to calculate orbits of double stars and to get information on their masses .
this is a foundation of the new astronomy astrophysics . as a by - product of the measurements of double stars
he also published his determination of the distance of a star vega . with this measurement
it was finally demonstrated that stars are distant suns .
struve double star catalogue is used even in present days , his achievements have found place in astronomy textbooks .
another important scientific - historic achievement of f.g.w .
struve was the astronomic - trigonometric measurement of tartu ( struve ) meridian arc ( 1816 - 1852 ) .
this measurement was made together with carl friedrich tenner ( 1783 - 1859 ) .
they succeeded in determining the almost 3000 km long section of the meridian arc between the mouth of danube and the arctic ocean with the accuracy of @xmath0 m @xcite .
the comparison of data of different arc sections indicated that the length of arc , corresponding to one degree of latitude , increases towards the pole , i.e. the earth is flattened .
the measurements were used by f.w .
bessel for determination of spheroidal earth s new parameters .
struve geodetic meridian arc is included to the unesco world heritage list . in 1839 f.g.w .
struve was appointed director of the new pulkovo observatory .
the next director of tartu observatory johann heinrich mdler ( 1794 - 1874 ) was interested in the dynamics of the milky way . in his book
centralsonne he tried to determine the center of the galaxy using proper motions of stars .
the accuracy of data was not sufficient for this task , the method itself is correct .
the founder of the modern astronomy school in tartu university was ernst pik ( 1893 - 1985 ) .
he started his astronomical career as student of the moscow university , and made his principal discoveries as observator of the tartu university observatory .
one of his first scientific papers was devoted to the question : what is the density of matter near the plane of the milky way stellar system ?
his calculations showed that the gravitating matter density can be fully explained by observable stars , and that there is no evidence for hypothetical matter near the symmetry plane of the galaxy .
this work seems to be the first attempt to address the dark matter problem@xcite .
next pik devoted his attention to spiral nebulae .
their nature was not known at this time : are they gaseous objects within the milky way or distant worlds similar in structure to our galaxy ?
immediately when relative velocity measurements of the central part of andromeda nebula m31 had been published , pik developed a method how to use this information to estimate the distance to m31 .
his result was 440 kpc ( about 1.5 million light years)@xcite . with this work he solved the problem of the nature of spiral nebulae , and showed that the universe is millions of times larger than our milky way system .
another unsolved problem at this time was the source of stellar energy and the evolution of stars .
already in early 1920s pik demonstrated , using very simple physical considerations , that gravitational contraction and radioactivity can not be the main sources of stellar energy .
he concluded , that some unknown subatomic processes in central regions of stars must be responsible for stellar energy .
when data on atomic structure were available , he developed a detailed theory of stellar evolution , based on nuclear reactions in stellar interiors . here under very high temperature hydrogen burns to helium , and huge amounts of energy will be released @xcite .
the efficiency of these reactions has just been estimated based on atomic theory , thus pik was able to calculate ages of stars .
he demonstrated , that hot giant stars are so luminous that their energy sources will be exhausted within several tens millions years .
their presence shows that they have been recently formed in other words , star formation is a process which takes place even today .
these results revolutionised our understanding of stars and their evolution .
presently they are accepted by the astronomical community .
professor of astronomy in tartu university rootsme ( 1885 - 1959 ) applied these ideas to kinematics of stars to find the sequence of formation of different stellar populations@xcite .
similar ideas were developed independently by eggen , lynden - bell and sandage@xcite .
pik s student grigori kuzmin ( 1917 - 1988 ) continued his mentors work on studying the structure of the galaxy . in his phd thesis he developed further pik s method to derive the density of matter in the galaxy and constructed a mathematical model of the galaxy ,
much more advanced than previous ones@xcite .
in contrast to earlier models by oort and schmidt he used ellipsoids of variable density , and applied the theory first to the andromeda galaxy and then to our own galaxy .
one of central problems in modeling the galaxy was the density of matter near the sun .
in 1930s famous dutch astronomer jan oort@xcite studied this problem and found , that in addition to ordinary stars there must be in the galaxy an unknown population , so that the total local density exceeds twice the density of visible matter .
kuzmin found that the method used by oort is not very accurate and that there is no indication for the presence of local dark matter in the galaxy .
later two students of kuzmin , heino eelsalu ( 1930 - 1998 ) @xcite and mihkel jeveer ( 1937 - 2006 ) @xcite , reanalysed the problem , using different data and methods , and confirmed his results .
the discrepancy between results of tartu astronomers and the rest of the world continued until 1990s , when finally modern data confirmed that kuzmin was right ( gilmore , wyse & kuijken@xcite ) .
thus we came to the conclusion that _ there is no evidence for the presence of large amounts of dark matter in the disk of the galaxy_. if there is some invisible matter near the galactic plane , then it consists probably of low mass stars or jupiters , which have been formed from the flat gas population .
kuzmin also developed a method how to describe more accurately the kinematics of stellar populations , using three integrals : the energy and mass conservation integrals , and a third integral , which allows the existence of three - axial velocity ellipsoids of stellar populations@xcite . from his model
it follows that orbits of stars in galaxies lie in a toroidal volume .
numerical modeling of star orbits in realistic mass models have confirmed kuzmin s prediction .
the work by pik , rootsme , kuzmin and their students formed the basis of a concept of the structure and evolution of stellar populations in galaxies , which is rather close to the presently accepted picture . in early 1960s
i was interested in the problem too .
as new observational data arrived , the need for a better and more accurate model of our galaxy and other galaxies was evident .
detailed local structure is known only for our own galaxy , and global information on stellar populations is better known for external galaxies , thus it is reasonable to investigate the structure of our galaxy and other galaxies in parallel . also there was a need for a more detailed method of the construction of composite models of galaxies .
this goal was realised in a series of papers in tartu observatory publications@xcite , a summary of the method was published in english@xcite .
a natural generalisation of classical galactic models is the use of all available observational data for spiral and elliptical galaxies , both photometric data on the distribution of colour and light , and kinematical data on the rotation and/or velocity dispersion .
further , it is natural to apply identical methods for modeling of galaxies of different morphological type ( including our own galaxy ) , and to describe explicitly all major stellar populations .
the main principles of model construction were : ( 1 ) galaxies can be considered as sums of physically homogeneous populations ( young flat disk , thick disk , core , bulge , halo ) ; ( 2 ) physical properties of populations ( mass to luminosity ratio @xmath1 , colour ) should be in agreement with models of physical evolution of stellar populations ; ( 3 ) the density of a population can be expressed as ellipsoids of constant flatness and rotational symmetry ; ( 4 ) densities of populations are non negative and finite ; ( 5 ) moments of densities which define the total mass and effective radius of the galaxy are finite .
it was found that in a good approximation densities of all stellar populations can be expressed by a generalised exponential law : @xmath2 $ ] , where @xmath3 is the central density , @xmath4 is the distance along the major axis , @xmath5 is the axial ratio of the equidensity ellipsoid , @xmath6 is the core radius ( @xmath7 is the harmonic mean radius ) , @xmath8 and @xmath9 are normalising parameters , depending on the structural parameter @xmath10 , which allows to vary the density behaviour with @xmath11 . the cases @xmath12 and @xmath13 correspond to conventional exponential and de vaucouleurs models , respectively .
this density law ( called einasto profile ) was first applied to find a composite model of the galaxy@xcite , based on the new system of galactic constants , using all available data ( einasto and kutuzov@xcite ) .
next the method was applied to the andromeda galaxy @xcite .
a central problem in galactic modeling is the correct estimation of the mass - to - luminosity ratio of populations .
this ratio depends on the evolutionary history of the population and on its chemical composition . in order to bring these ratios for different populations to a coherent system , a model of physical evolution of stellar populations
was developed @xcite .
the model used as input data the evolutionary tracks of stars of various composition ( metallicity ) and age ; the star formation rate as a function of stellar mass was accepted according to salpeter@xcite , with a low mass limit of star formation of @xmath14 .
the model yielded a continuous sequence of population parameters as a function of age ( colour , spectral energy distribution , @xmath1 ) , see fig .
[ fig : evolution ] .
the results of modeling stellar populations were calibrated using direct dynamical data for star clusters and central regions of galaxies ( velocity dispersions ) by einasto & kaasik@xcite .
these data supported relatively high values ( @xmath15 ) for old metal
rich stellar populations near centres of galaxies ; moderate values ( @xmath16 ) for discs and bulges ; and low values ( @xmath17 ) for metal poor halo type populations .
modern data yield lower values , due to more accurate measurements of velocity dispersions in clusters and central regions of galaxies , and rotation data on bulge dominated s0 galaxies .
i had a problem in the modeling of m31 .
if rotation data were taken at face value , then it was impossible to represent the rotational velocity with the sum of known stellar populations . the local value of @xmath1 increases towards the periphery of m31 very rapidly , if the mass distribution is calculated directly from rotation velocity . all known old metal poor halo type stellar populations have a low @xmath18 .
in contrast , on the basis of rotation data we got @xmath19 on the periphery of the galaxy near the last point with measured rotational velocity .
i discussed the problem with my collaborator enn saar .
he suggested to abandon the idea , that only stellar populations exist in galaxies .
instead it is reasonable to assume the existence of a population of unknown nature and origin , and to look which properties it should have using available data on known stellar populations .
so i calculated a new set of models for m31 , our galaxy and several other galaxies of the local group , as well as for the giant elliptical galaxy m87 in the virgo cluster .
in most models it was needed to include a new population in order to bring rotation and photometric data into mutual agreement . to avoid confusion with the metal - poor halo
the new hypothetical population was called corona
. results of these calculations were reported in the first european astronomy meeting in athens in september 1972@xcite .
the conclusions were : ( 1 ) there are two dark matter ( dm ) problems : the local dm near the galactic plane , and the global dm forming an extended almost spherical population ( corona ) ; ( 2 ) the local dark matter , if it exists , must be of stellar origin , as it is strongly concentrated to the galactic plane ; ( 3 ) the global dark matter is probably of non - stellar origin .
available data were insufficient to determine outer radii and masses of coronas .
preliminary estimates indicated that in some galaxies the mass and radius of the corona may exceed considerably the mass and radius of known stellar populations .
arguments for the non - stellar origin of galactic coronas were the following .
( 1 ) physical and kinematical properties of the stellar populations depend almost continuously on the age of the population , the oldest have the lowest metallicity and @xmath1-ratio , and there is no place where to put the corona into this sequence@xcite .
( 2 ) since the @xmath1 value and spatial distribution of the corona differ so much of similar properties of known stellar populations , the corona must have been formed much earlier than all known populations ; the total mass of the corona exceeds masses of known populations by an order of magnitude , thus we have a problem : how to transform in an early stage of the evolution of the universe most of gas to the coronal stars ?
it is known that star formation is a very inefficient process , as in a star - forming gaseous nebula only about 1 % of matter transforms to stars@xcite .
( 3 ) due to the large size of the corona , coronal stars must have in the vicinity of the sun much higher velocities than all other stars , but no extremely high - velocity stars have been found by jaaniste and saar@xcite .
( 4 ) luminosity decreases in outer regions rapidly , therefore , if the matter is in stars , they must be of very low luminosity .
the presence of low - luminosity stars in outer galactic regions without bright ones would require a process of large - scale segregation of stars according to mass ( low - luminosity stars have small masses ) , but this is highly improbable@xcite .
the hidden matter can not be in the form of neutral gas , since this gas would be observable@xcite .
for these reasons , i assumed that coronas may consist of hot gas .
soon it was clear that a fraction of coronal matter is indeed gaseous@xcite , however not all . to find the radii and masses of galactic coronas
more distant test objects are needed .
one possibility is the use of companion galaxies .
if coronas are large enough , then in pairs of galaxies the companion galaxy is moving inside the corona , and it can be considered as a test particle to measure the gravitational attraction of the main galaxy .
mean relative velocities , calculated for different distances from the main galaxy , can be used instead of rotation velocities to find the mass distribution of giant galaxies .
a collection of 105 pairs of galaxies yield following results : radii and masses of galactic coronas exceed radii and masses of stellar populations of galaxies by an order of magnitude ! together with a. kaasik and e. saar we calculated new models of galaxies including dark coronas .
results were reported in the caucasus winter school in january 1974 , and published in nature@xcite .
our data suggest that all giant galaxies have massive coronas of some unknown origin ( dark matter ) , the total masses of galaxies including dark coronas exceed masses of known populations by an order of magnitude .
it follows , that dark matter is the dominating component in the whole universe .
similar results have been obtained by ostriker , peebles and yahil@xcite .
additional arguments supporting the physical connection between main galaxies and their companions were found from the morphology of companions@xcite . in january 1975
the first conference on dark matter was held in tallinn , estonia .
the rumour on dark matter had spread around the astronomical community and , in contrast to conventional local astronomy conferences , leading soviet astronomers and physicists attended .
the main topics was the possible nature of the dark matter .
it was evident that a stellar origin is almost excluded@xcite , but a fully gaseous corona also has difficulties , as shown by komberg and novikov@xcite .
the problems with baryon nucleosynthesis constraints were discussed by zeldovich .
so the nature of dm was not clear .
the next dark matter discussion was in july 1975 during the third european astronomical meeting in tbilisi , georgia , where a full session was devoted to the dark matter problem . here
the principal discussion was between the supporters of the classical paradigm with conventional mass estimates of galaxies , and of the new one with dark matter .
the major arguments supporting the classical paradigm were summarised by gustav tammann@xcite .
the most serious arguments were : _ big bang nucleosynthesis suggests a low - density universe with the density parameter @xmath20 ; the smoothness of the hubble flow also favours a low - density universe . _
dark matter problem was also discussed during the iau general assembly in grenoble , 1976 . here
arguments for the non stellar nature of dark coronas were again presented@xcite .
it was clear that by sole discussion the presence and nature of dark matter can not be solved , new data and more detailed studies were needed .
a very strong confirmation of the dark matter hypothesis came from new extended rotation curves of galaxies .
vera rubin and her collaborators developed new sensitive detectors to measure optically the rotation curves of galaxies at very large galactocentric distances .
their results suggested that practically all spiral galaxies have extended flat rotation curves@xcite .
the internal mass of galaxies rises with distance almost linearly , up to the last measured point , see fig .
[ fig : bosma_rubin ] . at the same time measurements of a number of spiral galaxies with the westerbork synthesis radio telescope were completed , and mass distribution models were built , all - together for 25 spiral galaxies by bosma@xcite , see fig .
[ fig : bosma_rubin ] .
observations confirmed the general trend that the mean rotation curves remain flat over the whole observed range of distances from the center , up to @xmath21 kpc for several galaxies .
these new observations confirmed the presence of dark coronas of galaxies .
the nature of the coronas was still unclear , and the difficulties discussed in tallinn and tbilisi were not clarified . in late 1970s suggestions
were made that some sort of non - baryonic elementary particles may serve as candidates for dark matter particles .
there were several reasons to search for non - baryonic particles as a dark matter candidate .
first of all , no baryonic matter candidate did fit the observational data .
second , the total amount of dark matter is of the order of 0.20.3 in units of the critical cosmological density , whereas the nucleosynthesis constraints suggest that the amount of baryonic matter can not be higher than about 0.04 of the critical density .
a very important observation was made , which caused doubts to the baryonic matter as the dark matter candidate . in 1964 cosmic microwave background ( cmb )
radiation was detected .
initially the universe was very hot and all density and temperature fluctuations of the primordial gas were damped by very intense radiation . at a certain epoch called recombination the gas became neutral , and density fluctuations in the gas had a chance to grow by gravitational instability . but gravitational clustering is a very slow process . in order to have time to build up all observed structures the amplitude of initial density fluctuations at the epoch of recombination must be of the order of @xmath22 of the density itself .
density fluctuations are of the same order as temperature fluctuations , and astronomers started to search for temperature fluctuations of the cmb radiation .
none were found .
as the accuracy of measurement increased , lower and lower upper limits for the amplitude of cmb fluctuations were obtained . in late 1970s
it was clear that the upper limits are much lower than the theoretically predicted limit @xmath22 .
then astronomers recalled the possible existence of non - baryonic particles , such as heavy neutrinos .
this suggestion was made independently by several astronomers ( @xcite ) .
if dark matter is non - baryonic , then this helps to explain the paradox of small temperature fluctuations of cosmic microwave background radiation .
density perturbations of non - baryonic dark matter start growing already during the radiation - dominated era , whereas the growth of baryonic matter is damped by radiation .
if non - baryonic dark matter dominates dynamically , the total density perturbations can have an amplitude of the order @xmath22 at the recombination epoch , which is needed for the formation of the observed structure of the universe .
the evolution of perturbations in a neutrino - dominated dark matter medium was discussed in a conference in tallinn in april 1981 ( this conference was probably the first one devoted to the astro particle physics ) .
numerical simulations made for a neutrino - dominated universe were made by a number of astronomers .
these calculations demonstrated some weak points in the scenario : large - scale structures ( superclusters ) form too late and have no fine structure as observed in the real universe .
a new scenario was suggested , among others , by bond , szalay & turner@xcite ; here hypothetical particles like axions , gravitinos or photinos play the role of dark matter .
numerical simulations of structure evolution for neutrino and axion dominated universe were made and analysed by melott et al.@xcite .
all quantitative characteristics ( connectivity of the structure , multiplicity of galaxy systems , correlation function ) of this new model fit the observational data well .
this model was called subsequently the cold dark matter ( cdm ) model , in contrast to the neutrino based hot dark matter model .
presently the cdm model with some modifications is the most accepted model of the structure evolution ( blumenthal et al.@xcite ) .
after my talk in the caucasus winter school zeldovich turned to me and offered collaboration in the study of the universe .
he was developing a theory of the formation of galaxies the pancake theory@xcite ; an alternative whirl theory was suggested by ozernoy , and a third theory of hierarchical clustering by peebles .
zeldovich asked for our help in solving the question : can we find some observational evidence which can be used to discriminate between these theories ?
initially we had no idea how we can help zeldovich .
but soon we remembered our previous experience in the study of galactic populations : kinematical and structural properties of populations hold the memory of their previous evolution and formation ( rootsme@xcite , eggen , lynden - bell & sandage@xcite ) .
random velocities of galaxies are of the order of several hundred km / s , thus during the whole lifetime of the universe galaxies have moved from their place of origin only about 1 @xmath23mpc ( we use the hubble constant in units @xmath24 km s@xmath25 mpc@xmath25 ) . in other words if there exist some regularities in the distribution of galaxies , these regularities must reflect the conditions in the universe during the formation of galaxies . in our work to solve the zeldovich question we had a close collaboration with his team . in 1975 doroshkevich , shandarin and novikov@xcite
obtained first results of numerical simulations of the evolution of particles according to the theory of gravitational clustering , developed by zeldovich@xcite .
this was a 2dimensional simulation with @xmath26 particles ( see fig .
[ fig : zeld - nbody ] ) . in this picture
a system of high and low
density regions was seen : high density regions form a cellular network , which surrounds large under dense regions .
one of our challenges was to find out , whether the real distribution of galaxies showed some similarity with the theoretical picture .
now we had a guiding idea how to solve the problem of galaxy formation : _ we have to study the distribution of galaxies on larger scales_. both our galactic astronomy and theoretical cosmology teams participated in the effort to find the distribution of galaxies and their systems in space .
one approach was the study of the distribution of nearby zwicky clusters .
many bright galaxies of nearby zwicky clusters had at this time measured redshifts , so we hoped to determine the distribution of clusters and to find some regularities there . to see the distribution better , we built in the office of saar and jaaniste a 3dimensional model from plastic balls .
some regularity was evident : there were several clusters of zwicky clusters
superclusters , one of them in the perseus region .
but too many clusters had no galaxies with measured redshifts , so it was difficult to get an overall picture .
zwicky nearby clusters were used several years later by einasto , jeveer & saar@xcite , when more redshifts were available and for the rest photometric distances were estimated .
a different approach was used by mihkel jeveer .
he used wedge diagrams .
his trick was : he made a number of wedge diagrams in sequence for fixed @xmath27 and @xmath28 intervals , and plotted in the same diagram galaxies , as well as groups and clusters of galaxies , and markarian galaxies .
redshift data for clusters , groups and markarian galaxies were almost complete in the northern hemisphere up to a redshift about 15.000 km / s . two wedge diagrams of width @xmath29 , crossing the coma , perseus , hercules and local superclusters , are shown in fig .
[ fig : pers ] . in these diagrams a regularity was clearly seen : _ isolated galaxies and galaxy systems populate identical regions , and the space between these regions is empty_. after this success the whole tartu cosmology team continued the study using wedge diagrams and other methods to analyse the distribution of galaxies and their systems .
a detailed analysis of the perseus supercluster region was made ; here the number of foreground galaxies is very small . already in 1975 , after the tbilisi meeting , we discussed with zeldovich the possibility to organise a real international conference devoted solely to cosmology .
due to soviet bureaucratic system it was extremely difficult for soviet astronomers to attend international conferences in western countries ; thus the only possibility to have a better contact between cosmologists from east and west was to hold the conference within the soviet union .
zeldovich suggested to hold it in tallinn .
after some discussion we decided to devote it to `` large scale structure of the universe '' . when we started preparations we had no idea what this term could mean .
[ fig : pers ] the symposium was held in september 1977 .
our main results were presented in the talk by jeveer & einasto@xcite : ( 1 ) galaxies , groups and clusters of galaxies are not randomly distributed , but form chains , concentrated in superclusters ; ( 2 ) the space between galaxy chains contains almost no galaxies and form holes or voids of diameter up to @xmath30 @xmath23mpc ( see fig .
[ fig : pers ] ) ; ( 3 ) the whole pattern of the distribution of galaxies and clusters resembles cells of a honeycomb , rather close to the picture predicted by zeldovich .
a more detailed analysis was published separately by jeveer , einasto & tago@xcite .
the presence of voids in the distribution of galaxies was reported also by other groups : by tully & fisher@xcite , tifft & gregory@xcite , and tarenghi et al.@xcite in the local , coma and hercules superclusters , respectively .
theoretical interpretation of the observed cellular structure was discussed by zeldovich@xcite .
malcolm longair noted in his concluding remarks : _ the discovery of the filamentary character of the distribution of galaxies , similar to a lace tablecloth , and the overall cellular picture of the large scale distribution was the most exciting result presented at this symposium_. these results demonstrated that the pancake scenario by zeldovich@xcite has many advantages over other rivalling scenarios .
the term `` large scale structure of the universe '' got its present meaning . the first problem to solve was to find some explanation for the absence of galaxies in voids .
this was done by einasto , jeveer & saar@xcite .
saar developed an approximate model of the evolution of density perturbations in under and over dense regions based on zeldovich ideas .
he found that the matter flows out of under dense regions and collects in over dense regions until it collapses ( pancake forming ) and forms here galaxies and clusters .
in under dense regions the density decreases continuously , but never reaches zero : there must be primordial matter in voids . in these under - dense regions
the density is always less than the mean density , thus galaxy formation is not possible .
initially we believed that pancakes are 2dimensional surfaces as predicted by zeldovich@xcite . to our surprise
we did not find evidence for the presence of wall like structures between voids _ the dominating structural element was a chain ( filament ) of galaxies and clusters_. the absence of wall like pancakes and the dominance of filaments was explained theoretically by bond , kofman & pogosyan@xcite , through the effect of tidal forces .
the presence of filaments and voids in the galaxy distribution was met with some scepticism .
one objection against the new concept was raised by peebles : the human brain has a tendency to see regularity ( filaments in galaxy distribution ) even in the case if the actual distribution is almost random .
zeldovich@xcite addressed this problem in his talk during the iau symposium and suggested that objective criteria should be used to check this aspect of the galaxy distribution . together with the zeldovich team we developed several methods to characterise the filamentary distribution
one of these methods was the multiplicity function of galaxy systems , it is sensitive to the richness of galaxy systems .
the other test applied was the connectivity or percolation test , which makes difference between galaxy systems consisting of isolated clusters , clusters connected with filaments , and a random distribution of galaxies . applying these tests to real galaxy samples confirmed the presence of galaxy systems of very different richness , and a high connectivity due to galaxy filaments between superclusters .
the application of tests to simulated structures showed that the original zeldovich scenario ( where small - scale perturbations were absent ) , as well as the peebles scenario , have some problems .
the zeldovich scenario passes the percolation test , but not the multiplicity test there are no fine structures within superclusters .
the peebles scenario fails both tests , results of these tests were described by zeldovich , einasto & shandarin@xcite and einasto et al.@xcite .
difficulties with both scenarios were solved when the cold dark matter model of structure formation was applied , which unites best properties of both the zeldovich and the peebles scenarios ( melott et al.@xcite , blumenthal et al.@xcite ) .
galaxies form by clustering and merging of smaller galaxies as suggested by peebles , but their formation starts in future superclusters , where the density is highest , as it follows from the zeldovich pancake scenario . at the time of the tallinn symposium only relatively small areas of sky were covered by complete magnitude - limited galaxy samples , thus many astronomers were suspicious to the presence of the overall cellular distribution .
it was evident that wide area and deeper flux limited galaxy redshift surveys are needed .
harvard astronomers made for the whole northern hemisphere a survey up to limiting magnitude 14.5 , later the survey was extended to 15.5 magnitude , cfa1 and cfa2 surveys , respectively .
the second cfa surveys shows the filamentary character of the galaxy distribution very clearly , as seen from the first slice of their study by de lapparent , geller & huchra@xcite .
the presence of the cosmic web was confirmed .
a much deeper redshift survey up to the blue magnitude 19.4 was made using the anglo - australian 4-m telescope .
this two degree field galaxy redshift survey ( 2dfgrs ) covers an equatorial strip in the northern galactic hemisphere and a contiguous area in the southern hemisphere@xcite .
over 250 thousand redshifts have been measured .
presently the largest project to map the universe , the sloan digital sky survey ( sdss ) , has been completed by a number of american , japanese and european universities and observatories@xcite .
the goal was to map a quarter of the entire sky : to determine positions and photometric data in 5 spectral bands of galaxies and quasars , and redshifts of all galaxies down to red magnitude r = 17.7 ( about 1 million galaxies ) .
all 7 data releases have been made public .
this has allowed to map the largest volume of the universe so far .
[ fig : sdss ] shows the luminosity density field of a shell of thickness 10 @xmath23mpc at a distance @xmath31 @xmath23mpc from us .
the discovery of the non baryonic nature of the dark matter has resolved the first fundamental problem of the dark matter , discussed by tammann in tbilisi in 1975 .
the second problem , the smoothness of the hubble flow , was explained only recently with the discovery of dark energy , previously called also cosmological lambda term .
two teams , led by riess@xcite ( high - z supernova search team ) and perlmutter @xcite ( supernova cosmology project ) , initiated programs to detect distant type ia supernovae in the early stage of their evolution , and to investigate with large telescopes their properties .
these supernovae have an almost constant intrinsic brightness . by comparing the luminosities and redshifts of nearby and distant
supernovae it is possible to calculate how fast the universe was expanding at different times .
the supernova observations give strong support to the cosmological model with the @xmath32 term .
studies of the hubble flow in the nearby space , using observations of type ia supernovae with the hubble space telescope ( hst ) , were carried out by several groups .
the major goal of the study was to determine the value of the hubble constant .
as a by - product also the smoothness of the hubble flow was investigated .
one of these projects was led by allan sandage@xcite .
the analysis confirmed earlier results that the hubble flow is very quiet over a range of scales from our local supercluster to the most distant objects observed .
this smoothness in spite of the inhomogeneous local mass distribution requires a special agent .
sandage emphasises that no viable alternative to dark energy is known at present , thus the quietness of the hubble flow gives strong support for the existence of dark energy .
this effect has been investigated in detail by arthur chernin@xcite .
* the presence of dark matter in the universe and the cosmic web were established gradually by astronomers from many centers , in several cases tartu astronomers have pioneered in these studies . *
initially the dark matter concept had many problems , until its non baryonic nature and the presence of dark energy were found ( @xmath32cdm model ) .
also the concept of the web - like distribution of galaxies was initially met with scepticism , which disappeared when wide - area and complete redshift surveys were completed . * the discoveries of dark matter and cosmic web are connected : dark matter as the dominant population in the universe determines properties of the web , and the structure of the web gives information on properties of dm particles . * in the study of dark matter and the structure of the universe our tartu team benefited from the earlier experience in tartu by pik , rootsme and kuzmin , and especially from a close collaboration with the moscow cosmology team led by yakov zeldovich .
i thank maret einasto , gert htsi , juhan lauri liivamgi , enn saar , erik tago , elmo tempel , and our colleague volker mller in potsdam for collaboration and for the permission to use our common results in this review talk .
my special thank is to late mihkel jeveer for pioneering contributions both in the search for dark matter and of the structure of the universe , as well as to late yakov borissovich zeldovich and his team , who inspired us to study the large scale distribution of galaxies , and participated in the development of the ideas following the discovery of the cosmic web .
the present study was supported by the estonian science foundation grant etf 4695 , and by grant to 0060058s98 .
i thank icranet , and the astrophysikalisches institut potsdam ( dfg - grant mu 1020/15 - 1 ) where part of this study was performed , for hospitality .
w. g. tifft and s. a. gregory , observations of the large scale distribution of galaxies , in _ large scale structures in the universe _ , eds . m. s. longair and j. einasto , iau symposium , vol .
79 , 267 ( 1978 ) .
m. tarenghi , w. g. tifft , g. chincarini , h. j. rood and l. a. thompson , the structure of the hercules supercluster , in _ large scale structures in the universe _ , eds . m. s. longair and j. einasto , iau symposium , vol .
79 , 263 ( 1978 ) . | an overview is provided for 200 years of galactic studies at the tartu observatory .
galactic studies have been one of the main topics of studies in tartu over the whole period of the history of the observatory , starting from f.g.w .
struve and j.h .
mdler , followed by ernst pik and grigori kuzmin , and continuing with the present generation of astronomers .
our goal was to understand better the structure , origin and evolution of stars , galaxies and the universe . |
the science of nanostructures is one of the fields of growing interest in materials science .
nanotubes , in particular , are of both fundamental and technological importance ; being quasi - one - dimensional ( 1d ) structures , they possess a number of exceptional properties .
while the peculiar electronic structure metallic versus semiconducting behavior of carbon nanotubes depends sensitively on the diameter and the chirality , i.e. , on the way the graphene sheet is wrapped up into a tube @xcite , boron nitride ( bn ) tubes display a more uniform behavior with a wide band - gap ( larger than 4 ev ) , almost independent of diameter and chirality @xcite .
the potential applications of nanotubes in nanodevices are numerous @xcite : super - tough nanotube fibres @xcite , gas sensors @xcite , field effect displays @xcite , and electromechanical devices @xcite , to name only a few .
one of the most spectacular examples is the realization of field effect transistors both with carbon nanotubes @xcite and with bn - nanotubes @xcite .
since carbon and bn nanotubes are routinely produced in gram quantities , the challenge now consists in having fast experimental tools for characterization of nanotube samples and , if possible , single isolated nanotubes .
optical spectroscopic techniques provide us with this tool @xcite .
the final goal is to find a unique mapping of the measured electronic and vibrational properties onto the tube indices @xmath0 .
to that end , the electronic structure and dielectric properties of the tubes are two key areas to study .
one possible spectroscopic method is optical absorption spectroscopy with direct excitations from occupied to unoccupied states . for carbon tubes ,
the energy difference @xmath1 between corresponding occupied and unoccupied van hove singularities ( vhss ) in the 1-dimensional electronic density of states ( dos ) is approximately inversely proportional to the tube diameter @xmath2 . in resonant raman spectroscopy
@xcite and scanning tunneling spectroscopy @xcite , this scaling is employed for the determination of tube diameters ( and @xmath0 indexes in resonant raman @xcite ) .
a recent example is the fluorescence spectroscopy of single carbon tubes in aqueous solution , where @xmath3 is probed through the frequency of the excitation laser and @xmath4 is probed simultaneously through the frequency of the emitted fluorescent light @xcite .
also optical absorption spectroscopy of nanotubes in aqueous solution @xcite allows for the spectral resolution of peaks that can be associated with distinct types of nanotubes . for the distance between the first vhss in semiconducting carbon tubes , a simple @xmath5-electron tight - binding fit yields the relation @xmath6 , where @xmath7 is the distance between nearest - neighbor carbon atoms .
the value for the hopping matrix element @xmath8 varies between 2.4 ev and 2.9 ev , depending on the experimental context in which it is used .
this fact is a clear indication that the above relation gives only qualitative and not quantitative information on the tube diameter and/or chirality .
furthermore , for small - diameter tubes the band structure completely changes with respect to the band structure of large diameter tubes , including a reordering of the vhss in the density of states and displaying fine structure beyond the first and second vhss @xcite .
this structure sensitively depends on the tube indices and may be probed by optical absorption spectroscopy over a wider energy range , possibly extending into the uv regime .
the spectroscopical characterization of macroscopic tube samples is made difficult by the fact that in the bulk solid and even in bundles the tubes are not perfectly aligned and do not have a well defined diameter and helicity , rendering , in the case of carbon tubes , a random mixture of semiconducting and metallic tubes @xcite .
additionally , the role of intertube interaction in the spectra needs to be addressed since the tubes are close packed and could interact with each other via long - range forces induced by the excitations . only very recently optical absorption spectra were reported for aligned single - wall carbon tubes of a very narrow diameter distribution ( [email protected] ) grown inside the channels of a zeolite matrix @xcite .
geometric arguments predict three possible tubular helicities for the range of diameters around 4 : the armchair ( 3,3 ) , the zig - zag ( 5,0 ) and the chiral ( 4,2 ) configuration .
therefore , this particular case serves as an important case study where a direct comparison between experimental data and theoretical calculations becomes possible . indeed in ref .
@xcite we have shown the relevance of a first - principles calculation for these small - diameter carbon nanotubes by reproducing the polarization dependence of the measured optical spectra . at present , the dielectric response of tubes in the frequency range of the electronic interband transitions and the collective excitations ( plasmons ) of the valence electrons ( up to 50 ev ) is not well understood since previous studies focused on the low - energy regime and excitations @xcite . for higher frequencies , up to now , the majority of the theoretical studies of the response , besides model calculations @xcite , are mostly limited to summing over independent band - to - band transitions obtained within the semi - empirical tight - binding method @xcite or the density - functional theory ( dft ) framework @xcite , or to calculations of the joint density of states ( jdos ) @xcite which use the bandstructure with no explicit evaluation of the transition matrix elements . as we will show below , these approximations are not sufficient for a full interpretation of the experiments .
this shortcoming is not due to the quality of the bandstructure calculation itself but instead due to the neglect of the induced microscopic components in the response to the external field , the local field effects ( lfe ) @xcite .
these effects strongly modify the total response for certain polarizations of the external perturbation .
also , induced exchange and correlation ( xc ) components obtained beyond the random - phase approximation ( rpa ) might contribute @xcite .
therefore , important questions concerning the electron interaction , excitations and screening still remain unanswered .
our approach here is to determine the spectra by _ ab - initio _ calculations incorporating important ingredients of the electron response as in previous works @xcite .
the present paper is organized as follows : after an exposition of the theoretical framework in section [ theory ] , we present in section [ tubeabs ] _ ab initio _ calculations of optical spectra of small single - wall carbon and boron nitride nanotubes .
we identify the influence of crystal lfe , xc effects and intertube interaction in the spectra .
we also point out the similarities and differences between carbon and bn structures which are related to similarities and differences in the respective electronic band structures .
a certain number of comparisons with experiment allows us to verify that the chosen _ ab initio _ approach is an important improvement and well suited for the description of the spectra of these systems . in section [ sheetabs ] , we analyse the spectra of the building blocks of the tubes , i.e. the graphene and hexagonal bn sheets ( layers ) including the optical spectrum of graphite .
this gives information about the the interaction between objects ( sheet - sheet ) in the excited state , even when these objects can be considered to be isolated in the ground state .
we also discuss to which extent the dielectric response of the tubes can be understood in terms of the response of the sheets .
this will help to understand whether some aspects of the response of the tubes are _ inherent _ to the sheets , and if so , they could also be observed in other systems of more practical interest , e.g. samples comprising a mixture of tubes with different diameters and orientation or multiwall tubes . in section [ eelsec ]
, we present results for electron energy loss in order to have some additional validation from existing experimental data , and because a comparison between optical and loss spectra allows a deeper discussion and understanding of interaction effects .
finally , we conclude with section [ finsec ] , with an overall discussion of the results .
_ ab initio _ calculations in the framework of density functional theory ( dft ) have yielded high - quality results for a large variety of systems : from molecules to periodic solids and structural defects @xcite .
these results are however mostly limited to quantities related to the electronic _ ground state _ , whereas additional phenomena that occur in the excited state are not correctly described @xcite .
in particular , the self - consistency between the total perturbing potential and the charge response induces hartree and exchange - correlation potentials that have to be dealt with .
the former give rise to the so - called local field effects , whereas the latter can lead , for example , to excitonic effects .
today , in the solid state _
ab initio _ framework two main approaches can include both lfe and xc effects @xcite and can be therefore suitable for the description of electronic excitations in nanostructures .
first , green s functions approaches within many - body perturbation theory : here one adds self - energy corrections to the dft kohn - sham bandstructure and the electron - hole interaction is included via the solution of the bethe - salpeter equation @xcite .
this approach has given excellent results for various bulk and finite systems , but it is computationally very cumbersome and not ready yet to be applied systematically to more complex systems .
the time - dependent hartree - fock ( tdhf ) method represents a certain level of approximation within this approach , where correlation ( i.e. screening ) in the self energy is neglected .
lfe , on the other hand , are still retained .
tdhf has been employed , up to now in a semi - empirical way , to calculate optical spectra , including excitons , of carbon nanotubes @xcite and also for the 4 - diameter ones @xcite .
however , the obtained absorption peak assignments in the latter case were in disagreement with predictions based on the operative dipole selection rules for the specific tubular space groups @xcite .
this suggests that in such systems with important metallic character a neglect of screening may lead to problems .
the second way to calculate spectra is given by the time - dependent dft ( tddft ) @xcite , where all manybody effects are embodied in the exchange - correlation potential and kernel .
it is in practice an approximate way but always treats the variations of the hartree potential exactly ( right as in both the green s functions and tdhf approaches ) .
this approach , using the ( static ) adiabatic local density approximation ( tdlda ) for the xc effects , or even completely neglecting the density variations of the xc potential ( rpa , on the basis of an lda bandstructure ) has been applied successfully to many finite and infinite systems @xcite . in particular
, excellent results for the loss spectra of graphite have been obtained using this approximation @xcite .
this is the approach we follow in the present work . as a first step in our tddft approach
, we determined the electronic ground state of the systems .
the kohn - sham single - particle equations were solved self - consistently in the lda for exchange and correlation @xcite .
for the description of the valence - core interaction we have used norm - conserving pseudopotentials which were generated from free - atom all - electron calculations @xcite . for the `` isolated '' geometries of the tubes and of the sheets we calculated periodic arrays with a large distance between the building blocks in a supercell geometry , in order to minimize the mutual interaction .
the crystalline valence - electron wavefunctions were expanded using a plane - wave basis set ( with an energy cutoff of 62 rydberg for carbon systems , and 50 rydberg for bn systems ) .
a part of the calculations was carried out using the abinit code @xcite .
the next step is the linear response calculation of the independent particle polarizability @xmath10 @xcite .
it involves a sum over excitations from occupied bands to unoccupied bands : @xmath11 , \label{chi0}\end{aligned}\ ] ] where @xmath12 means that the indices @xmath13 and @xmath14 of the first term are exchanged .
the result is checked for convergence with respect to the number of bands @xcite and the discrete sampling of * k*-points within the first brillouin zone . within tdlda , the full polarizability @xmath15 is connected to @xmath16
via@xcite @xmath17 where v@xmath18 is the bare coulomb interaction and @xmath19 , the so - called exchange - correlation kernel , is the functional derivative of the lda exchange - correlation potential with respect to the electron density . by setting @xmath19 to zero ,
exchange and correlation effects in the electron response are neglected and one obtains the random - phase approximation ( rpa ) .
we have carried out our calculations in the rpa and also in the tdlda for certain cases .
the inverse dielectric function for a periodic system and momentum transfer @xmath20 is obtained from : @xmath21 with @xmath20 in the first brillouin zone and @xmath22 , @xmath23 are reciprocal lattice vectors .
the absorption spectrum is obtained as the imaginary part of the macroscopic dielectric function @xmath24 whereas the loss function for a transferred momentum @xmath25 is given by @xmath26 $ ] . in inhomogeneous systems , e.g. periodic solids , clusters and structural imperfections
the inhomogeneity in the electron response gives rise to local field effects ( lfe ) and the @xmath27 can not be considered as being purely a diagonal matrix .
therefore , its off - diagonal elements have to be included in the matrix inversion . making the approximation @xmath28 corresponds to neglecting the inhomogeneity of the response , i.e. to neglecting the lfe .
when these effects are neglected and , moreover , all transition matrix elements in @xmath16 are supposed to be constant , one arrives at the widely used approximation that the absorption spectrum is proportional to the jdos , i.e. proportional to the sum over interband transitions from occupied ( @xmath29 ) to empty ( @xmath30 ) states over the brillouin zone points @xmath31 , @xmath32 .
let us first look at the optical absorption spectra of small - diameter carbon nanotubes ( all three possible helicities ) , for which experimental results have recently become available .
the calculations of the spectra @xcite were done for the tubes arranged in a hexagonal lattice with an intertube distance ( distance between tube walls ) , equal to @xmath33 5.5 which leads to nearly isolated tubes . additionally , for the ( 3,3 )
armchair ones we repeated the calculations for a solid with a smaller intertube distance , @xmath33 3.2 , which is close to the interlayer distance in graphite ( @xmath34 3.4 ) . the optical absorption spectra for the small diameter tubes are diplayed in fig .
[ fig1 ] ( as well as jdos curves for two cases ) . in the upper panel of the figure ,
the jdos divided by the square of the excitation energy , for the ( 3,3 ) nearly isolated ( thin line ) and interacting ( in the solid ) ( thick line ) tubes are shown .
it can be seen that after an initial very steep decrease ( @xmath35 1 ev ) , in both jdos curves there is a gradual increase starting from 2 ev . in the case of isolated tubes a sequence of pronounced peak structures up to 5 ev
is observed : this can be explained from the occurrence of direct interband transitions between the van hove singularities of the density of states ( dos ) .
these peaks are smeared out in the jdos of the solid where the tubes are strongly interacting @xcite . in the lower panels of fig .
[ fig1 ] the calculated absorption spectra for light polarizations perpendicular and parallel to the tube axis are displayed ( as well as the experimental data plotted in the inset ) .
the dashed lines denote results in the rpa neglecting lfe , continuous lines including lfe . for parallel polarization
both lfe and lda - xc effects were found to be negligible and , therefore , in this case only the rpa results without lfe are presented .
the fact that both these effects turned out to be negligible can explain why the peak positions ( a , b and c ) predicted here for the parallel polarization match very well the ones found in two recent dft - rpa studies @xcite , which completely neglected lfe and xc effects in the response .
our calculated peak positions ( a , b and c ) in fig .
[ fig1 ] are in good agreement ( to within 0.2 ev ) with the experimentally observed peak structures for the parallel polarization ( inset ) .
it is also important to note that the present results support the same peak - to - helicity assignment as in previous works @xcite .
namely , the three observed peaks a , b and c are due to the ( 5,0 ) , ( 4,2 ) and ( 3,3 ) tubes , respectively .
such an assignment is also in accordance with the dipole selection rules for these tubular space groups @xcite . concerning the effect of intertube interaction in the absorption we found that it is rather small for this polarization : tube - tube interaction leads only to a broadening of the main absorption peak ( thick solid curve for the ( 3,3 ) tubes ) .
this can be explained from an increase of the energy range of the possible interband transitions brought about by the interaction .
all the discussion of the results up to now seems to indicate that lfe may not be needed . nonetheless , for light polarizations perpendicular to the tube axis
due to the presence of depolarization lfe play an important role which can not be ignored .
more specifically , for this polarization the experimental spectrum displays vanishing intensity for frequencies up to 4 ev @xcite ( see inset in fig .
[ fig1 ] , dotted curve ) .
clearly , neither the jdos calculation nor the calculated spectrum without lfe can capture this effect .
instead , they both predict a series of peaks from 2 to 5 ev .
the reason that rpa - without - lfe fails here is due to the depolarization effect @xcite which is created by the induced polarization charges .
the depolarization is only accounted for if lfe are included : as it can be seen in fig .
[ fig1 ] ( second panel ; continuous curve ) _ lfe suppress the low - energy absorption peaks _ and render the tubes almost transparent below 5 ev in agreement with the experiment .
it should not come , therefore , as a surprise that the recent dft - rpa studies @xcite ( which did not consider lfe ) did not reproduce this transparency for the perpendicular polarization . the tdlda result , displayed as the dotted curve in the second panel , turned out to be qualitatively similar to the rpa - with - lfe result : again the low - energy absorption peaks are suppressed .
this shows that the main effect comes from fluctuations of the hartree- , and not from those of the xc - potential ( as also obtained for the case of graphite @xcite ) .
lda - xc effects only cause a small ( 0.3 ev ) redshift of the remaining absorption peak to 5.5 ev , with respect to rpa - with - lfe .
this is a characteristic behavior of finite systems @xcite .
still there is an open question about the role of electron - hole interaction in the case of the small band - gap ( 4,2 ) tube @xcite , however the good agreement with experiment indicates that its contribution should be small for the parallel polarization .
lfe also have a similar drastic impact for the system of isolated zig - zag ( 5,0 ) tubes ( not shown ) .
the manner according to which lfe operate may be understood as follows : for perpendicular polarization the tubes form an assembly of almost isolated , but highly polarizable , objects .
an applied external field induces hence a local , i.e. microscopic , response - the lfe - which strongly weakens the total perturbation ( i.e. it is a `` depolarization '' ) . the macroscopic response to this weak perturbation is only very moderate , because the electrons are localized on the tubes .
this is totally different from the screening in a bulk metal or small - gap semiconductor , where even a very weak total perturbation still leads to a strong response at low frequencies .
for polarization parallel to the tube axis , the situation resembles rather this latter case , something which explains the absence of lfe for this polarization . on the other hand for the perpendicular polarization
when the tubes are interacting in the solid ( third panel ) the depolarization is much weaker : the tubes start to absorb ( they are no longer transparent ) because the electronic states start to delocalize and the system is now more similar to a bulk metal .
this different behavior of the response depending on the intertube distance leads to a very important consequence : it suggests that _ the inter - tube interaction can be detected experimentally _ in a qualitative study of the absorption spectra for perpendicular light polarizations . in fig .
[ bntubesfig ] we present optical absorption spectra for three different bn nanotubes .
first , we compare for the case of the bn(3,3 ) tube the spatially averaged spectrum with the joint density of states ( jdos ) , divided by the square of the transition energy
. if local field efffects are neglected , most peaks of the jdos are visible in the averaged absorption spectra while some peaks are suppressed due to small or vanishing oscillator strength in eq .
( [ chi0 ] ) .
this demonstrates that the fine structure in the spectra is not an artifact of low k - point sampling but is due to the presence of van - hove singularities in the 1-dimensional density of states of the tubes .
proper inclusion of lfe leads to a smoothing of the spectrum .
however , some fine structure survives and may be discernible in high - resolution optical absorption experiments .
the spectrum with perpendicular polarization demonstrates that for bn nanotubes , depolarization effects play a similarly important role as for carbon tubes : neglecting lfe , the onset of absorption would be around 4.5 ev .
lfe , however , lead to a redistribution of the oscillator strength to higher energies and render the tubes almost transparent up to 8 ev .
the band gap of bn nanotubes in dft - lda is 4 ev and is only weakly dependent on radius and chirality @xcite .
accordingly , the absorption spectra for polarization parallel to the tube axis are very similar for the ( 3,3 ) , ( 5,5 ) , and ( 6,0 ) tubes which all display a strong absorption peak around 5 ev and a second high peak around 14 ev .
we expect these structures to be stable also for tubes with larger diameters .
only the fine structure of the spectra ( e.g. , the absorption peak at 9 ev for the ( 3,3 ) tube ) depends on the details of the band - structure and varies for the different diameters and chiralities @xcite . for the ( 6,0 ) tube , the first high absorption peak is split and the dominant peak shifted towards lower energy .
this is a curvature effect which leads to a reduction of the band - gap for small zigzag bn tubes . in the next section
we will see to which extent the above findings for carbon and bn nanotubes can be understood by an analysis based on results for graphene and bn sheets and graphite .
hexagonal graphite has the aba bernal stacking sequence of the graphene sheets . in the present work we assumed the experimental lattice parameter a@xmath36 and ( c / a)@xmath36 ratio ( 2.46 and 2.73 , respectively @xcite ) .
the calculated rpa optical absorption spectra @xcite , with and without lfe , are shown in fig .
[ graphabs ] for in - plane light polarization ( * e * @xmath37 * c * ) and in fig .
[ fig4 ] ( a ) for polarization parallel to the * c * axis ( * e * @xmath38 * c * ) .
the in - plane spectrum is dominated by a very intense peak structure at low frequencies ( up to 5 ev ) and also another peak structure of broader frequency range which sets in beyond 10 ev and has a pronounced peak at 14 ev .
the origin of these peak structures is due to @xmath39 and @xmath40 interband transitions , respectively , according to the earlier interpretations by bassani and paravicini @xcite who assumed a two - dimensional approximation no interaction between the graphene sheets and the operative dipole selection rules for this polarization .
our calculations of the oscillator strength for specific transitions between bands in the brillouin zone ( bz ) are in agreement with their interpretation .
lfe are found to be nearly negligible for this polarization .
this is not surprising since for in - plane polarizations graphite is homogeneous in the long - wavelength limit ( @xmath41 ) .
the general aspects of the spectrum peak positions , their intensity and lineshape are in close agreement with the existing experimental results @xcite and the previous all - electron calculation of ahuja et al .
@xcite who neglected lfe .
* c*. the star symbol denotes the unoccupied electron states.,width=302 ] the absorption spectrum of graphite for the light polarization parallel to the * c * axis is shown in fig .
[ fig4 ] ( a ) .
it is characterized by a weak intensity in the low frequency range ( 0 - 5 ev ) and important peaks in the frequency range beyond 10 ev . for
this polarization the bandstructure does not play the exclusive direct role in defining the absorption spectrum .
now lfe are very important . when lfe are not considered the peak positions for this polarization are at 11 and 14 ev as in the earlier dft - rpa calculation @xcite .
however , when lfe are included transitions are mixed and the absorption spectrum is appreciably modified . the main effect of local fields is to shift the oscillator strength at 10 - 15 ev to higher frequencies .
they decrease the intensity of the 11 ev peak and are responsible for the appearance of the 16 ev peak in the spectrum .
the latter peak is seen in experiments as a shoulder suggesting that the inclusion of lfe is necessary .
* c * , for graphite and the graphene - sheet geometries with 2 ( c / a)@xmath36 and 3 ( c / a)@xmath36.,width=302 ] the non - zero oscillator strength found below 5 ev is attributed to the inter - layer interaction which is present in the solid .
it is also observed experimentally .
the dipole selection rules @xcite for an isolated graphene sheet ( i.e. complete two - dimensionality ) lead to vanishing matrix elements and oscillator strength at this frequency range .
lfe do not have any influence on the lower part of the spectrum ( less than 10 ev ) .
the existing experimental evidence is not conclusive for the dielectric function in the 11 ev frequency region .
the frequency - dependent im[@xmath42 obtained from electron energy loss data @xcite displays a very sharp and intense peak ( im[@xmath42=10 ) at 11 ev . on the other hand , on the basis of optical measurements @xcite the observed maximum at this frequency is of considerably smaller intensity .
the earlier interpretations @xcite were based on semi - empirical tight - binding bandstructure calculations in the two - dimensional approximation ( i.e. isolated graphene layers ) and they predicted peak structure between 13.5 and 16.5 ev .
it is therefore also important to understand the effect of the inter - layer interaction in the optical response and if this interaction is primarily responsible for the occurrence of the intense peak at 11 ev . for this purpose
, we progressively increased the inter - layer spacing ( doublying and tripling the ( c / a)@xmath36 ratio ) .
this yields stackings of graphene layers in the unit cell with much weaker mutual interaction .
the absorption spectra for these graphene - sheet geometries are displayed in fig .
[ fig4 ] ( b , c ) .
the first observation is that the oscillator strength vanishes completely in the frequency region below 10 ev in complete accordance with the predictions based on the dipole - selection rules for an isolated graphene layer .
therefore , the double inter - layer spacing leads to non - interacting graphene layers as far as the rpa absorption spectrum ( at this frequency range ) is concerned .
when lfe are neglected , the peak structure in the 10 - 15 ev range stays intact , now with a smaller intensity due to the larger volume . with increasing interlayer separation ,
lfe become progressively more important .
the shift of oscillator strength induced by lfe is so big that the absorption peaks at 10 - 15 ev are almost completely suppressed .
these findings demonstrate that both inter - layer interaction and lfe influence considerably the intensity of the absorption peak at 11 ev .
qualitatively , the depolarization effects found in the tubes can hence be explained by the local field effects observed in the graphitic response perpendicular to the graphene sheets .
likewise , the effect of intertube interaction in the spectra for polarizations perpendicular to the tube axis is also consistent with the increased propensity of lfe to shift oscillator strength at higher frequencies when the intersheet distance progressively increases ( see fig .
[ fig4 ] ) .
we conclude this section with a comparative presentation of the absorption spectra of the graphene and bn sheets using the same inter - sheet distance for both cases , shown in fig .
[ sheetsfig ] .
also we make comparisons with the spectra of the corresponding tubes .
the spectrum of the graphene sheet for the in - plane polarization resembles closely the in - plane polarization spectrum of graphite ( fig .
[ graphabs ] ) ( except for a scaling due to the change in the volume ) , confirming once more , as in the case of the tubes for the polarization parallel to the tube axis , that the position of the absorption peaks and their lineshape is only weakly influenced by the distance and inter - sheet interaction in this case .
the main difference between the graphene and bn - sheet spectra for the in - plane polarization is the complete absence of any feature below 4 ev in the bn spectrum .
this is clearly related to the 4 ev lda - band gap of bn .
the out - of - plane polarization spectra of the graphene and bn sheets are remarkably similar .
most importantly , both display a transparency up to 10 ev and both demonstrate a strong depolarization effect with shift of oscillator strength to higher energies @xcite .
we compare now the calculated sheet - absorption spectra for in - plane polarization with the corresponding tube - absorption spectra for polarization parallel to the tube axis ( fig .
[ fig1 ] for carbon and fig .
[ bntubesfig ] for bn ) . in the carbon case ,
the strong absorption feature of the sheet between 0 and 4 ev maps onto one or several absorption peaks of the tubes in this energy range .
the details of this mapping are , however , sensitively dependent on the diameter and the chirality of the tubes , since dipole selection rules play a strong role in these highly symmetric systems .
in particular , for tubes with larger diameter than the ones calculated in this article , the distance @xmath4 of the first van hove singularities strongly depends on whether the tube is metallic or semiconducting .
therefore , calculations on graphite or graphene alone will not be sufficient to predict absorption spectra of small - diameter carbon nanotubes .
for bn , in contrast , the first high absorption peak at 5.5 ev maps directly onto a corresponding peak in the tubes .
also the high energy absorption feature around 14 ev is very similar for the bn sheet and the bn tubes . due to the large band gap of bn ,
the comparison between sheet and tube spectra is much more favourable for bn than for carbon tubes . on the other hand , in the case of carbon , the extrapolation of results from graphite and graphene to the tubes
is much more straightforward for the electron energy loss spectra as discussed in the following section .
having seen the strong interaction effects that occur in optical spectra for a polarization perpendicular to the planes or tube axis , it is interesting to make an excursion to a different type of spectroscopy , namely electron energy loss ( eels ) @xcite .
although also in this case one can measure , like in the absorption experiment , a momentum transfer close to zero , there is a substantial difference in the definition of the response function : in the absorption measurement , one detects the response to the total macroscopic field , whereas in eels the response to the external field is reported .
therefore , as pointed out above , absorption is linked to the imaginary part of the macroscopic dielectric function , but eels is linked to the imaginary part of the inverse of the latter .
mathematically this translates into the fact that the crucial response function for eels is governed by equation ( [ eq : chi ] ) , whereas in the case of absorption one can use slightly different quantity : the macroscopic dielectric function can in fact be rewritten as @xmath43 where v@xmath44 is the long - range component of the coulomb potential , and it has a @xmath45 divergence for vanishing @xmath46 and the quantity @xmath47 obeys an equation similar to ( [ eq : chi ] ) : @xmath48 , but setting the @xmath49 component of @xmath50 to zero ( @xmath51 ) @xcite .
therefore , this seemingly tiny difference is responsible for the difference between , e.g. , the position of the main absorption peaks and that of the valence plasmons in solids , and one can also expect that it will be crucial when interaction effects on the spectra are discussed . in particular , eels spectra should show , due to the presence of this long - range term , stronger interaction effects than absorption spectra .
this is in fact the case , as we will illustrate in the following for the total @xmath52 plasmon in graphite .
this plasmon represents the collective excitation mode of all the valence electrons in graphite . before discussing the results we stress here that the tubes can not be considered as completely isolated objects in this calculation of the loss function with the 5.5 intertube distance
we refer to the tubes in the latter geometry as _ distant_. fig .
[ ctubeels ] shows the rpa loss function , @xmath53 , for the ( 3,3 ) tube , in the range 15 - 35 ev and for a vanishing momentum transfer * q * parallel to the tube axis .
for this orientation lfe are negligible .
a strong shift of the @xmath52 plasmon from 22 to 28 ev due to intertube interactions in the solid can be seen @xcite .
the magnitude of this shift reveals a strong dependence of the plasmon position upon the intertube distance ( hence the average valence electron density ) essentially following the plasmon - frequency dependence in the case of the homogeneous electron gas @xcite .
this shows that the tubes respond as homogeneous and highly polarizable objects for parallel * q * in the long - wavelength limit ( @xmath54 ) . a direct consequence
would then be that the atomic arrangement , orientation of bonds and helicity may play a secondary role in the response in this frequency and * q * range .
therefore , the result for the @xmath52 plasmon ( shown in fig .
[ ctubeels ] ) would be representative of either of the three tubes since all of them being of nearly the same diameter
possess the same average electron density .
this is indeed the case as it can be seen in the inset of fig .
[ ctubeels ] where an almost indistinguishable @xmath52 plasmon was also obtained for the ( 5,0 ) tube .
hence , in contrast to an optical absorption experiment , small - q loss measurements of _ the @xmath52 plasmon can not determine tubular helicities _ for a given tube diameter .
the governing factor for the @xmath52 plasmon must be traced to the in - layer graphitic response as it can be seen in fig .
[ ctubeels ] where the loss function for graphite and graphene is also shown ( for in - layer * q * orientation ) at comparable ( to the tubes ) average electron densities ( to within 10 % ) @xcite ( dashed curves in fig .
[ ctubeels ] ) .
this shows that the loss function of the tubes for parallel * q * in this frequency range is governed by the average - density - dependent part of the in - layer graphitic response .
similar plasmon shifts , therefore , can also be expected in other carbon systems with graphene - based structural blocks e.g. multiwalled tubes .
the present results outline the significance of the _ @xmath52 plasmon as a key measurable spectroscopic quantity _ which could gauge the _ intertube distances and interactions _ in real samples of carbon nanotubes . in order to gain a more complete understanding of the in - layer graphitic response at small q s and how the latter is influenced by the inter - layer interaction we determined both the loss and dielectric function for various graphene - like geometries i.e. varying the interlayer spacing or , equivalently , the ( c / a)@xmath36 ratio . in these calculations
we also looked at the lower - frequency @xmath5 plasmon .
bearing in mind the discussion in the previous paragraph , these calculations could then serve as benchmarks for predicting the position of the @xmath52 plasmon in nanotubes as a function of the intertube distances for low q s . ) for graphite and the graphene - like geometries with multiple ( c / a)@xmath36 ratios .
only the results without lfe are presented since lfe are negligible in this q range.,width=302 ] the loss and dielectric function for small in - plane * q * ( 0.22 @xmath55 ) is shown in fig .
[ grapheels ] for graphite and the graphene geometries with multiple ( c / a)@xmath36 ratios .
it can be seen that the peak positions of both plasmons have shifted to lower frequencies when the inter - layer separation is increased .
this effect is very pronounced for the @xmath52 plasmon position .
these results reaffirm that the total ( @xmath52 ) plasmon is extremely sensitive to the inter - layer interaction for small in - plane q s . this can be explained as follows : with increasing ( c / a)@xmath36 , i.e. interlayer spacing , the system becomes an assembly of nearly isolated graphene sheets .
this leads to a decrease of screening ( re[@xmath56 \to 1 $ ] ; see fig .
[ grapheels ] ) with the effect that @xmath57 \to im[\varepsilon_{m}]$ ] , namely a coincidence of the loss and absorption functions .
since for the in - plane polarization the positions of the absorption peaks do not change with increasing ( c / a)@xmath36 ( see im[@xmath42 in fig . [ grapheels ] ) , then the loss function undergoes important changes , in particular the @xmath52 plasmon .
the latter is displaced at a much faster rate to lower frequencies towards the 14 ev peak of the absorption spectrum whereas the @xmath5 plasmon peak is rather insensitive to ( c / a)@xmath36 since it is already located very close to the 0 - 5 ev peak structure of the @xmath39 transitions in @xmath58 $ ] . at present , existing measurements @xcite of the loss spectra of samples of single - wall carbon nanotubes have given a @xmath52 plasmon in the frequency range 2124 ev for small momentum transfer q. our predicted frequency of the @xmath52 plasmon for the case of the distant tubes is within this frequency range ( fig.[ctubeels ] ) .
however , a direct comparison of the present results for the loss function of ( 3,3 ) and ( 5,0 ) tubes with the measured loss data is not straightforward .
the difficulty stems from two factors , tube diameter and alignment , which have a competing effect on the @xmath52 plasmon position : a ) bulk samples of single - wall tube material possess a mean diameter of 14 @xcite , which is considerably larger to the range of 4 studied in the present work . assuming a common intertube spacing ,
then larger diameters would give rise to a displacement of the @xmath52 plasmon towards lower frequencies since the diametrically - opposed wall parts of the same tube are facing each other at larger distances , i.e. a situation resembling a stacking of graphene sheets with larger ( c / a)@xmath36 ratios ( see fig .
[ grapheels ] ) .
for instance , the 5 ( c / a)@xmath36 ratio corresponds to an intersheet separation of 16.8 and the corresponding @xmath52 plasmon peak is at 16 ev .
b ) the alignment of the tubes in the samples is not perfect ; therefore it is to be expected that also out - of - plane excitations of graphitic origin will contribute to the response .
these excitations @xcite should tend to produce a more diffuse shape for the loss spectrum , heavily dampening the @xmath52 plasmon and shifting the observed peak towards higher frequencies .
the type of response described just above can be clearly seen in fig .
[ ctubeels2 ] which shows the loss function of the distant ( 3,3 ) tubes for * q * orientation perpendicular to the tube axis . for this orientation in - plane as well as
out - of - plane graphene excitations contribute to the tube response .
the latter cause the diffuse shape of the loss function ( see inset of fig .
[ ctubeels2 ] ) .
lfe are now very important and the peak position of the @xmath52 plasmon is at 28 ev . ) of orientation perpendicular to the tube axis . in the inset the loss function of graphene @xcite for 2 ( c / a)@xmath36 and for a similar * q * of orientation perpendicular to the sheet is displayed .
continuous and dashed curves denote results obtained with and without lfe , respectively.,width=302 ] for completeness , we show in fig .
[ fig9 ] also the calculated eels spectra for small bn nanotubes with momentum transfer @xmath59 along the tube axis . as in the case of carbon tubes ,
two main features are clearly pronounced : the @xmath5-plasmon at 6 - 7 ev and the high energy collective oscillation ( @xmath60 plasmon ) around 20 ev .
the exact position of the peaks depends on the radius and chirality of the tubes and on the intertube distance . in order to compare with experimental eels - spectra on multi - wall bn tubes @xcite , an extrapolation to larger - diameter tubes
is needed and the inter - wall interactions have to be taken into account . along the direction of the tube axis .
the spectra are calculated with an inter - tube distance of 7.4 . since the spectra with and without lfe are almost indistinguishable , only the calculation without lfe is shown.,width=302 ]
in this closing section we should like first to comment on the validity of rpa and tdlda for the description of the optical spectra of the tubes .
it is well known that rpa and tdlda often give excellent results for _ loss _ spectra @xcite . however , in this work we have seen that also very good _ absorption _ spectra were obtained for the carbon tubes , even at the rpa level , despite the fact that both rpa and tdlda are known to fail badly in the description of absorption spectra of many bulk materials ( silicon and argon being two representative cases ) @xcite .
our explanation in this regard is the following : for the light polarization parallel to the tube axis , the ensuing screening is significant and , therefore , xc effects are damped . for the perpendicular polarization and for the larger inter - wall distance ( @xmath33 5.5 )
the tubes behave essentially like isolated systems , where strong cancellations are known to occur between self - energy corrections and the electron - hole interaction , i.e. between xc effects ( see ref.@xcite ) .
the experimental precision is then not high enough ( also due to the almost vanished absorption intensity in the relevant frequency range ) to discern to which extent xc effects should be better described by approximations beyond the tdlda . in view of these considerations
, we can be confident regarding the quality of the calculations .
on the other hand , the results for the strongly interacting tubes ( smaller intertube distance ) should be regarded as qualitative since the system becomes then more similar to a solid where the cancellations may be more incomplete , and since no direct comparison to experiment is possible , at present , in this case .
still , excitonic effects are not included in the calculations .
they could play a role in both carbon and bn nanotubes , leading to redistribution of oscillator strength and/or appearence of new peaks in the bandgap ( bound excitons ) .
they might be the reason for the anomalous @xmath61 ratio @xcite measured recently for carbon tubes @xcite .
we expect a stronger deviation of measured optical spectra from theoretical ones in the case of bn nanotubes . for this large band gap material , two effects will most likely play an important role : quasi - particle corrections will widen the band gap as in the case of hexagonal bn , where a gw calculation has demonstrated a band - gap increase from 4 ev to 5.5 ev @xcite .
excitonic effects , in contrast , will lead to isolated states in the band gap or to an overall reduction of the band - gap .
to which extent quasi - particle corrections and excitonic effects cancel each other is presently not clear .
work along these lines is in progress @xcite . for the role of quasi - particle corrections and excitonic effects in carbon tubes , we refer the reader to ref .
@xcite . in conclusion ,
the present _ ab initio _ calculations of the optical absorption of small - diameter carbon and bn nanotubes give good agreement with the available experimental data .
the inclusion of local field effects in the response to a perturbation with perpendicular polarization is necessary for a proper description of the depolarization effect leading to a suppression of the low - energy absorption peaks for both types of tubes .
this suppression can also explain recent findings in near - field raman microscopy of single - wall carbon nanotubes @xcite .
the proper analysis of the polarization dependence of the absorption cross section is very important in order to describe the surface enhanced raman scattering experiments in isolated carbon nanotubes @xcite . in carbon tubes
the position of the first absorption peak strongly varies with the tube indices while in bn tubes the first peak is determined by the band gap of bn and is therefore mostly independent of @xmath0 . for the bn tubes some of the fine - structure which distinguishes tubes of different chirality is only visible in the uv region which gives rise to the hope that this energy regime will be probed in the future .
the intertube interaction was also found to be very important .
for the carbon tubes this interaction is the governing factor which determines the position of the higher - frequency @xmath52 plasmon .
this plasmon , hence , may prove to be a very useful spectroscopic quantity probing intertube interactions and distances in real samples .
finally , the corresponding results for the layered constituents graphene and bn sheets revealed that some aspects of the tubular dielectric response can be explained even at a quantitative level from the in - layer and interlayer response of the sheets .
this work was supported by the european community research training networks nanophase ( hprn - ct-2000 - 00167 ) and comelcan ( hprn - ct-2000 - 00128 ) , by spanish mcyt(mat2001 - 0946 ) and university of the basque country ( 9/upv 00206.215 - 13639/2001 ) . the computer time was granted by idris ( project no . 544 ) , dipc and cepba ( barcelona ) .
the authors also gratefully acknowledge fruitful discussions with thomas pichler and nathalie vast . for an overview over the numerous works on resonant raman spectroscopy of carbon nanotubes see , e.g. , r. saito and h. kataura in _ carbon nanotubes : synthesis , structure , properties , and applications _ , m.s .
dresselhaus , g. dresselhaus , and ph .
avouris ( editors ) , springer verlag ( 2001 ) .
l. wirtz , v. olevano , a.g .
marinopoulos , l. reining and a. rubio , in proceedings of _ electronic properties of novel materials : xviith international winterschool _ , ed .
h. kuzmany , j. fink , m. mehring and s. roth , world scientific , singapore ( 2003 ) ; and unpublished results .
de heer _ et al .
_ , science * 268 * , 845 ( 1995 ) ; t. pichler , _ et al .
. lett . * 80 * , 4729 ( 1998 ) ; h. kataura _ et al .
_ synth . met . *
103 * , 2555 ( 1999 ) ; m. ichida _
jpn * 68 * , 3131 ( 1999 ) ; x. liu _
b * 66 * , 045411 ( 2002 ) . c.l . kane and e.j .
mele , phys .
lett . * 78 * , 1932 ( 1997 ) ; r. egger and a.o .
gogolin , _ ibid .
_ , * 79 * , 5082 ( 1997 ) ; c.l .
kane , l. balents and m.p.a .
fisher , _ ibid .
_ , * 79 * , 5086 ( 1997 ) ; r. egger , _ ibid .
_ , * 83 * , 5547 ( 1999 ) .
lin and k.w.k .
shung , phys .
b * 50 * , 17744 ( 1994 ) ; s. tasaki , _ et al . _ , _ ibid . _ * 57 * , 9301 ( 1998 ) ; m.f .
lin , _ et al .
_ , _ ibid .
_ * 61 * , 14114 ( 2000 ) ; f.l . shyu and m.f .
lin , _ ibid .
_ * 62 * , 8508 ( 2000 ) .
see for example , w. kohn , rev .
* 71 * , 1253 ( 1999 ) ; c. fiolhais , f. nogueira , m. marques ( eds . ) , _ a primer in density functional theory _ ( springer , berlin , 2003 ) ; reviews in modern quantum chemistry : a celebration of the contributions of r.g .
parr , k.d .
sen , ed(s ) , ( world scientific , 2002 ) ; and references therein .
l. hedin and s. lundqvist , in _ solid state physics _ , edited by f. seitz , d. turnbull and h. ehrenreich ( academic , new york , 1969 ) , vol .
23 , p. 1 ; m.s .
hybertsen and s.g .
louie , phys .
lett . * 55 * , 1418 ( 1985 ) ; phys .
b * 34 * , 5390 ( 1986 ) ; r.w .
godby , m. schlter and l.j .
sham , phys .
lett . * 56 * , 2415 ( 1986 ) ; phys .
b * 37 * , 10159 ( 1988 ) . @xmath10 is expanded in reciprocal lattice vectors @xmath62 and depends on @xmath63 and on the momentum transfer @xmath20 .
it describes the change of the density @xmath64 in response to the change of the total potential @xmath65 ( which in turn is composed of the hartree , the exchange - correlation and the external potentials ) : @xmath66 . typically , for the systems and excitation energies treated in this work , it was sufficient to compute bands up to @xmath67 , where @xmath68 is the number of occupied bands .
in certain cases , especially for evaluating the loss function , we included more unoccupied bands in the summations .
s. g. louie _
et al . _ , ( to be published ) ; and in proceedings of _ electronic properties of novel materials : xviith international winterschool _ , ed .
h. kuzmany , j. fink , m. mehring and s. roth , world scientific , singapore ( 2003 ) .
calculations are performed for a periodic array of nanotubes with an intertube distance of 14 a.u .
the absolute value of @xmath69 scales with the dimension of the employed super - cell , but we are interested only in the relative absorption cross section @xmath70 in arbitrary units .
the quasi - one dimensional brillouin zones of the tube are sampled by 20 irreducible * k*-points along the tube axis . a.g .
marinopoulos , l. reining , a. rubio and v. olevano ( to be submitted ) .
we have used up to 16464 * k*-points for the brillouin - zone sums , and 121 * g*-vectors in the matrices .
our results are fully converged with respect to the brillouin zone sampling for frequencies above 1 ev . a proper evaluation of the spectra at lower frequencies for this semi - metal would require an improved treatment of intraband transitions , and is beyond the scope of the present work .
model calculations for a linear array of carbon tubes also show an important shift of the higher - frequency plasmon due to intertube interaction : g. gumbs and g.r .
aizin , phys . rev .
b * 65 * , 195407 ( 2002 ) . | we present results for the optical absorption spectra of small - diameter single - wall carbon and boron nitride nanotubes obtained by _ ab initio _ calculations in the framework of time - dependent density functional theory .
we compare the results with those obtained for the corresponding layered structures , i.e. the graphene and hexagonal bn sheets .
in particular , we focus on the role of depolarization effects , anisotropies and interactions in the excited states .
we show that already the random phase approximation reproduces well the main features of the spectra when crystal local field effects are correctly included , and discuss to which extent the calculations can be further simplified by extrapolating results obtained for the layered systems to results expected for the tubes .
the present results are relevant for the interpretation of data obtained by recent experimental tools for nanotube characterization such as optical and fluorescence spectroscopies as well as polarized resonant raman scattering spectroscopy .
we also address electron energy loss spectra in the small - q momentum transfer limit . in this case , the interlayer and intertube interactions play an enhanced role with respect to optical spectroscopy . |
to maintain activities of central starbursts or active galactic nuclei in galaxies , fueling of molecular gas toward the center is necessary .
bar is one of the possible mechanisms , for the gas fueling .
many numerical simulations of gas motion in the bar potential show that bars can accumulate gas toward the central region ( e.g. , athanassoula 1992 ) .
observational evidence of gas accumulation toward the central kilo - parsec region has also been shown .
for example , sakamato et al . ( 1999 ) and sheth et al .
( 2005 ) showed that barred spiral galaxies have higher central concentration of molecular gas than non - barred spiral galaxies .
furthermore , kuno et al . ( 2007 ) showed that the higher central concentration in barred spirals is made by gas accumulation within a region on the order of the bar length and found a correlation between bar strength and the degree of central concentration . on the other hand , observations of molecular gas with high angular resolution sufficient to resolve
the central structure less than 100 pc are still limited ( e.g. , schinnerer et al .
2003 ; garca - burillo et al .
such observations are necessary to understand how the gas flows into the nucleus and relates with central activities .
for example , the size of the twin peaks often seen in the central region of barred galaxies is less than several hundred pc ( kenney et al .
therefore , we need such high angular resolution to investigate the distribution and dynamics of molecular gas in the central region .
maffei 2 is a barred spiral galaxy that belongs to the ic 342/maffei group .
since maffei 2 is located behind the galactic plane , its optical features are not clear .
however , the bar and spiral structures can be seen clearly in a near infrared image ( jarrett et al . 2003 ) .
many co observations have so far been done ( ishiguro et al . 1989 ; kawabe et al .
1991 ; israel , baas 2003 ; mason , wilson 2004 ; kuno et al .
the co map obtained with the nobeyama 45-m telescope shows structures typical of the distribution of molecular gas in barred spiral galaxies ; central peak , offset ridges along the leading side of the large - scale bar , condensations of molecular gas at the end of the bar , and spiral arms from the end of the bar .
they are symmetric against the center .
the motion of molecular gas in the bar is also consistent with that seen in barred spiral galaxies .
namely , the large velocity change is seen across the bar implying that the gas moves along the offset ridges toward the center .
since maffei 2 is one of the nearest barred spiral galaxies , we can investigate the distribution and dynamics of molecular gas in the center and the bar in detail .
furthermore , since maffei 2 has a moderate starburst at the center as suggested from ir and radio observations ( rickard , harvey 1984 ; turner , ho 1994 ) , observations of maffei 2 with high angular resolution are helpful to understand the mechanism of the gas fueling toward the galactic center by bars and relation with a starburst phenomenon in barred spiral galaxies . in this paper , we present the results of @xmath0co(10 ) , @xmath0co(21 ) , cs(21 ) lines and 103 ghz continuum observations of maffei 2 with the nobeyama millimeter array ( nma ) .
we discuss the central structure of molecular gas and star formation in maffei 2 .
we adopt 2.8 mpc as the distance of mafei 2 throughout this paper ( karachentsev et al .
observations in @xmath0co(10 ) , @xmath0co(21 ) , and cs(21 ) were made from dec .
2002 to jan .
2004 using nma .
103 ghz continuum data was also obtained simultaneously with @xmath0co(10 ) .
figure 1 shows the field of views of nma on the co map of maffei2 obtained with the nobeyama 45-m telescope ( kuno et al .
the receiver frontends were sis mixers .
system temperatures ( dsb ) were 100 200 k for cs(21 ) , 150 300 k for co(10 ) , and 200 350 k for co(21 ) .
the uwbc spectro correlator ( okumura et al . 2000 ) which covers 512 mhz with 256 channels was used as backend ( effective frequency resolution : 4 mhz ) .
the bandwidth and the frequency resolution correspond to 1563 km s@xmath8 and 12.2 km s@xmath8 at 98 ghz , 1322 km s@xmath8 and 10.3 km s@xmath8 at 115 ghz , 666 km s@xmath8 and 5.2 km s@xmath8 at 230 ghz .
the data of @xmath0co(10 ) and cs(21 ) were obtained with antenna configurations ab , c , and d , while only c and d were used for @xmath0co(21 ) .
the field center is ( @xmath9 , @xmath10 ) = ( , + ) for all observations . phase and amplitude calibrations were performed every 20 25 minutes using a nearby quasar nrao 150 ( 0359 + 509 ) .
the quasar was also used for band - pass calibration .
the absolute flux of nrao 150 was measured by comparing with uranus and neptune .
the uncertainty of absolute flux is estimated to be about 20 % .
we mapped the @xmath0co data with uniform weight , while natural weight was used for cs(21 ) and 103 ghz continuum .
we achieved @xmath11 synthesized beam in @xmath0co(10 ) , @xmath12 beam in @xmath0co(21 ) .
the beam sizes correspond to 23 pc @xmath13 20 pc and 31 pc @xmath13 19 pc at the distance of maffei 2 . for cs(21 ) and
103 ghz continuum , the beam sizes are @xmath14 and @xmath15 , which correspond to 68 pc @xmath13 49 pc and 67 pc @xmath13 50 pc , respectively .
the observational data are listed in table 1 .
the missing fluxes of the lines are estimated by comparing single dish observations .
flux within the field of view of nma ( @xmath16 ) in the @xmath0co(10 ) map is 3200 jy km s@xmath8 after correction for primary beam attenuation . on the other hand ,
the total flux within the @xmath17 beam measured with a single - dish telescope is 3513 jy km s@xmath8 ( sargent et al .
the missing flux is estimated to be about 10 % .
therefore , most of the total flux was recovered in our @xmath0co(10 ) map . for @xmath0co(21 ) , the flux within the field of view of nma ( @xmath18 ) is 6850 jy km s@xmath8 .
the total flux within the @xmath19 beam measured by sarget et al . ( 1985 ) is 14880 jy km s@xmath8 .
the total flux of nma is only 46 % of the single dish measurement .
however , the total flux within the @xmath20 and @xmath21 beams measured by israel and baas ( 2003 ) is 4700 jy km s@xmath8 and 8527 jy km s@xmath8 , respectively .
these results seem to be consistent with ours .
when we convolve our data to the @xmath20 beam , the peak flux is 3468 jy km s@xmath8 which corresponds to 74 % of the single dish measurement .
total flux of cs(21 ) line is 68 jy km s@xmath8 .
total flux density of 3 mm continuum is 38 mjy .
unfortunately , there are no single dish data of cs(21 ) and 103 ghz continuum .
figure 2 shows the integrated intensity maps of @xmath0co(10 ) and @xmath0co(21 ) .
although the @xmath0co(10 ) map is consistent with the previous map ( ishiguro et al . 1989 ; kawabe et al .
1991 ) , our map shows more detailed structures . in the bar region , the offset ridges along the leading side of the bar can be seen .
the width of the ridge ( fwhm ) deconvolved with the beam size is @xmath22 which corresponds to 50 pc at the distance of maffei2 .
the offset ridges are truncated at about @xmath23 from the center on both sides . at the edges of the offset ridges
, they have bifurcation toward the trailing side .
it is interesting that the same feature is seen in the @xmath0co(10 ) map of ic 342 ( meier et al . 2000 ) .
the offset ridges connect to strong peaks at about @xmath24 ( @xmath25 pc ) north and south of the center .
they can be regarded as the same feature as the twin peaks found by kenney et al .
( 1992 ) in some barred spiral galaxies .
the size of both peaks ( fwhm ) is about @xmath26 ( 54 pc @xmath13 136 pc ) .
the center derived from 2 mass data ( , + ( j2000 ) : jarrett et al .
2003 ) is located between the peaks . at the peaks , the offset ridges , especially the northern ridge ,
show large curvature .
since the inclination angle of maffei 2 is fairly large ( @xmath27 : hurt et al .
1996 ) , some structures overlap along the line of sight . such structures mixed in the peaks can be separated in the velocity domain , as shown in section 3.2 . the distribution of @xmath0co(21 ) emission is consistent with that of @xmath0co(10 ) emission .
the curve of the ridges at the peaks is more apparent in @xmath0co(21 ) , especially in the southern ridge .
although the size of the peaks is about the same as that in @xmath0co(10 ) , they are closer to the center . in both lines ,
the emissions are weak at the center . assuming a conversion factor from co(1 0 ) intensity to column density of h@xmath28 @xmath29h@xmath30 of @xmath31 @xmath32 [ k km s@xmath8]@xmath8 ( nakai , kuno 1995 ) , the molecular gas mass within the field of view ( @xmath34 pc ) is estimated to be about @xmath35 .
this is smaller than the values reported in the previous observations ( e.g. , ishiguro et al . 1989 ) because of the difference in the adopted distance and the conversion factor .
if we use the same distance and conversion factor , our result is consistent with previous studies .
the peak flux of @xmath0co(10 ) is 47.3 jy beam@xmath8 km s@xmath8 which corresponds to the column density of @xmath36 @xmath32 or the surface density of about 2700 @xmath37 pc@xmath38 .
since the width of the ridges ( fwhm ) in the map is about 50 pc , the thickness must be smaller than that .
if we assume that the thickness and inclination angle of the gas disk are 50 pc and 67 deg , respectively , a lower limit of the average volume density within the beam is derived to be @xmath39 @xmath7 .
the peak flux density of @xmath0co(10 ) at the twin peaks is about 600 mjy beam@xmath8 which corresponds to @xmath40 k. on the other hand , the peak flux density of @xmath0co(21 ) is about 2.8 jy beam@xmath8 which corresponds to @xmath41 k. although these values are the average within the beam and a lower limit of the kinetic temperature of the molecular gas , the temperature is higher than that of gmcs in our galaxy .
we will discuss the physical properties of molecular gas including observations of other lines in a forthcoming paper .
figure 3 shows the integrated intensity map of cs(21 ) .
the cs map also show twin peaks .
the peaks coincide with the co peaks .
the northern peak is much stronger than the southern side .
figure 4 shows the 3 mm continuum map .
there is a peak about @xmath42 north of the center with an extended structure along the declination .
there is a weak peak about @xmath43 south of the center .
the overall size is @xmath44 ( @xmath45 ) .
the 3 mm continuum is thought to trace star - forming regions , since it is dominated by free - free emission ( condon 1992 ) .
actually , these structures coincide with those in other tracers of star - forming activity , such as the 2 cm and 6 cm continua ( turner , ho 1994 ) , 10.8 @xmath46 m ( telesco et al . 1993 ) . however , the correlation with h@xmath47 is not good .
the h@xmath47 peak in ishiguro et al .
( 1989 ) is located about @xmath48 north of the 103 ghz peak .
the reason for the bad correlation between free - free emission and h@xmath47 is thought to be dust extinction in h@xmath47 .
a similar trend is seen in m82 ( matsushita et al . 2005 ) .
the total flux density of the 3 mm continuum is 38 mjy .
the peak flux density within the @xmath15 beam is about @xmath49 of the total flux density .
therefore , most of the emission is concentrated in the peak .
we estimate the production rates of lyman continuum photons , @xmath50 , using the flux according to condon ( 1992 ) .
the result derived from the total flux density is @xmath51 where @xmath52 and @xmath53 are electron temperature and distance , respectively .
this is about two times larger than that derived from the 6 cm continuum ( turner , ho 1994 ) after correcting the difference of the adopted distance .
as compared with the starburst galaxy m82 , the derived @xmath50 of maffei 2 is comparable with that of an individual peak in m82 measured with the @xmath54 beam ( matsushita et al . 2005 ) .
the massive star formation rate ( @xmath55 ) derived from the production rate of lyman continuum photons using the equation ( 24 ) in condon ( 1992 ) is @xmath56 . on the other hand ,
the star formation rate derived from the infrared luminosity of @xmath57 ( rickard , harvey 1984 ) using the equation ( 26 ) in condon ( 1992 ) is @xmath56 , if the distance we adopted is used .
they show good agreement .
figures 5 and 6 are the channel maps of @xmath0co(10 ) and @xmath0co(21 ) .
they show quite similar features . near the systemic velocity of maffei 2 ( -24 km
s@xmath8 : kuno et al .
2007 ) , two peaks are seen north and south of the center , shown by a cross in the figures . for a circular motion , the emission should be observed near the minor axis at systemic velocity .
the peaks , however , are located near the major axis , indicating large non - circular motion .
the peaks move toward the north with decreasing radial velocity keeping the separation of the peaks almost constant . on the other hand , they move toward the south with increasing radial velocity .
position - velocity diagrams are helpful to understand the complex velocity structure in maffei 2 .
figure 7 shows the position - velocity diagrams of @xmath0co(21 ) parallel to the major axis of maffei 2 ( p.a .
= @xmath58 : hurt et al .
it is apparent that the central structure is separated into two parallel components in the p v diagrams ( b - c - d , b-c-d ) .
it implies that there are two components along the line of sight .
the radial velocities at the center of both components are not systemic velocity of maffei 2 ( -24 km s@xmath8 ) .
there is no emission at systemic velocity at the center .
the parallel components are clearly separated in the p
v diagram along the minor axis as shown in figure 8 .
since the parallel structures in the p v diagrams are symmetric against the 2 mass center , the center is thought to coincide with the dynamical center .
for the northern component ( upper side ) in figure 7(c ) , a
b corresponds to the offset ridges of the large - scale bar ( a b in figure 7(a ) for the southern component ) .
if we assume that the spiral arms of maffie 2 are trailing arm , the eastern side against the major axis of the galaxy is far side and the large - scale bar approaches us closing to the center for the northern side of the galaxy ( figure 1 ) . therefore , if the gas moves along the large - scale bar , the observed blue - shift of the radial velocities from the systemic velocity means inward motion .
the offset ridge connects to the parallel structure at b ( b for the southern component ) .
the radial velocity along the parallel structure changes from blue - shift to red - shift before the structure crosses the minor axis ( b c ) , although the point should be on the minor axis for a circular motion .
the same feature but in the opposite sense can be seen for the southern component .
furthermore , the parallel components are connected by a weak oval structure , as indicated in figure 7b .
there are bifurcations in the oval structure at c and c. the radial velocity increases closing to the center from the points to d and d.
as mentioned in the previous section , the p v diagrams along the major axis ( figure 7 ) shows two components almost overlapping along the line of sight .
these structures can be regarded as a spiral structure embedded in a weak oval structure .
we made integrated intensity maps of each component of the parallel structure in figure 7 separately using the channel map .
figure 9 shows the integrated intensity map overlaid on the hst image ( f814w ) obtained from the hst archive .
they look like the two - arm spiral structure continues from the offset ridges along the large - scale bar in the co map obtained with the nro 45-m telescope .
the structure can be seen in both @xmath0co(10 ) and @xmath0co(21 ) .
the spiral structure is symmetric against the center determined from the 2mass data .
the northern arm coincides with the dust lane in the hst image very well near the center , while the dust lane along the southern arm is not clear .
this is because the northern arm is on the near side ( western side against the major axis of the galaxy ) near the center .
it is apparent that the oval structure in figure 7b is not an expanding ring .
the part of the oval structure that overlaps with the northern arm is on the near side near the center , as mentioned above .
the radial velocity of the part is red - shift from the systemic velocity of maffei 2 ( around c in figure 7c ) . on the other hand ,
if the ring is shrinking , the velocity toward the center must be higher than 50 km s@xmath8 . in that case , the oval structure collapses into the nucleus in less than @xmath59 yr .
this seems to be unlikely .
the oval structure can be naturally understood as an oval motion of molecular gas .
the oval structure connects the spiral structure .
the schematic figure of the spiral and oval structures is shown in figure 10 . here , we assume inclination and position angle derived for the galactic disk
. similar structures with the offset ridges and oval structure are often seen in numerical simulations ( e.g. , athanassoula 1992 ) .
these structures are made from @xmath11 and @xmath12 orbits in a bar potential .
similar structure has been found in ngc 4303 ( schinnerer et al .
2002 ; koda , sofue 2006 ) .
the complex p v diagrams can be understood with our interpretation .
the offset ridges along the large - scale bar contact with the x2 orbit at b and b in figure 7 .
the constant gradient of the radial velocity b
c and b c in figure 7 is attributed to the flow of the molecular gas along the spiral arms around the center .
the spiral arms diverge from the oval structure toward the center .
this corresponds to c d and c d in the p v diagrams .
the deviation from the systemic velocity becomes max there , since the gas stream line closes to the line of sight .
the sizes of the structure shown in figure 10 were derived from the p v diagrams along the major and minor axes and the channel map as follows : ( 1 ) the separation along the major axis between the spiral arms derived from figure 7(b ) ( @xmath25 pc ) .
( 2 ) the separation along the minor axis between the spiral arms derived from figure 8 ( @xmath60 pc ) . ( 3 ) the separation along the major axis direction between the positions where radial velocity equals to the systemic velocity .
if we assume that most of the gas moves along the spiral arms , the spiral arms are perpendicular to the line of sight at these positions .
this can be measured in channel map at the systemic velocity of maffei 2 ( @xmath61 -24 km s@xmath8 ) ( figure 5 and 6 ) .
( @xmath62 pc ) .
( 4 ) the separation along the minor axis between the positions where radial velocity equals the systemic velocity derived from figures 5 and 6 .
( @xmath63 pc ) .
( 5 ) the length of the oval structure along the major axis derived from figure 7(b ) ( @xmath64 pc ) .
martini et al .
( 2003a ; 2003b ) show from hst images that galaxies with strong bars often have a grand - design nuclear dust spiral structure , which appears to be the continuation of the dust lanes along the leading edges of a large - scale bar toward the center .
on the other hand , observations of such spiral structures of molecular gas are still rare .
the distribution of molecular gas in maffei 2 can be regarded as such a spiral structure . although maffei 2 is classified as sab(rs ) in rc3
, it seems to be due to the unclear optical image . according to buta and block ( 2001 ) , maffei 2
should be classified as sb(rs ) .
therefore , our result is consistent with martini(2004 ) s finding that grand - design spiral structure is only found in sb(s ) or sb(rs ) .
some numerical simulations show that when there is a massive concentration at the center , such as a super massive black - hole , a nuclear spiral structure is formed ( fukuda et al .
1998 ; ann , lee 2004 ; maciejewski 2004 ) .
this is because the mass concentration removes the rigid rotation part from the rotation curve . as a result ,
iilr ( inner inner lindblad resonance ) , where a circum - nuclear ring is formed , disappears or another resonance ( nuclear lindblad resonance ) appears inside the iilr .
the mass of the central concentration assumed in numerical simulations to make the spiral structure is 0.1 % 1 % of the total mass of the host galaxies .
the observational data of molecular gas has the advantage that it can be used to investigate the kinematics of the central structure . from the p
v diagram along the major axis ( figure 7 ) , the rotation velocity is estimated to be about 140 km s@xmath8 at @xmath65 pc .
the central mass within @xmath3 35 pc is estimated to be @xmath4 from the rotation velocity assuming spherical distribution of mass .
the mass is about 0.5 % of the total dynamical mass of maffei 2 ( mason , wilson 2004 ) .
even if the rotation velocity is overestimated due to the non - circular motion , it seems to be more than 0.1 % of the dynamical mass ( koda , wada 2002 ) .
the mass fraction of the central mass concentration in maffei 2 is comparable to the prediction of the simulations .
the rotation velocity still rising toward the center at @xmath65 pc .
it is desirable to obtain the inner rotation curve with higher spatial resolution and higher sensitivity observations to know the mass distribution in the inner region .
the molecular gas changes the direction of motion with a large velocity change when it enters the offset ridge along the large - scale bar .
then , the gas flows toward the center along the offset ridge .
since the radial velocity is almost constant along the straight ridges , as seen in figure 7 , the streaming velocity along the ridge seems to be constant .
( the component of the pattern speed of the bar is relatively small as compared with the velocity along the bar . )
the velocity along the ridge is estimated to be 70 - 120 km s@xmath8 assuming the inclination angle and position angle of the galactic disk . here , we assumed a pattern speed of the bar of 53.5 km s@xmath8 kpc@xmath8 which was adopted from hurt et al .
( 1996 ) and made a correction for the difference of the adopted distance .
the velocity is comparable with those in ngc 253 ( sorai et al . 2000 ) and ngc 1530 ( regan et al . 1997 ) . using the velocity and surface density of molecular gas at the radius of 200 pc in the offset ridge ,
the mass flow of molecular gas along the ridges is derived to be 712 @xmath66 .
this is comparable with the estimation for ngc 1530 by regan et al .
( 1997 ) .
if the molecular gas in the flow is accumulated in the center directly , it takes only @xmath67 yr to concentrate all of the molecular gas in the disk on the oval structure .
however , only a portion of the flowing gas must be accumulated in the central region as estimated by regan et al .
. they evaluated it to be 20 % of the flowing gas .
there are concentrations of molecular gas where the offset ridges connect to the spiral structure ( figures 7 and 9 ) .
the mass of the condensations within 65 pc @xmath13 94 pc is estimated to be @xmath68 from the @xmath0co(10 ) intensity using a conversion factor of @xmath31 @xmath32 [ k km s@xmath8]@xmath8 ( nakai , kuno 1995 ) .
the peaks correspond to @xmath11/@xmath12 orbit - crowding region .
we compare the position of the peaks with the peaks in the cs(21 ) and 3 mm continuum maps .
since the northern peak is much stronger than the southern peak in the cs(21 ) and 103 ghz maps , we discuss the northern peak here .
the peak of the 3 mm continuum which is a tracer of star - forming regions is located downstream of the co peak .
the offset between the co and 3 mm peaks is about 20 pc , while the cs peak coincides with the co peak .
we can estimate the time scale for the gas to move from the cs peak to the 3 mm peak along the ridge using the streaming velocity measured at the offset ridge , 70 120 km s@xmath8 , assuming that the peaks are near the point where the spiral structure is perpendicular to the line of sight ( cf .
figure 10 ) .
the time scale is about 1.6
2.8 @xmath69 yr . on the other hand , free fall time of the dense gas traced by cs ( @xmath6 @xmath7 ) is estimated to be @xmath70 yr .
this is comparable to the time for the gas to move from the cs peak to the 3 mm peak . from these results ,
the following scenario can be described : the molecular gas in @xmath11 orbit moves along the offset ridge in the large bar . at the @xmath11/@xmath12 orbit - crowding region , the gas contacts with @xmath12 orbit and is compressed by shock and the density rises abruptly there .
once the dense gas is formed , it evolves into stars with free - fall time .
it is interesting to compare the structure with that seen in ngc 6951 ( kohno et al .
kohno et al .
( 1999 ) show that the condensations of dense gas traced by hcn in ngc 6951 are located downstream of more diffuse gas traced by co. they showed that the time scale of gravitational instability and flowing time of the gas along the ridges are comparable ( @xmath71 yr ) .
this result supports the idea that the dense gas is made by gravitational instability during the flow along the ridge .
we speculate that the difference of our results from kohno et al .
( 1999 ) may be caused by the difference in the size of the structure .
the structure found in ngc 6951 is several times larger than that in maffei 2 .
the size of the co peak is more than hundreds pc , while the condensations in maffei 2 are comparable to the galactic gmcs . therefore , the molecular clouds should be treated as particles in ngc 6951 . in that case , since the viscosity of the gas is small , shock may not play an important role in dense gas formation at the @xmath11/@xmath12 orbit - crowding region , while the gravitational instability does . on the other hand ,
the clump of molecular gas whose size is comparable with a gmc is compressed at the crossing point in maffei 2 . as a result , strong shock may occur and dense gas is formed at a burst there .
we made @xmath0co(10 ) , @xmath0co(21 ) , cs(21 ) lines and 103 ghz continuum observations of maffei 2 . using the data set , we investigated the distribution and kinematics of molecular gas in the central region .
the results are summarized as follows : + ( 1 ) we revealed the spiral structure of molecular gas in the central region .
the offset ridges of molecular gas along the leading side of the large - scale bar continue to the spiral structure embedded in the weak oval structure which is regarded as a x2 orbit .
the size of these structures is less than @xmath3 100 pc .
the spiral structure continues toward the center at least until @xmath3 35 pc diverging from the oval structure .
+ ( 2 ) the central mass within @xmath3 35 pc is estimated to be @xmath4 from the rotation curve , which corresponds to 0.5 % of the dynamical mass of maffei 2 .
the percentage is consistent with the theoretical predictions to make such a spiral structure of molecular gas .
+ ( 3 ) the amount of the molecular gas which flows along the offset ridges of the large - scale bar is estimated to be 712 @xmath66 .
this is an upper limit of the mass flow of molecular gas into the nucleus .
+ ( 4 ) our results imply that dense gas is formed at the crossing points of @xmath11 and @xmath12 orbits .
massive stars are formed from the dense gas and star - forming regions appear after free - fall time of the dense gas ( @xmath72 yr ) .
we thank the nma staff for their kind support and encouragement .
this research used the facilities of the canadian astronomy data centre operated by the national research council of canada with the support of the canadian space agency . &
cs(21 ) & 103 ghz cont . &
@xmath0co(10 ) & @xmath0co(21 ) + array configuration & ab , c , d & ab , c , d & ab , c , d & c , d + velocity coverage ( km s@xmath8 ) & 1563 & & 1322 & 666 + velocity resolution ( km s@xmath8 ) & 30.5 & & 10.3 & 7.8 + synthesized beam ( @xmath73 , @xmath74 ) & @xmath75 & @xmath76 , -35 & @xmath77 , 27 & @xmath78 , -34 + synthesized beam ( pc ) & @xmath79 & @xmath80 & @xmath81 & @xmath82 + rms noise of channel map ( mjy beam@xmath8 ) & 8 & 2 & 40 & 130 + total flux within field of view ( jy km s@xmath8 ) & 68 & 38 & 3200 & 6850 + | distribution and kinematics of molecular gas in the central region of the barred spiral galaxy maffei 2 were investigated using a data set of @xmath0co(10 ) , @xmath0co(21 ) , cs(21 ) lines and 103 ghz continuum .
we found that the offset ridges along the kpc - scale bar continue to the central spiral structure embedded in the weak oval structure which is regarded as @xmath12 orbit in the bar potential .
the spiral structure continues toward the center diverging from the oval structure .
the size of these structures is less than @xmath2 pc . the mass concentration within @xmath3 35 pc is estimated to be @xmath4 .
the high mass concentration is consistent with theoretical predictions concerning the creation of such a nuclear spiral structure . a comparison with the tracers of dense gas and star - forming region suggests that the dense molecular gas traced by cs(21 ) line is formed at the crossing points of @xmath11 and @xmath12 orbits and the star - forming region appears after @xmath5 yr which is comparable with the free - fall time of the dense gas traced by the cs line ( @xmath6 @xmath7 ) . |
let us first explain the unconventional way of determining neutrino mass hierarchy , the `` @xmath0 disappearance '' method @xcite , to be explored in this paper .
suppose that we measure the atmospheric @xmath0 by doing disappearance measurements of @xmath21 and @xmath22 in vacuum . for clarity ,
we start with a simplified setting where the solar mixing angle @xmath23 vanishes . in this case
, it is obvious that @xmath24 and @xmath25 channels are governed by @xmath26 and approximately @xmath27 , respectively , since the two oscillations scales approximately decouple from each other .
notice that because @xmath28 , @xmath29 for the normal hierarchy and @xmath30 for the inverted hierarchy .
therefore , in the hypothetical simplified world of vanishing @xmath23 , one could , in principle , determine the mass hierarchy just by comparing the absolute values of the two @xmath0 , @xmath31 and @xmath32 . for non - zero @xmath23 ,
the same considerations apply except that the difference is reduced by approximately @xmath33 . to make the discussion transparent we introduce , following @xcite , the effective atmospheric @xmath34 determined by the disappearance measurement in the @xmath35 channel .
it is this effective @xmath36 , obtained by a two - flavor approximate description of the full three - flavor situation , which contains to high accuracy all the three - flavor mixing effects .
let us briefly review the derivation of @xmath34 . in the approximate two - flavor mixing framework ,
the survival probability is given by @xmath37 the energy @xmath38 at which @xmath39 is maximum is given by @xmath40 .
now we repeat the same computation with the full three - flavor expression of the survival probability ; @xmath41 can be obtained by solving @xmath42/de = 0 $ ] . by demanding that @xmath43 , an expression for @xmath44 can be derived .
assuming vacuum oscillations and ignoring the terms of the order of @xmath45 , we obtain @xcite , @xmath46 where @xmath47 the @xmath48 s are elements of the neutrino mixing matrix @xcite , for which we use the standard parametrization @xcite .
the physical meaning of @xmath49 is that it is the fraction of the @xmath50 flavor in the @xmath51 eigenstate over the sum of this fraction in the @xmath51 and @xmath52 eigenstates .
note that @xmath53 without further approximation and @xmath54 if @xmath55 .
the difference @xmath56 of the two effective @xmath0 can be easily computed from eq .
( [ dm2eff ] ) and ( [ defr ] ) to be @xmath57 where @xmath58 which is positive for small values of @xmath59 , and @xmath60 sign is for the normal ( inverted ) hierarchy
. implies @xmath61 .
] thus the _ sign _ of @xmath62 is the same as the _ sign _ of @xmath26 .
therefore , within the experimentally allowed region for the mixing parameters , the hierarchy is normal if @xmath56 is positive and inverted if @xmath56 is negative .
one should be reminded that _ both _ measurements , @xmath63 and @xmath2 , have to be accurate to the 1% level at least , because @xmath64 1% for the current best fitted parameters .
a more elaborate treatment in @xcite entailed a rough estimation that both accuracies must be better than 0.5% .
therefore , the issue is whether one can at least envisage experimental setups capable of measuring both @xmath0 to such a high accuracy .
it appears that precision determination of @xmath2 is more feasible because there exists a well known method : the spectrum measurement in the @xmath19 disappearance channel .
if it can be performed with high precision , it should be possible to determine @xmath2 with great accuracy .
unfortunately , there is a potential obstacle to accurate measurement of @xmath2 , the problem of _ absolute energy scale uncertainty_. for concreteness , we consider the phase ii of the t2k project ( t2k ii ) @xcite where the beam power from j - parc is upgraded to 4 mw and the far detector will be hyper - kamiokande with 0.54 mt fiducial volume . with huge number of events obtained in several years of running , the precision of @xmath2 would reach to @xmath65% level without the energy scale uncertainty @xcite .
to determine @xmath0 accurately , however , we have to know precisely the absolute value of the energy of the detected muons to reconstruct the neutrino energies .
since the oscillation probability depends on @xmath66 , the accuracy of @xmath2 measurement is limited by the energy scale error . at the moment
, the absolute energy of muons are known to the accuracy of 2% at energies @xmath67 gev in the super - kamiokande detector @xcite .
lacking better calibration sources , one expects that the energy scale uncertainty will be the major limiting factor in improving the @xmath2 determination in the t2k ii experiment @xcite .
notice that the difficulty is not specific to this experiment and similar statements would apply to other detection methods as well . if the 2% energy scale uncertainty still remains , our method for determining the mass hierarchy would not work because then the maximum precision we can expect on @xmath2 is 2% .
see appendix [ bound ] .
a precision better than 1% is needed for the method discussed in this paper to be useful . in this work ,
we take the optimistic attitude that this uncertainty can be reduced to the level of 0.5%-1% by the time t2k ii will be realized , or by development of alternative detection technologies .
we study to what extent one can determine the mass hierarchy with the method described in the previous section .
we believe that physics results we can achieve under such an assumption are worthwhile to report . for definiteness
, we assume that @xmath2 can be determined with 0.5% accuracy .
we also examine the case in which the accuracy is 1% to reveal how the capability of the mass hierarchy determination depends upon the accuracy of @xmath2 .
we note that due to the fact that a large oscillation effect in the @xmath68 mode is expected , the precision in the determination of @xmath2 essentially does not depend on the precise values of the unknown mixing parameters such as @xmath7 and @xmath69 , and possible deviation of @xmath70 from @xmath71 .
therefore , the fractional accuracy , @xmath72 can be regarded , to good approximation , as constant for a given systematic uncertainty . also , since matter effects change the size of @xmath2 by less than 0.1% they can be ignored for the t2k measurement of @xmath2 , see @xcite .
most probably , the most difficult part of the method for the determination of the neutrino mass hierarchy explored in this paper is to measure @xmath3 to the level of a few parts per mil .
the reactor @xmath7 experiments @xcite , as they stand , will not reach the accuracy of 1% level .
recently , the intriguing possibility of using the resonant absorption reaction @xcite @xmath74 to explore @xmath5 interaction below the threshold of the charged - current absorption reaction @xmath75 , has been proposed by raghavan @xcite . see ref .
@xcite for earlier suggestions .
the resonance condition is automatically satisfied if the @xmath5 beam is prepared with the use of the inverse reaction @xmath76 .
if both , the source and the target , atoms are placed in a metal lattice and can enjoy the same environment the resonant cross section can be enhanced by the mssbauer effect by a factor of @xmath77 @xcite .
see also @xcite where some potential difficulties of this experiment are discussed . because the energy of @xmath5 from the bound state beta decay is so low , @xmath78 kev
, the first oscillation maximum ( minimum of @xmath79 ) is reached at the baseline distance @xmath80 m. then , one can envisage @xmath7 experiments with @xmath8110 m baseline @xcite . with the mssbauer enhanced cross section of @xmath82 the rate @xmath83 , with @xmath84 being the number of target atoms and @xmath85 the neutrino flux , is given by @xmath86 that is , a million events a day by using 100 g of @xmath87he target mass @xmath88 and assuming a source strength @xmath89 of 1 mci .
it is natural to expect that the monochromatic nature of the @xmath5 beam as well as the extremely high statistics allow one to measure @xmath3 to a high accuracy , as demonstrated by the recent analysis in @xcite . for the purpose of the present analysis
, we use one of the particular settings discussed in that paper , the one called run iib , which is defined as follows : run iib : it consists of measurements at 10 different detector locations ; @xmath90 for @xmath91 , so that the entire period , from 0 to 2@xmath92 , is covered . at each location
an equal number of @xmath93 events is to be collected .
the authors of @xcite argued that if the direct counting of produced @xmath94 atoms works , a movable detector technique would allow for a relative systematic uncertainty as low as 0.2% , and if not it may be of the order of @xmath951% .
if the former uncertainty is realized , a sensitivity to @xmath3 of @xmath96% at 1@xmath97 cl is possible . for the latter uncertainty ,
the sensitivity is worse by about a factor of four @xcite . in this work ,
we consider this systematic uncertainty to be 0.2% , and also discuss the case where it is 1% .
given the accuracies of the measurements of @xmath2 and @xmath3 discussed in secs .
[ mumu ] and [ ee ] , it is straightforward to determine the region of mixing parameters in which the mass hierarchy can be resolved . to understand what are the relevant parameters , we show in fig .
[ 1sigma - region ] the 1@xmath97 cl determination of @xmath3 and @xmath2 as a function of @xmath98 for @xmath99 ( left panel ) , 0 ( middle panel ) and 1 ( right panel ) .
the physical parameters , @xmath100 and @xmath101 , are held fixed , which implies that @xmath3 is also fixed .
if the two strips of @xmath3 and @xmath2 do not overlap then the mass hierarchy is determined at better than 84% cl .
even though they overlap the mass hierarchy can be determined to the extent that one can discriminate if @xmath102 is positive ( normal hierarchy ) or negative ( inverted hierarchy ) . throughout this section
we use the following values for the solar oscillation parameters : @xmath103ev@xmath104 and @xmath105 @xcite , unless stated otherwise .
-0.5 cm ( shaded by orange color ) and @xmath2 ( bands delimited by two solid and dashed curves ) by measurement using the recoilless resonant @xmath5 absorption reaction and the t2k ii experiment , respectively , are plotted as functions of @xmath98 .
the input value of @xmath106 is assumed .
the red solid ( blue dashed ) curve for @xmath107 denotes the case of normal ( inverted ) mass hierarchy .
the left , the middle and the right panels are for the input values of @xmath108 , @xmath109 or @xmath110 , and @xmath111 or @xmath112 , respectively . , title="fig:",scaledwidth=102.0% ] -0.4 cm 0.2 cm a few remarks are in order : \(1 ) the dependence of the fractional uncertainty of @xmath3 which is proportional to @xmath113 @xcite is clearly visible in fig .
[ 1sigma - region ] .
\(2 ) @xmath2 varies as a function of @xmath98 because of the three - flavor effect in the disappearance probability @xmath114 , see eq .
( [ delta_emu ] ) .
note , however , that the relative uncertainty with respect to its central value is independent of @xmath7 .
\(3 ) the three panels in fig .
[ 1sigma - region ] , which correspond to different values of @xmath69 , indicate that the discriminating sensitivity of the mass hierarchy depends upon @xmath69 in an interesting way .
the sensitivity is highest ( lowest ) at @xmath115 ( 0 or @xmath112 ) , see eq .
( [ delta_emu ] ) . to quantify the sensitivity region for the resolution of the mass hierarchy we define the probability distribution function @xmath116 of the difference @xmath117 .
then the region of parameter which gives positive @xmath118 at @xmath1090 , @xmath1095 and @xmath1099% cl are determined by the condition @xmath119 assuming that @xmath3 and @xmath2 are gaussian distributed distribution of @xmath1 is gaussian as far as we exploit the setting discussed in @xcite .
] , @xmath120 and @xmath121 , with the average values @xmath122 and @xmath123 and widths @xmath124 and @xmath125 , respectively , @xmath126 is also a gaussian distribution with average value @xmath127 and width @xmath128 . using the precision for the determination of @xmath2 and @xmath3 obtained in secs .
[ mumu ] and [ ee ] , it is straightforward to determine the sensitivity regions . in fig .
[ sensitivity - region ] we present the sensitivity regions in the space spanned by @xmath98 and @xmath69 at @xmath10 90% ( green ) , @xmath10 95% ( yellow ) , and @xmath10 99% ( red ) cl in which the mass hierarchy can be resolved by the method of comparing these two disappearance measurements .
cm -0.5 cm plane in which the mass hierarchy can be resolved at @xmath1090% ( green ) , @xmath10 95% ( yellow ) , and @xmath1099% ( red ) cl by the method of comparing the two disappearance measurements .
the uncertainty on @xmath3 is roughly given by @xmath129 under the assumed 0.2% systematic error and the uncertainty on @xmath2 is assumed to be 0.5% . here the current best fit values @xmath130 ,
, title="fig:",scaledwidth=55.0% ] -2.0 cm as anticipated in the remark after fig .
[ 1sigma - region ] , the sensitivity depends significantly on the cp violating phase @xmath69 .
it is highest at @xmath108 , and lowest at @xmath111 or @xmath112 .
in fact , this behavior is easy to understand from eq .
( [ delta_emu ] ) , given that @xmath131 is positive definite , the highest ( lowest ) sensitivity is reached at @xmath99 ( + 1 ) . -0.50 cm -0.5 cm but for @xmath132 ( left panel ) and 0.34 ( right panel ) which are allowed by the solar neutrino data and kamland at 2@xmath97 cl . , title="fig:",scaledwidth=50.0% ]
but for @xmath132 ( left panel ) and 0.34 ( right panel ) which are allowed by the solar neutrino data and kamland at 2@xmath97 cl .
, title="fig:",scaledwidth=50.0% ] -1.8 cm the strong dependence of the sensitivity on @xmath69 suggests that similar significant dependences exist also on other quantities , @xmath23 , @xmath133 and @xmath134 . in fig .
[ sensitivity - region2 ] , the sensitivity regions for determining the mass hierarchy are plotted for two values of @xmath23 different from the best fit one but allowed by the current data at 2@xmath97 cl .
it is remarkable that the sensitivity regions depend so strongly on @xmath23 which corresponds to @xmath135 .
the smaller @xmath23 the better the sensitivity . at small @xmath7 the second term in the right - hand side of eq . ( [ delta_emu ] ) is small , and the difference between the two @xmath0 is governed by the first term .
the term is larger for smaller @xmath23 in the first octant , and hence the sensitivity is higher .
similar argument applies to @xmath136 .
if the true value of this ratio turns out to be larger ( smaller ) than the current best fit , it will be easier ( more difficult ) to determine the hierarchy .
we have presented the dependence of the sensitivity on @xmath23 because the significant improvement of the accuracy is not expected for this quantity , unless either precision measurement of low - energy solar neutrino flux is realized @xcite , or a dedicated reactor @xmath23 measurement is carried out @xcite . on the contrary
, the accuracy of @xmath137 determination will be improved to about 5% level by kamland alone as 10 times more statistics is accumulated by this experiment together with a better control of the systematic uncertainties @xcite . of course , there is no need to discuss the dependence on @xmath138 and @xmath139 because they must be determined with high accuracies to realize the method .
to determine @xmath140 accurately we need to solve the @xmath70 octant degeneracy by either the atmospheric @xcite or the reactor - accelerator combined methods @xcite .
-0.50 cm
-0.5 cm except : ( left panel ) the uncertainty on @xmath2 is increased by a factor of 2 to @xmath141 , ( right panel ) sytematic error for the @xmath142 determination was increased by a factor of 5 ( from 0.2% to 1% ) which implies that the uncertainty on @xmath142 is roughly described by @xmath143 with the uncertainty on @xmath144 returned to @xmath145 .
, title="fig:",scaledwidth=50.0% ] except : ( left panel ) the uncertainty on @xmath2 is increased by a factor of 2 to @xmath141 , ( right panel ) sytematic error for the @xmath142 determination was increased by a factor of 5 ( from 0.2% to 1% ) which implies that the uncertainty on @xmath142 is roughly described by @xmath143 with the uncertainty on @xmath144 returned to @xmath145 .
, title="fig:",scaledwidth=50.0% ] -1.8 cm let us make some comments about the possibility to determine the cp phase @xmath69 .
unfortunately , it will not be possible to establish cp violation ( @xmath146 ) by this method without even further improvement on the precision with which both @xmath0 s are measured .
this can be understood by looking at fig .
[ 1sigma - region ] .
we can see that any point within the allowed region of @xmath2 delimited by the red solid ( blue dashed ) lines for @xmath147 ( corresponding to @xmath109 or @xmath110 , i.e. , cp violation ) can also be within the allowed region either for @xmath148 or @xmath99 ( corresponding to no cp violation ) .
however , if cp is not violated , there is some possibility to distinguish @xmath149 from @xmath108 since the two allowed regions for @xmath148 and @xmath99 have little overlap for @xmath8 .
it is interesting to note that accelerator experiments like t2k ii which will operate at the first oscillation maximum will have difficulty to distinguish the two points @xmath149 and @xmath92 @xcite .
finally , let us also discuss the case where one of the assumption on the precision for @xmath0 is not valid . in the left panel of fig .
[ sensitivity - region3 ] , we show the same information as in fig . [ sensitivity - region ] but assuming that the precision for @xmath2 is 1% ( instead of 0.5% ) but keeping the accuracy for @xmath3 the same as before . in this case
the sensitivity is significantly reduced compared to the case presented in fig .
[ sensitivity - region ] , but there is still significant parameter space where the mass hierarchy can be determined at @xmath10 90% cl . on the other hand , in the right panel of fig .
[ sensitivity - region3 ] , we show the same information as in fig . [ sensitivity - region ] but assuming the systematic uncertainty for the mssbauer experiment is 1% ( instead of 0.2% ) , which implies about the factor four larger uncertainty @xmath150 , on @xmath138 , but keeping the accuracy for @xmath2 the same as before ( 0.5% ) . in this case , there is much less parameter space where we can determine the mass hierarchy .
we have shown that if @xmath2 and @xmath63 can be measured , respectively , with 0.5% and @xmath151 % accuracy , the neutrino mass hierarchy can be resolved down to @xmath152 for @xmath153 at @xmath1095% cl for the current best fit values of the oscillation parameters . for slightly different values of the oscillation parameters , still allowed by data , the sensitivity can be even higher . although it is not possible to determine the cp phase @xmath69 using these two measurements , it may be possible to distinguish the case @xmath111 from @xmath108 .
it should be stressed that high precision measurements of the muon and electron disappearance oscillation experiments considered in this paper have their own good physics motivation .
if these two experiments will be realized , then as a special bonus , there is a possibility that the mass hierarchy can be determined , even though neither of these experiments alone can accomplish this task .
existing technologies seem to allow us to realize the experiment such as the second phase of the t2k project , in which the determination of @xmath2 up to 2% precision looks quite feasible .
how much one will be able to improve this precision depends essentially on how much one can reduce the energy scale uncertainty . if it can be reduced to less than 1% , then it is worthwhile to consider the method studied in this paper . the method discussed in this paper paper , ``
@xmath0-disappearance '' method , is most sensitive when @xmath69 is near @xmath92 . whereas long baseline experiments are most sensitive when @xmath154 ( @xmath155 ) for the normal ( inverted ) mass hierarchy , @xcite .
thus , these two methods are complimentary in the cp violating phase variable , @xmath69 .
on the other hand , it is not yet clear if the oscillation experiment based on the mssbauer enhanced resonant @xmath5 absorption can be really performed .
if it will be realized , the direct counting of @xmath87h atoms in the target is quite essential to carry out the measurement of @xmath3 to the required accuracies . to measure both @xmath3 and @xmath2 to the accuracy less than 1% is challenging , but
if it is possible , this will provide a direct and alternative way to determine the neutrino mass hierarchy .
is supported by bilateral programs , scientist exchanges based on japan society of promotion of science and brazilian academy of sciences .
he is grateful to departamento de fsica , pontifcia universidade catlica ( puc ) do rio de janeiro for hospitality where this work was initiated and largely completed during his visit .
we thank e. lisi , t. kajita , t. nakaya , w. potzel and t. schwetz for useful discussions and correspondences .
this work was supported in part by the grant - in - aid for scientific research , no .
16340078 , japan society for the promotion of science , fundao de amparo pesquisa do estado de so paulo ( fapesp ) , fundao de amparo pesquisa do estado de rio de janeiro ( faperj ) and conselho nacional de cincia e tecnologia ( cnpq ) .
fermilab is operated under doe contract no .
de - ac02 - 76ch03000 .
we assume , as an idealistic situation , that the neutrino beam is monitored by a near detector with identical detection apparatus to a far detector for measuring modulation of the neutrino spectrum .
also , the neutrino source is assumed to be a point source for both the near and far detectors . with the energy scale uncertainty @xmath156 , the true and the reconstructed neutrino energies , @xmath157 and @xmath158 , are related in both detectors by @xmath159 where @xmath160 . for simplicity , we ignore all the other uncertainties except for the energy scale one and assume that the fractional uncertainty @xmath161 is constant over the energy range of interest . in this case
, it is clear that the number of events observed per unit of detector mass at the near and far detectors , @xmath162 and @xmath163 , respectively , are related to the neutrino disapperance probability @xmath164 by 99 for a recent review , see r. n. mohapatra and a. y. smirnov , arxiv : hep - ph/0603118 .
y. ashie _ et al . _
[ super - kamiokande collaboration ] , phys .
d * 71 * , 112005 ( 2005 ) [ arxiv : hep - ex/0501064 ] .
n. tagg [ the minos collaboration ] , arxiv : hep - ex/0605058 .
f. di capua [ opera collaboration ] , pos * hep2005 * , 177 ( 2006 ) .
y. itow _ et al .
_ , arxiv : hep - ex/0106019 .
for an updated version , see + http://neutrino.kek.jp/jhfnu/loi/loi.v2.030528.pdf v. d. barger , s. geer , r. raja and k. whisnant , phys .
b * 485 * , 379 ( 2000 ) [ arxiv : hep - ph/0004208 ] ; a. cervera , a. donini , m. b. gavela , j. j. gomez cadenas , p. hernandez , o. mena and s. rigolin , nucl .
b * 579 * , 17 ( 2000 ) [ erratum - ibid .
b * 593 * , 731 ( 2001 ) ] [ arxiv : hep - ph/0002108 ] ; h. minakata and h. nunokawa , jhep * 0110 * , 001 ( 2001 ) [ arxiv : hep - ph/0108085 ] .
p. huber , m. lindner and w. winter , nucl .
b * 645 * , 3 ( 2002 ) [ arxiv : hep - ph/0204352 ] ; v. barger , d. marfatia and k. whisnant , phys .
d * 65 * , 073023 ( 2002 ) [ arxiv : hep - ph/0112119 ] ; h. minakata , h. nunokawa and s. j. parke , phys .
d * 68 * , 013010 ( 2003 ) [ arxiv : hep - ph/0301210 ] ; o. mena and s. j. parke , phys .
d * 70 * , 093011 ( 2004 ) [ arxiv : hep - ph/0408070 ] ; o. mena requejo , s. palomares - ruiz and s. pascoli , phys .
d * 72 * , 053002 ( 2005 ) [ arxiv : hep - ph/0504015 ] .
m. ishitsuka , t. kajita , h. minakata and h. nunokawa , phys .
d * 72 * , 033003 ( 2005 ) [ arxiv : hep - ph/0504026 ] .
d. s. ayres _
[ nova collaboration ] , arxiv : hep - ex/0503053 .
h. nunokawa , s. parke and r. zukanovich funchal , phys . rev .
d * 72 * , 013009 ( 2005 ) [ arxiv : hep - ph/0503283 ] .
a. de gouvea , j. jenkins and b. kayser , phys .
d * 71 * , 113009 ( 2005 ) [ arxiv : hep - ph/0503079 ] .
r. s. raghavan , arxiv : hep - ph/0511191 ; arxiv : hep - ph/0601079 .
z. maki , m. nakagawa and s. sakata , prog .
* 28 * , 870 ( 1962 ) .
see also , b. pontecorvo , zh .
. fyz . * 53 * , 1717 ( 1967 ) [ sov .
jetp * 26 * , 984 ( 1968 ) ] .
s. eidelman _ et al .
_ [ particle data group collaboration ] , phys . lett .
b * 592 * , 1 ( 2004 ) .
b. aharmim _ et al . _
[ sno collaboration ] , phys .
c * 72 * , 055502 ( 2005 ) [ arxiv : nucl - ex/0502021 ] .
k. hiraide , h. minakata , t. nakaya , h. nunokawa , h. sugiyama , w. j. c. teves and r. zukanovich funchal , phys .
d * 73 * , 093008 ( 2006 ) [ arxiv : hep - ph/0601258 ] .
t. kajita , private communications .
y. kozlov , l. mikaelyan and v. sinev , phys . atom . nucl . * 66 * , 469 ( 2003 ) [ yad .
fiz . * 66 * , 497 ( 2003 ) ] [ arxiv : hep - ph/0109277 ] ; h. minakata , h. sugiyama , o. yasuda , k. inoue and f. suekane , phys .
d * 68 * , 033017 ( 2003 ) [ erratum - ibid .
d * 70 * , 059901 ( 2004 ) ] [ arxiv : hep - ph/0211111 ] .
white paper report on using nuclear reactors to search for a value of @xmath7 , arxiv : hep - ex/0402041 . h. minakata and s. uchinami , arxiv : hep - ph/0602046 .
m. nakahata , nucl .
suppl . * 145 * , 23 ( 2005 ) .
h. minakata , h. nunokawa , w. j. c. teves and r. zukanovich funchal , phys .
d * 71 * , 013005 ( 2005 ) [ arxiv : hep - ph/0407326 ] ; nucl .
suppl . * 145 * , 45 ( 2005 ) [ arxiv : hep - ph/0501250 ] .
a. suzuki , talk at the 3rd international workshop on neutrino oscillations in venice ( no - ve 2006 ) , venice , italy .
7 - 10 , february 2006 .
the idea was suggested by o. l. g. peres and a. y. smirnov , phys .
b * 456 * , 204 ( 1999 ) [ arxiv : hep - ph/9902312 ] ; nucl .
b * 680 * , 479 ( 2004 ) [ arxiv : hep - ph/0309312 ] , and explored e.g. , by m. c. gonzalez - garcia , m. maltoni and a. y. smirnov , phys .
d * 70 * , 093005 ( 2004 ) [ arxiv : hep - ph/0408170 ] ; t. kajita , talk at next generation of nucleon decay and neutrino detectors ( nnn05 ) , aussois , savoie , france , april 7 - 9 , 2005 .
t. kajita , h. minakata and h. nunokawa , phys .
b * 528 * , 245 ( 2002 ) [ arxiv : hep - ph/0112345 ] .
h. minakata and h. nunokawa , third reference in @xcite . o. mena and s. j. parke , phys .
d * 72 * , 053003 ( 2005 ) [ arxiv : hep - ph/0505202 ] . | recently a new method for determining the neutrino mass hierarchy by comparing the effective values of the atmospheric @xmath0 measured in the electron neutrino disappearance channel , @xmath1 , with the one measured in the muon neutrino disappearance channel , @xmath2 , was proposed .
if @xmath3 is larger ( smaller ) than @xmath4 the hierarchy is of the normal ( inverted ) type .
we re - examine this proposition in the light of two very high precision measurements : @xmath2 that may be accomplished by the phase ii of the tokai - to - kamioka ( t2k ) experiment , for example , and @xmath3 that can be envisaged using the novel mssbauer enhanced resonant @xmath5 absorption technique . under optimistic assumptions for the systematic uncertainties of both measurements , we estimate the parameter region of @xmath6 in which the mass hierarchy can be determined . if @xmath7 is relatively large , @xmath8 , and both of @xmath1 and @xmath2 can be measured with the precision of @xmath9 % it is possible to determine the neutrino mass hierarchy at @xmath10 95% cl for @xmath11 for the current best fit values of all the other oscillation parameters .
1.6 cm in spite of the great progress that has been made in recent years there still remains a number of important questions regarding the nature of neutrinos .
one such unknown is whether the neutrinos have a normal or inverted mass hierarchy .
this unresolved ambiguity in the neutrino mass pattern is usually phrased as follows : if the neutrino masses are labeled as @xmath12 ( @xmath13 ) for the mass of the neutrino state with the greatest ( least ) @xmath14 component then for the normal hierarchy @xmath15 and for the inverted hierarchy @xmath16 .
the determination of this pattern must shed light on the secret of how the lepton sector is organized and may testify to the underlying symmetries by which the structure of neutrino masses and lepton flavor mixing are prescribed @xcite .
it is widely recognized that it is difficult to determine the neutrino mass hierarchy by any of the ongoing and the near future neutrino oscillation experiments including the atmospheric neutrino observation by super - kamiokande @xcite , minos @xcite , opera @xcite , and t2k ( tokai - to - kamioka ) @xcite . the most studied method for determining
the mass hierarchy within the scope of the future neutrino experiments is to explore the earth matter effect in long - baseline accelerator experiments @xcite .
since the way the matter effect interferes with vacuum oscillation depends upon the sign of @xmath17 , the degeneracy between the normal and the inverted mass hierarchies can be lifted in matter .
the no@xmath18a @xcite and the t2kk ( tokai - to - kamioka - korea ) @xcite experiments are designed to have a significant matter effect so that the hierarchy may be determined under favorable values of the unknown neutrino parameters . although it is generally accepted that long - baseline experiments are the most promising method for determining the mass hierarchy , alternative ways should be explored especially if they have the most sensitivity for neutrino parameters which are complementary to the favorable parameters for long baseline experiments .
it was proposed in @xcite that accurate measurement of effective atmospheric scale @xmath0 both in electron and muon neutrino disappearance channels would determine the neutrino mass hierarchy , provided that precision of the measurement could reach sub percent level .
we call this the `` @xmath0 disappearance '' method . in this paper , we describe a setting by which such precision measurement could be achieved , and perform a semi - quantitative analysis to estimate the parameter region in which the mass hierarchy can be determined by this method . for the measurement of @xmath0 in @xmath19 disappearance channel , which will be denoted as @xmath2
, we assume a factor of 2 - 4 improvement with respect to what is currently expected for the phase ii of the t2k project . for the measurement of @xmath3
, we assume an experiment based on the resonant absorption of @xmath5 enhanced by the mssbauer effect as recently proposed by raghavan @xcite . after reviewing the `` @xmath0 disappearance '' method in sec .
[ concept ] , we discuss the accuracy of the determination of @xmath2 in sec .
[ mumu ] , and recollect the one for @xmath3 in sec .
[ ee ] . in sec .
[ analysis ] we present our analysis method and results . finally , in sec .
[ conclusion ] we draw our conclusions . |
in social sciences , people in social organizations ( e.g. , a country or a company ) can be categorized into different rankings of socioeconomic tiers based on factors like wealth , income , social status , occupation , power , etc . in this paper
, we will take `` company '' as an example and the internal hierarchical structure of employees in a company can be outlined with _ company organizational chart _ formally .
most company organizational charts are usually tree - structure diagrams with ceo at the root , executive vice presidents ( evps ) at the second level and so forth .
company organizational chart shows the company internal management structure as well as the relationships and relative ranks of employees with different positions / jobs , which is a common visual depiction of how a company is organized . nowadays , to facilitate the collaboration and communication among employees , a new type of online social networks named _ enterprise social networks _
( esns ) has been adopted inside the firewalls of many corporations .
a representative example of online esns is yammer .
over 500,000 businesses around the world are now using yammer , including 85% of the fortune 500 .
yammer provides employees with various enterprise social network services to help them deal with daily work issues and contains abundant heterogeneous information generated by employees online social activities .
* problem studied * : company internal organizational chart is usually confidential to the public for the privacy and security reasons . in this paper , we want to infer the organizational chart of a company based on the heterogeneous information in online esns launched in the company , and the problem is formally named as the _ nference of rganization hart _ ( ioc ) problem . to help illustrate the ioc problem more clearly , we also give an example in figure [ fig : example ] , where the left plot is about an online esn adopted in a company and the right plot shows the company s organizational chart . in the esn , users can have different types of social activities , e.g. , follow other users , join groups , write / reply / like posts , etc .
meanwhile , in the organizational chart , employees are connected by supervision links from managers to subordinates , who are organized into a rooted tree of depth @xmath1 with ceo `` adam '' at the root . in companies , managers
can usually supervise several subordinates simultaneously , while each subordinate only reports to one single manager .
for instance , in figure [ fig : example ] , ceo `` adam '' manages `` bob '' and `` candy '' concurrently , while `` david '' only needs to report to `` bob '' .
the ioc problem is an interesting yet important problem . besides inferring company organizational chart ,
it can also be applied in other real - world concrete applications : ( 1 ) identifying the command structures of terrorist organizations @xcite based on the communication / traffic networks of their members .
the command structures of terrorist organizations are usually pyramid diagrams outlining their support systems consisting of the _ leaders _ , _ operational cadre _ , _ active supporters _ and _ passive supporters _ @xcite .
uncovering their internal operational structure and determining roles of members will be helpful for conducting precise strikes against their key leaders and avoid the tragic events , like 9/11 @xcite .
( 2 ) inferring the social hierarchies of animals based on their observed interaction networks @xcite .
many animals ( like , mammals , birds and insect species ) are usually organized into dominance hierarchies .
identifying and understanding the organizational hierarchies of animals will be helpful to design and carry out effective conservation measures to protect them .
albeit its importance , ioc is a novel problem and we are the first to propose to study it based on online esns .
the ioc problem is totally different from existing works : ( 1 ) `` _ _ hierarchy detection in social networks _ _ '' @xcite , which only studies the division of the regular users of the social networks into different hierarchies , who are not actually involved in any organizations ; ( 2 ) `` _ _ organizational intrusion _ _ '' @xcite , which focuses on attacking organizations and attaining company internal information only ; and ( 3 ) `` _ _ inferring offline hierarchy from social networks _ _ '' @xcite , which merely infers fragments of offline hierarchical ties in _ homogeneous _ networks , instead of reconstructing the whole organizational chart .
different from all these works , in this paper , we aim at recovering the complete organizational chart of a company ( including both the hierarchical tiers of employees and the supervision links from managers to subordinates ) based on the _ heterogeneous _ information about the employees in online esns
. meanwhile , to guarantee the smooth operations of companies , the inferred organizational chart needs to meet certain structural requirements @xcite , including both ( 1 ) macro - level depth requirement , and ( 2 ) micro - level width requirement .
two classical organizational structures adopted by companies are the vertical structure and the horizontal structure @xcite .
vertical organizational structure with well - defined chains of command clearly outlines the responsibilities of each employee but will result in delays in information delivery @xcite .
meanwhile , horizontal organizational structure with flat command system involves everyone in decision making but will lead to difficulties in coordinating the activities of different departments @xcite .
for instance , based on the input social network shown in plot a of figure [ fig : balance_example ] , we give two extreme cases of the vertical and horizontal organizational charts without depth regulation in plots b and c respectively , both of which will lead to serious management problems for large companies involving tens of thousands employees .
proper regulation of the inferred organizational chart s depth ( i.e. , the macro - level depth requirement ) is generally desired . on the other hand ,
most employees in companies need good supervisors to coach and instruct their daily work , but the number of subordinates each manager can supervise is limited , which can be determined by their management capacities , available time and energy . rationally regulating the allocation of supervision workload among managers ( i.e. , the micro - level width requirement ) can improve the management effectiveness significantly . for instance , in plot d of figure [ fig : balance_example ] , we show an inferred organizational chart with depth regulation but no subordinate allocation regulation . in the plot ,
users in esn are stratified into @xmath0 tiers ( which is relatively reasonable compared to the extreme cases in plots b and c ) but the employees management workloads at tier @xmath0 are all assigned to one single manager , which may be beyond his / her management ability . despite its importance and novelty ,
the ioc problem is very hard to solve due to the following challenges : * _ regulated social stratification _ : _ effective social stratification _ to partition users into different hierarchical arrangements ( i.e. , identifying the relative manager - subordinate roles of employees ) while meeting the macro - level depth requirement is the prerequisite for addressing the ioc problem . * _ supervision link inference _ : supervision link is a new type of link merely existing from managers to their subordinates .
predicting the existence of potential _ supervision links _ with the heterogeneous information in esns is still an open problem . *
_ regulated supervision workload allocation _ : to maximize the management effectiveness and efficiency , the number of subordinates each manager can supervise is limited by the management threshold @xmath2 . in other words , supervision links in organizational chart have an inherent _ k - to - one _ constraint . to address all the above challenges ,
a new unsupervised organizational chart inference framework named create ( hr covr ) is proposed in this paper .
several new concepts ( e.g. , `` _ _ class transcendence social links _ _ '' , `` _ _ matthew effect based constraint _ _ '' and `` _ _ chart depth regulation constraint _ _ '' ) will be introduced and create resolves the _ regulated social stratification _
challenge by minimizing the existence of class transcendence social links in esns .
create tackles the _ supervision link inference _ challenge by aggregating multiple social meta paths in the esn between to consecutive social hierarchies .
finally , create handles the _ regulated supervision workload allocation _ challenge by applying _ network flow _ to match consecutive social hierarchies to preserve the _ k - to - one _ constraint on _ supervision links_. the remaining parts of the paper are organized as follows . in section [ sec : formulation ] , we will define some important terminologies and the ioc problem .
method create will be introduced in section [ sec : method ] .
extensive experiment results are available in section [ sec : experiment ] .
finally , we describe the related works in section [ sec : relatedwork ] and conclude this paper in section [ sec : conclusion ] .
in this section , we will introduce the formal definitions of `` _ _ heterogeneous social network _ _ '' and `` _ _ organizational chart _ _ '' at first and then define the ioc problem with these two concepts .
* definition 1 * ( heterogeneous social networks ) : a _ heterogeneous social network _ can be represented as @xmath3 , where @xmath4 and @xmath5 are the sets of different types of nodes and complex links among these nodes in the network respectively . as introduced in section [ sec : introduction ] , users in online esns ( e.g. , yammer ) have various types of social activities , e.g. , follow other users , join groups , write / reply / like online posts , etc . as a result , yammer can be represented as a _ heterogeneous social network _
@xmath3 , where @xmath6 is the set of user , group and post nodes in @xmath7 and @xmath8 denotes the set of social links among users , join links between users and groups , as well as write , reply and like links between users and posts respectively .
* definition 2 * ( organizational chart ) : the _ organization chart _ of a company can be represented as a _ rooted tree _
@xcite @xmath9 , where @xmath10 is the set of _ employees _ , @xmath11 denotes the set of directed _ supervision links _ from managers to subordinates in @xmath12 and _ root _ represents the ceo in the company .
based on the definitions of _ heterogeneous social network _ and _ organizational chart _ , we can define the ioc problem formally as follows : * definition 3 * ( organizational chart inference ( ioc ) ) : given an online esn @xmath3 launched in a company , the ioc problem aims at inferring the most likely organizational chart @xmath9 of the company , where @xmath13 ( @xmath14 is the user set in network @xmath7 ) .
furthermore , considering that the node set as well as the root node in @xmath12 are fixed , the ioc problem actually aims at inferring the @xmath15 most likely _ supervision links _
@xmath11 among employees . the inferred supervision links together with the node set @xmath10 as well as the @xmath16 node can recover the original organizational chart @xmath9 of the company
considering that supervision links exist merely between managers and subordinates , we propose to stratify users in enterprise social networks into different _ social classes _ to identify their relative manager - subordinate roles in subsection [ subsec : stratification ] .
macro - level depth requirement of the inferred chart is achieved with the _ depth regulation constraint _ in the social stratification objective function .
potential supervision links can be inferred between employees in consecutive classes by aggregating social meta paths among employees in the esn in subsection [ subsec : metapath ] . to preserve the _ k - to - one _ constraint on supervision links ( i.e. , the micro - level width requirement ) , redundant non - existing supervision links will be pruned in subsection [ subsec : networkflow ] .
generally , as shown in figure [ fig : framework ] , framework create has three steps : ( 1 ) regulated social stratification , ( 2 ) supervision link inference , and ( 3 ) regulated social class matching , which will all be introduced in this section .
supervision links merely exist between managers and subordinates .
division of users into hierarchies to identify their relative manager - subordinate roles can shrink the supervision link inference space greatly .
the process of hierarchizing users in online esn is called _ social stratification _ formally .
* definition 4 * ( social stratification ) : traditional _ social stratification _ concept used in social science denotes the ranking and partition of people into different hierarchies based on various factors , e.g. , power , wealth , knowledge and importance @xcite . in this paper , we define _ social stratification _ as the partition process of users in online esns into different hierarchies according to their management relationships , where managers are at upper levels , while subordinates are at lower levels .
the relative stratified levels of users in online esn are defined as their social classes .
* definition 5 * ( social class ) : _ social class _ is a term used by social stratification models in social science and the most common ones are the upper , middle , and lower classes @xcite . in this paper , we define _ social class _ of users in online esns as their management level in the company , where ceo belongs to social class 1 , evps belong to class 2 , and so forth . in social stratification
, users in online esns will be mapped to their _ social classes _ according to mapping : @xmath17 . for each user @xmath18 ,
his social class @xmath19 is defined recursively as follows : @xmath20 where @xmath21 represents the direct manager of @xmath22 . in social science ,
the working class are eager to get acquainted with and join the upper echelons of their class by either accumulating wealth @xcite , imitating their dressing styles @xcite , and mimicking their dialect and accents @xcite . meanwhile , the upper class are very cohesive and they tend to be friends who share similar background @xcite . so is the case for the social links in _ enterprise social networks_. by analyzing the yammer network data , we observe that the probability for users to follow upper - level managers is @xmath23 on average , while that of following subordinates is merely @xmath24 . as a result , in online esns
, subordinates tend to follow their managers , while people in management are reluctant to initiate the friendship with their subordinates @xcite .
based on such an observation , we introduce the concept of _ class transcendence _ social links and propose to stratify users by minimizing the existence of such links in esns .
* definition 6 * ( class transcendence social link ) : link @xmath25 ( i.e. , @xmath22 follows @xmath26 ) is defined as a _ class transcendence social link _ in online esn @xmath7 iff @xmath27 and @xmath28 ( where smaller social class denotes upper management level in the organizational chart ) . in social stratification , each introduced class transcendence social link in the result will lead to a _ class transcendence penalty_. let @xmath29 be the _ social stratification _
result of all users in the esn .
for any directed social link @xmath27 in the esn , the _ class transcendence penalty _ introduced by it can be represented as @xmath30 the _ class transcendence penalty _ introduced by all social links ( i.e. , @xmath31 ) in the esn can be represented as @xmath32 the rich get richer " ( i.e. , the matthew effect @xcite ) is a common phenomenon in social science literally referring to issues of fame or status as well as cumulative advantage of economic capital . by analyzing the yammer network data
, we have similar observations : `` people at higher management level can accumulate more followers easily '' .
such an observation provides important hints for inferring users relative management levels according to their in degrees in esn ( i.e. , the number of followers ) . * definition 7 * ( matthew effect based constraint ) : for any two given users @xmath22 and @xmath26 in the network , let @xmath33 and @xmath34 be the follower sets of @xmath22 and @xmath26 in the network respectively .
the _ matthew effect based constraint _ on users @xmath22 and @xmath26 can be represented as @xmath35 if @xmath36 . furthermore , to maximize the operation efficiency of companies , the inferred organizational chart needs to meet the _ macro - level depth requirement _ , which can be achieved with the following chart depth regulation constraint .
* definition 8 * ( chart depth regulation constraint ) : the _ chart depth regulation constraint
_ avoids obtaining organizational chart with too short command chains ( e.g. , the extreme horizontal structure ) and can be represented as @xmath37 where parameter @xmath38 is used to regulate the depth of the chart , whose sensitivity analysis will be given in section [ sec : experiment ] .
furthermore , term @xmath39 is also added to the minimization objective function to avoid obtaining charts with too long command chains ( i.e. , the extreme vertical structure ) .
based on all the above remarks , the _ optimal _ regulated social stratification @xmath40 of users in esn can be obtained by solving the following objective function : @xmath41 the integer programming objective function can be solved with open source toolkits , e.g. , glpk , pulp , etc .
, very easily and the obtained results of variables @xmath42 represent the inferred social classes of users in online esn .
it is a challenge to estimate the supervision relations between the esn members in consecutive social classes .
here we use the meta paths concept introduced in @xcite to identify and evaluate different types relationship in esn . * social meta paths in enterprise social networks * *
_ follow _ : user @xmath43 user , whose notation is `` @xmath44 '' or @xmath45 . *
follower of follower _ : user @xmath43 user @xmath43 user , whose notation is `` @xmath46 '' or @xmath47 . *
_ common followee _
: user @xmath43 user @xmath48 user , whose notation is `` @xmath49 '' or @xmath50 . * _ common follower _ : user @xmath48 user @xmath43 user , whose notation is `` @xmath51 '' or @xmath52 . * _ common group membership _ : user @xmath53 group @xmath54 user , whose notation is `` @xmath55 '' or @xmath56 . * _ reply post _ : user @xmath57 post @xmath58 post @xmath59 user , whose notation is `` @xmath60 '' or @xmath61 .
* _ like post _ : user @xmath57 post @xmath62 user , whose notation is `` @xmath60 '' or @xmath63 .
an existing user intimacy @xcite measure , path - sim , based on meta paths was introduced in @xcite , which can calculate the propagation probability between users via meta paths in _ undirected homogeneous networks_. to deal with _ directed heterogeneous networks _ , we extend it and introduce a new intimacy measure , dp - intimacy ( directed path - intimacy ) , based on social meta path @xmath64 : @xmath65 where @xmath66 denotes the instance set of meta path @xmath67 going from @xmath22 to @xmath26 in the esn .
different social meta paths capture the intimacy between users in different aspects and overall intimacy between users can be obtained by aggregating information from all these social meta paths .
let @xmath68 be the intimacy scores between users @xmath22 and @xmath26 calculated based on social meta paths @xmath69 respectively .
without loss of generality , we choose logistic function as the intimacy aggregation function @xcite , the overall intimacy between users @xmath22 and @xmath26 can be represented as @xmath70,\ ] ] where the value of @xmath71 denotes the weight of social meta path @xmath72 and @xmath73 .
meta path aggregation based supervision link inference method proposed in previous step calculate the intimacy scores of all the potential links between pairs of social classes .
however , to regulate the supervision workload allocation , the number of subordinates each manager can supervise is limited by the management threshold @xmath2 . in this section , we will prune the redundant non - existing supervision links with network - flow based regulated social class matching to preserve the _ k - to - one _ constraint on _ supervision links_. based on the _ social stratification _
results @xmath74 , we can stratify all the users into _ social classes _
@xmath75 $ ] and users in class @xmath76 $ ] can be represented as set @xmath77 . by aggregating information in various _ social meta paths _ , we can calculate the intimacy scores of all potential _ supervision links _ between consecutive _ social classes _ , which exist between @xmath78 and @xmath79 can be represented as set @xmath80 .
links in @xmath81 are associated with certain weights ( i.e. , the calculated intimacy scores ) , which can be obtained with the mapping @xmath82 .
users in social classes @xmath83 and @xmath84 ( i.e. , @xmath78 and @xmath79 ) together with all the potential supervision links between them ( i.e. , @xmath81 ) and their intimacy scores ( i.e. , @xmath85 ) can form a weighted bipartite preference graph . *
definition 9 * ( weighted bipartite preference graph ) : the _ weighted bipartite preference graph _ between users in @xmath78 and @xmath79 can be represented as @xmath86 .
an example of _ weighted bipartite preference graph _ is shown in the upper plot of figure [ fig : network_flow ] . in the example , all the potential supervision links between the upper - level and lower - level individuals are represented as the directed purple lines between them , whose weights are the numbers marked on the lines .
each employee in the figure is associated with multiple potential _
supervision links _ and the redundant ones can be pruned with the _ network flow _
method introduced in the next subsection . based on the _ bipartite preference graph _ @xmath87
, we propose to construct the following _ network flow graph _ first . * definition 10 * ( network flow graph ) : based on bipartite preference graph @xmath86 , the _ network flow graph _ can be represented as @xmath88 .
node set @xmath89 includes all nodes in @xmath87 and two dummy nodes : source node @xmath90 and sink node @xmath91 ( i.e. , @xmath92 ) . besides all the links in @xmath87 ,
we further add directed links from @xmath90 to all nodes in @xmath78 , as well as those from all nodes in @xmath79 to @xmath91 ( i.e. , @xmath93 ) . only the links in @xmath81 are associated with weights , which can be obtained with mapping @xmath94 .
for instance , based on the _ bipartite preference graph _ in the upper plot of figure [ fig : network_flow ] , we can construct its corresponding _ network flow graph _
( i.e. , the lower plot ) .
all the links in the network flow graph are directed denoting the flow direction .
* bound constraint of network flow * for each link @xmath95 , we allow a certain amount of flow going through within range @xmath96 $ ] , where @xmath97 and @xmath98 represent the _ lower bound _ and _ upper bound _ associated with link @xmath25 respectively and @xmath99 where @xmath100 is the flow amount going through link @xmath25 .
more specifically , for links from @xmath90 to the upper level individuals , i.e. , @xmath101 , we set its _ lower bound _ and _ upper bound _ to be @xmath102 and @xmath103 respectively and we can get @xmath104 where @xmath2 is the management threshold , whose sensitivity analysis is available in section [ sec : experiment ] .
it is actually the constrain to preserve micro - level width requirement . for link
@xmath105 , we set its _ lower bound _ and _ upper bound _ to be @xmath106 and @xmath107 , i.e. , @xmath108 which means exact amount @xmath109 flow goes through link @xmath110 ( i.e. , each subordinate needs to have exactly one manager ) .
links @xmath111 have lower and upper bounds @xmath112 , @xmath113 and the flow amount needs to be an integer , i.e. , @xmath114 denoting whether supervision links in @xmath81 are selected or not in the matching result . * mass - balance constraint of network flow * in network flow model , for each node , e.g. , @xmath22 , the amount of flow going into @xmath22 should be equal to that going out from @xmath22 , i.e. , @xmath115 * minimum cost network flow * all links going from @xmath78 to @xmath79 are associated with corresponding _ flow costs _ , which are negatively correlated to their intimacy scores .
for instance , in this paper , for link @xmath116 with weight @xmath117 , we can represent their _ flow cost _ as @xmath118 .
the optimal _ network flow _ with the _ minimum cost _ ( i.e. , the maximum intimacy ) can be obtained by addressing the following _ integer programming _ problem : @xmath119 similarly , the above integer programming problem can be addressed with open source toolkits and how to solve the equation will not be introduced here .
variables obtained by solving the above equation can lead to the minimum cost but can also meet the constraints as well .
these obtained variables denote the existence scores of the corresponding supervision links , where the selected links ( i.e. , those corresponding variable @xmath120 ) will be assigned with label @xmath121 while the rest are assigned with label @xmath122 .
to examine the performance of create in addressing the ioc problem , in this part , extensive experiments will be conducted on real - world _ enterprise social network _ : yammer .
we crawl all the microsoft employees information from yammer and obtain the complete organizational chart involving all these employees in microsoft during june , 2014 .
the social network data covers all the user - generated content ( such as posts , replies , topics , etc . ) and social graphs ( such as user - user following links , user - group memberships , user - topic following links , etc . ) by then that are set to be public . in summary , it includes more than @xmath123 microsoft employees , and millions of user - generated posts published and the social links .
all the users in yammer are registered with the official employment i d in microsoft , via which we can identify them in the organizational chart correspondingly . from microsoft ,
the complete organization structure of all employees is obtained . as introduced
before , the structure of the organizational chart is a rooted tree with the ceo at the top .
the create framework proposed in this paper is an unsupervised model , and the _ organizational chart _
is used for evaluation only in the experiments . to ensure the employee node set of organizational chart to be identical to that of esn ,
a fully aligned @xcite yammer network and organizational chart are sampled from the dataset .
initially , with the directed follow links among users in yammer , we achieve the regulated social stratification of users by minimizing the number _ class transcendence social links_. all the potential supervision links between pairwise consecutive social classes in the social stratification result are inferred by aggregating information from various social meta paths in yammer , whose existence likelihood is denoted as the intimacy score . for simplicity ,
the weights of different social meta paths in logistic function are assigned with identical values , i.e. , @xmath124 $ ] .
a subset of these inferred supervision links will be selected via the regulated social class matching based on the network - flow model to preserve the _ k - to - one _ constraint on supervision links .
meanwhile , to demonstrate the effectiveness of create , we compare create with many baseline methods , including both state - of - art and traditional methods in social stratification and organizational chart inference . * social stratification methods * : * _ regulated social stratification _ :
regulated social stratification is the first step of create proposed in this paper , which is also named as create for simplicity .
create exploits the concept of class transcendence social links and matthew effect based constraint to stratify users in esns into different social classes .
in addition , to regulate the depth of inferred social classes about employees , create further adds a _ chart depth regulation constraint _ into the objective function . * _ agony based social division _ : asd is a state - of - art social division method proposed in @xcite , which detects the social hierarchies of regular users in general online social networks .
asd is not designed for organizational chart inference and does nt consider the matthew effect based constraint nor the chart depth regulation constraint .
* organizational chart inference methods * : * _ social stratification + link prediction + matching _ ( create ) : create is the framework proposed in this paper and it has three steps : ( 1 ) regulated social stratification , ( 2 ) link inference and ( 3 ) regulated social class matching . *
_ social stratification + link prediction _
( create - sl ) : create - sl contains two steps : ( 1 ) _ social stratification _ , and ( 2 ) _ link prediction _ based on accumulated social meta paths . create - sl has no _ matching step _ to keep the micro - level width requirement and the outputs can not meet the _ k - to - one _ constraint . * _ social stratification + matching _
( create - sm ) : create - sm contains two steps : ( 1 ) _ social stratification _ , and ( 2 ) social class matching .
create - sm has no supervision link prediction step and social links from upper - level social class to the lower - level are regarded as the potential supervision links candidates . *
_ social stratification _
( create - s ) : create - s is identical to create - sl except that it has no matching step and outputs the all the _ social links _ between sequential hierarchies as the supervision links . *
_ traditional unsupervised link prediction methods _
: no existing supervised link prediction models can be applied as no labeled supervision link exist . for completeness
, we further compare _ our _ with traditional unsupervised baseline methods which include _ common neighbor ( cn ) _
@xcite , _ jaccard s coefficient ( jc ) _
@xcite and _ adamic adar ( aa ) _ @xcite between consecutive stratified social classes .
* social stratification evaluation metrics * in the _ social stratification _
step , the outputs are the inferred _ social classes _ of all the employees . by comparing them with individuals real - world _ social classes _
( i.e. , the ground truth ) , we can calculate the _
mean absolute error _
@xcite , _ mean squared error _
@xcite and _ coefficient of determination _
( i.e. , @xmath125 ) @xcite of the results .
in addition , the ratio of correctly stratified users ( i.e. , accuracy ) can also be used to measure the performance .
so , the metrics used to evaluate the performance of different social stratification methods include _ mean absolute error _ ( mae ) @xcite , _ mean squared error _ ( mse ) @xcite , @xmath125 @xcite and _ accuracy_. * organizational chart inference evaluation metrics * methods create - sl , create - l and the traditional unsupervised link prediction can only output the confidence scores of all potential supervision links without labels , whose performance will be evaluated by metrics @xmath126 and precision@100 . meanwhile , create and create - sm can output both labels and scores of potential supervision links and , besides @xmath126 and precision@100 , we will also evaluate their performance with precision , recall and f1-score . in social stratification , parameter @xmath38 is applied to maintain the macro - level depth requirement , which can control the depth of the organizational chart . before comparing the performance of create with asd
, we will analyze the sensitivity of parameter @xmath38 at first .
we select @xmath38 with values in @xmath127 and obtain the accuracy scores achieved by create as shown in figure [ fig : alpha ] .
.,scaledwidth=100.0% ] when parameter @xmath38 is very small ( e.g. , from @xmath109 to @xmath0 ) , we observe that it has no effects on the performance of create .
the possible reason can be that _
matthew effect based constraint _ can already effectively outline the relative hierarchical relationships among users in online esns , the average social class of users obtained based on which is already greater than @xmath0 .
when @xmath38 becomes larger ( from @xmath128 to @xmath129 ) , the _ structure regulation constraint _ starts to matter more and the social stratification accuracy goes up steadily and achieve the highest value at @xmath130 , i.e. , the default value of @xmath38 in later experiments .
create performs better as @xmath38 increases shows that the _ structure regulation constraint _ can stretch the organizational structure and stratify users in their correct social classes .
however , as @xmath38 further increases ( i.e. , from @xmath129 to @xmath131 ) , the accuracy achieved by create decreases dramatically .
the reason can be that larger @xmath38 stretches the organizational structure too much and put lots users into the wrong social classes .
for example , it is nearly impossible for users to achieve @xmath131 as the average social class , which is actually the largest social class in the sampled fully aligned organizational chart .
social stratification results of create and asd are given in figures [ eg_fig9_comp_level]-[eg_fig_comp1 ] , where figure [ eg_fig9_comp_level ] shows the results achieved by create and asd at each social class ( evaluated by precision and recall respectively ) and figure [ eg_fig_comp1 ] shows their overall performance ( evaluated by accuracy , mae , mse and @xmath125 ) .
from the microscopic perspective , we observes that create performs better than asd consistently at all social classes .
create achieves both @xmath132 precision and @xmath132 recall at social classes @xmath133 , i.e. , create identifies the top @xmath1 management levels of the company correctly .
the performance of create at other social classes is also very promising .
for instance , the precision scores achieved by create at social classes @xmath134 ( besides @xmath133 ) are either @xmath132 or close to @xmath132 and the recall scores of create at social classes @xmath135 are also very high , which all outperform those of asd with significant advantages . from a macroscopic perspective , the performance of create in stratifying the whole user set in esn is very excellent and much better than that of asd . the accuracy , mae , mse and @xmath125 scores achieved by create are @xmath136 , @xmath137 , @xmath138 and @xmath139 respectively , which all outperforms those achieved by asd .
for example , the accuracy achieved create is almost the triple of that obtained by asd , while the mae and mse obtained by create are merely the @xmath140 and @xmath141 of those achieved by asd . in addition
, asd gets negative @xmath125 scores in identifying social classes of users in esns , which denotes that the identified users social classes are massively disordered and have no linear correlation with the social class ground truth at all .
create has proved its excellent effectiveness in stratifying users in esns , based on which , we further study its performance in inferring the potential supervision links between pairs of consecutive social classes , whose results are evaluated by auc , precision@100 in table [ tab : result1 ] as well as by recall , precision and f1 in figure [ eg_fig_comp2 ] . in table [ tab : result1 ] , we compare create ( of different management thresholds @xmath2 ) with all the other baseline methods , where create ( with parameter @xmath142 and @xmath143 ) performs the best .
compared with create - sl ( or create - s ) , create ( or create - sm ) which has the matching step can identify supervision links more effectively .
for instance , create ( with @xmath144 ) outperforms create - sl by over @xmath145 in auc and @xmath146 in precision@100 , and create - sm ( with parameter @xmath144 ) outperforms create - s with remarkable advantages .
it demonstrates that matching step can effectively prune non - existing supervision links and preserve the micro - level width requirement ( i.e. , the _ k - to - one _ constraint ) .
compared with create - sm and create - s , create which infers the potential supervision links based on heterogeneous information in esns instead of merely regarding the social links as supervision link candidates achieves much better results .
for example , in table [ tab : result1 ] , the auc of create is @xmath147 higher than that of create - sm and create - s , while the precision@100 of create is also roughly @xmath148 higher as well .
in addition , in figures [ eg_fig_comp2 ] , the recall , precision and f1 obtained by create is almost triple of those achieved by create - sm .
it confirms the argument that heterogeneous information in esns can capture the relationships among colleagues ( especially between managers and subordinates ) .
in addition , we also compare create with traditional unsupervised link prediction methods , including cn , jc and aa , and the advantages of create are very obvious according to table [ tab : result1 ] : create can outperform all these unsupervised link prediction methods with significant advantages .
in social class matching , the management threshold parameter @xmath2 plays a key role in constraining the number of supervision links connected to each managers .
the sensitivity of parameter @xmath2 will be analyzed in this section , where the results achieved by create ( with different @xmath2s ) evaluated by different metrics are available in figure [ eg_fig_k ] .
small management threshold @xmath2 ( e.g. , @xmath149 ) limits each manager s subordinate number to @xmath149 and will preserve the supervision links with extremely high likelihood only but may miss many promising ones .
however , as threshold @xmath2 increases , more links with high likelihood will be preserved and the metric scores increase consistently . meanwhile , when the threshold @xmath2 goes to @xmath150 , the performance of create degrades dramatically .
the possible reason can be that , with larger threshold , each manager can have too many supervision links , which may exceed the subordinates they have in the real - world .
enterprise social networks are important sources for employees in companies to get reliable information .
ehrlich et al .
@xcite propose to search for experts in enterprise with both text and social network analysis techniques .
they propose to examine the users dynamic profile information and get the social distance to the expert before deciding how to initiate the contact .
enterprise social networks can lead to lots of benefits to companies and the motivations of enterprise social network adoption in companies are studied in details in @xcite .
users in enterprise social networks will connect and learn from each other through personal and professional sharing .
people sensemaking and relation building on an enterprise social network site is studied in by dimicco et al .
in addition , social connections among users in enterprise social networks usually have multiple facets .
@xcite propose to study the study the multiplexity of social connections among users in enterprise social networks , which include both professional and personal closeness . from social networks ,
some works have been done to infer the hierarchies of individuals @xcite .
a measure , agony , is proposed in @xcite , by minimizing which the authors propose a hierarchy detection method .
a random graph model and markov chain monte carlo sampling is proposed by clauset et al . in @xcite , which can address the problem of structural inference of hierarchies in networks .
maiya et al .
propose to identify the hierarchies in social networks to achieve the maximum likelihood in @xcite .
all the above three papers focus on dividing individuals into different hierarchies only .
a offline hierarchical ties inference method has been proposed by jaber et al . in @xcite to discover offline links among people based on a time - based model . however
, none of these papers can recover the whole organizational chart .
cross - social - network studies has become a hot research topic in recent years .
kong et al .
@xcite are the first to propose the concepts of `` anchor links '' , `` anchor users '' , `` aligned networks '' etc . a novel network anchoring method is proposed in @xcite to address the network alignment problem .
cross - network heterogeneous link prediction problems are studied by zhang et al .
@xcite by transferring links across partially aligned networks .
besides link prediction problems , jin et al
. proposes to partition multiple large - scale social networks simultaneously in @xcite and zhang et al .
study the community detection problem across partially aligned networks in @xcite .
zhan et al .
analyze the information diffusion process across aligned networks @xcite .
in this paper , we have studied the organizational chart inference ( ioc ) problem based on the heterogeneous online esns . to address the ioc problem ,
a new chart inference framework create has been proposed in section [ sec : method ] .
create consists of @xmath0 steps : ( 1 ) regulated social stratification , ( 2 ) supervision link inference with social meta paths aggregation , and ( 3 ) regulated social class matching .
experiments on real - world esn and organizational chart dataset have demonstrated the effectiveness of create . [
sec : ack ]
this work is supported in part by nsf through grants cns-1115234 , google research award , the pinnacle lab at singapore management university , and huawei grants .
organisation - organizational structure - organisational chart .
http://kalyan-city.blogspot.com/2010/06/organisation-organizational-structure.html [ http://kalyan-city.blogspot.com/2010/06/organisation-organizational-structure.html ] . | nowadays , to facilitate the communication and cooperation among employees , a new family of online social networks has been adopted in many companies , which are called the `` enterprise social networks '' ( esns ) .
esns can provide employees with various professional services to help them deal with daily work issues .
meanwhile , employees in companies are usually organized into different hierarchies according to the relative ranks of their positions .
the company internal management structure can be outlined with the organizational chart visually , which is normally confidential to the public out of the privacy and security concerns . in this paper
, we want to study the ioc ( nference of rganizational hart ) problem to identify company internal organizational chart based on the heterogeneous online esn launched in it .
ioc is very challenging to address as , to guarantee smooth operations , the internal organizational charts of companies need to meet certain structural requirements ( about its depth and width ) . to solve the ioc problem , a novel unsupervised method create ( hr covr )
is proposed in this paper , which consists of @xmath0 steps : ( 1 ) social stratification of esn users into different social classes , ( 2 ) supervision link inference from managers to subordinates , and ( 3 ) consecutive social classes matching to prune the redundant supervision links .
extensive experiments conducted on real - world online esn dataset demonstrate that create can perform very well in addressing the ioc problem . |
measurement of the @xmath0 spectrum gives an excellent test for studying the dynamical content of the configurations being used .
nrqcd can be used to give precise results which indicate how the light dynamical quarks feed in to the heavy meson properties .
the program at cleo - c will enable lattice results on @xmath0 spectroscopy to be experimentally verified @xcite . here
we present new results using dynamical configurations from the milc collaboration @xcite .
we have used the coarser set of available ensembles .
these were generated using the improved staggered asqtad action and have the advantage over previous configurations of the incorporation of 2 + 1 flavors of dynamical quarks .
these are therefore the most realistic simulations of the physical vacuum to date .
several 2 + 1 flavor ensembles are available with one dynamical quark being held at approximately the value of the strange quark mass @xmath1 and two degenerate flavors varying from @xmath1 to @xmath2 . also available
are a quenched ensemble and a two flavor ensemble with degenerate dynamical quark masses of @xmath3 .
we have used both the most chiral 2 + 1 flavor ensemble and the degenerate 3 flavor ensemble plus the quenched and 2 flavor ensembles to best demonstrate the effects of unquenching on the bottomonium spectrum .
we use the standard @xmath4 nrqcd lagrangian @xcite @xmath5 but with corrections to the @xmath6 and @xmath7 fields for @xmath8 errors .
all terms are tadpole improved using an estimate of the landau gauge link . .
open circles and fancy diamonds represent scales determined from the 1p-1s @xmath0 and 2s-1s @xmath0 splittings respectively . ]
for all ensembles , 210 configurations were used . for all but the degenerate 3 flavor ensemble , 4 different quark ( antiquark ) origins were used on each of 4 different timeslices . for the degenerate 3 flavor ensemble ,
4 different origins were used on each of 2 different timeslices . in fitting the meson correlators , the bayes technique @xcite was used .
@xmath9 matrix fits were done using local , ground and excited state smearings for the s states and @xmath10 matrix fits were done using ground and excited state smearings for p states .
the bayes parameters used were quite relaxed with @xmath11 , each of the amplitudes = 0.1(7 ) , and splittings between excited states @xmath12 .
in figure [ fig : ainv ] results are plotted for the ratio of @xmath13 obtained from both the 1p-1s and 2s-1s splittings in the @xmath0 system to that obtained from using @xmath14 , assuming that @xmath15 .
the results from the dynamical configurations are plotted against the ratio of the pseudoscalar to vector light - light meson mass where the valence quark mass is equal to the @xmath16 sea quark mass .
we will also be able to extract a reliable scale from the quantity @xmath17 once @xmath1 is well determined @xcite .
dependence on the quantity used to obtain @xmath13 with quenching gives ambiguities in calculating other physical observables .
the plot clearly illustrates these ambiguities disappearing as @xmath18 tends to its physical value . for the most chiral 2 + 1 flavor dynamical ensemble
the scales from @xmath0 have converged .
the @xmath0 results imply that @xmath19 .
these scales can be used to determine a value for @xmath20 @xcite .
figure [ fig : leprat ] shows the ratio of the leptonic width , i.e. the width of decay to @xmath21 , of the 1st radially excited @xmath0 to that of the ground state @xmath0 .
the experimental points for @xmath22 and @xmath0 are plotted to demonstrate that this ratio is to a large extent independent of the heavy quark mass . at leading order @xmath23 , where the ` wavefunction at the origin ' @xmath24 , the overlap of the local vector current with a given state .
therefore we can calculate this ratio using appropriate amplitudes from our @xmath9 fits .
the @xmath25 factor for this current , which must be calculated in perturbation theory , cancels in the ratio .
it can be seen that although even the most chiral result is still some way from experiment , the ratio is clearly falling with dynamical quark mass .
next , the currents must be improved by adding current correction operators as has been done for @xmath26 @xcite .
the hyperfine splitting between the @xmath0 and @xmath27 is a spin dependent quantity which is proportional to the square of the coefficient of the @xmath28 term in the action and also , in leading order in a potential model , to @xmath29 . in figure
[ fig : hyprat ] we show the ratio of hyperfine splitting of the 1st excited to that of the ground @xmath0 state . the uncertainty from the coefficient of the @xmath28 term , and , similarly to the leptonic width
, a lot of the heavy quark mass dependence cancels in the ratio .
the @xmath27 and @xmath30 are experimentally unseen as yet so the experimental result given is for the @xmath22 system from the belle collaboration @xcite .
the plot shows a tendency towards experiment with unquenching .
to that of the ground state .
the open circles represent results for dynamical configurations while the cross represents the quenched result .
the diamond and burst represent the current experimental results for @xmath0 and @xmath22 respectively . ] to that of the ground state .
the open circle represents the results for the most chiral dynamical configuration while the cross represents the quenched result .
the diamond represents the current experimental result for the @xmath22 system . ]
the dependence of the scale on the quantity used to obtain it has been a long standing problem of the quenched approximation . full convergence of @xmath13 obtained from different splittings in the @xmath0 spectrum is now clearly seen on the most chiral 2 + 1 flavor milc ensemble . we conclude that this ensemble has good dynamical content
. ratios of leptonic widths and hyperfine splitting have been measured and are seen to tend toward experiment wih unquenching and decreasing dynamical quark mass .
we are grateful to milc for use of their configurations .
this work was supported by nsf , doe , pparc , the eu , and nserc canada . c. w. bernard _
et al _ , phys .
d * 64 * ( 2001 ) 054506 . c. t. h. davies _
et al _ , phys .
d * 50 * ( 1994 ) 6963 .
g. p. lepage _ et al _ , nucl .
* 106 * ( 2002 ) 12 .
m. wingate _
et al _ , j. hein _ et al _ , these proceedings .
c. t. h. davies _ et al _ , these proceedings . c. j. morningstar and j. shigemitsu , phys .
d * 59 * ( 1999 ) 094504 .
belle collaboration , presented at fpcp conference , philadelphia , pa , may 16 - 18 , 2002 . | we describe the bottomonium spectrum obtained on the milc configurations which incorporate 2 + 1 flavors of dynamical quarks .
we compare to quenched and 2 flavor results also on milc configurations .
we show that the lattice spacing determination using different quantities shows clear signs of convergence with 2 + 1 flavors and give results for the leptonic width and hyperfine splitting , in the form of the ratio of the 1st excited state of the @xmath0 to that of the ground state . |
for any two non - zero integers @xmath0 and @xmath1 the baumslag
solitar group @xmath2 is given by the presentation @xmath3 .
these groups were introduced by baumslag and solitar in @xcite , who identified @xmath4 as the first example of a two - generator one - relator non - hopfian group and thus answered a question by b. h. neumann , see @xcite . later on , it was shown that @xmath2 is hopfian if and only if @xmath5 or @xmath6 or @xmath7 , where @xmath8 denotes the set of prime divisors of @xmath9 , see @xcite and @xcite .
after reviewing some fundamental properties of baumslag
solitar groups , we shall consider random walks on these groups .
such a random walk is constructed as follows .
first , we choose a probability measure @xmath10 on @xmath2 such that the support of @xmath10 generates @xmath2 as a semigroup .
then , the random walk starts at the identity element and proceeds with independent @xmath10-distributed increments each of which is multiplied from the right to the current state .
the poisson furstenberg boundary was introduced by furstenberg in @xcite and @xcite .
it is a probabilistic model for the long - time behaviour of the random walk , and simultaneously provides a way to represent all bounded harmonic functions on the state space . in the early 1990s ,
kaimanovich considered random walks on @xmath11 . under the assumption of finite first moment
, he identified their poisson furstenberg boundary geometrically , see ( * ? ? ?
* theorem 5.1 ) .
in particular , he showed that the latter is trivial if the random walk has no vertical drift . for random walks on non - amenable groups
the situation is different because their poisson furstenberg boundary can never be trivial .
this motivates the present paper , in which we study random walks on non - amenable baumslag
solitar groups and their poisson furstenberg boundary .
the paper is organised as follows . in section [ sec : bsgroups ] , we discuss some algebraic and geometric properties of baumslag
solitar groups @xmath2 with @xmath12 .
we explain how these groups can be understood through their natural projections to the bass serre tree @xmath13 and the hyperbolic plane @xmath14 .
afterwards , we recall the construction of the space of ends @xmath15 and the boundary @xmath16 . they shall later be used to associate a geometric boundary to @xmath2 . in sections
[ sec : randomwalks ] and [ sec : pfboundary ] , we discuss random walks on countable groups and define the notion of poisson furstenberg boundary .
we outline a few classical results and state kaimanovich s strip criterion , which is an important tool to identify the poisson furstenberg boundary geometrically . in section [ sec : identification ] , we consider random walks on @xmath2 with @xmath12 . in order to ensure that the natural projections of the random walk to @xmath14 and @xmath13 converge almost surely to random elements in @xmath16 and @xmath15 respectively , we need to make suitable assumptions on the moments . if the random walk has vertical drift , we need to assume finite first moment .
otherwise , the situation is much more subtle , and we need to assume finite ( @xmath17)-th moment .
the fact that the projections converge almost surely allows us to endow the cartesian product @xmath18 , or occasionally just its second component @xmath15 , with the borel @xmath19-algebra and a hitting measure .
finally , kaimanovich s strip criterion shows that the resulting probability space is isomorphic to the poisson furstenberg boundary .
for the part of the paper up to lemma [ lem : zerodrifttree ] , we will assume that the two non - zero integers @xmath0 and @xmath1 satisfy @xmath12 .
in lemma [ lem : zerodrifttree ] and in the subsequent results , we have to restrict ourselves to the non - amenable subcase @xmath20 .
the main result is theorem [ thm : identification ] on page . in the
, we will explain how to obtain similar results for the remaining non - amenable cases .
we would like to thank wolfgang woess .
he suggested this problem and supported us with references and ideas while the research was carried out .
moreover , we are grateful to vadim kaimanovich , who made a couple of comments that led to a substantial improvement of the results .
the structure of baumslag solitar groups can be studied by means of hnn extensions .
indeed , @xmath2 is precisely the hnn extension @xmath21 with isomorphism @xmath22 given by @xmath23 .
this fact allows us to use the respective machinery , such as britton s lemma , see @xcite , which implies that a freely reduced non - empty word @xmath24 over the letters @xmath25 and @xmath26 and their formal inverses can only represent the identity element @xmath27 if it contains @xmath28 with @xmath29 or @xmath30 with @xmath31 as a subword .
now , one can easily conclude that , if neither @xmath5 nor @xmath6 , the elements @xmath32 and @xmath33 generate a non - abelian free subgroup .
so , @xmath2 is non - amenable . on the other hand ,
if @xmath5 or @xmath6 , a simple calculation shows that the normal subgroup @xmath34 is abelian with quotient isomorphic to @xmath35 . in this case , @xmath2 is solvable and therefore amenable . as we will discuss in section [ sub : cratagi ] ,
the distinction between the two cases is of importance when working with random walks .
assume first that @xmath12 .
the cayley graph @xmath36 of the group @xmath37 with respect to the standard generators @xmath25 and @xmath26 is the directed multigraph with vertex set @xmath38 , edge set @xmath39 , source function @xmath40 given by @xmath41 , and target function @xmath42 given by @xmath43 . recall that a graph is just a pair consisting of a vertex set and an edge set with the property that every edge is a two - element subset of the vertex set .
every directed multigraph can be converted into a graph by ignoring the direction and the multiplicity of the edges and deleting the loops .
for the purpose of this paper it is sufficient to think of @xmath36 as a graph , and we shall tacitly do so .
consider the illustration of @xmath36 in figure [ fig : cayley ]
. intuitively speaking , we may look at it from the side to see the associated bass
serre tree .
formally , let @xmath44 and let @xmath13 be the graph with vertex set @xmath45 and edge set @xmath46 .
this graph is actually a tree ; it is obviously connected and , by britton s lemma , it does not contain any cycle .
note that the canonical projection @xmath47 given by @xmath48 is a weak graph homomorphism from @xmath36 to @xmath13 , i.e. whenever the vertices @xmath49 and @xmath50 are adjacent in @xmath36 , their images @xmath51 and @xmath52 either agree or they are adjacent in @xmath13 .
[ rem : levelmaps ] consider the infinite cyclic group @xmath35 and the map @xmath53 given by @xmath54 and @xmath55 .
the latter can be uniquely extended to a group homomorphism @xmath56 .
indeed , the equation @xmath57 holds in @xmath35 so that we can apply von dyck s theorem to extend @xmath58 , see e.g. ( * ? ? ? * , fn.2 ) . since @xmath59 ,
the group homomorphism @xmath56 is constant on the cosets from @xmath60 and therefore induces a well - defined map @xmath61 given by @xmath62 .
we shall think of @xmath58 and @xmath63 as level functions , they assign a level to every vertex of @xmath36 and @xmath13 , respectively . [
lem : neighbours ] every vertex @xmath51 of @xmath13 has exactly @xmath64 neighbours ; @xmath0 of them are one level below and @xmath1 of them are one level above the vertex . by construction , the levels of two adjacent
vertices always differ exactly by @xmath65 .
the defining relation @xmath66 ( `` @xmath67 '' ) and britton s lemma ( `` @xmath68 '' ) imply that @xmath69 if and only if @xmath31 , whence the vertex @xmath51 has exactly @xmath1 neighbours above .
similarly , it has exactly @xmath0 neighbours below because @xmath70 if and only if @xmath29 . of @xmath11 with respect to the standard generators @xmath25 and @xmath26 . ]
( 0,0 ) ( -180.6,30)@xmath25 ( -149,36.5)@xmath25 ( -174,12)@xmath26 ( -158,15.5)@xmath26 ( -157,48.5)@xmath26 the second projection captures the information that is obtained by looking at @xmath36 from the front .
in order to construct it , we introduce another group .
let @xmath71 be the set of all affine transformations of the real line that preserve the orientation , i.e. all maps @xmath72 of the form @xmath73 with @xmath74 and @xmath75 .
this set endowed with the composition @xmath76 forms a group . as in the construction of the level function @xmath56 in remark [ rem : levelmaps ] , consider the map @xmath77 given by @xmath78 and @xmath79 .
due to von dyck s theorem , it can be uniquely extended to a group homomorphism @xmath80 .
the group @xmath71 has a geometric interpretation . in order to describe it ,
let @xmath14 be the hyperbolic plane as per the poincar half - plane model , i.e. @xmath81 , endowed with the standard metric @xmath82 the elements of @xmath71 can be thought of as isometries of the hyperbolic plane @xmath14 , which are precisely the maps @xmath83 of the form @xmath84 see e.g. ( * ? ? ?
* theorem 7.4.1 ) .
correspond to the orientation - preserving isometries of @xmath14 that fix @xmath85 , which is defined in section [ sub : compact ] .
the orientation - reversing isometries of @xmath14 that fix @xmath85 will be crucial for the investigation of baumslag
solitar groups @xmath2 with @xmath86 in the . ]
now , we are ready to construct the second projection @xmath87 .
pick an element @xmath88 , map it via @xmath89 to @xmath71 , think of the latter as an isometry of @xmath14 , and evaluate it at @xmath90 .
the following lemma illustrates this construction .
[ lem : lookingfromfront ] for every @xmath88 the point @xmath91 is above the point @xmath92 ; the two points have the same real part and their distance is @xmath93 .
similarly , for every @xmath88 the point @xmath94 is right from the point @xmath92 ; the two points have the same imaginary part and their distance is @xmath95 .
so , in some way , we are actually looking at @xmath36 from the front .
this is clear for @xmath96 .
now , pick an arbitrary element @xmath88 .
the points @xmath91 and @xmath92 are obtained by applying @xmath97 to the points @xmath98 and @xmath99
. remains true when replacing @xmath100 by @xmath101 .
therefore , we may actually conclude that @xmath102 . ]
but since @xmath97 is the composition of a dilation @xmath103 and a translation @xmath104 , the relative position of the two points is preserved .
the same argument works for the second assertion , which completes the proof . here and throughout the paper
, we use the symbol @xmath105 to denote the non - negative integers and the symbol @xmath106 to denote the strictly positive integers . given a graph with vertex set @xmath107
, we consider finite paths @xmath108 , infinite paths @xmath109 , and doubly infinite paths @xmath110 . in any case , being a path means that for every possible choice of @xmath111 the vertices @xmath112 and @xmath113 are adjacent in the graph .
moreover , we say that a path is reduced if for every possible choice of @xmath111 the vertices @xmath112 and @xmath114 are distinct .
fix an ascending doubly infinite path @xmath115 in the tree @xmath13 .
ascending refers to the level function defined in remark [ rem : levelmaps ] , and it means that for every @xmath116 the vertex @xmath112 is located above the preceding vertex @xmath117 .
now , let @xmath118 be the full @xmath119-preimage of @xmath120 , i.e. the set consisting of all @xmath88 such that the image @xmath121 is traversed by @xmath122 .
the subgraph @xmath123 spanned by @xmath118 , see in figure [ fig : hyperbolic ] , is obviously connected so that the graph distance @xmath124 becomes a metric .
this subgraph is sometimes referred to as discrete hyperbolic plane or plane of bricks , which makes particular sense in light of proposition [ prop : discretehyperbolicplane ] .
variations of the latter have already been used in the literature , e.g. in @xcite , @xcite , and @xcite .
concerning @xcite , see also the remark in @xcite .
( right ) and its projection to @xmath14 ( left ) .
] ( 0,0 ) ( -255,32)@xmath125 ( -366,104)@xmath126 ( -189,27 ) ( -321,80 ) ( -243,92.5 ) [ prop : discretehyperbolicplane ] the restriction @xmath127 is a quasi - isometry , even a bi - lipschitz map , between the graph @xmath128 endowed with the graph distance @xmath124 and the hyperbolic plane @xmath14 endowed with the standard metric @xmath129 . we realise the edges of the graph @xmath128 geometrically .
whenever two vertices @xmath130 are adjacent , we connect their images @xmath92 and @xmath131 by a geodesic in @xmath14 . in order to avoid confusion
, we refer to these images as @xmath14-vertices and to the geodesics between them as @xmath14-edges . by the proof of lemma [ lem : neighbours ] and by lemma
[ lem : lookingfromfront ] , the @xmath14-vertices and @xmath14-edges yield a tessellation of the hyperbolic plane with isometric bricks of the following shape .
the @xmath14-vertices of each brick are located on two distinct horizontal lines ; on the upper one there are @xmath132 and on the lower one there are @xmath133 . in either case , the @xmath14-vertices are connected by @xmath14-edges of length @xmath134 to form a chain ( = piecewise geodesic curve ) . due to the curvature ,
both the two leftmost and the two rightmost @xmath14-vertices are located precisely above each other and connected by vertical @xmath14-edges of length @xmath135 , see in figure [ fig : hyperbolic ] and figure [ fig : approximate ] . since the bricks are uniformly bounded and cover the hyperbolic plane @xmath14 , the restriction @xmath127 is certainly quasi - surjective .
pick any two vertices @xmath130 .
we aim to estimate the distances @xmath136 and @xmath137 by multiples of each other .
first , choose a path of minimal length from @xmath49 to @xmath50 in @xmath128 .
it corresponds to a chain of @xmath14-edges from @xmath138 to @xmath139 , see in figure [ fig : hyperbolic ] .
this chain consists of @xmath136 many @xmath14-edges , each of which has length at most @xmath140 .
hence , @xmath141 for the converse estimate , let us make the following auxiliary definition .
every point @xmath142 that is not in the interior of a brick is either an @xmath14-vertex , in which case we define @xmath143 to be @xmath9 , or it is in the interior of an @xmath14-edge , in which case we define @xmath143 to be one of the endpoints of the @xmath14-edge , whichever is closer . in the case
that @xmath9 is precisely in the middle of the @xmath14-edge , we choose the left endpoint rather than the right one and the lower endpoint rather than the upper one . with this notion in mind ,
consider the geodesic @xmath144 from @xmath138 to @xmath139 , see figure [ fig : approximate ] .
whenever @xmath144 traverses the interior of a brick @xmath145 , it enters the interior at some point @xmath146 and leaves it at some other point @xmath147 .
in this situation , approximate the part of @xmath144 from @xmath9 to @xmath148 by a chain of @xmath14-edges from @xmath143 to @xmath149 .
we may choose this chain such that , whenever @xmath150 , the chain has no @xmath14-edge at all and , otherwise , the number of @xmath14-edges in the chain is at most @xmath151 .
but , by a compactness argument , there is an @xmath152 such that if the part of @xmath144 has length smaller than @xmath153 , then @xmath150 and the chain has no @xmath14-edge at all .
therefore , we may conclude that @xmath154 it is not hard to see that if we do this for every brick @xmath145 whose interior is traversed by @xmath144 , we finally obtain a chain of @xmath14-edges from @xmath138 to @xmath139 .
depending on whether a part of @xmath144 originally traversed the interior of a brick or ran along an @xmath14-edge , we may estimate the number of @xmath14-edges approximating it by @xmath155 , by @xmath156 , or by @xmath157 times its length .
hence , @xmath158 in the case @xmath159 and @xmath160 . ]
( 0,0 ) ( -251,109)upper ( -251,16)lower ( -9,69)@xmath135 ( -49,19)@xmath134 ( -162,5)@xmath144 ( -162,35)@xmath9 ( -141,34)@xmath143 ( -63,130)@xmath148 ( -111,126)@xmath149 [ rem : exponentialline ] note that the horizontal lines mentioned in the proof of proposition [ prop : discretehyperbolicplane ] are horospheres , and by no means geodesics .
for example , one may pick such a horizontal line and observe that the part of the line contained in the closed disc @xmath161 with centre @xmath90 and radius @xmath162 has a length growing exponentially in @xmath163 , see also figure [ fig : shortline ] on page .
[ rem : treebolic ] the level of a vertex @xmath88 can be recovered both from @xmath164 and from @xmath92 .
in fact , the image of @xmath165 is contained in the horocyclic product of the tree @xmath13 and the hyperbolic plane @xmath14 , which is sometimes referred to as treebolic space , see @xcite for details .
both the tree @xmath13 and the hyperbolic plane @xmath14 have a natural compactification . in case of @xmath13 , it is the end compactification , which can be constructed as follows . fix a base point , say @xmath166 , and consider the set @xmath167 of all reduced paths that start in @xmath145 , be they finite or infinite .
the endpoint map yields a one - to - one correspondence between the finite paths and the vertices @xmath60 .
we may therefore think of @xmath60 as a subset of @xmath167 .
the set @xmath167 can be endowed with the metric @xmath168 here , the symbol @xmath169 denotes the number of edges the two paths run together until they separate , i.e. @xmath170 , see in figure [ fig : compactifications ] .
hence , the later the paths separate the closer they are .
the set @xmath167 endowed with the metric @xmath171 is a compact metric space that contains @xmath60 as a discrete and dense subset .
the complement of @xmath60 is the set of infinite paths , it is usually denoted by @xmath15 and called the space of ends . in case of @xmath14 ,
we temporarily switch to the poincar disc model .
more precisely , instead of working in the half - plane @xmath81 , we consider the open unit disc @xmath172 .
the cayley transform @xmath173 given by @xmath174 is one possibility to convert between the two models .
since we are currently interested in the topological structure , let us highlight that the hyperbolic topology on @xmath175 is the one induced by the cayley transform , i.e. the one that turns the cayley transform into a homeomorphism .
it happens to agree with the standard topology on @xmath175 .
so , topologically speaking , the hyperbolic plane in the poincar disc model is just a subspace of the complex plane @xmath176 .
we may therefore compactify it by taking the closed unit disc @xmath177 , see figure [ fig : compactifications ] . in order to translate this compactification back to the poincar half - plane model , we first extend both the domain and the codomain of the cayley transform so that we obtain a bijection @xmath178 , and then apply its inverse .
the resulting space @xmath179 is our compactification .
it is , once again , endowed with the induced topology , and thus a compact space that contains @xmath14 as a dense subset .
the complement of @xmath14 is the union @xmath180 , it is usually denoted by @xmath16 and called the hyperbolic boundary .
having introduced the hyperbolic boundary this way , the following lemma gives us a helpful criterion for convergence .
its proof is elementary and we leave it to the reader .
[ lem : convergence ] a sequence @xmath181 in @xmath14 converges to @xmath85 if and only if the absolute values @xmath182 tend to infinity .
moreover , it converges to a point @xmath183 if and only if it does with respect to the standard topology on the complex plane @xmath176 .
( 0,0 ) ( -258,97)@xmath9 ( -250,82 ) ( -196,72)@xmath148 ( -240,45)@xmath145
the aim of the current work is to study random walks on baumslag solitar groups . before doing so
, we fix the notation . given a countable state space @xmath184 , an initial probability measure @xmath185 $ ] , and transition probabilities @xmath186 $ ] , we are interested in the markov chain @xmath187 that starts according to @xmath188 and proceeds according to @xmath0 . formally , we construct the probability space @xmath189 , where @xmath190 is the set of trajectories , @xmath191 is the product @xmath19-algebra , and @xmath192 is the probability measure induced by @xmath188 and @xmath0 .
the projections @xmath193 given by @xmath194 become random variables that constitute the markov chain . for details on the terminology
used above , see e.g. @xcite , and for a gentle introduction to discrete markov chains , see e.g. @xcite .
we will use the term random walk instead of markov chain .
next , let us assume that @xmath184 is a countable group @xmath38 , in which case we may study random walks whose transition probabilities are adapted to the group structure .
in order to do so , we first pick a probability measure @xmath195 $ ] whose support @xmath196 generates @xmath38 as a semigroup , see also remark [ rem : recurrenceandtransience ] below .
then , we consider the random walk given by the following data .
the initial probability measure @xmath197 $ ] puts all mass on the identity element @xmath198 and the transition probabilities @xmath199 $ ] are given by @xmath200 .
we could also have said @xmath201 , which leads to a handy interpretation .
the random walk starts at the identity element and has independent @xmath10-distributed increments each of which is multiplied from the right to the current state .
therefore , @xmath202 a.s .
( = almost surely ) and for every @xmath162 we may decompose @xmath203 , where @xmath204 is a sequence of independent @xmath10-distributed random variables , see the right - hand side of figure [ fig : randomwalk ] . .
] ( 0,0 ) ( -287,60)@xmath205 ( -406,94 ) ( -380.5,108.5 ) ( -347,7)@xmath189 ( -176.5,1)@xmath206 ( -163,148)@xmath207 ( -28,132)@xmath208 [ rem : recurrenceandtransience ] since we assume that @xmath209 generates @xmath38 as a semigroup , the random walk is irreducible , i.e. any two states can be reached from each other with positive probability .
in particular , the following dichotomy holds .
either every state is recurrent , i.e. the return probability is equal to 1 , or every state is transient , i.e. the return probability is smaller than 1 . in the latter case , the probability that every finite set of states
will eventually be left and the random walk escapes to infinity is equal to 1 .
we need to assume that the probability of huge jumps is sufficiently small .
the notion of moments helps us to make this assumption rigorous .
given a probability space , e.g. @xmath189 introduced in section [ sub : bn ] , and a real valued random variable @xmath210 , the latter has finite first moment if @xmath211 . in this case , both @xmath212 and @xmath213 , and we can define the expectation @xmath214 . of course , the difference would still make sense if only one of the two integrals was finite .
but this is not of relevance for us and when writing @xmath215 we implicitly mean that @xmath216 .
more generally , given any non - negative @xmath217 , a real valued random variable @xmath210 has finite @xmath111-th moment if @xmath218 . in our setting ,
the increments @xmath204 are not real valued random variables but take values in @xmath38 , whence we need to specify the hugeness of a jump before talking about finite moments .
if @xmath38 is a finitely generated group and @xmath219 is a finite generating set , then the word metric @xmath220 on @xmath38 is given by @xmath221 note that the word metric coincides with the distance in the respective cayley graph .
a random variable @xmath222 has finite @xmath111-th moment if the image @xmath223 has finite @xmath111-th moment in the classical sense , i.e. if @xmath224 .
we leave it to the reader to verify that this property does not depend on the choice of the finite generating set @xmath219 , see also ( * ? ? ?
* lemma 11.37 ) .
let us now return to the situation we are interested in , namely that @xmath225 with @xmath12 .
when working with the projection @xmath226 , we often consider the imaginary parts @xmath227 and the real parts @xmath228 separately , and it is convenient to abbreviate the former by @xmath229 and the latter by @xmath230 .
occasionally , we do not need to assume that @xmath231 has some finite moment but impose this assumption on the images @xmath232 and @xmath233 .
the following lemma relates the two situations .
[ lem : assumptions ] if @xmath231 has finite @xmath111-th moment , then @xmath232 and @xmath233 have finite @xmath111-th moment , too .
[ rem : multiple ] before we prove lemma [ lem : assumptions ] , let us note that for every @xmath88 the imaginary part @xmath229 can be expressed in terms of the level @xmath234 , namely as @xmath235 .
this formula can be shown using either the multiplication in @xmath71 or lemma [ lem : lookingfromfront ] .
taking the logarithm on both sides yields @xmath236 .
so , instead of thinking of@xmath237 we may think of a multiple of @xmath234 .
let @xmath238 be the standard generating set .
then @xmath239 concerning the second assertion , observe that @xmath240 , which can be shown by the same argument as in the proof of proposition [ prop : discretehyperbolicplane ] .
this observation allows us to estimate @xmath241 by a multiple of @xmath242 .
indeed , @xmath243 & \leq\max{\ensuremath{\{\ , \ell_{a},\ell_{b } \,\}}}\cdot\operatorname{\textit{d}_{s}}(1,g)+\ln\left(\frac{q}{p}\right)\cdot|\lambda(g)|\leq\max{\ensuremath{\{\ , \ell_{a},\ell_{b } \,\}}}\cdot\operatorname{\textit{d}_{s}}(1,g)+\ln\left(\frac{q}{p}\right)\cdot\operatorname{\textit{d}_{s}}(1,g)\,.\end{aligned}\ ] ] therefore , @xmath244 which proves the claim .
it is easy to construct examples showing that the converse of lemma [ lem : assumptions ] does not hold .
in addition to the moments of @xmath232 and @xmath233 , we will use the notion of vertical drift . consider a random walk @xmath245 on @xmath38 and its pointwise projection @xmath246 to the levels .
since @xmath247 , these projections constitute a random walk on the integers with i.i.d .
( = independent and identically distributed ) increments .
[ def : verticaldrift ] if@xmath232 has finite first moment , then @xmath248 has finite first moment and we can define the expectation @xmath249 .
the latter is called the vertical drift and denoted by @xmath250 .
we will distinguish between positive vertical drift , i.e. @xmath251 , negative vertical drift , i.e. @xmath252 , and no vertical drift , i.e. @xmath253 , which is the most subtle of the three cases .
the poisson furstenberg boundary is a probabilistic model for the long - time behaviour of a random walk . in order to define it
, we need to ensure that we are working with lebesgue rohlin spaces , which are also known as standard probability spaces . for definitions and basic examples
we refer to @xcite , @xcite , and @xcite .
moreover , let us mention the collection of facts in ( * ? ? ?
* appendix ) and the more informal introduction in @xcite .
the most prominent examples of lebesgue
rohlin spaces are discrete probability spaces and the unit interval @xmath254 $ ] endowed with the lebesgue @xmath19-algebra @xmath255 and the lebesgue measure @xmath58 .
in fact , every lebesgue
rohlin space is isomorphic .
so , we actually mean isomorphic @xmath256 .
recall that two probability spaces @xmath257 and @xmath258 are isomorphic @xmath256 if there are null sets @xmath259 with @xmath260 and a bijection @xmath261 which is measurable and measure preserving in both directions . ] either to one of these examples or to the disjoint union of an interval @xmath262 $ ] with @xmath263 and countably many atoms with total mass @xmath264 , see @xcite and ( * ? ? ?
* proposition 6 ) .
[ rem : polish ] a polish space is a topological space that is separable , i.e. contains a countable and dense subset , and completely metrisable , i.e. there is a metric that induces the topology and turns the space into a complete metric space .
all polish spaces endowed with the borel @xmath19-algebra @xmath265 and a borel measure @xmath10 become , after completion , examples of lebesgue rohlin spaces , see @xcite and ( * ? ? ?
* , example 1 ) . in light of remark [ rem : polish ]
, we may observe that the space of trajectories @xmath266 introduced in section [ sub : bn ] is the product @xmath267 and can therefore be endowed with the product topology .
one can show that the latter is actually a polish space , see e.g. ( * ? ? ?
* theorem 24.11 ) . since its borel @xmath19-algebra agrees with the product @xmath19-algebra @xmath191 ,
the completion of @xmath189 is a lebesgue rohlin space . from now on , let us assume that , as soon as a measurable space is endowed with a probability measure , we are working with its completion .
we may therefore say that @xmath189 is a lebesgue rohlin space . since we are interested in the long - time behaviour of the trajectories @xmath268 , we identify those pairs of trajectories whose tails sooner or later behave identically .
more precisely , we define an equivalence relation @xmath269 on @xmath266 by @xmath270 note that we allow the times @xmath271 and @xmath272 to be different .
if we did not , we would end up with the tail boundary instead of the poisson furstenberg boundary .
consider the partition @xmath273 of @xmath266 into equivalence classes mod @xmath269 , see in figure [ fig : randomwalk ] .
this partition induces a sub-@xmath19-algebra @xmath274 of @xmath191 , consisting of all @xmath275 which are compatible with the partition @xmath273 , i.e. which are unions of equivalence classes mod @xmath269 , see in figure [ fig : randomwalk ] .
the poisson furstenberg boundary @xmath276 is the quotient of @xmath189 with respect to the induced sub-@xmath19-algebra @xmath274 .
more precisely , it is the lebesgue rohlin space @xmath277 that consists of the measurable hull @xmath278 of @xmath273 , the induced sub-@xmath19-algebra @xmath279 , and the restriction @xmath280 of the probability measure @xmath192 to @xmath279 . concerning the measurable hull ,
see @xcite and @xcite .
moreover , compare ( * ? ? ?
* proposition 11 ) . the map from the trajectory space @xmath266 to the poisson furstenberg boundary @xmath145 that assigns to every trajectory @xmath281 the respective element of the partition @xmath278
is called the boundary map @xmath282 .
note that the above is not the only possible definition of the poisson furstenberg boundary , further equivalent ones are given in @xcite .
one important feature of the poisson furstenberg boundary is that it can be used to describe all bounded harmonic functions on the state space @xmath184 .
assume we are given a countable state space @xmath184 and transition probabilities @xmath283 $ ] as introduced in section [ sub : bn ] .
a function @xmath284 is called harmonic if for every element @xmath285 the equation @xmath286 holds . in other words , being at @xmath285 , the value of @xmath287 today is exactly as large as the expected value of @xmath287 tomorrow .
the initial probability measure of a random walk is denoted by @xmath185 $ ] .
first , we pick some reference measure @xmath188 with @xmath288 .
then , we consider the random walk @xmath245 that starts according to @xmath188 , has probability measure @xmath289 and poisson furstenberg boundary @xmath290 .
all other initial probability measures , in particular the dirac measures @xmath291 at points @xmath285 , are absolutely continuous with respect to @xmath188 .
therefore , the measures @xmath292 are absolutely continuous with respect to @xmath289 , which implies that we may endow @xmath293 with measures @xmath294 in order to obtain the respective poisson furstenberg boundaries . from this point of view , it would have made sense to define the poisson furstenberg boundary as a measurable space @xmath293 endowed with a family of measures .
a first step decomposition shows that for every two points @xmath295 the equation @xmath296 holds .
hence , given an essentially bounded function @xmath297 mapping from the poisson furstenberg boundary @xmath290 to the real numbers @xmath298 , we can construct a bounded harmonic function @xmath299 given by the poisson integral representation formula @xmath300 .
there is also a way back from @xmath287 to @xmath297 using martingale convergence so that , in the end , one obtains a one - to - one correspondence , even an isometry of banach spaces , between the space @xmath301 of equivalence classes of essentially bounded functions and the space @xmath302 of bounded harmonic functions , see e.g. ( * ? ? ?
* section 2.1 ) . given a random walk , be it on a generic state space or on a group ,
a challenging problem is to decide whether the poisson furstenberg boundary is trivial or not . in the latter case ,
one may wonder how to identify it geometrically .
we shall only outline a few results about the poisson furstenberg boundary of random walks on countable groups .
a recent survey has been given by erschler in @xcite .
as before , let @xmath245 be a random walk on a countable group @xmath38 driven by the probability measure @xmath10 .
we assume that the support @xmath209 generates @xmath38 as a semigroup , see section [ sub : bn ] .
if @xmath38 is abelian , then the poisson furstenberg boundary is always trivial , see @xcite and @xcite .
the same holds true for all groups of polynomial growth , and for groups of subexponential growth endowed with a probability measure @xmath10 with finite first moment . for the special case of probability measures with finite support ,
see @xcite , and for the general case , see e.g. ( * ? ? ?
* theorem 5.3 ) and @xcite .
moreover , it was shown in @xcite , that the assumption of finite first moment can not be dropped .
if @xmath38 is amenable , then one can show that there is at least one symmetric probability measure @xmath10 such that the poisson furstenberg boundary is trivial , see the conjecture in @xcite .
the proof of the conjecture has been announced in ( * ? ? ?
* theorem 4 ) and given in @xcite and @xcite . in case of the baumslag
solitar group @xmath303 , the poisson furstenberg boundary may or may not be trivial depending on the vertical drift , see definition [ def : verticaldrift ] . more precisely , for random walks on @xmath303 with finite first moment the poisson
furstenberg boundary is isomorphic to @xmath298 for @xmath252 and trivial for @xmath253 and isomorphic to @xmath304 for @xmath251 , see ( * ? ? ?
* theorem 5.1 ) .
we may think of @xmath304 as the space of upper ends of the corresponding bass serre tree @xmath13 .
further results about random walks on rational affinities are given in @xcite .
for the poisson furstenberg boundary of lamplighter random walks , see @xcite , @xcite , @xcite , and also @xcite .
if @xmath38 is non - amenable , then the poisson furstenberg boundary is always non - trivial , see @xcite .
in particular , this implies that the poisson furstenberg boundary of random walks on non - amenable baumslag
solitar groups can never be trivial , also when @xmath253 .
kaimanovich s strip criterion is a tool for identifying the poisson furstenberg boundary geometrically .
the strategy is to choose a suitable @xmath10-boundary as a candidate .
our one will be given in terms of the boundaries @xmath16 and @xmath15 .
the strip criterion then enables us to prove that our candidate is indeed isomorphic to the poisson furstenberg boundary .
let us first recall the strip criterion . for a proof
we refer to @xcite .
[ thm : strip ] let @xmath245 be a random walk on a countable group @xmath38 driven by a probability measure @xmath10 with finite entropy @xmath305 .
moreover , let @xmath306 and @xmath307 be @xmath308- and @xmath10-boundaries , respectively . if there exist a gauge @xmath309 on @xmath38 with associated gauge function @xmath310 and a measurable @xmath38-equivariant map @xmath311 assigning to pairs of points @xmath312 non - empty strips @xmath313 such that for every @xmath88 and @xmath314-almost every @xmath312 @xmath315 then the @xmath10-boundary @xmath307 is maximal .
[ rem : removeg ] the proof shows that it is not even necessary to verify the convergence for every @xmath88 .
it suffices to consider the special case @xmath96 as long as we can ensure that a random strip contains the identity element @xmath198 with positive probability , i.e. that @xmath316 .
the following four notions have not yet been introduced . *
( a ) `` entropy '' * the entropy of the probability measure @xmath10 is the expected amount of information contained in the outcome of a random variable that is distributed according to @xmath10 .
more precisely , it is the real number given by @xmath317 . here , as usual , one defines @xmath318 . for us ,
the assumption of finite entropy will be no issue because baumslag
solitar groups are finitely generated and the increments under investigation have finite first moment .
this implies that their probability measures @xmath10 have finite entropy , as shown by the following well - known lemma .
[ lem : entropy ] let @xmath38 be a finitely generated group and a probability measure @xmath10 with infinite entropy supported on the generating set @xmath319 . then
, @xmath184 has finite first moment with respect to the word metric @xmath320 , but @xmath10 has infinite entropy . ] and let @xmath321 $ ] be a probability measure . if a random variable @xmath222 distributed according to @xmath10 has finite first moment , then @xmath10 has finite entropy .
let @xmath219 be a non - empty finite generating set .
moreover , let @xmath322 , whence @xmath323 . in this proof , we shall use the shorthand notation @xmath324 instead of @xmath220 to denote the word metric on @xmath38 .
we have to show that the entropy @xmath325 is finite .
first , we change the base of the logarithm @xmath326 and split the summands appropriately @xmath327 then , we recall that the function @xmath328 is increasing on the interval @xmath329 $ ] , and conclude that @xmath330 & \leq\sum_{g\in g}\operatorname{\textit{d}}(1,g)\cdot b^{-\operatorname{\textit{d}}(1,g)}\leq\sum_{n=0}^{\infty}(2\cdot|s|)^{n}\cdot n\cdot b^{-n}<\infty\,.\end{aligned}\ ] ] on the other hand , since @xmath184 has finite first moment , @xmath331 so , both sums are finite , whence @xmath305 must be finite , too . * ( b ) `` @xmath10-boundary '' * two equivalent definitions of a @xmath10-boundary can be found in @xcite . for us , it suffices to record that every lebesgue
rohlin space @xmath307 endowed with a left @xmath38-action and a boundary map @xmath332 that is measurable , @xmath269-invariant , and @xmath38-equivariant is a @xmath10-boundary . *
( c ) `` @xmath308-boundary '' * while @xmath10 is the probability measure driving the random walk , the symbol @xmath308 denotes the reflected probability measure given by @xmath333 .
accordingly , a @xmath308-boundary is a space that satisfies the requirements of a @xmath10-boundary when replacing @xmath10 by @xmath308 . *
( d ) `` gauge '' * a gauge @xmath334 is an exhaustion @xmath309 of the group @xmath38 , i.e. a sequence of subsets @xmath335 which is increasing @xmath336 and whose union @xmath337 is the whole group @xmath38 . given a gauge @xmath334 and an element @xmath88 , we may ask for the minimal index @xmath338 with the property that @xmath339 . this index is the value of the associated gauge function @xmath310 at @xmath49 .
kaimanovich distinguishes between various kinds of gauges , see @xcite .
for example , a gauge @xmath334 is subadditive if any two group elements @xmath340 satisfy @xmath341 and it is temperate if all gauge sets @xmath342 are finite and grow at most exponentially . even though these two properties do play a crucial role in the corollaries to the strip criterion given in @xcite , they are not required in the strip criterion itself . and
, in fact , not all of our gauges will have these two properties .
* ( e ) `` measurable strips '' * the power set @xmath343 is naturally endowed with the product @xmath19-algebra , which enables us to talk about measurability of the map @xmath344 . *
( f ) `` maximal '' * the poisson furstenberg boundary @xmath276 inherits a left @xmath38-action from the space of trajectories , and every @xmath10-boundary @xmath307 is a @xmath38-invariant measurable quotient of @xmath276 . @xmath345_{\operatorname{bnd } } \ar[dr]^{\operatorname{bnd}_{+ } } \ar@{}[d]|/.28868\einheit/{\circlearrowleft } & \\ \makebox[30pt][r]{$(b,\mathcal{b},\nu)$ } \ar[rr]_{\vphantom{b}\pi } & & \makebox[30pt][l]{$(b_{+},\mathcal{b}_{+},\nu_{+})$ } } \end{xy}\ ] ] a @xmath10-boundary @xmath307 is called maximal if the projection @xmath346 is a measurable isomorphism between lebesgue rohlin spaces .
we still assume that @xmath12 and consider a random walk @xmath245 on @xmath225 . moreover , recall the abbreviations @xmath347 and @xmath348 introduced in section [ sub : rpipavd ] . the following lemmas concern the behaviour of the projections @xmath349 .
they seem to be well - known and we do not claim originality .
but , for the sake of completeness , we give rigorous proofs .
[ lem : positivedrifthyp ] assume that @xmath232 has finite first moment .
if the vertical drift is positive , i.e. @xmath251 , then the projections @xmath349 converge a.s . to @xmath85 .
we can use the strong law of large numbers to obtain @xmath350{~n\to\infty~}\,\mathbb{e}(\lambda(x_{1}))=\delta>0\,.\ ] ] hence , the projections @xmath351 tend a.s . to infinity and , by remark [ rem :
multiple ] , the imaginary parts @xmath352 do .
this , of course , implies that the absolute values @xmath353 tend a.s . to infinity .
now , we can use lemma [ lem : convergence ] to complete the proof .
[ lem : negativedrifthyp ] assume that both @xmath232 and @xmath233 have finite first moment .
if the vertical drift is negative , i.e. @xmath252 , then the projections @xmath349 converge a.s . to a random element @xmath183 .
note that the argument given in the proof of lemma [ lem : positivedrifthyp ] can be adapted to show that the imaginary parts @xmath352 converge a.s . to @xmath354 ,
whence we only need to understand the behaviour of the real parts @xmath355 .
the equation @xmath356 yields @xmath357 , and in light of the multiplication in @xmath71 we obtain @xmath358 & = \big(\pi_{\operatorname{aff\,^{+}}(\mathbb{r})}(x_{1})\circ\dotsc\circ\pi_{\operatorname{aff\,^{+}}(\mathbb{r})}(x_{n})\big)(i ) \\ & = a_{x_{1}}\cdot\dotsc\cdot a_{x_{n}}\cdot i+\sum_{k=1}^{n}a_{x_{1}}\cdot\dotsc\cdot a_{x_{k-1}}\cdot b_{x_{k}}\,.\end{aligned}\ ] ] hence , the real parts @xmath355 are partial sums of the infinite series @xmath359 with @xmath360 . in order to verify a.s
. convergence of the latter , we apply cauchy s root test , @xmath361{~k\to\infty~}\,\left(\frac{q}{p}\right)^{\delta}<1\,.\ ] ] for the convergence claimed in the first factor we can use the strong law of large numbers , for the one claimed in the second factor the borel cantelli lemma .
indeed , let us write @xmath362 for the quotient @xmath363 .
in order to show that @xmath364 a.s . , recall
that @xmath233 has finite first moment .
so , for every @xmath152 we may estimate @xmath365 therefore , the borel
cantelli lemma yields @xmath366 . replacing @xmath153 by @xmath367
, we obtain a countable family of null sets whose union is , of course , again a null set that consists of all @xmath368 with @xmath369 .
hence , @xmath364 a.s .
, see also ( * ? ? ?
* exercise 5.1.3 ) .
so , we have finally convinced ourselves that @xmath370 a.s .
, whence @xmath359 converges a.s . to a random element @xmath371 .
the natural question that remains is the one asking for the driftless case .
[ quote : brofferio ] an answer has been given by brofferio in ( * ? ? ?
* theorem 1 ) .
it says that under the same mild assumptions , namely that @xmath232 and @xmath233 have finite first moment , the projections @xmath349 converge a.s . to @xmath85 .
but , for us , a result of slightly different flavour will be of relevance .
[ lem : zerodrifthyp ] assume that @xmath232 has finite second moment and there is an @xmath152 such that @xmath233 has finite @xmath372-th moment .
if there is no vertical drift , i.e. @xmath253 , then the projections @xmath349 have sublinear speed , i.e. @xmath373{~n\to\infty~}\,0\,.\ ] ] the proof is based on ideas that go back to lie in ( * ? ? ?
* lemme 5.49 ) and have also been used by cartwright , kaimanovich , and woess in ( * ? ? ?
* proposition 4b ) .
we first adapt these ideas to our situation by stating and proving lemma [ lem : elie ] , and then deduce lemma [ lem : zerodrifthyp ] . by assumption , there is no vertical drift so that the pointwise projection @xmath246 is recurrent , see plya s theorem for the simple random walk and the chung
fuchs theorem in @xcite for the general case .
in particular , we know that there exists a.s . a strictly increasing sequence @xmath374 given by @xmath375 and by @xmath376 for all @xmath162 .
we call @xmath377 the @xmath163-th ladder time , see figure [ fig : upwardstopping ] for an illustration of the first ladder times @xmath378 and @xmath379 . the following lemma concerns the random variable @xmath380 with @xmath381 . and @xmath379 .
] ( 0,0 ) ( -116,26)@xmath382 ( -184,86)@xmath383 [ lem : elie ] under the same assumptions as in lemma [ lem : zerodrifthyp ] , namely that @xmath232 has finite second moment , there is an @xmath152 such that @xmath233 has finite @xmath372-th moment , and there is no vertical drift , i.e. @xmath253 , the random variable @xmath380 has finite first moment . adapting the proof of (
* lemme 5.49 ) , we begin with some preliminaries .
pick an @xmath152 that satisfies the requirements of lemma [ lem : elie ] and let @xmath384 . since @xmath232 has finite second moment , we know that also @xmath248 has finite second moment and @xmath385 , see @xcite referring to @xcite . using this asymptotics , we obtain @xmath386 in particular , there is a @xmath387 such that for all @xmath388 the summands @xmath389 are strictly smaller than @xmath390 . since @xmath391 , we know that @xmath392 .
moreover , note that , by construction of the ladder times @xmath374 , the differences @xmath393 are i.i.d .
, whence the fact that @xmath394 , which implies that @xmath395 , and the strong law of large numbers yield @xmath396{~n\to\infty~}\,\mathbb{e}\left(\tau^{\beta}\right)\ , , \\ \longrightarrow~\limsup_{n\to\infty}\frac{{\tau(n)}^{\beta}}{n}<\infty~~\text{a.\,s . }
\label{eqn : preliminary } \tag{$\ast$}\end{gathered}\ ] ] now , we are prepared for the main argument .
recall that we aim to show that @xmath380 has finite first moment .
the sums @xmath397 are i.i.d . and non - negative with the additional property that they are not a.s .
equal to @xmath354 .
hence , by ( * ? ? ?
* lemme 5.23 ) , @xmath398 it thus suffices to verify the right - hand side .
in order to do so , we would like to estimate @xmath399 a priori , it might be the case that @xmath400 and the second factor in the rightmost term is @xmath354 , in which case the product would not make sense .
we claim that @xmath401 is a.s .
finite , which does not only legitimate the above estimate but also completes the proof .
indeed , observe that @xmath402 & \leq\underbrace{\limsup_{n\to\infty}\frac{\ln\left(\tau(n)\right)}{{\tau(n)}^{\beta}}}_{=\;0}\,+\limsup_{n\to\infty}\frac{\ln\left(1+\max_{1\leq k\leq\tau(n)}{\ensuremath{\{\ , |b_{x_{k}}| \,\}}}\right)}{{\tau(n)}^{\beta } } \\ & = \limsup_{n\to\infty}\left(\frac{\max_{1\leq k\leq\tau(n)}\left\{\,\ln\left(1+|b_{x_{k}}|\right)^{\frac{1}{\beta}}\,\right\}}{\tau(n)}\right)^{\beta}\leq\limsup_{n\to\infty}\left(\underbrace{\frac{\sum_{k=1}^{\tau(n)}\ln\left(1+|b_{x_{k}}|\right)^{\frac{1}{\beta}}}{\tau(n)}}_{=:\;m_{n}}\right)^{\beta}\,.\end{aligned}\ ] ] now , recall that @xmath403 .
so , by the strong law of large numbers , @xmath404{~n\to\infty~}\,\mathbb{e}\left(\ln(1+|b_{x_{1}}|)^{\frac{1}{\beta}}\right)\,.\ ] ] this implies that @xmath405 a.s . , and completes the proof .
recall from @xcite and @xcite that @xmath385 with a strictly positive constant . in particular , there is a @xmath387 such that for all @xmath388 the summands @xmath406 are strictly larger than @xmath407 and we obtain @xmath408 as we have mentioned in the proof of lemma [ lem : elie ] , the differences @xmath393 are i.i.d . , and they are non - negative .
so , we may deduce from the strong law of large numbers that @xmath409{~n\to\infty~}\,\infty\quad\text{and}\quad\frac{n}{\tau(n)}\,\xrightarrow[\text{\raisebox{.5ex}{a.\,s.}}]{~n\to\infty~}\,0\,.\ ] ] this can be used to estimate the distance between @xmath410 and @xmath349 from above .
first , for every @xmath411 let @xmath412 be the unique element with @xmath413 .
this element exists a.s . because the ladder times @xmath414 do .
now , we may estimate @xmath415{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}}+\underbrace{\frac{\operatorname{\textit{d}_{\mathbb{h}}}(a_{z_{\tau(m)}}\cdot i , a_{z_{\tau(m)}}\cdot i+b_{z_{n}})}{n}}_{{\tikz[baseline=-3.8pt]{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 2};}}}+\underbrace{\frac{\operatorname{\textit{d}_{\mathbb{h}}}(a_{z_{\tau(m)}}\cdot i+b_{z_{n}},a_{z_{n}}\cdot i+b_{z_{n}})}{n}}_{{\tikz[baseline=-3.8pt]{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 3};}}}\,.\ ] ] the numbers refer to figure [ fig : distancedecomposition ]
. we will consider the three summands separately and show that each of them converges a.s . to @xmath354 . for and this is straightforward .
indeed , @xmath416{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}=\frac{|\ln(a_{z_{\tau(m)}})|}{n}\leq\frac{|\ln(a_{z_{\tau(m)}})|}{\tau(m)}=\ln\left(\frac{q}{p}\right)\cdot\left|\frac{\lambda(x_{1})+\dotsc+\lambda(x_{\tau(m)})}{\tau(m)}\right|\,\xrightarrow[\text{\raisebox{.5ex}{a.\,s.}}]{~n\to\infty~}\,\ln\left(\frac{q}{p}\right)\cdot|\mathbb{e}(\lambda(x_{1}))|=\ln\left(\frac{q}{p}\right)\cdot|\delta|=0\ ] ] and similarly @xmath416{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 3};}}\leq\frac{\operatorname{\textit{d}_{\mathbb{h}}}(a_{z_{\tau(m)}}\cdot i , i)}{n}+\frac{\operatorname{\textit{d}_{\mathbb{h}}}(i , a_{z_{n}}\cdot i)}{n}={\tikz[baseline=-3.8pt]{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}+\frac{|\ln(a_{z_{n}})|}{n}={\tikz[baseline=-3.8pt]{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}+\ln\left(\frac{q}{p}\right)\cdot\left|\frac{\lambda(x_{1})+\dotsc+\lambda(x_{n})}{n}\right|\,\xrightarrow[\text{\raisebox{.5ex}{a.\,s.}}]{~n\to\infty~}\,0\,.\ ] ] for recall from the proof of lemma [ lem : negativedrifthyp ] that @xmath417 and observe that for every @xmath418 with @xmath419 the following estimate holds @xmath420 & = \sum_{k=\tau(m)+1}^{\ell}\underbrace{a_{x_{\tau(m)+1}}\cdot\dotsc\cdot a_{x_{k-1}}}_{\leq\,1}\cdot\,|b_{x_{k}}|\leq\sum_{k=\tau(m)+1}^{\ell}|b_{x_{k}}|~~\text{a.\,s . } \tag{$\ast$}\end{aligned}\ ] ] hence , using that @xmath421 and @xmath422 , we obtain @xmath423{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 2 } ; } } & = \frac{1}{n}\cdot\operatorname{arcosh}\left(1+\frac{1}{2}\cdot\left(\frac{|b_{z_{n}}|}{a_{z_{\tau(m)}}}\right)^{2}\right)=\frac{1}{n}\cdot\ln\left(1+\frac{1}{2}\cdot\left(\frac{|b_{z_{n}}|}{a_{z_{\tau(m)}}}\right)^{2}+\sqrt{\left(1+\frac{1}{2}\cdot\left(\frac{|b_{z_{n}}|}{a_{z_{\tau(m)}}}\right)^{2}\right)^{2}-1}\;\right ) \\[3pt ] & \leq\frac{1}{n}\cdot\left(\ln(2)+\ln\left(1+\left(\frac{|b_{z_{n}}|}{a_{z_{\tau(m)}}}\right)^{2}\right)\right)\leq\frac{1}{n}\cdot\left(\ln(2)+2\cdot\ln\left(1+\frac{|b_{z_{n}}|}{a_{z_{\tau(m)}}}\right)\right ) \\[10pt ] & \leq\frac{1}{n}\cdot\left(\ln(2)+2\cdot\ln\left(1+\frac{|b_{z_{\tau(1)}}-b_{z_{\tau(0)}}|}{a_{z_{\tau(0)}}}+\frac{|b_{z_{\tau(2)}}-b_{z_{\tau(1)}}|}{a_{z_{\tau(1)}}}+\dotsc+\frac{|b_{z_{\tau(m)}}-b_{z_{\tau(m-1)}}|}{a_{z_{\tau(m-1)}}}+\frac{|b_{z_{n}}-b_{z_{\tau(m)}}|}{a_{z_{\tau(m)}}}\right)\right)\,,\end{aligned}\ ] ] which allows us to apply ( [ eqn : observation ] ) and finally conclude @xmath424 & \leq\underbrace{\frac{\ln(2)}{n\vphantom{()}}}_{\to\;0}+\,2\cdot\underbrace{\frac{\ln\left(1+\sum_{k=\tau(0)+1}^{\tau(1)}|b_{x_{k}}|\right)+\dotsc+\ln\left(1+\sum_{k=\tau(m)+1}^{\tau(m+1)}|b_{x_{k}}|\right)}{m+1\vphantom{()}}}_{\to\;\mathbb{e}(\ln(1+\sum_{k=1}^{\tau}|b_{x_{k}}|))~\text{a.\,s.~by lemma \ref{lem : elie}}}\cdot\underbrace{\frac{m+1}{\tau(m)}}_{\substack{\to\;0 \\ \text{a.\,s.}}}\cdot\underbrace{\frac{\tau(m)}{n\vphantom{()}}}_{\substack{\leq\,1 \\ \text{a.\,s.}}}\,\xrightarrow[\text{\raisebox{.5ex}{a.\,s.}}]{~n\to\infty~}\,0\,.\hspace*{38pt}\end{aligned}\ ] ] and @xmath349 . ] ( 0,0 ) ( -132,49 ) ( -74,80.5 ) ( -45,41 ) ( -109,94)@xmath425 ( -43,15)@xmath349 ( -114,31)@xmath410 unlike the ones considered in section [ sub : hyperbolicconvergence ] , the projections @xmath426 do not need to satisfy the markov property .
consider , for example , the random walk @xmath245 driven by the uniform measure on the standard generators and their formal inverses .
then , given @xmath427 and @xmath428 , the projection @xmath429 comes back to @xmath145 with probability @xmath430 .
on the other hand , coming back to @xmath145 in a single step would not be possible if the history was @xmath431 and @xmath432 . despite of this subtlety ,
the following lemmas yield almost sure convergence of the projections @xmath426 to a random end .
[ lem : nonzerodrifttree ] assume that @xmath231 has finite first moment .
if the vertical drift is different from @xmath354 , i.e. @xmath433 , then the projections @xmath426 converge a.s . to a random end @xmath434 . by lemma [ lem : assumptions ]
, the assumption that @xmath231 has finite first moment implies that @xmath232 has also finite first moment . for the proof of the lemma [ lem : nonzerodrifttree ] , we give an argument using the notion of regular sequences , see @xcite .
one difference to @xcite is that we do not fix any particular end @xmath435 .
therefore , we replace the busemann function @xmath50 , which depends on the choice of @xmath435 , by the graph distance to the basepoint @xmath145 .
the other difference is that we work with the limit inferior instead of the limit in order to be prepared to deal with the driftless case , too .
let @xmath436 be the graph distance in the tree @xmath13 .
accordingly , the symbol @xmath437 denotes the graph distance @xmath438 from the basepoint @xmath145 to the vertex @xmath9 .
we call a sequence @xmath439 of vertices regular if @xmath416{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}\quad\liminf_{n\to\infty}\;\frac{|x_{n}|_{\mathbb{t}}}{n}>0\quad\text{and}\quad{\tikz[baseline=-3.8pt]{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 2};}}\quad\frac{\operatorname{\textit{d}_{\mathbb{t}}}(x_{n},x_{n+1})}{n}\,\xrightarrow{~n\to\infty~}\,0\,.\ ] ] in order to prove lemma [ lem : nonzerodrifttree ] , we pursue a two - step strategy . first , we show that every regular sequence converges to an end and , second , that the projections @xmath426 constitute a.s . a regular sequence .
concerning the first part , we pick an arbitrary regular sequence @xmath439 and claim that there is an end @xmath434 such that for every @xmath152 the open ball @xmath440 contains infinitely many @xmath441 .
assume there was no such end .
then , we know that for every @xmath434 there is an @xmath442 such that @xmath443 contains only finitely many @xmath441 , whence there is also an @xmath444 such that @xmath445 does not contain any @xmath441 at all . the open balls @xmath445 with @xmath434 and the singletons @xmath446 of vertices @xmath9 form an open covering of @xmath167 . by compactness , it contains a finite subcovering .
but since the constants @xmath444 have been chosen in such a way that the sequence @xmath439 does not enter any of the open balls @xmath445 , it must remain in a finite subset of the tree , which contradicts .
next , we pick such an end @xmath434 and claim that the sequence @xmath439 converges to @xmath447 .
let @xmath152 .
we define @xmath448 . by
, the constant @xmath449 is strictly positive .
moreover , all but finitely many @xmath450 are strictly greater than @xmath451 .
the elements in @xmath452 are characterised by the property that the paths starting in @xmath145 and representing them must have a certain finite initial piece .
let @xmath453 be the length of this piece , i.e. @xmath454 , see figure [ fig : jumpaway ] . by the above and by , there is an @xmath455 such that for all @xmath456 the inequalities @xmath457 and @xmath458 hold .
since @xmath452 contains infinitely many @xmath441 , we can even find an @xmath459 such that @xmath460 .
it turns out that not just for @xmath461 but for all @xmath462 we have @xmath463 .
indeed , if there was an @xmath462 such that @xmath463 and @xmath464 , we would know that @xmath465 the latter , of course , contradicts @xmath458 , see figure [ fig : jumpaway ] .
so , the two claims show that every regular sequence converges to an end .
concerning the second part , we aim to prove that @xmath416{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}\quad\liminf_{n\to\infty}\;\frac{|\pi_{\mathbb{t}}(z_{n})|_{\mathbb{t}}}{n}>0~~\text{a.\,s.}\quad\text{and}\quad{\tikz[baseline=-3.8pt]{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 2};}}\quad\frac{\operatorname{\textit{d}_{\mathbb{t}}}(\pi_{\mathbb{t}}(z_{n}),\pi_{\mathbb{t}}(z_{n+1}))}{n}\,\xrightarrow[\text{\raisebox{.5ex}{a.\,s.}}]{~n\to\infty~}\,0\,.\ ] ] recall from remark [ rem : multiple ] that not just @xmath231 and @xmath232 have finite first moment but also @xmath248 has .
so , the strong law of large numbers yields @xmath466{~n\to\infty~}\,|\mathbb{e}(\lambda(x_{1}))|=|\delta|>0\ ] ] which implies that @xmath467 next , let @xmath238 be the standard generating set .
the numerators @xmath468 of the fraction considered in are i.i.d . ,
and the first one satisfies @xmath469 so , again , by the strong law of large numbers @xmath470{~n\to\infty~}\,\mathbb{e}(\operatorname{\textit{d}_{\mathbb{t}}}(\pi_{\mathbb{t}}(z_{0}),\pi_{\mathbb{t}}(z_{1})))\,,\ ] ] from where a simple calculation yields . .
] ( 0,0 ) ( -2,92)@xmath447 ( -71,107)@xmath452 ( -26,88.5)@xmath471 ( -37,35)@xmath472 ( -143,38)@xmath473 ( -198,52)@xmath145 ( -127,71)@xmath453 ( -221.5,110)eventually left for the driftless case , the situation is not as easy and in order to show almost sure convergence of the projections @xmath426 to a random end , we restrict ourselves to the non - amenable subcase @xmath20 .
[ lem : zerodrifttree ] let @xmath20 .
assume that @xmath231 has finite first moment , @xmath232 has finite second moment , and there is an @xmath152 such that @xmath233 has finite @xmath372-th moment .
if there is no vertical drift , i.e. @xmath253 , then the projections @xmath426 converge a.s . to a random end @xmath434 .
again , we claim that the projections @xmath426 constitute a.s . a regular sequence .
but since @xmath253 , we need to modify the argument from the proof of lemma [ lem : nonzerodrifttree ] that showed . by assumption , @xmath38 is non - amenable and , in particular , the spectral radius @xmath474 of the random walk @xmath187 is strictly smaller than 1 , see e.g. ( * ? ? ?
* corollary 12.5 ) .
this , together with the fact that the random walk is uniformly irreducible yields that @xmath475 for a proof of this statement , see e.g. ( * ? ? ?
* proposition 8.2 ) . in order to estimate the numerators @xmath476 from above
, we apply an auxiliary result : _ there are @xmath477 such that for every element @xmath88 the inequality @xmath478 holds .
_ let us postpone the proof and record that , using this auxiliary result and lemma [ lem : zerodrifthyp ] , we obtain @xmath479 this is . concerning , note that the respective argument from the proof of lemma [ lem : nonzerodrifttree ] did not use the assumption that @xmath480 , and therefore does also works for @xmath253 .
so , we know that the projections @xmath426 constitute a.s . a regular sequence , which converges to a random end by the proof of lemma [ lem : nonzerodrifttree ] . it remains to show the auxiliary result . in order to do so
, we construct a path from @xmath65 to @xmath49 in the cayley graph @xmath36 with at most @xmath481 many edges , where the values of @xmath449 and @xmath482 are to be determined uniformly , i.e. not depending on @xmath49 .
first , we aim to adjust the tree component .
either combinatorially using the defining relation @xmath66 or geometrically using the properties of the cayley graph @xmath36 , we can find a path from @xmath65 to the coset @xmath51 with at most @xmath483 many edges .
let @xmath484 be the endpoint of such a path .
next , recall the notion of discrete hyperbolic plane from section [ sub : dhp ] .
we pick an arbitrary ascending doubly infinite path @xmath485 in the tree @xmath13 that traverses the vertex @xmath51 , consider the discrete hyperbolic plane @xmath128 , and take a shortest path from @xmath50 to @xmath49 in @xmath128 . by the proof of proposition [ prop : discretehyperbolicplane ]
, its length @xmath486 can be estimated from above by @xmath487 with @xmath488 .
we may continue this estimate and finally obtain @xmath489 & \leq\kappa\cdot\operatorname{\textit{d}_{\mathbb{h}}}(\pi_{\mathbb{h}}(h),\pi_{\mathbb{h}}(1))+\kappa\cdot\operatorname{\textit{d}_{\mathbb{h}}}(\pi_{\mathbb{h}}(1),\pi_{\mathbb{h}}(g ) ) \\ & \leq\kappa\cdot\max{\ensuremath{\{\ , \ell_{a},\ell_{b } \,\}}}\cdot\left(\left\lfloor\frac{q}{2}\right\rfloor+1\right)\cdot|\pi_{\mathbb{t}}(g)|_{\mathbb{t}}+\kappa\cdot\operatorname{\textit{d}_{\mathbb{h}}}(\pi_{\mathbb{h}}(1),\pi_{\mathbb{h}}(g))\,.\end{aligned}\ ] ] so , the concatenation of the two paths considered above has at most @xmath490 many edges with @xmath491 and @xmath492 . resuming sections [ sub : hyperbolicconvergence ] and [ sub : treeconvergence ] , we may formulate the following theorem .
[ thm : convergence ] let @xmath187 be a random walk on a non - amenable baumslag
solitar group @xmath225 with @xmath20 .
suppose that the increment @xmath231 has finite first moment . 1 .
if the vertical drift is * positive * , i.e. @xmath251 , then the projections @xmath349 converge a.s . to @xmath85 and the projections @xmath426 converge a.s .
to a random element @xmath434 .
2 . if the vertical drift is * negative * , i.e. @xmath252 , then the projections @xmath349 converge a.s . to a random element @xmath183 and the projections @xmath426
converge a.s . to a random element @xmath434 .
3 . if there is * no vertical drift * , i.e. @xmath253 , and , in addition , @xmath232 has finite second moment and there is an @xmath152 such that @xmath233 has finite @xmath372-th moment , which is certainly the case if the increment @xmath231 has finite @xmath372-th moment , then the projections @xmath349 have sublinear speed and the projections @xmath426 converge a.s . to a random element @xmath434 .
the boundaries @xmath16 and @xmath15 are endowed with their borel @xmath19-algebras @xmath493 and @xmath494 . under suitable assumptions on the moments ,
the projections @xmath349 and @xmath426 converge a.s . to a random element in the respective boundary
and we may consider the boundary maps @xmath495 and @xmath496 , defined almost everywhere , assigning to a trajectory @xmath497 the limits @xmath498 even though the boundary maps are only defined almost everywhere , they are measurable in the sense that the preimages of measurable sets are measurable . given @xmath499 and @xmath500 , we may construct their product map @xmath501 .
it is measurable with respect to the product @xmath19-algebra @xmath502 .
since both @xmath16 and @xmath15 are metrisable and separable topological spaces , it is not hard to see that the product @xmath19-algebra @xmath502 agrees with the borel @xmath19-algebra @xmath503 , see e.g. ( * ? ? ?
* appendix m.10 ) . in the following
, one should keep in mind that @xmath499 , @xmath500 , and @xmath504 are only defined if the respective projections @xmath349 and @xmath426 converge a.s . to a random element in the boundary .
the three pushforward probability measures @xmath505 , @xmath506 , @xmath507 on the measurable spaces @xmath508 , @xmath509 , @xmath510 are called the hitting measures . note that we may again , tacitly , complete the probability spaces with respect to @xmath511 , @xmath512 , @xmath513 .
each of the boundaries @xmath16 and @xmath15 is endowed with a left @xmath38-action .
the one on @xmath16 is induced by the action @xmath514 on @xmath14 and the one on @xmath15 is induced by the action @xmath515 on @xmath13 .
let us describe them in more detail .
the former is an action by isometries on @xmath14 , and in light of their classification mentioned in section [ sub : ptthp ] , we can also evaluate them at @xmath16 . for the latter , recall that ends are infinite reduced paths that start in @xmath145 .
the coordinatewise action on the ends maps every such path @xmath434 to some other path that need not start in @xmath145 any more .
the end @xmath516 is obtained by connecting @xmath145 with the initial vertex of this path and reduce the concatenation .
this way , it is not hard to see that we can map every @xmath434 to an end with an arbitrarily chosen finite initial piece.consider such a finite initial piece , i.e. a finite reduced path from @xmath145 to some vertex @xmath51 . given the end @xmath434 , we construct its image @xmath516 .
it will have the correct finite initial piece unless cancellation takes place .
but then , consider the image @xmath517 instead . since @xmath518 and @xmath519
, cancellation will take place in at most one of the two cases . ] in particular , every orbit @xmath520 is infinite and dense in @xmath15 . by changing the initial probability measure of the random walk @xmath245 as in section [ sub : pfbasoip ] , we obtain stationarity of the measure @xmath512 . more precisely , for every measurable set @xmath521 @xmath522 the same result holds true for @xmath16 and the product @xmath18 , which is endowed with the componentwise left @xmath38-action .
these observations will be helpful in a moment , when we show that the hitting measures are either dirac measures or non - atomic .
our proof is based on ( * ? ? ?
* lemma 3.4 ) , which is much more general . the original idea for our special case might be older .
[ lem : nonatomic ] the hitting measure @xmath512 is non - atomic .
moreover , if @xmath251 , then the hitting measure @xmath511 is the dirac measure at @xmath85 and , if @xmath252 , then @xmath511 is again non - atomic .
let us first consider the hitting measure @xmath512 .
suppose , there were elements of positive measure .
then , we may choose such an element @xmath434 with maximal measure @xmath25 .
in particular , for every element @xmath523 we know that @xmath524 .
we claim that for every @xmath523 the equality @xmath525 holds .
indeed , let us first suppose that there was an element @xmath526 with @xmath527 . then
, @xmath528 this is a contradiction . due to the irreducibility of the random walk
, @xmath529 does not only hold for all @xmath526 but inductively for all @xmath530 , which proves our claim .
but the orbit @xmath520 is infinite , so @xmath531 . and
this is , again , a contradiction . if @xmath251 , the result that the hitting measure @xmath511 is the dirac measure at @xmath85 is an immediate consequence of lemma [ lem : positivedrifthyp ] . on the other hand , if @xmath252 , then @xmath532 by lemma [ lem : negativedrifthyp ] .
now , we can repeat the above argument .
suppose , there was an element of positive measure .
then , we may choose such an element @xmath183 of maximal measure .
again , all elements in its orbit @xmath533 must have the same measure and , since the orbit is infinite , this yields a contradiction .
[ lem : support ] the hitting measure @xmath512 has full support , i.e. every non - empty open subset @xmath521 has positive measure .
moreover , if @xmath252 , the hitting measure @xmath513 on the cartesian product @xmath18 has full support .
we classify the ends of the tree @xmath13 according to which neighbour of the vertex @xmath145 they first traverse .
this yields a partition of the space of ends into @xmath64 open subsets , i.e. the open balls of radius 1 .
at least one of them must have positive measure , call it @xmath534 .
now , let @xmath521 be an arbitrary non - empty open subset . in particular , there is a vertex @xmath51 , such that all ends traversing @xmath51 belong to @xmath535 .
similarly to the argument given in footnote [ fn : endcorrection ] , either @xmath536 or @xmath537 is contained in @xmath535 , see figure [ fig : hittingtree ] .
we may assume w.l.o.g .
that @xmath538 .
moreover , due to the irreducibility of the random walk , there is an @xmath162 such that @xmath539 , i.e. the probability to reach @xmath49 in exactly @xmath163 steps is positive .
hence , @xmath540 the proof of the second assertion is similar .
since we have to keep track of two components , it is slightly more technical so that we give only a proof sketch .
recall from above the set @xmath534 .
given the random variable @xmath500 takes a value in @xmath541 , at least one open interval @xmath542 with @xmath116 will be hit by the random variable @xmath499 with positive probability , call it @xmath543 .
so , we know that @xmath544
. now , let @xmath545 be an arbitrary non - empty open subset . by definition of the product topology ,
the open set @xmath535 contains a rectangle of open sets @xmath546 and @xmath547 .
we seek to construct an element @xmath530 such that both @xmath548 and @xmath549 , from where we may finally conclude as above that @xmath550 .
again , there is a vertex @xmath51 of the tree , such that all ends traversing @xmath51 belong to @xmath551 .
moreover , there are an @xmath371 and an @xmath152 such that the open interval @xmath552 is contained in @xmath553 . based on this data
, we will find an element @xmath530 of the form @xmath554 with the desired properties .
let us first look at the tree component .
the exponent @xmath555 is either @xmath354 or @xmath65 , whichever ensures that the reduced path from @xmath145 to @xmath556 traverses @xmath51 .
now , let us turn to the hyperbolic component .
the image @xmath557 is a bounded open interval .
the exponent @xmath558 is chosen in such a way that the length of the image @xmath559 is at most @xmath560 .
finally , there is an integer @xmath116 such that both images @xmath561 and @xmath562 are contained in the interval @xmath552 and therefore both belong to @xmath553 .
the exponent @xmath563 will be either @xmath111 or @xmath564 .
let us return to the tree and choose it in such a way that all ends in the image @xmath565 traverse @xmath566 .
then , by construction , they also traverse the vertex @xmath51 and belong to @xmath551 .
has full support . ]
( 0,0 ) ( -82,60)@xmath145 ( -56,35)@xmath51 ( 1,65)@xmath535 ( -161,120)@xmath541 ( -154,41)positive ( -150,29)measure [ thm : identification ] let @xmath187 be a random walk on a non - amenable baumslag solitar group @xmath225 with @xmath20 and suppose that the increment @xmath231 has finite first moment . 1 . if the vertical drift is * positive * , i.e. @xmath251 , then the poisson furstenberg boundary is isomorphic to @xmath567 endowed with the boundary map @xmath496 .
2 . if the vertical drift is * negative * , i.e. @xmath252 , then the poisson furstenberg boundary is isomorphic to @xmath568 endowed with the boundary map @xmath501 .
3 . if there is * no vertical drift * , i.e. @xmath253 , and , in addition , @xmath232 has finite second moment and there is an @xmath152 such that @xmath233 has finite @xmath372-th moment , which is certainly the case if the increment @xmath231 has finite @xmath372-th moment , then the poisson furstenberg boundary is again isomorphic to @xmath567 endowed with the boundary map @xmath496 .
[ rem : realnumbers ] in the case of negative vertical drift , we could also have used @xmath569 instead of @xmath18 .
as already mentioned , we seek to apply the strip criterion , see theorem [ thm : strip ] . by lemma [ lem : entropy ] , the probability measure @xmath10 driving the random walk has finite entropy .
moreover , it is not hard to see that @xmath16 and @xmath15 are polish spaces , and so is their product @xmath18 .
therefore , by remark [ rem : polish ] , the probability spaces @xmath567 and @xmath568 are lebesgue rohlin spaces . they are endowed with a left @xmath38-action and boundary maps @xmath496 and @xmath501 , defined almost everywhere . in order to show that they are @xmath10-boundaries , we have to ensure that the boundary maps are measurable , @xmath269-invariant , and @xmath38-equivariant .
but all three properties are immediate by construction , compare also ( * ? ? ?
* end of 1.5 ) .
if the vertical drift is negative , i.e. @xmath252 , let us take the @xmath10-boundary @xmath568 and the @xmath308-boundary @xmath570 . here
, @xmath571 denotes the hitting measure of the pointwise projection of the random walk @xmath572 driven by the reflected probability measure @xmath308 to the bass serre tree @xmath13 .
next , we need to define gauges and strips .
let @xmath238 be the standard generating set and define gauges @xmath573 .
in other words , the gauges exhaust the group @xmath38 with balls centred at the identity @xmath198 , and the gauge function @xmath310 is nothing but the distance to @xmath65 with respect to the word metric @xmath220 . by lemma [ lem : nonatomic ] , we know that @xmath574-almost every pair of points @xmath575 has distinct ends @xmath576 and a boundary value @xmath577 . in this situation , we may connect @xmath578 and @xmath579 by a unique doubly infinite reduced path @xmath580 and define the strip @xmath581 as follows .
it consists of all @xmath88 that are contained in the full @xmath119-preimage of @xmath120 , i.e. their image @xmath121 is traversed by @xmath122 , and have the property that the real part @xmath228 has minimal distance to @xmath577 among all real parts @xmath582 with @xmath484 , see the left - hand side of figure [ fig : strips ] . to all remaining pairs
we assign the whole of @xmath38 as a strip .
this way , the map @xmath311 becomes measurable and @xmath38-equivariant . by lemma [ lem : support ]
, a random strip contains the identity element @xmath198 with positive probability , i.e. the map @xmath311 satisfies the inequality of remark [ rem : removeg ] .
so , it suffices to verify the following convergence for an arbitrary pair @xmath575 with distinct ends @xmath576 and @xmath577 , ( left ) and @xmath253 ( right ) . ] ( 0,0 ) ( -286,-7)@xmath577 @xmath583{~n\to\infty~}\,0\,.\ ] ] but the strip @xmath581 intersects the gauge @xmath584 in at most @xmath585 many cosets of the form @xmath60 , and each of them contains at most two elements of the strip .
therefore , @xmath586{~n\to\infty~}\,0\,.\ ] ] in the final step of the above calculation , we use again that the increments @xmath231 have finite first moment .
indeed , @xmath587{~n\to\infty~}\,\mathbb{e}(\operatorname{\textit{d}_{s}}(1,x_{1}))\,,\ ] ] from where we may first conclude that the sequence @xmath588 is a.s . bounded and second that the sequence @xmath589 converges a.s .
so , we can finally apply the strip criterion and obtain that @xmath568 is isomorphic to the poisson furstenberg boundary .
vice versa , if the vertical drift is positive , i.e. @xmath251 , the same argument yields that @xmath567 is isomorphic to the poisson furstenberg boundary .
it remains to consider the driftless case , i.e. @xmath253 .
then , both @xmath10 and @xmath308 are driftless and there is no natural candidate for a real number that determines the horizontal position of the strip .
but the fact that the projections @xmath349 have sublinear speed allows us to solve this issue .
more precisely , take the @xmath10-boundary @xmath567 and the @xmath308-boundary @xmath570 .
now , define gauges @xmath590 again , we know that @xmath591-almost every pair of points @xmath592 has distinct ends @xmath576 , which we may connect by a unique doubly infinite reduced path @xmath580 .
let @xmath593 be the full @xmath119-preimage of @xmath120 , i.e. the set of all group elements @xmath88 such that the image @xmath121 is traversed by @xmath122 , see the right - hand side of figure [ fig : strips ] .
again , to all remaining pairs we assign the whole of @xmath38 as a strip .
this way , the map @xmath311 becomes measurable , @xmath38-equivariant , and satisfies the inequality of remark [ rem : removeg ] .
now , pick an arbitrary pair @xmath592 with distinct ends @xmath576 .
we claim that @xmath594{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}}+\underbrace{\frac{|z_{n}|+2}{n}}_{{\tikz[baseline=-3.8pt]{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 2};}}}\,.\ ] ] indeed , the inequality holds for a similar reason as above ; the strip @xmath593 intersects the gauge @xmath584 in at most @xmath595 many cosets of the form @xmath60 .
slightly more involved is the observation that each of them contains at most @xmath596 many elements of the gauge . fix a coset @xmath51 .
the projections @xmath139 of the elements @xmath484 are located on the horizontal line @xmath597 with imaginary part @xmath598 .
one necessary condition for such an element @xmath484 to be contained in the gauge @xmath584 is that the projection @xmath139 is contained in the closed disc @xmath599 . if @xmath600 is empty , then the coset @xmath51 does not contain any element of the gauge and we are done
otherwise , there is a unique @xmath100 with @xmath601 such that @xmath600 is the horizontal line between @xmath602 and @xmath603 , see figure [ fig : shortline ] .
the projections @xmath139 with @xmath484 have the property that the real parts @xmath582 and @xmath604 differ exactly by @xmath148 .
so , the horizontal line @xmath600 contains at most @xmath605 many of them .
let us now estimate @xmath605 in terms of @xmath606 .
since @xmath607 and @xmath608 are both contained in @xmath609 , their distance is at most @xmath610 .
therefore , @xmath611 and , in particular , @xmath612 so , the coset @xmath51 contains strictly fewer than @xmath596 elements of the gauge .
we will now show that both summands and converge a.s . to @xmath354 , which will complete the proof .
let us first observe that @xmath613 concerning , we deduce from ( [ eqn : gaugefunctionestimate ] ) and the proof of proposition [ prop : discretehyperbolicplane ] that @xmath614 , and finally obtain by the same argument as above @xmath416{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 1};}}=\frac{\ln\left(2\cdot|z_{n}|^{2}+1\right)}{n}\leq\frac{\ln\left(2\cdot(\max{\ensuremath{\{\ , \ell_{a},\ell_{b},1 \,\}}}\cdot\operatorname{\textit{d}_{s}}(1,z_{n})+1)^{2}+1\right)}{n}\,\xrightarrow[\text{\raisebox{.5ex}{a.\,s.}}]{~n\to\infty~}\,0\,.\ ] ] on the other hand , concerning , we apply ( [ eqn : gaugefunctionestimate ] ) and lemma [ lem : zerodrifthyp ] to obtain @xmath416{\node[shape = circle , draw , inner sep=0.5pt ] ( char ) { \scriptsize 2};}}=\frac{|z_{n}|+2}{n}\leq\frac{\max\left\{\,\operatorname{\textit{d}_{\mathbb{h}}}(\pi_{\mathbb{h}}(1),\pi_{\mathbb{h}}(z_{n})),\sqrt{\operatorname{\textit{d}_{s}}(1,z_{n})}\,\right\}+3}{n}\,\xrightarrow[\text{\raisebox{.5ex}{a.\,s.}}]{~n\to\infty~}\,0\,.\ ] ] , the closed disc @xmath609 , and their intersection @xmath600 . ]
( 0,0 ) ( -114,33)@xmath615 ( -77,58)@xmath148 ( -2,65)@xmath401 ( -89,25)@xmath609 ( -195,86)@xmath616 ( -97,86)@xmath617
recall from section [ sub : aobsg ] that a baumslag
solitar group @xmath2 is non - amenable if and only if neither @xmath5 nor @xmath6 . until now
, we have only identified the poisson furstenberg boundary for random walks on non - amenable baumslag
solitar groups @xmath2 with @xmath20 .
replacing one of the two generators by its inverse , it is easy to see that @xmath618 so , in order to investigate the remaining non - amenable baumslag solitar groups , the only cases that we have to consider are @xmath86 and @xmath619 . in this appendix
, we shall review our methods from the previous sections and explain how to adjust the arguments in order to obtain similar results for these cases .
let us first assume that @xmath225 with @xmath86 .
in order to define the projection @xmath87 back in section [ sub : ptthp ] , we considered the map @xmath77 given by @xmath78 and @xmath79 , and extended it to a homomorphism .
now , we are assuming that @xmath86 , in which case the transformation @xmath620 is not orientation preserving any more .
if we replaced @xmath1 by @xmath621 in the definition , then @xmath622 would be orientation preserving but @xmath623 , whence we could not apply von dyck s theorem any more .
so , we have to change the approach .
let @xmath624 be the set of all maps @xmath625 either of the form @xmath626 or of the form @xmath627 with @xmath74 and @xmath75 .
this set endowed with the composition forms again a group .
consider the map @xmath628 given by @xmath629 and @xmath630 . with this map ,
it is possible to apply von dyck s theorem and to extend it uniquely to a group homomorphism @xmath631 . finally , as in the case of @xmath71 , every @xmath632 can be thought of as an isometry of @xmath14 .
so , we may again consider the projection @xmath87 given by @xmath633 .
the following lemma illustrates this definition .
[ lem : lookingfromfrontnew ] for every @xmath88 the point @xmath91 is above the point @xmath92 ; the two points have the same real part and their distance is @xmath634 .
but for every @xmath88 the point @xmath94 is either right or left from the point @xmath92 depending on whether the level @xmath234 is even or odd ; in any case , the two points have the same imaginary part and their distance is @xmath635 .
the proof is similar to the one of lemma [ lem : lookingfromfront ] .
so , we only discuss the differences .
let us consider the two points @xmath99 and @xmath636 .
if @xmath637 is of the form @xmath638 , then it is again the composition of a dilation @xmath639 and a translation @xmath104 , whence the relative position of the two points is preserved . on the other hand , if @xmath637 is of the form @xmath640 , then it is the composition of a reflection at the imaginary axis @xmath641 , a dilation @xmath639 , and a translation @xmath104 , in which case the relative position of the two points is still preserved with the exception that right and left are switched . in order to decide whether @xmath637 is of the first or the second form , we can write the element @xmath88 as a product over @xmath642 and @xmath643 . since @xmath631 is a homomorphism ,
the image @xmath637 can be written as the respective product over @xmath644 and @xmath645 . but
each occurrence of @xmath644 yields one reflection .
so , the image @xmath637 is of the first form if and only if the number of occurrences of @xmath644 is even , which is the case if and only if @xmath234 is even . using this projection @xmath87 and , of course , replacing @xmath1 by @xmath621 wherever it is necessary
, we can repeat the arguments from the previous sections .
the definition of the tree @xmath13 and the level functions @xmath58 and @xmath63 , including lemma [ lem : neighbours ] , as well as the definition of the discrete hyperbolic plane @xmath128 , including proposition [ prop : discretehyperbolicplane ] , can be adapted .
recall that , in section [ sub : rpipavd ] , we considered the imaginary and real parts of @xmath92 separately , and introduced the shorthand notation @xmath347 and @xmath348 .
let us highlight that we have @xmath646 , which allows us to adapt the proof of lemma [ lem : assumptions ] . in order to identify the poisson furstenberg boundary geometrically
, we have to ensure convergence to the boundaries @xmath16 and @xmath15 .
let us first consider the boundary @xmath16 .
the proof of lemma [ lem : positivedrifthyp ] for @xmath251 can be adapted .
the proof of lemma [ lem : negativedrifthyp ] for @xmath252 , in turn , deserves a bit of work .
we have to show that the real parts @xmath355 converge a.s . to a random element @xmath183 .
in the original proof , we observed that @xmath647 and @xmath648 with @xmath360 .
while the first formula remains true , the second one does not .
we are now in a situation where not only the scaling but also the direction of the next horizontal increment depends on the current level .
however , instead of the above , we obtain that @xmath649 with @xmath650 if @xmath234 is even and @xmath651 if @xmath234 is odd .
this observation allows us to apply cauchy s root test precisely as in the proof of lemma [ lem : negativedrifthyp ] . for the same reason ,
namely because all the estimates are not in terms of the actual horizontal increments but of their absolute values , the proofs of lemmas [ lem : zerodrifthyp ] and [ lem : elie ] for @xmath253 can be adapted .
the same holds , concerning the boundary @xmath15 , for the proofs of lemma [ lem : nonzerodrifttree ] for @xmath480 and lemma [ lem : zerodrifttree ] for @xmath253 . from these observations
, we may deduce the following results . 1 .
if the vertical drift is * positive * , i.e. @xmath251 , then the projections @xmath349 converge a.s . to @xmath85 and the projections @xmath426 converge a.s . to a random element @xmath434 .
2 . if the vertical drift is * negative * , i.e. @xmath252 , then the projections @xmath349 converge a.s . to a random element @xmath183 and the projections @xmath426
converge a.s . to a random element @xmath434 .
3 . if there is * no vertical drift * , i.e. @xmath253 , and , in addition , @xmath232 has finite second moment and there is an @xmath152 such that @xmath233 has finite @xmath372-th moment , which is certainly the case if the increment @xmath231 has finite @xmath372-th moment , then the projections @xmath349 have sublinear speed and the projections @xmath426 converge a.s . to a random element @xmath434 . 1 .
if the vertical drift is * positive * , i.e. @xmath251 , then the poisson furstenberg boundary is isomorphic to @xmath567 endowed with the boundary map @xmath496 .
if the vertical drift is * negative * , i.e. @xmath252 , then the poisson furstenberg boundary is isomorphic to @xmath568 endowed with the boundary map @xmath501 .
3 . if there is * no vertical drift * , i.e. @xmath253 , and , in addition , @xmath232 has finite second moment and there is an @xmath152 such that @xmath233 has finite @xmath372-th moment , which is certainly the case if the increment @xmath231 has finite @xmath372-th moment , then the poisson furstenberg boundary is again isomorphic to @xmath567 endowed with the boundary map @xmath496 .
let us now assume that @xmath225 with @xmath619 .
this situation differs fundamentally from the ones discussed so far because the bricks introduced in the proof of proposition [ prop : discretehyperbolicplane ] would now have equally many @xmath14-vertices on their upper and lower level .
therefore , we use the euclidean plane @xmath652 instead of the hyperbolic plane @xmath14 . in order to construct a projection @xmath653 ,
let @xmath654 and consider the map @xmath628 given by @xmath655 ( ( x , y)\mapsto(-x , y+1))&\text{if}~q<0 \end{array}\right.\quad\text{and}\quad\pi_{{\ensuremath{m}}}(b):=((x , y)\mapsto(x+1,y))\,.\ ] ] in both cases , i.e. @xmath656 and @xmath657 , it is possible to apply von dyck s theorem and to extend the map uniquely to a group homomorphism @xmath658 .
now , we may consider the projection @xmath653 given by @xmath659 .
the definition of the tree @xmath13 and the level functions @xmath58 and @xmath63 , including lemma [ lem : neighbours ] , remain the same . but instead of the discrete hyperbolic plane , we now obtain a discrete euclidean plane @xmath128 . the proof of proposition [ prop : discretehyperbolicplane ] can be adapted to the new situation and shows that the graph @xmath128 endowed with the graph distance @xmath124 is quasi - isometric , and even bi - lipschitz , to the euclidean plane @xmath652 endowed with the standard metric @xmath660 .
we aim to show that , as soon as the projections converge to a random element in @xmath15 , independently of the vertical drift , the poisson furstenberg boundary is isomorphic to @xmath567 .
in particular , we do not need to introduce any boundary to capture the behaviour of the projections @xmath661 . concerning the projections @xmath426 , we distinguish between two cases . if the vertical drift is different from @xmath354 , i.e. @xmath480 , then the proof of lemma [ lem : nonzerodrifttree ] can be adapted and we obtain that the projections @xmath426 converge a.s . to a random end @xmath434 .
but if there is no vertical drift , i.e. @xmath253 , then the proof of lemma [ lem : zerodrifttree ] can not be adapted because it was based on the fact that the projections @xmath349 had sublinear speed ; and the projections @xmath661 do not need to have sublinear speed any more .
in this situation , the following lemma may be used instead of lemma [ lem : zerodrifttree ] .
let @xmath187 be a random walk on an arbitrary non - amenable baumslag
solitar group @xmath225 .
if the increment @xmath231 has finite support , then the projections @xmath426 converge a.s . to a random end @xmath434 .
recall from the beginning of section [ sub : treeconvergence ] that the projections @xmath426 do not need to satisfy the markov property .
despite of this , we first show that they leave a.s .
every finite ball with centre @xmath145 , i.e. every set of vertices of the form @xmath662 .
suppose they did not .
then , there is a ball such that the probability to visit this ball infinitely often is strictly positive .
now , it is not hard to see that the probability to visit the centre of this ball infinitely often is also strictly positive . in other words , @xmath663 but for every @xmath368
we obtain @xmath664 therefore , @xmath665 .
we may read the random sequence @xmath666 with @xmath667 as the left random walk on the left homogeneous space @xmath60 driven by the reflected probability measure @xmath308 . by the above
, @xmath401 is recurrent .
but since the group @xmath38 is non - amenable and the subgroup @xmath668 is amenable , one can show that the homogeneous space @xmath60 is non - amenable , see @xcite referring to ( * ? ? ?
* , exemple 2b ) . therefore , by ( * ? ? ? * satz ii.9 ) , @xmath401 must be transient , which is a contradiction .
now , we can repeat the argument given in the proof of lemma [ lem : nonzerodrifttree ] to show that for almost every trajectory @xmath368 there is an end @xmath669 such that for every @xmath152 the projections @xmath670 visit the open ball @xmath671 infinitely often .
we know that for almost every trajectory @xmath368 both the projections @xmath670 leave every finite ball with centre @xmath145 and the graph distance of any two subsequent projections @xmath670 and @xmath672 is bounded by a constant . hence , for every @xmath152 the projections
@xmath670 must eventually remain in the open ball @xmath673 .
in other words , they converge to @xmath669 .
now , we can show as in lemmas [ lem : nonatomic ] and [ lem : support ] that the hitting measure @xmath674 is again non - atomic and has full support .
moreover , we obtain the following version of the identification theorem . 1 .
if the * vertical drift * is different from @xmath354 , i.e. @xmath480 , then the poisson furstenberg boundary is isomorphic to @xmath567 endowed with the boundary map @xmath496 .
if there is * no vertical drift * , i.e. @xmath253 , and the increment @xmath231 has not just finite first moment but also finite support , then the poisson furstenberg boundary is again isomorphic to @xmath567 endowed with the boundary map @xmath496 .
again , we apply the strip criterion . as in the original proof of the identification theorem
, we take the @xmath10-boundary @xmath567 and the @xmath308-boundary @xmath570 .
next , we define gauges @xmath675 we know that @xmath591-almost every pair of points @xmath592 has distinct ends @xmath576 , which we may connect by a unique doubly infinite reduced path @xmath580 .
let @xmath593 be the full @xmath119-preimage of @xmath120 . to all remaining pairs
we assign the whole of @xmath38 as a strip .
this way , the map @xmath311 becomes measurable , @xmath38-equivariant , and satisfies the inequality of remark [ rem : removeg ] .
now , pick an arbitrary pair @xmath592 with distinct ends @xmath576 .
we claim that @xmath676 indeed , the inequality is easy to see .
the strip @xmath593 intersects the gauge @xmath584 in at most @xmath585 many cosets of the form @xmath60 , and each of them contains at most @xmath585 many elements of the gauge .
now , it suffices to consider the standard generating set @xmath238 and to observe that @xmath677 .
then , using the fact that @xmath588 is a.s .
bounded , we may conclude that @xmath678{~n\to\infty~}\,0\,,\ ] ] which allows us to apply the strip criterion .
to4em , _ poisson furstenberg boundaries , large - scale geometry and growth of groups _ , proceedings of the international congress of mathematicians , vol .
ii , hindustan book agency , new delhi , 2010 , pp . 681704 .
to4em , _ boundary theory and stochastic processes on homogeneous spaces _ , harmonic analysis on homogeneous spaces ( proc .
pure math .
xxvi , williams coll . ,
williamstown , mass . , 1972 ) ,
soc . , providence , r.i .
, 1973 , pp .
193229 .
to4em , _ boundaries of invariant markov operators : the identification problem _ ,
ergodic theory of @xmath680 actions ( warwick , 19931994 ) , london math . soc .
lecture note ser .
228 , cambridge univ .
press , cambridge , 1996 , pp . | the aim of this paper is to give a description of the poisson furstenberg boundary of random walks on non - amenable baumslag solitar groups . after a short introduction to baumslag
solitar groups and their geometry , we change our focus to random walks on these groups .
the poisson furstenberg boundary is a probabilistic model for the long - time behaviour of random walks . for random walks on non - amenable baumslag
solitar groups we identify the poisson furstenberg boundary in terms of the boundary of the hyperbolic plane and the space of ends of the associated bass serre tree using kaimanovich s strip criterion .
the precise statement can be found in theorem [ thm : identification ] on page .
* keywords : * random walk , poisson furstenberg boundary , strip criterion , baumslag solitar group , compactification .
* msc 2010 classes : * 60j50 ( primary ) , 60g50 , 20f65 , 50c63 . |
most of our knowledge of the universe at far infrared wavelengths has been obtained with photoconductive detectors , particularly as used in the infrared astronomy satellite ( iras ) and the infrared space observatory ( iso ) .
these detectors have been selected because they provide excellent performance at relatively elevated operating temperatures ( compared with those needed to suppress thermal noise in bolometers ) .
similar considerations led to development of high performance photoconductor arrays for the multiband imaging photometer for spitzer ( mips ) , namely a ge : ga array and a stressed ge : ga array operating at 70 and 160 respectively which were built at the university of arizona .
to provide complementary measurements at 24 , the instrument also includes a si : as blocked impurity band ( bib ) array , built at boeing north america ( bna ) under contract to the infrared spectrograph ( irs ) team . in the bib or impurity band conduction ( ibc ) architecture ,
the high impedance required to minimize johnson noise is provided by a thin , high - purity layer of silicon .
the infrared absorption occurs in a second layer , which can be relatively strongly doped . due to the separation of these two functions
, the detector layers can be optimized separately .
thus , these devices can be designed and built to have fast response , high resistance to cosmic ray irradiation induced responsivity shifts , high quantum efficiency , and good photometric behavior .
because the processing necessary for high performance silicon ibc devices has only relatively recently become possible there is relatively little experience with them in space astronomy missions .
an early - generation detector array was used in the short wavelength spectrometer ( sws ) in iso @xcite .
initially , the device showed degradation due to damage by large ionizing particle exposures when the satellite passed through the trapped radiation belts .
once the operating conditions were adjusted to minimize these effects , the sws detectors showed the expected virtues of this type of device even though they did not achieve their preflight sensitivity expectations @xcite . at wavelengths
longer than 40 , photoconductors are typically built in germanium because of the availability of impurity levels in this material that are much more shallow than those in silicon . achieving the appropriate structure and
simultaneously the stringent impurity control for germanium ibc devices has proven difficult . as a result
, all far infrared space astronomy missions have used simple bulk photoconductors .
mips uses bulk gallium doped germanium ( ge : ga ) detectors , both unstressed and stressed .
in such detectors , the same volume of material determines both the electrical and photo - absorptive properties , making the optimization less flexible than with si ibc detectors .
consequently , their behavior retains some undesirable properties that can be circumvented through the more complex architecture of ibc devices . nonetheless , generally satisfactory performance is possible and has been achieved in past space astronomy missions .
the 60 and 100 bands in iras @xcite utilized 15 ge : ga photoconductors each . the detectors were read out with transimpedance amplifiers that used junction field effect transistor ( jfet ) first stages mounted in a way that isolated them thermally .
this allowed these transistors to be heated , resulting in low noise and stable operation @xcite .
detector calibration was maintained by flashing reverse bolometer stimulators mounted in the center of the telescope secondary mirror , and cosmic ray effects were erased by boosting the detector bias to breakdown @xcite .
the intrinsic performance of the detectors was limited by the johnson noise of the transimpedance amplifier ( tia ) feedback resistors and by other noise sources associated with the readout .
the in - flight performance was similar to expectations from pre - flight calibrations .
the isophot instrument @xcite carried a @xmath0 array of unstressed ge : ga detectors operating from 50 to 105 and a @xmath1 array of stressed devices operating from 120 to 200 .
the readout was by a capacitive transimpedance metal oxide semiconductor field - effect transistor ( mosfet ) amplifier whose processing had been adjusted to improve its performance at low temperatures @xcite .
calibration was assisted with a stimulator built into the instrument , which could be viewed by adjusting the position of a scan mirror . in practice ,
the unstressed focal plane never achieved the performance level anticipated from laboratory measurements of its noise equivalent power ( nep ) .
the performance of the stressed devices was substantially better , due in part to the relatively large fast response component of these devices ( compared with the slow component ) and their better thermal isolation from the readout amplifiers .
the long wavelength spectrometer ( lws ) instrument on iso @xcite used a single ge : be detector , five ge : ga detectors , and four stressed ge : ga detectors .
the readouts were based on jfets , mounted with thermal isolation and heated to a temperature where they operated with good stability and low noise .
the readout circuit was an integrating source follower and neps of @xmath2 were measured in the laboratory @xcite .
calibration was assisted with built in stimulators that were flashed between spectral scans .
on orbit , it was found that frequent small glitches , probably associated with cosmic ray hits , limited the maximum integration times to shorter values than had been anticipated and also required a lower operating voltage @xcite . with these mitigations ,
the neps were found to be @xmath34 higher in orbit than expected from ground test data @xcite . in mips ,
the ge : ga detectors are carefully isolated thermally from their readouts and operated at sufficiently cold temperatures that their dark currents are low and stable .
the mosfet - based readouts use a specialized foundry process that provides them with good dc stability even at the low operating temperature of @xmath31.5k .
this feature , combined with the capacitive tia ( ctia ) circuit , maintains the detector bias accurately . a scan mirror ( based on a design provided by t. de graauw ) modulates the signals on a pixel so measurements can be obtained from the relatively well - behaved @xcite fast component of the detector response .
responsivity variations are tracked with the aid of frequent stimulator flashes . finally , the instrument operations force observers to combine many short observations of a source into a single measurement .
the high level of redundancy in the data helps identify outlier signals and also improves the calibration by simple averaging over variations .
the efficacy of this operational approach is confirmed by the on - orbit results .
details on the design and construction of mips can be found in @xcite .
the inflight performance of mips is described by @xcite .
this paper describes the approaches for reduction and calibration of the mips data .
section [ sec_challenge ] details the challenges of using si and ge detectors in a space astronomy mission .
section [ sec_design ] gives a summary of the design and operational features of mips that address these challenges .
section [ sec_overview ] gives an overview of the three stages of mips data reduction .
these stages are discussed in more detail in the following three sections .
section [ sec_ramps_to_slopes ] details the processing steps to turn the integration ramps into measured slopes .
section [ sec_slope_cal ] discusses the corrections to transform the slopes into calibrated fluxes . section [ sec_redund ] gives a brief overview of the use of the inherent redundancy in the observations to further improve the reduction .
section [ sec_inflight ] gives the results of initial testing of these reduction techniques with flight data .
finally , section [ sec_summary ] provides a summary .
at high backgrounds , such as might be encountered in an airborne instrument , far infrared photoconductors behave relatively well , with rapid adjustment of the detector resistance appropriate to a change in illumination level . as the background is decreased , the adjustment to equilibrium levels occurs in a multistep process with multiple associated time constants as discussed below .
thus , the detectors can be used in a straightforward manner at high backgrounds but precautions must be taken at low ones to track the calibration . for a more detailed discussion
see @xcite . the fast response component in these detectors results from the current conducted within the detector volume associated with the drift of charge carriers freed by absorption of photons .
the speed of this component is controlled by the propagation of a zone boundary with drift velocity @xmath4 , so that the time constant is given by the free carrier lifetime divided by the photoconductive gain .
this time is very fast ( microseconds or shorter ) in comparison to normal detection standards .
however , as charge moves within the detector , the electrical equilibrium must be maintained .
for example , charge carriers generated by photoionization are removed from the detector when they drift to a contact .
they are replaced by injection of new charge carriers from the opposite contact , but the necessity for new charge can only be conveyed across the detector at a characteristic time proportional to the `` dielectric relaxation time '' , basically its capacitive or @xmath5 time constant : @xmath6 here , @xmath7 is the dielectric constant of the material and @xmath8 is the mobility for the charge carrier of interest , @xmath9 is the permittivity of free space , @xmath10 is the density of free carriers , and @xmath11 is the charge of the electron .
the slow response components arise from this phenomenon .
the form of this time constant makes explicit the dependence on illumination level through the density of free charge carriers , @xmath12 . in fully illuminated detectors ( for example , the integrating cavities used for the 160 array ) and at the low backgrounds appropriate for space - borne operation , @xmath13 can be tens of seconds . in transverse contact detectors , such as those used for the mips 70 array ,
the part of the detector volume near the injecting contact may be poorly illuminated and have large resistance .
the detector therefore adjusts to a new equilibrium only at the large dielectric time constant of this layer , which can be hundreds of seconds at low backgrounds .
the initial shift of charge in the detector can set up a space charge that reduces the field in the bulk of the device , leading to a reduction of responsivity following the initial fast response . from its appearance on a plot of response versus time ,
this behavior is described as `` hook '' response .
as the field is restored at a characteristic rate of @xmath13 , the response grows slowly to a new equilibrium value .
see @xcite for detailed modeling of these effects . in space applications ,
ionizing particles such as cosmic rays also affect the calibration of these detectors .
the electrons freed by a cosmic ray hit can be captured by ionized minority impurities , reducing the effective compensation and increasing the responsivity .
the shifts in detector characteristics can be removed by warming it to a temperature that re - establishes thermal equilibrium , and then cooling it back to proper operating conditions . between such anneal cycles , the responsivity needs to be tracked to yield calibrated data .
all successful uses of far infrared photoconductors at low backgrounds have included local relative calibrators of reverse bolometer design that allow an accurately repeatable amount of light to be put on the detector .
these stimulators allow frequent measurement of the relative detector responsivity .
in general , this strategy is most successful when the conditions of measurement are changed the least to carry out the relative calibration .
the mips instrument includes such calibrators , which are flashed approximately every two minutes . based upon data obtained at a proton accelerator and in space ,
the average increase in response over a two minute period in the space environment can be 0.5% to 1% , so the calibration interval allows tracking the response accurately .
although the detectors in the silicon array are expected to perform well photometrically , the array as a system shows a number of effects that must be removed to obtain calibrated data .
the array is operated well below the freezeout temperature for the dopants in the silicon readout ( the readout circuit uses a different foundry process from that developed for the ge detectors ) .
therefore , the array must be operated in a continuous read mode to avoid setting up drifts in the outputs that would degrade the read noise .
the flight electronics and software are designed to maintain a steady read rate of once per half mips second ( see [ sec_data_collection ] ) . when the array is first turned on
, the transient effects of the readout cause a slow drift in the outputs .
much of this effect can be removed by annealing the array , which is the standard procedure for starting the mips 24 operations .
the array shows an effect termed `` droop . ''
the output of the device is proportional to the signal it has collected , plus a second term that is proportional to the average signal over the entire array .
in addition , the 24 array has a number of smaller effects ( e.g. , rowdroop , electronic nonlinearities , etc . ) which are described later in this paper .
the design and operation of mips is summarized here , paying special attention to those areas which answer the challenges outlined above and , therefore , produce data that can be reduced successfully .
mips has three instrument sections , one for 24 imaging , one for 70 imaging and low resolution spectroscopy , and one for 160 imaging .
light is directed into the three sections off a single axis scan mirror .
the 24 section uses a @xmath14 pixel si : as ibc array and operates in a fixed broad spectral band extending from 21 to about 27 ( the long wavelength cutoff is determined by the photo - absorptive cutoff of the detector array ) . after
light enters this arm of the instrument from a pickoff mirror , it is brought to a pupil on a facet of the scan mirror .
it is reflected off this mirror into imaging optics that relay the telescope focal plane to the detector array at a scale of @xmath15 per pixel corresponding to a @xmath16 sampling of the point spread function , where @xmath17 is the telescope aperture .
the surface area of a 24 pixel is @xmath18 @xmath19 .
the field of view provided by this array is @xmath20 .
a reverse bolometer stimulator in this optical train allows relative calibration signals to be projected onto the array .
the scan mirror allows images to be dithered on the array without the overheads associated with moving and stabilizing the spacecraft .
it also enables an efficient mode of mapping ( scan mapping ) in which the spacecraft is scanned slowly across the sky and the scan mirror is driven in a sawtooth waveform that counters the spacecraft motion , freezing the images on the detector array during integrations .
the 70 section uses a @xmath21 pixel ge : ga array sensitive from 53 to 107 .
a cable failure external to the instrument has disabled half of the array and the following description reflects this situation .
the light from the telescope is reflected into the instrument off a second pickoff mirror .
it is brought to a pupil at a second facet of the scan mirror and from there passes through optics that bring it to the detector array .
for this arm of the instrument , there are actually three optical trains that can relay the light to the array ; the scan mirror is used to select the path to be used for an observation .
one train provides imaging over a field @xmath22 , with a pixel scale of @xmath23 corresponding to a @xmath24 sampling of the point spread function .
the physical size of a 70 pixel is @xmath25 mm@xmath26 and 2 mm long in the direction of the optical axis .
this train provides imaging over a fixed photometric band from 55 to 86 .
the scan mirror feeds this mode when it is in position to feed the other two arrays , so imaging can be done on all three arrays simultaneously .
a second train also provides imaging in the same band , but with the focal plane magnified by a factor of two to @xmath27 per pixel .
this mode is provided for imaging compact sources where the maximum possible angular resolution is desired : the pixel scale corresponds to @xmath28 at the center wavelength of the filter band .
the third train brings the light into a spectrometer , with spectral resolution of @xmath29 from @xmath30 . in this spectral energy distribution ( sed ) instrument mode
, light is directed to a reflective `` slit '' and then to a concave reflective diffraction grating that disperses the light and images the spectrum onto a portion of the 70 array .
the slit is 16 pixels long and 2 pixels wide , corresponding to @xmath31 on the sky .
the dispersion is 1.73 pixel@xmath32 .
reverse bolometer stimulators are provided for calibration , and the scan mirror provides the dithered and scan mapping modes of operation at 70 as have been described for the 24 array .
the 160 section shares the pickoff mirror and scan mirror facet with the 24 band .
after the light has been reflected off the scan mirror , the telescope focal plane is reimaged and divided , with part going to the si : as array and part going to a stressed ge : ga array , operating in a fixed filter band from 140 to 180 .
this array has @xmath33 pixels , arranged to provide an imaging field @xmath20 long in the direction orthogonal to the scan mirror motion with the two rows of detectors spaced such that there is a gap one pixel wide between the two rows .
this pixel size provides @xmath16 sampling of the point spread function .
the physical size of a 160 pixel is @xmath34 mm@xmath35 .
reverse bolometer stimulators are included in the optical train , and the scan mirror provides modes similar to those with the other two arrays .
a key aspect of the calibration of the mips ge arrays is the frequent use of stimulators @xcite to track responsivity variations .
the emitters in these devices are sapphire plates blackened with a thin deposition of bismuth , which also acts as an electrical resistor .
the emitters are suspended in a metal ring by nylon supports and their electrical leads .
when a controlled current is run through the device , the sapphire plate is rapidly heated by ohmic losses in the metallized layer . the thermal emission is used to track changes in detector response in a relative manner ; hence these devices are described as stimulators rather than calibrators . because of the large responsivity of the detector arrays , it is necessary to operate these stimulators highly inefficiently to ensure accurate control without blinding the detectors .
they are mounted inside cavities that are intentionally designed to be inefficient ( e.g. , black walls , small exit holes ) .
this allows the stimulators to be run at high enough voltage to be stable and emit at a reasonable effective temperature .
the constant - amplitude stimulator flashes provide a means of tracking the responsivity drift inherent in the ge detectors .
figure [ fig_stim_repeat ] illustrates the importance of tracking the responsivity variations of ge detectors with as fine a time resolution as feasible .
the repeatability of the stimulator measurements is a function of both the background seen by the detector element as well as the amplitude of the stimulator ( stim ) signal above the background .
the repeatability of a measurement of the stim signal improves with decreasing background and increasing stim amplitude . for both the 70 and 160 arrays , in ground testing stim amplitudes of greater than @xmath37500 dn / s above the background yielded a repeatability of better than @xmath31% on most backgrounds . setting the stim amplitudes at this level provides a balance between repeatability of the stims and the range of backgrounds accessible to observation without saturation . at this level
, well over 95% of the sky should be observable without saturating stim flash measurements at both 70 and 160 .
additional complications at 160 include a strong illumination gradient in the stim flash illumination pattern from one end of the array to the other as well as a large increase in the responsivity of the array with exposure to cosmic rays .
it is not possible to set the stim amplitude at the optimum 7500 dn / s across the whole array due to a factor of four gradient in the stim amplitude across the array .
the on - orbit stim amplitude was set to provide an optimal amplitude over the majority of the 160 pixels .
the degradation in stim repeatability on the low illumination region can be mitigated by an observing strategy that dithers the image such that the same region of the sky spends equal amounts of time on both regions of the detector .
both of the ge arrays show calibration shifts with even small exposure to ionizing radiation .
the effects of ionizing particles were tested using characterization arrays ( see [ sec_lab_test ] ) at the university of california , davis accelerator .
the proton beam was attenuated to reduce the particle impact rate to a level similar to that expected on orbit .
the energy of the particles was such that each impact was strongly ionizing , depositing much more energy in the detector volume than is expected from a typical cosmic ray .
thus , these tests served as a worst - case model of the detector response to cosmic - rays on orbit .
the detector responsivity slowly increased with time under exposure to the proton beam .
the rate of responsivity increase on the 70 array was comparable to that observed under typical illumination conditions ( cf .
[ fig_stim_repeat ] ) without the proton beam , suggesting that accumulated transient response from the background and signals inside the cold test chamber and the photon flux at the accelerator contribute similarly to the responsivity increase .
if the particle impacts at the accelerator really represent a worst - case scenario , this suggests that the on - orbit responsivity increase of the 70 array may be dominated by photon flux rather than cosmic ray effects .
in contrast , the 160 array showed a large responsivity increase with increasing radiation dose .
if they are of modest size , such responsivity shifts can be determined and removed during calibration through use of the stimulator observations . however ,
when the shifts are large , they are also highly unstable and can result in substantial excess noise .
three methods were tested to remove such effects : re - thermalization of the detectors by heating them ( anneals ) , exposing the detectors to a bright photon source , and boosting the detector bias above breakdown .
our experiments indicated that the latter two methods produced little benefit .
although the instrument design permits use of all three techniques , we remove radiation damage to the ge arrays by periodically thermally annealing the detectors .
there are four mips observing modes , all of which have been designed to provide a high level of redundancy to ensure good quality data ( especially for the ge arrays ) .
the photometry mode is for point and small sources . as an example of the redundancy inherit in mips observations , a visualization of a single photometry mode cycle is shown in fig .
[ fig_phot70 ] for 70 .
the scan map mode provides efficient , simultaneous mapping at 24 , 70 , and 160 by using a ramp motion for the scan mirror to compensate for continuous telescope motion , effectively freezing images of the sky on the arrays .
a visualization of a small portion of a scan leg is shown in fig .
[ fig_scanmode ] .
the sed mode provides 53 to 107 spectra with a resolution r @xmath36 .
the total power mode ( tpm ) is for making absolute measurements of extended emissions .
the pacing of the mips data collection is based on a `` mips second . ''
a mips second is approximately 1.049 seconds , and has been selected to synchronize the data collection with potential sources of periodic noise , such as the computer clock or the oscillators in the power supplies .
to first order , this design prevents the down - conversion of pickup from these potential noise sources into the astronomical signals .
the data are taken in data collection events ( dces ) ; at the end of a dce , the array is reset before taking more data .
dces are currently limited to 3 , 4 , 10 , or 30 mips seconds for the 24 array and 3 , 10 , or 10 mips seconds for the 70 and 160 arrays . during a dce ,
each pixel generates a voltage ramp on the array output , as the charge from incoming photons is accumulated on the input node of its integrating amplifier .
these ramps are the basic data collected by all three arrays .
the 24 array is non - destructively read out every 1/2 mips second while the 70 and 160 arrays are non - destructively read out every 1/8 mips second .
all the samples are downlinked for the 70 and 160 arrays , but this is not possible for the 24 array due to bandwidth restrictions .
the 24 array has two data modes , sur and raw .
most 24 data are taken in sur mode in which the ramps are fitted to a line , and only the fitted slope and first difference ( the difference between the first two reads in the ramp ) are downlinked .
the raw mode downlinks the full 24 ramps , but this mode is used only for engineering observations .
there are three natural steps in reducing data from integrating amplifiers : ( 1 ) converting the integration ramps to slopes ; ( 2 ) further time - domain processing of the slope images ; and ( 3 ) processing of dithered images in the spatial domain . for detectors
that do not have time - dependent responsivities , only the first and last steps are usually important .
this is strongly not the case for the mips ge arrays and also mildly not so for the mips si array .
as a result , mips processing includes all three steps .
first , the integration ramps are converted into slopes ( dn / s ) while removing instrumental signatures with time constants on the order of the dce exposure times ( [ sec_ramps_to_slopes ] ) .
second , the slopes are calibrated and instrumental signatures with time constants longer than the dce exposure times are removed ( [ sec_slope_cal ] ) .
third , the redundancy inherent in the mips observing modes allows a second pass at removing instrumental signatures ( [ sec_redund ] ) .
the algorithms used in the first two steps have mostly been determined .
the main algorithms used by the third step are being optimized with actual data taken on orbit .
portions of the reduction algorithms described in this paper were presented in a preliminary form by @xcite .
we made extensive use of laboratory testing and theoretical investigations in choosing and ordering the relevant steps .
figure [ fig_flowchart ] is a graphical representation of the specific tasks in each of the three processing steps .
three versions of the 70 and 160 arrays were constructed : a flight array , a flight spare array , and a characterization array . before integration into the instrument ,
the basic performance of the flight and flight spare arrays was measured ( i.e. , read noise , dark current , nep , etc . ) .
the characterization arrays were then installed in the two specialized dewars previously used for the flight and flight spare array testing .
these arrays are used to determine the detailed behaviors of the 70 and 160 detectors .
this knowledge was then used to design observations with the flight arrays to remove specific ge detector effects .
the ability to do extensive testing on the characterization arrays has been crucial to the development of the data reduction algorithms for the ge arrays detailed in this paper .
in addition to testing at the array level , testing at the instrument level was carried out using the low background test chamber ( lbtc ) .
the lbtc was constructed to allow for testing of the full mips instrument and , thus , had a number of independently controlled stimulators including pinhole stimulators providing point sources for testing .
the lbtc allowed for the imaging performance of the full instrument to be tested as well as providing for extensive testing of the 24 array .
additional details of the laboratory testing can be found in @xcite .
we also carried out detailed numerical modeling of the behavior of the ge arrays .
this modeling allowed effects found in the laboratory testing to be investigated in more detail .
for example , the modeling was able to show that the difference in hook behaviors between the 70 and 160 arrays was due to their different illuminations @xcite .
the numerical modeling was also crucial to the understanding of the behavior of small signals on the detectors .
for example , this modeling was able to determine that the stim flash latents ( see [ sec_sflash_latent ] ) were additive , not multiplicative .
this understanding then guided the efforts to remove this signal .
the first part of the processing fits the ramps to produce slopes for each dce .
the processing for the ge ( 70 and 160 ) and si ( 24 ) raw mode data is similar , differing only in the instrumental signatures removed .
first , reads which should be rejected from the linear fits are identified .
reads are rejected if they represent missing data , autoreject reads , or saturated data .
second , the ramps are corrected for instrumental effects . these are dark current ( si only ) , rowdroop ( si only ) , droop ( si only ) , electronic nonlinearities , and stim flash latents ( ge only ) .
third , jumps in the ramps usually caused by cosmic rays are identified . in the process , reads that are abnormally noisy are identified as noise spikes . finally , all the continuous segments in each ramp are fit with lines and the resulting slopes averaged to produce the final slope for each pixel .
an example of a 24 ramp is given in fig .
[ fig_24ramp ] .
the 70 and 160 ramps are very similar to the 24 ramp , except they do not have droop .
the graphical representation of the data processing shown in fig . [ fig_flowchart ] gives the ordering of the reduction steps .
the processing for the si sur mode data is necessarily different as the ramps are fit on - board and only the slope and first difference images are downlinked .
the following subsections will describe the si and ge raw mode processing followed by a description of the necessary differences for the si sur mode processing .
there are two reasons to automatically reject ( autoreject ) reads ; to avoid reset signatures and to not use the ramps beyond 2 mips seconds for stim flash dces .
all mips arrays are reset at the beginning of a ramp , and this has been seen to leave a signature in the first few reads .
in general , this reset signature only affects the first read .
the first read is automatically rejected for all three arrays .
this is even true for sur data for which the line fit is done on - board spitzer .
the 70 and 160 arrays can be operated with a reset in the middle of the dce to improve performance .
when this mode is used , the reset signature has been seen to last for 4 reads and these 4 reads are then automatically rejected . in a stim flash dce
only the first 2 mips seconds of a ramp are valid .
after 2 mips seconds , the stim is turned off and after 2.5 mips seconds , a reset is applied .
finally , all reads that are below or above the allowed limits for the mips analog - to - digital converters ( adc ) ( soft saturation ) or saturating the 70 and 160 readout circuits ( hard saturation ) are flagged as low or high saturation , respectively .
all three mips arrays display nonlinearities that have been traced to the electronics . for the 24 array
these nonlinearities are mainly due to a gradual debiasing which occurs as charge accumulates in each pixel during an exposure . for the 70 and 160 arrays ,
the readout circuits have been constructed to keep the same bias voltage across the detectors even as charge accumulates . nevertheless , electronic nonlinearities arise due to the simplified ctia circuit .
the behavior of the electronic nonlinearities was determined from extensive ground - based testing on the flight arrays .
for the 24 array , the functional form was characterized from raw mode data ramps ; a typical case is shown in figure [ fig_24ramp ] .
the ramps for most of the pixels can be nearly perfectly described by quadratic polynomial fits ; the linear component of the fit gives directly the linearized signal .
for the 70 and 160 arrays , the electronic nonlinearities have been shown generally to have a quadratic shape with significant deviations .
corrections were tabulated as a lookup table to allow for the semi - arbitrary forms . for the 24 , 70 , and 160 arrays ,
the maximum nonlinearity at full well ( adc saturation ) ranges over the array from @xmath37 , @xmath38 , and @xmath39 , respectively .
the main reason discontinuities or jumps appear in mips ramps is cosmic rays .
cosmic rays strike the ge detectors ( 70 and 160 arrays ) at a rate of one per pixel per twelve seconds .
the rate on the si detector ( 24 array ) is much lower , due to its smaller pixels .
it is also possible to get a ramp jump due to an anomaly we have termed a readout jump .
ground - based testing has shown that the entire output of one of the 32 readouts ( 4 @xmath40 8 pixels ) on the 70 array occasionally jumps up and then jumps back down by the same dn amount approximately 1 second later .
jumps in the ramps are detected using a combination of two methods .
first , @xmath41
2-point differences are constructed from the @xmath42 reads and outliers are flagged as potential ramp jumps using an iterative sigma clipping algorithm .
these potential jumps are tested to see if they are noise spikes or actual ramp jumps by fitting lines to the segments on either side of the potential jump .
if the two fitted lines imply a jump that is smaller than the expected noise , then the jump is actually a noise spike , not a cosmic ray or readout jump .
second , a more sensitive test for ramp jumps is performed @xcite .
this method works by assuming each read in a ramp segment has a ramp jump after it and fitting lines to the resulting two subsegments on either side .
the most significant ramp jump in the segment implied from the two line fits is then tested to see if it is larger than the noise .
if so , then this read is labeled as a ramp jump .
the process can be repeated on the subsequent ramp segments until no more jumps are found or a preset number of iterations have been performed . as this second method is more sensitive than the first , but significantly more computationally intensive , we combine the two methods to achieve the best sensitivity to ramp jumps with the least computation time .
we explored the signatures of cosmic rays in ramps using several hours worth of 70 and 160 array data that were subject to constant illumination .
we then extracted those ramps where we detected ramp jumps ( assumed to be due to energetic particle impacts ) and assessed the effects on the ramp after the impact . on the 70 array we find two main effects : a steepening of the ramp that lasts for a few reads and a persistent responsivity increase of @xmath43 after a big hit ( see fig .
[ figure : slope - change - after - radhit ] ) .
this is consistent with the slow responsivity increase observed during the radiation run .
these results dictate our strategy for dealing with cosmic ray hits on this array : several reads after a hit should be rejected from slope fitting to ensure that the fast transient does not bias the slope measurement , while the small responsivity increase after large hits will be tracked by the stim flash measurements .
the 160 array response to cosmic rays is somewhat different .
we detected no fast transient within the ramp , but the slope of the ramp after a hit was often different from the slope before the hit .
this slope change typically did not persist into the next dce , after a reset had occurred , as shown in figure [ figure : slope - change - after - radhit ] .
thus , we were unable to detect a persistent responsivity increase due to particle impacts , in contrast to the accelerator data ( [ sec_anneals ] ) .
given that we are unable to predict how the slope will change after a particle hit and that the slope returns to its previous value after the next reset ( usually the next dce ) , the conservative strategy for dealing with particle impacts on this array is to simply ignore all data between a particle hit and the next reset .
slopes are determined for each ramp by fitting lines to all the good segments in a ramp .
the slope for a ramp is then the weighted average of the slopes of the ramp segments .
the weight of each segment is determined from the uncertainty in the segment slope as discussed in the next paragraph .
each good segment of a ramp is identified as containing only good reads and not containing any ramp jumps .
lines are fit to these segments with the standard linear regression algorithm . calculating the uncertainties on the fitted slope and zero point
is not as straightforward .
the uncertainties on each read have both a correlated and random component .
the correlated component is due to photon noise , as the reads are a running sum of the total number of photons detected .
the random component is the read noise .
we have derived equations for the linear fit uncertainties for the correlated component following the work of @xcite .
the details of this derivation are given in appendix [ app_fit_unc ] .
the slope and zero point uncertainties are calculated for the correlated and random read uncertainties separately and then combined in quadrature to get the final uncertainties . the calibration of the 70 and 160 arrays is directly tied to the stim flashes measured approximately every two minutes .
the brightness of these stim flashes is set as high as possible to ensure the best calibration ( cf .
[ sec_stims ] ) .
these stim flashes produce a memory effect , called a stim flash latent , that is persistent for a brief time .
intensive measurements of stim flash latents have been performed at the university of arizona on the 70 and 160 characterization arrays .
we determined the time constants , amplitudes , variations with the background , and repeatability of the stim flash latents as well as the accuracy of the correction and the effects on the calibration of sources observed during the latent . to characterize the decay behavior of the latents
, we fit an exponential law to the time signal of each array pixel .
each cycle is divided by the stim amplitude value , to have dimensionless data ( fraction signal / stim ) .
the function @xmath44 used to fit the latent is a double exponential : @xmath45 where @xmath46 is the time after the stim is turned off , @xmath47 is the background level , @xmath48 and @xmath49 give the component amplitudes , and @xmath50 and @xmath51 give the time constants . at 70 ,
only a single exponential is needed ( thus @xmath52 ) .
the amplitude @xmath48 is always less than 3% of the stim amplitude , and in most of the cases below 0.5% . the time constant @xmath50 ranges from 5s to 20s . as a function of increasing background , @xmath48 increases and @xmath50 decreases .
the latents are repeatable to 15% or better .
an example of the stim latent of one pixel on the 70 characterization array is given in fig .
[ fig_slatent ] . at 160 ,
the latency effect is more pronounced than at 70 .
the amplitude @xmath48 is less than 5% of the stim amplitude .
the time constant @xmath50 ranges from 5 to 20s .
the amplitude @xmath49 is less than 3% .
the time constant @xmath51 equals 20s at high background , and is negligible at low background .
the amplitude @xmath48 and time constant @xmath50 are almost insensitive to the background .
the latents are repeatable to 20% or better .
[ fig_slatent ] also gives an example of the stim latent of one pixel on the 160 characterization array .
in general , the stim flash latents are negligible @xmath330 seconds after the stim is turned off .
in the first 30 seconds , the calibration of a point source might be overestimated by 1% at 70 and 12% at 160 if no correction is applied . to correct for the stim latent contribution to the pixel signal
, we apply a time - dependent correction at the ramp level .
we subtract the latent contribution , which is obtained by integrating eq .
[ eq : latent ] . on pre - flight data ,
the amplitude of the latents after correction is reduced by a factor of @xmath32 at 70 and @xmath34 at 160 .
the rowdroop effect manifests itself as an additive constant to each individual pixel and is proportional to the sum of the number of counts measured by all pixels on its row , where a row is in the cross - readout direction .
this effect is not completely understood , and is similar to ( but separate from ) the droop phenomenon ( see [ sec_droop ] ) .
the additive signal imparted to each pixel on a row is constant and exhibits no gradient or dependence with pixel position , thus , it is not related to a charge
bleed , or `` mux bleed '' effect .
the rowdroop contributes a small amount to the flux of an individual pixel , and will only significantly affect pixels on rows with high - intensity sources .
an example of rowdroop from ground - based testing is shown in fig .
[ fig_rowdroop ] .
using images of pinhole sources obtained in ground testing , we have computed the row droop constant of proportionality , @xmath53 .
this is the factor that gives the fraction of the total counts in a row which is the result of row droop and should be subtracted from each pixel in that row .
we find that the constant of proportionality for the mips 24 array is @xmath54 .
thus , the rowdroop contributes @xmath55% of the total number of counts on a row .
the rowdroop is corrected for on a read - by - read basis .
droop is a constant signal added to each pixel by the readouts .
the exact cause of droop is unclear .
this extraneous signal , akin to a dc offset , is directly proportional to the total number of counts on the entire array at any given time .
we have measured the constant of proportionality from ground test data .
the droop coupling constant was measured to be @xmath56 , which agrees well with the 0.32 determined by bna .
the droop correction algorithm first computes the mean signal on the array , which is then multiplied by the droop coupling constant to derive the droop signal , as given by @xmath57 where @xmath58 is the droop signal , @xmath59 is the signal on each pixel ( comprising both the actual incident flux and the droop ) , @xmath60 is the number of pixels , and @xmath61 is the droop coupling constant .
the resultant droop signal is then subtracted from the original signal on each pixel . under normal circumstances ,
the uncertainty associated with this process is at the @xmath43 level , limited mainly by the uncertainty on the coupling constant .
however , greater uncertainties arise when pixels are saturated ; since adc saturation occurs well before hard detector saturation , droop signal will still accumulate for an incident flux above the adc saturation level . in this case
, the actual signal ramp must be extrapolated beyond the saturation point .
the droop signal is determined by extrapolating a fit to the unsaturated portion of the ramp . as with the rowdroop correction ,
the droop correction is done on a read - by - read basis for raw mode 24 data .
dark subtraction is done at each read using a dark calibration image containing the full dark ramp for each pixel .
this step serves both to remove the ( small ) dark current contribution and the offset ramp starting points , so that each ramp starts near zero .
the majority of the 24 data are taken in the sur mode instead of the raw mode . in the sur mode
, a line is fit to the data ramp on - board the spacecraft .
the resulting slope and first difference ( difference between the first two reads of the data ramp ) images are downlinked instead of the full ramp .
the first difference frame effectively increases the dynamic range of the sur mode as signals that saturate somewhere in the ramp , but after the second read , will have a valid measurement in the first difference frame . to reduce the data downlinked , any first difference value that is from a ramp which does not saturate is set to zero .
this increases the compressibility of the first difference frame .
there can be degeneracy of sur slope values due to the possibility of saturation .
the possible slope value for a given pixel reaches a maximum at full well , the point of adc saturation .
after that point , as the data ramp reaches saturation at the last few reads , the slope value will begin to decrease because the on - board sur algorithm does not reject saturated reads . in cases of extreme saturation , the slope becomes quite small , and can eventually become zero if saturation occurs within the first few reads .
the first difference value is provided to break this degeneracy .
we have employed a conservative threshold value for the first difference , above which a pixel is flagged as being likely saturated .
adc saturation occurs at + 32768 dn ( see figure [ fig_24ramp ] for an example of a saturated raw ramp ) . assuming a linear ramp , the first difference for a ramp that just saturates on the final read would be @xmath62 , where @xmath63 is the total number of reads in the data ramp .
for example , there are 60 reads in a 30 second dce , yielding an ideal saturation threshold of @xmath64 dn / read . to be more conservative
, we actually employ a threshold value of 1000 dn / read for a 30 second exposure time and scale this for other exposure times . since the data ramps are not linear , the actual first difference threshold is larger than our chosen default value , so most cases of saturation will be flagged .
the only exception being saturation at the first read , in which case both the slope and the first difference would be zero . for all pixels that have been flagged for saturation , the first difference value should be used in place of the slope . the rowdroop and droop subtraction is done in the same way for sur mode as for raw mode , except that the corrections are performed on the slope and first difference images . because the sur data do not preserve the actual data ramps , the linearity correction made somewhat complicated .
nevertheless , the quadratic behavior of the ramps can be used to analytically determine the linearization of the sur slope values .
this correction depends on the observed sur slope value , exposure time , and known quadratic nonlinearity .
note that saturation invalidates this method , as the sur slope - fitting algorithm does not reject saturated reads . in this case
, no linearity correction is applied .
the next step in the mips data reduction is to calibrate the slope images while removing instrumental effects with time constants longer than the dce exposure times .
the instrumental effects corrected at this stage include latents ( 24 ) , responsivity drift ( 70 and 160 ) , pixel - to - pixel responsivity variations , the telescope illumination pattern , and flux nonlinearities ( 70 and 160 ) . the graphical representation of the data processing shown in fig .
[ fig_flowchart ] gives the ordering of the reduction steps .
since the time dependent responsivity of the ge arrays requires additional calibration steps than is usual for more common array detectors , we give the mathematical basis of our ge slope calibration in [ sec_slope_math ] . ignoring the 70 and 160 flux nonlinearities ,
an uncalibrated slope image can be represented by @xmath65 r(i , j , t_n)\ ] ] where @xmath66 is the science image of interest , @xmath67 represents the telescope and instrument optics ( the mean of @xmath67 is one ) , @xmath68 is the dark current , and @xmath69 is the instantaneous responsivity of the array ; @xmath70 represent the pixel coordinates and @xmath71 the time of the @xmath42th dce .
calibration involves isolating @xmath66 , the flux from the sky@xmath72object in the above equation .
the term @xmath73 is the equivalent of a traditional flat - field term . as @xmath74 is a rather sensitive function of time for the 70 and 160 detectors , however , a global `` flat - field ''
can not be determined , but must be derived for each dce separately .
the stimulators provide the means to monitor @xmath74 and all science observations will be bracketed by stim flashes .
stim flash images will be equivalent to science frames with the addition of a stimulator illumination pattern : @xmath75 r(i , j , t_n)\ ] ] where @xmath76 is the illumination pattern introduced on the array by the stim flash with the mean of @xmath76 equal to one .
mips observations include the requirement that each stimulator flash will be preceded by a background exposure with the identical telescope pointing ; thus for the @xmath77 stimulator dce there exists a background dce taken at time @xmath78 , @xmath79 r(i , j , t_n - \epsilon).\ ] ] if we assume that the responsivity of the array @xmath69 does nt change dramatically between times @xmath80 and @xmath81 , i.e. @xmath82 , we can construct for each stimulator flash a background subtracted stim flash : @xmath83 with background subtracted stim flashes determined from eq .
[ eq : bksubstim ] for all stim flashes in the data set , an instantaneous stim can be determined for any time , @xmath71 , by interpolation from bracketing stim flashes : @xmath84\ ] ] where @xmath85 $ ] is some interpolating function on background subtracted stims for times @xmath80 bracketing @xmath71 .
analysis of ge characterization array data indicates that a weighted linear fit ( weighted by the uncertainty in the stim flash frames ) to two stim flashes on either side of the data frame ( a total of four stim flashes ) provides the optimal strategy for determining the instantaneous stim amplitude ( repeatability to @xmath31% on most backgrounds ) . dividing science frames , eq .
[ eq : sciframe ] , by the interpolated instantaneous stim , eq .
[ eq : inst_stim ] , produces @xmath86/s(i , j).\ ] ] while we have removed the time dependent responsivity variation , the data of interest , @xmath87 , are still modified by the optical response and the dark current ; in addition , we have introduced the stimulator illumination pattern into our data .
fortunately , since the time dependence has been removed , we can remove these other instrumental signatures through carefully accumulated calibration data .
first , the dark correction , @xmath68 , can be determined from a sequence of exposures as above , with the additional constraint that the scan mirror be positioned such that no light from the `` sky '' falls on the detector
. thus the data and stim flashes in a dark current data sequence are represented by @xmath88 and @xmath89 r(i , j , t_n),\ ] ] respectively .
the dark data are corrected for responsivity variations exactly as described above and the individual frames combined to produce an average dark current , @xmath90 .
subtracting this dark current from science frames that have been corrected for responsivity variations , eq .
[ eq : rescor ] yields @xmath91 our responsivity , dark corrected science frame .
what remains is to correct for the telescope optics , @xmath67 , and the stim illumination pattern , @xmath76 .
correcting for the combined illumination pattern of the telescope and stim involves a standard series of mips exposures , i.e. data frames interspersed with stim flashes . as such , they may be represented by equations of the form eq .
[ eq : sciframe ] , where the @xmath92 represent dithered images of `` blank '' sky fields . calibrating the sequence by correcting for responsivity variations and dark current as above results in a series of images @xmath93 since by construction , the @xmath92
are dithered images of `` smooth '' regions , if a large number of @xmath92 are acquired , they may be median combined to remove point sources ( and extended sources if dithered `` sufficiently '' ) , cosmic rays , etc .
resulting in @xmath94 where @xmath67 and @xmath76 are constant regardless of telescope pointing .
hence the median only affects the changing sky image as the telescope is dithered .
the constant @xmath95 in equation [ eq : illcor ] may be set to one resulting in the illumination correction frame @xmath96 the responsivity corrected , dark subtracted data ( equation [ eq : darkrescor ] ) are now divided by the illumination correction resulting in @xmath97 and we have recovered the quantity of interest , the astronomical sky @xmath66 .
suitable observations of standard stars can then be used to convert instrumental counts to physical units ( e.g. janskies ) .
the dark , flat field , and illumination correction calibration images described above will be obtained throughout the life of the mission .
example preflight calibration images are shown in figs .
[ fig_24cal]-[fig_160cal ] .
simulations indicate that high s / n flat field and illumination correction images ( @xmath98 rms ) can be obtained with dithered observations of `` smooth '' areas of the sky : @xmath360100 dces at 24 and @xmath3200 dces at 70 are required . at 160 , the situation is less ideal , with simulations indicating as many as 500 dces may be required to produce flats to better than 1% rms .
si ibc arrays are known to have considerable latency , where the signal induced by bright illumination persists after the illumination has terminated .
ideally , if one knows the position of a source exposed on the array and the latency decay behavior , these artifacts can be subtracted from an image .
we have characterized the latent behavior from ground test data .
several different conditions were explored , including varying brightnesses of the illuminating source , varying brightnesses of the background , initial bias boosts , and changing the number of resets via different exposure times .
a bias boost can flush out most of the trapped charge , but resets are not nearly as effective .
since bias boosts will only be done in the first dce of each observation , we correct for latent residuals in the data processing .
the latent decay curve can be described by single exponential , given by @xmath99 where @xmath100 is the slope in the absence of a latent , @xmath101 is the initial value of the latent , and @xmath102 is the latent time constant . based on the limited ground data , the latent parameters ( @xmath101 and @xmath102 ) appear to be functions of background levels , number of resets and possibly location on the array . in general , the latent contribution is about @xmath3 1% of the initial source brightness @xmath35 sec after that source has shut off .
higher background yield slightly higher values for @xmath101 and lower values of @xmath102 .
the value of @xmath102 is in the range of @xmath103 seconds .
both the 70 and 160 arrays exhibit nonlinearities that are dependent on the incident point source flux as well as the background .
these are termed flux nonlinearities and have been observed in data taken with the characterization array as well as the flight array . as is usual for the ge arrays , each pixel shows flux nonlinearities with a different dependence on source flux and background . correcting for this effect
can be broken into two pieces : ( 1 ) removing the pixel to pixel differences in the nonlinearity followed by ( 2 ) the application of a global nonlinearity correction as a function of the source brightness and background .
the pixel to pixel variations in the flux non - linearity may be mapped by analyzing the ratio of two stim flashes , where one is the standard on orbit calibrating stim flash .
measured differences in the ratio from pixel to pixel can be used to correct each pixel to the same flux non linearity for the given background and source ( second stim flash ) amplitude .
repeating the measurement for a variety of second stim flash amplitudes ( up to saturation for each pixel ) and backgrounds will map out the correction .
the second , global , stage of the correction can be characterized by observations of calibration stars with a range of known brightness ratios on similar backgrounds .
the combination of these two tasks outlined above should provide a good measurement of the flux nonlinearity correction for a range of backgrounds .
this correction will improve continuously during the mission as the range of backgrounds and calibration stars expands .
the absolute calibration of mips will rely on a well determined anchor at 10.6 using the fundamental calibrators @xmath104 boo , @xmath104 tau , and @xmath105 gem @xcite .
three independent methods will be used to extrapolate the calibration at 10.6 to the mips bands : ( 1 ) solar analogs , ( 2 ) a star atmospheric models , and ( 3 ) semi - empirical models of k giants .
grids of stars for each method have been observed from the ground and tied to the fundamental calibrators at 10.6 . for the solar analog stars , on orbit observations at 24 , 70 , and 160
are being compared with extrapolations of empirical measurements of the sun extrapolated into the mips bands .
a grid of a stars has been observed in all three bands on orbit and compared to extrapolations of a star atmosphere models to the mips bands .
while the solar analog and a star calibrators will be observed in the mips 160 band , the k giant calibrators will be the only ones detectable at high signal - to - noise in that band .
on orbit observations of the k giant calibrators are being compared to theoretical extrapolations of model atmospheres , eg .
@xcite extrapolated to longer wavelengths using the engelke function @xcite .
absolute flux calibrators will be observed throughout the lifetime of the mission .
the last step in the reduction of mips data is to use the redundancy inherent in the observing modes to improve the removal of instrumental signatures .
this step is mainly for the 70 and 160 data due to the challenging aspects of ge detector calibration .
we define the level of redundancy to be the number of different pixels that measure the same point on the sky .
our approach will be to look for known instrumental signatures ( as a function of time ) in the difference between what a particular pixel detects and what all the other pixels detected for the same sky locations .
this is possible because the observing strategy has been designed so that each point on the sky will be observed multiple times by different pixels .
table [ tab_redund ] shows the minimum level of redundancy for each mips observing mode .
many mips observations are taken with multiple cycles resulting in significantly higher redundancies .
it is recommended to have a minimum redundancy of four .
lccc photometry , compact & 14 & 10 & 2 + photometry , large & 10 & 6 & 1 + photometry , compact super resolution & 14 & 8 & 6 + photometry , large super resolution & 10 & 8 & + scan map , slow & 10 & 10 & 1 + scan map , medium & 10 & 10 & 1 + scan map , fast & 5 & 5 & 0.5 + sed & & 2 & + total power & 1 & 1 & 1 the basic algorithm for using redundancy to refine the instrumental signature removal is as follows . 1 .
create a mosaic of all the images in question . during the mosaic creation ,
use a sigma rejection algorithm to remove data that are deviant from the majority of the observations .
2 . use the mosaic as a `` truth '' image of what each image should have measured .
3 . for each pixel , subtract the actual from the `` truth '' measurements to create a measurement of the time history differences .
4 . examine the difference time history for known instrumental signatures .
while many instrumental signatures could be present , we plan to concentrate on stim latent residuals and systematic differences between `` extended '' sources and point sources .
actual on - orbit data will guide the details and number of instrumental signatures that are corrected using redundancy .
5 . correct for all instrumental signatures that are found to be significant .
iterate steps 1 - 5 until no new significant instrumental signatures are found .
the input to this algorithm is calibrated slope images .
the output product of this algorithm is enhanced images .
a useful side product will be the mosaicked image of the object . to use the redundancy to remove additional instrumental signatures we must first coadd all related observations into a single mosaic . because the mips optical train is made up of purely off - axis reflective elements there
exist scale changes and rotations across the re - imaged focal plane . to coadd images taken at different places on the array , it is crucial to correct the data for these distortions .
we used the code v optical models for spitzer / mips to estimate the distortions present in the images from the three mips detectors .
the results from code v allow us to determine distortion polynomials which can then be used to correct for the distortions .
we estimated the distortions by setting up a grid of equally spaced points in the field of view at a specific scan mirror angle .
the chief ray from each object point was traced through the system to where it was imaged on the focal plane . in a perfect optical system
the image points would map perfectly from the object with a possible change in magnification .
the difference between the ideal location and the actual location is the distortion .
for example , figure [ fig_70um_dist ] is a vector plot of the distortions present in the 70 narrow field array .
the equally spaced grid of points present the focal plane points and the ends of the vectors correspond to the object points , after a plate scale factor was applied .
the difference in the points ( the length of the vector ) is caused by the distortions .
lc 24 & 2.84 + 70 wide & 0.2 + 70 narrow & 7.70 + 160 & 7.78 + table [ tab_dist_det ] lists the scale change of the field of view of the different mips arrays . the scale change is defined as ( maximum length of distorted field - minimum length of the distorted field)/(minimum length of the distorted field ) . from a distortion standpoint , it is useful to look closely at individual pixels to see how distortion changes the area imaged on the pixel .
figure [ fig_70um_close ] is a plot of a distorted pixel in the 70 narrow field array .
one can see that the distorted pixel changes shape from a square to a somewhat trapezoidal shape .
the ratio of the distorted to undistorted pixel area is 1.19 .
table [ tab_dist_pix ] lists information on how distortion affects the area imaged on individual pixels on the different arrays .
the distorted pixel area ratio is defined as ( distorted pixel area)/(undistorted pixel area ) .
lccccc 24 & 0.9998 & 0.0282 & 0.9406 & 1.0613 + 70 wide & 1.0027 & 0.0042 & 0.9973 & 1.0148 + 70 narrow & 1.0129 & 0.0664 & 0.8913 & 1.1929 + 160 & 0.9781 & 0.0361 & 0.9007 &
1.0137 + following the procedure of converting the pixel coordinates to world coordinates outlined in @xcite , the distortion correction is applied to the pixel coordinates before any other transformations .
the distortion correction is accounted for by distortion polynomials .
the distortion polynomials give the additive correction to map the distorted pixel coordinates , @xmath106 to the distortion corrected pixel coordinates @xmath107 .
thus , @xmath108 and @xmath109 , where @xmath110 and @xmath111 the ability to remove additional instrumental signatures is dependent on creating a high resolution mosaicked image .
for the mosaicked image to be of sufficient resolution , the mosaicked pixel sizes must be smaller than the original input pixels .
while many mosaicking programs compensate for undersampling by making use of dithered data , the mips data are well sampled and do not require this compensation . instead , our focus is co - adding the related calibrated images into a single image without interpolating between pixels .
therefore , we always work on the coordinates of the corners of a pixel , transforming them from the input image coordinate system to the output mosaicked coordinate system .
the output mosaic image is on a single tangent plane . in the transformation of the pixel corners in the input image pixels to their location on the output mosaic image
the corners are corrected for distortion , converted to right ascension and declination , and then projected onto the tangent plane defined by the right ascension and declination of the mosaic center .
figure [ fig_mosaic ] is an example of three images which overlap each other on the mosaicked plane . in the process of establishing the location of the image pixel corners on the mosaicked plane ,
the link and overlap coverage between the input pixel and the output pixels it falls on is determined .
a critical step in removing residual instrumental signatures based on the co - added mosaic image depends on correctly linking each mosaic pixel with each image pixel that overlaps it ( and vice versa ) and accurately determining the degree of overlap .
essentially each output sub - sampled mosaic pixel becomes a cube of data , with each plane in this cube representing the information in each overlapping image pixel . the surface brightness and uncertainty associated with each mosaic pixel
is found by weighted averaging the overlapping planes of data . in the surface brightness case ,
the weighting is based on the overlap coverage and uncertainty associated with the input image pixel .
the information in each mosaic pixel is based on multiple observations of a single area on the sky .
this redundancy of data can be used to identify cosmic rays or any single image pixel measurement that deviates from the expected mean of multiple observations and expected noise .
for example , for a 70 photometry observing mode cycle , if one of the pixels suffers from a much larger stim latent than the other observations , it will stand out and be identified as an outlier . as an outlier , it will not be used in creating the mosaicked image .
after all the outliers have been determined , then the links between the output mosaic pixel and the input image pixels are used to tally the number of times an image pixel was flagged as an outlier .
if the majority of the time an image pixel was flagged as deviant , then this pixel is flagged in the original data as an outlier .
if a sufficiently large number ( about 1% ) of the input image pixels are flagged as outliers then the mosaic step is repeated .
the final output is a mosaic image that can then be used as the `` truth '' image of what each image should have measured . following the steps outlined in section [ sec_redund_algorithm ] this truth image
is used to remove residual instrumental signatures .
with the successful launch of the spitzer space telescope in august 2003 , these reduction algorithms were tested against mips flight data of astronomical sources . this testing has validated the algorithms described in this paper , but has also shown that a number of modifications will be needed to handle the realities of flight data .
the initial results of this testing are summarized here and in @xcite , but a full accounting will be a subject of a future paper when the final mips reduction algorithms are known . there were two significant changes in the instrument operations which are not easily correctable by reduction algorithms .
the 70 @xmath112 array was found to suffer from a cable short induced sometime between ground testing and flight .
this short injects a large amount of noise into one half of the 70 @xmath112 array resulting in a useful array of only @xmath113 pixels .
the 160 @xmath112 array was found to suffer from a `` blue - leak '' caused by an unintended reflection from the blocking filter which passes through the bandpass filter .
this `` blue - leak '' results in an approximately factor of 15 image leak for stellar sources .
this leak means that asteroids are now the primary calibrators for 160 @xmath112 .
other than bright stars , the leak signal is below the confusion limit for most science targets as they have much smaller blue/160 @xmath112 ratios . at 24 @xmath112
, the pre - flight reduction algorithms were found to work well with only three changes needed .
first , the row droop correction does not seem necessary , but extensive testing has yet to be completed .
second , an additive offset in the second read of every ramp was found which produced a low level ( 1 - 2% ) gradient in final mosaics .
a straightforward correction for this has been implemented using raw and sur data for calibration .
third , scan mirror angle dependent flat fields are needed due to contamination of the scan mirror by small particles .
this contamination is seen as dark spots in individual images which move with scan mirror angle , but not spacecraft offsets . with these three modifications to the preflight algorithms , mips
is producing high quality 24 @xmath112 images which are well calibrated . at 70 @xmath112 ,
flight data has validated the basic structure of the preflight reduction algorithms but significant modification is required to account for time dependent behaviors .
the stim flash latents were found to grow in amplitude quickly after anneals . with a similar timescale , the residual background time dependence ( after correction using the stim flash amplitudes ) was seen to grow .
these two facts required hand reductions to remove the stim flash latents and background variations to produce good quality mosaics at 70 @xmath112 . these two effects are prime candidates for removal using redundancy , but the effectiveness of automatic removal has not been demonstrated yet . at 160 @xmath112 ,
the basic preflight algorithms have been validated from comparison with flight data .
some differences in detector behavior were seen in flight data .
for example , the stim flash latents have a faster time constant than in preflight data . at this time
, the nonlinearities in the 70 and 160 @xmath112 arrays have not been well enough characterized with flight data to validate this section of the preflight algorithms .
finally , the cosmic ray rate seen in the ge arrays has been seen to be about a factor of two over preflight predictions , 1 cosmic ray every 12 or so seconds .
the ramp jump detection has been seen to work well and line segment fitting removes the majority of the effects of these cosmic rays .
some residual effects remain and additional characterization may lead to algorithms to remove these additional effects .
the effectiveness of the algorithms described in this paper as well as the design of mips is attested by the point - spread - functions ( psfs ) constructed from flight data at 24 , 70 , and 160 @xmath112 shown in fig .
[ fig_psf ] .
these psfs all clearly have a well - defined first airy ring with the 24 psf also exhibiting a well - defined second airy ring .
all three psfs are well represented by the predictions of tinytim models @xcite adapted to mips . in addition , there are many papers written using mips flight data for the spitzer special astrophysical journal supplement issue ( 2004 , apjs , 154 ) .
this paper has described the preflight data reduction algorithms for all three arrays for the mips instrument on spitzer .
these algorithms have been guided by extensive laboratory testing of the si ( 24 ) and ge ( 70 and 160 ) arrays .
in addition , numerical modeling of the ge arrays has provided important insights into their behavior .
the design and operation of the mips instrument has been summarized to give sufficient background for understanding the data reduction algorithms .
the design and operation of the mips instrument is mainly driven by the needs of the ge arrays .
as ge detectors display significant responsivity drift over time due mainly to cosmic ray damage , the mips observing modes include frequent observations of an internal illumination source .
in addition , most mips operating modes have been designed to provide significant redundancy to increase the robustness of the mips observations against detector effects .
the data reduction for the mips arrays is divided into three parts .
the first part converts the data ramps into slope measurements and removes detector signatures with time constants less than approximately 10 seconds .
these detector signatures at 24 include saturation , dark current , rowdroop , droop , electronic nonlinearities , and cosmic rays . at 70 and 160 ,
the detector signatures removed include saturation , electronic nonlinearities , stim flash latents , and cosmic rays .
the resulting slopes are determined from linear fits and their uncertainties are computed accounting for both the random and correlated nature of the data ramp uncertainties .
the second part of the mips data reduction converts the slopes to calibrated slopes and removes detector signatures with time constants larger than approximately 10 seconds . at 24
, this translates to applying a flat field , correcting for object latents , and applying the flux calibration . at 70 and 160
, this step includes subtracting the dark , flat fielding using an instantaneous flat field , correcting for the flux nonlinearities , and applying the flux calibration . a flat field specific to each 70 and 160 image is required to correct for the time - dependent responsivity of the ge arrays .
it is constructed from the frequent stim flashes and a previously determined illumination correction .
the third data reduction step is to use the spatial redundancy inherit in the mips observing modes to improve the removal of instrumental signatures .
this step is only applied to the ge data .
known instrumental signatures are searched for in the difference between what a specific pixel and what all other pixels from the same sky locations detected . if instrument signatures are detected , they are removed and the process is repeated .
this method is iterative in nature and will require care to avoid introducing spurious signals into the data .
the design of this portion of the data reduction algorithms is necessarily the least developed because only after spitzer launches will it be known which instrumental signatures are important to correct with this method .
finally , initial testing using flight data from mips has validated these data reduction algorithms , but some modification is necessary to account for the realities of flight .
a future paper will describe these modifications in detail once they have been devised and tested .
we wish to thank j. w. beeman and e. e. haller for their contributions to the design and building of the mips instrument .
this work was supported by nasa jpl contract 960785 .
when a detector is non - destructively read out multiple times before resetting the resulting data ramps represent correlated measurements .
this is because measurement @xmath114 is equal to @xmath115 where @xmath116 is number of photons detected in the time between @xmath117 and @xmath114 .
this statement ignores the effects of read noise , which produces uncorrelated uncertainties on the @xmath117 measurements .
while fitting lines to data with correlations is a complex subject , the form of the correlations in the case of non - destructive readouts allows analytic equations to be derived for the linear fit parameters and uncertainties .
we present a derivation of equations for linear fit parameters and their uncertainties for the case of a data ramp with correlated reads and no read noise .
this derivation is based on a similar derivation by @xcite for nicmos data ramps but is slightly more general . as part of this derivation , it can be seen that the linear fit parameters derived assuming either random or correlated uncertainties are equivalent .
this is not the case for the uncertainties on the fit parameters , which is the main motivation for this derivation .
the basics of fitting a line to data with random uncertainties are given in @xcite .
we repeat their results here , as the derivation for correlated uncertainties draws directly from this work . in fitting data to a line of the form @xmath118 the fit parameters and their uncertainties are @xmath119 where @xmath120 is the number of ( @xmath121 ) measurements , @xmath122 is the uncertainty on each measurement of @xmath117 , @xmath123 these equations assume that the measurements of @xmath117 are independent . to determine the linear fit terms for a line fit to correlated data ramps , the assumption that the @xmath117 measurements are independent is not correct .
the standard formulae need to be modified to sum over terms that are independent .
the modifications start with realizing that @xmath124 is the independent quantity in the absence of read noise .
any equation in the standard derivation that relies on the independence of @xmath117 needs to be modified to only depend on @xmath116 .
thus , @xmath125 and using a similar derivation , @xmath126 the standard equations ( [ eq_a_random ] & [ eq_b_random ] ) can then used to determine the best fit values of @xmath127 and @xmath47 for the case of correlated uncertainties .
in fact , the values of @xmath127 and @xmath47 derived assuming correlated or uncorrelated uncertainties are exactly the same . the differences between the two types of uncertainties arises in determining @xmath128 and @xmath129 . to derive @xmath128 and @xmath129 for a data ramp with correlated measurements we start with equations 6.19 and 6.20 of @xcite . converting from @xmath117 to @xmath116 as the independent variable gives @xmath130 \\ & = & \sum_{i=2}^n \left [ \sigma(p_i)^2 \left ( \frac{\partial z}{\partial p_i } \right)^2 \right ] \\\end{aligned}\ ] ] where @xmath131 is either @xmath127 or @xmath47 .
the partial derivatives needed are then @xmath132 and @xmath133 thus , @xmath134 the assumption that the uncertainties can be calculated separately for the correlated and random measurement uncertainties was tested via monte carlo simulations .
simulations for cases similar to that expected for the 70 array are plotted in figure [ fig_fit_cor_ex ] .
as can be seen from these plots , equations [ eq_sigma_a ] and [ eq_sigma_b ] give very good estimates of the actual uncertainties .
beeman , j. w. & haller , e. e. 2002 , , 4486 , 209 beichman , c. a. , neugebauer , g. , habing , h. j. , clegg , p. e. , & chester , t. j. 1988 , nasa rp-1190 , vol . 1 bevington , p. r. & robinson , d. k. 1992 , data reduction and error analysis for the physical sciences ( new york : mcgraw - hill , inc . ) burgdorf , m. j. , et al .
1998 , adv . in space research , 21 , 5 church , s. e. , griffin , m. j. , price , m. c. , ade , p. a. , emergy , r. j. , and swinyard , b. m. 1993 , proc .
spie , 1946 , 116 cohen , m. , walker , r. g. , barlow , m. j. , & deacon , j. g. 1992 , , 104 , 1650 cohen , m. , witteborn , f. c. , walker , r. g. , bregman , j. d. , & wooden , d. h. 1995 , , 110 , 275 cohen , m. , witteborn , f. c. , carbon , d. f. , davies , j. k. , wooden , d. h. , & bregman , j. d. 1996a , , 112 , 2274 cohen , m. , witteborn , f. c. , bregman , j. d. , wooden , d. h. , salama , a. , & metcalfe , l. 1996b , , 112 , 241 de graauw , t. et al .
1996 , , 315 , l49 dierckx , b. , vermeiren , j. , cos , s. , faymonville , r. , and lemke , d. 1992 , proc .
esa symposium on photon detectors for space astronomy ( see n94 - 15025 ) , pp .
405 - 408 engelke , c. w. 1992 , , 104 , 1248 gordon , k. d. et al .
2004 , , 5487 , 177 greisen , e. w. & calabretta , m. r. 2002 , , 395 , 1061 haegel , n. m. , simoes , j. c. , white , a. m. , & beeman , j. w. 1999 , , 38 , 1910 haegel , n. m. , schwartz , w. r. , zinter , j. , white , a. m. , & beeman , j. w. 2001 , , 40 , 5748 heim , g. b. et al .
1998 , , 3356 , 985 heras , a. m. , et al . 2000 , experimental astronomy , 10 , 177 hesselroth , t. , ha , e. c. , pesenson , m. , kelly , d. m. , rivlis , g. , & engelbracht , c. w. 2000 , , 4131 , 26 kessler , m. f. et al .
1996 , , 315 , l27 krist , j. 1993 , asp conf .
ser . 52 : astronomical data analysis software and systems ii , 2 , 536 lemke , d. , et al .
1996 , , 315 , l64 low , f. j. , beichman , c. a. , gillett , f. c. , houck , j. r. , neugebauer , g. , langford , d. e. , walker , r. g. , & white , r. h. 1984 , optical engineering , 23 , 122 neugebauer , g. , et al .
1984 , , 278 , 1 rieke , g. h. 2002 , detection of light , 2nd edition ( cambridge , england : cambridge university press ) rieke , g. h. , montgomery , e. f. , lebofsky , m. j. , & eisenhardt , p. r. 1981 , , 20 , 814 rieke , g. h. , lebofsky , m. j. , & low , f. j. 1995 , , 90 , 900 rieke , g. h. , et al .
2004 , , 154 , 25 schnurr , r. , thompson , c. l. , davis , j. t. , beeman , j. w. , cadien , j. , young , e. t. , haller , e. e. , & rieke , g. h. 1998 , , 3354 , 322 sparks , w. b. 1998 , nicmos instrument science report , space telescope science institute , 98 - 008 swinyard , b. m. et al .
1996 , , 315 , 43 swinyard , b. , clegg , p. , leeks , s. , griffin , m. , lim , t. , & burgdorf , m. 2000 , experimental astronomy , 10 , 157 valentijn , e. a. & thi , w. f. 2000 , experimental astronomy , 10 , 215 young , e. t. et al .
1998 , , 3354 , 57 young , e. t. et al .
2003 , , 4850 , 98 | we describe the data reduction algorithms for the multiband imaging photometer for spitzer ( mips ) instrument .
these algorithms were based on extensive preflight testing and modeling of the si : as ( 24 ) and ge : ga ( 70 and 160 ) arrays in mips and have been refined based on initial flight data .
the behaviors we describe are typical of state - of - the - art infrared focal planes operated in the low backgrounds of space .
the ge arrays are bulk photoconductors and therefore show a variety of artifacts that must be removed to calibrate the data . the si array , while better behaved than the ge arrays , does show a handful of artifacts that also must be removed to calibrate the data .
the data reduction to remove these effects is divided into three parts .
the first part converts the non - destructively read data ramps into slopes while removing artifacts with time constants of the order of the exposure time .
the second part calibrates the slope measurements while removing artifacts with time constants longer than the exposure time .
the third part uses the redundancy inherit in the mips observing modes to improve the artifact removal iteratively . for each of these steps
, we illustrate the relevant laboratory experiments or theoretical arguments along with the mathematical approaches taken to calibrate the data .
finally , we describe how these preflight algorithms have performed on actual flight data . |
with the development of electromagnetic and optics technology and the increasing demands in industrial and military applications , scattering from periodic structures @xcite have attracted much interest in recent years . a variety of numerical methods including finite difference methods @xcite , finite element methods @xcite , spectral and spectral element methods @xcite , integral equation methods @xcite , dirichlet - to - neumann map methods @xcite , and mode expansion method @xcite have been developed by the engineering community and the applied mathematical community for solving linear diffraction problems from periodic structures .
finite difference and finite element methods are easy to implement and result in sparse linear systems , which enable the use of sparse direct solvers .
however , these methods typically have significant dispersion errors for high - frequency problems and thus fail to provide accurate solutions .
if the medium is piecewise constant , boundary integral equations formulated on the interfaces of multilayer structures @xcite or the boundaries of the obstacles @xcite are natural and mathematically rigorous . by exploiting high - order quadratures , one obtains much higher efficiency and accuracy than finite difference and finite element methods . for problems with general medium , spectral methods @xcite can be employed .
they are simple to implement and typically require relatively small number of unknowns to attain a fixed accuracy .
we refer to @xcite for a brief survey of existing spectral methods for partial differential equations . in this paper , we consider 2d quasi - periodic scattering problems . by introducing transparent boundary conditions ,
the problem is governed by a helmholtz equation with variable coefficients defined on rectangular domains .
we propose a spectral collocation method and a tensor product spectral method .
the first one uses the spectral collocation techniques @xcite and the second one , based on separable representations of the differential operator , combines the fourier spectral method and the ultraspherical spectral method @xcite . for vertically layered medium case , both methods only require to solve a one dimensional problem .
the two spectral methods proposed in this paper can adaptively determine the number of unknowns and approximate the solution to a high accuracy .
we remark that the tensor product spectral method leads to matrices that are banded or almost banded . by employing the fast algorithm in @xcite
, a layered medium approximation problem can be used as a preconditioner for the problem with general media .
recently , chebfun software @xcite was extended to solve problems in two dimensions @xcite , both of which are nonperiodic .
we refer to @xcite for the summary of the mathematics and algorithms of chebyshev technology for nonperiodic functions .
extension to periodic functions in one dimension was proposed in @xcite .
our spectral methods can be used to solve the problems in two space dimensions , one of which is periodic .
implementing our methods with the adaptive strategies of chebfun software is easy .
the rest of this paper is organized as follows . in
2 the model problem is formulated and further reduced to a boundary value problem . in
3 we present the spectral collocation method for the problem .
section 4 is devoted to the tensor product spectral method .
numerical examples illustrating the accuracy and efficiency of the methods are reported in 5 .
we present brief concluding remarks in 6 .
assume that no currents are present and that the fields are source free .
then the electromagnetic fields in the whole space are governed by the following time - harmonic maxwell equations ( time dependence @xmath0 ) & & = h , [ me1 ] + & & = -,[me2]where @xmath1 is the imaginary unit , @xmath2 is the angular frequency , @xmath3 is the permeability , @xmath4 is the permittivity , @xmath5 is the electric field and @xmath6 is the magnetic field . in this paper
, we consider a simple quasi - periodic scattering model : @xmath3 is constant everywhere ; @xmath4 is periodic in @xmath7 variable of period @xmath8 , invariant in @xmath9 variable , and constant away from the region @xmath10 , i.e. , there exist constants @xmath11 and @xmath12 such that & & ( x , y , z)=^+ , y > 1- , + & & ( x , y , z)=^- , y < -1 . in the tm polarization ,
the electric field @xmath5 takes the simpler form @xmath13 the maxwell equations ( [ me1])-([me2 ] ) yield the helmholtz equation : u+ ^2u=0.[reduv1 ] consider the plane wave @xmath14 is incident from the above , where @xmath15 , @xmath16 , and @xmath17 is the angle of incidence with respect to the positive @xmath18-axis .
the incident wave @xmath19 leads to reflected wave @xmath20 and transmitted wave @xmath21 . for @xmath22
, we have @xmath23 , and for @xmath24 , @xmath25 .
we are interested in quasi - periodic solution @xmath26 with phase @xmath27 , i.e. , @xmath28 is @xmath8-periodic in @xmath7 variable . therefore , the reflected and transmitted waves can be written as & & u^r=_j r_j^_jx+_jy , y>1 , + & & u^t=_j t_j^_jx-_jy , y<-1 , where @xmath29 and @xmath30 are unknown complex scalar coefficients and & & _ j=_0+j , + & & _ j=,(_j)0 , + & & _ j= , ( _ j)0.note that if @xmath31 is real , then the @xmath32 s satisfying @xmath33 correspond propagating modes . throughout , we assume that @xmath34 and @xmath35 for all @xmath36 .
this assumption excludes the `` resonant '' cases , where waves can propagate along the @xmath7-axis . for a quasi - periodic function @xmath37 ,
define the linear operators @xmath38 and @xmath39 by @xcite & & ( sf)(x)=_j_j f_j^_j x,[mcals ] + & & ( t f ) ( x)=_j_j f_j ^_j x,[mcalt]where & & f ( x)=_jf_j ^_j x , + & & f_j = _ 0 ^ 2f ( x)^-_j xx.let @xmath40 denote the unit outward normal .
we obtain transparent boundary conditions : [ dtn ] & & _ u - su=-2_0^_0 x-_0,(0,2)\{1 } , + & & _ u - t u=0,(0,2)\{-1}.[dtnn ] the quasi - periodic scattering problem is to solve the helmholtz equation ( [ reduv1 ] ) in the rectangular domain @xmath41 subject to the transparent boundary conditions ( [ dtn])-([dtnn ] ) and the quasi - periodic boundary condition & & u(x+2,y)=^_0 2u(x , y).[qp ] define @xmath42
. then @xmath43 satisfies [ periodic]\ { lll__0 v+^2v=0 , & in & =(0,2)(-1,1 ) , + _ y v-^-_0 xs ( ^_0 xv)=-2_0 ^ -_0 , & on & ( 0,2 ) , + _
xt(^_0 xv)=0 , & on & ( 0,2 ) , + v(x+2,y)=v(x , y ) , & in & ^2 , .where the operator @xmath44 is defined by @xmath45 .
the solution to ( [ periodic ] ) is unique at all but a discrete set of frequencies @xmath2 when the incident angle @xmath46 is fixed .
existence and uniqueness of the solution is strictly proved in dobson @xcite by a variational approach . since the operators @xmath38 and
@xmath39 given in ( [ mcals])-([mcalt ] ) are defined by infinite series , the computation has to be truncated in practice . for simplicity , we truncate @xmath38 and @xmath39 as follows : & & ( s^n f)(x)=_j=1-q^n - q_jf_j^_j x , + & & ( t^n f ) ( x)=_j=1-q^n - q_j f_jt^n ^_j x , where @xmath47 is a sufficiently large integer and @xmath48 is an integer satisfying that @xmath49 and all propagating modes are contained in the middle of the truncated series .
next , we propose two spectral methods for the following problem [ tperiodic]\ { lll__0 v+^2v=0 , & in & =(0,2)(-1,1 ) , + _
y v-^-_0 xs^n(^_0 xv)=-2_0 ^ -_0 , & on & ( 0,2 ) , + _ y v+^-_0 xt^n(^_0 xv)=0 , & on & ( 0,2 ) , + v(x+2,y)=v(x , y ) , & in & ^2 .
.we refer to ( * ? ? ?
* section 3.1 ) for the discussion on the existence and uniqueness of the solution of the problem ( [ tperiodic ] ) . in this paper
, we focus on the accurate and efficient spectral methods for the problem ( [ tperiodic ] ) .
thus we always assume that the discrete problem has a unique solution .
we approximates the problem ( [ tperiodic ] ) by fourier discretization in @xmath7 variable and chebyshev discretization in @xmath18 variable .
let @xmath50 and @xmath51 with @xmath52 , @xmath53 , and @xmath54 . introduce the first - order fourier differentiation matrix @xcite , @xmath55_{i , j=1}^{n}$ ] , n , & & d_ij^x = \ { ll 0 , & i = j , + ( -1)^i - j , & ij , . +
n , & & d_ij^x = \ { ll 0 , & i = j , + ( -1)^i - j , & ij , .the second - order fourier differentiation matrix @xcite , @xmath56_{i , j=1}^{n}$ ] , n , & & d_ij^xx = \ { ll -+ , & i = j , + ( -1)^i - j+1 , & ij , . +
n , & & d_ij^xx = \ { ll -- , & i = j , + ( -1)^i - j+1 ^ 2 , & ij , .and the @xmath57 chebyshev differentiation matrix @xcite , @xmath58_{i , j=0}^m,$ ] & & d_00^y=,d_mm^y=- , + & & d_ii^y=-,i=1, ,m-1 , + & & d_ij^y= , ij , i , j=0,1, ,m , + & & c_i = \ { ll 2 , & i=0 m , + 1 , & .the discretization of the helmholtz equation in ( [ tperiodic ] ) takes the form [ discrete ] * p**v**d*_xx^+2_0*p**v**d*_x^-|_0|^2*p**v*+*p**d*_y^2*v*+^2()=*0*,where @xmath59 is the @xmath60 matrix obtained by deleting the first and last rows of the identity matrix @xmath61 , @xmath62_{m=0,n=1}^{m , n}$ ] , @xmath63 is the approximate solution at the point @xmath64 , @xmath65_{m=0,n=1}^{m , n}$ ] , @xmath66 , and @xmath67 denotes the componentwise product .
the operators @xmath68 and @xmath69 can be approximated by @xmath70 matrices : [ s ] & & * s*=*g*^*_:=*g*^*\{_1-q,_2-q,,_n - q}*g * , + [ t ] & & * t*=*g*^*_:=*g*^*\{_1-q,_2-q,,_n - q}*g*,where @xmath71 is the @xmath70 matrix with @xmath72 entry @xmath73 , @xmath74 .
the discrete forms of the transparent boundary conditions are given by [ beta1 ] & & * e*_0^_y*v*-*e*_0^^=-2_0 ^ -_0 * e*^ , + [ gamma1 ] & & * e*_m^_y*v*+*e*_m^^= * 0*,where & & * e*_0=[1 0 0]^^m+1 , + & & * e*_m=[0 0 1]^^m+1 , + & & * e*=[1 1 1]^^n .
let [ helm]*h*= ( * d*_xx+2_0*d*_x-|_0|^2 * i*_n)+*i*_n(*pd*_y^2)+^2(*i*_n ) \{()}.combining ( [ discrete ] ) , ( [ beta1 ] ) and ( [ gamma1 ] ) yields the linear system [ global ] * a**v*=*g*,where = [ c*i*_n(*e*_0^_y)-*s*_0^ + * h * + * i*_n(*e*_m^_y)+*t*_m^ ] , and @xmath75\in\mbbc^{mn+n}.\ ] ] the matrix @xmath76 in ( [ discrete ] ) is called the downsampling matrix @xcite obtained by using the inner chebyshev points of the second kind,@xmath77 as resampling points .
one might choose other points , such as the chebyshev points of the first kind , @xmath78 we refer to @xcite for explicit construction of rectangular differentiation matrix @xmath79 .
assume that the medium in @xmath80 is vertically layered , i.e. , @xmath81 .
then , the matrix @xmath82 in ( [ helm ] ) takes the form @xmath83 where @xmath84 let @xmath85 be the discrete fourier transform matrix with @xmath72 entry @xmath86 , @xmath87 we have the following propositions .
the proofs are straightforward .
[ pfft ] the matrices @xmath88 in ( [ s ] ) and @xmath89 in ( [ t ] ) satisfy @xmath90 where _ s&=&\{_1^,_2^,,_n^ } + & = & \{_0,_1,,_n - q,_1-q,_2-q,,_-1},and _ t&=&\{_1^,_2^,,_n^ } + & = & \{_0,_1,,_n - q,_1-q,_2-q,,_-1}. [ diag ] the first - order spectral differentiation matrix @xmath91 and the second - order spectral differentiation matrix @xmath92 can be diagonalized by @xmath85 .
specifically , we have @xmath93 where _ x&=&\{_1^x,_2^x,,_n^x } + & = & \ { ll\{0,1,,(n-1)/2,(1-n)/2,(3-n)/2,,-1 } , & , + \{0,1,,n/2 - 1,0,1-n/2,2-n/2,,-1 } , & , .and _ xx&=&\{_1^xx,_2^xx,,_n^xx } + & = & \ { ll-\{0,1,4,,((n-1)/2)^2,((1-n)/2)^2,,4,1 } , & , + -\{0,1,4,,(n/2)^2,(n/2 - 1)^2,,4,1 } , & . . multiplying the linear system ( [ global ] ) by @xmath94\ ] ] from the left yields [ uncoupled] [ c*i*_n(*e*_0^_y)-_s_0^ + ( _ xx+2_0_x-|_0|^2 * i*_n)+*i*_n(*p*(*d*_y^2+^2^y ) ) + * i*_n(*e*_m^_y)+_t_m^ ] = , where @xmath95,\qquad \wh{\bf e}={\bf
fe}=\l[\begin{array}{c}\sqrt{n}\\ { \bf 0}\end{array}\r].\ ] ] reordering the unknowns and equations in ( [ uncoupled ] ) , we obtain @xmath47 uncoupled linear systems of order @xmath96 : [ oned] [ c*e*_0^_y-_0*e*_0^ + * p*(*d*_y^2+^2^y)-|_0|^2*p * + * e*_m^_y+_0*e*_m^ ] _ 1= [ c
-2_0 ^ -_0 + * 0 * + 0 ] , and for @xmath97 , @xmath98\wh{\bf v}_i={\bf 0},\ ] ] where @xmath99 is the @xmath100-th column of @xmath101 . obviously , @xmath102 for @xmath103 . after solving the linear system ( [ oned ] )
, we obtain @xmath104 by @xmath105 .
we propose a tensor product spectral method ( see 4.3 ) for the problem ( [ tperiodic ] ) by combining the fourier spectral method and the ultraspherical spectral method @xcite .
we consider the second - order linear ordinary differential equation ( ode ) [ pode ] w:=a(x)w(x)+b(x)w(x)+c(x)w(x)=f(x),xwith periodic boundary conditions . here , @xmath106 , @xmath107 , @xmath108 and @xmath37 are periodic functions on @xmath109 $ ] .
the fourier spectral method finds an infinite vector @xmath110^\rmt,\ ] ] such that the fourier expansion of the solution of ( [ pode ] ) is given by @xmath111.\ ] ] note that we have @xmath112 the first - order differentiation operator is given by @xmath113,\ ] ] and the second - order differentiation operator is given by @xmath114 in order to handle variable coefficients of the forms @xmath115 , @xmath116 and @xmath117 in ( [ pode ] ) , we need to represent the multiplication of two fourier series as an operator on coefficients .
let @xmath118 $ ] denote the multiplication operator that represents multiplication of two fourier series , i.e. , if @xmath119 is a vector of fourier expansion coefficients of @xmath120 , then @xmath118{\bf w}$ ] returns the fourier expansion coefficients of @xmath121 .
suppose that @xmath106 is given by its fourier series @xmath122 then the explicit formula for @xmath118 $ ] is given by the following toeplitz operator @xmath123=\l[\begin{array}{ccc } \ddots & \vdots & \ddots\\ \ddots & a_{-2 } & \ddots\\ \ddots & a_{-1 } & \ddots\\ \ddots & a_0 & \ddots \\ \ddots & a_1 & \ddots \\ \ddots & a_2 & \ddots \\ \ddots & \vdots & \ddots \end{array}\r].\ ] ] this multiplication operator looks dense ; however , if @xmath106 is approximated by a _ trigonometric polynomial of degree @xmath124 _ , then @xmath118 $ ] is banded with a bandwidth of @xmath124 . combining the differentiation and multiplication operators yields @xmath125\mcald_x^2+\mcalt[b]\mcald_x+\mcalt[c]){\bf w}={\bf f},\ ] ] where @xmath126 and @xmath127 are vectors of fourier expansion coefficients of @xmath120 and @xmath37 , respectively .
we need truncate the operator to derive a practical numerical scheme .
let @xmath128 be the projection operator satisfying @xmath129{\bf w}=\l[\begin{array}{cccc } w_{1-q } & w_{2-q } & \cdots & w_{n - q } \end{array}\r]^\rmt.\ ] ] we obtain the following linear system @xmath130 where @xmath131\mcald_x^2+\mcalt[b]\mcald_x+\mcalt[c].\ ] ] the truncation parameter @xmath47 can be adaptively chosen so that the numerical solution approximates the exact solution to relative machine precision .
consider the second order linear ordinary differential equation [ code]w:=a(x)w(x)+b(x)w(x)+c(x)w(x)=f(x),where @xmath106 , @xmath107 , @xmath108 , @xmath120 and @xmath37 are functions defined on @xmath132 $ ] .
the ultraspherical spectral method finds an infinite vector @xmath133^\rmt\ ] ] such that the chebyshev expansion of the solution of ( [ code ] ) is given by @xmath134,\ ] ] where @xmath135 is the degree @xmath32 chebyshev polynomial .
note that we have the following recurrence relation @xmath136 where @xmath137 is the ultraspherical polynomial with an integer parameter @xmath138 of degree @xmath32 @xcite .
then the differentiation operator for the @xmath139th derivative is given by [ l0]_=2 ^ -1(-1)! [ ccccc * 0 * & & & & + & & + 1 & & + & & & + 2 & + & & & & ] , 1.here @xmath140 in ( [ l0 ] ) denotes a @xmath139-dimensional zero row vector .
the conversion operator converting a vector of chebyshev expansion coefficients to a vector of @xmath141 expansion coefficients , denoted by @xmath142 , and the conversion operator converting a vector of @xmath143 expansion coefficients to a vector of @xmath144 expansion coefficients , denoted by @xmath145 , are given by @xmath146,\quad \mcals_\lambda=\l[\begin{array}{ccccc}1 & 0 & -\frac{\lambda}{\lambda+2 } & & \\ & \frac{\lambda}{\lambda+1}&0&-\frac{\lambda}{\lambda+3}&\\ & & \frac{\lambda}{\lambda+2}&0 & \ddots\\ & & & \frac{\lambda}{\lambda+3}&\ddots\\ & & & & \ddots \end{array}\r],\ \lambda\geq 1.\ ] ] we also require the multiplication operator @xmath147 $ ] that represents multiplication of two chebyshev series , and the multiplication operator @xmath148 $ ] that represents multiplication of two @xmath143 series , i.e. , if @xmath126 is a vector of chebyshev expansion coefficients of @xmath120 , then @xmath147{\bf w}$ ] returns the chebyshev expansion coefficients of @xmath121 , and @xmath148\mcals_{\lambda-1}\cdots\mcals_0{\bf w}$ ] returns the @xmath143 expansion coefficients of @xmath121 .
suppose that @xmath106 has the chebyshev expansion @xmath149 then @xmath147 $ ] can be written as @xcite : @xmath150=\frac{1}{2}\l[\begin{array}{ccccc}2a_0 & a_1 & a_2 & a_3 & \cdots\\ a_1 & 2a_0 & a_1 & a_2 & \ddots \\ a_2 & a_1 & 2a_0 & a_1 & \ddots \\ a_3 & a_2 & a_1 & 2a_0 & \ddots \\ \vdots & \ddots & \ddots & \ddots & \ddots \end{array}\r]+\frac{1}{2}\l[\begin{array}{ccccc } 0 & 0 & 0 & 0 & \cdots\\ a_1 & a_2 & a_3 & a_4 & \cdots \\ a_2 & a_3 & a_4 & a_5 & \iddots \\ a_3 & a_4 & a_5 & a_6 & \iddots \\ \vdots & \iddots & \iddots & \iddots & \iddots \end{array}\r].\ ] ] the explicit formula for the entries of @xmath148 $ ] with @xmath138 is given in @xcite .
this multiplication operators @xmath148 $ ] with @xmath151 look dense ; however , if @xmath106 is approximated by a truncation of its chebyshev or @xmath143 series , then @xmath148 $ ] is banded . combining the differentiation , conversion and multiplication operators yields
[ dcode](_2[a]_2+_1_1[b]_1+_1_0_0[c])*w*=_1_0*f*,where @xmath126 and @xmath127 are vectors of chebyshev expansion coefficients of @xmath120 and @xmath37 , respectively .
we need truncate the operator to derive a practical numerical scheme .
let @xmath152 be the projection operator given by @xmath153.\ ] ] we obtain the following linear system @xmath154 where @xmath155\mcald_2+\mcals_1\mcalm_1[b]\mcald_1+\mcals_1\mcals_0\mcalm_0[c].\ ] ] we seek the approximate solution @xmath157 to the problem ( [ tperiodic ] ) , @xmath158 the linear differential operator in ( [ tperiodic ] ) can be expressed as @xmath159 where @xmath160 , @xmath161 , @xmath162 is the identity operator , and @xmath163 and @xmath164 are the corresponding multiplication operations .
the discretization of the helmholtz equation in ( [ tperiodic ] ) takes the form [ tpfc ] * q**c**v**x*^+*q**y**v*+^2^=*0*,where * q * is the @xmath165 matrix obtained by deleting the last two rows of the identity matrix @xmath166 , @xmath167 , @xmath168 , @xmath169 , @xmath170\mcalp_n^\rmt$ ] , @xmath171\mcalq_m^\rmt$ ] , @xmath172_{i=0,j=1-q}^{m-1,n - q}$ ] .
the discrete forms of the transparent boundary conditions are given by [ beta2 ] & & * b*_1 ^ -*a*_1^_=-2_0 ^ -_0 * e*_q^ , + [ gamma2 ] & & * b*_2^+*a*_2^_= * 0*,where & & _
= \{_1-q,_2-q,,_n - q } , + & & _
= \{_1-q,_2-q,,_n - q } , + & & * b*_1=[0 1 4 ( m-1)^2]^ , + & & * b*_2=[0 1 -4 ( -1)^m(m-1)^2]^ , + & & * a*_1=[1 1 1]^^m , + & & * a*_2=[1 -1 ( -1)^(m-1)]^^m , and @xmath173 is the @xmath48-th column of the identity matrix @xmath174 .
let [ helmfc]*h*= * x*(*qc*)+*i*_n(*qy*)+^2(*q * ) . combining ( [ tpfc ] ) ,
( [ beta2 ] ) and ( [ gamma2 ] ) yields the linear system [ globalfc ] * a**v*=*g*,where = [ c*i*_n_1 ^ -__1^ + * i*_n_2^+__2^ + * h * ] , and @xmath175\in\mbbc^{mn}.\ ] ] for general bivariate function @xmath176 , we can use its low rank approximant , i.e. , sum of functions of the form @xmath177 , where @xmath178 and @xmath179 are univariate functions .
an algorithm @xcite which is mathematically equivalent to gaussian elimination with complete pivoting can be used to construct low rank approximations . in this case , we have @xmath180 , i.e. , @xmath181 .
then @xmath182 the matrix @xmath82 in ( [ helmfc ] ) takes the form @xmath183 reordering the unknowns and equations in ( [ globalfc ] ) , we obtain [ onedfc] [ c*b*_1 ^ -_0*a*_1^ + * b*_2^+_0*a*_2^ + * q*(*y*+^2-|_0|^2*c * ) ] * v*_q = [ c -2_0 ^ -_0 + 0 + * 0 * ] , and [ ofc] [ c*b*_1 ^ -_j - q*a*_1^ + * b*_2^+_j - q*a*_2^ + * q*(*y*+^2-((j - q)^2 + 2_0(j - q)+|_0|^2)*c * ) ] * v*_j=*0*,jq , where @xmath184 is the @xmath32-th column of @xmath185 . obviously , @xmath186 for @xmath187 . note that the fast algorithm in @xcite can be used to solve the linear system ( [ onedfc ] ) , which requires @xmath188 operations . here
, @xmath189 is the number of chebyshev points needed to resolve the function @xmath179 .
the coefficient matrix resulting from a layered medium approximation problem can be used as a preconditioner for the problem with general medium .
the corresponding computational complexity for the preconditioner solve is @xmath190 .
we have performed numerical experiments in numerous cases . in this section
, we present a few typical results of these experiments to illustrate the accuracy and efficiency of the two spectral methods .
all computations are performed with matlab r2012a .
the parameters are chosen as @xmath191 , @xmath192 , @xmath193 , @xmath194 .
we consider three cases : & & _ 1(x , y)=(y)=1+^+4 , + & & _ 2(x , y)=1+^+4-((x/2 ) ) , + & & _ 3(x , y)=1+^+4-y((x/2 ) ) .
the medium characterized by @xmath195 is vertically layered .
the bivariate function @xmath196 is of rank @xmath197 .
we use the rank @xmath198 approximant obtained by the algorithm proposed in @xcite to approximate @xmath199 . in figure [ nr ]
, we plot the three media and the real parts of the corresponding waves obtained by our spectral methods .
we observe that the numerical results obtained by the two methods coincide well for all the three media .
we also test the performance of the problem with @xmath195 as a preconditioner for the problem with @xmath199 .
the fast algorithm in @xcite is used to solve the linear systems ( [ onedfc ] ) and ( [ ofc ] ) .
the gmres algorithm @xcite is used as the iterative solver .
the initial guess is set to be the zero vector .
gmres with the layered medium preconditioner obtained a solution with the relative residual norm less than @xmath200 at the @xmath201th iteration , while gmres without preconditioning almost stagnates ; see figure [ conhis ] for the convergence history .
we have proposed a spectral collocation method and a tensor product spectral method for solving the 2d quasi - periodic scattering problem . both of the methods can adaptively determine the number of unknowns and approximate the solution to a high accuracy .
the tensor product spectral method is more interesting because it leads to matrices that are banded or almost banded , which enable the use of the fast , stable direct solver @xcite .
based on this fast solver , a layered medium preconditioning technique is used to solve the problem with general media .
our methods also apply to solve general partial differential equations in two space dimensions , one of which is periodic .
extension of our methods to bi - periodic structure diffraction grating problem @xcite is being considered . | we consider the 2d quasi - periodic scattering problem in optics , which has been modelled by a boundary value problem governed by helmholtz equation with transparent boundary conditions . a spectral collocation method and a tensor product spectral method
are proposed to numerically solve the problem on rectangles .
the discretization parameters can be adaptively chosen so that the numerical solution approximates the exact solution to a high accuracy .
our methods also apply to solve general partial differential equations in two space dimensions , one of which is periodic .
numerical examples are presented to illustrate the accuracy and efficiency of our methods .
helmholtz equation , transparent boundary condition , spectral method , chebfun 65m70 , 65t40 , 65t50 , 78a45 |
we consider two - level systems as described by the following hamiltonian @xmath114 where @xmath1 ( @xmath2 ) represent pauli operators , noise can occur along all three axes @xmath115}\sigma_{\alpha}^k$ ] .
given the usual case of @xmath116 , the effect of transverse noise is suppressed by the energy gap @xmath8 .
thus the main decoherence comes from the longitudinal noise @xmath117 } \sigma_{z}^k,\ ] ] in the form of dephasing . and relative fluctuation @xmath118 .
( b ) the decay of coherence as a function of time for hybrid dressed spin ( hybrid ds ) .
the driving amplitude is @xmath30 .
, title="fig:",width=245 ] and relative fluctuation @xmath118 .
( b ) the decay of coherence as a function of time for hybrid dressed spin ( hybrid ds ) .
the driving amplitude is @xmath30 . , title="fig:",width=245 ] in the case of local environment noise , it is not helpful to use a pair of two - level systems to encode an effective spin-@xmath12 as there is no decoherence free subspace , see fig .
[ fig : lifetime - si](a ) .
one can introduce dressed state by applying a continuous driving field with amplitude and phase fluctuation @xmath119}{\left [ e^{-i\omega_0 t - i\delta_{p}(t)}\sigma_+^k+e^{i\omega_0 t + i\delta_{p}(t)}\sigma_-^k \right]},\ ] ] where @xmath120 and @xmath121 represent the amplitude and phase fluctuation of the driving field .
the effect of @xmath122 is suppressed by the driving field , as long as its power spectra is negligible at the frequency of @xmath7 . in the interaction picture with respect to @xmath9 ,
the effective hamiltonian including driving field fluctuations is @xmath123}{\left
[ \cos\delta_{p}(t ) \sigma_x^k+\sin \delta_{p}(t ) \sigma_y^k \right]}\\ & = & \frac{\omega}{2 } \sigma_x^k+\frac{\tilde{\delta}_{\omega}(t)}{2 } \sigma_x^k+{\left [ \frac{\omega}{2 } + \frac{\delta _ { \omega}(t)}{2 } \right]}\sin \delta_{p}(t ) \sigma_y^k , \label{eq : si - phase-1}\end{aligned}\ ] ] where @xmath124}. \label{eq : si - phase}\ ] ] the phase fluctuation @xmath121 in eq.([eq : si - phase-1 ] ) shows two effects .
first , it leads to additional fluctuation in the amplitude of driving , and the effective total amplitude fluctuation is given by eq.([eq : si - phase ] ) .
secondly , it also translates into the flip error ( @xmath125 ) of dressed states , which is suppressed if the fluctuation is small such that @xmath126 .
thus , decoherence mainly comes from the amplitude fluctuation of driving described as follows @xmath127 generally , the amplitude fluctuation becomes larger as the driving power increases . by increasing the driving field amplitude , on the one hand ,
it is beneficial for the suppression of environment noise , on the other hand , the accompanying increase in the driving field fluctuations would diminish the overall performance , as shown fig .
[ fig : lifetime - si](a - b ) .
the essence of the hybrid dressed state is using a pair of two - level systems under identical driving to carry one single effective spin-@xmath12 as @xmath128 because the following conditions are satisfied @xmath129}{\left\vert \tilde{\uparrow } \right\rangle}&=&0,\\ { \left [ \tilde{\delta}_{\omega}(t)(\sigma_x^{(k , l)_a}+ \sigma_x^{(k , l)_b } ) \right]}{\left\vert \tilde{\downarrow } \right\rangle}&=&0,\end{aligned}\ ] ] hybrid dressed states are thus robust not only to environment noise but also to the amplitude and phase fluctuation in driving . in our numerical simulation , we model the environment noise @xmath32 and the amplitude fluctuations of the driving fields @xmath120 as an ornstein - uhlenbeck process @xcite with the correlation function @xmath33 , where @xmath34 is the noise amplitude and @xmath35 is the noise correlation time .
we also include the fluctuating phase @xmath121 in our model where it takes a random value uniformly distributed in @xmath130}$ ] .
it can be seen from fig .
[ fig : lifetime - si](b ) that coherence times of hybrid dressed state are significantly prolonged , and is not susceptible to the fluctuations the continuous driving field .
the scaling of coherence times is shown in fig.2 of main text .
+ ) remains high for the time while the decay of coherence for a simple dressed spin ( simple ds ) happens in a much shorter time scale . *
( b ) * the coherent oscillation of hybrid dressed spin ( hybrid ds ) state population during the evolution .
the relevant parameters are as follows : the noise amplitude is @xmath131 , the amplitude of driving field is @xmath30 with a relative fluctuation is @xmath132 ; and the nearest - neighbor coupling is @xmath133 .
the noise and fluctuation correlation time is @xmath29.,width=438 ] we consider a pair of nv centers with addressable electronic transition frequencies @xcite to prepare ( entangled ) hybrid dressed states .
we drive both nv centers with the same rabi frequency , the interaction between nv centers is then given by @xmath134}+a \mathbf{s}_z^{(1)}\otimes \mathbf{s}_z^{(2 ) } , \label{eq : spin - spin_interaction - si}\ ] ] where @xmath135 represent the spin-@xmath12 operators in the subspace spanned by @xmath47 .
initially we prepare @xmath136 with microwave-@xmath42 pulses acting on the ground spin sublevels of nv centers , where @xmath137 and @xmath138 . after an interaction time @xmath48
, the system evolves into the state @xmath49 .
the state preparation fidelity is severely limited by local magnetic noise , particularly when the spin - spin coupling is even weaker than noise , namely @xmath139 . in fig.[fig : ising - coupling - si ] , we simulate the state preparation of ( entangled ) hybrid states , and shows a high fidelity ( @xmath140 ) under realistic noise and driving fluctuation . [ [ derivation - of - tunable - interaction - between - hybrid - dressed - states - with - nv - centers ] ] derivation of tunable interaction between hybrid dressed states with nv centers ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ the tunability of hamiltonian parameters is necessary for quantum simulation . in order to achieve tunable interaction between nv centers from the spin - spin interaction @xmath141 , we use a microwave field with rabi frequency @xmath142 , and frequency detuning @xmath143 , the effective hamiltonian in the interaction picture is @xmath144 } , \label{eq - si : hxm}\end{aligned}\ ] ] where @xmath145 and @xmath146 .
the above hamiltonian can be diagonalized as @xmath147 where the corresponding eigenstates are @xmath148 the spin - spin interaction @xmath149 can then be written in the basis of @xmath150 ( @xmath151 ) as @xmath152}\bigotimes { \left [ \cos{(\theta_l)}\sigma_{z}^{l}+\sin{(\theta_l)}\sigma_{x}^{l } \right]}.\label{eq - s17}\end{aligned}\ ] ] under the assumption that @xmath153 , the cross interaction terms @xmath154 and @xmath155 can be neglected , and thus in the interaction picture with respect to eq.([eq - si : hxm_d ] ) we obtain @xmath156}. \label{eq - si : samer}\end{aligned}\ ] ] we note that this microwave field is exploited to achieve tunable interaction between nv centers , thus we do not require that it should be identical on all nv centers .
for example , one can use @xmath157 and @xmath158 for two nv centers that carry a single hybrid spin-@xmath12 , and in this way the interaction becomes @xmath159}.\label{eq - si : altr}\end{aligned}\ ] ] according to the trotter decomposition , the effective interaction between two nv centers from choosing alternating rabi frequencies ( i.e. eq.[eq - si : samer]-[eq - si : altr ] ) becomes @xmath160},\label{eq - si : hkl - eff}\end{aligned}\ ] ] where @xmath161}/2 $ ] and we will see that the term @xmath162 can be eliminated in the subspace of hybrid dressed states .
therefore , we obtain the following effective hamiltonian for hybrid dressed state as @xmath163}.\label{eq - si : hkl - eff - xz}\end{aligned}\ ] ] to achieve robust features against both local magnetic noise and fluctuation in driving fields , we can introduce hybrid dressed states starting from eq.([eq - si : hxm_d ] ) , similar as eq.(1 ) in the main text . in the same interaction picture of eq.([eq - si : hxm ] ) , the noise and the driving field fluctuation is described by @xmath164 written in the rotated basis of @xmath150 , we obtain @xmath165}\sigma_x^m+{\left [ \frac{\delta\omega_m(t)}{2}\sin\theta_m+\frac{\delta b_m(t)}{2}\cos\theta_m \right]}. \sigma_z^m .
\label{eq - si : hnoise}\ ] ] in the interaction picture with respect to eq.([eq - si : hxm_d ] ) , the flip effect @xmath166 in @xmath167 ( see eq.[eq - si : hnoise ] ) is suppressed , and thus @xmath168 } \sigma_z^m.\label{eq - si : hnoise}\ ] ] in order to suppress this noise , we introduce continuous driving fields , which are presented in the lab frame as follows : @xmath169 moving to the interaction picture with respect to the bare energy structure yields @xmath170 in the rotating wave approximation ( rwa ) assuming @xmath171 .
we now transform to the rotated basis and move to the second interaction picture with respect to eq .
[ eq - si : hxm_d ] . therefore ,
in the rwa assuming @xmath172 , we obtain the continuous driving fields @xmath173}=-\frac{\omega'_m}{2}\sigma_x , \label{second}\ ] ] which would then suppress the noise in eq.([eq - si : hnoise ] ) .
the remaining noise comes from the fluctuation of the driving field @xmath174 the total effective hamiltonian is @xmath175 the corresponding dressed states ( i.e. the eigenstates of @xmath176 ) are @xmath177 and the hybrid dressed states are @xmath178 as we discuss in the main text , the above hybrid dressed states are robust against both local magnetic noise and fluctuations of driving fields . in the following , we will derive the tunable hamiltonian of coupled nv centers by writing the total effective hamiltonian in the basis of hybrid spin-@xmath12 .
for the intra - pair interaction ( between nv centers that carry one hybrid dressed spin-@xmath12 ) , because @xmath179 in eq.[eq - si : hkl - eff ] , is @xmath180}.\end{aligned}\ ] ] this gives the following local hamiltonian of hybrid dressed spin-@xmath12 @xmath181 with @xmath182},\ ] ] where @xmath183 .
for the inter - pair interaction between nv centers that carry different hybrid dressed spin-@xmath12 , @xmath184 for ( @xmath88 ) will flip the dressed states @xmath185 , and @xmath186 .
therefore , it can be effectively eliminated if @xmath187}$ ] .
this explains why we can obtain the effective hamiltonian for hybrid dressed states in eq.[eq - si : hkl - eff - xz ] .
similarly , @xmath87 can also be eliminated .
this gives the effective interaction between hybrid dress spin-@xmath12 as @xmath188}x_{k}x_{l},\ ] ] where @xmath189 .
thus , we obtain the following tunable effective hamiltonian @xmath190}x_{k}x_{l},\ ] ] an example of which is shown in eq.(7 ) of the main text for two hybrid dressed spin-@xmath12 .
note that in order to induce the trotter decomposition , by alternating @xmath191 , the driving fields should be changed correspondingly .
namely , we should alternate the detuned driving field s rabi frequency @xmath192 and detuning @xmath193 ( eq . [ eq - si : hxm ] ) , and the rabi frequency of the continuous driving fields @xmath194 ( eq . [ decoupling ] ) as well . to achieve @xmath195 interactions
, we apply counter - propagating raman beams in addition to a resonant carrier transition @xcite , as presented in the following hamiltonian : @xmath196 \label{raman1}\\ & + & \omega_0\left ( { \left\vert e \right\rangle}^i{\left\langle 0 \right\vert}+h.c\right ) \cos \left [ ( \omega_e+\frac{\omega_0}{2}-\delta-\delta)t -\eta_{i , n } ( b_n^\dagger + b_n ) \right]\label{raman2}\\ & + & \omega_c\sigma^i_x\cos \left({\omega_0}t \right)\label{dressing1}. \ ] ] here , eq.([phonons ] ) is the vibration hamiltonian that is described by the creation ( annihilation ) operators @xmath197 for the @xmath198 vibrational mode . eq.([bare ] ) is the bare energy structure where @xmath199 is the virtually excited state , in the optical regime , through which the raman transitions ( eq.[raman1]-[raman2 ] ) are operated . @xmath200 and @xmath201 , are the pauli matrices that operate on the qubit states of the @xmath202 ion , and @xmath203 is the lamb - dicke parameter of the @xmath202 ion and the @xmath198 vibrational mode . finally , eq.([dressing1 ] ) is the dressing field s term .
the hamiltonian in the interaction picture with respect to the bare energy structure ( eq .
[ bare ] ) becomes : @xmath204}+h.c \right)+\frac{\omega_0}{2 } \left ( { \left\vert e \right\rangle}^i{\left\langle 0 \right\vert}e^{i\left[(\delta+\delta)t + \eta_{i , n } ( b_n^\dagger + b_n ) \right]}+h.c\right ) \\ & + & \frac{\omega_c}{2}\sigma^i_x \label{dressed } , \ ] ] where we used the rwa assuming @xmath205 and @xmath206 . in the second order of perturbation theory , we obtain the following effective hamiltonian : @xmath207,\label{eff_xx } \end{split}\ ] ] where we neglected a.c .
stark shifts . since the a.c
stark shifts are described by @xmath208 operators , they can be eliminated by adjusting the detunings of the raman beams , and in any case , their first order contribution is suppressed by the dressing field @xmath209 ( eq.[dressed ] ) . in the next step , we expand the exponent of the displacement operator to the first order in the lamb - dicke parameter . the zeroth order , gives rise to a detuned carrier transitions , resulting in an a.c stark shift that is suppressed from the above reasons . in the rotating frame of the vibrational modes ( eq.[phonons ] ) and the dressed state energy gap ( eq.[dressed ] ) , the operator @xmath210 is effectively reduced to @xmath211 since @xmath212 is suppressed if @xmath213 ; thus the first order expansion yields the srensen - mlmer hamiltonian : @xmath214 which results in the effective ising hamiltonian , @xmath215 in the second order of perturbation assuming @xmath216 . to achieve the @xmath217 interactions
, we can introduce an additional rotated term , which breaks the ising coupling to obtain the xxz hamiltonian @xcite , similarly to the derivation for the nv centers ( eq.[eq - si : hxm ] ) . for realizing the this method we apply co - propagating focused raman beams , thus having single addressing ability .
the driving fields in the lab frame are presented as @xmath218 \label{raman'1}\\ & + & \omega^m_0\left ( { \left\vert e \right\rangle}^i{\left\langle 0 \right\vert}+h.c\right ) \cos \left [ ( \omega_e+\frac{\omega_0}{2}-\delta_2-\delta_m)t + \eta_{i , n } ( b_n^\dagger + b_n ) \right]\label{raman'2}.\end{aligned}\ ] ] after moving to the interaction picture with respect to the bare energy structure ( eq.[bare ] ) , and using the rwa where we assume @xmath219 , we obtain @xmath220}+h.c \right)+\frac{\omega^m_0}{2 } \left ( { \left\vert e \right\rangle}^i{\left\langle 0 \right\vert}e^{i\left[(\delta_2+\delta_m)t -\eta_{i , n } ( b_n^\dagger + b_n ) \right]}+h.c\right ) . \ ] ] in the second order of perturbation theory , this yields @xmath221 which is detuned from the dressed state energy structure ( eq.[dressed ] ) by @xmath222 .
now , if we move to the interaction picture with respect to @xmath223 , the only term of eq.[eff2 ] that preserves energy and thus survives in the rwa , is @xmath224 , with @xmath225 .
thus , together with @xmath226 , which is left from the dressed state energy gap ( eq.[dressed ] ) , we obtain the rotated term needed for breaking the ising coupling into the xxz one , similarly to eq.[eq - si : hxm ] @xmath227 } , \end{split}\ ] ] where @xmath228 and @xmath229 .
this hamiltonian can be diagonalized as @xmath230 where the corresponding eigenstates are @xmath231 eq . [ sm interaction ] can then be written in the basis of @xmath150 ( @xmath151 ) as @xmath232 thus , in the interaction picture with respect to eq.[second interaction ] we obtain the effective hamiltonian @xmath233 } , \label{xxz hamiltonian}\end{aligned}\ ] ] if we assume that @xmath234 . in order to suppress the leakage coming from an undesired @xmath235 coupling
, we can demand @xmath236 , where @xmath237 are the notations of different blocks .
the validity of the effective hamiltonian is supported by our numerical simulation with realistic experiment parameters , as shown in fig.[fig - si : ion_sim ] .
-[eff2 ] ) ( solid lines ) , and show the equivalence to eq.([xxz hamiltonian ] ) ( markers ) for different @xmath238 .
we initialize the system with @xmath239 and the motional ground state@xmath240 , and measure the reduced density matrix s elements of : @xmath241 ( cyan circles ) , @xmath242 ( cyan pluses ) , @xmath243 ( red exes ) , and @xmath244 ( green squares ) , as usually done when measuring entanglement . for that purpose ,
we use the following parameters : @xmath245khz , @xmath246 , @xmath247mhz , @xmath248mhz , @xmath249mhz . * ( a ) * for @xmath250 we use @xmath251 and @xmath252mhz . * ( b ) * for @xmath253 we use @xmath254khz and @xmath255khz . * ( c ) * for @xmath256 we use @xmath257khz and @xmath258khz . *
( d ) * for @xmath259 we use @xmath260khz and @xmath261.,width=566 ] f. dolde , i. jakobi , b. naydenov , n. zhao , s. pezzagna , c. trautmann , j. meijer , p. neumann , f. jelezko , j. wrachtrup , http://www.nature.com/nphys/journal/v9/n3/full/nphys2545.html[_nature physics _ * 9 * , 139 ( 2013 ) . ] | a fundamental goal of quantum technologies concerns the exploitation of quantum coherent dynamics for the realisation of novel quantum applications such as quantum computing , quantum simulation , and quantum metrology .
a key challenge on the way towards these goals remains the protection of quantum coherent dynamics from environmental noise .
here , we propose a concept of hybrid dressed state from a pair of continuously driven systems .
it allows sufficiently strong driving fields to suppress the effect of environmental noise , while at the same time being insusceptible to both the amplitude and phase noise in the continuous driving fields .
this combination of robust features significantly enhances coherence times under realistic conditions , and at the same time provides new flexibility in hamiltonian engineering that otherwise is not achievable .
we demonstrate theoretically applications of our scheme for noise resistant analog quantum simulation in the well studied physical systems of nitrogen - vacancy centers in diamond and of trapped ions .
the scheme may also be exploited for quantum computation and quantum metrology . _
introduction. _ quantum technologies hold the promise for the realisation of a wide variety of applications including quantum computing @xcite , quantum simulation @xcite , quantum metrology @xcite and precision measurements @xcite as well as quantum sensing @xcite to name just a few .
this motivates the considerable effort that is being invested in the creation of the technological basis for these devices with the ultimate goal of constructing quantum devices that can outperform their classical counterparts .
one of the main obstacles on this path is the effect of noise and decoherence due to interaction with an uncontrolled environment , the effect of which becomes increasingly severe as the number of system components grows .
this poses a considerable challenge for achieving the quantum control of such systems while maintaining the quantum coherence of the system .
hence noise control is central for the future development of scalable quantum technologies .
various theoretical concepts and proposals , e.g. quantum error correction @xcite , decoherence free subspace @xcite , and dynamical decoupling @xcite , have been developed for the suppression or avoidance of quantum decoherence .
each of these methods are best suited for specific scenarios . for example
, the decoherence free subspace approach is mainly useful in those cases in which noise exhibits a symmetry @xcite , which however is not always the case .
one important example among others that suffer from non - collective noise is the nitrogen - vacancy ( nv ) center in diamond @xcite as a promising physical system for quantum information processing and quantum sensing , the noise of which is dominated by the local nuclear and paramagnetic environment .
these noise sources exhibit relatively long memory times and thus can be suppressed using continuous driving .
this approach however suffers drawbacks for strong driving where intensity and phase fluctuations may become significant @xcite .
it is an important observation , that for decoupling fields that are derived from the same source , these fluctuations will be strongly correlated .
hence , we propose to exploit this feature by pairing two driven systems into hybrid dressed states .
the energy gap arising from continuous driving protects quantum coherence in non - markovian environments and makes it insusceptible to inhomogeneous local environment noise , while the pairing approach makes it robust against the driving fluctuation . in comparison with dynamical decoupling in decoherence free subspace , see e.g. @xcite , the present idea of hybrid dressed state provides new flexibility in the engineering of effective hamiltonian . with detailed numerical studies ,
we demonstrate that the present scheme can prolong coherence times by more than two orders of magnitude under realistic conditions .
we exploit hybrid dressed states to implement quantum gates and quantum simulation with significantly improved fidelity , e.g. using nv centers in diamond , which thus provides a route towards a large - scale reliable quantum simulation .
it is also feasible to verify the idea with other physical systems such as trapped ions as will be demonstrated here by detailed simulations . _
the proposal. _ we start from single two - level systems with the basic hamiltonian @xmath0 where @xmath1 ( @xmath2 ) represent pauli operators .
the coherence of a two - level system , e.g. nv center in diamond , would inevitably be affected by local environmental noise , such as magnetic field and/or charge fluctuation , which can be described by @xmath3}\sigma_{\alpha}^k$ ] .
generally , @xmath4 are independent for different two - level systems . the dressed states induced by a continuous driving field @xmath5 are protected from noise by an energy penalty @xcite , where @xmath6 , @xmath7 and @xmath8 are the amplitude and frequency of the resonant driving field respectively . in the interaction picture with respect to @xmath9 ,
the effective system hamiltonian becomes @xmath10 whose eigenstates @xmath11 can be used to represent a ( simple ) dressed spin-@xmath12 that is protected by the continuous driving as long as the power spectrum of the noise @xmath13 at the frequency @xmath7 is negligible .
the phase fluctuations in the continuous driving field have two effects , one to flip the dressed states , which would be suppressed by the energy penalty as well , while the other leads to effective amplitude fluctuations @xcite .
the main decoherence source is therefore due to the amplitude fluctuations of continuous driving fields and the residual effect of environment noise due to a finite continuous driving field strength , which causes dephasing of dressed states .
it is important to note that if , as is often the case , the decoupling fields are derived from the same source , their fluctuations will be correlated - a feature that we will exploit now .
system is carried by a pair of two - level systems ( with ising - like coupling strength @xmath14 ) under continuous driving ( with a rabi frequency @xmath7 ) . *
( b ) * the corresponding energy structure of hybrid dressed states .
, width=302 ] to overcome these challenges and making use of the correlated nature of the driving fluctuations , we propose to use a pair of dressed states to carry one effective spin-@xmath12 , see fig.[fig : model ] , represented by two of the eigenstates of @xmath15 as @xmath16 the states @xmath17 and @xmath18 that represent the hybrid dressed spin-@xmath12 are separated from the other two eigenstates @xmath19 and @xmath20 by an energy gap @xmath21 .
the dephasing noise would induce the spin flip as @xmath22 and vice versa , the effect of which on the hybrid dressed spin-@xmath12 can be suppressed by tuning the amplitude of the driving field @xmath7 to sufficiently high values so that the relevant noise power spectra become insignificant .
the amplitude fluctuations in the driving field , which are described as @xmath23}\sigma_x^{k}$ ] , would however cause dephasing of the simple dressed spin-@xmath12 as @xmath24}{\left\vert { \uparrow_x } \right\rangle}$ ] and @xmath25}{\left\vert { \downarrow_x } \right\rangle}$ ] .
thanks to the correlations in the amplitude fluctuations , the choice of hybrid dressed spin-@xmath12 states , renders the effective spin-@xmath12 instead insusceptible to such fluctuations , because @xmath26}{\left\vert \tilde{\uparrow } \right\rangle}({\left\vert \tilde{\downarrow } \right\rangle})=0 $ ] .
this allows us to apply strong driving while being immune to their fluctuations .
for the cases of no dynamical decoupling and for simple dressed spin ( simple ds ) with driving fluctuation of @xmath27 .
( b ) the coherence time of hybrid dressed spin ( hybrid ds ) as a function of the driving fluctuation as compared with a simple dressed spin .
the other parameters are : the noise amplitude @xmath28 , and the correlation time is @xmath29 , the driving amplitude @xmath30 , and the phase fluctuates in @xmath31$].,width=332 ] in fig.[fig : lifetime ] , we plot the decay of quantum coherence of hybrid dressed states under environment noise and driving fluctuations .
the noise ( fluctuation ) @xmath32 is modeled as an ornstein - uhlenbeck process @xcite with the correlation function @xmath33 , where @xmath34 is the noise amplitude and @xmath35 is the noise correlation time .
we initially prepare a coherent superposition of two basis states , namely @xmath36 for hybrid dressed spin , and @xmath37 for a simple dressed spin .
we characterize quantum coherence by @xmath38 , and define coherence time as @xmath39 when @xmath40 .
it can be seen from fig.[fig : lifetime](a ) that the fluctuation of the continuous driving field indeed would severely limit its efficiency .
we demonstrate in fig.[fig : lifetime](b ) that hybrid dressed states are very robust against the driving fluctuation and can prolong the coherence time by more than two orders of magnitude even for relatively large amplitude and phase fluctuations .
we illustrate the application of this basic principle by considering two spins , such as nv centers in nanodiamonds as well as shallowly implanted nv - centers for sensing applications .
we stress that the same principle is applicable for other physical systems where noise is local and strong driving fields are necessary for decoupling the systems from high level noise . with the ability to address individual spins
@xcite , we first prepare the initial dressed states @xmath41 with microwave-@xmath42 pulses , where @xmath43 and @xmath44 .
we drive both spins with the same rabi frequency and consider the interaction between spins given by @xmath45 , where @xmath46 represent the spin-@xmath12 operator in the subspace spanned by @xmath47 @xcite . the interaction for time
@xmath48 evolves the system into @xmath49 .
the robustness of hybrid dressed states allows us to measure inter - spin interactions and thus the distance between ( adjacent ) spins with a high precision .
this is similar in spirit to the experiment with trapped ions @xcite .
there , however , correlated noise dominated while local noise suffered by individual system is always non - collective and needs to be preliminarily addressed by dynamical decoupling .
we are thus able to prepare a maximally entangled hybrid dressed state with a very high fidelity , even when the required interaction time significantly exceeds the spin coherence times @xcite .
such a robust hybrid dressed state can also be exploited to construct a sensitive gradient magnetometer based on a pair of dressed spins , e.g. a pair of linked nanodiamonds @xcite . _
tunable coherent coupling and quantum simulation. _ for the purpose of quantum simulation , it is necessary to engineer _ tunable _ coherent interaction between hybrid dressed states .
we consider two pairs of spins under continuous driving , that interact with each other via @xmath50 @xcite . with appropriately tuning the amplitude @xmath7 and frequency ( @xmath51 ) of the continuous dressing field , one can obtain the following general form of dressed states
@xmath52 and @xmath53 , where @xmath54 .
to make the hybrid dressed states robust against noise and driving fluctuation , we introduce continuous decoupling fields with rabi frequencies @xmath55 acting on the simple dressed states @xmath56 , so that the protection part hamiltonian is @xmath57 . assuming @xmath58 ( @xmath59 )
, we obtain from the original spin - spin interaction the following effective interaction hamiltonian between two hybrid dressed spin-@xmath12 , which are carried by ( @xmath60 ) and ( @xmath61 ) respectively @xcite , @xmath62 where the pauli operators are written in the basis @xmath63 , and the coupling strength @xmath64 , @xmath65 .
this leads to the flexibility in tuning effective spin - spin interactions by controlling the driving parameters .
the system hamiltonian written in the basis of hybrid dressed states is @xcite @xmath66 where the effective spin-@xmath12 operators are defined as @xmath67 .
the local energy is given by @xmath68 , and the coupling strength is @xmath69 . in the particular case @xmath70 , namely @xmath71
, the dressing field also acts the role of dynamical decoupling , thus we have @xmath72 @xcite , which allows to efficiently realize high - fidelity entangling gates between hybrid dressed spins . assuming equal ( intra- and inter- pair ) nearest - neighbor interactions @xmath14 and non - nearest - neighbor couplings as @xmath73 , the effective nearest - neighbor coupling is @xmath74 , while @xmath75 . in fig.[fig : ising - coupling ] , it can be seen that the fidelity of maximal entanglement generation at time @xmath76 is high ( above @xmath77 ) even in the case where the noise / fluctuation is stronger than the coupling strength . ) under realistic noise and driving fluctuation , with respect to the ideal entangling evolution , remains high for the time while the coherence of individual simple dressed spin ( simple ds ) decays rapidly . *
( b ) * the coherent oscillation of state population during the entangling evolution . the maximally entangled hybrid dressed state is generated at time @xmath76 ( green star ) .
the relevant parameters are : the noise amplitude @xmath78 , the driving amplitudes @xmath79 , @xmath80 with a relative fluctuation @xmath81 ; and the nearest - neighbor coupling is @xmath82 ( corresponding to a distance of @xmath83 nm ) .
the noise and fluctuation correlation time is @xmath29.,width=321 ] the generalization of coupling in eq.([eq : effective_coupling ] ) to more two - level systems on a regular lattice , see fig.[fig : ising - model](a ) , e.g. self - assembled nanodiamond lattice @xcite and trapped ion lattice @xcite , thus leads to hybrid dressed states protected robust simulation of quantum ising model @xmath84 .
the coherence protection provided by @xmath85 is independent of the parameters of the simulating hamiltonian in @xmath86 .
long - range dipolar interaction , namely the interaction @xmath87 for @xmath88 in eq.([eq : hdd_coupling ] ) , should be effectively eliminated , as it would cause the leakage out of the protected subspace . with @xmath89 periodically spatial alternating driving amplitudes ( @xmath90^n$ ] ) for a one - dimensional chain so that @xmath91 , see the inset of fig.[fig : ising - coupling](a ) for the case of @xmath92 , the unwanted interactions can be suppressed up to the @xmath93-neighbor coupling .
for example , the residual interaction is @xmath94 with @xmath95 for @xmath96 ( @xmath97 ) scaling long range interactions .
analogous techniques can be employed for the general case of two - dimensional lattice . for one - dimensional chain , the effective coupling strength scales as @xmath98 where @xmath99 , see fig.[fig : ising - model](b ) , where @xmath100 is the original interaction range .
this enables us to achieve a shorter interaction length that otherwise is not possible @xcite .
furthermore , as we mentioned , the dependence of the interaction on the driving basis , see eq.([eq : hdd_coupling ] ) , provide additional flexibility in engineering the simulating hamiltonian parameters .
in contrast to the digital fashion of quantum simulation in decoherence free subspace @xcite , the present analog quantum simulation will save precious time in case of simulating a large spin system . of the index @xmath101
is represented by driven two - level systems @xmath102 and @xmath103 . *
( b ) * the effective spin - spin interaction range @xmath104 as a function of the original interaction range @xmath100 . *
( c ) * the spin structure factor @xmath105 with @xmath106 achieved via adiabatic preparation for a one - dimensional eight - site ising chain using hybrid dressed spins ( hybrid ds ) and simple two - level systems ( simple tls ) .
the parameters are @xmath107 , @xmath108 with @xmath109 and @xmath110 . , width=325 ] to show the performance of hybrid dressed states in adiabatic quantum simulation , we calculate the spin structure factor @xmath105 with @xmath106 for a one - dimensional ising chain . in fig.[fig : ising - model ] ( c ) , it can be seen that the transition of the order parameter @xmath111 across the critical point becomes much smoother due to the decoherence effect of noise .
one would expect that it will be difficult to observe quantum phase transition as the noise level and/or the system size increases .
in contrast , we show in fig.[fig : ising - model](c ) that , the effect of noise and driving fluctuation is much suppressed in a quantum simulator with the same number of hybrid dressed spins .
we remark that the present scheme is also feasible for the study of quantum quench dynamics , which has attracted intensive interest recently , as it is related to the fundamental problem of thermalization in closed many - body quantum systems @xcite .
in particular , high - fidelity quantum simulation may help to distinguish local thermalization from environment induced equilibrium @xcite .
_ experiment implementation. _ the present proposal can be implemented in various physical systems , e.g. nv centers in diamond and trapped ions . for nv centers in diamond ,
the tunable coherent interactions between hybrid dressed states can be engineered from the natural magnetic dipole - dipole interaction using appropriate microwave drivings @xcite . the proof - of - principle verification of the ideas
can also be achieved with trapped ions , in which we can induce the interaction term @xmath112 in the dressed state basis by applying counter - propagating raman beams for generating the detuned red sideband transition , additionally we drive the carrier dressing transition for shielding the magnetic noise @xcite . in order to induce the transverse field for the simulated hamiltonian @xmath113
, we can introduce an additional rotated term which transforms the ising coupling to the xxz hamiltonian @xcite . for that purpose
, we use the single addressing ability of the focused co - propagating raman beams , to drive the dressed states off resonantly @xcite . the feasibility of this approach is carefully verified by our numerical simulation with realistic experiment parameters @xcite . _
conclusion. _ in summary , we propose to implement high - fidelity quantum simulation using hybrid dressed state .
its key advantage is the possibility for applying strong driving to suppress non - collective environment noise while at the same time the inevitable driving fluctuations are suppressed .
additionally , it provides new flexibility in the engineering of tunable coherent coupling .
we demonstrate theoretically the applicability of our scheme for noise resistant quantum simulation with nv centers in diamond , but would like to stress that it is sufficiently versatile to apply to a wide variety of physical systems .
our simulations show that a proof - of - principle verification is readily achievable with trapped ions under state - of - the - art experiment capability .
the method can be naturally combined with a wide range of applications such as quantum metrology / sensing and quantum computing .
our work thus contributes a new path towards the development of quantum technologies , in which the effect of decoherence is currently a severe obstacle . _
acknowledgements. _ j .- m.c is supported by the startup grant of huazhong university of science and technology and the national natural science foundation of china ( grant no.11574103 ) .
m. b. p. is supported by the alexander von humboldt foundation , the dfg ( spp1601 , and sfb tr21 ) , the eu integrating projects siqs and the eu streps equam and diadems , and the erc synergy grant bioq .
a. r. is supported by the israel science foundation ( grant no.039 - 8823 ) , the european commission ( strep equam ) , eu project diadems , the marie curie career integration grant ( cig ) ionquansense(321798 ) and the niedersachsen - israeli research cooperation program .
99 t. d. ladd , f. jelezko , r. laflamme , y. nakamura , c. monroe , j. l. obrien , http://www.nature.com/nature/journal/v464/n7285/full/nature08812.html[_nature_ * 464 * , 45 ( 2010 ) ] .
j. i. cirac , p. zoller , http://www.nature.com/nphys/journal/v8/n4/full/nphys2275.html[_nature physics _ * 8 * , 264 ( 2012 ) ] .
i. m. georgescu , s. ashhab , franco nori , http://journals.aps.org/rmp/abstract/10.1103/revmodphys.86.153[_rev .
mod .
phys . _
* 86 * , 153 ( 2014 ) ] .
t. h. johnson , s. r. clark , d. jaksch , http://www.epjquantumtechnology.com/content/1/1/10[_epj quantum technology . _
* 1 * , 10 ( 2014 ) ] .
m. greiner , o. mandel , t. esslinger , t. w. hnsch , and i. bloch , http://www.nature.com/nature/journal/v415/n6867/abs/415039a.html[_nature_ * 415 * , 39 ( 2002 ) ] .
d. leibfried , b. demarco , v. meyer , m. rowe , a. ben - kish , j. britton , w. m. itano , b. jelenkovic , c. langer , t. rosenband , and d. j. wineland , http://journals.aps.org/prl/abstract/10.1103/physrevlett.89.247901[_phys .
rev .
lett .
_ * 89 * , 247901 ( 2002 ) ] .
a. friedenauer , h. schmitz , j. t. glckert , d. porras , and t. schtz , http://www.nature.com/nphys/journal/v4/n10/abs/nphys1032.html[_nature physics _ * 4 * , 757 ( 2008 ) ] .
k. kim , m .- s .
chang , s. korenblit , r. islam , e. e. edwards , j. k. freericks , g .- d .
lin , l .- m .
duan , and c. monroe , http://www.nature.com/nature/journal/v465/n7298/full/nature09071.html[_nature_ * 465 * , 590 ( 2010 ) ] .
s.f .
huelga , c. macchiavello , t. pellizzari , a.k .
ekert , m.b .
plenio and j.i .
cirac , http://journals.aps.org/prl/abstract/10.1103/physrevlett.79.3865[_phys .
rev .
lett . _
* 79 * , 3865 ( 1997 ) ] .
v. giovannetti , s. lloyd , l. maccone , http://www.nature.com/nphoton/journal/v5/n4/full/nphoton.2011.35.html[_nature photonics _ * 5 * , 222 ( 2011 ) ] .
j. khler , j. a. j. m. disselhorst , m. c. j. m. donckers , e. j. j. groenen , j. schmidt , w. e. moerner , http://www.nature.com/nature/journal/v363/n6426/abs/363242a0.html[_nature_ * 363 * , 242 ( 1993 ) ] .
j. wrachtrup , c. von borczyskowski , j. bernard , m. orritt , r. brown , http://www.nature.com/nature/journal/v363/n6426/abs/363244a0.html[_nature_ * 363 * , 244 ( 1993 ) ] .
p. w. shor , http://journals.aps.org/pra/abstract/10.1103/physreva.52.r2493[_phys .
rev .
a _ * 52 * , r2493 ( 1995 ) ] .
a. steane , _ proc .
roy .
soc .
lond .
a _ * 452 * , 2551(1996 ) .
g. m. palma , k .- a .
suominen and a. k. ekert , _ proc .
roy .
soc .
london ser .
a _ * 452 * , 567 ( 1996 ) . m. b. plenio , v. vedral and p.l .
knight .
http://journals.aps.org/pra/abstract/10.1103/physreva.55.67[_phys .
rev .
a _ * 55 * , 67 ( 1997 ) ] .
l .-
m duan and g .- c .
guo , https://dx.doi.org/10.1103/physrevlett.79.1953[_phys .
rev .
lett . _ * 79 * 1953 ( 1997 ) ] .
p. zanardi and m. rasetti , http://journals.aps.org/prl/abstract/10.1103/physrevlett.79.3306[_phys .
rev .
lett .
_ * 79 * , 3306 ( 1997 ) ] .
d. a. lidar , i. l. chuang , k. b. whaley .
http://journals.aps.org/prl/abstract/10.1103/physrevlett.81.2594[_phys .
rev .
lett .
_ * 81 * , 2594 ( 1998 ) ] .
l .- a .
wu , m. s. byrd , and d. a. lidar , http://journals.aps.org/prl/abstract/10.1103/physrevlett.89.127901[_phys .
rev .
lett .
_ ] * 89 * , 127901 ( 2002 ) .
d. a. lidar , http://onlinelibrary.wiley.com/doi/10.1002/9781118742631.ch11/summary[_adv .
chem .
phys .
_ ] * 154 * , 295 ( 2014 ) .
e. l. hahn , http://journals.aps.org/pr/abstract/10.1103/physrev.80.580[_phys .
rev . _ * 80 * , 580 ( 1950 ) ] .
l. viola , s. lloyd , http://journals.aps.org/pra/abstract/10.1103/physreva.69.032314[_phys .
rev .
a _ * 58 * , 2733 ( 1998 ) ] . p. facchi , d. a. lidar , s. pascazio , http://journals.aps.org/pra/abstract/10.1103/physreva.69.032314[_phys .
rev .
a _ * 69 * , 032314 ( 2004 ) ] . f. f. fanchini , j. e. m. hornos , r. d. j. napolitano , http://journals.aps.org/pra/abstract/10.1103/physreva.75.022329[_phys .
rev .
a _ * 75 * , 022329 ( 2007 ) ] . k. khodjasteh , d. a. lidar , http://journals.aps.org/prl/abstract/10.1103/physrevlett.95.180501[_phys .
rev .
lett . _ * 95 * , 180501 ( 2005 ) ] .
g. s. uhrig , http://journals.aps.org/prl/abstract/10.1103/physrevlett.98.100504[_phys .
rev .
lett .
_ * 98 * , 100504 ( 2007 ) ] .
p. rabl , p. cappellaro , m. v. gurudev dutt , l. jiang , j. r. maze , and m. d. lukin , http://journals.aps.org/prb/abstract/10.1103/physrevb.79.041302[_phys .
rev .
b _ * 79 * , 041302 ( 2009 ) ] .
g. gordon , n. erez , g. kurizki , j. phys .
b : at . mol .
opt .
phys .
40 s75 ( 2007 ) .
n. timoney , i. baumgart , m. johanning , a. f. varon , m. b. plenio , a. retzker , ch .
wunderlich , http://www.nature.com/nature/journal/v476/n7359/full/nature10319.html[_nature_ * 476 * , 185 ( 2011 ) ] .
j .- m .
cai , b. naydenov , r. pfeiffer , l. p. mcguinness , k. d. jahnke , f. jelezko , m. b. plenio , and a. retzker , http://dx.doi.org/10.1088/1367-2630/14/11/113023[_new journal of physics _ * 14 * , 113023 ( 2012 ) ] .
m. w. doherty , n. b. manson , p. delaney , f. jelezko , j. wrachtrup , l. c.l .
hollenberg , http://www.sciencedirect.com/science/article/pii/s0370157313000562[_phys .
rep .
_ ] * 528 * , 1 ( 2013 )
. see online supplemental material for the details of calculations .
daniel t. gillespie , http://journals.aps.org/pre/abstract/10.1103/physreve.54.2084[_phys .
rev .
e _ * 54 * , 2084 ( 1996 ) ] .
f. dolde , i. jakobi , b. naydenov , n. zhao , s. pezzagna , c. trautmann , j. meijer , p. neumann , f. jelezko , j. wrachtrup , http://www.nature.com/nphys/journal/v9/n3/full/nphys2545.html[_nature physics _ * 9 * , 139 ( 2013 ) . ]
s. kotler , n. akerman , n navon , y. glickman , r. ozeri , http://www.nature.com/nature/journal/v510/n7505/full/nature13403.html[_nature_ * 510 * , 376 ( 2014 ) . ] a. albrecht , g. koplovitz , a. retzker , f. jelezko , s. yochelis , d. porath , y. nevo , o. shoseyov , y. paltiel , m. b. plenio , http://iopscience.iop.org/1367-2630/16/9/093002/[_new j. phys . _
* 16 * , 093002 ( 2014 ) . ]
j. w. britton , brian c. sawyer , adam c. keith , c .- c .
j. wang , j. k. freericks , h. uys , m. j. biercuk , j. j. bollinger , http://www.nature.com/nature/journal/v484/n7395/full/nature10981.html[_nature_ * 484 * , 489 ( 2012 ) ] .
a. bermudez , j. almeida , f. schmidt - kaler , a. retzker and m. b. plenio , http://journals.aps.org/prl/abstract/10.1103/physrevlett.107.207209[_phys .
rev .
lett . _
* 107 * , 207209 ( 2011 ) ] .
a. bermudez , j. almeida , k. ott , h. kaufmann , s. ulm , f. schmidt - kaler , a. retzker and m.b .
plenio , http://iopscience.iop.org/1367-2630/14/9/093042/[_new j. phys . _
* 14 * , 093042 ( 2012 ) ] .
r. islam , c. senko , w. c. campbell , s. korenblit , j. smith , a. lee , e. e. edwards , c. c. j. wang , j. k. freericks , c. monroe , http://www.sciencemag.org/content/340/6132/583.abstract[_science_ * 340 * , 583 ( 2013 ) ] .
t. monz , k. kim , a. s. villar , p. schindler , m. chwalla , m. riebe , c. f. roos , h. hffner , w. hnsel , m. hennrich , r. blatt , http://journals.aps.org/prl/abstract/10.1103/physrevlett.103.200503[_phys .
rev .
lett .
_ ] * 103 * , 200503 ( 2009 ) .
p. calabrese , j. cardy , http://iopscience.iop.org/1742-5468/2007/06/p06008/[_j .
stat .
mech . _ * 0706 * , p06008 ( 2007 ) ] .
m. rigol , v. dunjko , m. olshanii , http://www.nature.com/nature/journal/v452/n7189/abs/nature06838.html[_nature_ * 452 * , 854 ( 2008 ) ] .
a. riera , c. gogolin , j. eisert , http://journals.aps.org/prl/abstract/10.1103/physrevlett.108.080402[_phys .
rev .
lett . _ * 108 * , 080402 ( 2012 ) ] .
s. trotzky , y .- a .
chen , a. flesch , i. p. mcculloch , u. schollwck , j. eisert , i. bloch , http://www.nature.com/nphys/journal/v8/n4/full/nphys2232.html[_nature physics _ * 8 * , 325 ( 2012 ) ] .
d. porras and j. i. cirac , http://journals.aps.org/prl/abstract/10.1103/physrevlett.92.207901[_phys .
rev .
lett . _ * 92 * , 207901 ( 2004 ) ] .
s .- l .
zhu , c. monroe , and l .- m .
duan , http://journals.aps.org/prl/abstract/10.1103/physrevlett.97.050505[_phys .
rev .
lett . _ * 97 * , 050505 ( 2006 ) ] .
a. bermudez , p. o. schmidt , m. b. plenio , and a. retzker , http://journals.aps.org/pra/abstract/10.1103/physreva.85.040302[_phys .
rev .
a _ ] * 85 * , 040302(r ) ( 2012 ) .
i. cohen and a. retzker http://journals.aps.org/prl/abstract/10.1103/physrevlett.112.040503[phys .
rev .
lett .
* 112 , * 040503 ( 2014 ) . ]
i. cohen , p. richerme , z .- x .
gong , c. monroe , a. retzker , http://arxiv.org/abs/1505.04695[arxiv:1505.04695 . ] |
on february 28 , 1997 , the first x - ray afterglow of a gamma - ray burst ( grb ) was detected , leading to the identification of its progenitor at cosmological distances @xcite . in a few days ,
the afterglow faded away with time as a power law .
this behavior is satisfactorily explained in the spherical ( isotropic ) fireball model involving relativistic ejecta decelerated by circumburst medium @xcite .
the introduction of collimated jets relaxes the energy requirement of grbs by a factor of several hundred , as well as explains the steeper temporal decay of afterglows @xcite .
grbs are extra - galactic sources of gev and probably higher energy photons .
evidences of distinct high - energy component have been accumulated by egret onboard the compton gamma - ray observatory : ( 1 ) @xcite reported the detection of long - duration mev
gev emission of grb 940217 lasting up to 1.5 hour after the kev burst , including an @xmath218 gev photon .
this burst is the longest and the most energetic among those grbs with detected high - energy emission so far ; ( 2 ) @xcite revealed a high - energy component of grb 941017 temporally and spectrally different from the low - energy component . in the fireball model ,
synchrotron emission of shock - accelerated electrons is commonly thought to produce prompt @xmath3-ray emission as well as afterglow emission at lower energies ( e.g. , * ? ? ?
it is natural to expect that these photons are inverse - compton up - scattered by electrons , giving rise to a higher energy component peaking at sub - gev to tev energies @xcite .
when electrons scatter the self - emitting synchrotron photons , synchrotron self - compton ( ssc ) emission is resulted . in the external shock scenario ,
the temporal profile of the ssc emission from forward shock electrons is similar to that of the low energy afterglow emission and no significant time lag is expected .
the _ swift _ satellite , thanks to its rapid response time and accurate localization , has started a new era of research on grbs .
different modifications to the standard afterglow model are put forward to explain the peculiar behaviors exhibited in the x - ray light curves , in particular the shallow declining phase @xcite .
recently , the ssc emission of the modified forward shock has been extensively discussed in the literature @xcite and applied to the case of grb 940217 @xcite .
the agile ( astro - rivelatore gamma a immagini leggero ) satellite , launched on april 23 , 2007 , is dedicated to high - energy @xmath3-ray astronomy .
the fermi gamma - ray space telescope ( fgst ) was launched on june 11 , 2008 .
the large area telescope ( lat ) onboard covers the energy range from 20 mev to 300 gev and its effective area is about 5 times larger than that of egret at gev energies .
the first grb observations with lat have resulted in detection of photons with energies larger than @xmath4 gev from several grbs @xcite .
@xcite , @xcite , and @xcite predicted promising and detectable ssc emission from the forward shock with fgst out to z@xmath21 .
most of the discussions in the literature have focused on the afterglow emission from tens of mev to gev .
lat can also detect very - high - energy ( vhe ; @xmath5100 gev ) afterglow emission .
however , with a small effective area @xmath6 , it is very hard to have a significant detection at such high energy .
imaging atmospheric cherenkov telescopes such as h.e.s.s . ,
magic , and veritas may serve better at energies above @xmath2100 gev because of their much larger effective area ( @xmath7 ) and a high rejection rate of hadronic background .
the effective collecting area of cherenkov telescopes increases with energy @xcite .
some of these large area cherenkov detectors have been used to set constraints on the possible vhe afterglow component of grbs @xcite .
it is thus desirable to see whether these results are consistent with the predictions of the fireball model .
our aim of this paper is also to investigate the prospect of significant detections in the future . to have a reliable estimate of the afterglow emission at energies above 100 gev
, one needs to calculate the forward shock emission ( both synchrotron and ssc emission of the shocked electrons ) carefully .
the attenuation of vhe photons by the cosmic infrared background is also taken into account.since the attenuation effect for photons with an energy @xmath5100 gev is more severe for high - redshift grbs , we limit our grb sample to nearby events .
this paper is organized as follows : in [ sect_afterglow_modeling ] , we describe the grb afterglow model , introduce the code that is used in the afterglow modeling and the calculation of the ssc emission from grb forward shock . in [ sect_model_prediction ] , we present the expected results of the ssc model using reasonable parameter values for grbs . in [ sect_vheafterglow_fromnearbygrbs ] , we describe the grb sample which includes six nearby grbs with sufficient multi - wavelength afterglow data and predict their corrected energy flux after the attenuation by the cosmic background during the afterglow phase , which is then compared with the available observational data .
we summarize our results and discuss their implications in
[ sect_discussion ] .
we conclude in [ sect_conclusions ] .
while synchrotron emission is widely considered to be responsible for the radio , optical , and x - ray afterglows ( e.g. * ? ? ?
* ) , inverse compton scattering of forward shock photons is considered in details by @xcite and @xcite .
inverse compton scattering may considerably change the temporal and spectral behavior of grb afterglows , and its cooling effect on electrons accelerated in external shocks will contribute to the photon spectra at sub - gev to tev energy range @xcite . in the afterglow model , both synchrotron emission and inverse compton emission
are taken into account .
it is assumed that : ( 1 ) the external medium is homogenous with a density @xmath8 or a wind profile @xmath9 ; ( 2 ) the relativistic jet is uniform , i.e. energy per solid angle is independent of direction within the jet ; ( 3 ) the shock parameters ( @xmath10 and @xmath11 , fractions of the shock energy given to the electrons and the magnetic field , respectively ) are constant ; ( 4 ) the energy distribution of electrons accelerated in shocks follows @xmath12 ; ( 5 ) the possible achromatic flattening in the afterglow light curve is due to energy injection in the form @xmath13 @xcite or @xmath14^{-1}$ ] with @xmath15 being the initial spin - down time scale @xcite .
the parameters involved in this afterglow model include : @xmath16 ( the initial isotropic outflow energy ) , @xmath17 ( the initial half - angle of the jet ) , @xmath8 ( the density of the homogeneous external medium ) or @xmath18 ( the wind parameter ) , @xmath19 ( the power - law index of energy distribution of shock - accelerated electrons ) , @xmath20 , and @xmath21 ( shock parameters ) . in the case where energy injection is necessary , three additional parameters : @xmath22 ( the injected luminosity in the rest frame ) , the timescale of energy injection and @xmath23 , are included .
the code used in our afterglow modeling and the prediction of the ssc emission is that developed by @xcite , who carry out numerical calculations of synchrotron and ssc emission of the external forward shock in the afterglow phase .
the reverse shock emission , predicted in the fireball model but not detected in most events @xcite , is not taken into account .
the key treatments ( see 3 of * ? ? ?
* for details ) are as follows : ( i ) the dynamical evolution of the outflow is followed using the formulae in @xcite , which describes the hydrodynamics in both relativistic and non - relativistic phases .
( ii ) the arbitrary assumption that the distribution of shocked electrons is always in a quasi - stationary state is considered to be unsatisfactory and the energy distribution of electrons is calculated by solving the continuity equation with the power - law source function q = k@xmath24 , normalized by a local injection rate @xcite .
( iii ) the observed flux is integrated over the equal - arrival surface " .
( iv ) the klein - nishina correction of the inverse compton emission has been included .
( v ) energy injection into the outflow is considered necessary in reproducing some multi - frequency afterglow data of some grbs .
this may change the dynamics significantly . by fitting the low - energy multi - waveband afterglows , from radio to x - ray band , parameters involved in the afterglow model
are gotten .
_ simultaneous afterglow data in at least two well - separated wavebands are needed to get a relatively well - constrained set of parameters . _
a rough estimate of the energy - integrated vhe afterglow flux ( without correction by the cosmic background ) is given by @xmath25 where @xmath26 is the total luminosity of the ssc emission , @xmath27 and @xmath28 are the typical ssc emission frequency and the ssc cooling frequency of the forward shock electrons ( see eq.(33 - 34 ) in * ? ? ?
* the case of @xmath29 , for the expressions ) and @xmath30 is the luminosity distance of the event .
high energy photons , especially those in the tev range , will be attenuated by the cosmic background light .
various models of the spectral energy distribution of the cosmic infrared background are proposed @xcite , but all these models give comparable opacities for low redshifts . in this work , a level consistent with a study of two distant blazars and galaxy counts is used ( p0.45 , * ? ? ?
we adopt reasonable values of parameters for nearby grbs and predict the spectra in high - energy to vhe range .
after corrected for the attenuation by extragalactic background , we compare them with the sensitivity levels of @xmath3-ray instruments .
parameters assumed and the time - averaged spectra , including both synchrotron and ssc components from the forward shocks , are shown in figure [ fig : spectrum ] .
for this fictitious burst , current imaging atmospheric cherenkov telescopes would be more likely than space satellites such as egret , agile / grid and _ fgst_/lat to detect the modeled emission , as seen in figure [ fig : spectrum ] .
= @xmath31@xmath32 , @xmath33 , @xmath34 , @xmath35 , @xmath36 , @xmath37 and @xmath38 .
the solid and dashed line are calculated with the same parameters represented above during the first epoch but occurring at larger redshift , @xmath39 and @xmath40 , respectively .
the sensitivity curves of egret @xcite , agile / grid @xcite , fgst / lat @xcite , veritas @xcite and hess ( assuming @xmath41 , * ? ? ?
* ) for an integration time of 0.5 hour are plotted as labeled .
magic 2-sigma upper limits derived from 30-min observations of grb 060206 are also plotted , taken from albert et al .
( 2007 ) . ]
for photons with energy higher than @xmath42 gev , the attenuation due to interaction with background photons is significant if the source has a high redshift .
therefore nearby bursts ( those with [email protected] ) are chosen in this study .
to predict the vhe afterglow emission of nearby grbs and compare the calculated results with sensitivity of different @xmath3-ray telescopes , grbs in our sample must satisfy : nearby events to alleviate the attenuation effect ; at least two independent waveband low - energy afterglows are recorded to get relatively constrained parameters ; the low - energy afterglow can be reproduced according to the afterglow model . in this work ,
we consider nearby grbs ( @xmath44 ) with relatively high luminosity and multi - wavelength afterglow data sufficient to meaningfully constrain the properties of the grbs ( i.e. the model parameter values as described in [ subsect_grb_afterglow_model ] ) up to march 2007 .
five grbs meet such criteria : grb 030329 , grb 050509b , grb 050709 , grb 060505 , and grb 060614 .
though having a relatively large redshift of [email protected] , grb 051221a is also considered in this work because it is one of the brightest short grbs detected so far .
grb 030329 triggered the high energy transient explorer , hete-2 @xcite .
very detailed bvri afterglow light curves , spanning from @xmath20.05 to @xmath2 80 days , were compiled by @xcite .
@xcite reported xmm - newton and rossi - xte late - time observations of this burst .
based on the emission and absorption lines in the optical afterglow , a redshift of z=0.1685 has been identified @xcite .
the x - ray telescope ( xrt ) onboard _ swift _ began observations of grb 050509b 62s after the trigger of the burst alert telescope ( bat ) @xcite .
optical and infrared data were reported in @xcite . @xcite and @xcite reported a redshift of [email protected] based on numerous absorption features and a putative host galaxy , respectively .
grb 050709 was discovered by hete-2
its prompt emission lasted 70 ms in the 3 - 400 kev energy band , followed by a weaker , soft bump of @xmath2100 s duration .
the optical counterpart of this burst was observed with the danish 1.5-m telescope at the la silla observatory .
the observations started 33 hours after the burst and spanned over the following 18 days @xcite .
observations with the chandra x - ray observatory revealed a faint , uncatalogued x - ray source inside the hete-2 error circle @xcite , which was coincident with a pointlike object embedded in a bright galaxy @xcite at z = 0.16 @xcite .
grb 051221a was localized by bat @xcite and also simultaneously observed by the konus - wind instrument .
the x - ray ( @xmath45s ; * ? ? ? * ) and the optical ( @xmath46s ; * ? ? ?
* ) afterglow light curves of grb 051221a were well detected , while in the radio band only one detection followed by several upper limits are available @xcite .
@xcite detected several bright emission lines , indicating a redshift of @xmath47 .
grb 060505 was detected by bat in the 15 - 150 kev band @xcite .
@xcite reported the detection of the optical transient , later confirmed by vlt fors2 observations @xcite .
xrt detected a source which was located about 4@xmath48 from a galaxy with z=0.0894 @xcite .
grb 060614 triggered both _
swift_-bat @xcite and konus - wind @xcite .
xrt found a very bright ( @xmath21300 counts s@xmath49 ) un - catalogued source inside the bat error circle .
ground - based optical and infrared follow - up observations were performed using several instruments ( e.g. , * ? ? ?
* ; * ? ? ?
based on the detection of the host galaxy emission lines , a redshift of z = 0.125 was proposed by @xcite and confirmed by @xcite . noted that the classification of grb 060614 is ambiguous in the commonly - used long / short burst scheme , since it has a long duration but no accompanying sn @xcite .
the available multi - frequency afterglow data are then used to obtain the model parameters . in this work ,
we have reproduced the multi - frequency afterglow data of grb 030329 and grb 060614 .
the well - sampled distinguishing afterglow behavior of grb 030329 has gained much attention .
some authors concentrated on the rebrightening occurring at 1.6 days after the trigger and considered different mechanisms to explain the rebrightening features seen in the optical light curves @xcite .
we concentrate on the multi - waveband emission , from radio @xcite , optical @xcite to x - ray band @xcite for the purpose in this paper .
we show in figure [ fig : fits ] that , with a certain set of parameters , the numerical results can describe the observed data in all three wavebands .
fluctuations were captured in @xmath50 band afterglow light curves after @xmath51s from the burst trigger , which may imply the multiple energy injection into the outflow @xcite or a two component jet @xcite .
we ignore these details and focus on the general trend of the optical emission ( particularly , in our calculation the energy of the relativistic ejecta is a constant ) .
as shown in figure [ fig : fits ] , the difference is that in the time range @xmath52 sec our approach gives a ( little bit ) brighter optical emission , so will be the high energy emission .
the modeled and observed afterglow light curves of grb 060614 are shown in figure [ fig : fit1 ] .
energy injection , starting around 30 minutes after the grb onset , is needed in the afterglow modeling to reproduce the increase in flux ( instead of simple power - law decay seen in other grbs ) .
the early x - ray flux before 500s from the burst trigger , which is much brighter than the modeled flux , results from the dominating contribution from the prompt emission .
table [ sample ] lists the physical parameters derived from the afterglow modeling for grb 030329 and grb 060614 .
parameters of grb 050509b , grb 050709 , grb 051221a and grb 060505 are taken from the literature , also listed in table 1 .
we are interested in vhe observations during the afterglow phase when the ssc is likely to dominate ( see [ sect_discussion ] ) .
vhe @xmath3-ray afterglow data of three of the grbs in the sample ( i.e. grb 030329 , grb 050509b , and grb 060505 ) are available .
@xcite reported a total of 4 hours of observations , which spanned five nights , using the whipple 10-m telescope .
no evidence for vhe @xmath3-ray signal was found during any of the observation periods .
when combining all data , a flux upper limit of @xmath53 was derived .
the first observation , lasting for about an hour , was started 64.6 hours after the burst .
the 99.7% c.l .
flux upper limit above an energy of @xmath54 gev derived from this observation is shown in table [ flux ] , as well as in figure [ fig : highenergy ] .
the 28-minute h.e.s.s .
observation of grb 030329 were taken 11.5 days after the burst @xcite .
since the burst position was located above the northern hemisphere , the zenith angle of the grb observation was relatively large , i.e. 60@xmath55 , resulting in an energy threshold of 1.36 tev .
no evidence for vhe @xmath3-ray signal was found . the 99% c.l .
flux upper limit ( @xmath56 tev ) is @xmath57 , assuming a photon index of @xmath58 .
the observations of this burst using the stacee detector employ an on - off observation mode and contain two 28-minute on / off pairs .
the first on - source observation started 20 minutes after the burst and the second 80 minutes after the burst .
after data quality cuts , about 18 minutes of useful on - source data remain in each observation .
no evidence for vhe @xmath3-ray signal above the energy threshold of 150 gev was reported by @xcite .
the 95% c.l flux upper limits ( above 150 gev , assuming a photon spectrum of @xmath59 ) were @xmath60 and @xmath61 for the first and second on - source observation , respectively ( a. jarvis private communication ) . the h.e.s.s .
observations began 19.4 hours after the burst and lasted for 2 hours @xcite .
no evidence for vhe @xmath3-ray signal was found .
the 99% c.l .
flux upper limit ( @xmath62 tev ) is @xmath63 , assuming a photon index of @xmath58 .
based on the parameters obtained in 4.2 , the gev - tev emission is obtained using the code described in 2.2 .
we depict the calculated high energy afterglow spectrum in figure [ fig : highenergy ] , which shows the time - integrated high energy afterglow spectrum of these six events .
the solid and dashed lines represent the intrinsic ssc spectra and corrected spectra for each grb , respectively .
the absorption is based on the cosmic infrared background model `` p0.45 '' @xcite .
such model is constrained by the upper limits provided by two unexpectedly hard spectra of blazars at optical / nir wavelengths and is close to the lower limit from integrated light of resolved galaxies . in order to compare with the vhe observational data which are usually given in integrated photon fluxes ,
we integrate the spectra above the energy threshold .
we consider first the grbs with vhe data .
these include grb 030329 , grb 050509b , and grb 060505 .
in table 2 we list the modeled integrated energy fluxes after correction due to interaction with photons from cosmic infrared background , as well as the vhe @xmath3-ray observations and the derived upper limits .
all predicted fluxes are below the upper limits derived from the vhe observations .
we then investigate whether a sensitive vhe instrument is expected to detect the predicted vhe signal from nearby grbs during the late afterglow phase .
we use h.e.s.s .
sensitivity as an example of an array of sensitive atmospheric cherenkov telescopes .
the sensitivity level of h.e.s.s .
detector is shown in figure [ fig : sensitivity ] , assuming a @xmath64=2.6 spectrum @xcite .
assuming softer spectra , the level is higher , and the difference is about 50% between @xmath64=2.0 and @xmath64=3.0 ( c.f .
we show the temporal evolution of energy fluxes ( @xmath5200gev ) of six grbs in our sample in figure [ fig : sensitivity ] , indicating only the vhe signal from grb 030329 may be above the h.e.s.s . sensitivity . for grb 030329 which is a bright burst with low redshift
, the expected energy flux would be high enough to be detected with a delayed observation time of @xmath65 hours if grb position was favorable , i.e. with zenith angle @xmath66 ( and thus an energy threshold of @xmath67 gev is attained ) .
in this paper , we have calculated the ssc emission from the forward shock electrons following @xcite .
we shall discuss here the importance of other radiation processes in the late afterglow phase .
possible vhe @xmath3-ray emission initiated from protons has been suggested @xcite . however , the proton - synchrotron component , as well as the hadron - related photo - meson electromagnetic components , is in most cases overshadowed by the ssc component of electrons in the afterglow phase .
this is especially the case for the parameter values of @xmath10 and @xmath11 used here in the modeling of these six grbs ( e.g. , * ? ? ?
additional ic component will play a role and will enhance any vhe emission .
ssc is , in general , only a lower limit .
another possible contribution to vhe emission is related to the central engine afterglow " emission , like the flares ( e.g. , * ? ? ? * ) or the plateaus followed by a sudden drop ( e.g. , * ? ? ?
the ssc radiation of the late internal shocks or the external inverse compton radiation in the external forward shock front may be able to give rise to some vhe emission signals ( see fan & piran 2008 for a review ) .
however , no central engine afterglow " emission has been reported in the late afterglow phase for the six bursts we studied .
so it is hard for us to estimate its ability to enhance the detectability of the vhe emission .
as shown in figure [ fig : sensitivity ] , we only expect detectable signal using a sensitive ground - based @xmath3-ray detector for a bright and nearby grb like grb 030329 .
the rate of such nearby and energetic grbs is very uncertain .
grb 940217 might have been another event of this kind @xcite .
several factors that reduce the chance of detecting vhe photons can be summarized as follows : firstly , as a result of large zenith angles ( e.g. , 60@xmath55 for grb 030329 ) , the energy thresholds of some observations are relatively high ( @xmath21.4 tev ) .
any vhe photon is severely attenuated by the cosmic infrared background , unless the level is very low .
secondly , the observations were taken at very late epochs , e.g. 11.5 days after the burst for h.e.s.s .
observations of grb 030329 , when expected vhe flux had largely decayed .
thirdly , the fraction of low - redshift grbs is small , e.g. see the catalogue in @xcite . for grb 051221a
( at @xmath68 ) studied here , the attenuation is severe at energies @xmath69200 gev .
despite these practical limitations , detection of vhe afterglow emission of grbs is probable .
those grbs close enough ( [email protected] ) and with an intrinsic high luminosity ( like grb 030329 ) , can be detected above @xmath2200 gev when the observation is taken within @xmath210 hours after the burst . from eq .
1 , grbs with low @xmath70 , large @xmath71 , large @xmath10 , and small @xmath11 are more likely to be detected in the vhe band .
besides , in general the expected vhe afterglow flux decays as @xmath72 where @xmath73 is at least one .
thus , observations with shorter delay are more likely to probe the predicted vhe emission . _
the main purpose of this work is to provide a relatively reliable prediction of the detectability of the vhe emission from nearby grbs_. it is also interesting to explore the role of the vhe afterglow emission detection in revealing the grb physics .
perhaps the most robust conclusion is that the vhe emission in late afterglow phase can not be attributed to the synchrotron radiation of the forward shock , for which the maximal photon energy is @xmath74 @xcite , where @xmath64 is the bulk lorentz factor of the decelerating outflow . in general , for @xmath75 , the inverse compton emission is expected to be very weak .
so the detection of vhe afterglow emission from grbs requires that @xmath76 , in support of the current radio / optical / x - ray afterglow modeling ( e.g. , see tab.1 ) . together with _
, agile , fgst and some ground - based optical telescopes , ground - based @xmath3-ray detectors can provide us continuous spectra in the optical to tev energy band during the afterglow phase .
a self - consistent modeling of these data in a very wide energy band , in principle , will impose very tight constraint on the physical parameters and on the environment .
however , given the small number of the vhe photons expected from a single burst , we do not expect that the vhe emission can help us a lot to achieve such goals .
in this work , we discuss the prospect of detecting vhe @xmath3-rays with current ground - based detectors in the late afterglow phase . during this phase ,
the dominant radiation process in the vhe @xmath3-ray regime is the ssc emission from the forward shock electrons .
klein - nishina effects and attenuation by the cosmic infrared background , both known to suppress the vhe @xmath3-ray spectra , were taken into account . to minimize the attenuation effect , we chose a sample of six nearby grbs in this study .
we have calculated the detailed ssc emission numerically using the model developed by @xcite , with a set of parameters which are able to reproduce the available multi - wavelength afterglow light curves .
the results are consistent with the upper limits obtained using vhe observations of grb 030329 , grb 050509b , and grb 060505 . moreover , assuming observations taken 10 hours after the burst , the vhe signal predicted from five grbs is below the sensitivity level of a current sensitive atmospheric cherenkov detectors mainly due to the low fluence of these outflows . for those bright and nearby bursts like grb 030329 , a vhe detection is possible even with a delayed observation time of @xmath210 hours .
we thank the anonymous referee for helpful comments and d. xu for providing us data in figure 3 .
this work is supported by the national science foundation ( grants 10673034 and 10621303 ) and national basic research program ( 973 programs 2007cb815404 and 2009cb824800 ) of china .
yzf is also supported by a grant from danish national research foundation .
pht acknowledges support from imprs - hd .
aharonian , f. , et al .
( h.e.s.s .
collaboration ) 2005 , , 437 , 135 aharonian , f. , et al .
( h.e.s.s .
collaboration ) 2006a , nature , 440 , 1018 aharonian , f. , et al .
( h.e.s.s .
collaboration ) 2006b , , 457 , 899 aharonian , f. , et al .
( h.e.s.s .
collaboration ) 2009 , , 495 , 505 albert , j. , et al .
( magic collaboration ) 2007 , , 667 , 358 berger , e. , et al .
2003 , nature , 426 , 154 , j. , blake , c. , prochaska , j. x. , hennawi , j. , gladders , m. , & koester , b. 2005 , gcn circ .
3386 bloom , j. s. , et al .
2006 , , 638 , 354 bttcher , m. & dermer , c. d. 1998 , , 499 , l131 burrows , d. n. , et al .
2006 , , 653 , 468 butler , n. r. , kocevski , d. , bloom , j. s. , & curtis , j. l. 2007 , , 671 , 656 cheng , k. s. , & wei , d. m. , 1996 , mnras , 283 , l133 cobb , b. e. , bailyn , c. d. , van dokkum , p. g. , & natarajan , p. 2006
, , 651 , l85 cohen , e. , & piran , t. 1999 , , 518 , 346 , m. l. , capalbi , m. , vetere , l. , palmer , d. , & burrows , d. 2006 , gcn circ .
5078 costa , e. , et al .
1997 , nature , 387 , 783 covino , s. , et al .
2006 , , 447 , l5 cummings , j. , et al .
2005 , gcn circ .
4365 dai z. g. , & lu t. , 1998 , , 333 , l87 dermer , c. d. , chiang , j. , & mitman , k. e. 2000 , 537 , 785 falcone , a. d. , et al .
2007 , , 671 , 1921 fan , y. z. , & piran , t. , 2008 , front .
phys . chin .
, 3 , 306 fan , y. z. , piran , t. , narayan , r. , & wei , d. m. , 2008 , , 384 , 1483 fan , y. z. , & xu , d. 2006 , , 372 , l19 , d. b. , frail , d. a. , cameron , p. b. , & cenko , s. b. 2005 , gcn circ . 3585
fugazza , d. , et al .
2006 , gcn circ .
5276 fynbo , j. p. u. , et al . , 2006 , nature , 444 , 1047 galli , a. , & piro . , l. 2007 , , 475 , 421 galli , a. , & guetta , d. 2008 , , 480 , 5 gehrels , n. , et al .
2005 , nature , 437 , 851 gehrels , n. , et al .
2006 , nature , 444 , 1044 , s. , aptekar , r. , mazets , e. , palshin , v. , frederiks , d. , & cline , t. 2006 , gcn circ . 5264 , m. m. , dingus , b. l. , kaneko , y. , preece , r. d. , dermer , c. d. , & briggs , m. s. 2003 , , 424 , 749 gou , l. j. , & mszros , p. 2007
, , 668 , 392 greiner , j. , et al .
2003 , gcn , 2020 hjorth , j. , et al .
2005 , nature , 437 , 859 horan , d. , et al .
2007 , , 655 , 396 huang , y. f. , gou , l. j. , dai , z. g. & lu , t. 2000 , , 543 , 90 huang , y. f. , cheng , k. s. , gao , t. t. 2006 , , 637 , 873 hullinger , d. , et al .
2006 , gcn circ .
5142 hurley , k. , et al .
1994 , nature , 372 , 652 jarvis , a. , et al .
2008 , proceedings of the 30th international cosmic ray conference ; rogelio caballero , juan carlos dolivo , gustavo medina - tanco , lukas nellen , federico a. snchez , jos f. valds - galicia ( eds . ) ; universidad nacional autnoma de mxico , mexico city , mexico , 3 , 1111 jensen , b. l. , et al .
2005 , gcn circ .
3589 kneiske , t. m. , mannheim , k. , & hartmann , d. h. 2002 , , 386 , 1 lipkin , y. m. , et al .
2004 , , 606 , 381 mszros , p. , & rees , m. j. 1994 , , 269 , l41 mszros , p. , & rees , m. j. 1997 , , 476 , 232 moderski , r. , sikora , m. , & bulik , t. , 2000 , , 529 , 151 nousek , j. a. , et al .
2006 , 642 , 389 , e. o. , cenko , s. b. , gal - yam , a. , peterson , b. , schmidt , b. p. , fox , d. b. , & price , p. a. 2006 , gcn circ .
5123 omodei , n. 2008 , gcn circ .
8407 , d. , cummings , j. , stamatikos , m. , markwardt , c. , & sakamoto , t. 2006 , gcn circ .
5076 panaitescu , a. 2006 , , 367 , l42 parsons , a. m. , et al .
2006 , gcn circ .
5252 price , p. a. , roth , k. , & fox , d. w. 2005 , gcn circ .
3605 price , p. a. , berger , e. , & fox , d. b. , 2006 , gcn circ .
5275 primack , j. r. , somerville , r. s. , bullock , j. s. , & devriendt , j. e. g. 2001 , apic proc . ,
558 , 463 , j. x. , bloom , j. s. , chen , h .- w . , & hurley , k. 2005 , gcn circ . 3399 rhoads , j. e. , 1999 , , 525 , 737 roming p. w. a. , et al . , 2006 , apj , 652 , 1416 rykoff e. s. , et al . , 2009 , apj submitted ( arxiv:0904.0261 ) sari , r. , piran , t. , & narayan , r. , 1998 , , 497 , l17 sari , r. , piran , t. , & halpern , j. p. , 1999
, , 519 , l17 sari , r. , & esin , a. a. 2001 , , 548 , 787 schmidt , b. , peterson , b. , & lewis , k. 2006 , gcn circ .
5258 soderberg , a. m. , et al .
2006 , , 650 , 261 stecker , f. w. , malkan , m. a. , & scully , s. t. 2006 , , 648 , 774 tajima , h. , et al .
2008 , gcn circ .
8246 tam , p. h. , chadwick , p. , gallant , y. , horns , d. , phlhofer , g. , rowell , g. , & wagner , s. 2008 , proceedings of the 30th international cosmic ray conference ; rogelio caballero , juan carlos dolivo , gustavo medina - tanco , lukas nellen , federico a. snchez , jos f. valds - galicia ( eds . ) ; universidad nacional autnoma de mxico , mexico city , mexico , 3 , 1119 tam , p. h. 2009 , ph.d .
thesis , ruprecht - karls - universitt heidelberg tavani , m. , barbiellini , g. , argan , a. , et al .
2008 , [ arxiv:0807.4254 ] thoene , c. c. , fynbo , j. p. u. , sollerman , j. , jensen , b. l. , hjorth , j. , jakobsson , p. , & klose , s. 2006 , gcn circ .
5161 thompson , d. j. , et al .
1993 , apjs , 86 , 629 tiengo , a. , mereghetti , s. , ghisellini , g. , tavecchio , f. , & ghirlanda , g. 2004 , , 423 , 861 totani , t. , 1998 , , 502 , l13 totani , t. , & takeuchi , t. t. 2002 , , 570 , 470 vanderspek , r. , et al .
2004 , , 617 , 1251 villasenor , j. s , et al .
2005 , nature , 437 , 855 wang , x. y. , dai , z. g. , & lu , t. 2001 , , 556 , 1010 wei , d.m . , & lu , t. , 1998 , , 505 , 252 wei , d.m . , & lu , t. , 2000 , , 360 , l13 wei , d.m . , & fan , y. z. , 2007 , chjaa , 7 , 509 xu , d. , starling , r. l. c. , fynbo , j. p. u. , et al .
2009 , apj , 696 , 971 yu , y. w. , liu , x. w. , & dai , z. g. 2007 , , 671 , 637 zhang , b. , & mszros , p. 2001a
, , 552 , l35 zhang , b. , & mszros , p. 2001b
, , 559 , 110 zhang , b. , fan , y. z , dyks , j. , kobayashi , s. , mszros , p. , burrows , d. n. , nousek , j. a. , & gehrels , n. 2006 , , 642 , 354 zhang , x. h. , 2009 , research in astronomy and astrophysics , 9 , 213 lccccccccccc [ sample ] 030329 & 0.1685 & 1.4@xmath77 & 0.31 & 100 & 2.01 & 0.1 & 0.001 & & & & this work + 050509b & 0.2248 & 2.75@xmath78 & 0.5 & 1 & 2.2 & 0.15 & 0.046 & & & & 1 + 050709 & 0.16 & 3.77@xmath79 & 0.5 & 6@xmath80 & 2.6 & 0.4 & 0.25 & & & & 2 + 051221a & 0.5465 & @xmath81 & 0.1 & 0.01 & 2.4 & 0.3 & 2@xmath82 & 2@xmath78 & magnetar wind & @xmath83 & 3 + 060505 & 0.089 & 2.6@xmath79 & 0.4 & 1 & 2.1 & 0.1 & 0.008 & & & & 4 + 060614 & 0.125 & 5@xmath79 & 0.08 & 0.05 & 2.5 & 0.12 & 2@xmath82 & @xmath84 & 0 & @xmath85 & this work + lccccccc 030329 & h.e.s.s . & 11.5 days & 28 min & 1360 & @xmath86 & @xmath87 & 1 + 030329 & whipple & 64.55 hours & 65.2 min & 400 & @xmath88 & @xmath89 & 2 + 050509b & stacee & 20 min/80 min & 28 min & 150 & 3.8@xmath90 & 2.2@xmath91/5.4@xmath92 & 3 + 060505 & h.e.s.s . & 19.4 hours & 2 hours & 450 & 8.8@xmath93 & 2.5@xmath94 & 1 + [ tab : obs_time ] | the synchrotron self - compton ( ssc ) emission from gamma - ray burst ( grb ) forward shock can extend to the very - high - energy ( vhe ; @xmath0100 gev ) range .
such high energy photons are rare and are attenuated by the cosmic infrared background before reaching us . in this work ,
we discuss the prospect to detect these vhe photons using the current ground - based cherenkov detectors .
our calculated results are consistent with the upper limits obtained with several cherenkov detectors for grb 030329 , grb 050509b , and grb 060505 during the afterglow phase . for 5 bursts in our nearby grb sample
( except for grb 030329 ) , current ground - based cherenkov detectors would not be expected to detect the modeled vhe signal . only for those very bright and nearby bursts like grb 030329 , detection of vhe photons is possible under favorable observing conditions and a delayed observation time of @xmath110 hours . |
in recent years a new paradigm for computation has been proposed by craig lent and coworkers@xcite , based on the concept of quantum cellular automata ( qca ) .
such a concept , although extremely difficult to implement from a technological point of view@xcite , has several interesting features that make it worth pursuing .
the basic building block is made up of a single cell , containing two electrons that can be localized in four different areas or `` dots , '' located at the vertices of a square , as shown in each of the cells represented in fig .
[ eins](a ) .
coulomb repulsion forces the two electrons to occupy dots that are aligned along one of the diagonals , and each of the two possible alignments is associated with a logic state . by placing cells next to each other , a wire can be formed ( binary wire ) , along which polarization enforced at one end will propagate@xcite , as a consequence of the system of charges relaxing down to the ground state .
we can see this also as the logic state of the first cell propagating down the chain until it reaches the last cell .
it has been shown that , by properly assembling two - dimensional arrays of cells , it is possible to implement any combinatorial logic function@xcite . the basic principle of operation of such circuits is therefore the relaxation of the system to the ground state , thus leading to the often used expression `` ground - state computation '' . even in the case of perfectly symmetric and identical cells , the configuration of the qca circuit may depart from the ground state as a consequence of thermal excitations .
if the energy separation between the ground state and the first few excited states is small , their occupancy will be nonnegligible even at low temperatures , and the logic output may be corrupted .
a complete understanding of the behavior of qca arrays as a function of temperature is thus essential for any practical application of the qca concept .
the problem of errors due to finite temperature operation was first addressed by lent@xcite , on the basis of entropy considerations .
our approach consists in a detailed study of thermal statistics for qca arrays , retrieving the results of ref.@xcite as a special case , and allowing treatment of cells with more than just two states .
we have developed both a numerical model , which enables us to study relatively short chains made up of six - state cells in full detail , and an analytical model , which can be used for arbitrarily long chains of two - state cells . in both cases ,
we have considered a semiclassical approximation , and computed the probability of the system being in the ground state and that of presenting the correct logic output , i.e. of having the last cell of the chain in the expected logic state . since
only one configuration corresponds to the ground state , while several different configurations are characterized by the correct logic output , the probability of having the correct output is always larger than that of being exactly in the ground state . in sec .
[ enesec ] we present the cell model we have considered for both approaches and the semi - classical approximation that we have chosen to adopt .
we also discuss the structure of the energy spectrum for the excited states of a chain of cells . in sec .
[ partisec ] we present the procedure that has been followed for the calculation of the partition function with the numerical method and the associated results for the probabilities of correct operation as a function of temperature .
the analytical model is described in sec .
[ gebbo ] , together with the associated results and a comparison with those from the numerical model .
our approach is semi - classical insofar as electrons are treated as classical particles , with the only additional property that they can tunnel between dots belonging to the same cell .
this is a reasonable approximation if the tunneling matrix elements between the dots of a cell are small enough to strongly localize the electrons , which therefore behave as well defined particles .
our model chains are characterized by two geometrical parameters : @xmath0 is the distance between neighboring cell centers , @xmath1 is the distance between two dots in a cell .
we represent the _ driver _ cell , i.e. the cell whose polarization state is externally enforced , with bold lines and indicating only the electron positions ( see fig .
[ eins](a ) ) ; _ driven _ cells are represented with solid lines and each dot is indicated with a solid circle if occupied or with an empty circle otherwise .
within each cell we consider a uniformly distributed ( @xmath2 per dot , where @xmath3 is the electron charge ) positive background charge , which makes each cell overall neutral and prevents anomalous behaviors in the nearby cells , due to the uncompensated monopole component of the electrostatic field .
in particular , the repulsive action of the uncompensated electrons in a driver cell can `` push '' the electrons in the nearby driven cell away , thus leading to the formation of an unwanted state in which electrons are aligned along the side further from the driver cell . in our calculations
we have considered the gaas / algaas material system and assumed a uniform relative permittivity of 12.9 : this is a reasonable approximation , since the permittivity of algaas does not differ significantly from that of the gaas layer , where the electrons are confined . for this study
we have neglected , for the sake of simplicity and of generality , the effects of the semiconductor - air interface and of the metal gates defining the dots , whose rigorous treatment would have required considering a specific layout@xcite . for silicon - on - insulator qca cells@xcite , materials with quite different permittivities come into play : silicon , silicon oxide and air , but reasonable estimates could be obtained by repeating our calculations with a relative permittivity corresponding to that of silicon oxide , since most of the electric field lines are confined in the oxide region embedding the silicon dots .
moreover , estimates of the performance obtained with this approximation would be conservative , since part of the field lines are actually in the air over the device , whose relative permittivity is unitary , thus leading to a stronger electrostatic interaction and therefore to a reduced importance of thermal fluctuations .
as we have already stated , the two minimum energy configurations of a cell are those with the electrons aligned along one of the diagonals , since these correspond to the maximum separation between the electrons . however , other configurations are also possible , and , depending on intercell spacing , they can appear in the first few excited states of a binary wire .
we consider all of the six configurations that can be assumed by two electrons in four dots , excluding only those with both electrons in the same dot , which correspond to too large an energy .
we define the two lowest energy configurations ( those with the electrons along the diagonals ) state 1 and state 0 as indicated in fig .
[ eins](b ) , while the corresponding polarization values are @xmath4 and @xmath5 , respectively .
polarization values are defined@xcite as @xmath6 where @xmath7 is the charge in the @xmath8-th dot , with the first dot being at the top right and the others numbered counterclockwise .
configurations with the two electrons along one of the four sides of the cell have higher energies , as stated before , and do not correspond to a well defined logic state .
for this reason , we define them as @xmath9 states .
the energy is computed as the electrostatic energy of a classical system of charges@xcite : @xmath10 since in our model the total charge in each dot is either the background charge ( empty dot ) or the algebraic sum of the background charge and the charge of an electron , it can take on only two values : @xmath11 or @xmath12 , which implies that @xmath13 if we write the interelectronic distance @xmath14 in terms of the ratio @xmath15 and of the configuration , the energy of a binary wire can be written as @xmath16 where @xmath17 is the number of cells between the cell containing dot @xmath8 and the cell containing dot @xmath18 , @xmath19 indicates the sign of @xmath20 , @xmath21 and @xmath22 , indicate the position of dots @xmath8 and @xmath18 inside the corresponding cells . in particular , @xmath23 is equal to 0 if both dots @xmath8 and @xmath18 are on the left side or on the right side of the cell , to -1 if dot @xmath8 is on the right side and dot @xmath18 is on the left side and to 1 if dot @xmath8 is on the left side and dot @xmath18 is on the right one .
furthermore , @xmath24 is equal to 0 if both dots @xmath8 and @xmath18 are on the top or on the bottom of a cell , to 1 if one dot is on the top and the other is on the bottom .
we have considered a binary wire made up of six cells ( one of which is a driver cell in a fixed polarization state ) with size @xmath25 nm and computed the energy values corresponding to all possible @xmath26 configurations .
the values thus obtained have been ordered with the purpose of studying the energy spectra for different parameter choices .
let us define @xmath15 . if @xmath27 , i.e. @xmath28 , the interaction between neighboring cells is substantially due to just the dipole component , and a discrete spectrum is observed already for @xmath29 ( see fig .
[ zwei ] ) , with clear steps : the ground state corresponds to configurations with all cells in the same logic state : either all 1 or 0 ; the first excited state , for @xmath29 , includes configurations with one `` kink , '' i.e. with one cell flipped with respect to the rest of the chain .
higher steps correspond to a larger number of kinks .
energy values are expressed with reference to the ground state energy and in kelvin , i.e. as the result of the division of the actual energies in joule by the boltzmann constant . if @xmath30 is decreased , the interaction between neighboring cells is incremented and made more complex , so that @xmath9 states do appear , as shown in fig .
[ zwei ] for @xmath31 .
the various plateaus start merging and a continuous spectrum is approached . in particular , if we decrease @xmath30 while keeping @xmath1 constant , and thereby reducing the separation between neighboring cells , the difference between the energy of the ground state and that of the first excited state is expected to increase as @xmath32 , due to the increased electrostatic interaction .
however , this is true only down to a threshold value of @xmath30 , below which the splitting between the first excited state and the ground state starts decreasing , as shown in fig .
[ drei ] , where the energy split is plotted as a function of @xmath30 for a cell size @xmath1 of 40 nm , for a wire with 2 ( dotted line ) , 3 ( dashed line ) , and 6 ( solid line ) cells . this sudden change of behavior can be understood on the basis of the previously discussed results : below the threshold value for @xmath30 , the configuration for the first excited state contains a cell in the @xmath9 state , thereby disrupting the operation of the wire and lowering the splitting between the first excited state and the ground state . in the inset of fig .
[ drei ] we report the dependence of the splitting between the two lowest energy states on the number of cells .
these results are for @xmath29 , i.e. for a condition in which no @xmath9 state appears . once the number of cells is larger than a few units , the splitting quickly saturates to a constant value .
this is easily understood if we consider that the first excited state is characterized by the cell at the end of the wire being polarized opposite to the others : the strength of the electrostatic interaction drops quickly along the chain , and hence no significant change is determined by the addition of cells beyond the first five or six .
the energy splitting has been computed for a cell size @xmath25 nm and , as can be deduced from eq.([eq1 ] ) , is inversely proportional to @xmath1 .
it can thus be increased by scaling down cell dimensions .
in order to compute the probabilities , at a finite temperature , for the various configurations , we introduce the partition function of the wire : @xmath33 where @xmath34 is the energy of the @xmath8-th configuration and @xmath35 , @xmath36 being the boltzmann constant and @xmath37 the temperature .
the summation is performed over all configurations with the first cell in a given input logic state .
the probability @xmath38 of the entire system being in the ground state can be evaluated by taking the ratio of the boltzmann factor for the ground state to the partition function : @xmath39 where @xmath40 , and the sum extends over all excited states . as already mentioned in the introduction
, @xmath38 is not the only quantity of interest . from the point of view of applications
, we are mainly interested in knowing the probability @xmath41 of obtaining the correct logic output , which is higher than @xmath38 , because several configurations , besides the ground state , exhibit the correct polarization for the output cell ( the cell at the end of the chain )
. we can compute @xmath41 by summing over the probabilities corresponding to all such configurations , that we label with the subscript @xmath18 : @xmath42 we have computed both @xmath38 and @xmath41 as a function of the ratio of the splitting @xmath43 between ground state and first excited state to @xmath44 . the results for a chain of 6 cells are presented in fig .
[ vier ] : in the limit @xmath45 all the configurations ( a total of @xmath46 , @xmath47 being the number of cells ) become equally probable and the probability @xmath38 reaches its minimum value @xmath48 .
the probability of correct logic output , instead , reaches a minimum value of @xmath49 , as a consequence of the six possible states of the output cell being equally probable .
it should be noted that an error probability of a few percent may appear unacceptable for any practical circuit application , but data readout must always be done via some detector@xcite , which is characterized by a time constant necessarily longer than the typical settling time of the qca circuit .
therefore , each reading will be the result of an averaging procedure , and will be compared to a threshold value . in such a case ,
an error probability of a few percent for the output state will lead in most cases to a vanishingly small error probability for the actual output of the readout circuit .
as already noted , the number of possible configurations for a circuit with @xmath47 cells ( one of which is assumed to be the driver cell and hence in a given , fixed configuration ) is @xmath46 .
thus the cpu time required to explore all such configurations grows exponentially with the number of cells , which limits the length of the binary wires that can be investigated with this approach in a reasonable time down to about ten cells . in order to assess the thermal behavior of long wires , we have developed the approximate analytical approach that will be described in the next section .
the development of an analytical model for the investigation of the thermal behavior of a qca chain requires a main simplifying assumption , in order to make the algebraic treatment possible : for each cell we consider only two configurations , the ones corresponding to the logic states 1 and 0 , and , thus , to polarization @xmath50 and @xmath5 . from the discussion in sec .
[ enesec ] it is apparent that the larger @xmath30 , the better this approximation is , because the role of the @xmath9 states is reduced .
let us consider a generic 1-dimensional chain consisting of @xmath47 cells and introduce the following 1-dimensional ising hamiltonian @xmath51 where for each cell labeled by the index @xmath8 the variable @xmath52 corresponds to the polarization and therefore assumes the two values @xmath53 , and the positive quantity @xmath54 ( which has the dimension of an energy ) is related to the splitting @xmath43 between the ground state and the first excited state energies of an @xmath47-cell system by @xmath55 .
let us point out that there is a twofold degeneracy of the ground state , corresponding to the two configurations @xmath56 and @xmath57 .
this degeneracy is removed by enforcing the polarization state of the driver cell , which corresponds to enforcing the configuration of one of the boundary sites ; our conventional choice is @xmath58 . in this case
, the lowest energy state corresponds to the configuration @xmath56 .
the partition function of the @xmath47-cell system described by the hamiltonian ( [ hising ] ) with the boundary condition @xmath58 is given , in analogy with eq.([partifun ] ) , by the following expression : @xmath59 where @xmath60 stands for the summation over all possible states , i.e. @xmath61 .
this last expression can be written as @xmath62 where @xmath63 . in order to compute the r.h.s .
( [ vtmp ] ) , the usual procedure consists in introducing the transfer matrix @xcite @xmath64 whose eigenvalues are @xmath65 the expression for the matrix @xmath66 is given by @xmath67 it then follows that the partition function ( [ vtmp ] ) reads @xmath68_{11}+[{\cal v}^{n-1}]_{12}= ( e^{\beta j}+e^{-\beta j})^{n-1 } , \label{zetaising}\ ] ] where the subscripts indicate specific elements of the @xmath66 matrix .
this explicit formula for the partition function allows us to derive an analytical expression for the probability of the system being in its ground state as a function of the temperature and of the energy splitting between the two lowest states .
since the ground state energy for the hamiltonian ( [ hising ] ) is @xmath69 , we obtain @xmath70 finally , we can derive an analytical expression also for the probability of obtaining the correct logic output , in analogy with what we have already done in the numerical case .
we need to determine the occupation probability of a generic state with @xmath71 , which corresponds to having the correct output , because the polarization of the @xmath72th cell ( output cell ) is the same as that of the first cell . to this purpose , we evaluate the following `` reduced '' partition function : @xmath73 where again @xmath63 . using the transfer matrix ( [ tram ] ) , it follows that @xmath74_{11}$ ] , and hence @xmath75_{11 } } { [ { \cal v}^{n-1}]_{11}+[{\cal v}^{n-1}]_{12}}= \frac{1}{2}\ , \left[1+\left(\tanh(\beta\delta e/2)\right)^{n-1}\right ] \label{porod2}\ ] ] the above derived analytical expressions have been used to compute @xmath76 and @xmath77 as a function of temperature for a chain of 6 cells , cell size @xmath25 nm , cell separation @xmath78 nm .
results are presented with dashed lines in fig .
[ fuenf ] , together with those obtained with the numerical technique ( solid lines ) . for temperatures below about 2 k ( those for which reasonably low error probabilities can be achieved ) the analytical model provides values that are in almost perfect agreement with those from the more detailed numerical approach .
the situation differs at higher temperatures , because higher energy configurations , containing cells in @xmath9 states , start being occupied and are properly handled by the numerical model , while they are not at all included in the analytical approach . in particular , while for large values of the temperature the numerical @xmath77 tends to @xmath49 , as previously discussed , the analytical @xmath77 approaches the value @xmath79 , because the output cell can be in one of two states with the same probability .
analogous considerations can be made for @xmath76 , which becomes extremely small ( @xmath80 ) for higher temperatures in the numerical case , while drops just to @xmath81 in the analytical case , since there are @xmath82 possible configurations . in fig .
[ sechs ] , @xmath77 and @xmath76 are reported as a function of the ratio @xmath83 ( with a semilogarithmic scale ) for the analytical model ( thick solid line ) , and for the numerical model with @xmath84 ( thin solid line ) , @xmath29 ( dashed line ) , and @xmath85 ( dotted line ) .
as expected , the agreement improves with increasing @xmath30 , because of the reduced relevance of the @xmath9 states . for @xmath83 of the order of a few units ,
the error probability becomes very small and the analytical expression can be reliably used to evaluate it . in particular , the analytical expression allows us to provide estimates of the maximum operating temperature for a qca chain formed by a given number of cells .
we have computed the maximum operating temperature allowing a given correct logic output probability , as a function of the number of cells : results are reported in fig .
[ sieben ] for @xmath86 ( solid line ) , @xmath87 ( dashed line ) , @xmath88 ( dotted line ) and cell size @xmath25 nm , intercell separation @xmath78 nm .
the maximum operating temperature , for a number of cells above a few tens , drops logarithmically , which leads to a linear behavior in the logarithmic representation of fig .
[ sieben ] .
we have developed both a numerical and an analytical approach to the investigation of the thermal dependence of qca wire operation .
both methods are based on a semiclassical approach , in which electrons are considered as classical particles interacting via the coulomb force , with , however , the possibility of tunneling between the quantum dots belonging to the same cell .
the electrostatic energy associated with each configuration has been evaluated and used for the calculation of the occupancies , via the partition function .
numerical results have been derived for wires with six - state cells , which realistically reproduce the behavior of qca systems , provided that the confinement in each quantum dot is strong enough .
the numerical procedure thus developed is general and is currently being applied to the investigation of thermal limitations for simple logic gates , including the effect of spurious , @xmath9 states .
the analytical approach has allowed a detailed analysis of the error probability due to thermal excitations in arbitrarily long wires , generalizing the findings of previous studies , and the possibility of extending it to selected basic gates is being investigated .
it is clear from our results that the operating temperature depends on the ratio of the energy splitting @xmath43 to @xmath44 .
it could therefore be raised by increasing @xmath43 , which means reducing the dielectric permittivity or scaling down cell dimensions .
as already mentioned , the silicon - on - insulator material system offers better perspectives of higher - temperature operation , due to the lower permittivity of silicon oxide .
however , scaling down in any semiconductor implementation is limited by the increasing precision requirements@xcite , therefore a trade - off between manufacturability and operating temperature has to be accepted .
implementations at the molecular level could provide better opportunities , due to the reduced dimensions , but their actual feasibility is still being assessed . c. s. lent , p. d. tougaw , and w. porod , appl .
. lett . * 62 * , 714 ( 1993 ) .
m. governale , m. macucci , g. iannaccone , c. ungarelli , j. martorell , j. appl .
phys . , * 85 * , 2962 ( 1999 ) .
p. d. tougaw and c. s. lent , j. appl
. phys . * 75 * , 1818 ( 1994 ) . c. s. lent , p. d. tougaw , and w. porod , in `` the proceedings of the workshop on physics and computing , '' nov .
17 - 20 , 1994 , dallas , tx , p. 1 .
g. iannaccone , c. ungarelli , m. macucci , e. amirante , m. governale , thin solid films , * 336 * , 145 ( 1998 ) .
f. e. prins , c. single , f. zhou , h. heidemeyer , d. p. kern , e. plies , nanotechnology , * 10 * , 132 ( 1999 )
. m. girlanda , m. governale , m. macucci , g. iannaccone , appl .
, * 75 * , 3198 ( 1999 ) .
j. d. jackson , _ classical electrodynamics _
( wiley , new york , 1962 ) , p. 20 . see e.g. j. r. baxter , _ exactly solved models in statistical mechanics _
( academic press , london , 1982 ) . | we investigate the effect of a finite temperature on the behavior of logic circuits based on the principle of quantum cellular automata ( qca ) and of ground state computation .
in particular , we focus on the error probability for a wire of qca cells that propagates a logic state . a numerical model and an analytical , more approximate , model are presented for the evaluation of the partition function of such a system and , consequently , of the desired probabilities .
we compare the results of the two models , assessing the limits of validity of the analytical approach , and provide estimates for the maximum operating temperature . |
our view of how our universe may be described within a higher - dimensional spacetime has been revolutionised via brane - world models @xcite ( see , c.g .
, @xcite for review ) . a simple way to study the effects of the higher - dimensional gravity on inflation models is to consider a 4d inflaton field confined to the brane in the context of the randall - sundrum model @xcite .
work to date has concentrated mainly on deriving corrections to the standard 4d result by assuming that 5d metric perturbations generated by the inflaton perturbations can be neglected .
then the modification of spectrum is coming from the modified friedmann equation .
although a lot of interesting results have been obtained in this direction @xcite , it remains to see the consistency of neglecting the bulk metric perturbations .
the analysis of the evolution of gravitational waves in this model has revealed that the creation of massive graviton modes in the bulk is as important as the modification of the friedmann equation @xcite .
in fact , these two higher - dimensional effects cancel out to give a nearly identical spectrum with the conventional 4d models for the stochastic background of gravitational waves in the radiation dominated universe @xcite .
this highlights the importance of taking into account the higher - dimensional effects consistently .
the study of the effects of bulk metric perturbations has been initiated in ref .
@xcite ( see @xcite for a different approach ) . using the approximation that the geometry of the brane is approximated by de sitter spacetime for a slow - roll inflation ,
the bulk metric perturbations sourced by inflaton fluctuations on the brane were calculated .
it turns out that the inflaton fluctuations excite an infinite ladder of massive modes of bulk metric perturbations . then ref .
@xcite tried to estimate the effects of the backreaction of bulk metric perturbations . assuming that the inflaton field behaves in the same ways as 4d models on sufficiently small scales at the 0-th order of slow - roll parameters , the corrections to the evolution of inflaton perturbations at the first order of slow - roll parameters
were calculated .
unlike in 4d models where the slow - roll corrections can be completely ignored on sub - horizon scales , the slow - roll corrections play a role even on sub - horizon scales , yielding the change of the amplitude at a horizon crossing
. this result may not be so surprising as the bulk metric perturbations are strongly coupled to inflaton perturbations on small scales where gravity becomes 5d .
however , this indicates that we should carefully check the validity of using the 4d formula just by modifying the background friedmann equations .
this paper focuses on the classical evolution of inflaton perturbations coupled to bulk metric perturbations .
we do not assume de sitter geometry on the brane and we take into account the backreaction of inflaton dynamics consistently .
this entails a numerical analysis to solve the coupled system directly .
we investigate whether the inflaton perturbations behave as free massless fields on small scales .
the structure of this paper is as follows . in section
ii , we set up basic equations to describe the evolution of inflaton perturbations coupled to metric perturbations .
we use the hawkins - lidsey model of the inflation on the brane where the background solution is derived analytically @xcite .
the gaussian - normal coordinate is used to solve the perturbations .
section iii is devoted to the numerical scheme . in section iv , we explain initial conditions and boundary conditions
. then we study the evolution of the comoving curvature perturbations , which are directly related to observables .
we consider the randall - sundrum single - brane model @xcite . in this model ,
a 3d spatial hypersurface is embedded in a 5d anti - de sitter ( ads@xmath0 ) space - time with a negative cosmological constant @xmath1 .
the 5d einstein equations in the bulk are given by @xmath2 the cosmological constant is related with the curvature scale of the ads@xmath0 space - time @xmath3 as @xmath4 .
the embedded brane has a tension @xmath5 given by @xmath6 . here
@xmath7 denotes the 5d gravitational constant which can be rewritten using the 5d planck mass @xmath8 as @xmath9 .
on the other hand , we use @xmath10 as 4d one defined as @xmath11 which is related with @xmath7 as @xmath12 , where @xmath13 is the 4d planck mass .
the induced metric on the brane is defined as @xmath14 where @xmath15 denotes the unit normal vector to the brane .
the energy - momentum tensor @xmath16 on the brane and a brane tension @xmath17 produce the jump in the extrinsic curvature at the brane ; @xmath18 , \label{eq : k_jump}\ ] ] where we imposed the @xmath19 symmetry on the brane and the extrinsic curvature is defined by @xmath20 . according to ref .
@xcite , the effective einstein equation for the induced metric on the brane is obtained as @xmath21 where @xmath22 is the quadratic terms of the energy - momentum tensor , @xmath23 and @xmath24 is the electric part of 5d weyl tensor @xmath25 , @xmath26 we assume that the matter content confined to the brane is described by a perfect fluid whose energy - momentum tensor is given by @xmath27 where @xmath28 and @xmath29 denote the energy density and the pressure of matter on the brane , respectively . in this paper , we use the gaussian - normal ( gn ) coordinate .
the metric is given by @xcite @xmath30 here the warp function @xmath31 and the lapse function @xmath32 are obtained in @xcite as @xmath33 where @xmath34 is the scale factor on the brane , and we can set @xmath35 without a loss of generality
. the scale factor is determined by the friedmann equation and the energy conservation law as @xmath36 in this coordinate , the brane is located at @xmath37 . note that , in this coordinate , there is a coordinate singularity at the finite distance from the brane , @xmath38 where the warp function vanishes @xmath39 .
the location of the coordinate singularity @xmath40 agrees with the cauchy horizon of the ads@xmath0 spacetime @xcite .
we will come back to this issue later . in this paper , we consider a scalar field on the brane .
the scalar field satisfies the equation of motion @xmath41 where @xmath42 denotes the potential of the scalar field .
we use a model proposed by hawkins - lidsey as an inflaton model on the brane to avoid numerical integrations of the background equations @xcite .
the potential for inflaton is given by @xmath43 where @xmath44 is constant .
an advantage of this potential is that an exact solution for the scale factor is known @xmath45^{1/c^2}.\ ] ] the energy density and pressure of the scalar field can be calculated as @xmath46 thus the effective equation of state @xmath47 becomes @xmath48 then for small @xmath49 , a slow - roll inflation is realized , which is a generalisation of the power - law inflation .
now let us consider the scalar perturbations .
there are several ways to calculate the scalar perturbations in a ads@xmath0 spacetime . here
we describe an approach based on a master variable @xcite . using the generalized 5d longitudinal gauge ,
the perturbed spacetime is given by @xmath50 it was shown that the solution for metric perturbations can be derived from a master variable @xmath51 as @xmath52 as long as the master variable @xmath51 in the bulk satisfies a wave equation given by @xmath53 where a prime and a dot denote the derivatives with respect to @xmath54 and @xmath55 , respectively . in this section ,
we derive the evolution equation for inflaton perturbations coupled to the master variable .
see refs .
@xcite and references therein for earlier works .
we begin with the effective einstein equations on the brane @xcite .
the perturbed effective einstein equations are given by @xmath56 the perturbed einstein tensor is given by @xcite @xmath57 where @xmath58 the subscript b denotes a quantity evaluated on the brane .
we can parameterize the projected weyl tensor @xmath59 as an effective fluid ; @xmath60 where @xmath61 the weyl fluid satisfies the radiation - like equation - of - state @xmath62 .
the perturbed energy - momentum tensor @xmath63 is defined as @xmath64 throughout this paper , we focus on the perturbations of the scalar field .
thus the anisotropic stress @xmath65 on the brane vanishes .
the quadratic energy - momentum tensor @xmath66 is calculated as @xcite @xmath67 all perturbative quantities can be expanded by scalar harmonic functions @xmath68 , which enables us to replace the operator @xmath69 by @xmath70 . then , from the effective einstein equations ( [ eq : pert_effective_einstein ] ) , we can derive the evolution equations for the perturbative quantities as @xcite @xmath71 where we defined @xmath72 and used the relation @xmath73 derived from eqs .
( [ eq : friedmann_equation ] ) and ( [ eq : conservation_law ] ) . ref .
@xcite showed that the components of the perturbative weyl tensor @xmath74 can be written in terms of the master variable @xmath51 as @xmath75 the perturbed energy - momentum tensor is calculated as @xmath76 the equation of motion for the scalar field perturbations is derived from the conservation law @xmath77 as @xmath78 now we introduce a gauge - invariant variable , the mukhanov - sasaki variable , defined as @xcite @xmath79 substituting this definition into eq .
( [ eq : brane_equation ] ) and using the effective einstein equations , we can derive the evolution equation for @xmath80 as @xmath81 where @xmath82 .
\label{eq : mukhanov_sasaki_source}\ ] ] here the left - hand side of eq .
( [ eq : mukhanov_sasaki_equation ] ) is completely the same as the 4d cosmology while @xmath83 describes the the effect of the bulk metric perturbations . using eq .
( [ eq : master_equation ] ) , the curvature perturbations ( [ eq : metric_r ] ) evaluated at the brane @xmath84 can be expressed in terms of @xmath51 and its first - order derivatives as @xmath85 from the @xmath86 component of the effective einstein equation ( [ eq : effective_einstein_3 ] ) , we can relate @xmath80 and @xmath51 as @xmath87_{\rm b } , \label{eq : junction-2}\ ] ] which gives a non - local boundary condition for @xmath51 .
alternatively , we can combine eqs.([eq : effective_einstein_1 ] ) and ( [ eq : effective_einstein_3 ] ) to get @xmath88 then using eqs.([eq : scalar_energy_momentum_1 ] ) and ( [ eq : scalar_energy_momentum_3 ] ) and rewriting @xmath89 by @xmath80 , we obtain @xmath90 in summary , the master equation ( [ eq : master_equation ] ) , the mukhanov - sasaki equation ( [ eq : mukhanov_sasaki_equation ] ) with a source ( [ eq : mukhanov_sasaki_source ] ) and the junction condition ( [ eq : junction-2 ] ) or ( [ eq : junction ] ) form a closed system for the coupled inflaton and metric perturbations . note that , combining eqs .
( [ eq : effective_einstein_1 ] ) - ( [ eq : effective_einstein_3 ] ) , one can derive a boundary condition for @xmath51 which does not include @xmath80 @xcite .
this boundary condition is equivalent to choose ( [ eq : mukhanov_sasaki_equation ] ) and ( [ eq : junction-2 ] ) as boundary conditions .
in this subsection , we briefly discuss the numerical scheme to solve the equations .
first of all , in order to avoid the effect from the coordinate singularity at @xmath91 in the gn coordinate , we introduce an artificial cutoff ( regulator ) boundary in the bulk at @xmath92 . the constant @xmath93 controls the position of the regulator brane . since the regulator brane should be far from the physical brane and @xmath94 , we set @xmath95 in our numerical simulations .
we discuss the boundary condition at the regulator brane in the next section . in this paper ,
the numerical calculations of this system were carried out by employing the pseudo - spectral method @xcite .
the master variable @xmath96 is decomposed by the tchebychev polynomials defined as @xmath97 which is a polynomial function of @xmath98 of order @xmath99 . here
the variable @xmath98 is related to the gn coordinate @xmath54 . to implement the pseudo - spectral method , we must transform a spatial coordinate @xmath54 in the gn coordinate by @xmath100 so that the locations of both the physical and the regulator branes are kept fixed and the spatial coordinate @xmath54 is projected to the compact domain @xmath101 . adopting this coordinate system ,
the quantity @xmath102 is first transformed into the tchebychev space through the relation , @xmath103 where we set @xmath104 or @xmath105 .
we then discretize the @xmath98-axis to the @xmath106 points ( collocation points ) using the inhomogeneous grid @xmath107 called gauss - lobatto collocation points . with this grid
, the fast fourier transformation can be applied to perform the transformation between the amplitude @xmath102 and the coefficients @xmath108 . then the master equation ( [ eq : master_equation ] ) rewritten in the new coordinates
are decomposed into a set of ordinary differential equations for @xmath108 . for the temporal evolution of @xmath108 and @xmath109
satisfying eq.([eq : mukhanov_sasaki_equation ] ) , we use adams - bashforth - moulton method with the predictor - corrector scheme .
further technical details of the numerical scheme is summarized in the appendix of ref . @xcite .
the boundary condition at the brane is given by the junction condition eq .
( [ eq : junction-2 ] ) or eq .
( [ eq : junction ] ) . at the regulator brane
, we use a dirichlet boundary condition @xmath110 this is because the analysis in de sitter brane background @xcite suggests that the inflaton fluctuation excites the bound states that decays quickly away from the brane .
if we focus on this bound state the result should not be sensitive to the location of the regulator brane .
we will check this fact numerically later .
different boundary conditions from eq .
( [ eq : dirichlet ] ) may produce reflection waves from a regulator brane , which affect the evolution of q. a proper boundary condition should be imposed on the past cauchy horizon .
this requires a quantisation of 5d metric perturbations in the bulk , which will turn out to be a highly non - trivial problem due to the coupling to the inflaton perturbations . in 4d cases
, @xmath80 satisfies @xmath111 on small scales since @xmath112 term is dominant over the mass term induced by the dynamics of the inflaton [ see ( [ eq : mukhanov_sasaki_equation ] ) ] .
thus the solution on small scales is given by simple plane waves and we can choose the usual 4d bunch - davis vacuum state , which specifies an appropriate phase and normalization . on the other hand , in the present 5d case , it is not guaranteed that the perturbed inflaton field obeys the usual 4d klein - gordon equation due to the coupling to the bulk metric perturbations , which is described by @xmath83 in the right - hand side of eq .
( [ eq : mukhanov_sasaki_equation ] ) . in order to obtain the appropriate initial conditions for @xmath109
, we again need a quantisation of the coupled system of the inflaton perturbations on a brane and 5d metric perturbations in the bulk .
the aim of this paper is to check whether we can neglect @xmath83 on small scales or not .
if this term could be neglected on the small scales , the quantity @xmath113 would behave just as plane waves with a constant amplitude as in the standard 4d cases .
we call the solution of eq .
( [ eq : mukhanov_sasaki_equation ] ) with @xmath114 or , equivalently , eq .
( [ eq : evolution_4d ] ) on small scales 4d solutions .
if the solutions for the coupled equations agree with 4d solutions , we can assume that the inflaton perturbation behaves as a free massless field without taking into account the effect of 5d metric perturbations , which is the assumption made in literatures . for this purpose ,
we take the simplest possible initial conditions for @xmath115 and then specify initial conditions for @xmath109 so that they are consistent with the junction condition ( [ eq : junction ] ) .
we set @xmath116 , @xmath117 and @xmath118 .
then the time - derivative of @xmath109 is determined from the junction condition ( [ eq : junction ] ) .
note that our initial conditions do not give standard qunatum vacuum state for 4d solutions .
they are merely references to see the effects of @xmath83 on a classical evolution of @xmath109 .
we solve numerically the wave equation for the bulk metric perturbations ( [ eq : master_equation ] ) in the gn coordinates with the junction condition ( [ eq : junction ] ) supplemented by the evolution equation for inflaton perturbations ( [ eq : mukhanov_sasaki_equation ] ) .
we set the parameter of the hawkins - lidsey model as @xmath119 and the initial time for the simulations as @xmath120 . at the initial time , the hubble scale is given by @xmath121 thus we are considering high - energy region @xmath122 .
the observable is the comoving curvature perturbation defined by @xmath123 we focus on the dynamics of this quantity .
[ fig : full_r ] shows the behaviour of curvature perturbations for @xmath124 which corresponds to @xmath125 .
the curvature perturbations @xmath126 becomes constant on super - horizon scales , which has been shown to be valid even in brane world models @xcite . on sub - horizon scales , while @xmath127 in the 4d cosmology , a suppression of the amplitude is observed in the brane - world model , which is due to the coupling to the bulk metric perturbations . in fig .
[ fig : full_ar ] , we compare the amplitude of @xmath126 obtained in numerical simulations with the one obtained by neglecting the coupling to the bulk metric perturbations , i.e. @xmath114 in eq .
( [ eq : mukhanov_sasaki_equation ] ) .
while the difference is very small for the long - wavelength modes ( left panel ; @xmath128 ) , the suppression becomes significant for the short - wavelength modes ( right panel ; @xmath129 ) . due to this , the spectrum of @xmath130 just before the horizon crossing acquires a scale dependence as is shown in fig .
[ fig : spectrum ] .
perturbations with larger wavenumber stay on sub - horizon scales for a longer time than those with smaller wavenumber , so they receive more suppression . the suppression of the amplitude @xmath131 under horizon may be understood as the excitation of metric perturbations .
the suppression of the amplitude of @xmath80 is transferred into the enhancement of the metric perturbations @xmath51 in the bulk , which is shown in the left panel of fig . [
fig : full_omega ] .
here we comment on the reason why we evaluated @xmath131 just before the horizon crossing . in the 4d case
, we can choose the bunch - davis vacuum as an initial condition .
hence we can determine the amplitude and phase of the inflaton perturbations and obtain the spectrum of @xmath131 without any ambiguities .
on the contrary , in the present 5d case , the initial conditions can not be determined without quantising the coupled system . even in the classical level
, the junction condition ( [ eq : junction-2 ] ) or ( [ eq : junction ] ) restricts our choice of the initial conditions .
on superhorizon scales , the initial phase of the perturbations essentially determines its final amplitude after the horizon crossing because the phase determines the amplitude of the growing mode solution on large scales .
then we get an unphysical k - dependence of the curvature perturbations evaluated on superhorizon scales due to the restricted choice of @xmath132 for given @xmath133 . on the other hand ,
the initial phase is not important when we evaluate the amplitude of the curvature perturbations on small scales .
our aim is to see how @xmath83 term changes the amplitude of @xmath80 in comparison with the 4d solution for various @xmath133 on small scales .
therefore we chose to calculate the spectrum below the horizon scales . in this section
, we first check the assumption of introducing a regulator .
as expected from the previous work , @xmath51 decays fast away from the brane as is shown in the right panel of fig .
[ fig : full_omega ] .
this justifies the use of boundary condition @xmath134 .
in addition , thanks to this behaviour of @xmath51 , the evolution of the curvature perturbations does not depend on the location of the regulator .
in fact , even at late times when the information from the regulator brane comes into the brane , the behaviour of the curvature perturbations does not change .
thus our result is not obscured by the regulator brane nor the existence of the coordinate singularity of the gn coordinate .
finally , we check the accuracy of our numerical simulations by checking the constraint equations derived from the einstein equations .
these equations can be generally expressed as @xmath135 where @xmath136 and @xmath137 is a set of their coefficients .
we calculated the relative error for the constraint equations ( [ eq : effective_einstein_1 ] ) ( [ eq : effective_einstein_3 ] ) and ( [ eq : effective_einstein_4 ] ) as @xmath138 note that the constraint equation ( [ eq : effective_einstein_2 ] ) contains higher derivative terms of @xmath51 like @xmath139 , which are not solved in our simulations .
hence it is impossible to evaluate the numerical accuracy of ( [ eq : effective_einstein_2 ] ) from our numerical simulation .
in fact , if we rewrite the higher - derivative terms of @xmath51 by @xmath80 using eq .
( [ eq : junction-2 ] ) , eq .
( [ eq : effective_einstein_2 ] ) becomes an identity .
[ fig : constraint ] shows that the relative error of eq .
( [ eq : effective_einstein_1 ] ) is suppressed to less than @xmath140 and eqs .
( [ eq : effective_einstein_3 ] ) and ( [ eq : effective_einstein_4 ] ) are very precisely satisfied , indicating that our numerical scheme for solving the coupling system ( [ eq : master_equation ] ) and ( [ eq : mukhanov_sasaki_equation ] ) is reliable .
note that the constraint equation ( [ eq : effective_einstein_4 ] ) has a different character from the others , which measures the numerical accuracy of the master equation ( [ eq : master_equation ] ) on the brane . )
[ _ upper - left _ ] , ( [ eq : effective_einstein_3])[_upper - right _ ] and ( [ eq : effective_einstein_4])[_lower _ ] .
these plots are results of calculations for @xmath141 .
[ fig : constraint],width=283 ]
in this paper , we studied the evolution of the curvature perturbations in a brane - world inflation model where an inflaton is living on a single brane in a 5d ads spacetime .
we used the hawkins - lidsey model of the inflaton potential which enables us to derive the background solutions analytically .
we solved the full coupled evolution equations for the inflaton perturbations described by the mukhanov - sasaki variable and the bulk metric perturbations described by the master variable .
this is the first numerical result for the evolution of scalar curvature perturbations in a brane - world that consistently takes into account the backreaction of the metric perturbations .
we used the gn coordinate to describe the bulk spacetime that has a coordinate singularity .
then we are forced to introduce a regulator brane to cut the spacetime .
we have checked that the evolution of curvature perturbations is insensitive to the location of the regulator .
this is because the bulk metric perturbations decay fast away from the brane .
we adopted the simplest possible initial condition for @xmath51 by taking @xmath142 and @xmath143 at an initial time .
then we fix @xmath80 so that the boundary condition is consistent . in order to check
whether we can consistently neglect the coupling to bulk metric perturbations described by @xmath83 in mukhanov - sasaki equation ( [ eq : mukhanov_sasaki_equation ] ) or not , we followed the subsequent evolution of curvature perturbations numerically .
the evolution of curvature perturbations showed a suppression of the amplitude compared with the conventional 4d model even on sub - horizon scales .
this means that it is impossible to neglect a coupling to gravity even on small scales due to the coupling to the higher - dimensional gravity through @xmath83 .
the suppression of the amplitude may be understood as a loss of energy due to the excitation of the bulk metric perturbations as we took the initial condition that @xmath144 and @xmath145 . on super - horizon scales ,
the curvature perturbations become constant , which confirms the fact that the constancy of the curvature perturbations is independent of gravitational theory for adiabatic perturbations @xcite .
our result suggests that an usual assumption that the inflaton perturbations ( the mukhanov - sasaki variable ) approach to a free massless field on small scales can not be applied in a brane world models on small scales at high energies .
then the spectrum becomes sensitive to initial conditions not only for the inflaton perturbations but also for bulk metric perturbations . in this sense
, we should not take the result fig .
[ fig : spectrum ] at face value because this result is based on a special choice of initial conditions . in order to determine the initial conditions without ambiguity
, it is required to quantise the coupled brane ( inflaton)-bulk ( metric perturbations ) system in a consistent manner along the line of @xcite .
this is an outstanding open question to be addressed in order to derive the spectrum of the inflaton perturbations in a brane - world .
e. ramirez and a. r. liddle , 2004 _ phys .
d _ * 69 * 083522 [ arxiv : astro - ph/0309608 ] .
d. seery and a. taylor , 2005 _ phys .
d _ * 71 * 063508 [ arxiv : astro - ph/0309512 ] ; g. calcagni , 2003 _ jcap _ * 0311 * 009 [ arxiv : hep - ph/0310304 ] ; 2004 _ jcap _ * 0406 * 002 [ arxiv : hep - ph/0312246 ] . a. liddle and a. n. taylor , 2002 _ phys . rev .
d _ * 65 * 041301 ; a. r. liddle and a. j. smith , 2003 _ phys .
d _ * 68 * 061301 ; s. tsujikawa and a. r. liddle , 2004 _ jcap _ * 0403 * 001
. t. hiramatsu , k. koyama and a. taruya , 2004 _ phys .
b _ * 578 * 269 [ arxiv : hep - th/0308072 ] ; t. hiramatsu , k. koyama and a. taruya , 2005 _ phys .
b _ * 609 * 133 [ arxiv : hep - th/0410247 ] ; t. kobayashi and t. tanaka , 2005 _ phys . rev .
d _ * 71 * 124028 [ arxiv : hep - th/0505065 ] . t. kobayashi and t. tanaka , 2006 _ phys .
d _ * 73 * 044005 [ arxiv : hep - th/0511186 ] ; t. kobayashi , 2006 _ phys .
d _ * 73 * 124031 [ arxiv : hep - th/0602168 ] ; k. ichiki and k. nakamura , 2004 _ phys .
d _ * 70 * 064017 [ arxiv : hep - th/0310282 ] ; s. s. seahra , [ arxiv : hep - th/0602194 ] . m. sasaki , 1986 _ prog . theor
_ * 76 * 1036 ; v. f. mukhanov , 1988 _ sov . phys
. jetp _ * 67 * 1297 [ 1988 _ zh . eksp
. teor .
fiz_. * 94n7 * 1 ] ; v. f. mukhanov , h. a. feldman and r. h. brandenberger , 1992 _ phys . rept .
_ * 215 * 203 . | we study the evolution of scalar curvature perturbations in a brane - world inflation model in a 5d anti - de sitter spacetime .
the inflaton perturbations are confined to a 4d brane but they are coupled to the 5d bulk metric perturbations .
we numerically solve full coupled equations for the inflaton perturbations and the 5d metric perturbations using hawkins - lidsey inflationary model . at an initial time , we assume that the bulk is unperturbed .
we find that the inflaton perturbations at high energies are strongly coupled to the bulk metric perturbations even on subhorizon scales , leading to the suppression of the amplitude of the comoving curvature perturbations at a horizon crossing .
this indicates that the linear perturbations of the inflaton field does not obey the usual 4d klein - gordon equation due to the coupling to 5d gravitational field on small scales and it is required to quantise the coupled brane - bulk system in a consistent way in order to calculate the spectrum of the scalar perturbations in a brane - world inflation . |
recently a renormalizable theory of gravity in 3 + 1 dimensions was proposed by hoava , inspired from the lifshitz theory in solid state physics , now known as hoava - lifshitz(hl ) theory@xcite .
the theory is a potential candidate of quantum field theory of gravity .
it assumes a lifshitz - like anisotropic scaling between space and time at short distances , characterized by a dynamical critical exponent z = 3 and thus breaking the lorentz invariance .
while in the ir limit it flows to z = 1 , retrieving the einstein s general relativity(gr ) .
the discussions on the consequences of theory are going on@xcite . as a new theory , it is interesting to investigate its various aspects in parallel . since the theory has the same newtonian and post newtonian corrections as those of gr ,
systems of strong gravity , like black holes , are needed to get observable deviation from the standard gr .
various black hole solutions are found in hl theory@xcite of which one with asymptotically flat minkowski spacetime is the ks black hole obtained by applying deformation in the original theory@xcite .
various aspects of ks black hole were explored in the past@xcite .
perturbation of black hole is the best way to study the properties of these exotic objects .
as it is well established , the evolution of perturbations in black hole space times involves three stages@xcite .
the first one is an initial response containing the information of the initial form of the original wave field followed by a region dominated by damped oscillation of the field called quasinormal modes , which depends entirely on the background black hole spacetimes .
finally the field decays at late time as a power - law fall off of the field .
qnms play an important role in astrophysical processes involving black holes .
these interests are associated with their relevance in gravitational wave analysis@xcite .
apart from the observational interest of detection of quasinormal ringing by gravitational wave detectors@xcite , study of qnm of ads black holes are of particular interests in the study of ads / cft correspondence@xcite .
it was conjectured that a large static black hole in ads space corresponds to a thermal state in conformal field theory ( cft ) .
the fundamental link between qnms of a large ads black hole and the dual field theory were shown in the work of horowitz who computed the qnms of ads black holes@xcite .
motivated by this , a series of studies on the qnms in asymptotically ads backgrounds were done for various fields and dimensions@xcite .
see ref.@xcite for a recent review of qnms of the black hole .
the late - time relaxation of field perturbation is also an interesting topic of study since it reveals the actual physical mechanism by which a perturbed black hole sheds its hairs .
it was first demonstrated by price that in the background of the schwarzschild spacetime the massless neutral field at late time dies off as @xmath1 or @xmath2 depending on the initial conditions and the multipole order @xmath3 @xcite .
the evolution of field and the spectra of qnms may be different in various theories of gravity and would help us to distinguish these theories
. dynamical evolution of scalar field in horava black hole spacetimes is studied using horowitz - hubeny approach@xcite .
the spectrum of entropy / area is discussed from the viewpoint of qnms of scalar field in hl gravity@xcite .
qnms of various fields around ks black hole spacetime were calculated using wkb method@xcite . in earlier
work@xcite we have studied time evolution of scalar field using the time domain integration method@xcite . a noticeable deviation of the evolution behavior of scalar field in the ringdown region from the standard schwarzschild black hole was found .
however the late time evolution of field fails to show any distinction from the schwarzschild case .
the late time behavior of massless scalar field depends only on the multipole order @xmath3 and decays as @xmath4 .
where as massive field in the intermediate range the decays as @xmath5 , but in the asymptotic late time the decay is dominated by @xmath6 tail .
it is interesting to see how various other field perturbations decay in ks black hole spacetime . in this paper
we study the evolution of electromagnetic and massless dirac fields in the ks black hole spacetime and probe the signature of the now theory by comparing the present results with those of the standard theory of gr .
the paper is organized as follows . in sec.[sec2 ] we derive the radial wave equation for various field perturbations around ks black hole and discuss the numerical method used . in sec.[sec3 ]
the evolution of electromagnetic and massless dirac field are studied using the method of time domain integration and the qnms are evaluated and the results are discussed in sec.[sec4 ] .
the ir vacuum of pure hl gravity is found to be anti - de sitter@xcite .
even though hl gravity could recover gr in ir at the action level for a particular value of the parameter @xmath7 , there found a significant difference between these black hole solutions and the usual schwarzschild ads solution .
the asymptotic fall - off of the metric for these black hole solutions is much slower than that of usual schwarzschild ads black holes in gr .
meanwhile kehagias and sfetsos@xcite could find a black hole solution in asymptotically flat minkowski spacetimes by applying deformation in hl theory by adding a term proportional to the ricci scalar of three - geometry , @xmath8 while the cosmological constant @xmath9 .
this will not alter the uv properties of the theory but it does the ir ones leading to minkowski vacuum analogous to schwarzschild spacetime in gr .
ks solution is given as@xcite , @xmath10 where @xmath11 has the form , @xmath12 the event horizons are at , @xmath13 when @xmath14 the solution reduces to the schwarzschild spacetime case .
the evolution of massless scalar field @xmath15 , electromagnetic field @xmath16 and massless dirac field @xmath17 in the spacetime @xmath18 , specified by eq.([eq1 ] ) are governed by the klein - gordon@xcite , maxwells@xcite and the dirac equations@xcite respectively , @xmath19 @xmath20 @xmath21 \psi = 0 .
\label{deqn}\ ] ] the radial part of the above perturbation equations can be decoupled from their angular parts and reduced to form , @xmath22 where @xmath23 is the tortoise coordinate defined by @xmath24 and the effective potentials @xmath25 for different fields are given by , @xmath26 , \qquad \ell=0,1,2 , ... , \label{scpot}\ ] ] @xmath27 , \qquad \ell=1,2,3 , ... , \label{empot}\ ] ] @xmath28,\qquad |k|=1,2,3 , ... , \label{dpot}\ ] ] here @xmath29 is positive or negative nonzero integer related to the total orbital angular momentum by @xmath30 and @xmath31 and @xmath32 are the super symmetric partners and give same spectra .
so we choose @xmath31 by omitting the subscript .
the perturbation equations ( eq.([waveqn ] ) ) can be written in the null coordinates @xmath33 and @xmath34 as , @xmath35 in order to get the time evolution picture of the field we integrate this equation numerically using the following finite difference scheme suggested in@xcite , @xmath36 where the points
n , s , e and w form a null rectangle with an overall grid scale factor of @xmath37 having relative positions as @xmath38 , @xmath39 , @xmath40 and @xmath41 . since we are interested in the late - time behavior of the wave function which is found to be independent of the initial shape , we set @xmath42 and a gaussian profile @xmath43 $ ] . in all our calculations we set the initial gaussian with width @xmath44 centered at @xmath45 . in order to progress on the above numerical scheme one has to get the value of the potential at @xmath46 at each step .
here we use the runge - kutta method to numerically integrate the equation for the tortoise coordinate and find the values of @xmath47 at each step by cubic spline interpolation as suggested in ref.@xcite .
after the integration is completed the values of @xmath48 on line @xmath49 is extracted and plotted as a function of @xmath50 ; here , c is the value of @xmath23 at which the field is evaluated .
the evolution of scalar field around ks black hole was studied in ref .
the typical nature of time evolution of a massless scalar field is shown in fig.[sfig1 ] .
the plot shows how the generic behavior of wave function ( initial outburst , quasi - normal oscillations , and power - law decay ) in hl theory is differed from pure schwarzschild spacetime .
the qnm phase ends at a later time in hl theory and the oscillation frequency and the damping time have a higher values in hl theory .
we can calculate the qnm frequencies from the numerically integrated data by nonlinear @xmath51 fitting .
the calculated values of qnms are given in table[ta1 ] and table[ta2 ] .
( top curve ) and the schwarzschild black hole(bottom curve ) for @xmath52 . ] .qnm frequencies of massless scalar field for @xmath52 , calculated using wkb and numerical integration data [ cols= " < , < ,
< , < , < " , ] [ ta6 ] finally we compare the time evolution of different fields in hl gravity . in fig.[sed](a )
massless scalar , dirac and electromagnetic fields are plotted for @xmath53 .
the only difference between evolution of these three fields is in the qnm phase whereas the late time tails follow the same decaying pattern with same power law exponent . in fig.[sed](b )
the variation of real and imaginary part of qnm versus @xmath0 are plotted .
all the three fields show the same dependence on the hl parameter @xmath0 .
evolution of massless scalar , dirac and electromagnetic fields around ks black hole with @xmath53 and @xmath52 .
@xmath54 qnms of different field as a function of @xmath0 with @xmath52 .
, title="fig:"]evolution of massless scalar , dirac and electromagnetic fields around ks black hole with @xmath53 and @xmath52 .
@xmath54 qnms of different field as a function of @xmath0 with @xmath52 .
, title="fig : " ]
the evolution of electromagnetic and massless dirac perturbations in ks black hole in the deformed hoava - lifshitz gravity is studied in this work .
time domain integration method is used to obtain the evolution picture .
the question raised was whether we can distinguish the fundamental nature of the hl theory from a knowledge of the evolution of fields around corresponding black holes . comparing with the schwarzschild case
, we have found that the qnm phase has been extended for a longer time in hl theory before the power - law tail begins . also the oscillation frequency and the damping time
have higher values in hl theory .
the variation of the field evolution on the horava parameter @xmath0 are also studied . in the time domain picture
we find that the length of qnm region increases with @xmath0 .
the late time decay of field is found to be independent of nature of field and the hl parameter @xmath0 and follows the same power - law tail behavior as in the case of schwarzschild black hole .
these results will help us to check the correctness of the theory provided our gravitational detectors can trap these signals with high accuracy .
nv wishes to thank ugc , new delhi for financial support under rfsms scheme .
vck is thankful to csir , new delhi for financial support under emeritus scientistship scheme and wishes to acknowledge associateship of iucaa , pune , india .
the authors are grateful to dr .
alexander zhidenko for the help and clarifications received in developing the code for the numerical integration .
p. hoava , _ phys .
d _ * 79 * 084008(2009 ) .
p. hoava , _ jhep _ * 03 * 020(2009 ) .
p. hoava , _ phys .
lett . _ * 102 * 161301(2009 ) .
a padilla , _ j. phys
ser . _ * 259 * 012033(2010 ) .
t. p. sotiriou , _ j. phys
ser . _ * 283 * 012034(2011 ) .
h. lu , j mei and c. n. pope , _ phys .
lett . _ * 103 * 091301(2009 ) .
mu - in park , it jhep * 09 * 123(2009 )
. h. nastase , arxiv:0904.3604 [ hep - th ] .
r. g. cai , y. liu and y. sun , _ jhep _ * 06 * 010(2009 ) .
r. g. cai , l. cao and n. ohta , _ phys .
d _ * 80 * 024003(2009 ) .
a. ghodsi and e. hatefi , _ phys .
d _ * 81 * 044016(2010 ) .
h. w. lee y. kim and y. s. myung , arxiv:0907.3568 [ hep - th ] . j. z. tang and b. chen , _ phys .
d _ * 81 * 043515 ( 2010 ) .
e. kiritsis and g. kofinas , arxiv:0910.5487 [ hep - th ] . t. harada , u. miyamoto and n. tsukamoto , _ int . j. mod .
d _ * 20 * 111(2011 ) .
e. kiritsis , _ phys .
d _ * 81 * 044009(2010 ) .
d. capasso and a. polychronakos , arxiv:0911.1535v3 [ hep - th ] .
j. tang , arxiv:0911.3849[hp - th ] . h. w. lee , y. kim , y. s. myung , _ eur .
j. c _ * 70 * 367(2010 ) .
g. koutsoumbas and p. pasipoularides , arxiv:1006.3199v3 [ hep - th ] .
j. greenwald , a. papazoglou and a. wang , _ phys .
d _ * 81 * 084046(2010 ) .
r. g. cai and a. wang , _ phys .
b _ * 686 * 166(2010 ) .
g. koutsoumbas , e. papantonopoulos , p. pasipoularides and m. tsoukalas , arxiv:1004.2289v2 [ hep - th ] .
a. n. aliev and c. senturk , _ phys .
d. _ * 82 * 104016(2010 ) .
e. gruss , arxiv:1005.1353v1 [ hep - th ] .
r. g. cai , , l. m. cao and n. ohta , _ phys .
b _ * 679 * 504(2009 ) .
y. s. myung , arxiv:0905.0957v2 [ hep - th ] .
s. chen , j. jing , _ phys .
d _ * 80 * 024036(2009 ) .
y. s. myuang , _ phys .
d _ * 81 * 064006(2010 ) .
j. j. peng and s. q. wu , _ eur .
j. c _ * 66 * 325(2010 ) .
l. iorio and m. l. ruggiero , _ int .
a _ * 25 * 5399(2010 ) .
l. iorio and m. l. ruggiero , _ open astron .
j. _ * 3 * 167(2010 ) .
t. harko , z. kovacs and f. s. n. lobo , _ proc .
_ * a467 * 1390(2011 ) .
t. harko , z. kovacs and f. s. n. lobo , _ phys .
d _ * 80 * 044021(2009 ) .
m. r. setare and m. jamil , _ int .
phys . _ * 50 * 511(2011 ) .
w. janke , d. a. johnston and r. kenna , arxiv:1005.3392v2 [ hep - th ] .
b. gwak and b. h. lee , _ jcap _ * 09 * 031(2010 ) .
m. r. setare and d. momeni , _ int .
phys . _ * 49 * 106(2011 ) .
k. d. kokkotas and b. g. schmidt , _ living rev .
relativ . _ * 2 * 2(1999 ) .
h. p. nollert , _ class .
quantum grav . _
* 16 * r159(1999 ) b. wang , _ braz .
_ , * 35*,1029 ( 2005 )
. v. ferrari and l. gualtieri _ gen .
_ * 40 * 945(2008 ) .
e. berti , v. cardoso and c. m. will , _ phys .
* 73 * 064030(2006 ) .
j. m. maldacena , _ adv .
* 2 * 231(1998 ) .
j. m. maldacena , _ int .
phys . _ * 38 * 1113(1999 ) . g. t. horowitz and v. e. hubeny , _ phys .
d _ , * 62 * , 024027(2000 ) .
b. wang , c. y. lin , and e. abdalla , _ phys .
b _ , * 79 * , 481(2000 ) .
v. cardoso , r konoplya and p. s. lemos _ phys .
d _ * 68 * , 044024(2003 ) .
b. wang , c. molina , and e. abdalla , _ phys .
d _ * 63 * , 084001(2000 ) . v. cardoso and p. s. lemos _ phys .
* 64 * , 084017(2001 ) .
e. berti , v. cardoso and a. o. starinets , _ class .
quantum grav .
_ 26 163001(2009 ) .
r. h. price , _ phys .
d _ * 5 * , 2419(1972 ) . c. ding .
s. chen and j. jing,_phys .
d _ * 81 * 024028(2009 ) .
b. r. majhi , _ phys .
b _ * 686 * 49(2010 ) .
m. r. setare and d. momeni , arxiv:1002.0185v1 [ hep - th ] .
s. chen and j. jing , _ phys .
b _ * 687 * 124(2010 ) .
konoplya r. _
b _ * 679 * 499(2009 ) . c. wang and y. gui , _ astrophys and space sci _ * 325 * , 85(2010 )
k. lin , n. yang and j. li , _ int .
. phys _ * 50 * 48(2010 ) .
n. varghese and v c kuriakose , arxiv:1011.6608[gr - qc ] . c. gundlach , r. h. price , and j. pullin , _ phys .
d _ * 49 * , 883 ( 1994 ) .
r. a. konoplya and a. zhidenko , _ phys .
d _ * 77 * 104004(2008 ) .
r. h. price , _ phys .
d _ * 5 * , 2439(1972 ) . | the evolution of electromagnetic and massless dirac perturbations in the space - time geometry of kehagias - sfetsos(ks ) black hole in the deformed hoava - lifshitz(hl ) gravity is investigated and the associated quasinormal modes(qnms ) are evaluated from time domain integration data .
we find a considerable deviation in the nature of field evolution in hl theory from that in the schwarzschild space - time and qnm region extends over a longer time in hl theory before the power - law tail decay begins .
the dependence of the field evolution on the hl parameter @xmath0 is also studied . in the time domain picture
we find that the length of qnm region increases with @xmath0 .
but the late time decay of field follows the same power - law tail behavior as in the case of schwarzschild black hole . |
recently there has been a great deal of interest in studying a novel class of media that has become known as the left - handed materials ( lhms)[1 - 20 ] .
these materials are characterized by simultaneous negative permittivity @xmath0 and negative permeability @xmath1 .
properties of such materials were analyzed theoretically by veselago over 30 years ago[1 ] . as was shown by veselago
, the lhms possess some peculiar electromagnetic properties such as inverse snell s law , reversed doppler shift , and reversed cherenkov radiation .
it had also been suggested that slab of the lhm could be employed as an unconventional flat lens . due to the absence of naturally occurring materials having both negative @xmath0 and negative @xmath1 ,
veselago s prediction did not receive much attention until recently , when a system consisting an array of split - ring resonators and metallic wires was prepared and demonstrated to have negative refractive index experimentally [ 3,4 ] .
subsequently , some physical properties of the lhms were analyzed by many authors [ 5 - 20 ] .
pendry [ 5 ] predicted that the lhm slab can amplify the evanescent waves and the flat lens constructed from such a material with @xmath2 could in principle work as perfect " lens ( superlens ) .
although the concept of superlens was questioned by a number of authors [ 21 ] , more detailed physical considerations had shown that the construction of almost perfect " lens is indeed possible [ 22 - 27 ] . recently , such image behaviors have been observed by some numerical simulations [ 22 - 24 ] and experimental measurements [ 24 - 27 ] .
however , only near - field images were demonstrated and extensive applications of such a phenomenon were limited [ 22 - 27 ] . it was shown that the negative refraction could also occur in photonic crystal ( pc)[28 - 39 ] .
the physical principles that allow negative refraction in them arise from the dispersion characteristics of wave propagation in a periodic medium , which can be well described by analyzing the equifrequency surface ( efs ) of the band structures [ 28 - 37 ] . in the pc structures ,
there are two kinds of cases for negative refraction occuring [ 31 ] .
the first is the left - handed behavior as being described above[29 - 32 ] .
in this case , @xmath3 , @xmath4 and @xmath5 form a left - handed set of vectors ( i.e. , @xmath6 , where @xmath7 is the poynting vectors ) .
another case is that the negative refraction can be realized without employing a negative index or a backward wave effect[33 - 37 ] . in this case , the pc is behaving much like a uniform right - handed medium(i.e .
@xmath8 ) . recently , luo @xmath9 @xmath10[33 ] have shown that all - angle negative refraction could be achieved at the lowest band of two - dimensional(2d ) pc in the case of @xmath8 .
the advantages of the negative refraction in the lowest valence band are single - mode and high transmission .
these can help us to design microsuperlens and realize the focusing of the wave .
very recently , the subwavelength focusing and image by 2d pc slab have been observed experimentally [ 34 - 36 ] .
absolute negative refraction and imaging of unpolarized electromagnetic wave by 2d pc slabs have also been obtained [ 37 ] . however , due to the anisotropy of dispersion in 2d pc , such images only appear in near - field region[38,39 ] . how to realize a good - quality non - near - field image becomes an important issue .
the prerequisite condition to realize such a phenomenon is the negative refraction which possesses the single - mode and high transmission . although some works[29 ] have shown that the left - hand behavior and focusing exist in the 2d pc ,
the multiple - mode and low transmission in high frequencies affect the features of focusing[31,32 ] .
it is well known that the electromagnetic ( em ) wave can be decomposed into tm modes ( s wave ) and te modes ( p wave ) for the 2d pc structures [ 40 ] .
however , the investigations[29 ] have shown that air - hole - type 2d pcs possess good left - handed behavior for tm modes , and a pillar type 2d pc prefers te modes . in this paper , we will demonstrate that coated cylinder pcs with triangular lattice posses double features .
they have not only good left - handed behavior for the tm modes , but also for the te modes .
most interestingly , the all - angle single - beam left - handed behaviors with relative refractive index of -1 for both polarized waves have been found in these systems .
thus , high - quality focusing and imaging behavior have been obtained .
in contrast to the images in the near - field region for the lowest valence band , non - near - field images , explicitly following the well - known wave - beam negative refraction law , have been demonstrated . the absorption and compensation for the losses by introducing optical gain in these systems
have also been discussed .
the rest of this paper is arranged as follows . in sec .
ii , we demonstrate the all - angle single - beam left - handed behaviors for both te and tm modes in a two - dimensional coated photonic crystals . the good - quality non - near - field imaging behaviors are discussed in sec .
iii . in sec .
iv , we analyze the effect of absorption and gain .
the conclusions are given in sec .
we consider a 2d triangular lattice of coated cylinders immersed in a air background with lattice constant @xmath11 .
the coated cylinders have metallic cores coated with a dielectric coating .
the radii of metallic core and coated cylinder are 0.25a and 0.45a , respectively .
the dielectric constants of dielectric coating are taken as @xmath12 for the s wave and @xmath13 for the p wave .
for the metallic component , we use the frequency - dependent dielectric constant [ 41 ] , @xmath14 where @xmath15 and @xmath16 are the plasma frequency and the absorption coefficient . following ref.41 , for all numerical calculations carried out in this work , we have chosen @xmath17thz and @xmath18thz , which corresponds to a conductivity close to that of ti .
however , our discussion and conclusions given below can apply to other metal parameters as well . in order to simplify the problem
, we first consider the cases without absorption ( @xmath19 ) .
the effect of absorption will be discussed at the latter part . for all calculations throughout this paper ,
we adopt the multiple - scattering korringa - kohn - rostoker method [ 42 ] as our main computational tool , both to calculate the photonic band structure in the reciprocal space and to perform numerical simulations for wave propagating in the finite real space .
the multiple - scattering method is not only success in the calculations of band structure , it is also best suited for a finite collection of cylinders with a continuous incident wave of fixed frequency . for circular cylinders ,
the scattering property of the individual cylinder can be obtained analytically , relating the scattered fields to the incident fields
. the total field , which includes the incident plus the multiple - scattered field , can then be obtained by solving a linear system of equations , whose size is proportional to the number of cylinders in the system .
both near field and far field radiation patterns can be obtained straightforwardly .
so , such a method is a very efficient way of handling the scattering problem of a finite sample containing cylinders of circular cross sections , and it is capable of reproducing accurately the experimental transmission data , which should be regarded as exact numerical simulation .
the detailed description of this method has been given in ref .
. the calculated results of band structure for the s wave and the p wave are plotted in the fig.1(a ) and ( b ) , respectively .
we focus on the problems of wave propagation in the second bands marked by dotted lines in fig.1(a ) and ( b ) . owing to the strong scattering effects ,
it is generally difficult to describe the propagation behavior of em wave in the pc in a simple yet accurate way . however , a lot of theoretical and experimental practices[28 - 39 ] have shown that the overall behavior of the wave propagation within a pc can be well described by analyzing the equifrequency surface ( efs ) of the band structures , because the gradient vectors of constant - frequency contours in k - space give the group velocities of the photonic modes . thus , the propagation direction of energy velocity of em wave can be deduced from them .
the efs contours of the above system for the s wave and the p wave at several relevant frequencies are demonstrated in fig.2(a ) and ( b ) , respectively .
it is clear from the figures that some efs contours such as @xmath20 for the s wave and the p wave are very close to a perfect circle , indicating that the crystal can be regarded as an effective homogeneous medium at these frequencies . at the same time
, we also notice that the frequencies increase inwards for both cases , meaning that @xmath21 and the group velocities ( @xmath22 ) are opposite to the phase velocity . here ,
@xmath7 and @xmath23 represent the poynting vector and wave vector , respectively .
these indicate that the transmitting features of the wave in the above pc structures are the left - handed behavior .
the conservation of the surface - parallel wave vector would result in the negative refraction effect in these cases , which the direction of the refracted wave inside the pc can also estimated from the efs .
then , we apply snell s law in these cases , the negative indexes of refraction can also be obtained .
fig.3(a ) and ( b ) show the negative effective indexes as function of frequencies for the s wave and the p wave , respectively .
it is interesting that the cases with relative refractive index of -1 for the s wave at @xmath24 and the p wave at @xmath25 have been found . in order to test the above analysis , we do numerical simulations in the present systems .
we take the slab samples which consist of 13-layer coated cylinders in the air background with a triangular arrays .
the surface normal of the pc slab is along @xmath26 direction .
the parameters of coated cylinders are the same to the cases in fig.1 .
when a slit beam with a half width @xmath27 goes through the slab material , it will be refracted two times by two interfaces of the slab .
the shape of the sample and a snapshot of the refracted process are shown on the top of fig.4 .
there are different ray - traces for the wave transmitting through the slab sample with various effective indexes , when the wave does not incident on the interface with normal direction . from the ray - traces
, we can deduce the effective refraction index of the slab material .
the simulations are based on a highly efficient and accurate multiple - scattering method [ 42 ] . in our calculations , the widths of the samples are taken enough large , such as 40a , to avoid the edge diffraction effects .
the calculated results for the s wave and the p wave are plotted in fig.4(a ) and ( b ) , respectively .
the field energy patterns of incidence and refraction are shown in the figures .
the arrows and texts illustrate the various beam directions .
the geometries of the slab are also displayed .
it can be clearly seen that the energy fluxes of refraction wave travel following the negative refraction law for both polarized waves .
the negative refraction indexes for the s wave and the p wave obtained from the exact numerical simulation are marked as dark dot in fig.3(a ) and ( b ) , respectively . comparing them with the estimated results ( solid lines in fig.3(a ) and ( b ) ) from the efs , we find that the agreements between them are well around the region of effective refractive index of -1 for both polarized waves .
when the effective refractive indexes deviate largely from -1 , some differences can be found .
this is due to the goos - hanchen effect , which had been discussed in ref.[43 ] .
varying the angle of inclination of the sample , we have checked the cases with various incident angles .
the calculated and analytical results of refracted angle @xmath28 versus incident angle @xmath29 at @xmath24 for the s wave and @xmath25 for the p wave are summarized in fig.5 by the circle and triangular dark dots , respectively . because the frequencies are below 0.5@xmath30 , all - angle single - beam negative refractions
have been observed at these frequencies .
more interestingly , @xmath28 is linearly proportional to @xmath29 in the whole angle region for both polarized waves .
this feature is close to the ideal lhm system that can serve as a perfect superlens [ 5 ] .
it is well known that an important application of negative refraction materials is the microsuperlens [ 5 ] .
ideally , such a superlens can focus a point source on one side of the lens into a real point image on the other side even for the case of a parallel sided slab of material .
it possesses some advantages over conventional lenses . for example
, it can break through the traditional limitation on lens performance and focus light on to an area smaller than a square wavelength . in order to model such a superlens
, we take a slab sample with 40a width and 11a thickness .
a continuous - wave point source is placed at a distance 5.5a ( half thickness of the sample ) from the left surface of the slab .
we first discuss the case for the point source of the s wave .
the frequency of the incident wave emitting from such a point source is @xmath31 , which is corresponded to the case with relative refractive index of -1 .
if the wave transmits in such a 2d pc slab according to the well - known wave - beam refraction law , one should observe the focusing point in the middle of the slab and the image at the symmetric position in the opposite side of the slab , as being depicted by the simple picture on the top of fig.6 . to see whether or not such a phenomenon exists , we employ the multiple - scattering method [ 42 ] to calculate the propagation of waves in such a system . a typical field intensity pattern for the s wave across the above slab sample is plotted in fig . 6(b ) . x and y present vertical and transverse direction of wave propagating , respectively .
the field intensity in figure is over @xmath32 region around the center of the sample .
the geometry of the pc slab is also displayed for clarity of view .
the high quality image in the opposite side of the slab and the focusing in the middle of the slab according to the wave - beam refraction law are observed clearly
. a closer look at the data reveals a transverse size ( full size at half maximum ) of the image spot as @xmath33 ( or @xmath34 ) , which is well below the conventional diffraction limit . in order to clarify the sample thickness dependence of the image and focusing , we have also checked a series of slab samples with various thickness .
similar phenomena have also been observed .
for example , fig .
6(a ) shows the calculated field energy pattern for a 7a thick sample . a monochromatic point source with @xmath24
is placed at a distance of half thickness of the sample ( 3.5a ) from the left surface of the slab and its image is found again near the symmetric position in the opposite side of the slab . to have a more complete vision on the imaging effect of this type of superlens
, we move the light source and see what happens to the imaging behavior .
we first put a point source near the left surface of the slab . in this case , the refracted process following wave - beam negative refraction law is shown on the top of fig.7 .
the calculated intensity distribution is plotted in fig.7 . in our simulations ,
the point source is placed at a distance of 2.0a from the left surface of the sample and the image is found near 9a from the right surface .
the excellent agreements between the simulations and the estimated results from the rules of geometric optics are obvious .
the corresponded result that a point source is far from the left surface is displayed in fig.8 . in this case , the point source is placed at a distance of 9.0a from the left surface of the sample and the image is found near 2a from the right surface . comparing the simulations with the snapshot of refracted process according to the wave - beam negative refraction law shown on the top of fig.8 , we again find the excellent agreements between them .
the above results are only for the s wave , in the following , we will investigate the case of the p wave .
we take the pc slab samples with 40a width and various thicknesses .
a point source of the p wave is placed at a distance of half thickness of the sample from the left surface of the slab .
the frequency of the incident wave emitting from such a point source is @xmath35 , which is corresponded to the case with relative refractive index of -1 for the p wave .
the propagation behaviors of the p waves in such systems are still calculated by the multiple - scattering method [ 42 ] .
fig.9(a ) and ( b ) show the cases with 7a and 13a thick slab , respectively .
similar features to fig.7 for the s wave are found .
if we move the source position , the effect of source position on the image for the p wave can also be checked .
the calculated results under two kinds of source position for the p wave are plotted in fig.(10)(a ) and ( b ) , respectively .
the cases for the p waves are similar to those of the s waves in fig.8 .
these observations indicate clearly that the imaging behaviors depend on the slab thickness and the object distance , explicitly following the well - known wave - beam negative refraction law .
therefore , such pc slabs in some frequencies can be considered as homogeneous effective medium with effective refraction index of -1 .
the high - quality focusing and images can be realized in these pc systems for both s wave and p wave .
the above investigations have shown that the pc slab consisting of coated cylinders is actually considered as a good superlens for both polarized waves .
the common features of these coated systems are that they all include metal components .
therefore , the absorption for these systems is inevitable .
lou @xmath9 @xmath10 [ 38 ] have pointed that the central image peak disappear and the image degrade gradually with the increase of absorption .
however , fortunately , the loss can be overcome by introducing the optical gain in the systems .
recently , ramakrishna and pendry [ 19 ] have suggested a method to remove the absorption by introducing optical gain into the lens made from a multilayers stack of thin alternating layers of silver and dielectric medium .
here , we borrow their idea and introduce the optical gain in the 2d pc superlens .
fig.11(a ) shows the intensity distribution as a function of transverse coordinate ( @xmath36 ) for the s wave at the image plane ( 5.5a away from the second interface ) . curve _ a _ is corresponded to the perfect case without absorption and curve _ b _ to that with absorption , in this case @xmath16 is taken as @xmath37thz . comparing the curve _ a _ with the curve _ b _ , we find that the central peak of image decrease with the introducing of absorption , which is agree with the analysis of ref.[38 ] .
this is also consistent with the numerical studies of left - handed structures constructed from split - ring resonators in ref.[7,8 ] .
the result by introducing gain to remove the absorption for the corresponding case is plotted in fig.11(b )
b _ in the fig.11(a ) and ( b ) are the same one , and curve _ c _ in the fig.11(b ) is the result with the dielectric constant @xmath38 for the dielectric part of coated cylinder .
we do not find any difference between curve _ a _ in the fig.11(a ) and curve _ c _ in the fig.11(b ) .
similar phenomenon can also be found for the p wave .
fig.12(a ) and ( b ) show the corresponded case of the p wave .
fig.12(a ) represents the intensity distribution as a function of transverse coordinate ( @xmath36 ) for the p wave at the image plane ( 5.5a away from the second interface ) . comparing the curve _ a _ without absorption with the curve _ b _ in presence of absorption in fig.12(a ) ,
we find that the absorption decreases the central peak of the image as the case of the s wave . however , with the introducing of gain such as @xmath39 for the dielectric part of coated cylinder , the lose due to absorption can be compensated completely .
curve c in fig.12(b ) represents such a case .
in fact , for any cases of absorption , the losses can always be compensated by introducing fitted gain for both polarized waves .
thus , the lens based on the above 2d pc can work well even in presence of absorption .
through the exact numerical simulation and physical analysis , we have demonstrated all - angle single - beam left - handed behavior for both te and tm modes in the 2d coated photonic crystals .
more interestingly , the relative refractive index of -1 for both polarized waves have been found .
furthermore , the refracted angle is linearly proportional to the incident angle in the whole angle region for both polarized waves have been demonstrated .
these feature are close to the ideal lhm system that can serve as a perfect superlens .
the imaging behaviors by 2d coated photonic - crystal - based superlens have been investigated systematically .
good - quality images and focusing , with relative refractive index of -1 and explicitly following the well - known wave - beam negative refraction law , have been observed in these systems for both polarized waves .
our above results are in contrast to the previous investigations about the image behaviors for the lhm - based superlenes and the pc - based microsuperlenes at the lowest valence band . for these cases ,
the images only appear in the near - field region , which does not follow rules of geometric optics [ 22 - 27,38,39 ] . according to pendry s analysis [ 5 ]
, the perfect " image arises from the enhancement of evanescent components of incoming waves , and surface plasmons play an important role in the perfect " imaging [ 5,21 ] .
so , the quality of the image is affected by many factors such as interface feature , finite - size and cavity resonance [ 22 - 27 ] .
these will limit further applications . here
, our superlens based on the coated pc systems with triangular lattice can overcome these restrictions .
thus , extensive applications of such a phenomenon to optical devices are anticipated .
in addition , the absorption and compensation for the losses by introducing optical gain in these systems have also been discussed . in general , for the pc structures with metal components , increased absorption in metals prohibits the scaling of these structures to the optical wavelengths . however , since the losses by absorption can always be compensated by introducing fitted gain in our coated pc systems , many negative refraction phenomena that have been observed in the microwave regime can also be found in the optical wavelengths .
these features make the pc slabs consisting of coated cylinders promising for application in a range of optical devices , such as a superlens for visible light .
this work was supported by the national natural science foundation of china ( grant no.10374009 ) and the national key basic research special foundation of china under grant no.2001cb610402 . the project sponsored by srf for rocs , sem and the grant from beijing normal university .
v.g.veselago , sov.phys.uspekhi * 8 * , 2854 ( 1967)[sov .
10 , 509(1968 ) ] . j. b. pendry , a.j.holden , d.j.robbins , and w.j.stewart , ieee trans . microwave theory tech . * 47 * , 2075 ( 1999 ) .
d.r.smith , w.j.padilla , d.c.view , s.c.nemat-nasser , and s.schultz , phys .
* 84 * , 4184 ( 2000 ) ; d.r.smith and n. kroll , phys .
* 84 * , 2933 ( 2000 ) .
r. a. shelby , d.r.smith , and s.schultz , science * 292 * , 77 ( 2001 ) .
j. b. pendry , phys .
85 * , 3966 ( 2000 ) .
p.markos and c.m.soukoulis , phys .
e * 65 * , 036622(2002 ) ; phys .
b * 65 * , 033401(2002 ) .
markos , i. rousochatzakis and c. m. soukoulis , phys .
e 66 045601(r ) ( 2002 ) .
greegor , c.g.parazzoli , k.li and m.h.tanielian , appl .
82 , 2356(2003 ) .
s. foteinopoulou , e.n .
economou and c.m.soukoulis , phys .
* 90 * , 107402(2003 ) .
j. pacheco , jr .
, t.m.grzegorczyk , t b.i.wu , y.zhang and j.a.kong , phys . rev .
lett . * 89 * , 257401(2002 ) .
y. zhang , b. fluegel , and a. mascarenhas , phys .
lett . * 91 * , 157404(2003 ) .
d.r.smith and d.schurig , phys .
lett . * 90 * , 077405 ( 2003 ) .
a.a.houck , j. b. brock and i. l. chuang , phys .
lett . * 90 * , 137401(2003 ) ; c.g.parazzoli , r. b. greegor , k. li , b. e. c. koltenbah , and m. tanielian , phys .
* 90 * , 107401 ( 2003 ) .
j. li , lei zhou , c.t.chan and p.sheng , phys .
* 90 * , 083901(2003 ) .
r. ziokowski and e. heyman , phys .
e * 64 * , 056625(2001 ) .
focus issue negative refraction and metamaterials " , optics express * vol .
11*,no.7 ( 2003 ) .
g. shvets , phys .
b * 67*,035109(2003 ) . v. a. podolskiy , a.k.sarychev and v.m .
shalaev , optics express * 11*,no.735 ( 2003 )
. s. a. ramakrishna and j.b .
pendry , phys .
b * 67 * , 201101(r)(2003 ) .
a.a.zharov , i.v.shadrivov and y. s. kivshar , phys .
* 91 * , 037401(2003 ) ; v. m. agranovich , y.r.shen , r.h.baughman and a.a .
zakhidov , phys .
b * 69 * , 165112 ( 2004 ) .
g.w.t hooft , phys .
lett . * 87 * , 249701(2001 ) ; j. m. williams , phys .
* 87 * , 249703(2001 ) ; n.garcia and m.nieto-vesperinas , phys .
88 * , 207403(2002 ) ; a.l.pokrovsky and a. l. efros , phys .
lett . * 89 * , 093901 ( 2002 ) .
r.merlin , appl.phys.lett .
* 84 * , 1290(2004 ) .
l. chen , s. he and l. shen , phys .
* 92 * , 107404 ( 2004 ) .
smith , d. schurig , j. j. mock , p. kolinko , and p. rye , appl .
84 , 2244 ( 2004 ) .
zhaowei liu , nicholas fang , ta - jen yen , and xiang zhang , appl.phys.lett .
* 83 * , 5184(2003 ) .
a. grbic and g.v.eleftheriades , phys .
* 92 * , 117403 ( 2004 ) . a.n.lagarkov and v.n.kissel , phys .
* 92 * , 07701 ( 2004 ) .
h. kosaka , t. kawashima , a. tomita , m.notomi , t.tamamura , t.sato , and s.kawakami , phys . rev .
b * 58 * , 10096(1998 ) .
m.notomi , phys . rev .
b * 62 * , 10696(2000 ) .
b.gralak , s.enoch , and g.tayeb , j. opt .
a * 17 * , 1012(2000 ) .
s. foteinopoulou and c.m.soukoulis , phys .
b * 67 * , 235107(2003 ) .
p.v.parimi , w.t.lu , p.vodo , j.sokoloff , j.s .
derov and s. sridhar , phys .
lett . * 92 * , 127401 ( 2004 ) .
c.luo , s.g.johnson , j.d.joannopoulos , j.b.pendry , phys .
b * 65 * , 201104(r)(2002 ) ; optics express , * 11 * , 746(2003 ) ; c.luo , s.g.johnson , j.d.joannopoulos , appl.phys.lett . *
83 * , 2352(2002 ) .
e. cubukcu , k. aydin , e. ozbay , s. foteinopoulou and c.m.soukoulis , nature ( london ) * 423 * , 604(2003 ) .
p. v. parimi , w.t.lu , p. vodo and s. sridhar , nature * 426 * , 404(2003 ) .
e. cubukcu , k. aydin , e. ozbay , s. foteinopoulou and c.m.soukoulis , phys .
* 91 * , 207401(2003 ) .
xiangdong zhang , phys .
b ( accepted ) .
c.luo , s.g.johnson , j.d.joannopoulos , j.b.pendry , phys .
b * 68 * , 045115(2003 ) .
z.y . li and l.l .
lin , phys .
b * 68*,245110(2003 ) .
j.d.joannopolous , r.d.meade , and j.n.winn , photonic crystals ( princeton university , princeton , 1995 ) .
m. m. sigalas , c. t. chan , k. m. ho and c.m .
soukoulis , phys .
b * 52 * , 11744 ( 1995 ) ; l.m .
zhang and x. zhang , phys .
b * 58 * , 15589 ( 1998 ) . l.m.li and z.q.zhang,phys .
b * 58 * , 9587 ( 1998 ) ; x.zhang , z.q.zhang , and l.m.li , c. jin , d. zhang , b. man and b. cheng , phys .
b * 61 * , 1892 ( 2000 ) .
d. felbacq and r. smaali , phys .
rev . lett .
* 92 * , 193902(2004 ) .
fig.1 , the calculated photonic band structures of a triangular lattice of coated cylinder in air for s wave ( a ) and p wave ( b ) .
the radii of the dielectric cylinder and inner metallic cylinder are @xmath40 and @xmath41 , respectively .
the dielectric constants are @xmath42 for the s wave and @xmath43 for the p wave .
dotted lines mark the region for negative refraction .
fig.2,several constant - frequency contours for s wave ( a ) and p wave ( b ) of the second band of the 2d pcs which are corresponded to the cases in fig.1(a ) and ( b ) , respectively . the numbers in the figure mark the frequencies in unit of @xmath44 .
@xmath45 and @xmath22 represent the wave vector and the group velocity , respectively .
fig.3 , effective indexes versus frequencies for s wave ( a ) and p wave ( b ) .
the crystals and parameters are identical to those in fig.1 .
the positions of relative refractive index of -1 are marked by dotted lines .
fig.4 , simulation of negative refraction .
the shape of the sample and a snapshot of refraction process are shown on top of the figure .
the intensities of electric field for s wave ( a ) and magnetic field for p wave ( b ) for incidence and refraction are shown .
the 2d pc slabs with 13-layer are marked as dark dots in figures .
the frequencies of incident wave are @xmath46 for the s wave and @xmath25 for the p wave .
the crystals and parameters in ( a ) and ( b ) are corresponded to those in fig.1(a ) and ( b ) , respectively .
fig.5 , the angles of refraction ( @xmath28 ) versus angles of incidence ( @xmath29 ) at @xmath24 for s wave and @xmath25 for p wave .
circle dots are corresponded to the s wave and triangular dots to the p wave .
fig.6 , ( a ) the intensity distributions of point source and its image across a 7a 2d pc slab at frequency @xmath24 for s wave .
( b ) the corresponding case for a slab with 11a thickness .
schematic picture depicting the lensing of a source by a pc slab to an image are shown on top of the figure .
fig.7 , the intensity distributions of point source and its image across a 11a 2d pc slab at frequency @xmath24 for s wave .
the point source is placed at 2a distance from the left surface of the slab .
schematic picture depicting the lensing of a point source by a pc slab to an image are shown on top of the figure .
fig.8 , the intensity distributions of point source and its image across a 11a 2d pc slab at frequency @xmath24 for s wave .
the point source is placed at 9a distance from the left surface of the slab .
schematic picture depicting the lensing of a point source by a pc slab to an image are shown on top of the figure .
fig.9 , ( a ) the intensity distribution of point source and its image across a 7a 2d pc slab at frequency @xmath25 for p wave .
( b ) the corresponding case for a slab with 11a thickness .
the point source is placed at a distance with half thickness of the sample from the left surface of the slab .
fig.10 , the intensity distribution of point source and its image across a 11a 2d pc slab at frequency @xmath25 for p wave .
( a ) and ( b ) represent the cases with different source positions , 2a and 9a distances from the left surface of the slab , respectively .
fig.11 , intensity distribution along the transverse ( y ) direction at the image plane for s wave .
( a ) the case with absorption ( b ) and that without absorption ( a ) .
( b ) the case with absorption ( b ) and that with absorption and gain ( c ) .
the crystal and parameters are identical to those in fig.6 .
fig.12 , intensity distribution along the transverse ( y ) direction at the image plane for p wave . ( a ) the case with absorption ( b ) and that without absorption ( a ) .
( b ) the case with absorption ( b ) and that with absorption and gain ( c ) .
the crystal and parameters are identical to those in fig.9 . | based on the exact numerical simulation and physical analysis , we have demonstrated all - angle single - beam left - handed behavior and superlens for both te and tm modes in a two - dimensional coated photonic crystals .
the imaging behaviors by two - dimensional photonic - crystal - based superlens have been investigated systematically .
good - quality images and focusing , with relative refractive index of -1 , have been observed in these systems for both polarized waves .
in contrast to the images in near - field region for the lowest valence band , non - near - field images , explicitly following the well - known wave - beam negative refraction law , have been demonstrated .
the absorption and compensation for the losses by introducing optical gain in these systems have also been discussed . thus , extensive applications of such a phenomenon to optical devices are anticipated . |
the presence of disorder and impurities in strongly correlated systems offers a good opportunity to better understand the role played by quantum fluctuations in such materials @xcite .
either intrinsically present or explicitly added by doping , impurities in condensed matter systems can rarely be ignored , in particular when they induce new physics as compared to the disorder - free situation .
prominent examples are the kondo effect @xcite , anderson localization @xcite , dirty bosons physics in disordered superconductors @xcite , impurities in magnetic semiconductors @xcite , spin glasses @xcite ... in the context of antiferromagnetic ( af ) mott insulators , parent compounds of high temperature superconducting cuprates for instance , spin ladder materials @xcite have been shown to display very interesting features , in particular when the number of legs is an even number .
for example , a finite energy gap @xmath3 appears in the excitation spectrum of two - leg af spin-1/2 ladders @xcite , as seen for instance in srcu@xmath1o@xmath4 @xcite .
furthermore , defects in gapped ladders induce very interesting effects @xcite , in particular the apparition of effective gapless modes below the bare spin gap @xmath3 . having in mind that the ground - state of a two - leg ladder displays short - range resonating valence bond like physics @xcite
, a non - magnetic dopant is expected to break such a short - distance singlet , inducing a quasi - free spin-@xmath5 , strongly localized in the vicinity of the impurity .
interesting questions arise then when a finite concentration of impurities is introduced in a spin ladder , as studied in a large number of theoretical works @xcite .
similar physics is also at play in other spin - gapped materials : spin-@xmath6 ( haldane ) chains such as y@xmath1banio@xmath7 @xcite , or pbni1@xmath1v@xmath1o@xmath8
@xcite , spin - peierls chains such as cugeo@xmath4 @xcite for instance .
indeed , a universal behavior can be observed for the impurity - induced three - dimensional ordering mechanism in such weakly coupled chains or ladder materials @xcite . nevertheless , several aspects of impurity effects in ladder and more generally spin gapped
materials remain to be explored in order to better understand and interpret experimental results . regarding the effective pair - wise interaction between released moments ,
it is believed to remain non - frustrated even when the underlying spin systems is frustrated @xcite , but it is not clear to which extend such a result is robust when strong frustration leads to incommensurability @xcite , as expected for instance in the ladder material bicu@xmath1po@xmath2 @xcite . a natural question then arises regarding which effective model is able to quantitatively describe the low energy physics of randomly doped ladders .
indeed , it was believed since the seminal work of sigrist and furusaki @xcite that a simple model of random ( in sign and magnitude , reflecting the random locations of impurities in a ladder ) nearest - neighbors couplings between effective spin-1/2 ( describing impurity degrees of freedom ) was able to correctly capture the low temperature physics of depleted ladders .
this so - called random f - af chain model @xcite displays some universal behavior for various quantities such as uniform and staggered susceptibilities or the specific heat in the low temperature regime , with an interesting large spin phase occurring at very low temperature .
however , in the context of depleted ladders , universality for such thermodynamic quantities has been first questioned using quantum monte carlo ( qmc ) simulations by miyazaki and co - workers @xcite where no clear signature of universal low temperature scalings were found , in agreement with a more recent qmc study @xcite . despite the large number of works devoted to such systems in the absence of external magnetic field ,
much less is known regarding finite field effects .
indeed , as recently reviewed by giamarchi and co - workers @xcite , applying a finite external field on gapped af systems leads to the analog of a bose - einstein condensation ( bec ) of magnetic excitations @xcite ( hard - core bosons triplets ) when the field is sufficiently strong to close the spin gap @xmath3 .
note that a true bec is only expected for dimension @xmath9 , which occurs at low enough temperature below some energy scale controlled by three dimensional couplings .
nevertheless for low-@xmath10 a quasi - bec is expected , as observed in ultra - cold atom physics @xcite . in solid state physics
, triplet bec has been observed in several quantum magnetic compounds , such as coupled dimers tlcucl@xmath4
@xcite , frustrated bilayers bacusi@xmath1o@xmath2 @xcite , coupled haldane chains in dtn @xcite , and also spin ladder materials like ( c@xmath7h@xmath11n)@xmath1cubr@xmath12 @xcite .
however , when disorder is present in such spin gapped systems , a new phenomenology is expected with the interesting possibility to address bose - glass ( bg ) physics , as recently found in br - doped ipacucl@xmath4
@xcite or dtn @xcite ( see also ref . for a very recent review ) . while several issues remain unsolved regarding bg physics , _
e.g. _ for the excitation spectrum @xcite , the case where disorder comes from ligand substitution seems easier to understand from a microscopic point of view .
indeed , such doping will essentially generate disorder in the af couplings without inducing local moments . on the other hand , doping on the magnetic sites
is expected to be more complicated as gapless states will populate the clean gap .
therefore the magnetic response will display non - trivial brillouin - like behaviors in most of the experimentally relevant situations .
such cases have been studied theoretically by a few authors @xcite , showing a rich physics and various scenarios that demand further analysis . in this work ,
we focus on the two - leg ladder model to provide a systematic analysis of the physics of interacting impurities , building on both analytical and numerical arguments .
in particular , we are interesting in the following issues : ( i ) the effective interaction between impurities for commensurate and incommensurate backgrounds ; ( ii ) the low energy emergence of large spins due to random signs in effective couplings in a realistic context including finite size effects due to chain breaking ; ( iii ) the temperature scaling of the curie constant of the uniform susceptibility , as obtained from both effective and realistic doped ladder models ; ( iv ) the deviations of the magnetic curve from the brillouin response as a probe of the magnitude of interactions .
the ladder model used throughout this study is the one studied in refs .
@xmath13\\ \nonumber + & j_{2 } \left[\mathbf{s}_{i,1}\cdot\mathbf{s}_{i+2,1 } + \mathbf{s}_{i,2}\cdot\mathbf{s}_{i+2,2}\right]\\ \label{eq : microh } + & j_{\perp } \,\mathbf{s}_{i,1}\cdot\mathbf{s}_{i,2 } \;,\end{aligned}\ ] ] where @xmath14 is the spin-@xmath0 operator acting at site @xmath15 of leg @xmath16 and the @xmath17s are the magnitude of the various couplings which are here taken to be antiferromagnetic ( @xmath18 ) . in the rest of the paper , the only parameter coming with the presence of impurities is their concentration @xmath19 .
the doped microscopic model is numerically solved with two state - of - the - art methods : the density - matrix renormalization group ( dmrg ) technique @xcite and the stochastic series expansion ( sse ) quantum monte - carlo ( qmc ) technique @xcite .
the paper is organized as follow : in a first part , we discuss in details the effective model describing two - body interactions between impurities .
the resulting effective model is then compared to the solution of the microscopic model in the second part .
the latter is dedicated to the study of the magnetization curve at field below the spin gap @xmath3 , i.e. in the region dominated by the impurity spins response .
this region is itself divide in two regimes : ( i ) the small field regime @xmath20 , featuring a temperature - dependent curie constant @xmath21 , ( ii ) the intermediate field regime @xmath22 displaying again deviations from brillouin through an approximate power - law behavior .
we do not investigate fields @xmath23 as the physics involves triplet bosons in a disordered medium which is exciting but beyond the scope of the present manuscript .
in this first section , the emphasis is put on the quantitative analysis of the effective interaction between impurities from arguments similar to rkky theory .
this provides an effective hamiltonian which couplings distribution is essential for understanding the magnetic responses .
previous works along this direction are found in refs . .
we start with the derivation of the low - energy effective hamiltonian accounting for effective interactions between impurities . for a generic heisenberg spin model with @xmath24 spins ,
the clean hamiltonian takes the general form @xmath25 where @xmath26 are the microscopic couplings , which depend only on the relative distance @xmath27 .
we now consider that a few non - magnetic impurities occupy sites @xmath28 of the lattice .
then , the hamiltonian reads @xmath29 or @xmath30 , where @xmath31 notice that effective spins operators are introduced at sites @xmath28 where the impurities live , while these sites are actually vacant .
hamiltonian takes the form @xmath32 , in which @xmath33 is an effective magnetic field operator . assuming the perturbation @xmath34 can be treated using linear response theory
, we may write the fourier transform of @xmath35 : @xmath36 with @xmath37 is the static susceptibility at wave - vector @xmath38 , and @xmath39 one can thus write the perturbation @xmath34 as @xmath40 in which @xmath41 is the effective two - body interaction between impurities .
when the clean system possesses a spin gap associated to a spin correlation length @xmath42 , the susceptibility @xmath43 , and therefore the effective interaction @xmath44 , decreases exponentially with the distance @xmath45 . for a sufficiently small impurity concentration @xmath19 ( @xmath46 in one dimension ) , effective interactions remain much smaller than the spin gap . at temperatures smaller than this gap
, the clean part of the doped system can be considered to be in the ground - state of @xmath47 while the impurities dynamics is governed by , in which one can take the zero - temperature behavior for the susceptibility @xmath37 .
the static susceptibility of the ground - state of can be computed using the bond - order mean - field ( bomf ) approximation @xcite ( see appendix .
[ app : bomf ] ) . in the strong - coupling limit @xmath48 ,
the spin gap is in the @xmath49 sector and the magnon branch is well separated from the two - magnons continuum . on
can thus neglect the @xmath50 contribution and keep only the single magnon one .
the details of the calculations are given in appendix [ app : bomf ] and show that the susceptibility displays the same singularities as the spin structure factor . in the large @xmath51 regime , the result reads @xmath52 before turning to the interaction between two impurities , it is first instructive to consider the magnetization pattern induced by a single impurity , which is also of interest for nuclear magnetic resonance ( nmr ) experiments .
the impurity is located at site @xmath53 and the corresponding effective magnetic field defined by is simply given by @xmath54 .
the expectation value of the spin operator @xmath35 is then given by linear response theory which , in fourier transform , reads @xmath55 this perturbative response is a priori valid far from the impurity .
the magnetization profile in sector @xmath56 is then @xmath57 in which the general expression of the coupling of the frustrated ladder hamiltonian is @xmath58 after computing the integral limit of the sum over the brillouin zone , one obtains two different situations , depending on the behavior of the residues : in the commensurate regime @xmath59 , the profile is given by @xmath60 } + \frac{e^{-x/\xi_\text{spin}^-}}{\sinh\left(1/\xi_\text{spin}^-\right ) p'\left[-\cosh\left(1/\xi_\text{spin}^-\right)\right ] } \bigg ) \;,\ ] ] where @xmath61 are the spin correlation lengths defined in eqs .
and @xmath62 is the derivative of the polynomial @xmath63 defined in eq . .
in the incommensurate regime @xmath64 , one has @xmath65 } \bigg ] \;,\ ] ] where @xmath66 and @xmath42 are defined by eqs . and .
notice that there is no unknown constant in these expressions .
the key point of this result is that the transition from commensurate to incommensurate correlations induces a discontinuity in the features of the magnetization profile , which will show up in the effective interaction too . indeed , at each side of the transition , both residues diverge but their sum tends to zero .
notice that exactly at the transition , the denominator factorizes , having a single pole of order 2 for which the residue is zero , corresponding to @xmath67 . on the contrary , the amplitude of the magnetization @xmath68 goes to @xmath69 at each side of the transition .
but this does not mean that the magnetization profile diverges at short distance .
of course , @xmath68 remains always bound by @xmath0 .
however , the fact that the amplitude of the asymptotic behavior diverges which makes the perturbative analysis of the linear response fail . induced by a non - magnetic impurity at site @xmath70 , for an isotropic ladder ( @xmath71 ) with @xmath72 .
the incommensurate regime displays oscillations at wave - vector @xmath66 .
fit is done using . ] on fig .
[ fig : aimantation_impurete ] , we compare the magnetization profile in the sector @xmath73 obtained by dmrg to the mean - field predictions . in practice , expressions and provide good estimates of the behaviors , but it is preferable to fit the magnetization profiles using the following ansatz : @xmath74 in the commensurate regime , and @xmath75 in the incommensurate one .
remarkably , except near the onset of incommensurability where the amplitude diverges , these expressions remain correct at small distances , down to @xmath76 . in nmr experiments
, an incommensurate @xmath66 would give a narrowing of the peak w.r.t . the commensurate case with the same @xmath77 since the magnetization will display smaller values even close to the impurity . and
wave - vector @xmath66 extracted from the magnetization profile induced by a single impurity .
dmrg results are compared to bomf predictions in the large @xmath51 regime . ]
these profiles give a simple way to numerically access the fit parameters and compare them to bomf predictions .
indeed , the values of @xmath42 and @xmath66 extracted from the magnetization profiles agree qualitatively well with the mean - field predictions , as shown on fig .
[ fig : xi_q ] .
in particular , we checked that the amplitude @xmath78 possesses a maximum close to the transition from commensurate to incommensurate .
physically , these calculations provide an explicit illustration of the fact that the impurity generates a spinon that is confined close to it through an effective attractive potential acting over a typical length - scale @xmath42
. induced by a non - magnetic impurity at site @xmath70 , for an isotropic ladder ( @xmath71 ) for @xmath79 in the commensurate regime .
the decay of the magnetization profile and effective interaction between two impurities are compared to the decay of spin correlations .
these three quantities display the same length - scale @xmath42 in the exponential . yet
, only correlations display a power - law correction ( see app . [
app : bomf ] ) . ] last , we stress that there is a qualitative difference between the magnetization profile and the spin correlation function ( see app .
[ app : bomf ] for discussion of spin correlations in the model ) .
one does not expect a power - law correction in the decay of the magnetization .
this is clearly visible on fig [ fig : correction ] where fitting the envelope using @xmath80 gives an exponent @xmath81 while the exponent found for the fit of the correlations is rather @xmath82 , as expected from the usual arguments recalled in app .
[ app : bomf ] .
these quantitative results on the magnetization profiles and their sensitivity to the commensurate - incommensurate transition share similarities with those on the effective interaction between impurities which we now discuss . within the bomf approximation , valid in the strong - coupling limit , the effective interaction between impurities of eq
. takes the following form in the thermodynamical limit : @xmath83 where @xmath84 is the polynom @xmath85 as for the magnetization profile , one can evaluate the integral using the residue theorem to obtain two cases : in the commensurate regime , one has @xmath86}{\sinh\left(1/\xi_\text{spin}^+\right ) p'\left[-\cosh\left(1/\xi_\text{spin}^+\right)\right]}e^{-x/\xi_\text{spin}^+ } + \frac{q^2\left[-\cosh\left(1/\xi_\text{spin}^-\right)\right]}{\sinh\left(1/\xi_\text{spin}^-\right ) p'\left[-\cosh\left(1/\xi_\text{spin}^-\right)\right ] } e^{-x/\xi_\text{spin}^- } \bigg ) \;,\end{gathered}\ ] ] while in the incommensurate regime , one has @xmath87}{\sin\left(q+i\xi_\text{spin}^{-1}\right)p'\left[\cos\left(q+i\xi_\text{spin}^{-1}\right)\right]}e^{iqx } \bigg ] \;.\ ] ] similarly to the magnetization profile , the amplitude of @xmath88 diverges close to the transition between the two regimes but is strictly zero at the transition point .
this divergence does not mean that the effective interaction gets stronger but rather that the applicability of the long - distance result is limited to large distances .
the effective interaction remains always bonded by the maximum of @xmath89 and @xmath90 ( see short distances behavior hereafter ) .
further , no power - law corrections are expected in this quantity , contrarily to what is commonly proposed @xcite .
yet , this result is valid in the strong - coupling limit and we observe that it remains correct down to the isotropic ladder regime . in the weak - coupling limit ,
it is possible to have power - law or logarithmic corrections but we have not studied this case quantitatively . between to impurities as a function of their relative distance for @xmath91 and @xmath92 .
dmrg results ( symbols ) are fitted using the expression . ]
numerically , we compute the effective interaction using dmrg by targeting the lowest energies in the singlet and triplet sectors .
we assume that the lowest triplet excitation is due to the interaction between the two impurities ( the spin gap is large enough in this system ) so that one can use the relation @xmath93 only the total @xmath94 is fixed in dmrg calculations .
thus , since the triplet sector has a contribution for @xmath95 , one accesses to the amplitude @xmath96 from the energy difference of the first two energies in this sector . in order to get the sign of @xmath97 ,
one compares the obtained energies in sector @xmath95 with the lowest in sector @xmath98 . on figure
[ fig : jeff ] , we fit the curves using the function @xmath99 where @xmath100 and @xmath101 in the commensurate regime .
as expected , the wave - vector @xmath66 and length - scale @xmath42 exactly correspond to the ones of the magnetization profile and correlations function .
the behavior of the amplitude @xmath102 is not quantitatively predicted by the bomf theory as , for instance , the behavior of @xmath42 is not in perfect agreement , due to the approximations made in the bomf .
still , as one sees on figure [ fig : j0 ] , that the amplitude @xmath102 displays a sharp increase in the vicinity of the commensurate - incommensurate transition , reminiscent from the divergence expected in eqs . - .
of the effective interaction ( a ) evolution without frustration ( @xmath79 ) for increasing coupling @xmath103 .
comparison with other energy scales is given : the mean antiferro magnetic couplings @xmath104 and the maximum value of the effective interaction couplings @xmath105 .
( b ) evolution in the strong - coupling regime @xmath91 for increasing frustration @xmath106 , showing the divergence at the transition to the incommensurate regime . ] the short distances effective interactions computed with dmrg , and displayed on fig . [ fig : jeff ] , do not follow the prescription .
in fact , although linear response theory fails at these distances , one can guess the sign and magnitude of @xmath107 by looking at each configuration ( see fig [ fig : interaction_courte_distance ] ) .
the first thing to notice is that a configuration with impurities on the same rung breaks the ladder and the elementary excitation is then a magnon in the largest piece which energy cost is slightly larger than the spin gap . such effect does not enter in the effective model since no spin-@xmath0 is located at the vicinity of an impurity in that case . without frustration ( @xmath79 ) , the effective interaction oscillates at @xmath108 , even at short distances , except when @xmath109 for which the effective coupling is almost zero within dmrg accuracy .
indeed , we observe on figure [ fig : interaction_courte_distance](a ) that this configuration breaks the ladder .
two spinons are then generated on disconnected fragments and behave independently , making the triplet and singlet states degenerate and the effective interaction equal to zero .
the largest value of the effective interaction , which we write @xmath105 in the following , is obtained when two impurities are neighbour on the same chain .
the magnitude is then controlled by @xmath89 and shown on fig .
[ fig : j0](a ) . in the presence of frustration ,
the configuration with @xmath109 no longer breaks the ladder .
figure [ fig : interaction_courte_distance](b ) shows that spinons freed by the impurities should anti - align due to @xmath90 , as for the configuration with @xmath110 sketched on [ fig : interaction_courte_distance](c ) . in both cases
, the corresponding effective interaction is expected to be antiferromagnetic ( positive ) , in agreement with the dmrg result of figure [ fig : jeff ] .
if @xmath90 is larger than @xmath89 , it typically sets the scale of the maximum coupling @xmath105 in the effective interaction .
with @xmath79 , ( b ) @xmath111 with @xmath112 and ( c ) @xmath110 with @xmath112 . ]
last , one can recall that two - body interactions is just an approximation and that terms involving more than two partners should be included to improve the comparison with ab - initio calculations involving many impurities .
the validity of two - body interaction has been discussed in ref .
to which we refer to for further details on this question .
we here discuss the nature of the couplings distribution @xmath113 resulting from doping the ladder and which is a central quantity for the understanding of the magnetic responses .
we use the following notation from now on : @xmath24 is the total number of sites , @xmath114 the length of the ladder , @xmath115 the number of impurities and @xmath116 the impurities concentration or doping .
the latter corresponds to the probability for a site to be occupied by an impurity .
the lattice spacing is taken to be one in both directions .
the relative distance between two points on the ladder is written as @xmath117 with @xmath118 and @xmath119 .
consider impurities that are randomly distributed on the ladder .
the probability @xmath120 of having a distance @xmath121 between two impurities is given by a geometric law @xmath122 to understand this formula , one can scan all intermediate sites between the impurities following a zig - zag path .
thus , within the ladder geometry which has the peculiarity to differ from a chain because of the possibility to put two impurities on the same rung ( case with @xmath123 and @xmath124 ) , we have the following results for the mean longitudinal and transverse distances : @xmath125 one recovers the intuitive behaviors @xmath126 and @xmath127 in the dilute limit @xmath128 .
thus , in this limit , the typical average distance @xmath10 between impurities , as if they were on a chain , is given by the effective doping @xmath129 as @xmath130 and one has to keep in mind the presence of this factor two in qualitative reasoning . to obtain the distribution of couplings , we use for analytical calculations the simplified and generic relation @xmath131 with @xmath66 a dimensionless wave - vector which accounts for a possible incommensurability and @xmath132 the spin correlation length ( in a shortened notation ) ,
@xmath133 a phase - shift and @xmath102 an energy scale .
their typical behavior with microscopic parameters was discussed in the previous subsections .
formally , one obtains the distribution of couplings using the definition @xmath134 and use for the discrete case @xmath135 as we will see , the magnetic curve will be deeply connected to the coupling repartition function that we denote by @xmath136 we also introduce the repartition function of antiferromagnetic couplings only : @xmath137 indeed , negative @xmath17s corresponding to ferromagnetic couplings will yield polarized impurities as soon as the field is turn on .
a correct way of defining an energy scale corresponding to a magnetic field in the problem is thus to average only the positive @xmath17s .
then , we take the following definition @xmath138 for the typical energy scale of the antiferromagnetic couplings . in this case , the interaction is purely antiferromagnetic corresponding to @xmath100 . changing variables is done using @xmath139 with @xmath140 . [ [ continuous - distribution ] ] continuous distribution + + + + + + + + + + + + + + + + + + + + + + + + + we first consider the most elementary situation where @xmath141 is approximated by a continuous function , which requires @xmath128 and @xmath142 with fixed @xmath143 .
then , what enters in the distance probability is the effective chain doping @xmath144 , giving the exponential law @xmath145 .
the calculation yields a symmetric power - law distribution @xmath146 for @xmath147 $ ] , featuring the exponent @xmath148 .
the corresponding repartition function reads @xmath149 \;. \label{eq : rep - com}\ ] ] the energy scale @xmath150 takes the simple form @xmath151 a similar expression has been used to interpret experiments with 3d effects @xcite . here
, the effective volume of the interaction boils down to @xmath152 .
[ [ exact - distribution - and - lattice - effects ] ] exact distribution and lattice effects + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the continuous distribution ansatz is not justified for systems with very short correlation length such as the isotropic ladder .
we here carry out the calculation in the discrete case to obtain the exact formula that can be compared to numerical histograms of the couplings used in numerical simulations .
we have @xmath153 and @xmath154 for @xmath147 $ ] .
one recovers the result of the continuous approximation under its assumption .
in particular , one can see that lattice effects decouple the effect of the correlation length and of the impurity concentration , i.e. the exponent is not a function of @xmath143 only , but a function of both @xmath19 and @xmath132 , which makes the results not so universal .
last , we notice that the distribution becomes flat for the particular value @xmath155 and we will see that this can have consequences on the shape of the magnetic curve . for a realistic value of the correlation length @xmath156 .
the continuous approximation result is compared to the exact discrete result , showing lattice effects discussed in the text . ] for the typical energy scale @xmath104 , the exact result is computed by directly summing upon the @xmath157 and one obtains @xmath158 here again , one can check that is recovered provided @xmath142 and @xmath128 while keeping @xmath159 finite .
otherwise , deviations from occur at all doping . in particular , the low doping regime @xmath128 at fixed @xmath132 is @xmath160 these lattice effects are illustrated in fig .
[ fig : meanj ] for a realistic case with @xmath161 , which is characteristic of the isotropic ladder limit . in this case
the @xmath162 function is not bijective which changes qualitatively the distribution of the @xmath17s .
the presence of the cosine significantly lowers the weight of the largest @xmath17s while the smallest @xmath17s will see their weight increase .
second , in the presence of fractional @xmath163 , commensurate effects happen while an irrational @xmath163 has qualitatively the behavior of a true quasi - periodic signal
. in particular , for rational @xmath163 and @xmath101 , a fraction of couplings can be zero .
yet this situation is unphysical for the model under consideration which generic case is non - zero phase shift and irrational @xmath163 . in order to illustrate the typical behavior of the repartition function in the commensurate and incommensurate regimes ,
we have sampled numerically the distribution of couplings .
results are gathered in fig .
[ fig : repartition ] .
the essential features are the following : ( i ) up to discrete effect , the exponent @xmath164 in eq
. captures well the power - law in the commensurate case ; ( ii ) for irrational @xmath163 , the repartition function is qualitatively very close to the commensurate case , with a similar exponent , and up to the weight redistribution towards lower @xmath17 which translates into a smaller energy scale @xmath150 .
( iii ) for rational @xmath163 and @xmath101 , plateaus appear at @xmath165 and @xmath166 in the figure , corresponding the many zero couplings , together with cusps in @xmath150 . yet , the latter situation being unphysical , the main conclusion is that incommensurability hardly affects the coupling distribution .
the fact that frustration , throughout incommensurability , only lowers the energy scale @xmath150 but hardly affects the distribution is essential to understand that frustration will have only minor effects on the local magnetic responses studied in the next sections . for @xmath167 and @xmath156 with @xmath101 .
energy scale @xmath150 vs the incommensurate wave - vector @xmath66 . ]
in this section , we improve on the work of sigrist and furusaki@xcite for the calculation of the averaged total spin and zero temperature curie constant of a doped system on a bipartite lattice .
the results that are obtained are more general than for the special case of a ladder , and can be useful in checking numerical simulations and understanding finite - size corrections .
we assume a finite size sample containing @xmath24 sites and doped with @xmath168 impurities , where @xmath19 is the impurity concentration which is fixed .
the impurities are assumed not to break the lattice into disconnected sublattices ( see discussion section [ sec : chainbreak ] ) . on a bipartite lattice , with two sublattices @xmath169 and
@xmath170 which have the same number of sites @xmath171 , applying marshall s theorem yields that the total spin @xmath172 of a given impurity configuration reads : @xmath173 where @xmath174 ( resp .
@xmath175 ) is the number of impurities on sublattice @xmath169 ( resp .
@xmath170 ) .
the probability of having a configuration with @xmath174 impurities on sublattice @xmath169 , is qualitatively similar to the result on a ferro - antiferromagnetic ( f - af ) chain @xcite , and given by @xmath176 then , the probability of having a total spin @xmath172 on a sample is @xmath177 where @xmath178 $ ] .
this result is exact and can be used to compute numerically the mean total spin and the curie constant . for large @xmath115 and fixed @xmath19 ,
according to the central limit theorem , @xmath179 converges towards a gaussian .
a saddle - point calculation gives the asymptotic behavior : @xmath180 of variance @xmath181 .
one then obtains the average spin and the average square - spin in the @xmath182 limit as @xmath183 the total zero - temperature curie constant matches exactly @xmath184 from , we obtain the following asymptotic behavior @xmath185\;. \label{eq : c - exact - ladder}\ ] ] these asymptotic results are compared to numerical calculations using the exact distribution on fig .
[ fig : finite - size - c ] . in the thermodynamical limit @xmath186
, we thus find that the curie constant _ per impurity _ is @xmath187 at @xmath188 , while the curie constant _ per spin _ is @xmath189 .
notice that one can also compute exactly the average of the squared spin : @xmath190/\binom{n}{n_i}\;,\end{aligned}\ ] ] which is useful to cross - check the statistical convergence of averaging over samples , but we did not manage to compute exactly @xmath191 . , for four different doping @xmath19 , obtained from exact calculations ( points ) and compared to eq .
the finite - size correction of the zero - temperature curie constant , compared to eq .
[ eq : c - exact - ladder ] .
the @xmath192 line shows the usually admitted result .
the incommensurate case corresponds to an isotropic ladder with @xmath193 . ]
these results remain correct in the commensurate regime because the effective model is still unfrustrated .
but in the incommensurate regime , expressions are not valid anymore .
the total spin and curie constant at zero - temperature can be computed from exact diagonalization on the effective model using the exact couplings extracted from dmrg data .
the results for a frustrated isotropic ladder in the incommensurate regime are shown on fig .
[ fig : finite - size - c ] .
the frustration induced by the incommensurability yields a appreciable reduction of both the total spin and the curie constant .
note that this reduction is essentially due to the short distance behavior of the effective interaction , hence the necessity to use the exact effective couplings computed in dmrg rather than the asymptotic law .
the doping dependence of the prefactors and finite - size corrections were missed in previous work .
they originate from the dilution of the lattice .
they actually play a crucial role in the quantitative understanding of the numerics that usually work with a restricted number of impurities .
last , the precise value at finite - size is essential in extracting the low - temperature exponent , as we will see in the next section .
results in eqs . and are exact assuming that there is a fully connected cluster containing @xmath115 impurities over @xmath194 sites . of course
, the network can be disconnected in many sub - clusters by the presence of impurities .
then , each cluster , which is a finite size system , will eventually contribute to @xmath195 and @xmath78 in the thermodynamical limit , which was pointed out by sigrist and furusaki who computed an evaluation of the correction in the ladder case @xcite .
thus , we expect that , in general , @xmath196 in the thermodynamical and that @xmath78 is not exactly given by .
lastly , it has been proposed in ref . that these breaks provide an natural cut in the length scale ( and an energy scale ) which should affect the behavior of the correlations .
this was confirmed numerically in ref . in
which results on depleted two - leg ladders are consistent with an upper bound of the order of @xmath197 reached for very low temperatures .
the probability and size distribution of these clusters are governed by percolation theory .
the percolation transition distinguishes two main regimes : ( i ) there exist an infinite cluster of sites below a certain critical doping @xmath198 , ( ii ) for @xmath199 , only finite - size clusters exist and their size distribution is typically exponential , associated with a mean cluster size that will be denoted by @xmath200 in the following . at the critical doping @xmath201 , there is still an infinite cluster and scaling is expected for the mean cluster size .
the value of @xmath198 is very sensitive to dimensionality and connectivity of the lattice .
these percolation regimes induce important finite size effects and are essential for experiments and numerical simulations . on a chain , it is clear that @xmath202 , i.e. any finite doping will break the lattice and the mean cluster size is easily related to the doping @xmath203 . on a ladder ,
the situation is similar in the sense that any finite doping breaks the lattice into sub - clusters .
this chain breaking effects have been shown to have quantitative results on the magnetic response of doped chains @xcite . yet
, computing @xmath200 in the case of ladder is not trivial and the remaining clusters are themselves doped with various concentration of impurities which makes the predictions more involved .
we give below an exact discussion of the cluster sizes in the ladder and apply the results to the chain breaking effects on ladders .
we consider a ladder with nearest - neighbour only .
connectivity of the network is broken if ( i ) two impurities fall on the same rung , or ( ii ) two impurities fall on diagonal positions on a plaquette ( see figure [ fig : interaction_courte_distance](a ) ) .
if @xmath204 is the impurity position , there are three positions at which a second impurity can break the ladder : @xmath205 , @xmath206 et @xmath207 . occupying a site with an impurity
has a probability @xmath19 . in the diluted limit @xmath128 ,
the density of cuts is then @xmath208 and the corresponding mean cluster size is given by @xmath209 ( the factor 3 was missing in ref . ) .
notice that in the presence of frustration , chain - breaking requires at least four neighboring impurities which would make a different scaling @xmath210 . for large enough distances ,
breaks are uncorrelated
so a fair description of the distribution law is that of a poissonian process @xmath211 with @xmath212 in the diluted and continuum limit .
an exact calculation of the cluster sizes distribution is carried out in appendix [ app : breaks ] and supports this phenomenological approach .
the exact distribution reaches very quickly an asymptotic behaviour given by a geometric law @xmath213 of parameter @xmath214 consequently , one recovers in the continuum limit and @xmath215 .
in particular , this provides finite-@xmath19 corrections to the @xmath216 scaling which turn out to be quantitative even for a few percent doping as we see now . averaging the total spin and curie constant over clusters is not trivial since the doping of each cluster can now be distributed between zero and approximately @xmath0 . to handle a correct estimate
, one would have to average using the joint distribution of cluster sizes and doping .
we give below a rough estimate that consists in assuming a fix doping @xmath19 for all cluster and averaging only over cluster sizes @xmath217 using @xmath218 .
neglecting the doping fluctuations should yield a good approximation in the diluted limit where cluster sizes diverge . averaging the equations
is performed using @xmath219 using the approximate law for the size distribution and a continuous approximation for the average spin , we get @xmath220 to obtain the density of spin and curie constant , one has to multiply them by the clusters density @xmath221 . the mean spin density @xmath222 and curie constant density @xmath223 now read @xmath224 and @xmath225 which gives in the diluted limit @xmath226 @xmath227 in agreement with the results of sigrist and furusaki @xcite up to a prefactor . in the opposite limit of high - density for impurities @xmath228 ,
the system is equivalent to a few independent spin-@xmath0s which essentially behave as in a paramagnetic phase .
therefore , we have that @xmath229 and @xmath230 . in particular
, we infer that there is an optimal doping @xmath231 which maximizes the curie constant and a slightly different one which maximizes the total spin .
the value of @xmath231 is non - trivial since it occurs at the crossing of the two asymptotes . and ( b ) curie constant density @xmath232 averaged . ]
these predictions are a lower bound for @xmath233 and @xmath232 since very small clusters with a few sites have a total spin and susceptibility larger than the random - walk result . in order to show the quantitative role of chain breaking and the validity of the results ,
we plot on fig .
[ fig : chainbreaking ] the limiting behaviors at low and high doping @xmath19 together with the formulas and .
although these formula are not exact , they capture well the existence of a maximum at an optimal doping .
we now turn to the effect of interactions on the magnetic curve of two - leg ladders doped with impurities .
the magnetic excitation is denoted by @xmath234 .
we choose to define the total magnetization density as @xmath235 so that the high field saturation density is @xmath236 . with this normalization ,
the contribution of non - interacting impurities carrying a spin-@xmath0 to the magnetization is the brillouin formula at temperature @xmath237 ( we set @xmath238 in the following ) @xmath239 \label{eq : brillouin}\ ] ] where @xmath19 thus corresponds to the _ impurity _ saturation magnetization , assuming that each impurity frees a spin-@xmath0 .
the latter assumption is actually affected by chain breaking and will be discussed in more details in sec .
[ sec : magnetization ] . in the low - field limit at finite - temperature ,
the curie constant density @xmath232 is defined as @xmath240 in the case of independent impurities , we would have the total curie constant @xmath241 corresponding to @xmath242 , as expected from .
if one uses the exact result for correlated impurities , then the curie constant density reads @xmath243 without chain breaking or with it .
we first consider a simple model , dubbed `` random dimer model '' in which impurities are assumed to build dimers with their neighbour .
dimers are independent but have couplings randomly distributed according to @xmath113 .
the magnetization of a single dimer of coupling @xmath17 is given by @xmath244}{1 + e^{j / t } + 2 \cosh[h / t]}\;.\ ] ] the total number of dimers is @xmath245 so that the total magnetization density averaged over the coupling distribution reads : @xmath246}{1 + e^{j / t } + 2 \cosh[h / t ] } \;. \label{eq : rdm - magnetization}\ ] ] the brillouin formula is recovered when @xmath247 or more physically , in the high - temperature limit when @xmath248 .
taking the zero - field limit at finite temperature in yields a temperature - dependent curie constant @xmath21 : @xmath249 which reaches the free spins result @xmath250 in the high - temperature regime .
the effective hamiltonian of interacting impurity spins is given by @xmath251 and is solved numerically using either ed for @xmath252 or qmc up to @xmath253 and provided there is no incommensurability , i.e. for @xmath100 .
ed provides all energies @xmath254 in a sector of total spin @xmath255 so finite - temperature predictions are accessible .
the couplings @xmath256 are obtained by sampling impurities configurations on a ladder and using either the approximate formula with chosen @xmath66 , @xmath102 and @xmath132 , or the exact couplings computed from dmrg .
two `` ab - initio '' methods are also used to compute observables directly on the original microscopic hamiltonian : the dmrg technique , which gives accurate results for the zero - temperature magnetization curve , and quantum monte - carlo sse calculations well suited for finite - temperature dependence .
in this section , we focus on the limit of vanishing magnetic excitation @xmath257 at finite @xmath237 .
the order of limits matters and the situation @xmath258 for fixed @xmath234 will be studied in the next section . in this limit
, a modified curie law is generically expected , written as @xmath259 where @xmath21 is a temperature - dependent curie constant , corresponding to a static susceptibility @xmath260 .
the goal of this section is to investigate quantitatively the whole @xmath21 curve and analyze the effect of interactions , doping and frustration on its behavior .
the various regimes of @xmath21 in depleted ladders were first sketched by sigrist and furusaki @xcite who gave the following picture : starting at high - temperatures , the spins are essentially independent because of thermal fluctuations so that @xmath261 . assuming that the spin gap @xmath3 is larger enough than the maximum coupling @xmath105 ( implicitly corresponding to the strong - coupling regime ) , lowering the temperature below the spin gap freezes all magnon excitations . only remain spin-@xmath0s freed by impurities which should behave independently for a range of temperatures @xmath262 .
this yields a plateau around @xmath263 if one neglects chain breaking and @xmath264 if they are taken into account to first order corrections .
lowering again temperature enables one to reach the zero - temperature plateau discussed above and which is approximately given by @xmath265 ( the @xmath189 plateau within ref .
conventions ) . in the regime governed by impurity - spins interactions ,
real - space renormalization group ( rsrg ) arguments@xcite generally gives low - temperature corrections of the form @xmath266 with @xmath267 a non - universal constant and @xmath268 an exponent which generally depends on doping @xmath19 and that captures the interesting physics about impurities interactions .
we now check and analyze this scenario using our various models and methods .
the random dimer model formula for the curie constant already displays a non - trivial temperature dependence due to the coupling distribution . in the case of the power - law distribution
for which the exponent @xmath269 is the one of the couplings distribution and @xmath270 ( see below eq . )
, the high - temperature expansion leads to @xmath271\end{aligned}\ ] ] while , at low-@xmath237 , a sommerfeld - like expansion , in which the constant three appearing the denominator of has to be carefully taken into account , yields a power - law : @xmath272 with the constant @xmath273 @xmath274 is of the order of @xmath275-@xmath276 and matches some simple numbers for specific values of @xmath268 : @xmath277 , @xmath278 .
the curves for various @xmath268 are represented in fig .
[ fig : dimermodel ] and show that the prediction works for a wide range of temperatures . in the limit of small @xmath268 , one has the expansion @xmath279 with @xmath280 . .
_ inset _ : behavior of the constant @xmath281 from . ]
it is clear that , even though there is a power - law , there is nothing universal in this result .
the scaling originates only from the fact that the distribution is a power - law .
last , the zero - temperature result @xmath282 , which already differs from free spins , is always expected in the case of a symmetric @xmath283 distribution . in ab - initio ed and qmc calculations ,
the temperature - dependent curie constant is computed exactly using the average over thermal states and disorder configurations : @xmath284 since @xmath285 when @xmath286 for both the microscopic and effective models due to su(2 ) symmetry .
notice that , on the effective model , chain breaking effects discussed in sec . [
sec : chainbreak ] are not included . within the effective model description for @xmath287 impurities .
] we present on fig .
[ fig : ed - coft ] the results obtained from exact diagonalization with @xmath287 impurities and averaged over 10000 samples .
the first remarkable result is that the zero - temperature plateau is very well approximated on finite sizes by using or its exact numerical evaluation .
in fact , the effective model is not a bipartite lattice model to which the theorem applies , but the fact that it originates from a bipartite model to which the theorem applies ( without frustration ) seems to make it hold even in the effective model .
the reason for that is certainly that the sign of the couplings satisfy the bipartite nature of the original lattice .
the low - temperature departure from the @xmath188 plateau is very sensitive to finite - size effects and disorder averaging .
this makes it hard to capture the hypothetical thermodynamical behavior with this data .
yet , for the intermediate temperatures regime up to the high - temperature saturation plateau , we obtain a very good fit of the @xmath288 data using a power - law , as one can see from fig .
[ fig : ed - coft](b ) . collecting the fitted exponents fig .
[ fig : ed - coft](c ) shows a very good agreement with the @xmath164 prediction of the random dimer model . however , the behavior in the thermodynamical limit within the effective model is difficult to address . to sketch a possible scenario ,
we refer to the works done on the f - af random chain @xcite . indeed , for reasonably short correlation lengths @xmath132
, the effective model should fall into the f - af universality class in the rsrg sense .
this universality class has been dubbed as the large - spin phase , which is of the griffith s type , and for which it has been found that the total spin follows the random - walk scenario discussed above , and that a power - law correction to the zero - temperature curie constant is expected .
as regards the possible universal exponents of this phase , it was found numerically @xcite that it is strongly dependent on the singular nature of the initial coupling distribution . by denoting @xmath289 the initial distribution of the couplings ,
the following scenario is proposed : ( i ) when @xmath290 with @xmath291 , the rsrg flows towards a non - universal fixed point with non - universal value of @xmath268 .
this exponent should depend on @xmath19 and @xmath132 but is not necessarily equal to @xmath164 ; ( ii ) when @xmath292 ( initial distribution `` not too singular '' ) , the rsrg flows towards a universal fixed point with @xmath293 .
qmc calculations @xcite have demonstrated the following typical behavior for the curie constant on the f - af random chain : at high - temperatures below the saturation plateau , @xmath21 strongly depends on the initial distribution coupling . yet , at low enough temperatures , the various @xmath21 curves collapse very close to the rsrg prediction with @xmath294 where @xmath295 and @xmath296 .
coming back to the situation of doped ladders , we may propose the following scenario . provided the rsrg picture is applicable to the ladder , something certainly true for @xmath297 but hard to justify when @xmath298 , we first expect from refs .
that the high - temperature regime is always dependent on the distribution .
interestingly , in the doped ladder situation and within the random dimer picture , we found that this regime displays a power - law behavior with an exponent @xmath164 which is simply related to the coupling distribution exponent .
then , one expects that the rsrg picture develops at low - temperatures with two possibles cases .
the @xmath299 criteria translates on ladder to a critical doping @xmath300 such that ( i ) if @xmath301 , the low - temperature exponent is non - universal , dependent on @xmath19 and @xmath132 and could differ from the high - temperature exponent expected to be @xmath164 ; ( ii ) if @xmath302 , one can fall into the rsrg universality class and the low - temperature exponent should become independent of @xmath19 and @xmath132 and reaches @xmath303 .
one must notice that the second situation can actually be realistic for doped ladders in the weakly coupled regime , since , for instance , @xmath304 for @xmath305 , giving @xmath306 .
impurities and two correlation length @xmath161 ( a ) and @xmath307 ( b ) .
disorder averaging has been done over a few thousands of random configurations .
the curie constant per impurity ( from which the asymptotic value @xmath308 has been subtracted ) is shown _ vs. _ temperature for various concentrations ( symbols ) , together with power - law fits ( dashed lines ) of the form @xmath309 where @xmath310 is a varying exponent indicated on the plot .
the lines in the high - temperature region are with exponent @xmath311 .
the @xmath312 line is the universal regime found in the f - af chain by frischmuth _
et al . _
@xcite , including the same prefactor @xmath313 . ] in order to test this scenario which relies on several questionable assumptions , although it looks plausible within the usual pictures discussed in random one - dimensional magnets , we have carried out qmc simulations on the effective model up to @xmath253 impurities .
the results for the curie constant are plotted in fig .
[ fig : qmc - effective - model](a - b ) for two values of the correlation length @xmath161 and @xmath307 which respectively correspond to @xmath314 and @xmath315 .
we observe on these data a crossover from a fast decaying high - temperature regime , roughly controlled by the exponent @xmath164 and a smaller doping - dependent exponent at lower temperatures . for large values of @xmath143 ,
the deviation is even clearer and the exponent does not seem to exceed the rsrg universal result of @xmath316 .
we also show the f - af universal result on the same plot showing that data at large @xmath317 qualitatively saturates on this limit .
these results give good confidence that the above scenario is plausible . to further test the scenario ,
we have extracted the low - temperature exponent and plot it against @xmath19 and @xmath317 on fig .
[ fig : qmc - effective - model-2 ] .
we observe that the low - doping regime is consistent with the @xmath311 limit while intermediate dopings display significant deviations and a tendency to saturate around the rsrg universal regime for @xmath318 . yet
, the validity of the random f - af rsrg picture can be questioned for two main reasons : when @xmath143 becomes large , the dilute short range interaction limit fails and it is not guaranteed that the rsrg is still under control with longer range interaction .
second , as we will see on the magnetic curve , the discretized nature of the distribution can play a quantitative role .
the criteria for the initial distribution exponent @xmath319 is valid within a continuous description but the discretized nature of the coupling can make the distribution more singular .
interpretation in that sense was proposed on the same model through the study on correlation lengths@xcite .
still , we observe that the rsrg argument does capture a lowering of the @xmath320 curve w.r.t . the @xmath311 naive expectation .
having an accurate quantitative description of this curve yet remains a challenging question .
frustration makes the lattice non - bipartite so that the exact results do not apply .
still , within the effective model , if frustration is not too strong , the system remains commensurate and the above results remain valid and shows that the behavior is the same .
when frustration is large enough to induce incommensurability in the system , the effective model is affected and next - nearest neighbor couplings can become frustrating .
in this situation , qmc calculations are not possible due to the sign problem and we carry out ed calculations , limited to a @xmath287 impurities .
we do not show the data because the picture remains essentially and quantitatively the same as for the commensurate regime .
this absence of strong qualitative differences is certainly due to the fact that the spin correlation length is small and prevents frustrating effects to develop on large scales .
furthermore , rsrg arguments tell that the commensurate and incommensurate cases should fall into the same f - af random chain picture so that incommensurability does not actually plays a fundamental role in this model , at least on its one - dimensional version .
we now discuss the overall behavior of @xmath21 computed on the original microscopic model doped with impurities in order to test the scenario by sigrist and furusaki discussed above .
we take the situation of an isotropic ladder which has @xmath321 but in which the energy scales @xmath102 , @xmath105 and @xmath89 are very close to each other ( see fig . [
fig : j0 ] ) .
the first significant effect as seen on fig .
[ fig : qmc - coft - full](a ) is thus the absence of an intermediate plateau of independent impurity spins .
thermal magnons are activated before impurity spins become uncorrelated by thermal excitations so that the contributions of both can never be separated .
this absence of plateau in the isotropic ladder will have its counterpart in the magnetic curve while scanning the energy scales with the magnetic field rather than with the thermal energy ( see sec .
[ sec : magnetization ] ) . in the large @xmath322 limit
, the separation of energy scales suggests that the plateau could be visible but we have not checked it numerically . at temperatures slightly below the temperature corresponding to the spin gap @xmath323 ,
a power - law behavior is clearly visible showing the regime in which the effective model accounts for the physics .
the exponent is found to depend on doping , with very small exponents at low dopings which could give the impression of the presence of a plateau , although this is not correct .
a systematic extraction of the exponent ( see fig . [
fig : qmc - coft - full](b ) ) gives the results plotted on fig .
[ fig : qmc - effective - model-2 ] against the effective model results .
slightly larger exponents are found but the agreement can be viewed as correct considering the low values of the exponents and the difficulty to tackle this low - temperature regime numerically . consequently , the effective model seems to capture the physics at low - energy of the interacting impurity spins .
although the convergence towards the universal rsrg regime is plausible at small @xmath322 and low doping from our results on the effective model , the numerical challenge it represents on the microscopic model is beyond the scope of this paper ( the spin gap becomes significantly smaller ) .
another way to probe the effective interactions between impurities is to scan the energies using a magnetic field rather than temperature . as we are studying the part of the magnetic curve which typically lies below the spin gap @xmath3
, the results correspond to accessible magnetic fields , and are then particularly relevant to experimental measurements .
we aim at proposing some possible relevant fits of this regime . above the spin gap ,
the elementary excitations involve magnons which can localize in the disordered environment . there , the physics becomes quite different and we do not address these questions related to bose - glass physics .
we now turn to the generic behavior of the magnetic curve @xmath324 . in the previous section , the non - trivial behavior when
@xmath257 at finite @xmath237 was discussed .
physically , it corresponded to susceptibility measurements performed with @xmath325 .
strictly speaking , we must have @xmath326 when @xmath286 due to the su(2 ) symmetry .
if one now considers a finite - system with @xmath188 and a small but finite magnetic field , the degeneracy within a sector of total spin @xmath172 will be lifted to favor the state @xmath327 . then , there exists a disorder averaged magnetization jump @xmath328 which matches @xmath329 .
this is typical of a partially ferromagnetic state .
if one does not take into account chain breaking effects , as we do for the effective model , the scaling of @xmath191 yields a magnetization jump that vanishes in the thermodynamical limit as @xmath330 interestingly , we notice that , due to the random walk argument , the prefactor is actually related to the zero - temperature curie - constant @xmath232 by @xmath331 which makes a connection between the two non - commutating limits of the magnetic responses under study .
if we take chain breaking effects into account , then there exists a jump _ even in the thermodynamical limit _ which reads @xmath332 in the diluted limit @xmath333 .
lastly , one expects that , within a picture of impurities bringing each exactly one spin , the saturation plateau corresponding to the polarization of all these spins equals @xmath334 . yet ,
chain breaking effects should lower this value since configurations where two impurities are on the same rung do not bring any free spin .
taking this effect into account gives an expected saturation at @xmath335 .
this effect matters for dmrg or qmc data as well as experiments .
using the random dimer model , the low - temperature magnetization curve ( for @xmath336 ) takes a fermi - dirac form to a good approximation @xmath337 where @xmath234 plays the role of the chemical potential .
this is physically transparent as the system is equivalent in this limit to a collection of two - level systems with only the singlet and triplet @xmath338 states contributing to the low - energy physics which naturally maps onto fermionic statistics . in particular , the @xmath188 limit of this model
gives that the magnetic curve is simply related to the repartition function @xmath339 of the couplings through @xmath340 . in the case of the continuous distribution , this yields a power - law behavior @xmath341 with @xmath269 and for @xmath342 , which already deviates significantly from the brillouin picture .
it is important to notice that situation where @xmath343 from is physical in the case of systems with a large correlation length @xmath344 .
then , the curvature of the magnetic curve is expected to change from concave to convex . and finite temperature .
@xmath345 represents the magnetization jump . ]
still , we see that the random dimer model fails to reproduce the correct @xmath257 limit and gives for the jump @xmath346 ( @xmath347 for eq . ) .
one can incorporate the exact result in the rdm by stretching the repartition function of antiferromagnetic couplings @xmath348 .
we thus define the phenomenological stretched random dimer ansatz as @xmath349 physically , the issue of the random dimer model is that it works with total spins @xmath350 and @xmath351 and can not capture the large - spin formation . these features and ansatz of the random dimer model
are represented on fig .
[ fig : magnetization - rdm](a ) .
lastly , this rough understanding of the shape of the curve leads to the following simple power - law fit which could be useful for experiments or numerical calculations : @xmath352 in which one can leave free the three parameters @xmath345 , @xmath102 and @xmath268 . interestingly ,
rsrg arguments @xcite have also proposed a power - law behavior for the magnetic curve when @xmath353 based on energy scales phenomenology .
we compute numerically the magnetization curves at zero temperature using dmrg on the microscopic model and average over many configurations .
the results for the isotropic ladder @xmath354 are displayed on fig .
[ fig : magnetization - t0](a - b ) for both a system without frustration and with frustration in the incommensurate regime .
qualitatively , the two curve are essentially governed by the coupling distribution and frustration does not have a drastic qualitative effect .
interestingly , the simple approximations described in the preceding section account rather well of the behavior of the curve .
first , the ed on the effective model captures the power - law like behavior and even underlines the discrete nature of the coupling distribution .
this discrete nature is transparent from the ansatz using the exact effective couplings .
the dmrg does show faded steps corresponding to the larger couplings , and ed too .
the envelope of the random dimer model is captured by the continuous version of the coupling distribution .
yet , we see that one really needs the discretized version to be quantitative .
last , we show that a fit of the form captures the mean power - law behavior of the curve in a satisfactory way .
this is all the more relevant as we will see that temperature tends to fade the steps due to the discrete couplings .
one can notice the slight difference between the ed and dmrg results .
we attribute these to two main possible effects .
first , as the systems are chosen to have the same total number of impurities , the limitation of the two - body interaction effective model can play a role .
many - impurity interactions could become relevant even though these are subdominant effects .
second , we have seen that chain breaking effects must make the saturation plateau occur at @xmath355 , but it also has the effect of averaging magnetic curves over various dopings .
indeed , in the presence of chain breaking effects , each piece has a different doping which approaches @xmath19 on average but can be lower or higher .
this should have significant effects compare to the fixed @xmath19 curve of the ed on the effective model .
the last important remark is that no saturation plateau is reached in the isotropic ladder . as for the curie constant plateau ,
this is due to the fact that the typical energy scales are of the same order of magnitude @xmath356 ( see fig .
[ fig : j0 ] ) .
then , magnons become activated by the magnetic field before all impurities are truly polarized . the energy scales separation in the strong - coupling limit suggests that such a plateau could be possible at large @xmath322 but we have not investigated this situation in details . in particular , the microscopic model displays small couplings in this limit which are harder to capture with dmrg .
the approximation of eq . , naturally yields low - temperature corrections from a sommerfeld expansion , valid provided @xmath357 , which reads @xmath358 , where @xmath359 is the derivative of the coupling distribution . for instance , in the case of the continuous distribution and taking the approximation @xmath360 , the temperature corrections depend on doping and magnetic field through @xmath361 here again , the response to a small temperature is expected to strongly depend on the side of the limiting case @xmath362 , displaying a change in sign on the corrections . in the particular situation
where @xmath362 , for which @xmath113 is flat , the magnetization curve of the random dimer model can actually be computed exactly @xmath363}{1 + 2\cosh[h / t ] } \\ & \times \frac{t}{j_0}\ln\left\ { \frac{1+e^{j_0/t}(1 + 2\cosh[h / t])}{1+e^{-j_0/t}(1 + 2\cosh[h / t ] ) } \right\}\;. \end{split } \label{eq : m - exact - z1}\ ] ] following the previous remark on the inability of the random dimer model to account for the large - spin formation , we can devise an extension of the stretched dimer model at finite temperatures using the following ansatz : @xmath364 where the first part accounts for the contribution of ferromagnetic couplings , while the second accounts for the magnetization process of antiferromagnetic dimers .
the first term should in principle correspond to a brillouin function of spin @xmath195 but this version already gives satisfactory results
. of depleted ladders of size @xmath365 sites , averaged over @xmath366 disordered samples , at finite temperature @xmath367 .
different impurity concentrations @xmath368 are shown , together with the clean case at the same temperature for comparison .
horizontal dashed lines show the expected saturation value for the impurities @xmath369 , and the full lines are ed results obtained with pbc on 1000 random clusters of 10 impurities from which one clearly sees that the saturation value is only reached when @xmath370 . ]
the effect of temperature is first discussed on fig .
[ fig : magnetization - finitet ] by showing the comparison between the brillouin response to the ed and random dimer model predictions for four increasing temperatures .
all curves should collapse at high temperatures @xmath248 .
we see that the zero - temperature steps are rapidly faded as temperature is turn on .
still , the deviation from the brillouin curve due to the interaction remains well visible for finite temperature and actually makes the random dimer model almost exact . in order to validate the above comparison
, we have compared the ed on the effective model to qmc , which is the appropriate method for finite temperature calculations on the microscopic model . in fig .
[ fig : magnetization ] , one observes a rather good agreement for several realistic dopings .
the larger the doping , the larger the deviation from the brillouin curve is and the larger the distance from saturation is when the magnons set in .
the slight difference between ed and qmc could here again be attributed to many - impurities interactions not taken into account in the effective model and also to the effective doping averaging induced by chain breaking effects , as for the zero - temperature curve .
then , the following message is almost quantitatively correct from the comparison between all different approaches : the low - part of the magnetic curve probes the couplings distribution between the impurities .
this is evident in the random dimer model and the picture survives to the microscopic model rather well .
this simple analysis is certainly due to the fact that we are discussing a simple observable ( density of magnetization ) which is little affected by low - energy behavior or correlations in the system .
therefore , it could be accessible and interesting to test such phenomenology in experiments working with quasi one - dimensional systems .
our theoretical study could in principle apply to several realistic spin gapped materials .
however , as seen above , a clear separation between different energy scales the spin gap @xmath3 below which free local moments are expected , and the largest effective coupling @xmath105 below which they start to correlate upon random f - af exchanges would be difficult to achieve in systems close to the isotropic ladder limit .
the separation remains plausible in the strong - coupling limit , although we have not investigate this point quantitatively in this paper . in the isotropic case
, a saturation regime of impurity spins will be hardly detectable .
nevertheless , the regime of large spin could be detected in curie tails at low temperature , provided the three dimensional ordering of induced moments ( expected below temperatures set by three dimensional couplings ) occurs at low enough temperature .
in such a respect , a new analysis of susceptibility data of zn ( @xmath350 ) or ni ( @xmath351 ) doped bicu@xmath1po@xmath2 form ref . may give interesting results , although the three dimensional ordering of induced moments occurs below a few kelvins @xcite .
perhaps more promising is the doped haldane chain system y@xmath1banio@xmath7 @xcite where a very small inter - chain coupling @xmath371 k is expected from neutron scattering @xcite despite a very large spin gap @xmath372 k. more generally , our study clearly shows that curie tails , present in all af materials even for undoped ones , because of intrinsic defects or imry - ma domain formation with random couplings @xcite , have to be analyzed perhaps more carefully than what is usually done . in particular , the assumption of free impurities leading to the extraction of their concentration @xmath19 through the simple form @xmath373 is not expected to be valid in many experimental situations . regarding the magnetization curve , our work can potentially apply to many materials where brillouin - like responses are observed upon increasing the external field . for spin - gapped systems , the effective couplings between local moments can strongly renormalize downwards the brillouin - like magnetization , and pushes the saturation towards larger magnetic fields , possibly larger than the spin gap @xmath3 .
this means that , at the critical field where magnon excitations start to appear , not all impurity - induced moments have been saturated .
such a phenomenology is expected for bicu@xmath1po@xmath2 in a field @xcite . nevertheless , for this ladder material @xcite , and also for other systems such as the herbertsmithite kagom compound @xcite , the presence of non - negligible dzyaloshinskii - moriya ( dm )
anisotropies make the situation much more difficult to analyze since dm terms induce a finite magnetic response also below the spin gap .
the modification of the brillouin - like response due to the competition between impurity physics and dm interactions in an external field at finite temperature is a very interesting subject , relevant for many realistic systems , that we leave for future studies .
the physics of randomly depleted ladder , studied initially in the seminal work of sigrist and furusaki @xcite , offers a remarkable playground for studying the effect of impurity disorder in gapped systems without and with frustration . in this contribution to the field , we improved on several intuitive results of ref . to provide quantitative predictions and comparison to numerics and adressed the shape of the magnetization curve .
based on a detailed analysis of the effective couplings between impurities and of the corresponding coupling distribution , we focussed the main two magnetic responses : the zero - field susceptibility , through the temperature - dependent curie constant , and the magnetization curve .
the first one is shown to have a non - trivial power - law behavior at very - low temperature in qualitative agreement with a rsrg scenario . the high - temperature deviation from free impurity spins
is well captured by a simple random dimer model .
this model also accounts qualitatively well for the magnetization curve for which we give several phenomenological fits at zero and finite - temperature which are in good agreement with accurate numerical calculations .
one of the key outcome of this study is that incommensurability ( induced by frustration ) plays little role in the local quantities we looked at .
indeed , the main consequence of incommensurability is a mere reduction of the zero - temperature spontaneous magnetization and of the low - temperature limit of the curie constant .
the situation might be different in higher dimensional system but the one - dimensional version seems to be in the same universality class , as expected from rsrg arguments .
these predictions on the magnetic responses could motivate experiments in that direction since the required temperatures and magnetic fields are accessible for several compounds .
al thanks ccile delaporte and antoine channarond for insightful discussions on markov chains calculations .
al and gr acknowledge support from the french anr program anr-2011-bs04 - 012 - 01 qudec .
nl acknowledge genci ( grants 2012-x2012050225 and 2013-x2013050225 ) and calmip for numerical resources and the french anr program anr-11-is04 - 005 - 01 . 10 h. alloul , j. bobroff , m. gabay , and p. j. hirschfeld , rev .
phys . * 81 * , 45 ( 2009 ) .
a. c. hewson , _ the kondo problem to heavy fermions _ no . 2 ( cambridge university press , 1997 ) .
f. evers and a. d. mirlin , rev .
phys . * 80 * , 1355 ( 2008 ) .
m. p. a. fisher , p. b. weichman , g. grinstein , and d. s. fisher , phys .
b * 40 * , 546 ( 1989 ) .
k. sato , l. bergqvist , j. kudrnovsk , p. h. dederichs , o. eriksson , i. turek , b. sanyal , g. bouzerar , h. katayama - yoshida , v. a. dinh , t. fukushima , h. kizaki , and r. zeller , rev .
82 * , 1633 ( 2010 ) .
k. binder and a. p.
young , rev .
phys . * 58 * , 801 ( 1986 ) .
e. dagotto and t. m. rice , science * 271 * , 618 ( 1996 ) . d. c. johnston , j. w. johnson , d. p. goshorn , and a. j. jacobson , phys
b * 35 * , 219 ( 1987 ) .
t. barnes , e. dagotto , j. riera , and e. s. swanson , phys .
b * 47 * , 3196 ( 1993 ) .
s. gopalan , t. m. rice , and m. sigrist , phys .
b * 49 * , 8901 ( 1994 ) ; b. normand , ch .
regg , phys .
b * 83 * , 054415 ( 2011 ) .
m. azuma , z. hiroi , m. takano , k. ishida , and y. kitaoka , phys .
rev . lett . * 73 * , 3463 ( 1994 ) .
m. azuma , y. fujishiro , m. takano , m. nohara , and h. takagi , phys .
b * 55 * , r8658 ( 1997 ) .
m. azuma , m. takano , and r. s. eccleston , j. phys .
* 67 * , 740 ( 1998 ) .
n. fujiwara , h. yasuoka , y. fujishiro , m. azuma , and m. takano , phys .
lett . * 80 * , 604 ( 1998 ) .
s. ohsugi , y. tokunaga , k. ishida , y. kitaoka , m. azuma , y. fujishiro , and m. takano , phys .
b * 60 * , 4181 ( 1999 ) .
b. koteswararao , a. v. mahajan , l. k. alexander , and j. bobroff , j. phys . : condens .
matter * 22 * , 035601 ( 2010 ) .
f. casola , t. shiroka , s. wang , k. conder , e. pomjakushina , j. mesot , and h .-
ott , phys .
. lett . * 105 * , 067203 ( 2010 ) .
s. r. white , r. m. noack , and d. j. scalapino , phys .
73 * , 886 ( 1994 ) .
h. fukuyama , n. nagaosa , m. saito , and t. tanimoto , j. phys .
. jpn . * 65 * , 2377 ( 1996 ) .
m. sigrist and a. furusaki , j. phys .
. jpn . * 65 * , 2385 ( 1996 ) .
n. nagaosa , a. furusaki , m. sigrist , and h. fukuyama , j. phys .
jpn . * 65 * , 3724 ( 1996 ) .
y. motome , n. katoh , n. furukawa , and m. imada , j. phys .
65 * , 1949 ( 1996 ) .
y. iino and m. imada , j. phys .
. jpn . * 65 * , 3728 ( 1996 ) .
m. imada and y. iino , j. phys .
. jpn . * 66 * , 568 ( 1997 ) .
mikeska , u. neugebauer , and u. schollwck , phys .
b * 55 * , 2955 ( 1997 ) . a. w. sandvik , e. dagotto , and d. j. scalapino , phys . rev .
b * 56 * , 11701 ( 1997 ) .
t. miyazaki , m. troyer , m. ogata , k. ueda , and d. yoshioka , j. phys .
* 66 * , 2580 ( 1997 ) .
g. b. martins , m. laukamp , j. riera , and e. dagotto , phys .
78 * , 3563 ( 1997 ) .
m. laukamp , g. b. martins , c. gazza , a. l. malvezzi , e. dagotto , p. m. hansen , a. c. lpez , and j. riera , phys .
b * 57 * , 10755 ( 1998 ) . m. greven and r. j. birgeneau , phys
81 * , 1945 ( 1998 ) . m. arlego , w. brenig , d. c. cabra , f. heidrich - meisner , a. honecker , and g. rossini , phys .
b * 70 * , 014436 ( 2004 ) . f. anfuso and s. eggert , europhys . lett .
* 73 * , 271 ( 2006 ) .
k. trinh , s. haas , r. yu , and t. roscilde , phys .
b * 85 * , 035134 ( 2012 ) .
f. tedoldi , r. santachiara , and m. horvati , phys .
* 83 * , 412 ( 1999 ) .
j. das , a. v. mahajan , j. bobroff , h. alloul , f. alet , and e. s. srensen , phys .
b * 69 * , 144404 ( 2004 ) .
s. imai , t. masuda , t. matsuoka , and k. uchinokura , eprint arxiv : cond - mat/0402595 ( 2004 ) .
m. hase , i. terasaki , and k. uchinokura , phys .
lett . * 70 * , 3651 ( 1993 ) .
h. fukuyama , t. tanimoto , and m. saito , j. phys .
. jpn . * 65 * , 1182 ( 1996 ) .
k. manabe , h. ishimoto , n. koide , y. sasago , and k. uchinokura , phys .
b * 58 * , r575 ( 1998 ) .
b. grenier , j .-
renard , p. veillet , c. paulsen , g. dhalenne , and a. revcolevschi , phys .
b * 58 * , 8202 ( 1998 ) .
d. augier , e. srensen , j. riera , and d. poilblanc , phys .
b * 60 * , 1075 ( 1999 ) .
d. g. shelton and a. m. tsvelik , phys .
b * 57 * , 14242 ( 1998 ) . b. normand and f. mila , phys . rev .
b * 65 * , 104411 ( 2002 ) .
n. laflorencie , d. poilblanc , and a. w. sandvik , phys .
b * 69 * , 212412 ( 2004 ) .
n. laflorencie , d. poilblanc , and m. sigrist , phys .
b * 71 * , 212403 ( 2005 ) .
j. bobroff , n. laflorencie , l. k. alexander , a. v. mahajan , b. koteswararao , and p. mendels , phys .
lett . * 103 * , 047201 ( 2009 ) .
a. dobry , p. hansen , j. riera , d. augier , and d. poilblanc , phys . rev .
b * 60 * , 4065 ( 1999 ) .
n. laflorencie and d. poilblanc , phys .
lett . * 90 * , 157202 ( 2003 ) .
e. c. andrade and m. vojta , europhys
. lett . * 97 * , 37007 ( 2012 ) .
o. mentr , e. janod , p. rabu , m. hennion , f. leclercq - hugeux , j. kang , c. lee , m .- h .
whangbo , and s. petit , phys .
b * 80 * , 180413 ( 2009 ) . a. a. tsirlin , i. rousochatzakis , d. kasinathan , o. janson , r. nath , f. weickert , c. geibel , a. m. luchli , and h. rosner , phys . rev .
b * 82 * , 144426 ( 2010 ) .
l. k. alexander , j. bobroff , a. v. mahajan , b. koteswararao , n. laflorencie , and f. alet , phys .
b * 81 * , 054438 ( 2010 ) .
a. lavarlo , g. roux , and n. laflorencie , phys .
b * 84 * , 144407 ( 2011 ) .
t. sugimoto , m. mori , t. tohyama , and s. maekawa , phys .
b * 87 * , 155143 ( 2013 ) .
k. w. plumb , z. yamani , m. matsuda , g. j. shu , b. koteswararao , f. c. chou , and y .- j .
kim , e - prints arxiv:1301.5324 ( 2013 ) .
choi , j. w. hwang , p. lemmens , d. wulferding , g. j. shu , and f. c. chou , phys .
lett . * 110 * , 117204 ( 2013 ) .
a. furusaki , m. sigrist , p. a. lee , k. tanaka , and n. nagaosa , phys .
lett . * 73 * , 2622 ( 1994 ) .
a. furusaki , m. sigrist , e. westerberg , p. a. lee , k. b. tanaka , and n. nagaosa , phys .
b * 52 * , 15930 ( 1995 )
. e. westerberg , a. furusaki , m. sigrist , and p. a. lee , phys .
b * 55 * , 12578 ( 1997 ) . b. frischmuth and m. sigrist , phys . rev . lett . * 79 * , 147 ( 1997 ) . b. frischmuth , m. sigrist , b. ammon , and m. troyer , phys . rev .
b * 60 * , 3388 ( 1999 ) .
t. hikihara , a. furusaki , and m. sigrist , phys .
b * 60 * , 12116 ( 1999 ) .
e. yusuf and k. yang , phys .
b * 67 * , 144409 ( 2003 ) . j. a. hoyos and e. miranda , phys . rev .
b * 69 * , 214411 ( 2004 ) .
t. giamarchi , c. regg , and o. tchernyshyov , nature physics * 4 * , 198 ( 2008 ) .
i. affleck , phys .
b * 43 * , 3215 ( 1991 ) .
t. giamarchi and a. m. tsvelik , phys .
b * 59 * , 11398 ( 1999 ) .
d. s. petrov , m. holzmann , and g. v. shlyapnikov , phys .
* 84 * , 2551 ( 2000 ) .
s. dettmer , d. hellweg , p. ryytty , j. j. arlt , w. ertmer , k. sengstock , d. s. petrov , g. v. shlyapnikov , h. kreutzmann , l. santos , and m. lewenstein , phys .
. lett . * 87 * , 160406 ( 2001 ) .
t. nikuni , m. oshikawa , a. oosawa , and h. tanaka , phys .
lett . * 84 * , 5868 ( 2000 ) . c. regg , n. cavadini , a. furrer , h .- u .
gdel , k. krmer , h. mutka , a. wildes , k. habicht , and p. vorderwisch , nature * 423 * , 62 ( 2003 ) .
m. jaime , v. f. correa , n. harrison , c. d. batista , n. kawashima , y. kazuma , g. a. jorge , r. stern , i. heinmaa , s. a. zvyagin , y. sasago , and k. uchinokura , phys . rev .
lett . * 93 * , 087203 ( 2004 ) .
v. s. zapf , d. zocco , b. r. hansen , m. jaime , n. harrison , c. d. batista , m. kenzelmann , c. niedermayer , a. lacerda , and a. paduan - filho , phys .
lett . * 96 * , 077204 ( 2006 ) .
m. klanjek , h. mayaffre , c. berthier , m. horvati , b. chiari , o. piovesana , p. bouillot , c. kollath , e. orignac , r. citro , and t. giamarchi , phys .
* 101 * , 137207 ( 2008 ) . c. regg , k. kiefer , b. thielemann , d. f. mcmorrow , v. zapf , b. normand , m. b. zvonarev , p. bouillot , c. kollath , t. giamarchi , s. capponi , d. poilblanc , d. biner , and k. w. krmer , phys .
* 101 * , 247202 ( 2008 ) .
b. thielemann , c. regg , k. kiefer , h. m. rnnow , b. normand , p. bouillot , c. kollath , e. orignac , r. citro , t. giamarchi , a. m. luchli , d. biner , k. w. krmer , f. wolff - fabris , v. s. zapf , m. jaime , j. stahn , n. b. christensen , b. grenier , d. f. mcmorrow , and j. mesot , phys .
b * 79 * , 020408 ( 2009 ) .
t. hong , a. zheludev , h. manaka , and l .- p .
regnault , phys .
b * 81 * , 060410 ( 2010 ) . r. yu , l. yin , n. s. sullivan , j. s. xia , c. huan , a. paduan - filho , n. f. oliveira , jr .
, s. haas , a. steppke , c. f. miclea , f. weickert , r. movshovich , e .-
mun , b. l. scott , v. s. zapf , and t. roscilde , nature * 489 * , 379 ( 2012 ) .
a. zheludev and t. roscilde , e - prints arxiv:1305.1194 ( 2013 ) .
g. roux , a. minguzzi , and t. roscilde , new j. phys .
* 15 * , 055003 ( 2013 ) .
m. vojta , phys .
* 111 * , 097202 ( 2013 ) .
j. p. lvarez ziga and n. laflorencie , e - prints arxiv:1304.7636 ( 2013 ) .
mikeska , a. ghosh , and a. k. kolezhuk , phys .
lett . * 93 * , 217204 ( 2004 ) .
t. roscilde , phys .
b * 74 * , 144418 ( 2006 ) . r. yu , s. haas , and t. roscilde , europhys . lett . *
89 * , 10009 ( 2010 ) .
r. yu , o. nohadani , s. haas , and t. roscilde , phys .
b * 82 * , 134437 ( 2010 ) .
t. vekua and a. honecker , phys .
b * 73 * , 214427 ( 2006 ) .
s. r. white , phys .
lett . * 69 * , 2863 ( 1992 ) ; phys .
b * 48 * , 10345 ( 1993 ) ; u. schollwck , rev .
phys . * 77 * , 259 ( 2005 ) .
a. w. sandvik and j. kurkijrvi , , 5950 ( 1991 ) ; a. w. sandvik , , 14157 ( 1999 ) ; o. f. syljusen and a. w. sandvik , , 046701 ( 2002 ) ; h. g. evertz , advances in physics * 52 * , 1 ( 2003 ) . j. sirker , n. laflorencie , s. fujimoto , s. eggert , and i. affleck , phys .
lett . * 98 * , 137205 ( 2007 ) . c. payen , e. janod , k. schoumacker , c. d. batista , k. hallberg , and a. a. aligia , phys .
b * 62 * , 2998 ( 2000 ) .
g. xu , j. f. ditusa , t. ito , k. oka , h. takagi , c. broholm , and g. aeppli , phys .
b * 54 * , r6827 ( 1996 ) . a. lavarlo and g. roux , phys
lett . * 110 * , 087204 ( 2013 ) .
f. casola , t. shiroka , a. feiguin , s. wang , m. s. grbi , m. horvati , s. krmer , s. mukhopadhyay , k. conder , c. berthier , h .- r .
ott , h. m. rnnow , c. regg , and j. mesot , phys .
lett . * 110 * , 187201 ( 2013 ) .
f. bert , s. nakamae , f. ladieu , d. lhte , p. bonville , f. duc , j .- c .
trombe , and p. mendels , phys .
b * 76 * , 132411 ( 2007 ) .
s. r. white and i. affleck , phys .
b * 54 * , 9862 ( 1996 ) .
in this appendix , we detail the calculation of the spin correlation functions within the bond - order mean - field theory developed for the frustrated ladder in ref . .
we will use the same notations as in this reference .
the dynamical and static structure factors of the model has also been addressed recently in ref . . within bomf theory in which the singlet operators on rungs are assumed to condense @xmath374 ,
the mean - field hamiltonian is solved through a bogoliubov transformation on the triplet operators in @xmath375-space @xmath376 @xmath377 in which @xmath378 and @xmath379 satisfy @xmath380 .
this leads to the diagonal version of the hamiltonian @xmath381 where @xmath382 is the dispersion relation which depends on @xmath383 and the chemical potential @xmath384 used to enforce the hard - core nature of the triplets on rungs .
these two parameters are usually computed self - consistently with numerical methods . in this paper , in order to have tractable analytical formulas , we use the following approximations , which turn out to be good in the strong - coupling limit , @xmath385 and @xmath386 the zeros of @xmath382 extended to the complex plane will control the singularities of most physical quantities . to this end , we introduce the polynomial @xmath387 such that @xmath388 in ref .
, we obtained that the spin structure factor defined by @xmath389 where @xmath390 are real - space spin correlations , is given in the bomf approximation by @xmath391 we now give more details on the two commensurable and incommensurable regimes , limited to the strong - coupling regime .
we recall that the transition occurs at @xmath392 .
* for @xmath393 , @xmath394 has two real roots , lower than @xmath395 .
consequently , @xmath396 has branch cuts and four branching points on the axis @xmath397=\pi$ ] , with imaginary parts denoted by @xmath398 that define two correlation lengths @xmath61 ( @xmath399 ) , such that @xmath400 in the strong - coupling limit . * for @xmath401
, @xmath394 factorizes exactly and the square root disappears in the denominator of @xmath396 .
there is no longer branch cuts and the branching points merge to give two poles on the axis @xmath397=\pi$ ] , with imaginary part @xmath402 , where @xmath403 * for @xmath404 , the roots of @xmath394 have a non - zero imaginary part .
consequently , the branching points leaves the axis @xmath397=\pi$ ] .
there coordinates can be written as @xmath405 where @xmath66 is the incommensurate wave - vector associated to the real - space correlations and @xmath42 is the spin correlation length . in the large @xmath51 limit ,
we obtain @xmath406 real - space behavior of the correlation function is recovered after a fourier transform of the static structure factor : @xmath407 one can not easily compute this integral using the theorem of residues , because of the branch cuts , but one can argue that the behavior in @xmath408 is essentially controlled by @xmath409 et @xmath410 with @xmath411 and @xmath412 the singularities of @xmath396 in the upper half plane .
furthermore , due to the presence of the square - root in the denominator , one may guess the following asymptotic behavior @xmath413 indeed , as for the @xmath89-@xmath90 chain @xcite , the @xmath414 correction yields better fits of the numerical results .
we thus have the following scenario for the correlation functions : * for @xmath415 , in the commensurate regime : @xmath416 where @xmath61 are given by and @xmath169 and @xmath170 are two constants that depend on @xmath61 .
* for the transition point @xmath401 , we remark that the factorization of the denominator makes the decay purely exponential .
then , one expects @xmath417 with @xmath42 given by and @xmath78 a constant depending on@xmath42 . * for @xmath404 , in the incommensurate regime :
@xmath418 where @xmath66 and @xmath42 are given by and , and @xmath419 and @xmath133 are constant depending on @xmath66 and @xmath42 . in order to compute the magnetic susceptibility at @xmath420
, one applies a magnetic field corresponding to the wave - vector @xmath421 @xmath422 in the bomf approximation , the hamiltonian then reads @xmath423 the energy correction is obtained from second order perturbation theory in @xmath234 as @xmath424 by definition , the susceptibility @xmath425 enters in the expression through linear response theory @xmath426 from which we deduce the following expression for the static susceptibility : @xmath427 where @xmath394 is the polynom defined in .
using the result for the susceptibility , on gets for the magnetization profile the prediction @xmath428 where the position @xmath429 are relative to the impurity site .
the integral can be computed with help of the residues theorem applied over a rectangle of base between @xmath430 and @xmath431 and infinite in the vertical direction .
performing the change of variables requires the calculation of the derivative @xmath432 \\ & = -\frac{j(r)}{\xi}\left [ 1 + q\xi \sqrt{\left(\frac{j_0}{j(r)}\right)^2 e^{-2r/\xi } -1 } \right ] \label{eq : derivative}\end{aligned}\ ] ] the zeros of the derivative of denoted by @xmath433 and satisfy the equation @xmath434 with @xmath435 $ ] .
consequently , we have the zeros @xmath436 one can take by convention @xmath437 to define the intervals in which the sign of the derivative is constant @xmath438 $ ] .
it is clear that the intervals have the same size @xmath439 .
formally , one can write @xmath440 the solutions @xmath441 of the equation @xmath442 are not analytically computable in general . for a given @xmath17 ,
there is at most one solution in each interval @xmath443 and there is a least one solution for @xmath444 .
let us denote by @xmath445 the number of solutions at a given @xmath17 so that the index ranges @xmath446 .
using , we have @xmath447 we restrict the discussion to the continuous distribution case since the analytical formula for the discrete version do not help with respect to a direct numerical sampling . in this case , the weighting by the continuous approximation for @xmath141 gives @xmath448 the reduction of probability of large @xmath17 is understood by studying the situation where @xmath449 so that there is only a single solution @xmath450 .
then , one can write @xmath451 with @xmath452 since the effect of @xmath66 is to decrease the position of the solution w.r.t . the @xmath100 result .
this gives @xmath453 although it is not obvious in the formula , one may convince one - self graphically that @xmath454 corresponding to a decrease of the weight at large @xmath17 . consequently , the weight of small @xmath17s increase since the signal can approach zero at any distance .
in this section , we study the distribution of sizes of disconnected ladders @xmath455 for a given impurity doping @xmath19 . if we consider an impurity at position @xmath204 , there are three positions for a second impurity that break the ladder : @xmath205 , @xmath206 and @xmath207 . in the diluted limit @xmath333 ,
the density of cuts is then @xmath208 and the average length of disconnected ladders @xmath456 . as the cuts are not correlated
( at least at large enough distances ) , it is reasonable to assume that the number of cuts follows a geometric law of parameter @xmath457 : @xmath458 in fact , the distribution can be calculated exactly . for this
, the ladder is described by a markov chain @xmath459 , where @xmath460 represents the configuration of the plaquette made of the two consecutive rungs @xmath461 and @xmath462 ( see fig .
[ fig : markov ] ) .
the markov property is verified : the configuration on plaquette @xmath462 only depends of the configuration on plaquette @xmath461 as they have a rung in common .
the @xmath463 configurations on a plaquette are classified as follows : 1 .
the plaquette dos not break the ladder and there is no impurity on the second rung , 2 .
the plaquette dos not break the ladder and there is one impurity on the second rung , 3 .
the plaquette breaks the ladder .
the transition matrix @xmath84 , whose elements are the probabilities to go to configuration @xmath16 from configuration @xmath15 , writes @xmath464 where @xmath465 the transition probabilities from configuration @xmath466 are not needed and we can take this configuration as a trap state . we want to calculate the distribution of distances to reach configuration @xmath466 @xmath467 starting from an initial distribution for @xmath468 @xmath469 equation can be expanded as @xmath470 where@xmath471 one can easily show by mathematical induction that @xmath472 as a result , we have @xmath473 the distribution is not exactly a geometric law but @xmath474 converges really quickly to a constant @xmath475 independent of @xmath476 , that is , independent of the initial distribution @xmath477 . in the limit @xmath333 ,
one recovers @xmath457 . | the magnetic responses of a spin-@xmath0 ladder doped with non - magnetic impurities are studied combining both analytical and numerical methods .
the regime where frustration induces incommensurability is taken into account .
several improvements are made on the results of the seminal work by sigrist and furusaki [ sigrist and a. furusaki , j. phys .
soc .
jpn .
, * 65 * , 2385 ( 1996 ) ] , and deviations from the brillouin magnetization curve due to interactions are also analyzed .
we first discuss the magnetic profile around a single impurity and the effective interactions between impurities within the bond - operator mean - field theory .
the results are compared to density - matrix renormalization group calculations . in particular ,
these quantities are shown to be sensitive to the transition to the incommensurate regime .
we then focus on the behavior of the zero - field susceptibility through an effective curie constant . at zero - temperature ,
we give doping - dependent corrections to the results of sigrist and furusaki on general bipartite lattices , and compute exactly the distribution of ladder clusters due to chain breaking effects .
solving the effective model with exact diagonalization and quantum monte - carlo gives the temperature dependence of the curie constant .
its high - temperature limit is understood within a random dimer model , while the low - temperature tail is compatible with a real - space renormalization group scenario .
interestingly , solving the full microscopic model does not show a plateau corresponding to the saturation of the impurities _ in isotropic ladders_. the second magnetic response which is analyzed is the magnetic curve .
below fields of the order of the spin gap , the magnetization process is controlled by the physics of interacting impurity spins .
the random dimer model is shown to capture the bulk of the curve , accounting for the deviation from a brillouin behavior due to interactions .
the effective model calculations agree rather well with density - matrix renormalization group calculations at zero temperature , and with quantum monte - carlo at finite temperature . in all , the effect of incommensurability does not display a strong qualitative effect on both the magnetic susceptibility and the magnetic curve .
consequences for experiments on the bicu@xmath1po@xmath2 compound and other spin - gapped materials are briefly mentioned . |
phantom dark energy is a special kind of theoretical models with the negative kinetic energy@xcite , which has been investigated extensively in cosmology because it can provide a mechanism to interpret the accelerating expansion of the current universe @xcite . due to its negative kinetic energy ,
the phantom dark energy owns the super negative equation of state @xmath3 , which leads to that the null energy condition is violated for phantom field .
the universe dominated by phantom dark energy will blow up incessantly and arrive at a future singularity named big rip at where anything in the universe including the large galaxies will be torn up .
although the phantom dark energy owns such exotic properties , it is still supported by recent precise observational data @xcite .
however , all of these observational examinations and constraints on phantom dark energy are taken in the scope of cosmology . in black hole physics , the behaviors and properties of phantom field
have been studied in last few years .
e. babichev @xcite found that after absorbing phantom field , the mass of black hole decreases , which implies that the cosmic censorship conjecture could be challenged severely since the charge of a reissner - nordstrm - like black hole could be larger than its mass as it accretes the phantom energy .
we studied dynamical evolution of the phantom scalar perturbation in the schwarzschild black hole spacetime and found that the phantom scalar perturbation grows with an exponential rate in the late - time evolution , which differs from the decay of the usual scalar perturbations in the background of a black hole @xcite .
moreover , some black hole solutions describing gravity coupled to phantom scalar fields or phantom maxwell fields have been found and the corresponding geometric structure and thermodynamic properties are also studied in @xcite .
the strong gravitational lensing of such kind of black hole with phantom hair has been investigated in @xcite .
these investigations discloses some effects of phantom field on the black hole physics .
however , the observation examinations and constraints on phantom dark energy from black hole physics are still lacking .
quasi - periodic oscillations is one of very promising tools to test theories of gravity in the strong field regime and to constrain black hole parameters .
the investigations show that quasi - periodic oscillations are a common feature in the x - ray power density spectrum of black hole binaries @xcite . according to the types and the properties of quasi - periodic oscillations ,
their low frequencies are distributed in the range @xmath4 hz and high - frequencies are in the range @xmath5 hz .
the relativistic precession model is proposed as an exact mechanism to explain twin high - frequency quasi - periodic oscillations as well as a low - frequency mode in low - mass x - ray binaries @xcite . in this precession model ,
the twin higher frequencies are identified respectively with the azimuthal frequency @xmath6 and the periastron precession frequency @xmath7 of a test - particle moving in quasi - circular orbits at the innermost disk region in the background spacetime .
the low - frequency mode in quasi - periodic oscillations is regarded as the nodal precession frequency @xmath8 , which is emitted at the same radius where the twin higher frequencies signals are generated .
the possibility of applying quasi - periodic oscillations to constrain the black hole parameters in various theories of gravity has been already investigated in @xcite .
the main purpose of this paper is to make a constraint on the parameters of a regular and slowly rotating phantom black hole by using of quasi - periodic oscillations in relativistic precession model and the observation data of gro j1655 - 40 .
meanwhile , we want to see whether phantom theoretical model can pass an observation test from black hole physics .
the paper is organized as follows : in the following section we will extend the regular phantom black hole solution @xcite to the slowly rotating black hole case and then study the effect of the phantom parameter on the angular velocity at the event horizon . in sec.iii , we make a constraint on the parameters of a regular and slowly rotating phantom black hole by using of quasi - periodic oscillations and the observation data of gro j1655 - 40 .
finally , we end the paper with a summary .
in this section , we will extend the regular phantom black hole solution to the slowly rotating black hole case .
let us first introduce briefly the regular and static phantom black hole obtained in literature @xcite .
the action for phantom scalar field @xmath9 in the curve spacetime is @xmath10.\label{act1}\end{aligned}\ ] ] after solving the equation of motion of phantom scalar field and einstein field equation , bronnikov _ et al _ @xcite obtained a regular and static phantom black hole described by @xmath11 with @xmath12,\label{f0}\\ \phi&\equiv&\sqrt{2}\psi=\sqrt{2 } \arctan \frac{r}{b},\\ v&=&\frac{3m}{b^3}\bigg[(\frac{\pi}{2}-\psi)(3 - 2\cos^2\psi)-3\sin\psi\cos\psi\bigg ] .
\label{ff1}\end{aligned}\ ] ] here @xmath13 is the mass of black hole and @xmath0 is a positive constant related to the charge of phantom scalar field .
when @xmath0 tends to zero , the phantom scalar field @xmath9 becomes a constant and the corresponding potential @xmath14 approaches to zero , which means the action reduces to the usual action without material field and then the black hole solution ( [ metric0 ] ) recovers the known schwarzschild black hole one .
for the phantom black hole spacetime with non - zero value of @xmath0 , it is regular and has no singularity because the curvature scalar @xmath15\nonumber\\ & & + 36b^2m^2r^2(6r^4 + 14b^2r^2 + 11b^4)+12b^6(2mr^3 + 8b^2mr+b^4)\bigg\},\end{aligned}\ ] ] is neither divergent nor zero at anywhere . near the region @xmath16
, one can find @xmath17 which means that as the phantom charge @xmath0 disappears @xmath18 of the regular phantom black hole spacetime can reduce to that of schwarzschild black hole one .
we are now in position to obtain a phantom rotating black hole solution . for the electro - vacuum case , one can obtain a rotating black hole solution from a static black hole one by the newman - janis algorithm @xcite .
for example , one can obtain kerr black hole solution from schwarzschild one by this technique .
however , for the cases with scalar field , one must improve the usual newman - janis algorithm so that the rotating counterpart obeys to einstein equation with scalar field .
in general , it is difficult how to improve the newman - janis technique for the non - electro - vacuum cases . therefore , we here focus on obtaining a regular slowly rotating phantom black hole solution originated from the static and spherical symmetric solution ( [ metric0 ] ) by solving einstein equation of the gravity system ( [ act1 ] ) . the metric of a regular slowly rotating phantom black hole can be assumed as @xmath19 where @xmath20 is the rotation parameter associated with its angular momentum . in the case of a slowly rotating spacetime ( [ metric3 ] ) , we can suppose both of the phantom scalar field @xmath9 and the metric function @xmath21 depend only on the radial coordinate @xmath22 .
and then the equation of the phantom scalar field in spacetime ( [ metric3 ] ) can be expressed as @xmath23 -\frac{dv(\phi)}{d\phi}=0,\end{aligned}\ ] ] which is similar to that in the static and spherical symmetric spacetime ( [ metric0 ] ) because the phantom scalar field @xmath9 is supposed to only a function of @xmath22 and all of the higher order terms @xmath24 are neglected .
inserting the metric ( [ metric3 ] ) into the einstein s field equation , we find that the non - vanishing components of field equation can be expanded to first order in the angular momentum parameter @xmath20 as @xmath25 = \frac{1}{2}[u(r)\phi'^2 - 2v]+\mathcal{o}(a^2),\label{gt1}\\ rr : & & \frac{1}{u(r)(r^2+b^2)^2}\bigg[u'(r)r(r^2+b^2)+u(r)r^2-(r^2+b^2)\bigg ] = -\frac{1}{2u(r)}[u(r)\phi'^2 + 2v]+\mathcal{o}(a^2)\label{gt2},\\ \theta\theta:&&\frac{1}{2(r^2+b^2)}\bigg[u''(r)(r^2+b^2)^2 + 2u'(r)r(r^2+b^2 ) + 2u(r)b^2\bigg]=\frac{1}{2}(r^2+b^2)[u(r)\phi'^2 - 2v]+\mathcal{o}(a^2)\label{gt3},\\ \phi\phi:&&\frac{\sin^2\theta}{2(r^2+b^2)}\bigg[u''(r)(r^2+b^2)^2 + 2u'(r)r(r^2+b^2 ) + 2u(r)b^2\bigg]=\frac{\sin^2\theta}{2}(r^2+b^2)[u(r)\phi'^2 - 2v]+\mathcal{o}(a^2)\nonumber,\\ t\phi:&&\frac{1}{2(r^2+b^2)^2}\bigg\{(r^2+b^2)^2u(r)\frac{\partial^2f(r,\theta)}{\partial r^2}-f(r,\theta)\bigg[(r^2+b^2)^2u''(r)+2ru'(r)(r^2+b^2)+2u(r)(r^2 + 2b^2)\bigg]\nonumber\\ & & + ( r^2+b^2)\bigg[2f(r,\theta)+ \frac{\partial^2f(r,\theta)}{\partial \theta^2}-\frac{\partial f(r,\theta)}{\partial \theta}\cot\theta\bigg]\bigg\}+ \frac{f(r,\theta)}{2}[u(r)\phi'^2 - 2v]=0+\mathcal{o}(a^2)\label{gt5}.\end{aligned}\ ] ] solving the einstein equations ( [ gt1])-([gt3 ] ) , one can obtain the metric function @xmath26.\label{fba22}\end{aligned}\ ] ] separating @xmath27 , we find that the angular part @xmath28 satisfies @xmath29 and the radial part @xmath30 in eq.([gt5 ] ) obeys to @xmath31=0,\label{radial}\end{aligned}\ ] ] where @xmath32 is a separation constant . in order
that the coefficient @xmath33 can be reduced to that in the usual slowly rotating black hole without phantom field , here we set @xmath34 and then find that @xmath35 in this case .
this yields that the radial part of equation ( [ radial ] ) can be simplified as @xmath36 which leads to @xmath37 . \label{hr}\end{aligned}\ ] ] like in a usual slowly rotating black hole , the event horizon of black hole ( [ metric3 ] ) is given by @xmath38 , which is the same as that in the static and spherical symmetric case ( [ metric0 ] ) since we here expand the metric only to first order in the angular momentum parameter @xmath20 .
when the parameter @xmath39 , one can find that @xmath40 , which recovers that of a usual slowly rotating black hole without phantom scalar field .
the angular velocity of the horizon @xmath41 is an important quantity for a rotating black hole , which affects the region at where the super - radiance occurs in the black hole background . in the spacetime of a regular and slowly rotating phantom black hole ( [ metric3 ] )
, the angular velocity @xmath41 has a form @xmath42 with the parameter @xmath0 in a regular and slowly rotating phantom black hole spacetime.,width=264 ] which depends on the phantom charge @xmath0 .
we plot the change of the angular velocity @xmath41 with the parameter @xmath0 in fig.(1 ) , which tells us that @xmath41 decreases monotonically with the phantom charge @xmath0 and then the super - radiation is suppressed by phantom field in this case .
in this section , we will make a constraint on parameters of above regular and slowly rotating phantom black hole by quasi - periodic oscillations . in the general stationary and axially symmetric spacetime @xmath43 one can find that for the geodesic motion of particle there exist the conserved specific energy at infinity @xmath44 and the conserved @xmath45-component of the specific angular momentum at infinity @xmath46 since the metric coefficients are independent of the coordinates @xmath47 and @xmath48 .
and then the timelike geodesics can be expressed as @xmath49 with the effective potential @xmath50 where the overhead dot represents a derivative with respect to the affine parameter . for a circular orbit in the equatorial plane @xmath51
, we have @xmath52 solving above equations , one can obtain @xmath53 where @xmath54 is the angular velocity of particle moving in the circular orbits and @xmath6 is its corresponding azimuthal frequency . considering a small perturbation of a circular , equatorial orbit ,
i.e. , @xmath55 one can find that the perturbations @xmath56 and @xmath57 are governed by the following differential equations @xmath58 with @xmath59 and @xmath60 the radial epicyclic frequency @xmath61 and the vertical epicyclic frequency @xmath62 can be written as @xmath63 and @xmath64 , respectively .
finally , the periastron and nodal precession frequencies can be expressed as @xmath65 , @xmath7 and @xmath8 with the parameter @xmath0 in a regular and slowly rotating phantom black hole spacetime . here
we set @xmath66 and @xmath67.,title="fig:",width=204 ] , @xmath7 and @xmath8 with the parameter @xmath0 in a regular and slowly rotating phantom black hole spacetime . here
we set @xmath66 and @xmath67.,title="fig:",width=204 ] , @xmath7 and @xmath8 with the parameter @xmath0 in a regular and slowly rotating phantom black hole spacetime . here
we set @xmath66 and @xmath67.,title="fig:",width=204 ] in fig.(2 ) , we plot the dependence of the frequencies @xmath6 , @xmath7 and @xmath8 on the parameter @xmath0 in a regular and slowly rotating phantom black hole spacetime . with the increase of the phantom charge @xmath0 , the azimuthal frequency @xmath6 decreases and the periastron precession frequency @xmath7 increases . for the non - zero @xmath20 case
, the nodal precession frequency @xmath8 also decreases with the parameter @xmath0 .
however , for the non - rotation case , the nodal precession frequency @xmath8 becomes zero and is independent of the phantom parameter @xmath0 , which means that the vertical epicyclic frequency @xmath62 is identical to the azimuthal frequency @xmath6 for the spacetime ( [ metric3 ] ) with @xmath68 .
if three quasi - periodic oscillations frequencies are observed simultaneously , one can associate them to orbital and precession frequencies in the relativistic precession model . for the usual kerr black hole spacetime
, one can solve the corresponding three equations and determine the three variables ( @xmath22 , @xmath13 , and @xmath20 ) .
however , for the regular and slowly rotating phantom black hole , there are extra parameter @xmath0 related to the phantom scalar field , which means that we can not get values of the four variables through solving three equations directly . here
, we must resort to the @xmath69 analysis to best - fit the values of four unknown variables ( @xmath22 , @xmath13 , @xmath20 and @xmath0 ) in the phantom black hole spacetime ( [ metric3 ] ) with the three observed frequencies . from the current observations of gro j1655 - 40 ,
there are two set of data about these frequencies ( @xmath70 ) @xcite : @xmath71 moreover , there also is an independent dynamical measurement of the mass of the black hole @xcite : @xmath72 .
therefore , there are five free parameters : mass @xmath13 , rotation parameter @xmath20 , the phantom scalar charge @xmath0 , the radius @xmath73 and @xmath74 correspond the observations with three frequencies and two frequencies , respectively . with these data ,
the gravity of kerr black hole was tested in the relativistic precession model @xcite . for the phantom black hole ( [ metric3 ] )
, we obtain the minimum @xmath75 and constrain the black hole parameters @xmath76 at the @xmath77 confidence level . , @xmath78 and @xmath79 , respectively .
the red dot in the panels correspond the best - fit values of parameters : @xmath80 , @xmath81 and @xmath82.,title="fig:",width=264 ] , @xmath78 and @xmath79 , respectively .
the red dot in the panels correspond the best - fit values of parameters : @xmath80 , @xmath81 and @xmath82.,title="fig:",width=264 ] the best - fit values of the radius of circular orbital corresponding two sets of quasi - periodic oscillations are @xmath83 and @xmath84 , respectively . here
@xmath85 is the innermost stable circular orbit in the regular and slowly rotating phantom black hole spacetime with the best - fit values ( @xmath80 , @xmath81 and @xmath82 ) .
it means that the circular orbit of quasi - periodic oscillations are located at the strong gravitational - field region of the black hole because the black hole event horizon is at @xmath86 .
moreover , the contour levels of @xmath1 , @xmath78 and @xmath79 for the mass @xmath13 , rotation parameter @xmath20 and the phantom scalar charge @xmath0 are shown in fig.(3 ) . from the left panel in fig.(3 ) , we find that the contour lines with levels of @xmath1 , @xmath78 and @xmath79 are not closed curves since the parameter @xmath0 is non - negative for the regular phantom black holes ( [ metric0 ] ) and ( [ metric3 ] ) .
our results show that although the best - fit value of the phantom parameter @xmath0 is small , the allowed region of @xmath0 at the @xmath77 confidence level is @xmath2 , which means that the deviations from kerr metric is possible and the phantom theoretical model can not be excluded by the constraint from quasi - periodic oscillations with the observation data of gro j1655 - 40 .
in this paper we extend firstly the regular phantom black hole solution to the slowly rotating case and find that the presence of phantom field depresses the angular velocity of the event horizon and suppresses the super - radiation of black hole . and
then , we study the dependence of quasi - periodic oscillations frequencies in relativistic precession model on the phantom parameter @xmath0 . with the increase of the phantom charge @xmath0 , the azimuthal frequency @xmath6 decreases and the periastron precession frequency @xmath7 increases .
the nodal precession frequency @xmath8 also decreases with the parameter @xmath0 for the non - zero @xmath20 case , but it becomes zero in the case @xmath68 . with the observation data of gro j1655 - 40 , we constrain the parameters of the regular and slowly rotating phantom black hole .
our results show that although the best - fit value of the phantom parameter @xmath0 is small , the allowed region of @xmath0 at the @xmath77 confidence level is @xmath2 , which means that it is possible for a black hole has a phantom scalar hair . in other words ,
the phantom theoretical model can not be excluded by the constraint from quasi - periodic oscillations with the observation data of gro j1655 - 40 .
this work was partially supported by the national natural science foundation of china under grant no.11275065 , no .
11475061 , the construct program of the national key discipline , and the open project program of state key laboratory of theoretical physics , institute of theoretical physics , chinese academy of sciences , china ( no.y5kf161cj1 ) .
r. r. caldwell , m. kamionkowski and n. n. weinberg , phys .
lett . * 91 * 071301 ( 2003 ) .
s. nesseris and l. perivolaropoulos , phys .
d * 70 * , 123529 ( 2004 ) , arxiv : astro - ph/0410309 . s. nojiri and s. d. odintsov , phys .
b * 571 * , 1 ( 2003 ) , arxiv : hep - th/0306212 .
p. singh , m. sami and n. dadhich , phys .
d * 68 * , 023522 ( 2003 ) . j. g. hao and x. z. li , phys .
d * 70 * 043529 ( 2004 ) , arxiv : astro - ph/0309746 .
e. n. saridakis , p. f. gonzalez - diaz and c. l. siguenza , class .
grav . * 26 * , 165003 ( 2009 ) .
g. w. gibbons and d. a. rasheed , nucl .
b * 476 * 515 ( 1996),(arxiv : hep - th/9604177 ) .
g. clment , j. c. fabris and m. e. rodrigues , phys .
d * 79 * 064021 ( 2009 ) , arxiv:0901.4543 .
m. azreg - ainou , g. clment , j. c. fabris and m. e. rodrigues , phys .
d * 83 * 124001 ( 2011 ) , arxiv:1102.4093 . c. ding , c. liu , y. xiao , l. jiang and r. g. cai , phys .
d * 88 * , 104007 ( 2013 ) .
e. f. eiroa and c. m. sendra , phys .
d * 88 * , 103007 ( 2013 ) .
g. n. gyulchev and i. z. stefanov , phys .
d * 87*,063005 ( 2013 ) .
m. sharif and s. iftikhar , adv
. high energy phys . * 2015 * , 854264 ( 2015 ) .
s. e. motta , t. muoz - darias , a. sanna , r. fender , t. belloni and l. stella , mon . not .
. soc . * 439 * , 65 ( 2014 ) [ arxiv:1312.3114 [ astro-ph.he ] ] .
p. casella , t. belloni , and l. stella , l. astrophys .
j. * 629 * , 403 ( 2005 ) .
l. stella and m. vietri , astrophys .
j. * 492 * , l59 ( 1998 ) [ astro - ph/9709085 ] .
l. stella and m. vietri , phys .
* 82 * , 17 ( 1999 ) [ astro - ph/9812124 ] .
l. stella , m. vietri and s. morsink , astrophys .
j. * 524 * , l63 ( 1999 ) [ astro - ph/9907346 ] . c. bambi , eur .
j. c * 75 * , 162 ( 2015 ) .
c. bambi and s. nampalliwar , arxiv:1604.02643 .
z. stuchlik and a. kotrlova , gen .
. grav . * 41 * , 1305 ( 2009 ) .
t. johannsen and d. psaltis , astrophys .
j. * 726 * , 11 ( 2011 ) [ arxiv:1010.1000 [ astro-ph.he ] ] . | = 0.6 cm * abstract * we extend firstly the regular phantom black hole solution to a slowly rotating black hole case and find that the phantom field depresses the angular velocity of the event horizon and suppresses the super - radiation of black hole .
we also probe the dependence of quasi - periodic oscillations frequencies in relativistic precession model on the phantom parameter . with the observation data of gro j1655 - 40
, we make a constraint on the parameters of the regular and slowly rotating phantom black hole .
our results show that although the best - fit value of the phantom parameter @xmath0 is small , the allowed value of @xmath0 in the @xmath1 region is @xmath2 , which means that the phantom theoretical model can not be excluded by the constraint from quasi - periodic oscillations with the observation data of gro j1655 - 40 .
= 0.8 cm |
quantum phases are the most beautiful example of quantum physics and essential to understand physics in any material . for example ,
berry s phase , the accumulated phase in the eigenfunction acquired by evolving the quantum system adiabatically around a closed loop in the parameter space of the hamiltonian @xcite , has been shown to be responsible for the aharonov - bohm effect @xcite , the half - integer quantum hall effect @xcite , etc .
another important example is the sign of the hopping matrix element ( or hopping integral ) @xmath4 of the hamiltonian between two non - equivalent localized orbitals @xmath5 and @xmath6 in a tight - binding formalism .
this phase , a fundamental quantity in determining the electronic structure of a system , is dictated by the characteristics of the atomistic interaction , e.g. , whether it is attractive or repulsive .
both of these are important to directly extract fundamental quantum electronic information on a variety of materials @xcite . in graphene
, the berry s phase is theoretically extracted from the spinor eigenstate , which are @xmath7 for single- and @xmath8 for double - layer graphene @xcite . from recent studies ,
the @xmath9 berry s phase in the commonly used form of the spinor states is , in fact , related to the pseudospin winding number @xmath10 of a particle as it travels in a loop in @xmath11-space which encloses the dirac point @xcite .
these values have been measured through magneto - transport experiments @xcite that are typically neither capable of measuring berry s phase of a specific electron bandstructure nor free from spin / valley degeneracy of the electron bandstructure of a system under study .
additionally , this method requires a strong magnetic field , which breaks time - reversal symmetry in graphene .
meanwhile , the signs of hopping integrals between non - equivalent orbitals for graphene ( graphite ) have only been determined by _ ab initio _
calculations @xcite , e.g. , using maximally localized wannier functions @xcite .
since the sign of hopping integral depends on the characteristics of the localized orbitals and the interaction between them , it is crucial in determining the electron bandstructure within a tight - binding formalism .
however , the absence of methodology has led to use the signs following the well - known slonczewski - weiss - mcclure model @xcite without experimental verification for the past few decades .
additionally , the sign of hopping integral between non - equivalent orbitals has never been determined experimentally for any material .
given the high momentum - resolving power of angle - resolved photoemission spectroscopy ( arpes ) , arpes is an ideal candidate to solve the above issues on the determination of quantum phases .
for example , the phase difference between the matrix elements describing two different optical transitions at the ( 110 ) surface of platinum was extracted from a combined study of a spin - resolved arpes experiment and a theoretical model @xcite .
also , arpes has been employed to study the characteristics of the spinor eigenstates in graphite @xcite and graphene @xcite , which revealed an interference effect between photo - excited electrons @xcite .
however , these theoretical studies , within a tight - binding formalism , have incorrectly treated the interaction hamiltonian , which is the key part in the photoemission process as it describes the interaction between photons and electrons .
moreover , it has not been clear how berry s phase enters in arpes intensity and the sign of hopping integral has only been speculated without any comparison with experiments @xcite , which naturally leads to incorrect values . here we report that arpes can indeed provide information on these quantum phases , e.g. , the berry s phase and the sign of hopping integral between non - equivalent orbitals . the phase factor in the spinor eigenstate of graphene
@xcite gives rise to strong intensity variation around a constant energy contour .
upon rotating light polarization by @xmath0 , we found that graphene with a berry s phase of @xmath1 ( @xmath2 1 for single- and @xmath2 2 for double - layer graphene ) exhibits the rotation of arpes intensity maxima by @xmath3 , which gives important advantages compared to the conventional magneto - transport method @xcite .
additionally , we found that full polarization - dependence of arpes signal reveals the sign of hopping integrals in both single- and double - layer graphene ( graphite can also be understood ) , e.g. , @xmath12 and @xmath13 , which is the first experimental determination of the sign of hopping integral between non - equivalent orbitals for any material by any method .
single- and double - layer graphene samples were grown epitaxially on @xmath10-doped 6@xmath14-sic(0001 ) surfaces by electron - beam heating , as detailed elsewhere @xcite .
an sic sample was mounted in a prep - chamber with a base pressure of 5@xmath1510@xmath16 torr to remove a thick oxide layer from the sample by heating at 600 @xmath17c for a few hours .
the clean sample was then transferred to a custom - designed chamber equipped with low - energy - electron microscopy ( leem ) with a base pressure of 2@xmath1510@xmath18 torr and heated to 1000 @xmath17c under si flux to improve the surface conditions for graphene growth .
the sample temperature was raised to 1400 @xmath17c or 1600 @xmath17c ( determined by an optical pyrometer ) to make single- or double - layer graphene , respectively
. m@xmath154@xmath19 m range .
( b ) reflectivity spectra for the three regions ( 1 , 2 , and 3 ) specified in ( a ) .
( c ) post - processed image of ( a ) showing the regions covered by single- , double- , and triple - layer graphene .
( d ) a histogram showing the fractions of single- , double- , and triple - layer graphene in our sample used for double - layer graphene measurements . ] the surface morphology was monitored _ in situ _ during the sample growth by leem at the national center for electron microscopy at lawrence berkeley national laboratory .
the thickness of fabricated graphene samples was determined by leem measurements performed at room temperature following the standard procedure @xcite .
in particular , the electron reflectivity versus kinetic energy curve varies significantly with the number of graphene layers providing position - dependent measurements on the number of graphene layers . a typical bright field image for double layer graphene is shown in fig .
[ fig1](a ) over 4 @xmath19m@xmath154 @xmath19 m range , recorded with electron beam of kinetic energy 3.5 ev denoted as the dashed line in fig .
[ fig1](b ) . in order to determine the number of graphene layers at each position , the electron reflectivity is plotted as a function of electron kinetic energy , as shown in fig .
[ fig1](b ) , where the number of dips is the same as the number of graphene layers .
regions 1 , 2 , and 3 in fig .
1(a ) show 1 , 2 , and 3 dips , respectively , corresponding to single- , double- , or triple - layer graphene , respectively .
these regions are painted in black , white , and grey , respectively , in fig .
[ fig1](c ) .
the fractions of regions in the sample covered by different numbers of graphene layers were determined from the areal fractions of differently colored regions in fig .
[ fig1](c ) .
in particular , we find that the double - layer graphene sample contains @xmath2074 % of double - layer and @xmath2022 % of single - layer graphene .
we have performed polarization - dependent arpes experiments on single- and double - layer graphene at 10 k using a photon energy of 50 ev at beam - lines 10.0.1 and 12.0.1 of advanced light source at lawrence berkeley national laboratory . in figs .
2(a ) and 2(b ) , we show the typical geometry of arpes experiments : a beam of monochromatized light with energy @xmath21 and polarization vector @xmath22 is incident on a sample , resulting in the emission of photoelectrons in all directions .
the polarization vector of light is referenced with respect to the sample normal . in the experiment presented here ,
two different geometries were employed as shown in figs .
[ fig2](a ) and [ fig2](b ) . in one geometry shown in fig .
[ fig2](a ) , the polarization of light is almost parallel to the _ x _ axis , while in the other shown in fig .
[ fig2](b ) to the _ y _ axis ; hence , we define these two geometries as x- and y - polarization , respectively .
these geometries have the advantage with respect to the conventional _ s_- and _
p_-polarizations used in previous studies @xcite , to measure the whole two - dimensional variation of the intensity maps around a singular ( dirac ) point and not just the intensity distributions along two characteristic lines in momentum space .
this aspect becomes particularly relevant for some experimental conditions , e.g. , photon energy and sample orientation ( i.e. , the mixture of light polarizations ) , when the intensity maps ( or initial electronic states ) are neither symmetric nor anti - symmetric with respect to the reflection plane . under this condition in fact , the conventional notations would not give appropriate information on the symmetry of the initial states .
figures [ fig2](c ) and [ fig2](d ) show the experimental photoelectron intensity maps at the fermi level , @xmath23 , versus the two - dimensional wavevector * k * for single - layer graphene , for the two polarization geometries . here
, @xmath23 is 0.4 ev above the dirac point energy , @xmath24 @xcite .
the main feature in the intensity maps of both geometries is an almost circular fermi surface centered at the k point as shown in figs .
[ fig2](c ) and [ fig2](d ) , as expected for a conical dispersion .
this is in good agreement with a recent polarization - dependent arpes study on single - layer graphene when using photons with energy lower than 52 ev @xcite .
surprisingly we find that the the angular intensity distribution is quite different for the two polarizations : for the x - polarization geometry , the minimum intensity position is in the first brillouin zone , whereas for the y - polarization geometry , the maximum intensity position is in the first brillouin zone , suggesting a @xmath7 rotation of the maximum intensity in the @xmath25-@xmath26 plane around the k point upon rotating the light polarization by @xmath0 , from x to y ( see the white arrows ) .
the exact rotation angle is extracted from the direct comparison between the raw momentum distribution curves ( mdcs ) in fig .
[ fig2](e ) and the angular dependence of the photoelectron intensity maps integrated over the radial direction around the * k * point in fig .
[ fig2](f ) .
there is an overall shift of the intensity maxima ( minima ) by @xmath27 upon changing the light polarization from x ( black solid line ) to y ( black dashed line ) , although the latter appears to be slightly shifted by @xmath28 with respect to @xmath7 . as we will show later ,
this is due to the presence of a finite polarization component along the @xmath25 direction in our experimental geometry .
we note that not only the angular position of the intensity maximum , but also the absolute value of it changes upon changing light polarization .
the maximum intensity ratio from experiments is x - polarized / y - polarized=21.4 .
however , this number itself is not very meaningful , because the measured arpes intensity is affected by the difference in the experimental geometries for x- and y - polarized lights ( the difference in the out - of - plane component of light polarization , photon flux per unit area , etc .
, which are the factors that can not be controlled to be the same in different experimental geometries ) . on the other hand
, the ratio from our theory that will be discussed later provides x - polarized / y - polarized=0.83 , assuming that the experimental parameters for two geometries are the same except for the in - plane light polarization . a similar study on bernal stacked double - layer graphene reveals a strong and complicated momentum- , band- , and polarization - dependence as shown in figs . [ fig3](a ) and [ fig3](b ) , that is qualitatively different from that of single - layer graphene .
like single - layer graphene , the double - layer sample is slightly @xmath10 doped @xcite , therefore , only three of the four @xmath7 bands are occupied and hence detectable with arpes as shown in the cartoon of fig .
[ fig3](a ) .
the most prominent feature is that , when the light polarization is changed from x to y , the maximum intensity positions around the k point in the @xmath25-@xmath26 plane are rotated by @xmath29 ( see white arrows at @xmath23 ) .
this is in striking contrast with the single - layer case where the rotation is @xmath27 as seen from the raw mdcs in fig .
[ fig3](c ) and the photoelectron intensity maps integrated over the radial direction around the * k * point shown in fig .
[ fig3](d ) . due to trigonal warping effects
@xcite , however , the rotation for higher - energy states is not exactly @xmath0 as shown in figs .
[ fig3](a ) and [ fig3](b ) .
to the best of our knowledge , the only models in the literature describing the polarization - dependence of the arpes intensity in graphite @xcite and graphene @xcite are substantially different from our results .
previous studies , in fact , predict a small polarization - dependence @xcite and no polarization - dependence @xcite of the photoelectron intensity maps , respectively . therefore , to be able to reproduce our experimental findings and understand what lies behind this nontrivial polarization dependence , we have developed a new model .
in particular , we first consider the hamiltonian using the tight - binding model based on the @xmath30 orbital of each carbon atom using two parameters : @xmath31 and @xmath32 for the in - plane nearest - neighbor ( a - b or a@xmath33-b@xmath33 ) and the inter - layer vertical ( b - a@xmath33 ) hopping integrals , respectively , as schematically drawn in fig .
[ figs1](a ) .
the parameters @xmath31 and @xmath32 correspond to @xmath34 and @xmath35 , respectively , in the well - known slonczewski - weiss - mcclure ( swmc ) model @xcite . in our calculation , we have used @xmath36 ev and @xmath37 ev , which are the values in table ii of grneis _ et al_. @xcite , but we do not fix the signs of them . note that all four possible choices of the signs give exactly the same electron energy band structure within this two - parameter tight - binding model . , @xmath38 and @xmath39 are the three vectors connecting the in - plane nearest neighbor atoms where the inter - carbon distance @xmath40 , and the lattice constant is @xmath41 . the positions of the k and k@xmath33 points are @xmath42 and @xmath43 , respectively . ] with this setup , the tight - binding hamiltonian of a double - layer graphene for two - dimensional wave vector @xmath44 using a basis set composed of bloch sums of localized orbitals on each of the four sublattices ( a , b , a@xmath33 , and b@xmath33 ) reads @xmath45 here , @xmath46 with @xmath47 s defined as in fig .
[ figs1](a ) , and @xmath48 @xmath49 etc .
we note that often the * k * dependence of the basis function is suppressed in the spinor notation for simplicity . in eq .
( [ eq : h_bi_0 ] ) , we have considered a phase difference @xmath50 arising from the finite inter - layer distance _ d _ and the perpendicular component of electron wave vector @xmath51 . here , @xmath52 is not part of the bloch momentum , but the @xmath53 component of the photoelectron wave vector , which is determined by the photon energy .
this quantity plays a crucial role in determining the arpes intensity of double - layer graphene as will be discussed later and also of multi - layer graphene as previously reported @xcite .
the additional interaction hamiltonian coupling to electromagnetic waves of wavevector * q * for a double - layer graphene @xmath54 can be obtained by using the velocity operator @xmath55/i\hbar$ ] , where @xmath56 in the * k*-representation and @xmath57 is the planck s constant , as @xmath58 @xcite [ _ e _ is the charge of an electron , _ c _ is the speed of light , and the external vector potential is given by @xmath59 @xcite , i.e. , @xmath60 the transition matrix elements in eq .
( [ eq : h_bi_int ] ) are those taken between basis functions of bloch sums of @xmath30 orbitals of wavevectors @xmath61 and * k*. equation ( [ eq : h_bi_int ] ) is valid when @xmath62 is much larger than the distance between the nearest - neighbor atoms @xmath63 , i.e. , when the variation in @xmath64 over a length scale of _ b _ is much smaller than @xmath64 itself .
we shall eventually take the @xmath65 limit in our discussion because the momentum of light is negligible for photon energies considered . for single - layer graphene , performing a similar type of analysis
, we obtain @xmath66 and @xmath67 @xmath68 is critical to explain the polarization dependence of @xmath69 , because it describes the interaction between electrons and photons .
the lack of polarization dependence in previous studies @xcite is indeed due to the way in which @xmath68 is incorrectly treated . in one case @xcite ,
the light interaction is completely neglected by setting @xmath70 , while in the earlier study on graphite @xcite , the velocity operator @xmath71 is replaced by the momentum @xmath72 , where @xmath57 is the planck s constant and @xmath73 the free - electron mass .
this replacement works @xcite only when the hamiltonian is local , whereas a tight - binding hamiltonian , which has been used in the previous studies @xcite as well as our study , is intrinsically non local .
the experimental finding of a strong polarization dependence of @xmath69 in figs .
[ fig2 ] and [ fig3 ] clearly shows the need for a more complete theoretical treatment .
we have developed a theory using the widely adopted tight binding model with one @xmath74-like localized orbital per carbon atom , but employing the appropriate interaction hamiltonian with the velocity operator .
a very good agreement between our model and the experimental results is obtained for all polarizations and for both single- and double - layer graphene , when we compare fig .
[ fig2 ] with fig .
[ fig4 ] and fig .
[ fig3 ] with fig .
[ fig5 ] as will be discussed later . in order to understand what lies behind the observed non - trivial and unexpected wavevector - dependent photoelectron intensity @xmath75 ,
we need to calculate the absolute square of the transition matrix element @xmath76 , where @xmath77 is a single- or double - layer graphene eigenstate with @xmath78 the band index , @xmath79 is the plane - wave final state projected onto the @xmath30 orbitals of graphene [ both @xmath77 and @xmath80 are expressed using the basis set of bloch sums of localized @xmath74 orbitals at sublattices a and b in fig .
[ fig4](a ) ] and @xmath81 @xcite , which should not be neglected in photoemission process @xcite .
the use of a projection of the final plane - wave state onto the bloch sum , which when using plane - waves basis is effectively composed of multiple plane - waves @xcite , allow to explain the non - trivial polarization dependence of the arpes intensity distribution in figs .
2(c ) and 2(d ) .
since the polarization of * a * is in the _ x - y _ plane , the projection of @xmath82 onto the @xmath83-states of graphene will result in zero contribution to the transition matrix elements and hence are neglected in this analysis . for simplicity of notation , and without any loss of generality , in the rest of this section , we shall take the limit of @xmath65 and denote @xmath84 and @xmath85 . for single - layer graphene
, we may use @xmath86 and for double - layer graphene @xmath87 for photons ( with energy @xmath88 ev ) used in the experiment , @xmath52 of the planewave final state is much larger than the variation of two - dimensional bloch wavevector * k * around a single dirac point , leading to only a small variation in @xmath52 with any change in * k * around a single dirac point . for light with a nonzero polarization component along the _ z _ direction
, it will only give rise to an additive isotropic term to the photoelectron intensity that is independent of the in - plane polarization of the light .
now , let us consider the case where * k * is very close to the dirac point k as shown in fig .
[ figs1](b ) , and define @xmath89 ( @xmath90 ) . according to eq .
( [ eq : g_def ] ) , @xmath91 where _ a _ is the lattice parameter . for single - layer graphene , therefore , @xmath92 and @xmath93 where @xmath94 and @xmath95 are the pauli matrices . the energy eigenvalue and wavefunction of eq .
( [ eq : h_mono_0_k ] ) are given by @xmath96 and @xmath97 respectively , when @xmath98 is the angle between * q * and the @xmath99 direction .
using eqs .
( [ eq : pw_mono ] ) , ( [ eq : h_mono_int_k ] ) , and ( [ eq : wfn_mono ] ) , the transition matrix element is given for light polarized along the _ x _ direction by @xmath100 it follows that for @xmath101 ( states above the dirac point energy ) , @xmath102 and @xmath103 with @xmath104 and @xmath105 , respectively .
similarly , for light polarized along the _ y _ direction , the transition matrix element is given by @xmath106 it follows that for @xmath101 ( states above the dirac point energy ) , @xmath107 and @xmath108 with @xmath104 and @xmath105 , respectively . in both cases
, we can explain the rotation of the intensity maximum in the photoelectron intensity map around the k point by @xmath7 upon the change from x- to y - polarized light .
comparing eqs .
( [ eq : mtx2_mono_1 ] ) and ( [ eq : mtx2_mono_2 ] ) with eqs .
( [ eq : mty2_mono_1 ] ) and ( [ eq : mty2_mono_2 ] ) , irrespective of the sign of @xmath31 , the maxima of the photoemission intensity map of single - layer graphene is rotated by @xmath7 when the light polarization is rotated by @xmath0 , in agreement with experiment shown in fig .
moreover , the theoretical results with @xmath105 shown in figs .
[ fig4](a ) and [ fig4](b ) agree with the measured angular spectral intensity shown in figs .
[ fig2](c ) and [ fig2](d ) ; especially , the choice of @xmath36 ev ( fitted to previous experiments @xcite ) reproduces quite well the salient features in the experimentally measured intensity maps .
this is even more clear from the angular dependence of theoretical photoelectron intensity drawn with the red solid lines compared to experimental results drawn with the black solid lines in figs .
[ fig4](c ) and [ fig4](d ) .
note that experimental intensity maximum for y - polarized light shows additional shift by @xmath28 in fig .
[ fig2](f ) .
this additional shift is well understood by a finite polarization component along the @xmath25 direction .
when we assume the actual light polarization used in the experiment shown in fig .
[ fig2](b ) , the theoretical intensity exactly matches with the experimental result . on the other hand ,
when we assume the ideal y - polarization , the intensity maximum appears at @xmath7 as shown with the blue dashed line in in fig .
[ fig4](d ) .
therefore , we determine through experiment that the inter - orbital hopping matrix element @xmath31 between two in - plane nearest - neighbor carbon @xmath30 orbitals is negative ; we will come back to this point later .
recent theoretical study on the matrix element in single - layer graphene @xcite has found that , in order to describe the matrix element for y - polarized light from first - principles calculations using plane - wave basis , one needs multiple plane - wave components for the final photo - emitted electron state .
since we consider a projection of the final state onto the bloch sum , which when using the plane - waves basis is effectively composed of multiple plane - waves @xcite , our approach can explain the non - trivial polarization dependence of the arpes intensity distribution in figs .
2(c ) and 2(d ) .
additionally , our result obtained by using photons with energy 50 ev is in good agreement with the recent study @xcite based on first - principles calculations using photons with energy lower than 52 ev .
this suggests that , at photon energy below 52 ev and only when the correct interaction hamiltonian is employed , the projection of the final state onto the tight - binding bloch sum may describe the true final state qualitatively .
however , since our theoretical framework is within tight - binding formulation , and not from first - principles calculations with plane - wave basis , convergence tests with respect to the number of plane - waves and the character of the true final state are beyond the scope of this paper and , in fact , has been done in a recent study @xcite . in the case of double - layer graphene , for simplicity of the analysis
, we confine our discussion to the inner parabolic bands ( bands 1 and 2 in the cartoon in fig .
[ fig3 ] ) , although we considered all the four bands in our theoretical calculations shown in figs .
[ fig5](a ) and [ fig5](b ) . in double - layer graphene , for electronic states with energy @xmath109 of the hamiltonian in eq .
( [ eq : h_bi_0 ] ) , the energy and wavefunction are given by @xmath110 where @xmath111 respectively @xcite .
the phase difference @xmath112 between the two graphene layers in eq .
( [ eq : h_bi_0 ] ) appears here as well .
using eqs .
( [ eq : h_bi_int ] ) , ( [ eq : pw_bi ] ) , and ( [ eq : wfn_bi ] ) , the transition matrix element is given for light polarized along the _ x _ direction by @xmath113 it follows that for @xmath101 ( states above the dirac point energy ) , @xmath114 and @xmath115 with @xmath116 and @xmath117 , respectively .
the perpendicular component of the wavevector @xmath52 reads @xcite , @xmath118 where @xmath119 is the kinetic energy of the photo - electron and @xmath120 the inner potential .
note here that * k * is a two - dimensional bloch wavevector , i , e . , it has no out - of - plane component .
the inner potential has been measured for graphite by analyzing the energy dispersion at normal emission ( i.e. , @xmath121 ) @xcite .
according to eq .
( [ eq : k_z ] ) , @xmath122 .
similarly , for light polarized in the _ y _ direction , the transition matrix element for a double - layer graphene is given by @xmath123 it follows that for @xmath101 ( states above the dirac point energy ) , @xmath124 and @xmath125 with @xmath116 and @xmath117 , respectively . in both cases
, we can explain the rotation of the intensity maximum in the photoelectron intensity map around the k point by @xmath0 upon the change from x- to y - polarized light .
comparing eqs .
( [ eq : mtx2_bi_1 ] ) and ( [ eq : mtx2_bi_2 ] ) with eqs .
( [ eq : mty2_bi_1 ] ) and ( [ eq : mty2_bi_2 ] ) , irrespective of the sign of @xmath32 , the maxima of the photoemission intensity map of a double - layer graphene is rotated by @xmath0 when the light polarization is rotated by @xmath0 , in agreement with experiment shown in fig .
if we assume that @xmath116 , which is qualitatively in agreement with experiment shown in figs .
[ fig3](a ) and [ fig3](b ) , @xmath126 and @xmath127 for the upper band ( @xmath101 ) .
therefore , we have determined that the vertical inter - layer hopping matrix element @xmath32 between two carbon @xmath128 orbitals sitting on top of each other is positive ; we will come back to this point later .
although this model can overall account for the experimental data of double - layer graphene , there is a discrepancy : the measured photoemission intensity at @xmath129 ev along the @xmath130 direction for the x - polarization geometry is finite , whereas the theory predicts this value to vanish as shown in the @xmath131 ev map of the ideal dg in fig .
[ fig5](a ) .
we believe that this discrepancy arises from the finite size of the light beam spot ( @xmath2080@xmath1540 @xmath19m@xmath132 ) which covers not only the double - layer graphene portion but also some single - layer graphene , as discussed in fig .
in fact , double - layer graphene samples inevitably contain a finite amount of single - layer graphene @xcite .
the fraction of single - layer coverage can be obtained by leem measurements @xcite : from this analysis shown in fig .
[ fig1 ] , we find that the double - layer graphene sample used here contains @xmath2074% double - layer and @xmath2022% single - layer graphene . when the theoretical photoelectron intensity maps of single- and double - layer graphene are correspondingly weighted and averaged , the results denoted by sg@xmath133dg in figs .
[ fig5](a ) and [ fig5](b ) are in excellent agreement with the experimental data shown in figs .
[ fig3](a ) and [ fig3](b ) .
this is even more clear from the angular dependence of the photoelectron intensity maps shown in the red solid lines in fig .
[ fig5](c ) and [ fig5](d ) .
note that the presence of single - layer graphene does not affect the results for the y - polarization , which is obvious when we compare the red solid and green dashed lines in fig .
[ fig5](d ) , because both the intensity maxima of single- and double - layer graphene occur near @xmath134 = @xmath7 .
we have shown that , when the light polarization is rotated by @xmath0 , the maximum intensity position in @xmath135 in the @xmath25-@xmath26 plane of single- and double - layer graphene is rotated by @xmath3 , where @xmath136 and @xmath137 for single- and double - layer graphene , respectively .
the physical meaning of these rotations , whose origins rests on the phase factor @xmath138 of the sublattice amplitude of the wavefunctions @xcite , becomes clear upon a complete circulation of * q * around the dirac point k ( @xmath139 ) , which directly gives a berry s phase @xmath140 @xcite .
recently the berry s phase interpretation of @xmath10 @xcite has been given by a different interpretation in terms of pseudospin winding number @xcite .
the fact that @xmath141 enters in @xmath135 in the form of either @xmath142 or @xmath143 with some constant @xmath144 , demonstrates that the matrix elements directly contain information on the berry s phase .
the rotation of light polarization gives an additional phase @xmath0 to the phase factor , i.e. , @xmath145 .
thus , the photoelectron intensity @xmath146 is modified accordingly from @xmath147 or @xmath148 upon the rotation of light polarization by @xmath0 , i.e. , the @xmath3 rotation of @xmath135 .
this prediction is exactly realized in our experimental results .
the power of our method is that it can be extended in a straightforward way to other materials with berry s phase @xmath149 ( not necessarily @xmath7 or @xmath8 ) . in this case , the photoemission intensities for x- and y - polarization geometries are given by @xmath150 and @xmath151 , where @xmath144 is a system - dependent constant .
the important feature is that the rotation of light polarization by @xmath0 results in a rotation of intensity maxima by @xmath3 for a system with @xmath149 regardless of the constant @xmath144 .
thus , we have demonstrated here that the berry s phase can directly be measured from polarization - dependent arpes .
unlike methods based on magneto - transport experiments @xcite , our new method has three important advantages .
( i ) the berry s phase of a specific electronic band can be measured , because arpes has the angle - resolving power and also because one can set up a tight - binding hamiltonian focussing on only the electronic states of interest : those two results can directly be compared with each other .
( ii ) due to the angle - resolving power , the measured result is free from valley degeneracy for the case of graphene .
( iii ) our method does not need electric gating , which is essential for the transport measurements .
another important finding of our study is that we can directly extract , for the first time , the sign and the absolute magnitude of the inter - orbital hopping integrals ( iohis ) between non - equivalent localized orbitals of a tight - binding hamiltonian from experiment . until now ,
in fact , the sign determination of an iohi has resorted no to any experimental method , but to _ ab initio _ calculations , e.g. , using maximally localized wannier functions @xcite . in order to understand an ambiguity related with the experimental sign determination
, we take the simplest one - dimensional example , and extend the discussion to a more complicated tight - binding model of graphitic systems than the one described previously .
we consider simple tight - binding models having _ s_-like localized states , the values of which are all positive in real space ( we can arbitrarily set this gauge without losing generality . ) if there is only one localized orbital per unit cell in a one - dimensional tight - binding model as drawn in fig .
[ fig6](a ) , the energy bandstructure varies with the sign of the iohi @xmath152 as shown in fig .
[ fig6](b ) .
hence , the sign of iohis between `` equivalent '' orbitals can always be trivially determined @xcite . .
( b ) calculated electron energy band structure of the system depicted in ( a ) with different choices for the sign of @xmath152 .
( c ) schematic of a one - dimensional crystal having two _ nonequivalent _
_ s_-type orbitals per unit cell ( i.e. , having different on - site energies ) .
we assume that all the distances between the centers of the nearest neighbor orbitals are the same and hence the corresponding hopping integrals , denoted by @xmath153 .
( d ) calculated electronic band structure of the system depicted in * c * with different choices for the sign of @xmath153 . ] however , if we consider the case where there are two non - equivalent localized orbitals @xmath154 and @xmath155 whose value in real space is positive and if we denote the iohi between the nearest neighboring orbitals by @xmath153 as drawn in fig .
[ fig6](c ) , the actual band structure is invariant when we change the sign of @xmath153 as shown in fig .
[ fig6](d ) . therefore ,
even when the actual electronic band structure is empirically determined , the sign of @xmath153 can not be determined . in general , an empirical tight - binding model with more than one orbital per unit cell has this sign ambiguity problem for iohis between `` non - equivalent '' orbitals , thus preventing an experimental measurement of iohi from just the energy band dispersions .
we could understand this degeneracy as follows .
if we denote the bloch sums of the two localized orbitals by @xmath156 and @xmath157 , the tight - binding hamiltonian @xmath14 using this basis set reads @xmath158 note here that the matrix elements @xmath159 , in which @xmath160 , involve not only the on - site or nearest - neighbor hopping processes but also all the other possible hopping processes .
now , it is obvious that the following hamiltonian @xmath161 has exactly the same eigenvalues as @xmath162 : @xmath163 what we have done by going from @xmath162 to @xmath161 is to change the sign of the iohi between the two non - equivalent localized orbitals . in fact
, the two matrices @xmath162 and @xmath161 are related by a unitary transform , which does not change the eigenvalues of a matrix , @xmath164 with @xmath165 on the other hand , if we change the signs of the diagonal matrix elements @xmath166 , in which @xmath167 , we get a different eigenvalue spectrum .
thus , there is no ambiguity in the sign of the iohi between equivalent orbitals .
this simple example illustrates that one can not determine the signs of the iohis between non - equivalent orbitals just by looking at the measured electron energy bandstructure .
the tight - binding model for double - layer graphene that we used in our calculations is based on the @xmath74 orbitals of carbon atoms with two parameters : @xmath31 and @xmath32 for the nearest - neighbor in - plane and the vertical inter - layer hopping integrals , respectively .
the parameters @xmath31 and @xmath32 correspond to @xmath34 and @xmath35 in the well - known slonczewski - weiss - mcclure model , respectively @xcite , and we have used @xmath168 ev and @xmath169 ev @xcite .
the photoemission intensity map in the @xmath25-@xmath26 plane is strongly dependent on the signs of both @xmath31 and @xmath32 as shown in fig .
because the four different choices of the signs produce exactly the same electronic band structure , it has not been known from experiments which choice of the signs of the iohis is physically correct , although the absolute values have been experimentally estimated @xcite .
this sign - ambiguity problem still exists even when we include more complicated hopping processes in the model , especially the non - vertical inter - layer hopping integrals @xmath170 and @xmath171 , there still exist unitary transforms that leave the energy eigenvalues unchanged . in a tight - binding model having four hopping integrals ( @xmath172 , @xmath35 , @xmath170 , and @xmath171 ) , the following four different sets of parameters give exactly the same electron energy bandstructure : ( @xmath172 , @xmath35 , @xmath170 , @xmath171 ) , ( @xmath172 , @xmath173 , @xmath174 , @xmath175 ) , ( @xmath34 , @xmath35 , @xmath170 , @xmath175 ) , and ( @xmath34 , @xmath173 , @xmath174 , @xmath171 ) , assuming that the first set is composed of the values currently accepted and used when the swmc model is considered . in principle , there should be eight different sets of parameters giving the same energy bandstructure ; however , from our knowledge that the nearest - neighbor intralayer hopping integrals in different graphene layers are the same , we have reduced the number of candidates to four .
this is also the reason why we considered only the four cases in fig .
[ fig7 ] .
we have shown that the choice of @xmath105 and @xmath116 , i.e. , @xmath12 and @xmath13 in fig .
[ fig7](c ) , reproduces the experimental results shown in figs .
[ fig3](a ) and [ fig3](b ) , thus experimentally determined the signs of these iohis .
the signs of the iohis used for graphite in the conventional model @xcite is indeed correct .
previous theoretical study has also tried to determine the signs @xcite , but inappropriate theoretical approaches ( as previously mentioned ) and the lack of full polarization - dependent experimental data have led to incorrect speculations , which do not agree with the conventional model @xcite as well as our results .
our method can generally be used to determine the sign of the hopping integrals in complicated materials such as cuprates as well as in simple materials such as gallium arsenide and one - dimensional crystals having two atoms per unit cell .
we have shown that arpes can be used as a powerful tool to directly measure quantum phases such as the berry s phase of a specific electronic band in graphene with advantages compared to the interference type of measurements @xcite which do not give any information on the band - specific berry s phase , and the sign of the hopping integral between non - equivalent orbitals , never measured for any material before .
the experimental and theoretical procedures developed here can be applied in studying the electronic , transport , and quantum interference properties of a huge variety of materials .
we gratefully acknowledge d .- h .
lee , j. graf , c. m. jozwiak , s. y. zhou and h. zhai for helpful discussions .
this work was supported by the director , office of science , office of basic energy sciences , materials sciences and engineering division , of the u.s .
department of energy under contract no .
de - ac02 - 05ch11231
. 1 a. bohm , a. mostafazadeh , h. koizumi , q. niu , and j. zwanziger _ the geometric phase in quantum systems _
( springer - verlag , berlin heidelberg , 2003 ) . y. aharonov and d. bohm , _ phys
. rev _ * 115 * , 485 ( 1959 ) .
e. mccann and v. i. falko , _ phys .
lett . _ * 96 * , 086805 ( 2006 ) .
y. zhang , y. -w .
tan , h. l. stormer , and p. kim , _ nature _ * 438 * , 201 ( 2005 ) .
k. s. novoselov , e. mccann , s. v. morozov , v. i. falko , m. i. katsnelson , u. zeitler , d. jiang , f. schedin , and a. k. geim , _ nature phys . _
* 2 * , 177 ( 2006 ) .
y. taguchi , y. oohara , h. yoshizawa , n. nagaosa , and y. tokura , _ science _ * 291 * , 2573 ( 2001 ) , and references therein . s. murakami and n. nagaosa , _ phys .
lett . _ * 90 * , 057002 ( 2003 ) .
d. hsieh _ et al .
_ , _ science _ * 323 * , 919 ( 2009 ) .
a. grneis _
_ , _ phys .
b _ * 78 * , 205425 ( 2008 ) .
n. marzari and d. vanderbilt , _ phys . rev .
b _ * 56 * , 12847 ( 1997 ) .
j. c. slonczewski and p. r. weiss , _ phys .
* 109 * , 272 ( 1958 ) .
j. w. mcclure , _ phys .
rev . _ * 108 * , 612 ( 1957 ) .
_ , _ surf .
sci . _ * 416 * , 396 ( 1998 ) .
e. l. shirley , l. j. terminello , a. santoni , and f. j. himpsel , _ phys .
b _ * 51 * , 13614 ( 1995 ) .
m. mucha - kruczyski _ et al .
_ , _ phys .
b _ * 77 * , 195403 ( 2008 ) .
e. rollings _ et al .
_ , _ j. phys
solids _ * 67 * , 2172 ( 2006 ) .
d. a. siegel _
* 93 * , 243119 ( 2008 ) .
_ , _ new j. phys . _ * 10 * , 023034 ( 2008 ) .
d. pescia , a. r. law , m. t. johnson , and h. p. hughes , _ solid state commun . _
* 56 * , 809 ( 1985 ) .
k. c. prince , m. surman , th .
lindner , and a. m. bradshaw , _ solid state commun . _ * 59 * , 71 ( 1986 )
. s. y. zhou _
_ , _ nature mater .
_ * 6 * , 770 ( 2007 ) .
seyller , k. v. emtsev , f. speck , k. -y .
gao , and l. ley , _ appl .
lett . _ * 88 * , 242103 ( 2006 ) .
i. gierz , j. henk , h. hchst , c. r. ast , and k. kern , _ phys .
b _ * 83 * , 121408(r ) ( 2011 ) . a. h. castro neto , f. guinea , n. m. r. peres , k. s. novoselov , and a. k. geim , _ rev .
* 81 * , 109 ( 2009 )
. t. ohta _
_ , _ phys .
* 98 * , 206802 ( 2007 ) .
p. y. yu and m. cardona , _ fundamentals of semiconductors : physics and materials properties _ ; springer - verlag : berlin heidelberg , 1999 .
a. f. starace _ phys .
rev . a _ * 3 * , 1242 ( 1971 ) . s.
ismail - beigi , e. k. chang , and s. g. louie , _ phys .
_ * 87 * , 087402 ( 2001 ) . c. -h
park and s. g. louie , _ nano lett . _
* 9 * , 1793 ( 2009 ) .
ando , t. _ j. phys .
_ * 2007 * , _ 76 _ , 104711 .
zhou , s. y. ; gweon , g. -h . ;
lanzara , a. _ ann .
_ * 2006 * , _ 321 _ , 1730 - 1746 . c. -h .
park and n. marzari , arxiv:1105.1159v4 .
m. z. hasan _ et al .
_ , _ phys .
* 92 * , 246402 ( 2004 ) .
w. w. toy , m. s. dresselhaus , and g. dresselhaus , _ phys .
b _ * 15 * , 4077 ( 1977 ) .
a. misu , e. e. mendez , and m. s. dresselhaus , _ j. phys
_ * 47 * , 199 ( 1979 ) .
s. a. werner , r. colella , a. w. overhauser , and c. f. eagen , _ phys .
lett . _ * 35 * , 1053 ( 1975 ) .
a. tomita and r. y. chiao , _ phys .
lett . _ * 57 * , 937 ( 1986 ) .
t. bitter and d. dubbers , _ phys .
_ * 59 * , 251 ( 1987 ) . | quantum phases provide us with important information for understanding the fundamental properties of a system .
however , the observation of quantum phases , such as berry s phase and the sign of the matrix element of the hamiltonian between two non - equivalent localized orbitals in a tight - binding formalism , has been challenged by the presence of other factors , e.g. , dynamic phases and spin / valley degeneracy , and the absence of methodology . here ,
we report a new way to directly access these quantum phases , through polarization - dependent angle - resolved photoemission spectroscopy ( arpes ) , using graphene as a prototypical two - dimensional material .
we show that the momentum- and polarization - dependent spectral intensity provides direct measurements of ( i ) the phase of the band wavefunction and ( ii ) the sign of matrix elements for non - equivalent orbitals . upon rotating light polarization by @xmath0
, we found that graphene with a berry s phase of @xmath1 ( @xmath2 1 for single- and @xmath2 2 for double - layer graphene for bloch wavefunction in the commonly used form ) exhibits the rotation of arpes intensity by @xmath3 , and that arpes signals reveal the signs of the matrix elements in both single- and double - layer graphene .
the method provides a new technique to directly extract fundamental quantum electronic information on a variety of materials . |
in figure 1 ( left ) , different central line indices are plotted against @xmath0 for our sample ( bedregal et al .
bright and faint s0s lay in two separate clumps , each one in opposite extremes of the @xmath0 range .
when models from @xcite are used to estimate stellar population parameters , the observed trends seem to be driven by age and @xmath1-element relative abundances .
this is opposite to what is usually found in e , where [ fe / h ] is an important driver of the correlations . by parametrising line indices using @xcite models we found that age can explain the observed slopes of h@xmath5-log(@xmath0 ) and fe - log(@xmath0 ) relations , while additional differences in @xmath1-element relative abundances would be required to explain the mg - log(@xmath0 ) trends . on the other hand
, [ fe / h ] by itself can not explain any of the observed index@xmath3-log(@xmath0 ) relations .
( left ) and @xmath6 ( right ) for 9 s0s in the fornax cluster .
continuous lines : best fits to the data .
dashed lines : best fits found by @xcite for es in fornax . triangle : fornax a , a merger remnant .
two dots with arrows : pressure supported systems , for which the deprojected azimuthal velocity was used as a lower limit of @xmath2.,title="fig : " ] ( left ) and @xmath6 ( right ) for 9 s0s in the fornax cluster .
continuous lines : best fits to the data .
dashed lines : best fits found by @xcite for es in fornax . triangle : fornax a , a merger remnant .
two dots with arrows : pressure supported systems , for which the deprojected azimuthal velocity was used as a lower limit of @xmath2.,title="fig : " ]
in figure 1 ( right ) we present the index@xmath3 vs. log(@xmath2 ) relations for our sample of s0s .
clear trends appear in all the panels .
it is interesting to notice that the standard deviation of the linear fits performed is smaller for almost all the indices in comparison to its counterpart from the index@xmath3-log(@xmath0 ) fits . assuming that mass is the fundamental physical parameter governing the properties of these galaxies , the improvement in the fits may be interpreted in the sense that @xmath2 is a better estimator of the dynamical mass than @xmath0 . to test this hypothesis ,
the mass of these galaxies was parametrised as a function of @xmath8 and @xmath2 in order to compare how central indices correlate with both possible descriptions . @xmath9 and @xmath10 are the disk scale and effective radius of the bulge respectivelly and the mean @xmath8 was estimated within @xmath10 ( see figure 2 ) .
( left ) and from @xmath2 ( right ) .
open symbols : two pressure supported systems ( arrows as fig 1 ) .
open stars : dynamical mass of these two galaxies estimated from @xmath8 by assuming an isothermal sphere in hydrostatic equilibrium.,title="fig : " ] ( left ) and from @xmath2 ( right ) .
open symbols : two pressure supported systems ( arrows as fig 1 ) .
open stars : dynamical mass of these two galaxies estimated from @xmath8 by assuming an isothermal sphere in hydrostatic equilibrium.,title="fig : " ] clearly , both @xmath8 and @xmath2 are tracing the mass of these systems . however ,
when mass is estimated using @xmath8 ( figure 2 , left panel ) , it is underestimated in rotationally supported systems with respect to pressure supported ones . on the other hand , the right panel of figure 2 shows that galaxies follow a common sequence when @xmath2 is used .
the tighter trends in this figure agree with the idea that @xmath2 is a better tracer of dynamical mass than @xmath8 for this sample of galaxies . | based on fors2-vlt long - slit spectroscopy , the analysis of the central absorption line indices of 9 s0 galaxies in the fornax cluster is presented .
central indices correlate with central velocity dispersions ( @xmath0 ) as observed in ellipticals ( e ) .
however , the stellar population properties of these s0s indicates that the observed trends are produced by relative differences in age and @xmath1-element abundances and not in metallicity ( [ fe / h ] ) as previous studies have found in e galaxies .
the observed scatter in the line indices versus @xmath0 relations can be partially explained by the rotationally - supported nature of many of these systems .
the presence of tighter line indices vs. maximum ( circular ) rotational velocity ( @xmath2 ) relations confirms this statement .
it was also confirmed that the dynamical mass is the driving physical property of all these correlations and in our fornax s0s it has to be estimated assuming rotational support . |
topological defects constitues an important class of solutions in field - theoretical models with degenerate vacua .
they play very important role in several branches of physics .
let us mention here field - theoretical cosmology and the cosmic strings hypothesis ( see @xcite , @xcite , @xcite ) , dynamics of superconductors , superfluids and liquid crystals in condensed matter physics ( see @xcite - @xcite ) as well as a flux tube in qcd .
+ this short and incomplete list shows the necessity of having effective computing methods to study the dynamics of topological defects . in spite of the increasing development of mathematical techniques to solve nonlinear equations , exact solutions
seem not to be the rule and numerical methods have been the most common approach to study properties of topological defects
. therefore analytical and perturbative methods are of great interest and importance .
in this paper we study domain wall type solution in the @xmath2 model . in paper
@xcite excited domain wall of this kind was considered and the radiation emitted from the domain wall was found . in our approach we further develop the method of a so - called polynomial approximation ( see @xcite , @xcite ) , which is used to construct the domain wall type solution with time - dependent thickness .
we show how to compute corrections to the polynomial solution . the corrections consist of two parts : the static one and the time - dependent one .
the time - dependent part of the correction contains radiation emitted by domain wall with oscillating thickness .
the plan of our paper is the following . in the next section we present the model . in section [ pa ]
we derive the time - dependent planar domain wall solution in the polynomial approximation .
section [ corr ] is devoted to the detailed analysis of this solution .
we present the method of finding correction to the polynomial solution . in sections
[ static ] and [ dyn ] we calculate static and time - dependent part of this correction . in section [ back ]
we analyze the backreaction of the radiation emitted on the domain wall . in section [ sum ]
we shortly summarize the main points of our work .
we consider the model with single , scalar , real - valued field @xmath3 , defined by the action : @xmath4 , \label{lagr}\ ] ] where @xmath5 and @xmath6 , @xmath7 are positive constants .
the corresponding equation of motion for the field @xmath8 has the form : @xmath9 the energy functional for the model is given by the formula : @xmath10 = \frac{1}{2 } \int d^{3 } \vec{x } [ \partial_{\mu } \phi \partial^{\mu } \phi + \lambda ( \phi^{2}-v^{2})^{2 } ] .
\label{energia}\ ] ] it is convenient to rescale the space - time coordinates and the scalar field as follows : @xmath11 where @xmath12 .
the new variables are dimensionless .
the vacuum values of @xmath13 in the considered model are equal to @xmath14 .
configuration of the field which smoothly interpolates between these two vacua is called the domain wall .
our goal is to construct the domain wall configuration , localised on the @xmath15 plane , with the time - dependent width .
the planar domain wall distinguishes the direction perpendicular to the wall plane , given in our case by the coordinate lines of @xmath16 .
as we consider the configuration with @xmath13 independent of the coordinates @xmath17 and @xmath18 , we can restrict our approach to the @xmath19 dimensional model .
equation ( [ rr ] ) then takes the form : @xmath20 static solutions of this equation are well - known .
they have the form : @xmath21
in this section we construct approximate domain wall type solution of eq .
( [ prr ] ) with time - dependent width . in order to realize this we use the method of a polynomial approximation , whose detailed description is given in @xcite .
the basic idea of this approach is to approximate the scalar field inside the domain wall by the polynomial in the variable @xmath16 with time - dependent coefficients .
thus , inside an interval @xmath22 $ ] ( @xmath23 and @xmath24 are positive ) we have : @xmath25 the domain wall solution is characterized by the fact , that for sufficiently large @xmath26 the field approaches its vacuum values @xmath14 : @xmath27 it is possible ( see @xcite ) to adopt more accurate asymptotics for large @xmath26 , with exponential correction @xmath28 to the vacuum values @xmath29 .
for simplicity of further calculations we use expressions ( [ proz ] ) .
+ we can tune accuracy of our approximation changing degree of the polynomial ( [ ansatz ] ) .
one can easily notice , that the cubic ansatz presented above is the simplest , nontrivial choice .
the ansatz ( [ ansatz ] ) should be smoothly matched with the vacuum solutions at @xmath30 and @xmath31 .
the matching conditions follow in a standard manner from eq .
( [ prr ] ) .
one integrates eq .
( [ prr ] ) over @xmath16 in arbitrary small intervals @xmath32 $ ] ; @xmath33 $ ] and lets @xmath34 .
this implies : @xmath35 from eq .
( [ proz ] ) one gets : @xmath36 the ansatz ( [ ansatz ] ) with matching conditions ( [ zsz1 ] ) - ( [ zsz2 ] ) allows us to find solution of eq .
( [ prr ] ) in the proposed form . inserting the expansion ( [ ansatz ] ) into the matching conditions we get , analogously to the case of cylindrical domain wall discussed in @xcite , the following conditions for functions @xmath37 , @xmath38 , @xmath39 and parameters @xmath23 , @xmath24 : @xmath40 then we insert the expansion ( [ ansatz ] ) into ( [ prr ] ) and equate to zero coefficients in front of successive powers of @xmath16 . after some easy algebra , considering conditions ( [ wr ] ) we get the following equation : @xmath41 function @xmath39 is related to @xmath37 ( see formula ( [ wr ] ) ) and the last step in our approach is to solve eq . ( [ rna ] ) .
we construct an approximate solution of eq .
( [ rna ] ) , which is convenient for the further analysis . with variables redefined as follows : @xmath42 eq .
( [ rna ] ) takes the form : @xmath43 where now dot denotes the derivative with respect to @xmath44 . numerical analysis of eq .
( [ rna1 ] ) shows that there exist periodic solutions of two kinds .
periodic solutions of the first kind oscillate around static solution @xmath45 and take only positive values ; periodic solutions of the second kind oscillate around static solution @xmath46 and take only negative values .
we can easily analyse the oscillating solutions of eq .
( [ rna1 ] ) rewriting it in the form of the two equations of the first order by substitution @xmath47
. then periodic solutions are equivalent to closed trajectories in ( a , b ) configuration space .
@xmath48 , @xmath49 are the central critical points .
the ( a , b ) configuration space is presented in fig .
in the further part of our discussion we consider only the periodic solutions oscillating around @xmath45 , because of the connection between @xmath50 and the domain wall thickness parameter , @xmath23 .
when @xmath51 for all @xmath44 we have also @xmath52 , which is consistent with the interpretation of these parameters .
the periodic solutions oscillating around @xmath46 are symmetric ( in the sense that when one changes @xmath23 and @xmath24 by each other it has no influence on the dynamics of our system ) , so we do nt discuss this situation .
considering all these remarks we restrict our investigation to solutions oscillating with small amplitude around @xmath45 .
we can write : @xmath53 where @xmath54 is small , periodic function . inserting ( [ omega ] ) into eq .
( [ rna1 ] ) one gets : @xmath55 expression ( [ rna2 ] ) has the form of oscillator equation with nonlinear terms . our goal is to find its perturbative solution with initial condtions : @xmath56 , @xmath57 , which describes initially static , sqeezed wall .
one can solve the nonlinear oscillator equation of type ( [ rna2 ] ) using krylov - bogolubov method ( see @xcite).the general solution has the form : @xmath58 where functions @xmath59 can be calculated in a standard manner ( see @xcite ) with initial condition @xmath60 . in our case we find : @xmath61 - \frac{3 \delta^{2}}{4 } + \frac{\delta^{2}}{4 } \cos \left [ 2 \left ( 1+\frac{2 \delta^{2}}{16 } \right ) \tau \right ] + o(\delta^{3}),\ ] ] where @xmath62 is a small parameter .
thus we get solution @xmath37 : @xmath63 - \frac{3 \delta^{2}}{4 } + \frac{\delta^{2}}{4 } \cos \left [ 4 \left ( 1 + \frac{3 \delta^{2}}{16 } \right ) t \right ] \right ) + o ( \delta^{3 } ) .
\label{a_od_t}\ ] ] formula ( [ a_od_t ] ) agrees well with the numerical solutions of eq .
( [ rna ] ) .
+ in the last step we have to insert the solution ( [ a_od_t ] ) into the ansatz ( [ ansatz ] ) considering conditions ( [ wr ] ) .
the final expression for the field of the oscillating domain wall in the polynomial approximation has the form : @xmath64 \theta ( \xi_{0 } ( t ) - \mid \xi \mid ) \nonumber \\ & & + \theta ( \xi - \xi_{0 } ( t ) ) - \theta ( -\xi_{0 } ( t ) - \xi ) + o ( \delta^{3}),\end{aligned}\ ] ] where @xmath65 the function @xmath66 can be regarded as the half - width of the domain wall in the @xmath16 coordinate .
it is given by the formula : @xmath67 from eq .
( [ oa ] ) one can get period of oscillation of the domain wall @xmath68 : @xmath69 . in the case of linear oscillations around static solution
( one can regard this situation neglecting nonlinear terms in ( [ rna2 ] ) - it is admissible because @xmath70 is small by definition ) one gets @xmath71 and @xmath72 .
the formula ( [ wiel ] ) can be adopted to the special , static case of domain wall solution , taking @xmath73 .
it is equivalent to the @xmath48 central , critical point of eq .
( [ rna2 ] ) . the general solution @xmath74 can be then treated as a small oscillation ( with amplitude given by the parameter @xmath62 ) around the static solution @xmath75 : @xmath76 accordingly to eq .
( [ wiel ] ) initially the domain wall is a bit sqeezed .
then it oscillates with small amplitude around the static solution .
in this section we propose a simple extension of the pure polynomial solution obtained above .
formula ( [ wiel ] ) gives us the simplest , approximaty domain wall - type solution in our model .
one can study wider range of phenomena ( e.g. nontrivial asymptotics of the domain wall solution as well as radiation , which can be emitted by the oscillating domain wall ) considering higher - order corrections to this solution .
our goal is to calculate perturbatively corrections to the zeroth order polynomial solution .
let us denote it by @xmath77 .
thus , we can write the field satisfying eq .
( [ rr ] ) as : @xmath78 in order to find the correction @xmath77 let us insert expression ( [ p_pola ] ) into the field equation ( [ prr ] ) .
neglecting nonlinear terms in the field @xmath79 we get : @xmath80 \phi^{(1 ) } = - \partial_{\mu } \partial^{\mu } \phi^{(0 ) } + 2[\phi^{(0 ) 2 } - 1 ] \phi^{(0)}. \label{nrr}\ ] ] we expect that global character of the domain wall solution during time evolution wo nt change because of topological stability ( in the other words : the solution remains the planar domain wall type all the time , only small corrections can occur ) .
that s why the linearization in ( [ nrr ] ) may be done .
( [ nrr ] ) is a linear one with source term @xmath81 given by the formula : @xmath82 \phi^{0}. \label{zrodlo}\ ] ] as it was done for the polynomial solution @xmath83 we can split @xmath81 into statical and time - dependent part : @xmath84 from eq .
( [ wiel ] ) one gets : @xmath85 \theta(1 - \mid \xi \mid ) - \left ( \delta ' ( \xi - 1 ) + \delta ' ( \xi + 1 ) \right ) , \label{js}\ ] ] where prim denotes derivative with respect to @xmath86 , and : + + @xmath87 \theta ( \xi_{0}(t ) - \mid \xi \mid ) \nonumber \\ & & + \left [ \frac{27}{4 } \left ( \xi - \frac{1}{3 } \xi^{3 } \right ) + \xi^{3 } \right ] \left [ \theta ( \xi_{0}(t ) - \mid \xi \mid ) - \theta ( 1 - \mid \xi \mid ) \right ] \nonumber \\ & & + \left [ \frac{3}{2 } \delta \omega^{2 } \left ( \xi - \frac{1}{3 } \xi^{3 } \right ) \cos \omega t - 3 \delta \ , sgn \ , \xi ( 1 - 3 \xi^{2 } ) \cos \omega t \right ] \delta ( \xi_{0}(t ) - \mid \xi \mid ) \nonumber \\ & & + \frac{3}{2 } sgn \ , \xi ( 1-\xi^{2 } ) \left ( \delta ( \xi_{0 } ( t ) - \mid \xi \mid ) - \delta ( 1 - \mid \xi \mid)\right ) \nonumber \\ & & + \frac{3}{2 } \delta ( \xi - \xi^{3 } ) \cos \omega t \ ; \delta ' ( \xi_{0 } ( t ) - \mid \xi \mid ) \nonumber \\ & & - \frac{3}{2 } \left ( \xi - \frac{1}{3 } \xi^{3 } \right ) \left [ \delta ' ( \xi_{0 } ( t ) - \mid \xi \mid ) - \delta ' ( 1 - \mid \xi \mid ) \right ] \nonumber \\ & & + \delta \omega^{2 } \cos \omega t \left [ \delta ( \xi + \xi_{0 } ( t ) ) - \delta ( \xi - \xi_{0 } ( t ) ) \right ] \nonumber \\ & & - \left [ \delta ' ( \xi - \xi_{0 } ( t ) ) - \delta ' ( \xi - 1 ) \right ] - \left [ \delta ' ( \xi + \xi_{0 } ( t ) ) - \delta ' ( \xi + 1 ) \right ] . \label{jd}\end{aligned}\ ] ] eq . ( [ nrr ] )
is complicated due to unpleasant form of the polynomial solution given by eq .
( [ wiel ] ) . we can simplify the operator on the left - hand side of eq .
( [ nrr ] ) noticing , that @xmath83 can be split into statical and time - dependent part , and predominant component @xmath88 ( which is exact , statical solution of the initial field equation ) can be extracted from the static part as follows : @xmath89 inserting expression ( [ podz ] ) into the evolution equation and rewriting some terms on the right - hand side we get finally : @xmath90 \phi^{(1 ) } = j_{s}(\xi ) + j_{d } ( t , \xi ) } \nonumber \\ & & + 6 \left ( \phi^{(0)2}_{s } - \tanh^{2 } \xi
\right ) \phi^{(1 ) } \nonumber \\ & & + 6 \ , \delta \phi^{(0)}_{d } \left ( \delta \phi^{(0)}_{d } + 2 \ , \delta \phi^{(0)}_{s}+ 2 \tanh \xi \right ) \phi^{(1)}. \label{pert}\end{aligned}\ ] ]
the component in front of @xmath79 in the third term on the right - hand side is nonzero only on the border of the domain wall and quickly tends to zero when @xmath91 .
analogously , the component in front of @xmath92 in the fourth term is of order of the small parameter @xmath93 .
thus , we have the following , strongly suggested method of solving eq .
( [ pert ] ) : in the first step we put @xmath94 on the right - hand side of eq .
( [ pert ] ) and solve linear equation for @xmath79 with source term @xmath95 .
the solution obtained in this way can be inserted to the right - hand side of eq .
( [ pert ] ) , so in the second step we have to solve equation with the same linear operator but a new source term .
we find full solution as a result of such as iterating procedure .
nevertheless if components in front of @xmath79 on the right - hand side are small ( and that is our case ) , we can find predominant part of full solution in the first step . due to the form of source term ( which consist of static and time - dependent part ) it seems natural to split @xmath79 into two components : the static one and the time - dependent one .
they are denoted respectively by @xmath96 and @xmath97 . from eq .
( [ pert ] ) we get : @xmath98 for the static part of solution , and : @xmath99 \phi^{(1)}_{d } = j_{d}(t , \xi ) } \nonumber \\ & & + 6 \ , \delta \phi^{(0)}_{d } \left ( \delta \phi^{(0)}_{d } + 2 \ , \delta \phi^{(0)}_{s } + 2 \tanh \xi \right ) \left ( \phi^{(1)}_{d } + \phi^{(1)}_{s } \right ) \nonumber \\ & & + 6 \left ( \phi^{(0)2}_{s } - \tanh^{2 } \xi
\right ) \phi^{(1)}_{d } , \label{e2 } \end{aligned}\ ] ] for the time - dependent part . in the last equation we use the standard notation : @xmath100
our goal in this section is to solve eq .
( [ e1 ] ) and find static part of the function @xmath101 .
we adopt the method of solving , which was discussed in the previous section . in the first step we put @xmath102 on the right - hand side of eq .
( [ e1 ] ) and solve it in a standard way as a linear , inhomogeneous equation . then solution @xmath103 found in this way is inserted to the r.h.s .
( [ e1 ] ) and in the next step equation with a new source term should be solved .
nevertheless , if expression which occur in a new source term as a result of this procedure is small we get pretty good approximation in the first step .
the term in front of @xmath96 on the r.h.s . of eq ( [ e1 ] ) is nonzero only on the border of the domain wall and quickly tends to zero when @xmath91 .
eventually we expect to have a good approximation in the region outside the wall .
this result seems to be acceptable - inside the domain wall the accuracy can be improved by higher degree of the polynomial in ansatz ( [ ansatz ] ) .
finally we are going to solve the linear equation : @xmath104 \phi^{(1)}_{s } = j_{s } ( \xi ) .
\label{e3}\ ] ] the solution can be found by the standard green s function technique .
there exist two linearly independent solutions of the homogenous part of eq .
( [ e3 ] ) : @xmath105 as the green s function we take : @xmath106 .
\label{green}\end{aligned}\ ] ] the green s function @xmath107 obeys the condition : @xmath108 the general solution of eq .
( [ e3 ] ) has then the form : @xmath109 where @xmath50 and @xmath110 are arbitrary constants .
inserting the green s function ( [ green ] ) into the formula ( [ rgreen ] ) we get : @xmath111 we have to put @xmath112 and @xmath113 because we are looking for a solution generated by source term , not for the homogeneous equation solution .
another reason for keeping @xmath114 is quick ( exponential ) growth of the function @xmath115 for @xmath91 . in the case
@xmath116 the solution @xmath117 grows exponentially and does nt meet requirements of the perturbative calculus .
combining expressions ( [ js ] ) , ( [ green ] ) and ( [ rgr1 ] ) we get finally : @xmath118 f_{1 } ( \xi ) , \label{s1}\ ] ] and @xmath119 f_{1 } ( \xi ) } \nonumber \\ & & + \left [ \int_{-1}^{\xi } f_{1 } ( x ) \left ( \frac{27}{4 } \left ( x - \frac{1}{3 } x^{3 } \right)^{3 } + x^{3 } \right ) dx \right ] f_{2 } ( \xi ) .
\label{s2}\end{aligned}\ ] ] solution for @xmath120 has opposite sign .
+ integration in the formula ( [ s1 ] ) can be done numerically , yielding : @xmath121 where @xmath122 .
as one can expect @xmath117 tends to zero when @xmath123 .
the static solution with the correction @xmath124 is presented in fig .
2 , where one can compare it with the pure polynomial solution @xmath125 and the strict , static solution @xmath126 . as it is easy to notice the correction obtained above forces the polynomial solution to the well - known strict , static solution .
it gives us a strong argument , that spliting the static part of polynomial solution into two parts ( one of them is @xmath127 ) is acceptable .
this fact will be used in the next section .
( [ e1 ] ) can be also solved in a straightforward way using numerical methods .
numerical analysis confirms the results obtained above .
in this section we find time - dependent correction to the polynomial approximation .
we have to solve eq .
( [ e2 ] ) using the iterative method proposed in section [ corr ] . the second term on the r.h.s . of the eq .
( [ e2 ] ) is of the order of the small parameter @xmath93 and can be neglected in the first approximation .
the coefficient in front of @xmath97 in the third term on the r.h.s . is the same as in eq .
( [ e1 ] ) discussed in the previous section .
now it is necessary to specify more precisely the scenario of evolution of the discussed domain wall . for @xmath128
the domain wall is static and , as was discussed in the section [ stat ] has the form of the strict , static solution @xmath129 ( see ( [ stat ] ) ) . in @xmath130
the domain wall is squeezed by @xmath131 as a result of action of an external force .
the evolution at later times is governed by the field equation ( [ e2 ] ) . +
this scenario helps us to understand the role of the last term on the r.h.s .
( [ e2 ] ) . in the simplest case we can state that it is small for @xmath132 , and by neglecting it we obtain good approximation in the first step in this region . nevertheless from the discussion presented above we can conclude that we get quite good solution in the whole range of @xmath133 . to prove this we rewrite the third term on the l.h.s .
thus we have : @xmath134 \phi^{(1)}_{d } = j_{d}(t , \xi ) } \nonumber \\ & & + 6 \ , \delta \phi^{(0)}_{d } \left ( \delta \phi^{(0)}_{d } + 2 \ , \delta \phi^{(0)}_{s } + 2 \tanh \xi \right ) \left ( \phi^{(1)}_{d } + \phi^{(1)}_{s } \right ) .
\label{ap}\end{aligned}\ ] ] due to the results of section [ stat ] the static part of the polynomial solution is corrected by the function @xmath135 , which has been calculated above .
when we take some terms of higher order ( which can be obtained in the next steps of calculation ) we can change @xmath136 on the r.h.s . of eq .
( [ e2 ] ) by `` improved '' expression @xmath137 . from the discussion presented above and illustrated in fig .
2 we have : @xmath138 in the first step we are then going to solve the linear , inhomogeneous equation : @xmath139 \phi^{(1)}_{d } = j_{d}(t , \xi ) .
\label{e4}\]]this can be rewritten as a wave equation : @xmath140 \phi^{(1)}_{d } ( t , \xi ) = - j_{d } ( t , \xi ) , \label{e5}\ ] ] where operator @xmath141 has the form : @xmath142 .
\label{opr}\ ] ] eq .
( [ e5 ] ) can be solved by the standard green s function technique .
we calculate green s function for opearator @xmath141 using expansion in eigenfunctions .
+ we are looking for the green s function @xmath143 , which fulfill the equation : @xmath140 g(\xi , t;\xi',t ' ) = \delta(t - t ' ) \delta(\xi - \xi'),\ ] ] and obeys the condition : @xmath144 if the set of eigenfuctions @xmath145 is given , we can construct green s function in a standard manner .
detailed description of this procedure is given in @xcite .
the green s function can be written as : @xmath146 where propagator @xmath147 has the form : @xmath148}{\sqrt{\lambda_{n}}}. \label{falpr}\ ] ] the solution of the inhomogenous equation is given by the formula : @xmath149 because @xmath150 we can simplyfy ( [ fal1 ] ) as folows : @xmath151 the problem of calculating time - dependent correction to the polynomial solution is then reduced to finding the system of eigenfunctions of operator @xmath141 .
this system is well - known ( see e.g. ) and consists of two discrete eigenvalues and continous spectrum with corresponding eigenfunctions : @xmath152 + @xmath153 , \label{wcgle}\ ] ] where : @xmath154 the eigenfunctions from the continous part of the spectrum correspond to the real eigenvalues and can be split into two sets of eigenfunctions , orthogonal to each other : @xmath155 , \label{wcgle1}\ ] ] and : @xmath156 .
\label{wcgle2}\ ] ] + inserting formulae ( [ wdys1 ] ) - ( [ wcgle2 ] ) into ( [ falgr ] ) - ( [ fal2 ] ) we get the expression for the retarded green s function : @xmath157 \psi_{1}(\xi ) \psi_{1}^{*}(\xi ' ) \nonumber \\ & & + \left . \sum_{i = 1,2 } \int_{0}^{\infty } dk \frac{\sin [ \sqrt{k^{2}+4}(t - t ' ) ] } { \sqrt{k^{2}+4 } } \psi^{(i)}_{k } ( \xi ) \psi^{(i)*}_{k } ( \xi ' ) \right ) .
\label{g1}\end{aligned}\ ] ] we get the solution @xmath158 integrating @xmath159 with the source term @xmath160 in the whole range of @xmath161 . as @xmath160 is odd in variable @xmath162 , part of @xmath163 even in this variable does not give any contribution to the final solution .
the green s function can be rewritten as : @xmath164 \psi_{1 } ( \xi ) \psi_{1}^ { * } ( \xi ' ) \right . }
\nonumber \\ & & + \left .
\int_{0}^{\infty } dk \frac{\sin [ \sqrt{k^{2}+4}(t - t ' ) ] } { \sqrt{k^{2}+4}}\psi_{k}^{(1)}(\xi ) \psi_{k}^{(1 ) * } ( \xi ' ) \right ) .
\label{g2}\end{aligned}\ ] ] finally , inserting eq .
( [ g2 ] ) into ( [ fal2 ] ) we can write solution @xmath165
. for convenience we split it into two parts due to the parts of green s function ( [ g2 ] ) .
it reads : @xmath166 where @xmath167 and @xmath168 are given by formulae : @xmath169 } { ( k^{2}+1 ) ( k^{2}+4)^{3/2 } } \ , j_{d}(t',\xi ' ) } \nonumber \\ & & \left ( ( 2-k^2 ) \sin k \xi - 3k \cos k \xi \tanh \xi - \frac{3 \cos k \xi}{\cosh^{2 } \xi } \right ) \nonumber \\ & & \left ( ( 2-k^2 ) \sin k \xi ' - 3k \cos k \xi ' \tanh \xi ' - \frac{3 \cos k \xi'}{\cosh^{2}\xi ' } \right ) , \label{fid1}\end{aligned}\ ] ] @xmath170\frac{\sinh \xi ' } { \cosh^{2 } \xi ' } .
\label{fid2}\ ] ] + let us give an interpretation of these two parts of the final solution .
@xmath168 is connected with the @xmath171 mode in the system of eigenfunctions of the operator @xmath141 .
the interaction with @xmath171 generates in the time - dependent correction @xmath97 the component , which exponentialy vanishes for large @xmath86 and oscillates .
it can be treated as a form of excitation of the domain wall given by @xmath172 .
the crucial role which this component plays in dynamics of our system is more transparently visible in the process of collision of kinks ( see @xcite and @xcite ) .
+ the component @xmath167 has the form of a wavepacket - the oscillations of the domain wall generate radiation .
thus we have the following scenario of the discused phenomena : in @xmath173 the external force sqeezed the domain wall by @xmath174 .
the wall oscillates around the static configuration .
the oscillations generate radiation as in the formula ( [ fid1 ] ) .
it is natural to restrict @xmath175 to waves going out from the wall : @xmath176 \left [ 3k \tanh \xi \left ( 2-k^{2 } - \frac{3}{\cosh^{2 } \xi ' } \right ) \right .
\nonumber \\ & & \left . -
3k \tanh \xi ' \left ( 2-k^{2 } -\frac{3}{\cosh^{2 } \xi } \right ) \right ] \nonumber \\ & & + \sin [ k ( \xi - \xi ' ) - \sqrt{k^{2}+4}(t - t ' ) ] \left [ -2 \left ( 2 - k^{2 } - \frac{3}{\cosh^{2 } \xi } \right ) \right .
\nonumber \\ & & \left .
\left ( 2 - k^{2 } - \frac{3}{\cosh^{2 } \xi ' } \right ) - 9 k^{2 } \tanh
\xi \tanh \xi ' \right ] .
\label{comp } \end{aligned}\ ] ] for @xmath177 the result is analogous .
+ the solution @xmath178 has the form of a wavepacket : @xmath179 the waves have the frequency @xmath180 so they satisfy the dispersion relation : @xmath181 it agrees well with our expectations , bacause for large @xmath182 eq .
( [ e4 ] ) reads : @xmath183 to plot the function @xmath178 given by the formula ( [ comp ] ) we calculate it numerically for some fixed values of @xmath184 and @xmath185 , with the step @xmath186 . the final result is presented in fig .
one can easily see the causal character of obtained solution . in the region outside the light - cone
the field @xmath178 is nearly equal to @xmath187 .
the fluctuations are caused mainly by computation errors .
this fact was checked by decreasing the step of integration - the fluctuations decrease then too .
the field @xmath178 is small - on the figures it is multiplied by factor @xmath188 .
this once again confirms the approximation used in our approach .
in the previous section we have shown that the considered , oscillating domain wall radiates .
thus , we have the energy flowing out from the domain wall
. we can then expect dumping of oscillations of the domain wall . in this section
we construct a simple model of this process .
we start from equation of energy balans : @xmath189 where @xmath190 is the energy per unit area of the domain wall for the polynomial solution and @xmath191 is the energy emitted from unit square in the moment @xmath192 . from eq .
( [ energia ] ) one can calculate @xmath190 . neglecting terms of the second and higher order in @xmath62
we get : @xmath193 where @xmath194 , @xmath195 .
@xmath196 is the energy of the static , polynomial domain wall solution
. one can easily compare it with the energy of the strict , static solution @xmath197 : @xmath198 \approx 1.33 $ ] . as we can expect
the energy @xmath196 is a bit greater than @xmath199 $ ] .
( [ e0 ] ) @xmath190 is time - dependent .
it is caused by limitted accuracy of the polynomial solution - in the case of a strict solution energy should be independent on time .
in @xmath200 we have @xmath201 as the energy of our configuration .
for @xmath202 it oscillates as a result of approximation in the polynomial solution , nevertheless for the strict solution it should be constant .
accordingly we take : @xmath203 the parameter @xmath62 describes the amplitude of the domain wall oscillations . in reality
it is time - dependent : @xmath204 ( and we expect that it decreases as a result of radiation emitted from the wall ) .
the energy flux can be easily calculated from the energy - momentum tensor for the action ( [ lagr ] ) : @xmath205 the energy flux density @xmath206 is given by the formula : @xmath207 from the continuity equation we can get in a standard manner expression for @xmath208 : @xmath209 as @xmath210 we can re - define the field as follows : @xmath211 thus , we have : @xmath212 collecting the formulae ( [ bilans ] ) , ( [ e01 ] ) and ( [ fl ] ) one gets the following differential equation for @xmath213 : @xmath214 the solution of eq .
( [ e02 ] ) with the initial condition @xmath215 reads : @xmath216 eventually to find the solution @xmath217 one needs the value of the field @xmath218 on the border of the domain wall , which was calculated numerically in the previous section .
+ we have performed the numerical computations for three different values of the initial oscillation amplitude , namely : @xmath219 .
the final result is presented in figs .
8 , 9 and 10 .
the dumping of oscillations , as we can expect , is the quickest when we have the greatest amplitude .
+ the plots of function @xmath217 have characteristic bends at multiplicities of half - period of oscillations @xmath220 .
we can interprete them as follows : the energy is most strongly radiated from the wall when the velocity of oscillations is the greatest - near the returning points the velocity is small and the dumping is slower .
this coincides with the half - period of oscillations .
+ for small , initial amplitudes of the oscillations the dumping is very weak . in the case of @xmath221 during the time of approximately 4 oscillation periods the amplitude decreases only by @xmath222 . for @xmath223 and @xmath224
the corresponding values are @xmath225 and @xmath226 .
these two cases can however be outside the region of application of our approximation ( it demands @xmath227 ) .
let us briefly summarize the main points of our work .
we have derived the planar domain wall solution with the time - dependent thickness in the polynomial approximation .
the planar domain wall oscillates with small amplitude .
we have then proposed an approach enabling us to calculate the correction to the pure polynomial solution .
we have split this correction into two parts .
the time - dependent part contains the component which is interpreted as the radiation from the oscillating domain wall .
there are several possible extensions and generalizations of our approach .
it can be applied without much trouble to the other field - theoretical models . in our paper
we discuss only the case of small oscillations around the static solution .
as one can easily see in fig . 1
there exist also solutions oscillating with amplitude which is not small .
this case should be discussed separately . in accordance with our discussion in section [ back ]
we can expect very quick dumping of the oscillations . on the other hand
, we should remember that our method is based on the linear approximation and is reliable for small amplitudes of oscillations only .
thus , the process of creating the kink - antikink pair is also possible . 9 t.w.b .
kibble , j.phys .
, 1387 ( 1976 ) .
a.l.vilenkin , phys.rep . , 263 ( 1985 ) .
w.h.zurek , phys.rep .
, 177 ( 1996 ) .
r.p.huebner , magnetic flux structures in superconductors .
springer verlag , berlin - heidelberg - new york , 1979 .
r.j.donnelly , quanized vortices in helium ii .
cambridge university press , cambridge , 1991 .
solitons and condensed matter physics , ed .
a.r.bishop , t.schneider , springer verlag , berlin - heidelberg - new york , 1978 .
h.arod , j.stelzer , phys .
, 1784 ( 1997 ) .
r.peka , acta phys.pol .
b , 1981 ( 1997 ) .
h.arod , a.l.larsen , phys.rev . 4154 ( 1994 ) .
h.arod , phys.rev .
, 1082 , 1995 .
h.arod , nucl.phys . , 174 ( 1995 ) .
h.arod , nucl.phys . , 273 ( 1998 ) .
j.karkowski , z. wierczyski , acta phys.pol .
b @xmath228 , 234 ( 1996 ) .
a.a.kapustnikov , a.pashnev , a.pichugin , phys .
rev . , 2257 ( 1997 ) .
v.g.kiselev , ya.m.shnir , phys .
, 5174 ( 1998 ) .
d.k.campbell , j.f.schonfeld , c.a.wingate , physica , 1 ( 1983 ) .
m.peyrard , d.k.campbell , physica , 33 ( 1983 ) .
r. rajaraman , solitons and instantons .
north - holland publishing company , amsterdam - new york - oxford , 1982 .
g.korn , t.korn , sprawocznik po matematikie .
mir , moskwa , 1970 .
g.barton , elements of green s functions and propagation .
clarendon press , oxford , 1989 .
+ + fig .
1 . the configuration space of the equation ( [ e1 ] ) .
+ + fig .
the statical polynomial solution , the improved polynomial solution and the strict solution @xmath229 .
+ + fig .
3 . the radiation emitted by the oscillating domain wall for @xmath230 .
+ + fig .
the radiation emitted by the oscillating domain wall for @xmath231 . + + fig .
the radiation emitted by the oscillating domain wall for @xmath232 . + + fig . 6 . the radiation emitted by the oscillating domain wall for @xmath233 .
+ + fig . 7 . the radiation emitted by the oscillating domain wall for @xmath234 .
+ + fig . 8 .
the dumping of oscillations for the initial amplitude @xmath235 .
+ + fig .
9 . the dumping of oscillations for the initial amplitude @xmath236 .
+ + fig . 10 .
the dumping of oscillations for the initial amplitude @xmath237 . | domain wall - type solution with oscillating thickness in a real , scalar field model is investigated with the help of a polynomial approximation . we propose a simple extension of the polynomial approximation method . in this approach
we calculate higher order corrections to the planar domain wall solution , find that the domain wall with oscillating thickness radiates , and compute dumping of oscillations of the domain wall .
dynamics of a planar domain wall with oscillating thickness in @xmath0 model + by + maciej lusarczyk@xmath1 + jagiellonian university , institute of physics + reymonta 4 , 30 - 059 cracow , poland + @xmath1 e - mail : [email protected] |
it is widely believed that most millisecond pulsars ( msps ) are spun up by accretion from a close binary companion .
this recycling ( radhakrishnan & srinivasan , 1981 ) can only take place when the neutron star magnetic field has decayed to a value @xmath4 g. if accretion subsequently stops for some reason , the neutron star appears as a millisecond pulsar with a very low spindown rate , as its low magnetic field makes dipole radiation very weak . in line with these ideas a large proportion of millisecond pulsars
are members of binary systems .
often these pulsars undergo very wide eclipses , with an obscuring object much larger than the companion star s roche lobe .
this must be an intense wind from the companion star , driven in some way by the pulsar emission ( fruchter et al . , 1988 ) .
whenever such eclipses are seen the binary eccentricity and the pulsar mass function are extremely small , implying companion masses @xmath5 .
there is another group of binary millisecond pulsars with systematically lower mass functions , and it is natural to assume that these are also evaporating systems with orbital inclinations which prevent us seeing the eclipses ( freire et al . , 2001 ) . in a recent paper ( king et al .
2003 ; hereafter kdb ) we pointed out that the incidence of black widow pulsars ( bwps ) is far higher in globular clusters than in the field .
we identified a favoured formation mechanism for bwps in globulars in which turnoff mass stars exchange into wide binaries containing recycled millisecond pulsars ( msps ) and eject their helium white dwarf companions .
once angular momentum loss or nuclear expansion bring the companion into contact with its roche lobe the pulsar is able to expel the matter issuing through the inner lagrange point @xmath6 .
thus mass is lost on the binary evolution timescale .
bwps are observable only when this evolution and thus the mass loss are slow . here
we consider the consequences of this picture for the subsequent evolution of bwps .
our main aim is to understand the distribution of bwps in the plane of minimum companion mass @xmath7 and orbital period @xmath8 ( fig .
[ figmsp ] ) .
bwps appear to fall into two distinct groups which we shall call high mass and low mass , depending on whether @xmath9 or @xmath10 .
the high mass group have noticeably more irregular eclipses than the low mass group ( e.g. scott ransom , talk at the aspen meeting on binary millisecond pulsars in january 2004 ) . in the standard way we may associate with these two groups other binary msps which do not eclipse but whose values of @xmath11 are even smaller than those of the eclipsing systems .
the interpretation is that these are bwps seen at low orbital inclination . with this association the distribution of fig .
[ figmsp ] puts the dividing line between high and low mass bwps at @xmath12 .
we consider the low mass group first .
the small secondary mass ( close to @xmath7 for eclipsing systems ) implies a thermal time much longer than the binary evolution timescale for mass loss causing bwp behaviour .
the star therefore reacts adiabatically . as it is either fully convective or degenerate we can model approximately it as an @xmath13 polytrope , with radius @xmath14 ( cf e.g. king 1988 ; for more accurate representations see nelson & rappaport , 2003 ) . here
@xmath15 is the fractional hydrogen content , and @xmath16 measures the deviation from the fully degenerate radius given by @xmath17 .
we assume that low
mass bwps have a range @xmath18 few for reasons we will discuss in the next section . assuming a roche - lobe filling donor gives the mass - period relation @xmath19 where @xmath20 is the orbital period measured in hours .
we assume further that loss of orbital angular momentum via gravitational radiation ( gr ) drives orbital evolution , so that @xmath21 where @xmath22 , @xmath23 is the pulsar mass , and @xmath24 is the specific angular momentum of the lost mass relative to that of the secondary ( van teeseling & king 1998 ) .
thus @xmath25 where @xmath26 is the binary separation and @xmath27 is the distance from the centre of mass at which the matter is ejected from the binary .
this is between the circularization radius of the infalling matter and the @xmath6 point .
as all bwps have @xmath28 , @xmath29 except for very small @xmath30 ( see below ) the first term on the right hand side of ( [ dotm ] ) is @xmath31 and we can rewrite ( [ dotm ] ) as @xmath32 where @xmath33 and @xmath34 . according to kdb
the system is only visible as a bwp if this rate is less than the critical value @xmath35 for free free absorption at typical ( 400 1700 mhz ) observing frequencies , where @xmath36 is the temperature of the lost mass near @xmath6 in units of @xmath37 k. the resulting constraint @xmath38 is plotted on fig .
[ figbwp ] as ` visibility line ' . in a similar way we can plot various other constraints on this figure .
for very small mass ratios @xmath39 the @xmath24 term in the denominator of ( 3 ) becomes significant .
this signals dynamical instability , as the roche lobe moves inwards wrt the stellar surface ( cf stevens , rees & podsiadlowski , 1992 ) . at such masses
the companion must be broken up and sheared into a large disc surrounding the pulsar .
this is presumably the origin of the planets observed around psr b1257 + 12 ( wolszczan & frail 1992 ; konacki & wolszczan 2003 ) .
direct numerical calculation of @xmath27 from roche geometry shows that dynamical instability occurs when @xmath40 if matter is ejected from the @xmath41 point .
however , matter can be ejected from anywhere between the @xmath41 point and the circularization radius ( which even for small mass ratios is not comparable with the distance to the @xmath41 point ) .
if matter reaches in 5 per cent of the distance from the centre of mass to the @xmath41 point this gives @xmath42 i.e. @xmath43 .
this line is labelled ` dynamical instability ' on fig .
[ figbwp ] .
we also plot the line ( ` hubble line ' ) on which the binary evolution has a characteristic timescale longer than a hubble time .
finally we plot the binary evolution of a system whose secondary is fully degenerate ( ` degenerate ' ) and one whose radius is 3 times as large for the same mass ( @xmath44 ) as given by equation ( [ mp ] ) .
arrows on the degenerate sequences indicate the direction of evolution . if this is a viable picture of the evolution of low
mass bwps , we expect them to have combinations of @xmath45 lying inside the various constraints shown in fig . [ figbwp ] .
given that eclipsing systems ( solid circles ) have @xmath46 , while non eclipsing systems have @xmath47@xmath48@xmath7 , we see that the data plotted on fig .
[ figbwp ] are indeed reasonably consistent with adiabatic evolution under mass loss driven by gravitational radiation .
we now consider the remaining systems on fig . [ figbwp ] ,
i.e. those with @xmath49 .
nelson ( talk at the aspen meeting on binary millisecond pulsars in january 2004 ) has studied the evolution of the long period system j17405340 , which has a subgiant companion ( damico et al .
2001 ; ferraro et al . 2001 ) and shown that it is consistent with a location near a bifurcation point , at which nuclear evolution and angular momentum loss are comparable . the remaining 4 systems have @xmath50hr and @xmath51 . except for implausibly small orbital inclinations these companions have main sequence radii far too small to fill their roche lobes .
there are various ways around this difficulty , including competition between nuclear and orbital evolution , or attributing the eclipses to strong stellar winds of detached companions .
however the most likely explanation appears to us to follow from the turbulent history these binaries must have had .
the process of gaining a new companion must involve considerable disturbance to that star , probably leading to extensive mass loss .
this is clearly indicated for tidal capture ( podsiadlowski 1996 ) , but is probably true in any picture . for an exchange encounter to give a short period system a large post - exchange eccentricity
is required ( see section [ discuss ] ) .
when this eccentric system circularizes the tidal effects on the secondary can in some cases be similar to tidal capture if the periastron distance is comparable to the stellar radii .
the globular
cluster x ray binary ac211 ( van zyl et al .
2004 ; charles , clarkson & van zyl 2002 , and references therein ) may be an example where the companion star is oversized because of the capture process , the only difference being that the neutron star in this system accretes the overflowing matter rather than expelling it , presumably because no previous partner recycled it . for an exchange encounter the lowest mass star is ejected and so the captured star , in a high - mass bwp formed through this route , must have lost a significant portion of its initial mass .
this mass loss , however , occurs during the tidal circularization ( not afterwards ) and so the companion could still be thermally bloated . after the tidal disturbance and consequent mass loss , the companion attempts to reach its new main sequence radius on a thermal timescale .
this process competes with orbital shrinkage via gr and reduces the mass loss ( already weak cf equation ( [ dotm2 ] ) for the relatively long orbital periods of most of this group ) .
it is therefore plausible that bwps ultimately emerge from the tidally induced mass loss with the parameters of the high mass bwps .
the thermal timescale of some of these bwps may be less than the gr timescale and so eventually these would shrink from contact with their roche lobes .
however , they would still be observed as high - mass bwps for the length of their thermal timescale .
if the companion has a sufficiently strong stellar wind , this can produce orbital eclipses through free free absorption .
this appears to happen in psr 171819 ( wijers & paczyski 1993 ; burderi & king 1994 ) .
here @xmath52hr , and modelling of the absorption light curve ( burderi & king 1994 ) shows that @xmath53 .
we note that psr 171819 is probably a cluster member ( wijers & paczyski 1993 ) and that the stellar wind must be the eclipsing agent as the pulsar is not an msp , and thus incapable of driving mass loss .
although the companion must relax towards the main sequence , its radius ultimately shrinks only slowly , whereas orbital shrinkage via gr accelerates .
depending on the initial separation , the binary reaches contact with the companion somewhat oversized compared with its thermal equilibrium radius .
this is probably the origin of the range of radii ( @xmath54values ) inferred for low
mass bwps above .
grindlay ( talk at aspen meeting on binary millisecond pulsars in january 2004 ) has found 108 x - ray sources in the globular cluster 47 tuc .
a number of these are claimed to be quiescent low - mass x - ray binaries ( lmxbs ) and have measured periods from x - ray dips / eclipses ( eg w37 at 3hrs ) , power law components ( possibly from a wind ) and variable absorption .
these may be bwps , rather than quiescent lmxbs , where the absorption is too large to observe the radio emission and the stellar wind is responsible for the x - ray emission .
fig . [ figmsp ] reveals the striking fact that _ all binary msps with orbital periods _
@xmath3hr _ are bwps : 10 out of 16 actually eclipse .
_ there are two obvious possible explanations for this : \(a ) fan beam .
the pulsar beam of a recycled msp is so wide that it always includes the orbital plane , whatever the relative orientation of spin and orbit . for a wide fan beam
such systems are detectable at any spin inclination .
then all binary msps become bwps as soon as their companions fill their roche lobes .
\(b ) pencil beam .
the beam of a recycled msp is narrowly confined .
clearly the only plausible geometry for making bwps has the beam axis orthogonal to the spin , with the latter roughly aligned with the binary orbit .
it is quite difficult to break the degeneracy between these two possibilities .
however the pencil beam requires alignment , and thus that the pulsar has accreted @xmath55 from its _ current _ companion .
this requires it to have been an lmxb and then somehow broken contact , and seems harder to reconcile with the picture of high mass bwp evolution we have sketched above .
we thus tentatively conclude that a fan beam offers a simple explanation for the universality of eclipsing behaviour in msps with short orbital periods .
it is generally agreed that dynamical encounters must be invoked to explain the overproduction of bwps in globular clusters ( kdb ; rasio et al .
rappaport , putney & verbunt ( 1989 ) considered the problem of exchanging out the white dwarf companion of the neutron star in a binary msp .
their fig .
1 shows that the timescale ( in 47 tuc ) for ejecting the white dwarf from systems with periods of 10 , 100 and 1000 days is @xmath5610 , 3 , and 1 gyr , respectively .
thus an exchange encounter is likely only for orbital periods @xmath57 days .
the post exchange period is generally not substantially shorter , and most of these systems would not become bwps .
however the post
exchange eccentricity @xmath58 is evenly distributed in @xmath59 ( heggie 1975 ) . for large @xmath59 tides
circularize the orbit at much tighter separations than the initial post - exchange value .
the post - exchange period after tidal circularization can thus be much shorter ( hours ) in a few cases , producing a bwp .
the need for the exchange into the binary msp is set by the assumption that radio pulsars do not turn on as long as there is a binary companion in near contact ; thus the need for removing it altogether after the initial recycling .
burgay et al .
( 2003 ) , however , claim that the lack of detection of radio pulsations from quiescent lmxbs could result from pulsar radiation driving a large outflow from the system which then attenuates the radio emission .
a problem with this scenario is that the pulsar must remove the accretion disc before it can blow away matter flowing through the inner lagrangian point .
burderi et al .
( 2001 ) show that this can only occur in some wide systems , and only then when the magnetic pressure due to the neutron star overcomes the disc s internal pressure .
consequently it is unlikely that the companion in a bwp is the one that spun the pulsar up , unless the latter has somehow contracted well inside its roche lobe .
if this occurs through evolutionary loss of the envelope one then has the problem of bringing the system to short periods in a reasonable time . converting wide orbits to narrow ones was the main motivation for rasio et al .
( 2000 ) in considering an alternative scenario .
this invokes binary formation through exchange interactions , producing neutron star systems with companions which are more massive .
this must occur in a relatively short interval of @xmath60 gyr before all the potential extended companions with masses larger than that of the neutron star have finished their evolution .
given such a system , a common envelope phase results once the companion evolves and overfills its roche lobe .
the envelope of the red giant is ejected and the core spirals in , leaving a he rich white dwarf in a relatively close orbit with the neutron star . if contact is established , the pulsar could presumably begin to evaporate the companion .
however because the donor star has no hydrogen , the orbital periods of systems made this way are much shorter than those of observed bwps ( see equation [ r2 ] and fig . [ figbwp ] ) .
the evolutionary outcome for bwps depends on whether they have evolved to a period where they have become dynamically unstable .
a number of bwps exist which have reached or are evolving towards the ` hubble line ' and so are not yet old enough to have been dynamically disrupted . the systems which have undergone dynamical instability are observed as isolated msps . in the globular 47 tuc 7 out of 21 of msps are isolated ( see http://www.naic.edu/~pfreire/gcpsr.html[http://www.naic.edu/@xmath56pfreire/gcpsr.html ] for an up to date list of pulsars in globular clusters ) while 5 out of 21 are low - mass bwps .
objects may undergo evolution through the low - mass evolutionary route and disruption of the companion .
however , the number of resulting isolated msps should not be significantly greater than the number of bwps unless there is a process which strongly favours short periods ( less bloated secondaries ) .
the similar number of low - mass bwps and isolated pulsars in 47 tuc supports this conclusion .
we thank the referee saul rappaport for comments which have improved this paper .
we are also grateful to the organisers of the aspen meeting on binary millisecond pulsars in january 2004 which provided much stimulus for this work .
dr and ks are supported by the leicester pparc rolling grant for theoretical astrophysics , and js by a pparc studentship .
ark gratefully acknowledges a royal society wolfson research merit award .
meb acknowledges the support of a ukaff fellowship . burderi l. , king a.r . , 1994 ,
apj , 430 , l57 burderi l. et al . , 2001 ,
apj , 560 , l71 burgay m. , burderi l. , possenti a. , damico n. , manchester r.n . ,
lyne a.g . , camilo f. , campana s. , 2003 , apj , 589 , 902 charles p.a . , clarkson w.i . , van zyl l. , 2002 , new ast , 7 , 21 damico n. , possenti a. , manchester r.n . , sarkissian j. , lyne a.g . , camilo f. , 2001 , apj , 548 , l171 ferraro f.r . , possenti a. , damico n. , sabbi e. , 2001 , apj , 561 , l93 freire , p. c. ; camilo , f. c. ; kramer , m. ; lorimer , d. r. ; lyne , a. g. ; manchester , r. n. ; damico , n. , 2001 , aas abstract 199 , # 159.04 fruchter , a.s . , stinebring , d.r . ,
taylor , j.h . , 1988 ,
312 , 255 grindlay j.e . , 2004 , in rasio f.a . ,
stairs i.h .
, eds , binary radio pulsars , asp conf .
ser . , san francisco heggie d.c .
, 1975 , mnras , 173 , 729 king a.r . , 1988 ,
qjras , 29 , 1 king a.r . , davies m.b . ,
beer m.e . , 2003 , mnras , 345 , 678 konacki m. , wolszczan a. , 2003 , apj , 591 , l147 nelson l.a . , 2004 , in rasio f.a . , stairs i.h .
, eds , binary radio pulsars , asp conf .
, san francisco nelson , l.a . ,
rappaport , s.a .
2003 , apj 598 , 431 podsiadlowski ph . , 1996 , mnras , 279 , 1104 radhakrishnan , v. & srinivasan , g. , 1981 , paper presented at 2nd asian pacific regional iau meeting , bandung ransom s.m . , 2004 , in rasio f.a . , stairs i.h .
, eds , binary radio pulsars , asp conf .
, san francisco rappaport s. , putney a. , verbunt f. , 1989 , apj , 345 , 210 rasio f.a . ,
pfahl e.d . , rappaport s. , 2000 , apj , 532 , l47 stevens i.r . , rees m.j . ,
podsiadlowski ph . , 1992 , mnras , 254 , 19 van zyl l. , charles p.a . , arribas s. , naylor t. , mediavilla e. , hellier c. , 2004 , mnras , 350 , 649 van teeseling a. , king a.r . , 1998 ,
a&a , 338 , 957 wijers r.a.m.j . ,
paczyski b. , 1993 , apj , 415 , l115 wolszczan a. , frail d.a .
, 1992 , nat , 355 , 145 | we consider the population of black widow pulsars ( bwps ) .
the large majority of these are members of globular clusters . for minimum companion masses
@xmath00.1m@xmath1 , adiabatic evolution and consequent mass loss under gravitational radiation appear to provide a coherent explanation of all observable properties .
we suggest that the group of bwps with minimum companion masses @xmath2 are systems relaxing to equilibrium after a relatively recent capture event .
we point out that all binary millisecond pulsars ( msps ) with orbital periods @xmath3hr are bwps ( our line of sight allows us to see the eclipses in 10 out of 16 cases ) .
this implies that recycled msps emit either in a wide fan beam or a pencil beam close to the spin plane .
simple evolutionary ideas favour a fan beam .
[ firstpage ] binaries : close |
the collapsar model @xcite predicts that long - duration , soft spectrum , gamma - ray bursts ( grbs ) would be accompanied by a supernova ( sn ) explosion .
this was first confirmed by the spectroscopic association of grb 980425 with sn 1998bw and later followed by another four events ( grb 030329/sn 2003dh , grb 031203/sn 2003lw , grb 060218/sn 2006aj and grb
100316d / sn 2010bh ; see ref .
* for a recent review ) .
there is less conclusive , but substantial evidence of grb / sn association in another two - dozen events . nonetheless , most gamma - ray bursts have no observed supernova counterpart , most likely because the greater fraction of them occur at large red shifts and thus make a supernova detection unlikely .
most supernovae associated with gamma - ray bursts are type ic ( broad line ) , i.e. , they show no hydrogen nor helium lines , and large velocities are inferred for the ejected material ( @xmath0 km / s ) .
as pointed out by @xcite , long gamma - ray bursts are also associated with star forming regions , thus strengthening the relation with the collapse of massive stars given that such stars have lifetime scales that span only a few million years and are , thus , incapable of being displaced long distances from their birthplaces before they collapse into a black hole . in the same paper
, @xcite suggests that the central engine of these long - soft gamma - ray bursts is powered up by a compact object with large angular momentum and magnetic fields through the blandford - znajek ( bz ) mechanism @xcite . almost in parallel ,
@xcite ( and collaborators ) worked on the formation of binaries with black holes ( bhs ) , blandford - znajek mechanism @xcite and gamma - ray bursts @xcite . in this work
a summary is provided of some of the most relevant results resulting from the idea of using the formation of a black hole in a close binary as the source of grb / hn explosions . in the next section a brief explanation of some key ideas of the collapsar model
this is followed by a description of the stellar binary model leading to case c mass transfer and the spin up of massive stars in the third section .
next , some relevant results from the blandford - znajek model are covered in the fourth section .
this leads to a description of the best known black - hole binaries in our galaxy and nearby ones in the fifth section .
to finish the description of black - hole binaries , in the sixth section we discuss some massive systems some of which may harbor future gamma - ray bursts .
the conclusions are shown in the last ( seventh ) section of this work .
massive single stars ( @xmath1 , where @xmath2 stands for the zero - age - main - sequence mass ) evolve while burning through their nuclear - fuel resources and eventually explode as supernovae leaving behind a compact object . if the zams mass of the star is somewhere below the range between the 18 to 25 @xmath3 it is likely that the remnant will be a neutron star ( ns ) .
but if the zams mass is above this limit , the remnant will likely exceed the maximum mass for a stable neutron star ( which is likely between 3 and @xmath4 and depends on the equation of state of matter at density values above the nuclear one ) and will further collapse and form a black hole .
these mass limits are somewhat more vague if the stars evolve in binaries given that mass transfer and tides may influence the evolution of the star .
the collapsar model considers a massive star that collapses and fails to launch a supernova , thus forming a black hole a few seconds after the initial collapse .
material in the core of the star as well as along the rotation axis of the star , having little to no angular momentum , collapses directly , clearing a path along this axis .
most of the material outside the central core of the star possesses enough angular momentum to prevent it from directly falling in and , thus , forms an accretion disk . before the collapse
, the core of the star , where the silicon ashes burn into iron , has a temperature of @xmath5 k , and thus all the material is completely ionized .
this results in a magnetic field which is _ frozen _ into the material . as the collapse develops the magnetic flux as well as the available angular momentum
are conserved .
thus , during the contraction phase , the proto - neutron star spins more and more rapidly and its magnetic field increases .
eventually , the proto - neutron star becomes too massive to prevent the formation of a black hole and it collapses into one .
the interaction between the rapidly spinning black hole at the center of the star and the accretion disk is expected to produce two jets that will finish removing the remaining material along the rotation axis of the star .
this will be observed as a gamma - ray burst .
the outer layers of the accretion disk will be heated up until they either are expelled via a strong wind or they explode .
these will be observed as an extremely luminous and rapidly expanding supernova , often termed _ hypernova _ ( hn ) .
as it has been shown by , e.g. , @xcite and , a star which is born with more - than - enough angular momentum in its core to produce a collapsar , may lose most of this angular momentum by the time it is finished burning hydrogen ( i.e. , at the end of the main sequence ) .
even if this were not the case and enough angular momentum was still left in the star , during the helium - burning stage the envelope of a massive star expands from a few to several - hundred solar radii carrying out most of the angular momentum of the star . even though the helium core contracts and spins up , viscosity and/or magnetic fields slow it down ( a viscous or an alfvn timescale are usually much shorter than the nuclear timescale ) until solid - body rotation is restored , thus draining angular momentum from the stellar core . during the helium - burning stage and
later ones most of the envelope is lost due to strong stellar winds carrying away most of the remaining angular momentum left in the star .
such estimates are made with the assumption that the spruit - tayler dynamo is at work inside the star and , thus , the angular momentum is transported due to magnetic torques ; this may not be the complete picture , nonetheless the predictions for the rotation of pulsars and white dwarves seem to be consistent with their corresponding observed populations .
if most massive stars are drained out of their original angular momentum during their evolution then it is clear that collapsars ( gamma - ray burst progenitors ) may not evolve through this channel given that , both , a strong magnetic field and a very large amount of angular momentum in the collapsing core are needed , and they seem to be exclusive of each other .
at least two different solutions have been proposed around this problem .
one of them is to have rapidly - rotating ( above @xmath6 keplerian rotation ) massive stars which may evolve chemically homogeneously and , thus , avoid the giant phase . by doing this
they do not carry most of their angular momentum into an extended envelope ( of up to hundreds of solar radii ) which is later lost to winds .
instead , the winds remove angular momentum from a much more compact surface ( a few tens of solar radii ) , allowing the stars to preserve a substantial fraction of their original angular momentum . such scenario may occur for single stars , or tidally - locked binaries .
the other solution is the one we have put forward in @xcite and will address in the remainder of this work .
it involves a binary in an initially wide orbit ( @xmath7 , where @xmath8 stands for the orbital separation ) with a massive star ( @xmath9 ) and a non - massive companion star ( @xmath10 to @xmath4 ) .
the massive star evolves as a single star through its main sequence and helium - core burning stages , expanding from a few solar radii to @xmath11 during the red giant stage .
it is until the massive star begins helium - shell burning , when it further expands to @xmath12 , that it fills its roche lobe and starts an unstable mass transfer onto the companion star .
mass transfer at or after helium - shell burning is known in the literature as case c mass transfer . as
a consequence the orbit decays rapidly and a common envelope ensues .
the orbital separation keeps decreasing at the expense of removing the hydrogen envelope of the primary .
the common envelope ends when the envelope is fully removed , leaving the massive helium core exposed . at this stage
the orbit has been reduced to merely a few solar radii .
it is worth noting here that the situation is already better than that of a single star because by removing all the hydrogen envelope the gamma - ray burst will not have to struggle its way out of it risking being completely quenched .
furthermore , it naturally explains the lack of hydrogen lines in the spectra of hypernovae - related to gamma - ray bursts ( i.e. , they are not type ii ) . in this new configuration ,
the exposed core is close to filling its roche lobe and thus the ratio between the stellar radius , @xmath13 , and the orbital separation is not far from unity . from
we know that the tidal synchronization timescale depends on @xmath14 where @xmath15 is a high power ( either 6 or @xmath16 depending on whether the equilibrium tide or the dynamical tide , respectively , are most relevant ) .
thus tidal synchronization allows for a quick spin up of the massive star . at this stage , it is important for the massive star to have a substantial magnetic field ( @xmath17 g ) in order to have rapid transfer of angular momentum into the core of the star .
later in the evolution of the star , an extremely large magnetic field may also steal away part of the angular momentum of the iron core , as estimated by @xcite . as we mentioned earlier ,
the exact transport mechanism for the angular momentum is still not fully understood .
one can estimate values for the internal magnetic field that provide alfvn timescales appropriate to transport inwards angular momentum from the tidal interaction , and where little is lost outwards later , during the shorter lifetimes of the post - carbon - burning stages .
if the star preserves this amount of angular momentum until it collapses , the left plot of fig . [ fig:1 ] ( from *
* ) shows the expected spin or kerr parameter ( @xmath18 where @xmath19 is the angular momentum , and @xmath20 is the mass of the collapsed object ) of the rotating black hole as a function of the orbital period of the pre - collapse binary . [ cols="^,^,^ " , ]
during the formation of a stellar - mass black hole , the angular momentum as well as the magnetic flux that threads the core of the star are conserved .
if a substantial kerr parameter ( maybe @xmath21 ) is achieved , an accretion disk forms early on and a large magnetic field is produced ( @xmath22 g ) the conditions are ripe for a central engine of the blandford - znajek @xcite type to power a grb / hn explosion .
a spinning black hole of mass @xmath23 and kerr parameter , @xmath24 , contains a large amount of rotational energy which can be quantified by using the following formulas : e_rot = f(a _ ) m_bh c^2 , where f(a _ ) = 1- . here
@xmath25 is a function of @xmath24 which varies between 0 ( for @xmath26 , i.e. , a schwarzschild black hole ) and 0.29 ( for @xmath27 , a maximally spinning or kerr black hole )
. this rotational energy can be tapped out of the black hole through interaction with the accretion disk via the strong magnetic field that permeates both .
essentially , the plasma that conforms the accretion disk can not freeze in the magnetic field that threads it as the later is dragged by the rapidly - rotating black hole , thus producing a poynting flux along the rotational axis and heating up the plasma in the accretion disk . at maximum impedance ,
when the spin frequency of the magnetic field ( @xmath28 ) threading the material at the isco ( the innermost stable circular orbit ) and further out is half of the spin frequency of the black hole ( @xmath29 ) , or @xmath30 = 0.5 , one obtains the maximum energy - extraction efficiency . an analytical expression that provides the energy extracted by the blandford - znajek mechanism
is given by : e_bz = 1.8 10 ^ 54 _ f(a _ ) .
as we will show in the following sections , this result can be directly applied to several black - hole binaries where the masses and orbital periods are well constrained to obtain the available energies in their black holes .
the more reliable estimates for kerr parameter and available energy will come from those systems where modeling the evolution of the binary back to the pre - collapse situation can be better constrained because , as we have mentioned above , the pre - collapse orbital period is what determines the kerr parameter of the black hole .
@xmath31 & @xmath32 & @xmath33 & model & measured & @xmath34 & @xmath35 + & [ @xmath3 ] & [ @xmath3 ] & [ @xmath3 ] & [ @xmath3 ] & @xmath37 & @xmath38 & [ days ] & [ @xmath39 ergs ] + + j1118@xmath40480 & @xmath41 & @xmath42 & @xmath43 & @xmath44 & @xmath45 & - & @xmath46 & @xmath47 + vel 93 & @xmath41 & @xmath42 & @xmath48 & @xmath49 & @xmath45 & - & @xmath50 & @xmath47 + gro j0422@xmath4032 & @xmath51 & @xmath42 & @xmath52 & @xmath53 & @xmath45 & - & @xmath54 & @xmath55 + xte j1859@xmath40226 & @xmath51 & @xmath42 & @xmath56 & & @xmath45 & - & @xmath57 & @xmath55 + gs1124@xmath58683 & @xmath51 & @xmath42 & @xmath59 & @xmath60 & @xmath45 & - & @xmath61 & @xmath55 + h1705@xmath58250 & @xmath51 & @xmath42 & @xmath62 & @xmath63 & @xmath45 & - & @xmath64 & @xmath55 + a0620@xmath58003 & @xmath65 & @xmath42 & @xmath66 & @xmath67 & @xmath68 & @xmath69 & @xmath70 & @xmath71 + gs2000@xmath40251 & @xmath65 & @xmath42 & @xmath72 & @xmath73 & @xmath68 & - & @xmath74 & @xmath71 + + gro j1655@xmath5840 & @xmath41 & @xmath75 & @xmath76 & @xmath77 & @xmath45 & @xmath78 & @xmath79 & @xmath47 + 4u 1543@xmath5847 & @xmath41 & @xmath75 & @xmath80 & @xmath81 & @xmath45 & @xmath82 & @xmath83 & @xmath47 + xte j1550@xmath58564 & @xmath65 & @xmath75 & @xmath84 & @xmath85 & @xmath10 & @xmath86 & @xmath87 & @xmath88 + gs 2023@xmath40338 & @xmath65 & @xmath75 & @xmath89 & @xmath90 & @xmath10 & - & @xmath91 & @xmath88 + xte j1819@xmath58254 & @xmath51 & @xmath65 & @xmath92 & @xmath93 & @xmath94 & - & @xmath95 & @xmath96 + grs 1915@xmath40105 & @xmath51 & @xmath65 & @xmath97 & @xmath98 & @xmath94 & @xmath99 & @xmath100 & @xmath96 + cyg x@xmath581 & @xmath51 & @xmath101 & @xmath102 & @xmath103 & @xmath104 & @xmath105 & @xmath106 & @xmath107 + + lmc x@xmath581 & @xmath108 & @xmath109 & @xmath110 & @xmath111 & @xmath112 & @xmath113 & @xmath114 & @xmath115 + lmc x@xmath583 & @xmath116 & @xmath117 & @xmath118 & @xmath119 & @xmath120 & @xmath121 & @xmath122 & @xmath123 + m33 x@xmath587 & @xmath124 & @xmath125 & @xmath126 & @xmath127 & @xmath112 & @xmath128 & @xmath129 & @xmath130 + [ tab - results ] as early as @xcite but more in form in @xcite and @xcite , we studied seven galactic sources with nuclear - evolved ( i.e. , beyond main sequence ) companion stars ( in all cases @xmath131 ) , eight galactic sources with main sequence companion ( all of them with @xmath132 ) .
we further studied three extragalactic sources @xcite with massive companions .
we reconstructed the configuration at the pre - collapse time from the present - day configuration in order to obtain the kerr parameters present at black hole formation .
this was done with the idea of obtaining the energies available to the central engine which may have produced a grb / hn - like event .
it is important to notice that this reconstruction can be tricky in more ways than one . in systems like the eight galactic black - hole binaries with main - sequence companions ,
it is hard to know how much the orbit may have decayed ( due to gravitational waves , magnetic torques , etc . ) since the formation of the black hole because such low mass companions live for several giga - years and thus the black - hole binary may be that old . if such is the case , we can expect our results to be upper limits to the kerr parameters and available energies of the black - hole binaries . in the case of systems like nova sco ( gro j1655@xmath5840 ) , il lupi ( 4u 1543@xmath5847 ) , cyg v404 ( gs 2023@xmath40338 ) , xte j1550@xmath58564 , sag v4641 ( xte j1819@xmath58254 ) , grs 1915@xmath40105 or lmc x@xmath583 the story may be a bit more intricate but easier to reconstruct . in these systems
considerable mass transfer may have taken place after the black hole was formed , altering the original @xmath24 . nonetheless , the plot on the right - hand side of fig .
[ fig:1 ] provides help in obtaining the corrected value of the kerr parameter and with some simplifications ( e.g. , conservative mass transfer ) one can estimate the pre - collapse orbital period of the binary . in general , we find that the pre - collapse orbital period of the binary , @xmath133 ( in days ) , does not depend strongly on the mass of the primary but it does depend strongly on the mass of the companion : p_b ( ) ^0.83 ( ) ^3/2 .
thus , in general , a smaller companion results in a shorter orbital period , a larger kerr parameter and more available rotational energy . this situation does not , however , necessarily imply a more energetic explosion or a brighter gamma - ray burst during the formation of the black hole in a system with a low - mass companion star . in fact ,
a `` goldilocks '' scenario may be at play .
in such a scenario , a system with a massive companion will not evolve into a very short - orbital period binary and so it will not develop a rapidly spinning black hole nor an accretion disk , thus all or most of the massive primary will collapse into the black hole ( remember that as we are using the collapsar model we assume that the initial supernova is not successful ) . on the other side of the companion - mass spectrum , a system with a rather small companion ( but still massive enough to remove the envelope and spiral in ) , may produce an extreme kerr black hole . thus , with an extremely large amount of rotational energy the kerr black hole may blow away the accretion disk too quickly to launch the gamma - ray burst . such black hole should be observed with a kerr parameter close to what it was born with given that it will not tap into its rotational - energy reserves for too long as the central blandford - znajek engine will be rapidly blown apart by an energetic hypernova .
hence , only a black hole born with the right heritage of rotational energy will be able to tap into its energy resources such as to launch a long gamma - ray burst and deplete a substantial amount of the kerr parameter from the black hole .
as recent measurements of kerr parameters @xcite confirm , our results in at least three of our more trustworthy studied systems ( gro j1655@xmath5840 , 4u 1543@xmath5847 , xte j1550@xmath58564 and maybe even lmc x@xmath583 ) , show that black holes with large rotational energies may easily blow away their accretion disks , and thus , prevent the central engine from triggering a gamma - ray burst
. it may also be that the gamma - ray burst is not fully prevented but it becomes , instead , subluminous .
such situation may be similar to that of grb 980425/sn 1998bw where the gamma - ray burst was subluminous but the hypernova was extremely energetic .
further evidence of this may be that , as @xcite have found , the companion of nova sco seems to have been polluted by material from the hypernova explosion at the time of black hole formation but still , the black hole seems to have little of its original spin . in @xcite
we make the case that a companion with mass near @xmath4 may be on the right ballpark in order to produce a cosmological ( i.e. , a bright ) gamma - ray burst .
thus , lmc x@xmath583 may be the best local candidate of a black - hole binary which is a remnant of a grb / hn explosion . and
indeed , our estimate for the kerr parameter of this black hole is lower but not too distant from the observational estimate of @xcite . as mentioned above
, our estimate is for the initial kerr parameter , before powering a grb / hn explosion ; it is reasonable to expect a measurement below this value if the black hole powered such an event .
further tests that such event occurred could come from measuring the system velocity of the lmc x@xmath583 binary .
this would be useful because we know the binary lost mass asymmetrically given that , even if the explosion were spherically symmetric , it occurs off the center of mass of the binary .
hence the system received a blaauw - boersma kick @xcite and thus the system was kicked with a velocity proportional to the mass loss ( see , e.g. , the discussion in app .
a of * ? ? ?
* ) . from the energetics ( obtained from the kerr parameter ) ,
our estimate is that somewhere between 3 and @xmath134 were lost during the explosion and the system acquired a velocity of @xmath135 km s@xmath136 . to conclude with the set of seven black - hole binaries with nuclear evolved companions the difference between the observed and predicted values for the kerr parameter in grs 1915@xmath40105 must be explained . in @xcite
it is proposed that cyg x@xmath581 , v4641 sag ( xte j1819@xmath58254 ) and grs 1915@xmath40105 are all binary systems at different stages of a similar evolutionary track .
that is , the companion star is more massive than the primary at the time of the collapse into the black hole ( although xte j1819@xmath58254 and grs 1915@xmath40105 may not necessarily have ever had a companion star as massive as the one in cyg x@xmath581 ) .
as the companion star evolves it eventually starts transferring mass to the black hole .
there can be a couple of short phases but with substantial stable - mass - transfer rates ( @xmath137 yr@xmath136 and @xmath138 ) as modeled by @xcite , thus allowing for hypercritical accretion of several solar masses into the black hole .
in these binaries even conservative mass transfer will expand the orbit when the mass ratio is inverted ( as is the case of xte j1819@xmath58254 and grs 1915@xmath40105 ) , that is , when the less massive companion transfers mass to the more massive black hole . when the original mass of the black hole ( @xmath139 ) is doubled , the kerr parameter of the black hole reaches its maximum value @xmath140 ( see the right plot of fig .
[ fig:1 ] ) . at the same time , the orbital period becomes longer , growing from a couple of days to over 30 days which is the situation presently observed in grs 1915@xmath40105 .
this can be easily seen from the relation of the orbital periods to the system masses for conservative mass transfer : = = ( ) ^3 = ( ) ^3 , where @xmath23 is the mass of the black hole , @xmath141 is the mass of the donor ( companion ) star and the subindices 1819 and 1915 represent the corresponding present orbital periods and masses for the xte j1819@xmath58254 and grs 1915@xmath40105 systems , respectively .
the actual mass of the companion star of grs 1915@xmath40105 is actually not @xmath142 but rather @xmath143 , but our restriction of conservative mass transfer may also be somewhat loosened to account for some mass loss .
it must also be noted that the natal mass of the black hole is @xmath144 while the companion star has @xmath145 in this example , whereas in xte j1819@xmath58254 the ratio is inverted .
nonetheless , for conservative mass transfer , during this inversion of the masses , the orbital period would initially decrease and later it would increase leaving it unchanged .
thus , grs 1915@xmath40105 makes a good case in favor of a black hole being spun up by hypercritical mass accretion .
on the other end of the companion - mass spectrum , we have three black - hole binaries with massive companions .
in fact , in all these hmxbs ( high - mass x - ray binaries ) the companions are presently more massive than the black holes themselves .
these binaries are the eclipsing system m33 x@xmath587 , lmc x@xmath581 ( both of them extra - galactic ) and cyg x@xmath581 ( galactic ) . from the discussion above
, it is expected that such systems will have wider orbits at the time of the collapse .
this is also rather intuitive , as more massive stars will have larger radii and , therefore , will only fit in binaries with larger orbital separations .
we would expect that massive binaries which end up in much shorter orbits would merge rather quickly and would not be observable for long .
hence , one expects the binaries which survive the mergers and , therefore , are observable to have rather large orbital separations , large black hole masses ( due to the lack of energy to produce an explosion after the initial supernova fails ) and small values on their natal kerr parameters .
however , it has been found that these three systems ( m33 x@xmath587 , lmc x@xmath581 and cyg x@xmath581 ) possess black holes with rather large kerr parameters ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* who measure the values of @xmath146 , @xmath147 and @xmath105 , respectively for these sources as shown in table [ tab - results ] ) .
this is rather interesting on several aspects .
first , the necessary orbital periods for these binaries to transfer enough angular momentum to their black hole progenitors so they are born with these kerr parameters are below @xmath148 days ( for @xmath149 ) and @xmath150 days ( for @xmath151 ; see the left plot of fig .
[ fig:1 ] ) .
this implies orbital separations smaller than the stellar radii of any of the two stars in these binaries : @xmath152 , where @xmath153 is the zams radius of a star of at least 20 @xmath3 .
second , if we want to assume the black holes acquired these kerr parameters after they were formed , one must assume that several solar masses were transferred into the black hole in a few million years ( given the short lifetime of such massive companions ) .
this implies that the accretion onto the black hole was at least a factor of @xmath154 above the eddington limit , most likely hypercritical ( hypercritical accretion occurs for @xmath155 where @xmath156 is the eddington - luminosity - limited accretion rate ) . and ,
given that roche - lobe overflow ( rlof ) mass transfer would be unstable , the mass transfer likely occurred through a focused wind ( or wind roche - lobe overflow ; * ? ? ?
* ) as suggested in @xcite and @xcite . indeed ,
cyg x@xmath581 has been observed @xcite to have both , radial and bh - focused wind components .
third , if one assumes that black holes can not be born with such a large kerr parameter nor can they accrete such a large amount of mass and angular momentum in order to obtain it after they are formed ( due to eddington - limit restrictions and/or the instabilities in the mass transfer ) , then one concludes that the spin must have been acquired during the formation of the black hole .
this is the case in @xcite .
they propose the mechanism of @xcite as a way to explain the large kerr parameter in cyg x@xmath581 .
however moreno mndez and cantiello ( 2012 , in preparation ) seem to find that such scenario is unlikely in massive black holes .
this is because conservation laws ( for energy and angular - momentum ) pose extreme requirements on the conversion efficiency of gravitational binding energy into rotational energy .
fourth , if these kerr parameters were natal , the available energies for the blandford - znajek engine would be on the order of 1,300 to over 3,000 bethes ( 1 b @xmath157 1 bethe @xmath157 1 foe @xmath158 erg , i.e. the typical luminous energy of a supernova ) for these black holes .
most likely , such energies would be rapidly delivered to the accretion disk , quickly blowing it away , preventing the central engine from working long enough to launch a gamma - ray burst but probably producing an extremely luminous and energetic hypernova .
most likely the companions would show metal enrichment , not unlike that of nova sco .
given that these binaries should be younger than nova sco , they should have not had the time to _ hide _ the traces of this material by mixing it in .
it is also likely that black holes formed in such energetic fashion would lose a large amount of material early during their formation and , thus , would not grow as massive as they have .
were this the case it is likely they would break the binary apart ( if half the mass of the pre - collapse binary was lost ) , or , at the very least , acquire large system velocities , unlike what seems to be the case for cyg x@xmath581 according to the measurements by @xcite and @xcite .
& @xmath159 & model & @xmath160 & z & references .
+ & @xmath161 $ ] & @xmath161 $ ] & [ days ] & @xmath37 & [ bethes ] & @xmath162 $ ] & + + ic 10 x@xmath581 & @xmath163 & @xmath109 & @xmath164 & @xmath165 & @xmath166 & @xmath167 & 1 , 2 , 3 + ngc 300 x@xmath581 & @xmath168 & @xmath169 & @xmath170 & @xmath171 & @xmath166 & @xmath68 & 3 + + cyg x@xmath583 & @xmath172 ? & @xmath173 ? & @xmath174 & @xmath175 & @xmath176 & @xmath177 ?
& 4 , 5 , 6 , 7 + [ tab : wr - bh ] in any case , it results interesting to know whether other black - hole binaries with massive companions have such large kerr parameters .
so far , two more such systems have been reported , mainly differing from the above three by having wolf - rayet ( wr ) companions and smaller orbital separations .
it was found by @xcite and later confirmed by @xcite , that ic 10 x@xmath581 has a wolf - rayet companion in a @xmath178-hour orbit .
the corresponding masses estimated at @xmath179 and @xmath180 ( from spectral fitting ; see table [ tab : wr - bh ] ) . even for a more conservative mass estimate of the wolf - rayet companion star of @xmath181 , the black hole appears to have a mass @xmath182 .
similarly , according to @xcite the system formed by ngc 300 x@xmath581 and wr # 41 consists of either a @xmath183 black hole and a companion wolf - rayet star of @xmath184 , or , if wr # 41 provides half of the visual continuum flux , then the masses are @xmath185 and @xmath186 respectively . the black hole and the wolf - rayet orbit each other with an orbital period of @xmath187 hours . even assuming conservative - mass transfer between the companion stars and the black holes of ic 10 x@xmath581 and ngc 300 x@xmath581 , the orbital separations could not have been much smaller than what they presently are given that both systems possess such massive companions .
thus one would expect to measure kerr parameters for these two black holes around @xmath188 ( for the lower estimates of the black hole masses ) .
that is , of course , unless substantial mass has been accreted by the black holes through some form or another of roche - lobe overflow . in which case it would not be too surprising to find them with large kerr parameters like m33 x@xmath587 , lmc x@xmath581 and cyg x@xmath581 .
another interesting question is whether any of these wr - bh systems may produce a grb / hn ( and a bh - bh binary ) from the collapse of the massive companion star .
were the wolf - rayet stars tidally locked to the black holes , and were they to collapse in the near future , they may produce gamma - ray bursts . for the present orbital period values we could expect an @xmath189 for the formation of a black hole with @xmath190 , or @xmath191 for a black hole with @xmath192 .
these numbers translate into available energies between @xmath193 and 285 bethes .
however , if the wolf - rayet stars take too long to collapse , they may lose / transfer too much mass , thus increasing the orbital separation and slowing the rotation of the wolf - rayet ( both , through mass / angular momentum loss , and through tidal synchronization ) .
there is another wolf - rayet binary which may also be a possible source from where a grb / hn may be observed .
the wolf - rayet ( v 1521 cyg ) in cyg x@xmath583 orbits a compact object with an orbital period of @xmath194 hours .
cygnus x@xmath583 has recently gone through a series of studies which suggest that it harbors a black hole ( rather than a neutron star ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the present day orbital period translates to a kerr parameter @xmath195 almost regardless of the mass collapsing into a black hole , assuming such an object collapses into a black hole without losing a substantial amount of angular momentum .
the corresponding available energy would be @xmath196 bethes . if we consider that a black hole with a mass of @xmath197 forms , this implies an energy of @xmath198 bethes .
nonetheless , with such a large amount of available rotational energy , a substantial amount of the star may be blown away before producing a gamma - ray burst or forming a black hole .
nonetheless , the hypernova that may result from such a system may be extremely energetic .
in this proceeding the work presented in @xcite and @xcite is summarized . in these papers , a model of stellar binary evolution is put forward where the primary star evolves as a single star until helium - shell burning begins .
this allows the formation of a massive stellar core which eventually collapses to form a black hole . at this point
, the red giant star expands an extra 10 to @xmath199 and begins a phase of unstable case - c mass transfer ( via roche - lobe overflow ) to the companion star .
the binary orbit shrinks until the companion star plunges into the envelope of the primary star and , thus , a phase of common - envelope evolution begins .
the hydrogen envelope of the primary star is ejected at the expense of losing energy from the orbit of the binary . when the orbital separation , @xmath8 , has been reduced from @xmath200 to only a few solar radii
the common envelope is finally removed .
the exposed massive core of the primary star then becomes tidally synchronized with the companion star , hence it is spun up .
a strong magnetic field allows for transport of angular momentum to the core resulting in a star that rotates as a solid body .
knowledge of the orbital period before the collapse of the primary star into a black hole allows us to produce an estimate of the available angular momentum and , therefore , the kerr parameter and the rotational energy of the black hole .
this is the same knowledge which we use to asses whether an observed black - hole binary may be a relic of a grb / hn explosion .
our results seem to get observational confirmation from the measurements by @xcite and @xcite for the kerr parameters of gro j1655@xmath5840 ( nova sco ) , 4u 1543@xmath5847 ( il lupi ) and xte j1550@xmath58564 . for these three black - hole binary systems ,
our predictions fall well within the margin of error of the measurements .
our results are also consistent with the measurement of the kerr parameter estimated by @xcite for lmc x@xmath583 .
even when our predicted value is somewhat above the measurement estimate , we propose this system as a candidate of a relic of a cosmological - type of grb / hn explosion , the missing angular momentum from the observations being compensated by the rotational energy spent by the black hole in producing the explosive event at its birth .
our results seem to fall short from the observations on all the hmxbs , where we predict low natal kerr parameters and , instead , the observations show large kerr parameters . nonetheless , we propose that these systems have undergone a phase of substantial mass transfer from the companion star to the black hole where the kerr parameter of the later has been dramatically increased . in at least one system ,
grs 1915@xmath40105 , the companion star seems to have lost most of its original mass to the black hole and , as a result , the orbital period of the binary , as well as the kerr parameter of the black hole , have increased dramatically from the time of the collapse .
we propose that xte j1819@xmath58254 may be a binary in an intermediate phase between a system similar to cyg x@xmath581 and one similar to grs 1915@xmath40105 .
it appears that the natal kerr parameter of stellar mass black holes can be altered by two mechanisms .
the first one is during the gravitational collapse phase ; the rotational energy of the forming black hole can be tapped through the blandford - znajek mechanism to produce a grb / hn event , thus lowering the value of the kerr parameter one observes in a black hole as seems to be the case in lmc x@xmath583 ( where , nonetheless , we do not expect a drastic decrease in rotation due to the fact that the energy necessary to produce even a cosmological grb / hn kind of event is only a small fraction of the available rotational energy ) .
the second mechanism involves accreting a substantial amount of mass and angular momentum into the black hole once the companion star or its wind fill their roche lobe . in some of the black - hole binary systems
there are hints of past energetic events consistent with grb / hn explosions .
the case of nova sco and its polluted companion is one such interesting example .
close observation of the composition of black hole companions in other systems remains highly desirable .
lmc x@xmath583 may well be another interesting object as it may well be the only local relic we may study from a cosmological grb / hn event . even if the distance to this black - hole binary may prevent a detailed study of the composition of the companion , further constraints on the explosion of this system may be placed through the blaauw - boersma kick @xcite as well as the fact that the binary system remained bound .
it is intriguing to consider whether other observable binaries ( in particular black - hole binaries such as those of ic 10 x@xmath581 , ngc 300 x@xmath581 or even the galactic system cyg x@xmath583 ) may harbor progenitors of near - future , nearby grbs / hne .
j. a. orosz , d. steeghs , j. e. mcclintock , m. a. p. torres , i. bochkov , l. gou , r. narayan , m. blaschak , a. m. levine , r. a. remillard , c. d. bailyn , m. m. dwyer , and m. buxton , _ _ * 697 * , 573591 ( 2009 ) , .
j. a. orosz , `` inventory of black hole binaries , '' in _ a massive star odyssey : from main sequence to supernova _
, edited by k. van der hucht , a. herrero , & c. esteban , 2003 , vol .
212 of _ iau symposium _ ,
p. 365 , .
s. mohamed , and p. podsiadlowski , `` wind roche - lobe overflow : a new mass - transfer mode for wide binaries , '' in _
15th european workshop on white dwarfs _ , edited by r. napiwotzki & m. r. burleigh , 2007 , vol .
372 of _ astronomical society of the pacific conference series _ , p. 397 . | in this work i summarize a model of binary stellar evolution involving case c mass transfer followed by a common envelope that strips away the hydrogen from the core of the primary star at the cost of shrinking the orbital separation and then , through tidal interaction , spins it up .
this model is then used to produce the possible progenitors of long gamma - ray burst / hypernova ( grb / hn ) explosions . as the core collapses with the newly supplied angular momentum it produces a kerr black hole surrounded by an accretion disk .
energy is extracted from the rotation of the black hole ( bh ) through the blandford - znajek ( bz ) mechanism to power both , the long gamma - ray burst and the accompanying hypernova ( supernova type ic broad line ) . if the binary survives the asymmetric mass loss its remnant is a black - hole binary that may eventually be observed as a soft x - ray transient ( sxt ) when the companion evolves and starts to transfer mass back to the black hole . a comparison with a sample of black - hole binaries where the masses and orbital periods are well constrained is performed . |
all simplicial complexes considered in this paper are finite and abstract .
all homologies are simplicial homologies with coefficients in a field @xmath0 .
the vertex set of a simplicial complex @xmath4 will be denoted by @xmath5 . for @xmath6 ,
the induced subcomplex @xmath7 $ ] of @xmath4 on the vertex set @xmath8 is defined by @xmath7 : = \{\alpha\in x \ , : \ , \alpha\subseteq a\}$ ] .
a simplicial complex @xmath4 is said to be a _ triangulated manifold _
if it triangulates a manifold , i.e. , if the geometric carrier @xmath9 of @xmath4 is a topological manifold .
a triangulated closed @xmath10-manifold @xmath4 is said to be _
@xmath0-orientable _ if @xmath11 .
so , for a field @xmath0 of characteristic two , any triangulated closed manifold is @xmath0-orientable .
taking his cue from pre - existing notions of tightness in the theory of smooth and polyhedral embedding of manifolds in euclidean spaces , khnel @xcite introduced the following precise notion of tightness of a simplicial complex with respect to a field .
[ def : tight ] let @xmath4 be a simplicial complex and @xmath0 be a field .
we say that @xmath4 is tight with respect to @xmath0 ( in short , @xmath0-tight ) if ( a ) @xmath4 is connected , and ( b ) for every induced subcomplex @xmath12 of @xmath4 , the @xmath0-linear map @xmath13 ( induced by the inclusion map @xmath14 ) is injective .
recall that , if @xmath4 is a simplicial complex of dimension @xmath10 , then its _ face numbers _ @xmath15 are defined by @xmath16 . for @xmath17 ,
a simplicial complex @xmath4 is said to be _
@xmath18-neighbourly _ if any set of @xmath18 vertices of @xmath4 form a face , i.e. , if @xmath19 .
a 2-neighbourly simplicial complex is called _ neighbourly_. [ def : sminimal ] a simplicial complex @xmath4 is said to be strongly minimal if , for every triangulation @xmath12 of the geometric carrier @xmath9 of @xmath4 , we have @xmath20 for all @xmath21 , @xmath22 .
thus , a strongly minimal triangulation of a topological space , if it exists , is the most economical among all possible triangulations of the space .
unfortunately , there are very few criteria available in the literature which ensure strong minimality .
the notion of tightness is of great importance in combinatorial topology because of the following tantalizing conjecture @xcite .
[ conj : kl ] every @xmath0-tight triangulated closed manifold is strongly minimal .
intuitively , @xmath0-tightness of a triangulated manifold @xmath4 means that all parts of @xmath4 are essential in order to capture the @xmath0-homology of the topological space @xmath9 . in view of this intuition ,
conjecture [ conj : kl ] appears to be entirely plausible .
however , example [ exam : nsmtight ] below shows that this intuition is not correct for arbitrary simplicial complexes .
recall the following from @xcite .
[ def : stacked ] a triangulated manifold @xmath23 of dimension @xmath24 is said to be _ stacked _
if all its faces of codimension ( at least ) two are in the boundary @xmath25 .
a triangulated closed manifold @xmath26 of dimension @xmath10 is said to be _ stacked _ if there is a stacked triangulated manifold @xmath23 of dimension @xmath24 such that @xmath27 . in particular , a _ stacked sphere
_ is a triangulated sphere which may be realized as the boundary of a stacked triangulated ball .
[ def : lstacked ] a triangulated manifold is said to be _ locally stacked _ if all its vertex links are stacked spheres or stacked balls .
clearly , all stacked triangulated manifolds are locally stacked , but the converse is false ( see example [ exam : lutz ] below ) . due to
* theorem 2.24 ; case @xmath28 ) we have the following . [ prop : bdth2.24 ] let @xmath26 be a locally stacked @xmath0-orientable , neighbourly , triangulated closed @xmath29-manifold
. then the following are equivalent 1 .
@xmath26 is @xmath0-tight , 2 .
@xmath26 is stacked , and 3 .
@xmath30 .
thus , all orientable , neighbourly , stacked triangulated closed 3-manifolds are @xmath0-tight ( for any field @xmath0 ) .
even more , we know from @xcite [ prop : sminimal ] every @xmath0-tight , locally stacked , triangulated closed manifold is strongly minimal .
the first main result of this paper ( in section 4 ) is a converse of proposition [ prop : bdth2.24 ] : if a triangulated closed 3-manifold is tight with respect to a field of characteristic @xmath31 , then it must be ( orientable , neighbourly and ) stacked .
this result answers question 4.5 of @xcite affirmatively , in the case of odd characteristic . as a consequence of proposition [ prop : sminimal ]
it follows that the khnel - lutz conjecture is true in a special case , namely , if @xmath32 then any @xmath0-tight triangulated closed 3-manifold is strongly minimal .
let @xmath33 and @xmath34 be two triangulated @xmath10-manifolds intersecting in a common facet ( @xmath10-face ) @xmath35 .
that is , @xmath36 , where @xmath37 denotes @xmath35 together with all of its subfaces .
then @xmath38 is said to be the _ connected sum of @xmath33 and @xmath34 along @xmath35 _
( for more general definition , see the end of section 2 ) .
let @xmath39 be the boundary complex of the icosahedron .
thus , @xmath40 is a triangulated 2-sphere on 12 vertices .
it is well known that @xmath40 is the unique triangulation of @xmath41 in which each vertex is of degree 5 .
we introduce : [ def : icosian ] a triangulated @xmath42-sphere is said to be _ icosian _ if it is a connected sum of finitely many copies of @xmath40 .
a triangulated @xmath29-manifold is said to be _ locally icosian _ if all its vertex links are icosian . in @xcite ,
the third author proved the following interesting upper bound theorem for tight triangulations of odd dimensional manifolds .
[ prop : spreer ] let @xmath26 be an @xmath43-connected triangulated closed @xmath44-manifold and @xmath0 be a field . if @xmath26 is @xmath0-tight , then @xmath45 . in section 5 , we consider fields of characteristic 2 .
according to proposition [ prop : bdth2.24 ] , every neighbourly and stacked closed 3-manifold @xmath26 is @xmath46-tight .
we do not know whether the converse is true or not .
but in this paper we prove that if @xmath26 is a @xmath46-tight triangulated closed 3-manifold with @xmath47 or @xmath48 , then @xmath26 must be stacked and neighbourly ( and therefore @xmath49 ) .
we also show that , in general , each vertex link of a @xmath46-tight closed 3-manifold must be a connected sum of @xmath40 s and @xmath50 s .
further , we prove that any @xmath46-tight triangulated closed 3-manifold @xmath26 satisfies the following upper bound on @xmath51 : @xmath52 equality holds here if and only if @xmath26 is locally icosian . in conjunction with the fact that @xmath49 when @xmath53 , this inequality improves upon the upper bound of proposition [ prop : spreer ] in case @xmath54 .
we also prove that , if there is a non - stacked @xmath0-tight triangulated 3-manifold @xmath26 , then its integral homology group @xmath55 must have an element of order 2 .
the results of section 5 were largely suggested by extensive machine computations using simpcomp @xcite .
altogether , these results impose severe restrictions on the topology of @xmath29-manifolds admitting tight triangulations ( cf .
corollary [ coro:5.14 ] ) . in section 6 , we present some examples to show that the converses / generalizations of several results proved here are not true .
in this section , we gather together a few easy ( and mostly known ) consequences of tightness . for completeness
, we include their proofs .
we shall use : [ notation1 ] if @xmath56 is a vertex of a simplicial complex @xmath4 , then @xmath57 and @xmath58 will denote the antistar and the link ( respectively ) of @xmath56 in @xmath4 .
thus , @xmath59 , \nonumber \\ x_x & : = \{\alpha\in x : x\not\in\alpha , \alpha\sqcup\{x\}\in x\}. \nonumber\end{aligned}\ ] ] we denote a face @xmath60 in a simplicial complex by @xmath61 .
if @xmath4 is a simplicial complex and @xmath62 is an element then the _ cone _ with apex @xmath63 and base @xmath4 is the simplicial complex @xmath64 and is denoted by @xmath65 . for a simplicial complex @xmath4 of dimension @xmath10 , and for @xmath66 ,
the _ @xmath18-skeleton _
@xmath67 is defined by @xmath68 [ lemma:2.2 ] every @xmath0-tight simplicial complex is neighbourly .
suppose , if possible , @xmath69 are two vertices of an @xmath0-tight simplicial complex @xmath4 such that @xmath70 is not an edge of @xmath4 .
let @xmath12 be the induced subcomplex of @xmath4 on the set @xmath71 .
then @xmath72 , so that @xmath73 can not be injective .
this is a contradiction since @xmath4 is @xmath0-tight .
[ lemma:2.3 ] every induced subcomplex of an @xmath0-tight simplicial complex is @xmath0-tight .
let @xmath12 be an induced subcomplex of an @xmath0-tight simplicial complex @xmath4 .
by lemma [ lemma:2.2 ] , @xmath4 and hence also @xmath12 are neighbourly .
so , @xmath12 is connected .
let @xmath74 be an induced subcomplex of @xmath12 .
then @xmath74 is an induced subcomplex of @xmath4 also .
since the composition of the linear maps @xmath75 is injective , the first of them must be injective .
thus , @xmath76 is injective for all induced subcomplexes @xmath74 of @xmath12 .
therefore , @xmath12 is @xmath0-tight .
[ lemma:2.4 ] if @xmath4 is an @xmath0-tight simplicial complex of dimension @xmath10 , then @xmath67 is @xmath0-tight for @xmath77 .
since @xmath78 and @xmath4 is neighbourly by lemma [ lemma:2.2 ] , it follows that @xmath67 is neighbourly and hence connected .
let @xmath12 be an induced subcomplex of @xmath67 .
then @xmath79 , where @xmath74 is an induced subcomplex of @xmath4 .
since @xmath4 is @xmath0-tight , it follows that , for @xmath80 , @xmath81 is injective . clearly , @xmath82 .
since both @xmath12 and @xmath67 are of dimension @xmath83 , @xmath84 .
these imply that @xmath85 is injective .
[ lemma:2.5 ] every @xmath0-tight triangulation of a closed manifold is @xmath0-orientable .
if @xmath86 then there is nothing to prove .
so , assume that @xmath32 .
let @xmath4 be an @xmath0-tight triangulated closed @xmath10-manifold .
we can assume that @xmath87 .
since @xmath4 is a connected triangulated closed @xmath10-manifold , it is easy to see ab initio that , for any proper subcomplex @xmath12 of @xmath4 , @xmath88 .
now , fix a vertex @xmath56 of @xmath4 , and consider the induced subcomplex @xmath57 of @xmath4 on the complement of @xmath56 . then @xmath89 and @xmath90 . since @xmath91 , @xmath92 , @xmath93 and @xmath94 is injective
, it follows from the exactness of the mayer - vietoris sequence @xmath95 that @xmath96 .
let @xmath4 be a simplicial complex of dimension @xmath10 .
in @xcite we have defined the sigma vector @xmath97 of @xmath4 with respect to a field @xmath0 as @xmath98)\left/\binom{f_0(x)}{\#(a)}\right . , \
, 0\leq i \leq d , \nonumber\end{aligned}\ ] ] and @xmath99 for @xmath100 . in this paper , we introduce the following normalization of the sigma vector which will be referred to as the _ sigma - star _ vector .
@xmath101 ( see theorem [ theorem:5.7 ] below for a justification of this normalization . ) in @xcite the first author introduced the mu - vector @xmath102 w.r.t .
@xmath0 of a @xmath10-dimensional simplicial complex @xmath4 as follows .
@xmath103 ( here @xmath104 is kronecker s symbol .
thus , @xmath105 if @xmath106 , and @xmath107 otherwise . )
notice that @xmath108 is independent of the field @xmath0 .
therefore , we will write @xmath109 for @xmath108 .
we have : [ lemma:2.6 ] let @xmath26 be an @xmath0-orientable , neighbourly triangulated closed @xmath29-manifold
. then @xmath110 .
equality holds here if and only if @xmath26 is @xmath0-tight .
the inequality ( as well as the fact that equality holds if @xmath26 is @xmath0-tight ) is a very special case of ( * ? ? ?
* corollary 1.8 ) .
suppose @xmath111 .
since @xmath26 is 2-neighbourly , it also satisfies @xmath112 . by (
* theorem 1.6 ) , @xmath113 and @xmath114 . also , since @xmath26 is @xmath0-orientable , poincar duality implies @xmath115 and @xmath116 .
therefore , @xmath117 and @xmath118 .
hence ( * ? ? ?
* corollary 1.8 ) implies that @xmath26 is @xmath0-tight .
recall that , a triangulated closed @xmath10-manifold @xmath4 is called _ orientable _ if @xmath119 .
it follows from the universal coefficient theorem that , for a field @xmath0 of odd characteristic , a triangulated closed manifold is @xmath0-orientable if and only if it is orientable . [ coro:2.7 ] let @xmath120 be a prime and let @xmath26 be an orientable triangulated closed @xmath29-manifold . if @xmath26 is @xmath121-tight but not @xmath122-tight , then @xmath120 divides the order of the torsion subgroup of @xmath123 .
the hypothesis and lemma [ lemma:2.6 ] imply that @xmath124 .
hence the result follows from the universal coefficient theorem .
the following lemma is immediate from the definition of tightness .
[ lemma:2.8 ] ( a ) a simplicial complex is tight w.r.t .
a field of characteristic @xmath120 if and only if it is @xmath121-tight .
( b ) a simplicial complex is tight w.r.t .
a field of characteristic zero if and only if it is @xmath121-tight for all primes @xmath120 . for any non - empty finite set @xmath35 , @xmath125 will denote the simplicial complex whose faces are all the subsets of @xmath35 .
thus , if @xmath126 , @xmath125 is the standard triangulation of the @xmath10-ball ( namely , it is the face complex of the geometric @xmath10-simplex ) .
if @xmath127 , then the boundary @xmath128 of @xmath125 is the standard triangulation of the @xmath10-sphere ; it will be denoted by @xmath129 .
( more generally , @xmath130 usually denotes an @xmath1-vertex triangulation of the @xmath10-sphere . ) from the definition of a stacked sphere , one can deduce the following ( this also follows from ( * ? ? ?
* lemmas 4.3 ( b ) & 4.8 ( b ) ) ) .
[ lemma : stacked_sphere ] a simplicial complex @xmath131 is a stacked @xmath10-sphere if and only if @xmath131 is a connected sum of finitely many copies of the @xmath132-vertex standard sphere @xmath129 .
let @xmath4 be a triangulated closed manifold and let @xmath133 and @xmath134 be facets of @xmath4 . for a bijection @xmath135 ,
let @xmath136 be the simplicial complex obtained from @xmath137 by identifying @xmath138 and @xmath139 for @xmath140 . if @xmath141 for each vertex @xmath140 , then @xmath136 is a triangulated manifold .
if @xmath133 and @xmath134 belong to different connected components , say @xmath142 , @xmath143 and @xmath144 , then @xmath145 is said to be the _ connected sum _ of @xmath33 and @xmath34 and is denoted by @xmath146 . if @xmath133 and @xmath134 belong to the same connected component of @xmath4 , then @xmath136 is said to be obtained from @xmath4 by a _
combinatorial handle addition_. we know from @xcite [ prop : walkup_class ]
let @xmath23 be a connected triangulated closed manifold of dimension @xmath87 .
then @xmath23 is stacked if and only if @xmath23 can be obtained from a stacked @xmath10-sphere by a sequence of successive combinatorial handle additions .
let @xmath58 and @xmath57 be as in notation [ notation1 ] . if @xmath69 are two vertices of a simplicial complex @xmath4 , then we shall also use notations such as @xmath147 for @xmath148 .
we say that two vertices in a simplicial complex @xmath4 are adjacent ( or , that they are neighbours ) if they form an edge of @xmath4 .
we now introduce : [ notation:3.1 ] if @xmath69 are vertices of a simplicial complex @xmath4 , then @xmath149 will denote the number of distinct connected components @xmath150 of @xmath147 such that @xmath56 is adjacent in @xmath151 with some vertex in @xmath150 .
[ lemma:3.2 ] if @xmath4 is an @xmath0-tight simplicial complex then for all @xmath152 , we have @xmath153 . clearly , @xmath89 and @xmath90 .
therefore , the mayer - vietoris theorem yields the exact sequence ( noting that the cone @xmath154 is homologically trivial ) @xmath155 since @xmath57 is an induced subcomplex of the @xmath0-tight complex @xmath4 , the map @xmath156 is injective .
lemma [ lemma:2.2 ] implies that @xmath57 is connected and hence @xmath157 .
so , we get the short exact sequence @xmath158 hence the result
. lemmas [ lemma:2.2 ] and [ lemma:2.3 ] certainly indicate that tightness is a severe structural constraint on a simplicial complex .
so it is surprising that , beyond these two lemmas , no further structural ( combinatorial ) consequence of tightness seems to have been known .
the following lemma establishes a strong structural restriction on the 2-skeleton of an @xmath0-tight simplicial complex .
[ lemma:3.3 ] let @xmath4 be an @xmath0-tight simplicial complex for some field @xmath0 .
then , for any two distinct vertices @xmath159 of @xmath4 , we have @xmath160 . by lemma [ lemma:3.2 ] ,
since @xmath57 is also tight by lemma [ lemma:2.3 ] , applying lemma [ lemma:3.2 ] to the vertex @xmath162 of @xmath57 , we get @xmath163 .
therefore , @xmath164 .
interchanging the vertices @xmath56 and @xmath162 in this argument yields @xmath165 . since @xmath166 , we get @xmath167 but the two sides of this equation are just one less than @xmath149 and @xmath168 . hence the result .
notice that the _ graphs _ of graph theory are just the simplicial complexes of dimension @xmath169 . in this paper
, we do not use any non - trivial results from graph theory , but the language and the geometric intuition of graph theory will be useful . recall that the _ degree _ of a vertex @xmath138 ( denoted by @xmath170 ) in a simplicial complex is the number of edges ( 1-faces ) through @xmath138 .
a graph is said to be _ regular _ , if all its vertices have the same degree . for @xmath171 , the _ cycle of length @xmath1 _ ( in short , _
@xmath1-cycle _ ) is the unique connected regular graph of degree two on @xmath1 vertices .
it is the unique @xmath1-vertex triangulation of the circle @xmath172 .
an @xmath1-cycle with edges @xmath173 will be denoted by @xmath174 .
for @xmath175 , the _ path of length @xmath1 _ ( the @xmath1-path ) is the antistar of a vertex in the @xmath176-cycle . by an _ induced cycle _
, path ) in a simplicial complex @xmath4 , we mean an induced subcomplex of @xmath4 which is a cycle ( resp . , path ) .
notice that , in particular , a 3-cycle is induced in @xmath4 if and only if it does not bound a triangle ( 2-face ) in @xmath4 .
when @xmath177 , an @xmath1-cycle is induced in x if and only if it is induced in the graph @xmath178 .
a connected acyclic graph is called a _
tree_. [ lemma:3.4 ] let the link of some vertex @xmath56 in a @xmath42-dimensional @xmath0-tight simplicial complex @xmath4 be a cycle .
then @xmath4 is a triangulation of a closed @xmath42-manifold .
let @xmath179 be a cycle .
fix a vertex @xmath180 of @xmath4 .
it suffices to show that the link @xmath151 is also a cycle .
note that , since @xmath4 is neighbourly ( lemma [ lemma:2.2 ] ) , @xmath162 is a vertex of @xmath181 .
let @xmath182 and @xmath183 be the two neighbours of @xmath162 in @xmath181 .
it follows that @xmath182 and @xmath183 are the only two neighbours of the vertex @xmath56 in the graph @xmath151 .
therefore , it suffices to show that @xmath147 is a path joining @xmath182 and @xmath183 .
since @xmath184 is a cycle and @xmath185 is a path , they are both connected .
so , @xmath186 .
therefore , by lemma [ lemma:3.3 ] , @xmath187 .
that is , the vertices @xmath182 and @xmath183 ( being the two neighbours of @xmath56 in @xmath151 ) belong to the same component of @xmath147 .
thus , there is a path in @xmath147 joining @xmath182 to @xmath183 .
let @xmath188 be a shortest path in the graph @xmath147 joining @xmath182 to @xmath183 .
then , @xmath188 is an induced path in @xmath147 .
take any vertex @xmath189 in @xmath4 .
( if there is no such vertex then @xmath147 is the edge @xmath190 , and we are done . )
we will show that @xmath191 .
look at the induced subcomplex @xmath192 of @xmath4 . then @xmath193 is a path in which @xmath162 is an interior vertex .
so , @xmath194 is the disjoint union of two paths .
the vertex @xmath182 belongs to one of these two paths and @xmath183 belongs to the other .
therefore , @xmath195 . by lemma [ lemma:2.3 ]
, @xmath12 is also @xmath0-tight .
so , by lemma [ lemma:3.3 ] , @xmath196 .
that is , the neighbours @xmath182 and @xmath183 of @xmath56 in @xmath197 belong to different components of @xmath198 .
therefore , @xmath138 belongs to the path @xmath188 ( or else @xmath188 would be a path in @xmath198 joining @xmath182 and @xmath183 ) .
since @xmath199 was an arbitrary vertex of @xmath198 , this shows that the path @xmath188 is a spanning path in @xmath198 ( i.e. , it passes through all the vertices ) .
since @xmath188 is also an induced path in @xmath198 , it follows that @xmath200 is a path joining @xmath182 and @xmath183 .
now , the following result is an easy consequence of lemma [ lemma:3.4 ] .
notice that , in this result , there is no restriction on the dimension of @xmath26 , and @xmath26 need not be a triangulated manifold .
[ theorem:3.5 ] let the simplicial complex @xmath26 be tight with respect to some field .
let @xmath181 be an induced cycle in the link of a vertex @xmath56 in @xmath26 .
then the induced subcomplex of @xmath26 on the vertex set of the cone @xmath201 is a triangulated closed @xmath42-manifold .
let @xmath202 $ ] and let @xmath203 .
then , by lemmas [ lemma:2.3 ] and [ lemma:2.4 ] , @xmath4 is @xmath0-tight . clearly , @xmath4 is two dimensional and @xmath184 is a cycle .
thus , by lemma [ lemma:3.4 ] , @xmath4 is a triangulated closed 2-manifold . to complete the proof
, it suffices to show that @xmath204 , i.e. , that @xmath205 .
suppose , if possible , @xmath206 .
take a 3-face @xmath207 .
then the induced subcomplex of @xmath208 on the vertex set @xmath35 is a 4-vertex triangulated 2-sphere .
since @xmath4 is a connected and triangulated closed 2-manifold , it follows that @xmath209 , and @xmath210 .
thus @xmath181 is the 3-cycle on the vertex set @xmath211 .
but , the 2-face @xmath211 is in @xmath212 .
this contradicts the assumption that @xmath181 is an induced cycle in the link @xmath213 .
[ coro:3.6 ] let @xmath131 be the link of some vertex in an @xmath0-tight simplicial complex @xmath26 . 1 .
if @xmath214 , then @xmath131 has no induced cycle of length @xmath215 @xmath216mod @xmath217 .
2 . if @xmath32 , then @xmath131 has no induced cycle of length @xmath218 or @xmath219 @xmath216mod @xmath220 . let @xmath221 and let @xmath181 be an induced @xmath1-cycle in @xmath222 . by theorem [ theorem:3.5 ] ,
the induced subcomplex @xmath223 $ ] is an @xmath224-vertex triangulated closed 2-manifold . by lemma [ lemma:2.2 ]
, @xmath4 is neighbourly .
so , it has @xmath225 vertices , @xmath226 edges and hence @xmath227 triangles .
so , 3 divides @xmath228 , i.e. , @xmath229 ( mod 3 ) .
this proves part ( a ) .
if @xmath32 then , by lemmas [ lemma:2.3 ] and [ lemma:2.5 ] , @xmath4 is an orientable triangulated 2-manifold .
so , its euler characteristic @xmath230 is an even number .
thus , @xmath231 or 10 ( mod 12 ) .
this proves part ( b ) .
in this section , we prove that any triangulated @xmath29-manifold is tight with respect to a field of odd characteristic if and only if it is neighbourly , orientable and stacked . for this , we first need some additional preliminary results on @xmath42-sphere triangulations .
[ lemma:4.1 ] let @xmath131 be a triangulation of @xmath232 .
if @xmath131 has no induced cycle of length @xmath233 then @xmath234 .
let @xmath56 be a vertex of minimum degree in @xmath131 .
it is well known ( and easy to prove ) that the minimum degree of any triangulation of @xmath232 is at most five .
so , @xmath235 or 5 .
if @xmath236 or 5 , then @xmath237 is a 4-cycle or a 5-cycle in @xmath131 , and hence it is not induced .
so there are vertices @xmath238 in @xmath237 such that @xmath239 is an edge in @xmath131 but not in @xmath237 .
so @xmath240 is an induced 3-cycle in @xmath131 , a contradiction .
thus , @xmath241 .
say @xmath237 is the 3-cycle @xmath242 .
since this 3-cycle is not an induced cycle in @xmath131 , it follows that @xmath243
. then @xmath244 .
since a triangulation of @xmath232 can not be a proper subcomplex of another triangulation of @xmath232 , it follows that @xmath245 .
we now introduce : [ def : primitive ] a triangulated @xmath10-sphere @xmath131 is said to be _ primitive _ if it can not be written as a connected sum of two triangulated @xmath10-spheres . clearly , every triangulated @xmath10-sphere is a connected sum of finitely many primitive @xmath10-spheres .
the following lemma is more or less obvious .
[ lemma:4.3 ] let @xmath131 be a triangulated @xmath10-sphere
. then @xmath131 is primitive if and only if @xmath131 has no induced subcomplex isomorphic to @xmath246 . if @xmath131 is not primitive , then @xmath247 , where @xmath248 and @xmath249 are triangulated @xmath10-spheres .
let @xmath35 be the unique common facet of @xmath248 and @xmath249 .
then the boundary of @xmath125 is an induced @xmath246 in @xmath131 .
conversely , suppose @xmath131 has an induced @xmath246 , say with vertex set @xmath35 .
this @xmath246 divides @xmath131 into two triangulated @xmath10-balls @xmath250 and @xmath251 such that @xmath252 , @xmath253 .
put @xmath254 , @xmath255 .
then @xmath248 , @xmath249 are triangulated @xmath10-spheres and @xmath247 .
so , @xmath131 is not primitive .
the following lemma and definition clarify the meaning of connected sums of several primitive triangulated spheres .
[ notation : c_sum ] let @xmath131 be a triangulation of @xmath256 . put @xmath257
\cong s^{\hspace{.15mm}d-1}_{d+1}\}$ ] , @xmath258 , and let @xmath259 be the collection of all the induced subcomplexes of the simplicial complex @xmath260 which are primitive triangulations of @xmath256 .
let @xmath261 be the graph with vertex set @xmath259 such that @xmath262 are adjacent in @xmath261 if @xmath263 for some @xmath264 .
[ lemma : c_sum ] for any triangulated @xmath10-sphere @xmath131 we have the following . 1
. any two members of @xmath259 have at most one common facet ; if they have a common facet @xmath35 then @xmath264 .
2 . each member of @xmath265 belongs to exactly two members of @xmath259 .
3 . there is a natural bijection from @xmath265 onto the set of edges of @xmath261 .
it is given by @xmath266 , where , for @xmath267 , @xmath248 and @xmath249 are the two members of @xmath259 containing @xmath35 .
the graph @xmath261 is a tree . in consequence , @xmath268 .
induction on @xmath269 . if @xmath270 , @xmath265 is empty , so that @xmath131 is primitive by lemma [ lemma:4.3 ] .
so , let @xmath271 , and suppose that the result holds for all smaller values of @xmath272 . in this case , @xmath265 is non - empty .
take @xmath273 . by the proof of lemma [ lemma:4.3 ]
, there are triangulated @xmath10-spheres @xmath274 , @xmath275 such that @xmath276 , @xmath277 .
thus , @xmath278 .
hence it is easy to see that @xmath279 , @xmath280 and the graph @xmath261 is obtained from the disjoint union of the graphs @xmath281 and @xmath282 by adjoining a single edge from a vertex of @xmath281 to a vertex of @xmath282 . since , by induction hypothesis , the result is valid for @xmath274 and @xmath275 , its validity for @xmath131 follows .
this completes the induction .
[ def : c_sum ] let @xmath131 be a triangulated @xmath10-sphere .
any tree has a leaf ( a vertex of degree one ) and the deletion of a leaf from a non - trivial tree leaves a subtree .
therefore , the members of @xmath259 may be ordered as @xmath283 in such a way that , for each @xmath21 , @xmath284 , @xmath285 is a leaf of the induced subtree of @xmath261 on @xmath286 ( when @xmath287 , this ordering is not unique ) . for any such ordering , we write @xmath288 , and say that @xmath131 is the _ connected sum _ of the @xmath285 s .
clearly , we have @xmath289 . as a special case of ( * ? ? ?
* theorem 8.5 ) , kalai proved a nice characterization of stacked 2-spheres .
a triangulated 2-sphere @xmath131 is stacked if and only if @xmath131 has no induced cycle of length @xmath290 .
the following result is a dramatic improvement on this characterization .
[ theorem:4.7 ] let @xmath131 be a triangulated @xmath42-sphere .
then @xmath131 is stacked if and only if it has no induced cycle of length @xmath291 or @xmath292 .
write @xmath293 , where the @xmath285 s are primitive 2-spheres .
if @xmath131 is stacked then , by lemma [ lemma : stacked_sphere ] , each @xmath285 is a copy of @xmath294 .
let @xmath181 be an induced cycle of length @xmath290 in @xmath131 .
since an induced cycle of length @xmath290 in a connected sum of two triangulated manifolds must be an induced cycle in one of the summands , it follows inductively that @xmath181 is an induced cycle in one of the @xmath285 s .
since @xmath294 has no induced cycle at all , it follows that @xmath131 has no induced cycle of length @xmath290 .
this proves the only if " part . for the converse , assume that @xmath131 has no induced cycle of length 4 or 5 .
it follows that no @xmath285 has any induced cycle of length 4 or 5 .
being primitive , @xmath285 has no induced cycle of length 3 either ( lemma [ lemma:4.3 ] ) .
thus , by lemma [ lemma:4.1 ] , each @xmath285 is a copy of @xmath294 .
therefore , by lemma [ lemma : stacked_sphere ] , @xmath131 is stacked .
this proves the if " part .
now , we are ready to prove one of the main results of this paper .
[ theorem:4.8 ] a triangulated closed @xmath29-manifold @xmath26 is tight with respect to some field @xmath0 with @xmath32 if and only if @xmath26 is orientable , neighbourly and stacked . the if " part follows from proposition
[ prop : bdth2.24 ] . to prove the
only if " part , let @xmath26 be an @xmath0-tight triangulated closed 3-manifold , @xmath32 . by lemmas [ lemma:2.2 ] and [ lemma:2.5 ]
, @xmath26 is neighbourly and orientable .
it remains to show that @xmath26 must be stacked .
corollary [ coro:3.6 ] ( b ) shows that , for each vertex @xmath56 of @xmath26 , the vertex link @xmath213 is a triangulated 2-sphere with no induced 4-cycle or 5-cycle .
so , theorem [ theorem:4.7 ] implies that each @xmath213 is a stacked 2-sphere .
thus , @xmath26 is locally stacked .
the result now follows by proposition [ prop : bdth2.24 ] .
[ coro:4.9 ] let @xmath26 be a triangulated closed @xmath29-manifold .
if @xmath26 is tight w.r.t . a field @xmath0 , where @xmath295 , then @xmath26 is strongly minimal .
this result follows from theorem [ theorem:4.8 ] and proposition [ prop : sminimal ] .
by lemma [ lemma:2.8 ] , a simplicial complex is tight with respect to a field of characteristic two if and only if it is @xmath46-tight .
so , without loss of generality , we take @xmath296 in this section .
however , all the results apply equally well to arbitrary fields of characteristic two . here
, we characterize the links of @xmath46-tight triangulated @xmath29-manifolds .
this characterization is important since it leads to ( a ) severe restrictions on the size and topology of such a triangulation ( see theorem [ theorem:5.1 ] and the tables at the end of this section ) and ( b ) a polynomial time algorithm to decide tightness of @xmath29-manifolds which is described in detail in @xcite . by proposition [ prop : bdth2.24 ] ,
all stacked and neighbourly triangulated closed 3-manifolds are @xmath46-tight . here
is a partial converse .
( in case @xmath26 is orientable , this result follows from corollary [ coro:2.7 ] . )
[ theorem:5.1 ] let @xmath26 be a @xmath46-tight triangulated closed @xmath29-manifold .
if the torsion subgroup of @xmath123 is of odd order @xmath216possibly trivial@xmath297 , then @xmath26 is stacked @xmath216and neighbourly@xmath297 .
if possible , assume that @xmath26 is not stacked .
then , by proposition [ prop : bdth2.24 ] , @xmath26 is not locally stacked .
so , there exists a vertex @xmath138 whose link @xmath298 is not a stacked 2-sphere . by theorem [ theorem:4.7 ] and corollary [ coro:3.6 ] ( a )
, @xmath298 has an induced cycle @xmath181 of length 5 .
then , by theorem [ theorem:3.5 ] , the induced subcomplex @xmath299 $ ] of @xmath26 is a triangulated closed 2-manifold . since @xmath300 is an induced subcomplex of @xmath26 , by lemmas [ lemma:2.2 ] and [ lemma:2.3 ] , @xmath300 is neighbourly .
thus , @xmath300 is a 6-vertex neighbourly triangulated 2-manifold and hence is the 6-vertex triangulation @xmath301 of @xmath302 .
take a non - triangle @xmath303 of @xmath304 ( there are 10 of them ) .
then @xmath305 can be viewed as an 1-cycle of @xmath300 with @xmath46-coefficient as well as with @xmath306-coefficient . in both view
, it is not a boundary .
however @xmath307 is a boundary with @xmath306-coefficient .
since the map @xmath308 , induced by the inclusion map @xmath309 , is injective , it follows that @xmath310 \neq 0 $ ] as an element of @xmath311 and hence also as an element of @xmath123 .
but , @xmath312 = 0 $ ] in @xmath123 .
so , @xmath310 $ ] is an element of order 2 in @xmath123 .
this is a contradiction to the assumption on @xmath123 .
[ lemma:5.2 ] let @xmath131 be a primitive triangulation of @xmath232 .
suppose @xmath313 and @xmath131 has no induced @xmath291-cycle .
then , 1 .
all vertex links of @xmath131 are induced cycles in @xmath131 .
any two adjacent vertices of @xmath131 have exactly two common neighbours in @xmath131 .
let @xmath314 .
since @xmath313 is primitive , lemma [ lemma:4.3 ] implies that @xmath237 can not be a 3-cycle .
let the vertices @xmath238 of the cycle @xmath237 be neighbours in @xmath131 . since @xmath131 is primitive , the 3-cycle @xmath315 can not be induced in @xmath131 .
so , the triangle @xmath316 is in @xmath131 .
therefore , @xmath162 and @xmath182 are neighbours in @xmath237 .
this implies that @xmath237 is induced in @xmath131 .
this proves part ( a ) .
let @xmath159 be adjacent vertices of @xmath131 , and let @xmath182 be a common neighbour of @xmath56 and @xmath162 .
then , as the 3-cycle @xmath240 is not induced in @xmath131 , the triangle @xmath316 is in @xmath131 .
so , @xmath182 is the third vertex of one of the two triangles of @xmath131 through @xmath70 .
thus , @xmath56 and @xmath162 have exactly two common neighbours .
this proves part ( b ) .
[ lemma:5.3 ] let @xmath131 be a primitive triangulation of @xmath232 such that @xmath131 has no induced cycle of length @xmath215 @xmath216mod @xmath217 . then
, either @xmath317 or all the vertices of @xmath131 have degree @xmath42 @xmath216mod @xmath217 . by assumption and lemma [ lemma:5.2 ] ( a ) , @xmath318 , where @xmath319 consists of the vertices of degree @xmath21 ( mod 3 ) , @xmath320
. if possible , let @xmath321 and @xmath322 be such that @xmath70 is an edge of @xmath131 . by lemma [ lemma:5.2 ]
, the cycles @xmath237 and @xmath323 are induced in @xmath131 ( of length 0 ( mod 3 ) and 2 ( mod 3 ) , respectively ) with exactly two common vertices , say @xmath324 and @xmath138 . then @xmath324 and @xmath138 are the two neighbours of @xmath56 in @xmath323 and of @xmath162 in @xmath237 .
_ no vertex in @xmath325 is adjacent to any vertex in @xmath326 . indeed ,
if @xmath327 is a neighbour of @xmath328 , then @xmath131 has the 4-cycle @xmath329 .
since @xmath131 has no induced 3-cycle or 4-cycle , it follows that one of the triangles @xmath330 and @xmath331 is in @xmath131 . hence either @xmath63 or @xmath332 is in @xmath333 .
this proves the claim .
therefore , if @xmath181 is the cycle obtained from @xmath334 by deleting the two vertices @xmath56 , @xmath162 and the four edges @xmath335 , @xmath336 , @xmath337 , @xmath338 , then @xmath181 is an induced cycle in @xmath131 of length @xmath339 ( mod 3 ) .
this is a contradiction .
therefore , no vertex in @xmath340 is adjacent to any vertex in @xmath341 .
since @xmath131 is connected and @xmath342 , it follows that @xmath343 or @xmath344 .
if @xmath344 then the degree of each vertex is 0 ( mod 3 ) and hence @xmath131 has a vertex @xmath182 of degree 3 . since @xmath131 is primitive , @xmath345 bounds a triangle .
then , @xmath131 contains an @xmath294 and hence @xmath317 . otherwise , @xmath346 .
[ theorem:5.4 ] up to isomorphism , @xmath294 and @xmath347 are the only two primitive triangulations of @xmath232 with no induced cycle of length @xmath215 @xmath216mod @xmath217 .
clearly , @xmath294 has no induced cycle whatsoever .
it is easy to see that all the induced cycles of @xmath347 are 5-cycles .
( indeed , these are precisely the twelve vertex links . )
thus , these two triangulations of @xmath232 are primitive with no induced cycle of length 1 ( mod 3 ) .
conversely , let @xmath131 be a primitive triangulation of @xmath232 with no induced cycle of length 1 ( mod 3 ) .
assume @xmath313 .
by lemma [ lemma:5.3 ] , the minimum degree of the vertices of @xmath131 is five .
if all its vertices have degree 5 , then @xmath348 .
so , suppose there is a vertex @xmath324 with @xmath349 .
by lemma [ lemma:5.3 ] , @xmath350 .
so , @xmath351 is a cycle of length @xmath352 .
therefore , we may choose vertices @xmath353 in @xmath351 such that @xmath354 are at distance 2 , @xmath355 are at a distance 2 , and @xmath356 , @xmath357 , where all the distances are graphical distances measured along the cycle @xmath351 .
let @xmath358 .
figure 1 is a picture " of @xmath359 .
the boundary of @xmath359 is the union of six paths in @xmath131 ( drawn as circular arcs ; edges of @xmath131 are drawn as straight line segments ) .
since @xmath131 has no induced cycle of length @xmath360 , it is easy to see that these six paths intersect pairwise at most at common end points , and the six end points marked in fig . 1 are distinct vertices of @xmath131 .
thus , the boundary of @xmath359 is a cycle and hence @xmath359 is a disc .
figure 1 : disc @xmath359 in the proof of theorem [ theorem:5.4 ] fix an index @xmath361 .
consider the boundary @xmath362 of the disc @xmath363 .
then the vertex set of @xmath362 is the union of the vertex sets of @xmath351 , @xmath364 , @xmath365 minus the three vertices @xmath366 . by lemma [ lemma:5.2 ] ( b ) , @xmath351 and @xmath364 ( as also @xmath351 and @xmath365 )
have exactly two common vertices .
since @xmath359 is a disc , it follows that @xmath364 and @xmath365 have a unique vertex ( namely , u ) in common .
also , by lemma [ lemma:5.3 ] , each of @xmath351 , @xmath364 , @xmath365 have length 2 ( mod 3 ) .
therefore , the inclusion exclusion principle shows that the length of @xmath362 is @xmath367 ( mod 3 ) .
therefore , @xmath362 is not an induced cycle of @xmath131 .
thus , there is an edge @xmath368 in @xmath131 such that @xmath369 and @xmath370 are non - consecutive vertices in the cycle @xmath362 . by the proof of lemma [ lemma:5.3 ] , no vertex of @xmath371 is adjacent in @xmath131 with any vertex of @xmath372 ; also no vertex of @xmath373 is adjacent in @xmath131 with any vertex of @xmath374
. therefore , @xmath375 , @xmath376 ( see fig .
1 ) . that is
, @xmath369 is an interior vertex of the path @xmath377 and @xmath370 is an interior vertex of the path @xmath378 .
now , consider the disc @xmath379 in @xmath131 complementary to the disc @xmath359 ( i.e. , @xmath380 $ ] ) .
then , @xmath381 and @xmath382 are in @xmath379 .
clearly , @xmath381 separates @xmath383 from @xmath384 in @xmath379 ( i.e. , @xmath383 and @xmath384 are in different connected components of @xmath385 ) .
therefore , the geometric edges @xmath386 and @xmath387 intersect at an interior point .
this contradicts the very definition of the geometric realization of a simplicial complex .
thus , there is no vertex @xmath324 of @xmath131 with @xmath388 .
this completes the proof .
[ coro:5.5 ] for triangulations @xmath131 of @xmath232 , the following conditions are equivalent : 1 .
@xmath131 has no induced cycle of length @xmath389 @xmath216mod @xmath217 , 2 .
all the induced cycles of @xmath131 have length @xmath29 and @xmath292 , and 3 .
@xmath131 is a connected sum of @xmath294 s and @xmath347 s .
( the statement in ( c ) includes the possibility that @xmath131 is either icosian or stacked . )
let @xmath131 be as in ( c ) .
it follows that the only possible induced cycles in the summands are 5-cycles .
therefore , the only induced cycles in @xmath131 are 3-cycles and 5-cycles . thus , ( c ) @xmath390 ( b ) .
trivially , ( b ) @xmath390 ( a ) .
now suppose ( a ) holds .
write @xmath391 , where each @xmath285 is primitive
. since @xmath131 has no induced cycle of length 1 ( mod 3 ) , it follows that no @xmath285 has an induced cycle of length 1 ( mod 3 ) .
therefore , by theorem [ theorem:5.4 ] , each @xmath285 is @xmath294 or @xmath347 . hence @xmath131 is as in ( c ) .
thus , ( a ) @xmath390 ( c ) .
now , we can prove the second main result of this paper .
[ theorem:5.6 ] let @xmath26 be a @xmath46-tight triangulation of a closed @xmath29-manifold .
then each vertex link of @xmath26 is a connected sum of @xmath294 s and @xmath347 s .
this is now immediate from corollary [ coro:3.6 ] ( a ) and corollary [ coro:5.5 ] .
the following result provides a recursive procedure for the computation of the sigma - star vector .
[ theorem:5.7 ] let @xmath33 and @xmath34 be induced subcomplexes of a simplicial complex @xmath4 and @xmath0 be a field .
suppose @xmath392 and @xmath393 . if @xmath12 is @xmath18-neighbourly , @xmath17 , then @xmath394 for @xmath395 .
let us write @xmath396 , @xmath397 , @xmath398 .
thus , @xmath399 . for notational convenience ,
we write @xmath400 for @xmath401 ; { \ensuremath{\mathbb{f}}})$ ] , @xmath402 .
note that any subset @xmath8 of @xmath5 can be uniquely written as @xmath403 , where @xmath404 , @xmath405 and @xmath406 .
since @xmath12 is @xmath18-neighbourly , it follows from the exactness of the mayer - vietoris sequence that @xmath407 therefore , we can compute @xmath408\\ & = \frac{1}{m_1+m_2-m+1}\left[\sum_{k=0}^{m_1}\sum_{\stackrel{a_1 , b}{\#(a_1\sqcup b)=k } } \hspace{-6mm}\tilde{\beta}_i(a_1 \sqcup b ) \sum_{\ell = 0}^{m_2-m } \frac{\binom{m_2-m}{\ell}}{\binom{m_1+m_2-m}{\ell + k } } \right .
\\ & \hspace{45 mm } + \sum_{k=0}^{m_2}\sum_{\stackrel{a_2 , b}{\#(a_2\sqcup b)=k } } \hspace{-6mm}\tilde{\beta}_i(a_2 \sqcup b ) \sum_{\ell = 0}^{m_1-m } \frac{\binom{m_1-m}{\ell}}{\binom{m_1+m_2-m}{\ell + k } } \\ & \hspace{50 mm } \left .
- \sum_{k=0}^{m}\sum_{\stackrel{b}{\#(b)=k } } \hspace{-3mm}\tilde{\beta}_i(b ) \sum_{\ell = 0}^{m_1+m_2 - 2 m } \frac{\binom{m_1+m_2 - 2m}{\ell}}{\binom{m_1+m_2-m}{\ell + k}}\right]\\ & = \frac{1}{m_1 + 1}\sum_{k=0}^{m_1}\sum_{\stackrel{a_1 , b}{\#(a_1\sqcup b)=k } } \hspace{-5 mm } \frac{\tilde{\beta}_i(a_1\sqcup
b)}{\binom{m_1}{k } } + \frac{1}{m_2 + 1}\sum_{k=0}^{m_2}\sum_{\stackrel{a_2 , b}{\#(a_2\sqcup b)=k } } \hspace{-5 mm } \frac{\tilde{\beta}_i(a_2 \sqcup b)}{\binom{m_2}{k } } \\ & \hspace{60 mm } - \frac{1}{m+1}\sum_{k=0}^{m}\sum_{\stackrel{b}{\#(b)=k } } \hspace{-3 mm } \frac{\tilde{\beta}_i(b)}{\binom{m}{k}}\\ & = \sigma_i^{\ast}(x_1 ; { \ensuremath{\mathbb{f } } } ) + \sigma_i^{\ast}(x_2 ; { \ensuremath{\mathbb{f } } } ) - \sigma_i^{\ast}(y ; { \ensuremath{\mathbb{f } } } ) .
\end{aligned}\ ] ] in the penultimate step of the above computation , we have used the following well known identity to compute the inner sums . for any three non - negative integers @xmath409 , we have @xmath410 as a particular case of theorem [ theorem:5.7 ] , we have a formula for the sigma - star vector of a connected sum .
( since @xmath411 is independent of the field @xmath0 , we denote it by @xmath412 . ) [ coro:5.8 ] for any two triangulated @xmath10-spheres @xmath248 , @xmath249 and any field @xmath0 , we have 1 .
@xmath413 , and 2 .
@xmath414 for @xmath415 .
let @xmath416 , where @xmath35 is a @xmath10-face . then @xmath417 , @xmath418 , @xmath419 , @xmath420 satisfy the hypothesis of theorem [ theorem:5.7 ] ( with @xmath421 ) , and trivially @xmath422 , @xmath423 for @xmath415 .
[ notation:5.9 ] for @xmath424 ; @xmath425 , by @xmath426 we denote a triangulated @xmath42-sphere which can be written as a connected sum of @xmath427 triangulated 2-spheres , of which @xmath18 are copies of @xmath347 and the remaining @xmath428 are copies of @xmath294 @xmath216 in some order@xmath297 .
[ coro:5.10 ] let @xmath424 and @xmath429 .
then @xmath430 .
trivially , @xmath431 .
a computation shows that @xmath432 .
thus the result holds when @xmath433 .
the general result follows by an induction on @xmath427 , where the induction leap uses corollary [ coro:5.8 ] ( with @xmath434 ) .
our next result lists a set of necessary conditions that a @xmath46-tight triangulated 3-manifold must satisfy .
note that the statement of the main result of @xcite implies that the inequality in part ( a ) of theorem [ theorem:5.11 ] holds , more generally , for all triangulations of closed 3-manifolds .
equality holds in this more general setting if and only if the triangulation is neighbourly and locally stacked .
therefore , proposition [ prop : bdth2.24 ] implies that equality holds if and only if the triangulation is @xmath46-tight and stacked .
still we have included this inequality for the sake of completeness , and because its proof arises naturally in the course of proving the rest of the theorem .
[ theorem:5.11 ] let @xmath26 be a @xmath46-tight triangulated closed @xmath29-manifold .
then the parameters @xmath435 and @xmath436 must satisfy the following .
1 . @xmath437 @xmath216mod @xmath438 .
also , @xmath439 , with equality if and only if @xmath26 is stacked .
2 . @xmath440 .
equality holds here if and only if each vertex link of @xmath26 is a triangulated @xmath42-sphere of the form @xmath441 . by theorem [ theorem:5.6 ] , for each @xmath221 there are numbers @xmath442 and @xmath443 such that @xmath444 since @xmath26 is neighbourly by lemma [ lemma:2.2 ] , equating the number of vertices in the two sides , we get @xmath445 .
hence @xmath446 , and we have @xmath447 .
equality holds in the lower bound if and only if @xmath213 is stacked , and equality holds in the upper bound if and only if @xmath213 is the connected sum of @xmath448 copies of @xmath347 and @xmath449 copies of @xmath450 .
let us put @xmath451 . then @xmath452 .
the lower bound holds with equality if and only if @xmath26 is locally stacked ( hence , by proposition [ prop : bdth2.24 ] , if and only if @xmath26 is stacked ) and the upper bound holds with equality if and only if each vertex link of @xmath26 is the connected sum of @xmath448 copies of @xmath347 and @xmath453 copies of @xmath450 .
then , by corollary [ coro:5.10 ] , @xmath454 now , lemma [ lemma:2.6 ] implies @xmath455 .
therefore , adding over all @xmath221 , we get @xmath456 since 776 is relatively prime to 429 , the result follows from and the above discussion on the bounds on @xmath18 .
[ coro:5.12 ] let @xmath26 be a @xmath46-tight closed @xmath29-manifold with @xmath435 , @xmath457
. then @xmath458 .
equality holds here if and only if @xmath26 is locally icosian . since @xmath459 , theorem [ theorem:5.11 ] ( b ) implies that @xmath460 .
this simplifies to the given inequality .
clearly , equality holds here if and only if @xmath461 ( mod 9 ) and equality holds in theorem [ theorem:5.11 ] ( b ) .
[ coro:5.13 ] let @xmath26 be a @xmath46-tight triangulated closed @xmath29 manifold . if @xmath53 or @xmath462 , then @xmath26 is stacked .
suppose , if possible , @xmath26 is not stacked .
then , by theorem [ theorem:5.11 ] ( a ) , there is an integer @xmath463 such that @xmath464 .
hence corollary [ coro:5.12 ] implies @xmath465 this inequality simplifies to @xmath466 _ case 1 .
_ @xmath467 .
hence , by , @xmath468 .
thus , @xmath469 . therefore , @xmath470 and @xmath471 .
thus , @xmath472 .
but , @xmath473 ( mod 20 ) . hence @xmath474 or 17 ( mod 20 )
but , @xmath472 .
so , @xmath475 , a contradiction .
so , the result is true in this case .
_ @xmath462 .
then , by corollary [ coro:5.12 ] , @xmath476 . hence yields @xmath469 .
thus @xmath477 , and therefore @xmath1 is congruent to 12 or 17 ( mod 20 ) . hence @xmath478 ,
and if @xmath479 then @xmath480 .
therefore , @xmath481 .
the result now follows by case 1 .
[ coro:5.14 ] let @xmath26 be a closed topological @xmath29-manifold admitting a tight triangulation .
if @xmath482 , then @xmath26 is homeomorphic to one of the following manifolds @xmath483 where @xmath484 or @xmath485 .
let @xmath4 be an @xmath1-vertex tight triangulation of @xmath26 .
then , by corollary [ coro:5.13 ] , theorem [ theorem:5.11 ] ( a ) and proposition [ prop : walkup_class ] , @xmath26 is homeomorphic to @xmath486 , @xmath487 or @xmath488 , where @xmath489 .
the result follows from this .
the following table gives a list of small values for the parameters @xmath490 of a locally icosian @xmath46-tight triangulation of a closed 3-manifold .
indeed , the number @xmath1 of vertices in any such triangulation must be congruent modulo 15444 to one of the eight values of @xmath1 listed in this table .
the following tables list the small values for parameters @xmath490 satisfying the conclusion of theorem [ theorem:5.11 ] .
table [ tbl : z2tight.3mfd ] for @xmath490 with strict inequality in part ( b ) of the theorem .
table [ tbl : z2tight.3mfd.c= ] for @xmath490 with equality in part ( c ) of the theorem .
in this section , we present some examples which help to put the results of this paper in proper perspective .
let @xmath491 be the pure 4-dimensional simplicial complex with vertex set @xmath492 and an automorphism @xmath493 ( mod 10 ) .
modulo this automorphism , there is only one representative facet ( maximal face ) of @xmath491 , namely @xmath494 .
the face vector of @xmath491 is @xmath495 .
each vertex link of @xmath491 is an 8-vertex stacked 3-ball and hence @xmath491 is a locally stacked triangulated 4-manifold with boundary .
the boundary @xmath496 has face vector @xmath497 and triangulates @xmath498 .
since each 2-simplex of @xmath491 is in the boundary , @xmath491 is stacked and hence @xmath499 is a stacked triangulated closed 3-manifold . since @xmath499 is not neighbourly , it is not tight with respect to any field
. let @xmath12 be a triangulation of @xmath498 .
since the only closed @xmath29-manifolds with at most @xmath500 vertices are @xmath501 and @xmath502 ( see @xcite ) , it follows that @xmath503 .
then , by ( * ? ? ?
* theorem 5.2 ) , @xmath504 . since @xmath505 and @xmath506
, it follows that @xmath507 .
hence @xmath508 .
thus , @xmath509 for @xmath510 . therefore , @xmath499 is strongly minimal .
the following example shows that a neighbourly and locally stacked triangulated closed 3-manifold need not be stacked .
equivalently , it need not be tight .
thus , the hypothesis ` stacked ' in theorem [ theorem:4.8 ] can not be relaxed to ` locally stacked ' .
[ exam : lutz ] let @xmath511 be the pure 3-dimensional simplicial complex with vertex set @xmath492 and an automorphism @xmath493 ( mod 10 ) .
modulo this automorphism , a set of representative facets of @xmath511 are : @xmath512 then , each vertex link is a 9-vertex stacked 2-sphere and hence @xmath511 is a locally stacked , neighbourly , triangulated closed 3-manifold .
it triangulates @xmath513 and hence @xmath514 for any field @xmath0 .
therefore , by proposition [ prop : bdth2.24 ] , @xmath511 is not stacked and not tight w.r.t . any field . since there are non - neighbourly 10-vertex triangulations of @xmath513 ( cf .
@xmath499 in example [ exam : walkup ] ) , @xmath511 is not strongly minimal . by propositions [ prop : sminimal ] and [ prop :
bdth2.24 ] , this also shows that @xmath511 is not stacked and is not @xmath0-tight for any field @xmath0 .
the following example shows that an arbitrary @xmath0-tight simplicial complex need not be strongly minimal ( we do not know if it must be minimal in the sense of having the fewest number of vertices among all triangulations of its geometric carrier ) .
[ exam : nsmtight ] consider the neighbourly @xmath42-dimensional simplicial complex @xmath4 on the vertex set @xmath515 whose maximal faces are @xmath516 , @xmath517 and @xmath518 .
it is easy to see that @xmath4 is @xmath0-tight for any field @xmath0 .
observe that @xmath9 is also triangulated by the simplicial complex @xmath74 ( on the same vertex set ) whose maximal faces are @xmath516 , @xmath518 , @xmath519 .
thus , @xmath4 is not strongly minimal
. therefore , conjecture [ conj : kl ] is not true for arbitrary simplicial complexes .
recall that a @xmath10-dimensional simplicial complex is a _
if ( i ) each maximal face is @xmath10-dimensional , ( ii ) each @xmath520-face is in at most two @xmath10-faces , and ( iii ) for any two @xmath10-simplices @xmath35 and @xmath521 , there exists a sequence @xmath522 of @xmath10-simplices such that @xmath523 is a @xmath520-face for @xmath524 .
we now extend the definitions of stackedness and locally stackedness to pseudomanifolds as follows .
[ def : st - pmfd ] a pseudomanifold @xmath525 of dimension @xmath24 is said to be _ stacked _
if all its faces of codimension ( at least ) two are in the boundary @xmath526 . a pseudomanifold @xmath188 without boundary of dimension @xmath10
is said to be _ stacked _ if there is a stacked pseudomanifold @xmath525 of dimension @xmath24 such that @xmath527 .
a pseudomanifold is said to be _ locally stacked _ if all its vertex links are stacked pseudomanifolds ( with or without boundaries ) .
consider the @xmath29-dimensional pseudomanifold @xmath528 with vertex set @xmath529 and the facet - transitive automorphism group @xmath530 @xmath531 .
modulo this group , a facet representative is @xmath532 .
the link of each vertex is isomorphic to the 7-vertex torus @xmath533 .
its face vector is @xmath534 .
since @xmath528 is a @xmath29-neighbourly pseudomanifold with euler characteristic @xmath535 , it follows that its integral homologies are torsion free and its vector of betti numbers is @xmath536 . using its doubly transitive automorphism group ,
one may calculate easily that the sigma vector of @xmath533 ( w.r.t .
any field ) is @xmath537 .
hence the mu - vector of @xmath528 ( w.r.t .
any field ) also equals @xmath536 .
therefore , ( * ? ? ?
* theorem 1.6 ) implies that @xmath528 is tight w.r.t . all fields .
now , if @xmath525 is a pseudomanifold of dimension 4 such that @xmath538 and @xmath539 then for any vertex @xmath138 of @xmath525 , we have @xmath540 and @xmath541 .
thus , @xmath542 is a stacked 3-manifold whose boundary is the 7-vertex torus @xmath543 .
but it is easy to see that the 7-vertex torus bounds exactly three ( distinct but isomorphic ) stacked 3-manifolds , each with seven 3-faces .
this implies that @xmath544 .
then , @xmath545 , which is not possible .
therefore , @xmath528 is not stacked .
thus , theorems [ theorem:4.8 ] and [ theorem:5.1 ] are not true for @xmath29-dimensional pseudomanifolds .
[ exam : end ] consider the @xmath546-vertex triangulation @xmath301 of @xmath302 whose facets are @xmath516 , @xmath547 , @xmath548 , @xmath549 , @xmath550 , @xmath551 , @xmath552 , @xmath553 , @xmath554 , @xmath555 .
let @xmath556 .
so , @xmath4 is a 5-dimensional simplicial complex obtained from the cone @xmath557 by adding a 5-face .
the simplicial complex @xmath4 is isomorphic to an induced subcomplex of the 13-vertex triangulation @xmath26 of @xmath558 obtained by lutz in @xcite .
( indeed , for each vertex @xmath138 of @xmath26 , @xmath298 contains two induced @xmath301 s , say @xmath559 $ ] and @xmath560 $ ] , where @xmath561 .
both the induced subcomplexes @xmath562 $ ] and @xmath563 $ ] are isomorphic to @xmath4 . )
thus , @xmath26 is @xmath46-tight @xcite , @xmath559 = \mathbb{rp}^2_6 $ ] is a surface and yet the induced subcomplex @xmath564 $ ] is not a pseudomanifold .
* acknowledgements : * the second and third authors are supported by diiccsrte , australia and dst , india , under the australia - india strategic research fund ( project aisrf06660 ) .
the second author is also supported by the ugc centre for advanced studies . | it is well known that a triangulation of a closed 2-manifold is tight with respect to a field of characteristic two if and only if it is neighbourly ; and it is tight with respect to a field of odd characteristic if and only if it is neighbourly and orientable .
no such characterization of tightness was previously known for higher dimensional manifolds . in this paper , we prove that a triangulation of a closed 3-manifold is tight with respect to a field of odd characteristic if and only if it is neighbourly , orientable and stacked . in consequence ,
the khnel - lutz conjecture is valid in dimension three for fields of odd characteristic .
next let @xmath0 be a field of characteristic two .
it is known that , in this case , any neighbourly and stacked triangulation of a closed 3-manifold is @xmath0-tight . for triangulated closed 3-manifolds with at most 71 vertices or with first betti number at most 188
, we show that the converse is true .
but the possibility of an @xmath0-tight non - stacked triangulation on a larger number of vertices remains open .
we prove the following upper bound theorem on such triangulations . if an @xmath0-tight triangulation of a closed 3-manifold has @xmath1 vertices and first betti number @xmath2 , then @xmath3 .
equality holds here if and only if all the vertex links of the triangulation are connected sums of boundary complexes of icosahedra . 57q15 , 57r20 , 05c15 .
_ keywords : _ stacked spheres ; stacked manifolds ; triangulations of 3-manifolds ; tight triangulations ; icosahedron . |
the only known path to rendering qcd - like theories perturbative at all energy scales is to increase the number @xmath0 of color charges@xcite , so that @xmath5 itself becomes the small expansion parameter .
while mesons in large @xmath0 continue to exhibit the quantum numbers of a single quark - antiquark pair , the large-@xmath0 baryon requires @xmath0 valence quarks , since su(@xmath0 ) group theory requires a minimum of @xmath0 fundamental representation indices to form a color singlet .. ] however , physical baryons consist also of myriad gluons and sea quark - antiquark pairs ; does this then imply that large @xmath0 baryons have a meaning only within the context of the valence quark model ?
in this paper we claim that this is not the case , and indeed argue that it is possible to use the very existence of baryons boasting well - defined quantum numbers and large-@xmath0 arguments to derive a _ rigorous _ constituent quark picture .
these assumptions are clearly independent of the momentum transfer scale , and therefore this constituent picture holds from the low - energy to deep - inelastic scattering regimes .
, the individual coefficients of the terms in the @xmath5 expansion might grow large for high momentum transfers , spoiling the utility of the expansion .
it is not known where this transition occurs . ]
this is actually the same picture , in a somewhat different language , used to derive an effective hamiltonian @xmath5 operator expansion for baryon observables .
the operator expansion has been used to analyze phenomenologically the baryon mass spectrum of the ground - state@xcite , orbitally - excited@xcite , and heavy - quark@xcite baryons , as well as magnetic moments@xcite , axial - vector couplings@xcite , and photoproduction@xcite and pionic@xcite transitions of @xmath6s in large @xmath0 .
we then apply this knowledge to a study of the charge radii of the nonstrange baryons @xmath1 and @xmath2 .
we first present the generic expansion demanded by @xmath5 when no other physical input is included , and then specialize to include physical restrictions , such as the statement that the operators representing the charge radii must be proportional to the constituent quark charges .
we find that there are actually two independent contributions at the leading order , @xmath7 , and one at @xmath8 .
since there are six baryons in the @xmath9 multiplets , this implies a number of relations between the charge radii that are expected to be satisfied particularly well , as we explore below .
for example , we show that a relation found previously in an @xmath10 quark model with two - body currents@xcite holds for arbitrary @xmath0 with @xmath4 corrections .
the paper is organized as follows . in sec .
[ const ] , we elucidate the promised relation between constituent quarks and baryon symmetry properties . in sec .
[ oper ] we restrict to the two - flavor case and exhibit the complete @xmath5 operator expansion for scalar observables such as @xmath11 charge form factors .
we then consider this expansion in the `` general parametrization method''@xcite generalized to large @xmath0 , which places additional restrictions on the allowed operators based on the observable at hand .
we present and discuss results in sec .
[ results ] and conclude in sec .
[ concl ] .
we begin with the quantum numbers of the current quarks themselves . to obtain the electric charge and hypercharge of the quarks for arbitrary @xmath0 , we require only that @xmath12 , @xmath13 , and @xmath14 remain weak isospin doublets with @xmath15 and @xmath16 , respectively , that under strong isospin the up quark and down quark still form a doublet with @xmath15 and @xmath16 , respectively , while the strange and all other quarks are isosinglets , and that all quarks in the electroweak interaction and @xmath17 quarks in the strong interaction satisfy the gell - mann nishijima condition @xmath18 then the cancellation of the su(@xmath0)@xmath19su(2)@xmath19u(1 ) standard model chiral anomalies imposes @xmath20 while under strong hypercharge one finds @xmath21 it is interesting to note that these results maintain for arbitrary @xmath0 the usual electric charge and hypercharge assignments familiar in @xmath22 , such as the proton quantum numbers @xmath23 .
baryons in large @xmath0 have masses of @xmath24 , owing to both the intrinsic @xmath25 masses of the quarks and interaction terms which also scale as @xmath0@xcite .
the emergence of an exact spin - flavor symmetry in the large-@xmath0 limit for any number of flavors was first demonstrated in ref .
@xcite , so that it is meaningful to classify baryons into spin - flavor representations at leading order in @xmath5 .
the ground - state multiplet of baryons for arbitrary @xmath0 fills , by assumption , a spin - flavor multiplet described by a tensor completely symmetric on @xmath0 indices ( fig . 1 ) .
for three flavors ( @xmath17 ) , this is an su(6 ) multiplet that for @xmath10 reduces to the familiar positive - parity @xmath26-plet containing the spin-1/2 su(3 ) octet and spin-3/2 decuplet . when @xmath27 , these multiplets are much larger
. then each multiplet possesses , in general , a number of states whose quantum numbers reduce to those of the familiar baryons in @xmath10 . for example , the spin - flavor multiplet of fig .
1 decomposes into @xmath0 distinct flavor multiplets with spins 1/2 , 3/2 , , @xmath28 : is the @xmath2 to be identified as a spin-3/2 or spin-@xmath28 state ? in this case , one finds that@xcite @xmath29 , compared to @xmath30 .
the observed relatively small @xmath2-@xmath1 mass splitting suggests that one should take @xmath31 rather than @xmath32 .
similar considerations@xcite lead one to take the large-@xmath0 analogues of the familiar baryons to have the usual spins , isospins , and hypercharges of @xmath25 rather than @xmath24 . in particular , this identifies the proton as a state with @xmath33 , @xmath34 , and valence quark content consisting of the usual triple of @xmath35 in an @xmath36 combination , augmented by @xmath37 @xmath38 pairs , each in a spin - singlet , isosinglet combination .
then @xmath39 and @xmath40 , and one may verify the previous claim that @xmath23 .
obtaining a rigorous constituent picture for baryons requires that each baryon truly resides in a unique spin - flavor multiplet . in the case of the familiar su(3 ) octet and decuplet baryons , this is the completely symmetric * 56 * -plet of su(6 ) .
such an assumption is subject to two conditions : 1 .
the baryons are stable under strong interactions , so that they are true narrow - width eigenstates of the strong hamiltonian .
this is true in large @xmath0 , since the production of each meson costs one power of @xmath41 in the amplitude .
it is also true for physical nucleons , where only weak decays are permitted , and to a lesser extent for the other ground - state baryons , where phase space suppresses such decays .
configuration mixing between the dominant ground - state multiplet and higher multiplets is suppressed .. ] this is also true in large @xmath0 , where such mixing requires the exchange of gluons to excite the ground state into an overlap with the higher state .
these gluon couplings introduce additional @xmath5 suppressions .
for example , consider flipping the spin of one of the @xmath0 quarks in a proton to form a @xmath42 .
dynamics tells us that the spatial wavefunction of the baryon should adjust itself to the new spin configuration ; however , since only one of the quarks in @xmath0 has changed , one expects this effect to be suppressed by some power of @xmath5 .
once these conditions are satisfied , it becomes a matter of mathematics alone to identify individual `` constituent '' quarks within the baryon .
this is seen from the spin - flavor young tableau for the ground state ( fig .
1 ) ; the spin - flavor wavefunction is a completely symmetric tensor with @xmath0 indices , represented by @xmath0 boxes in the tableau .
each index corresponds to a fundamental representation of the spin - flavor group , and carries precisely the same quantum numbers as one of the current quarks within the baryon .
one may use spin - flavor projection operators to isolate these representation `` quarks '' ( which we call _ r - quarks _ ) , the collective action of which is to reproduce the _ entire _ baryon spin - flavor wavefunction . in terms of field theory ,
the r - quarks are interpolating fields carrying the spin - flavor quantum numbers of current quarks , such that an appropriately symmetrized set of @xmath0 boast complete overlap with the baryon wavefunction ( fig . [ merc ] ) .
the r - quarks are then true `` constituent '' quarks , in that the baryon is constituted entirely of them and nothing else . to reiterate , _ the rigorous constituent quark is the r - quark , which is defined as the interpolating field associated with a single box in the baryon young tableau .
_ it turns out that the `` naive quark model for an arbitrary number of colors '' presented in ref .
@xcite , based on the constituent quark model , is not so naive after all .
we hasten to add that this is not a revolutionary idea .
it was understood , at least implicitly , in a number of large-@xmath0 analyses where knowledge of the completeness of sets of spin - flavor operators acting upon particular baryon multiplets is important , such as in refs .
@xcite . indeed , the `` quark representation '' presented in ref .
@xcite is mathematically equivalent to the r - quark construction .
our purpose in introducing the r - quark is to give such analyses a firm physical interpretation as well as to probe the limitations of this picture , as detailed above .
obviously , such a manipulation can not possibly tell us everything about the baryon structure . as an explicit example , consider the strangeness content of the proton .
we have argued that the flavor structure of the proton for arbitrary @xmath0 consists of the usual valence @xmath35 triple and @xmath37 @xmath38 pairs each in a spin - singlet , isosinglet combination .
but if these are all of the r - quarks , how can the proton have strange content ?
the answer is that @xmath43 pairs are present , as are other sea quarks and gluons , but all of these have been incorporated into the r - quarks . in terms of field theory , these other components have been integrated out in favor of the r - quark fields .
pairs in the proton appear only in vacuum loops , which introduce @xmath5 suppressions compared to pure glue interactions .
the same is not necessarily true for @xmath44 or @xmath45 pairs in the proton , which can appear in @xmath46-graphs . ] thus , the proton may have strange content even if it has no strange r - quarks .
the r - quark decomposition clearly does not indicate in detail how constituent quarks are formed from the fundamental degrees of freedom in the baryon .
but it does give , by construction , values for observable matrix elements that an arbitrarily good constituent quark model , i.e. , one that gives all of the correct baryon observables , must satisfy . in this way it serves as a means to improve explicit quark model calculations . as an example given in the next section
, one can extract the r - quark masses and interaction energy terms from the @xmath11 spectrum .
finally , it should be pointed out that this decomposition has nothing to do with large @xmath0 _ per se _ , except that identifying physical baryons with distinct irreducible spin - flavor representations for larger @xmath0 is on somewhat more solid theoretical ground because of the conditions listed above .
if one declares that the proton lies entirely in an su(6 ) * 56 * -plet in the physical case of @xmath10 , there is no problem in defining the three r - quarks .
the analysis of any observable with given spin - flavor quantum numbers in the @xmath5 expansion may be carried out in essentially the same way : one simply writes down all operators with the same spin - flavor transformation properties as the observable , weighted with the appropriate suppression power of @xmath5 .
the number of such operators is finite since the number of spin - flavor structures connecting the initial- and final - state baryons is finite . as a trivial example , consider the problem of mass operators of the @xmath47 nucleon states .
the wigner - eckart theorem tells us that only operators with isospins @xmath48 or 1 can connect the states .
indeed , the most general decomposition , as was done for the ground - state baryon masses@xcite , or the orbitally - excited baryons@xcite , or here for the charge radii , may be considered the application of the wigner - eckart theorem in spin - flavor space .
we also see from this example that there are precisely as many operators ( 2 ) as independent mass observables , which permits arbitrary masses for the @xmath49 and @xmath50 states . in the given example , the @xmath51 and @xmath52 operators contribute to @xmath53 and @xmath54 , respectively . unless some of the operators in a given expansion may be eliminated or suppressed , the operators merely provide a reparametrization of the data , i.e. , a different basis for the same vector space of observables .
however , we have not yet taken into account suppressions of operators by powers of @xmath5 . in order to identify these suppressions for baryons ,
it is most convenient to work with r - quarks .
let us define an _
n - body operator _ as one that requires the participation of @xmath50 r - quarks ; that is , the feynman diagram has a piece that is @xmath50-particle irreducible .
since r - quarks each carry a fundamental color index , they exchange gluons just like current quarks and hence obey the same large-@xmath0 counting rules .
indeed , it is not difficult to see that an @xmath50-body operator requires the exchange of a minimum of @xmath55 gluons and hence a suppression of @xmath56 , since@xcite @xmath57 .
the most general possible @xmath50-body operators can be built from @xmath50th - degree polynomials in 1-body operators , whose members fill the adjoint representation of the spin - flavor group .
we denote these @xmath58 where @xmath59 are the usual pauli spin matrices , @xmath60 denote gell - mann flavor matrices , and the index @xmath61 sums over all @xmath0 quark lines in the baryons . in the two - flavor case considered here , @xmath62 is replaced with the isospin operator @xmath63 .
one then builds polynomials in @xmath64 , @xmath65 , and @xmath66 with the same spin - flavor quantum numbers as the observable in question
. however , there are still three important points to take into account before the analysis is complete . first , the operators in eq .
( [ basis ] ) sum over all the r - quarks in the baryon and may add coherently to give combinatoric powers of @xmath0 that compensate some of the @xmath5 suppressions .
generally , this occurs for @xmath66 and not @xmath65 or @xmath64 , since we have chosen baryons to have spins and isospins of @xmath25 rather than @xmath24 .
second , there exist relations , called operator reduction rules@xcite , between some combinations of operators due to the spin - flavor symmetry or the symmetry of the baryon representation .
for example , one particular combination of @xmath67 , @xmath68 , and @xmath69 is the quadratic casimir of the spin - flavor algebra , and just gives the same number when applied to all baryons in the same representation . in the two - flavor case with scalar operators , the operator reduction rules of ref .
@xcite tell us that the @xmath70 never need appear , since every possible contraction of its spin index leads to a reducible combination .
likewise , @xmath71 in the two - flavor case .
third , the most complicated operator necessary to describe a baryon with @xmath0 r - quarks is an @xmath0-body operator .
however , ultimately we are interested in the subset of these baryons that persist when @xmath10 , and by the same logic , these are completely described by expanding only out to 3-body operators .
the 4- , 5- , , @xmath0-body operators would be linearly independent when acting upon the _ full _ baryon representation , but must be linearly dependent on the 0- , 1- , 2- , and 3-body operators when acting upon the baryons that persist for @xmath10 .
since we are not taking the strict @xmath72 limit but rather @xmath0 large and finite , the question of losing information due to noncommutativity with the chiral limit@xcite does not arise . using these rules , it is straightforward to write down the expansion for an arbitrary scalar quantity with possible isospin breaking but preserving @xmath73 ( as in electromagnetic interactions or masses ) .
our example is the derivative of the baryon charge ( sachs ) form factor @xmath74 at @xmath75 , but note that the same expansion would hold for the whole @xmath76-dependent form factor , as well as masses@xcite : @xmath77 the brackets indicate that the operators are to be evaluated for a particular baryon state ; anticommutators are used to remind one that the commutator combinations are reducible , owing to the spin - flavor symmetry . here ,
each of the coefficients @xmath78 possesses a @xmath5 expansion starting at order @xmath79 ; they play the role of reduced matrix elements in the wigner - eckart theorem .
we make the naturalness assumption that any dimensionless coefficient appearing in the analysis is of order unity , unless one can think of a reason it is suppressed ( additional symmetry or chance dynamical cancellation ) or enhanced ( additional dynamics ) .
an example of the first case is the neutron - proton mass difference , where one would find an anomalously small coefficient unless the approximate symmetry of isospin is recognized .
an example of the second case is that the neutron - proton scattering lengths are much larger than `` natural size , '' pointing to shallow bound ( the deuteron ) or nearly bound ( @xmath80 ) states . to illustrate the r - quark picture for the baryons ,
consider the case of @xmath1 and @xmath2 masses using the r.h.s . of eq .
( [ matel ] ) .
the operator @xmath81 clearly gives a common mass @xmath82 to each r - quark , while the @xmath73 term differentiates @xmath83 and @xmath84 r - quarks .
the remaining operators require interactions of the r - quarks and may be considered matrix elements of the potential . using breit - wigner masses for the @xmath2 states ( note ,
however , ref .
@xcite for a treatment using pole masses ) , one finds @xmath85 and the interaction energy terms for nucleons and @xmath2 s amount to about 73 and 366 mev , respectively .
these values for the quark masses are consistent with those used in ordinary constituent quark models .
the r - quark masses thus account for the bulk of baryon masses , underscoring the economy of this picture .
it is convenient to rewrite eq .
( [ matel ] ) in terms of the global quantum numbers @xmath86 and @xmath87 , which equal @xmath88 and @xmath89 , respectively , in the two - flavor case
. then the expansion reads @xmath90 where again each @xmath91 possesses a @xmath5 expansion starting at order @xmath79 .
note in either case that there are 6 independent operators , reflecting that there are 6 observables , corresponding to the isodoublet of @xmath1 s and the isoquartet of @xmath2 s .
equation ( [ opoldn ] ) is therefore the most general expansion one can write down , modified only by the @xmath5 suppression factors .
for the particular case of the charge form factor , one can go a bit further . despite their @xmath24 masses ,
baryons in large @xmath0 nevertheless have a finite size@xcite , so @xmath92 in eq .
( [ opoldn ] ) should actually be replaced by @xmath93 .
one can see this by noting that no interaction diagram in the baryon is larger than @xmath94 , so that the interaction energy per quark is no larger than @xmath79 , and thus the wavefunction of each quark has a spatial extent of @xmath7 .
thus , the most general expansion based solely upon symmetry and the grossest features of large @xmath0 reads @xmath95 this operator method lies at one extreme end of possible analyses , in that it includes _ only _ symmetry information . at
the other end lie phenomenological models , in which not only the structure of the individual operators but also their coefficients are provided . as an intermediate choice , one may impose mild physical constraints on the allowed operators ; this is the approach of the `` general parametrization ( gp ) method''@xcite .
it was applied to the case of baryon charge radii@xcite in order to check relations appearing in a quark model calculation@xcite that includes two - body exchange currents .
here we extend the analysis to arbitrary @xmath0 .
pion - baryon couplings are studied using the gp and compared with results of a @xmath5 approach in ref .
@xcite .
it should be stressed that these `` mild physical constraints '' do indeed impose some model dependence on the gp , meaning that its predictivity follows not from qcd alone but requires additional dynamical assumptions .
however , as argued next and in the first paragraph of sec .
[ results ] , these assumptions have a firm dynamical basis and are more mild than those of an arbitrary model .
the assumptions of the gp method for charge form factors are quite minimal : all scalar operators are allowed that couple to the quarks ( r - quarks in our case ) through precisely one factor of the quark charges , which is what one expects from a single photon vertex .
then one has @xmath96 the rules for assigning @xmath5 suppressions in the coefficients are the same as above : @xmath50-body operators have a factor @xmath56 , and @xmath97 each possess @xmath5 expansions starting at order @xmath79 .
note that this expression , unlike eq .
( 5 ) in ref .
@xcite , has no strange quark term : as discussed above , the @xmath1 s and @xmath2 s have no strange r - quarks ; the @xmath43 contributions appear as @xmath8 corrections to the dynamical coefficients already presented .
it is straightforward to evaluate matrix elements of these three operators .
the sums are re - expressed in terms of the casimirs @xmath87 , @xmath98 , @xmath99 , and @xmath100 . to evaluate the final two casimirs , note that the spin - flavor wavefunction is completely symmetric .
thus , all of the @xmath83 quarks , for example , are in a symmetric state , and one then has total @xmath83-quark spin @xmath101 . after simplifying all terms , one finds @xmath102 \left [ n_c - 2j \right ] / 2n_c , \nonumber \\
\sum_{i \neq j \neq k } q_i \ , \mbox{\boldmath $ \sigma$}_j \cdot \mbox{\boldmath
$ \sigma$}_k & = & q \left [ 4j(j+1 ) + 2 - 5n_c \right ] + \left [ n_c + 2(j+1 ) \right ] \left [ n_c - 2j \right ] / n_c .\end{aligned}\ ] ] the gp expansion then reads @xmath103 \left [ n_c - 2j \right ] \right\ } \nonumber \\ & & + \frac{c}{n_c^3 } \left\ { q n_c \left [ 4j(j+1 ) + 2 - 5n_c \right ] + \left [ n_c + 2(j+1 ) \right ] \left [ n_c - 2j \right ] \right\ } .\end{aligned}\ ] ] the charge radii , defined as @xmath104 if @xmath105 , and neglecting the @xmath87 factor if @xmath106 , are presented for the @xmath1 and @xmath2 states in table i. it is interesting to compare the two expressions eqs .
( [ opold ] ) and ( [ opnew ] ) .
first , one sees that the latter is , as it must be , a special case of the most general possible expression eq .
( [ opold ] ) .
specifically , the two expressions are related by @xmath107 meaning that in gp the coefficients @xmath108 are independent and of natural [ @xmath25 ] size , @xmath110 is dependent and subleading in @xmath5 , and @xmath111 .
note also that the coefficient @xmath112 can appear at @xmath25 and @xmath113 at @xmath114 ) , a factor @xmath0 larger than naively expected from eq .
( [ opers ] ) , a result arising from the combined spin ( ) and flavor ( @xmath115 ) structure of the corresponding operators .
since the @xmath87 operator , containing a piece transforming as @xmath52 , is the sole source of isospin breaking in the gp , one expects that the @xmath116 and 3 contributions , first appearing in @xmath117 and @xmath118 terms , are absent . by the wigner - eckart theorem
, one can see that these relations involve only @xmath2 states .
we have pointed out that the gp expression eq .
( [ opnew ] ) is not the most general possible expansion for the charge radius .
the other terms in eq .
( [ opold ] ) but not ( [ opnew ] ) can appear if subleading effects are taken into account .
for example , in the gp expression , the only source of isospin quantum numbers is the quark charge operator @xmath115 .
explicit isospin breaking due to , say , the @xmath83-@xmath84 quark mass difference introduces factors of the operator @xmath119 , which do not conform to the expression ( [ opnew ] ) , but appear with an additional small ( @xmath120 ) coefficient .
similar statements are expected for loop corrections ; for example , one can see how electromagnetic loop corrections induce a @xmath118 and possibly other suppressed terms in the expansion , at the cost of an @xmath121 suppression .
inasmuch as these additional effects are dynamically suppressed , the gp expansion should give an excellent expansion for the charge form factors .
since the neglected coefficients are small , they would make little numerical difference if included in the analysis below . one interesting feature of the gp expression eq .
( [ opnew ] ) is that the terms not proportional to the total baryon charge @xmath87 are all proportional to @xmath122 , and in particular , vanish for @xmath32 .
that is , all charge radii ( and other electromagnetic matrix elements ) are proportional to @xmath87 for @xmath32 , which was pointed out by coleman@xcite for the case @xmath10 .
the symmetry reason for this feature is not hard to see : the charge operator @xmath87 transforms according to the adjoint representation of the spin - flavor group . the @xmath32 flavor representation , unlike that of @xmath34 , 3/2 ,
, @xmath123 , is completely symmetric , and has the same young tableau as the spin - flavor representation in fig . 1 .
in the product of this representation with its conjugate ( relevant to baryon bilinears ) there is only one adjoint representation , and since one already has one such operator , @xmath87 , its matrix elements must be proportional to the eigenvalue @xmath87 . for the flavor representations with @xmath124 ( such as that of spin-3/2 for @xmath125 ) ,
the corresponding product has two or more adjoints , and exact proportionality to @xmath87 no longer holds .
as discussed above , @xmath116 and 3 terms are absent in eq .
( [ opnew ] ) .
the following relations ( or any combination thereof ) hold in the gp : @xmath126 one also sees from eq .
( [ opnew ] ) and table i that both @xmath127 and @xmath112 terms are of leading order ( @xmath79 ) for generic @xmath1 s and @xmath2 s in large @xmath0 , despite the fact that the former comes from one - body and the latter from two - body operators .
this is due to the coherence effect in the two - body operator .
similarly , the three - body operator ( @xmath113 term ) is suppressed only by @xmath5 .
it is only special combinations of the charge radii in which these leading effects cancel . a particularly interesting combination of this type is @xmath128 in which the full one- and two - body terms , as well as the coherent part of the three - body term , cancel for all @xmath0 .
this cancellation also holds for the completely generic expansion ( [ opold ] ) , in which the r.h.s . of eq .
( [ newreln ] ) reads @xmath129 .
thus , if three - body operators are neglected , one has @xmath130 for all @xmath0 .
the only other such combinations are obtained by adding linear combinations of eqs .
( [ i23 ] ) .
if we still demand an @xmath4 combination but allow a two - body operator ( which would serve to distinguish large @xmath0 from the straightforward gp approach ) , one finds the separate relations @xmath131 , \nonumber \\ r_n^2 - r_{\delta^0}^2 & = & -\frac{6}{n_c^2 } \left [ b - \frac{2c}{n_c } \right ] .\end{aligned}\ ] ] one may combine these relations with eqs .
( [ i23 ] ) to predict all the @xmath2 charged radii in terms of @xmath132 good to @xmath4 . alternately , if one allows @xmath0-dependent coefficients , the only relation in addition to eqs .
( [ i23 ] ) with no corrections in the gp is @xmath133 which is trivial for @xmath10 . for completeness
, the isovector and isoscalar charge radii are given by @xmath134
the experimental values @xmath135 @xmath136@xcite and @xmath137 @xmath136@xcite , together with table i , suggest that @xmath138 if @xmath113 is neglected .
while this is somewhat larger than one would expect from a pure naturalness assumption , dynamical models for @xmath127 and @xmath112 must be studied to decide whether this ratio is unnatural . moreover , using eqs .
( [ i23 ] ) and ( [ deltas ] ) with these experimental values and estimating @xmath4 terms to be about @xmath139 @xmath136 ( which overwhelms statistical uncertainties on @xmath132 ) , one finds @xmath140 = 0.85 \pm 0.09 \ , { \rm fm}^2 , \nonumber \\ r_{\delta^+}^2 & = & r_p^2 + \frac{6}{n_c^2 } \left [ b + 2c \left ( 1 - \frac{1}{n_c } \right ) \right ] = 0.79 \pm 0.09 \ , { \rm fm}^2 , \nonumber \\ r_{\delta^0}^2 & = & r_n^2 + \frac{6}{n_c^2 } \left [ b - \frac{2c}{n_c } \right ] = -0.11 \pm 0.09 \ , { \rm fm}^2 , \nonumber \\ r_{\delta^-}^2 & = & r_p^2 - 2 r_n^2 - \frac{6}{n_c^2 } \left [ b - 2c \left ( 1 + \frac{1}{n_c } \right ) \right ] = 1.02 \pm 0.09 \ , { \rm fm}^2.\end{aligned}\ ] ]
we have seen that a rigorous constituent quark picture for baryons , in that all spin - flavor matrix elements are reproduced by construction , follows from the assumption that the physical baryons are narrow - width eigenstates of distinct spin - flavor representations .
both of these requirements hold in the large-@xmath0 limit . to improve systematically upon these assumptions ,
baryon strong decay amplitudes and configuration mixing must be accommodated , opening up new possibilities for large-@xmath0 quark models .
the analysis of observables is possible in this simplified scheme . in particular , here we have studied @xmath11 charge radii , and showed 1 ) that the one - body and part of the two - body operator are of leading order in @xmath5 , and 2 ) that a number of useful relations follow from a simple parametrization ( gp ) representing the most important physical effects
. it will be interesting to test which of these relations are supported by experiment .
ajb thanks w. broniowski for useful discussions and the deutsche forschungsgemeinschaft for support under title bu 813/2 - 1 and jefferson lab for their hospitality .
rfl thanks c.e .
carlson , c.d .
carone , and j.l .
goity for valuable comments , and the department of energy for support under contract no .
de - ac05 - 84er40150 . | we show how one may define baryon constituent quarks in a rigorous manner , given physical assumptions that hold in the large-@xmath0 limit of qcd .
this constituent picture gives rise to an operator expansion that has been used to study large-@xmath0 baryon observables ; here we apply it to the case of charge radii of the @xmath1 and @xmath2 states , using minimal dynamical assumptions .
for example , one finds the relation @xmath3 to be broken only by three - body , @xmath4 effects for any @xmath0 .
epsf 255=255 by 60 255 by-60255 by 0.1 in .1 in .1 in |
the long - standing hunt for the inhomogenous superconducting state predicted by fulde and ferrell @xcite and by larkin and ovchinnikov @xcite ( the so - called fflo state ) has recently known a revival in the study of fermionic cold atoms @xcite , quantum chromodynamics color superconductivity @xcite and heavy fermion superconductor cecoin@xmath0 @xcite .
the fflo phase consists in a condensate of finite momentum cooper pairs in contrast to the zero momenta pairs of the usual bcs state .
hence the fflo superconducting order parameter acquires a spatial variation .
such a state may be induced when a chemical potential difference is applied between two species of fermions with mutual pairing attraction . in superconductors
the different chemical potentials are obtained by a zeeman splitting @xmath4 between spin up and down states , @xmath5 and @xmath6 being respectively the bohr magneton and an external magnetic field ( in magnetic superconductors @xmath7 is the internal exchange field ) .
a characteristic feature of the field - temperature phase diagram is the existence of a tricritical point ( tcp ) which is the meeting point of three transition lines separating the normal metal , the uniform superconductor and the fflo state . for clean @xmath8-wave superconductors
the tcp is located at @xmath9 and @xmath10 , @xmath11 being the zero field critical temperature .
unfortunately , the fflo state is only energetically favorable in a small part of the phase diagram , located at low temperatures @xmath12 and high fields , see fig.[fig_phase_diagram ] . up to now
, there is no compelling experimental evidence for this mysterious fflo phase although some heavy fermion superconductors like cecoin@xmath0 @xcite , organic superconductors like @xmath1-(bets)@xmath2fecl@xmath3 @xcite , and rare - earth magnetic superconductors like errh@xmath3b@xmath3 @xcite are promising candidates . in the neighborhood of the tcp
, the fflo transition may be described within a modified ginzburg - landau approach ( mgl ) .
the mgl free energy functional , which follows from bcs theory with zeeman pair breaking , was first obtained for @xmath8-wave superconductors @xcite and then generalized to @xmath13-wave pairing with / without impurities @xcite .
the salient features of the mgl functional are the presence of higher order derivatives of the order parameter @xmath14 than the usual @xmath15 term , and the fact that the coefficient @xmath16 of the latter term changes sign at the tcp being positive for @xmath17 and negative otherwise . when the stiffness @xmath16 becomes negative , the spatially varying configurations of the order parameter are favored leading to the fflo transition . as a result of
the anisotropy contained in higher order derivatives , the degeneracy over the fflo modulation is removed even in the cubic lattice .
another source of anisotropy is related to momentum - dependent pairing . in the case of a @xmath13-wave
pauli limited superconductor both lattice and pairing anisotropies are of primary importance for the fflo @xcite . beside these mean - field studies ,
the fluctuations have rarely been considered in the context of the fflo transition , although they may be accounted for in a quite simple and accurate manner within the framework of the mgl . using the standard ginzburg - landau functional , the effect of fluctuations on the properties of the normal state were extensively studied for the homogeneous bcs superconductors @xcite and high-@xmath18 superconductors @xcite . though relatively small , thermodynamical and transport properties were actually measured in the gaussian regime . up to now , a detailed investigation of the fluctuations near the fflo transition was missing . a first attempt to fill this flaw
was realized recently by marenko and samokhin who addressed the issue of the quantum critical fluctuations at the zero temperature fflo transition @xcite .
the aim of this paper is to perform the complementary investigation of the fluctuations in the neighborhood of the tricritical point . in this letter , we study the effect of the fluctuations close to the fflo transition .
we focus on the specific heat @xmath19 and on the excess conductivity @xmath20 , the so - called paraconductivity , induced by the superconducting fluctuations within the normal state .
we have identified three regimes in the vicinity of the tricritical point @xmath21 which differ by the nature of the soft fluctuation modes . within the first regime , depicted as region i in fig.[fig_phase_diagram ] ,
the soft modes have the usual quadratic dispersion and are located around the origin of momentum space .
this familiar situation yields the well - established power law behaviors @xmath22 reviewed in @xcite . phase diagram around the tricritical point @xmath23 .
the solid lines represent the transitions between the normal metal , the spatially uniform bcs superconductor , and the non uniform fflo state . in region
i where @xmath24 ( resp .
region ii where @xmath25 ) , the normal state behavior is dominated by uniform ( resp .
non uniform ) superconducting modes which propagate quadratically . in region
iii where @xmath26 , the preponderant fluctuations propagate as @xmath27 .
region iii is defined by @xmath28 there are also crossover regimes ( cr ) separating the region iii from i and ii .
, width=288 ] in contrast , the second regime represented as region ii in fig.[fig_phase_diagram ] , is characterized by anomalous power laws in isotropic or weakly anisotropic materials .
here anomalous simply means distinct from @xmath22 .
this discrepancy is related to the fact that the minima of the dispersion are attained on a finite dimensional surface , a sphere or a ring , see fig.[fig_crossovers].@xmath29 .
hence the available phase space volume of the low energy fluctuations is larger than in the case of the standard bcs transition towards a uniform ground state , as it was first pointed out by brazovski @xcite .
on the other hand , among the host of promising candidates for the fflo state , crystal or gap symmetry effects are important .
in such case the sphere or ring containing the low energy modes is partially removed and the low energy fluctuations are located within small islands centered at a few number of isolated points in momentum space , as it is shown in fig.[fig_crossovers].@xmath30 . thus for large anisotropy , the anomaly disappears and one recovers the usual power laws @xmath31 . in summary the exponents in regime
ii depend drastically on the anisotropy of the system .
the most interesting regime , region iii of fig.[fig_phase_diagram ] is at the immediate vicinity of the tcp where the fflo and the uniform superconducting phases compete .
the power laws of the specific heat @xmath32 and paraconductivity @xmath33 are anomalous and quite universal since they pertain in both isotropic and anisotropic models .
these predictions rely on the fact that the dispersion of the soft fluctuation modes becomes quartic within this regime .
indeed , these three regimes and the related crossovers between them may serve as a powerfull tool to identify fflo state .
we investigate the fluctuation properties at the transition between normal and fflo states using a quite general approach based on the mgl functional @xcite .
note that we assume a second order phase transition to the fflo state . in the framework of isotropic model
this transition is of second order for one and two dimensional systems while it is of first order in three dimensional case @xcite .
however in the real superconductors the crystalline anisotropy ( even a cubic one ) is of primary importance and can modify this conclusion .
nevertheless , we may expect that in the case of weakly first order fflo transition , the fluctuation regimes studied here would be still observable .
such description is adequate near the tricritical point where the most striking behavior is expected and where the wave - vector of the fflo modulation is small .
looking for gaussian fluctuations in the normal state it is sufficient to consider mgl functional only in the quadratic @xmath34 approximation @xmath35 = \sum_{\mathbf{k}}\text { } \underset{i , j=1}{% % tcimacro{\dsum ^{d}}% % beginexpansion { \displaystyle\sum^{d } } % endexpansion } \left ( \alpha+gk_{i}^{2}+\gamma_{ij}k_{i}^{2}k_{j}^{2}\right ) \left\vert \psi_{\mathbf{k}}\right\vert ^{2 } \label{eq_our_model}%\ ] ] where @xmath36 are the fourier components of the complex order parameter and @xmath13 the dimensionality of the sample .
the coefficient @xmath37 , the coefficients @xmath16 and @xmath38 depend on the magnetic field and temperature , @xmath39 being the critical temperature for the second order transition between normal and uniform superconducting states .
although qualitative conclusions are quite general , explicit calculations are performed within the cubic anisotropy model defined by @xmath40 , where @xmath41 .
the case @xmath42 corresponds to the isotropic model since then the general fluctuation spectrum @xmath43 @xmath44 depends only on @xmath45 and on @xmath46 .
hence @xmath47 is a measure of the deviation from the ideally isotropic material with a spherical fermi surface . integrating out the gaussian modes @xmath36
, one obtains the fluctuation free energy density @xmath48 where @xmath49 is the size of the system .
the specific heat may be readily deduced from the free energy ( [ free ] ) via @xmath50 in evaluating the paraconductivity within the kubo formalism , one should use the time - dependent ginzburg - landau equation to obtain the current - current correlator at different times@xcite . moreover attention
should be paid to the fact that , owing to the presence of high order derivatives @xmath51 in eq.([free ] ) , the expression for the current operator differs from the usual one .
however the general expression@xcite relating the paraconductivity to the fluctuation spectrum,@xmath52 still holds for the fflo case although expressions for the velocities @xmath53 and for the current are very different from the usual quantum mechanical formula . owing to the denominators in eqs.([eq_heat_capa],[eq_sigmagene ] ) the main contributions to the specific heat and conductivity originate from the modes with the lowest energies @xmath54 .
the above expressions lead to singular power laws which diverge at the transition with characteristic exponents .
the exponents depend on the nature of the low energy modes : their location in momentum space and their dispersion . in the usual case of a uniform superconducting state , namely for @xmath24 and
@xmath55 the low energy modes are concentrated around the center of the brillouin zone and propagate with a quadratic dispersion .
the resulting power - law divergencies of the thermodynamical and transport properties are well - known @xcite . in the vicinity of the tricritical point ,
the situation is much more interesting since three different kinds of spectra @xmath54 , namely ( [ eq_spectreiso ] ) , ( [ eq_spectreaniso ] ) or ( [ eq_spectrum_g_0 ] ) , lead to different power laws . and anisotropic @xmath30 and @xmath56 models in the regime @xmath57 .
shaded areas represent the low energy modes leading to the various power laws.,width=288 ]
we first consider the isotropic case @xmath42 . at low fields and high temperatures @xmath58
, the positive stiffness @xmath16 leads to the standard spatially uniform superconducting state ( referred as bcs state on fig.[fig_phase_diagram ] ) .
conversely when @xmath59 , the stiffness becomes negative and the fluctuation spectrum @xmath60 may be rewritten as @xmath61 where @xmath62 and @xmath63 .
below @xmath64 the normal state is instable towards the formation of the fflo state with modulation wave vector @xmath65 .
owing to the isotropy of the model , all directions of @xmath65 are degenerate . above @xmath18 the minima of @xmath66
are attained on a finite dimensional surface : a sphere for @xmath67 or a ring for @xmath68 whose radius is exactly @xmath69 the one dimensional case is special in the sense that the minima consist in two points with coordinates @xmath70 . at the vicinity of the sphere / ring defined by @xmath71 the approximation@xmath72 may be used to evaluate eqs.([eq_heat_capa],[eq_sigmagene ] ) leading to the following singular behaviors for the specific heat @xmath73 and for the paraconductivity @xmath74 with @xmath75 for @xmath76 respectively .
both @xmath77 and @xmath78 diverge at the fflo critical temperature @xmath18 with the same @xmath13-independent exponent in contrast to the usual uniform superconductivity case .
in particular for the quasi - two and three dimensional cases , the divergencies of the fluctuation contributions near the fflo transition are much stronger than the @xmath79 power law characterizing the usual superconducting transition . in this sense ,
the anomalous behavior of the gaussian fluctuations is a signature of the fflo state in such ideal isotropic systems .
the previous results @xmath80 rely deeply on the perfect isotropy of the model .
nevertheless real superconductors are anisotropic owing to either crystalline effect or @xmath13-wave pairing .
hence the minima of the dispersion occur at a few number @xmath81of isolated points @xmath82 which are located at a finite distance from the origin .
these points belong to high symmetry lines of the crystal .
more specifically in the model @xmath83 the wave vectors @xmath84 are along the diagonals when @xmath85 and along the crystalline axis for @xmath86 , whereas @xmath87 corresponds to the isotropic case .
the relevant low energy fluctuations form small islands around the points @xmath82 , as shown in fig.[fig_crossovers].@xmath30 . in the case of a very weak anisotropy @xmath88
, these pockets may collapse within a finite dimensional surface and one recovers the isotropic model discussed above , see fig.[fig_crossovers].@xmath89 .
here we consider the opposite case of significant anisotropies @xmath90 which is expected in real compounds .
then the favored modes belong to @xmath81disconnected islands .
hence one may focus on each of the @xmath82 obtained by solving @xmath91 and expand the fluctuation spectrum of @xmath92 around @xmath82 as@xmath93 where @xmath94 is the effective mass tensor for the fluctuating modes located around @xmath82 .
owing to the symmetry of the crystal , the eigenvalues @xmath95 of the tensors @xmath96 are the same for the @xmath97 islands .
evaluating @xmath98 and @xmath99 with the spectrum @xmath100 and collecting the contributions from the @xmath101minima yield the specific heat @xmath102 and the diagonal paraconductivity @xmath103 in the plane @xmath104 @xmath105 where @xmath106 and @xmath107 are the eigenvalues of @xmath108 whose eigenvectors span the @xmath104 plane . for the conductivity @xmath109 one should replace @xmath110 by @xmath111 .
in sharp contrast to the isotropic model results ( [ heatiso],[condiso ] ) , the temperature dependence of both @xmath112 and @xmath113 becomes sensitive to the dimension with the same exponents as in the homogeneous case @xcite .
thus in the anisotropic model , the exponents themselves no longer provide a valuable hallmark for the fflo state .
naturally when @xmath114 one should recover the isotropic results ( [ heatiso],[condiso ] ) .
the crossover from anisotropic to isotropic models is provided by the @xmath115 dependence of @xmath116 the latter determinant vanishes in the limit @xmath114 , as an example like @xmath117 in the quasi - two dimensional case .
let us check now the validity of the approximations ( [ eq_spectreiso ] ) and ( [ eq_spectreaniso ] ) used so far in deriving the specific heat and paraconductivity .
we aim to identify the region of the phase diagram where the previous results ( [ heatiso],[condiso ] ) and ( [ heataniso],[condaniso ] ) apply .
the modes which participate to the singular parts of the specific heat and conductivity are spread within a shell of width @xmath118 around the circle @xmath71 ( isotropic case ) or in islands of size @xmath118 centered at isolated points ( anisotropic case ) . the width @xmath119 , which is defined by the condition @xmath120 , must be smaller than @xmath121 in order to have well defined shell or islands ( like in fig.[fig_crossovers].@xmath30 ) .
hence the common range of validity of eqs.([heatiso],[condiso ] ) and ( [ heataniso],[condaniso ] ) is given by @xmath122 or equivalently @xmath123 this corresponds to the area of the phase diagram which is sufficiently apart from the tricritical point ( large enough @xmath16 ) and close to critical line ( small enough @xmath124 ) , schematically the region ii on fig.[fig_phase_diagram ] .
conversely when @xmath125 the relevant low energy modes merge towards the center of the brillouin zone and the approximations ( [ eq_spectreiso ] ) and ( [ eq_spectreaniso ] ) fail . hence the specific heat and paraconductivity
must be reevaluated in the regime @xmath126
. then the physics is dominated by extremely soft fluctuation modes having a quartic dispersion instead of the more usual quadratic one , yielding very special behaviors in the region iii of fig.[fig_phase_diagram ] .
finally we investigate the crossover regime between region i and iii , see fig.[fig_phase_diagram ] .
when @xmath24 the specific heat is given by @xmath127 in the isotropic case , @xmath128 being the surface of the @xmath13-dimensional unit sphere .
the parameter @xmath16 controls the relative importance of quartic and quadratic terms . for large @xmath16
the usual bcs results are recovered while the result eq.([heattri ] ) is obtained in the opposite limit @xmath129 both quartic and quadratic terms are equally important when @xmath130 yielding to the same criterion @xmath131 for the borderline between regions i and iii than between regions ii and iii obtained above .
we now consider the model ( [ eq_our_model ] ) in the regime @xmath126 where the most salient features are expected .
_ approaching the tricritical point from the line defined by
_ @xmath132 _ _ , see fig.[fig_phase_diagram ] _ _ , one has @xmath133 and the spectrum of the fluctuations @xmath134 becomes purely quartic inducing the following anomalous temperature dependences of specific heat and paraconductivity : @xmath135 with @xmath136 for @xmath76 respectively , and @xmath137 with @xmath138 for @xmath76 respectively .
the functions @xmath139 and @xmath140 are weakly dependent on the anisotropy factor and are equal to unity in the isotropic case @xmath141 interestingly , the @xmath13-dependent exponents differ from those which are typical of the uniform superconductivity ( region i on fig.[fig_phase_diagram ] ) and of the fflo transition at larger @xmath16 ( region ii ) . consequently , the regime @xmath142 ( region iii ) and the related crossovers with regions i and ii should reveal the existence of the tricritical point and fflo phase . while they were derived along
the line @xmath143 the very unusual power laws @xmath144 and @xmath145 must emerge independently on the way the tcp is approached . _ following the normal / fflo transition line _
@xmath146 _ to reach the tcp _
, one has @xmath147where @xmath148 .
we shall demonstrate that the expressions ( [ heatiso],[condiso],[heataniso],[condaniso ] ) applying in region ii yield naturally the power laws ( [ heattri],[condtri ] ) accounting for the region iii when @xmath149 .
first from eqs.([heatiso],[heataniso ] ) , the specific heat is proportional to the product @xmath150 in the isotropic model and to @xmath151 @xmath152 in presence of sensible anisotropy , since the eigenvalues @xmath95 of the effective mass tensor scale as @xmath16 and @xmath153 taking in mind that @xmath154 along the transition line , one obtains the more divergent @xmath155 behavior of the specific heat already predicted as eq.([heattri ] ) in region iii for both isotropic and anisotropic models .
we have used the fact that the fflo and the second order uniform transition lines intersect themselves at the tricritical point . furthermore similar scaling on eqs.([condiso],[condaniso ] ) reveals that the conductivity in region ii is proportional to @xmath156 for the isotropic model and to @xmath157 for anisotropic models . in both situations
, the substitution @xmath154 leads to the same power ( @xmath158 , predicted above as eq.([condtri ] ) , for the paraconductivity . in summary and more generally ,
quartic terms in @xmath54 dominate the quadratic ones when @xmath159 and @xmath160 , see region iii in fig.1 .
up to now the orbital effect was neglected to emphasize the influence of the zeeman pair - breaking effect on the superconducting fluctuations .
this is exact in ultracold fermionic atoms since there is no orbital effect at all for neutral objects .
this is also a good approximation in superconductors with large maki parameter which are the ones where the fflo state is expected .
such a situation is generally encountered for i ) layered superconductors under in plane magnetic field when the weakness of the interplane coupling quenches the orbital motion of the electrons ( or for thin films of 3d superconductors ) , and ii ) magnetic superconductors where the field originates from internal ordered magnetic moments . beside these favorable situations
, one should evaluate the effect of orbital pair - breaking on the results obtained previously , especially for three dimensional superconductors under external magnetic field . under strong magnetic field ,
the electronic motion is described by the lowest landau level wherein the kinetic energy is totally quenched except along the direction of the field .
this provides a realization of an effective one - dimensional system . coming from the bcs side with @xmath24 we evaluate how the fluctuation specific heat and paraconductivity scale as one approaches the tricritical point along the critical line @xmath146 , namely in the limit @xmath161 for strong enough fields or sufficiently close to the transition line , the main contribution to the specific heat
is given by the lowest landau level as @xmath162 transforming the three - dimensional fluctuations into one dimensional fluctuations characterized by @xmath163 , @xmath164 being the landau level spacing and @xmath165 yielding the transition line under magnetic field . in a similar manner one may demonstrate that the magnetoconductivity along @xmath166-axis is given by @xmath167
. in sum one encounters a divergency of the specific heat while the magnetoconductivity @xmath109 is not singular in the limit @xmath168 performed along the bcs transition line . approaching the critical point along the line
@xmath133 provides an interesting field - induced crossover . in the isotropic model the specific heat
is then given by @xmath169 ^{2}}\text{\ , } \\
& = \dfrac{2eb}{\hslash}\frac{a^{2}k_{b}t_{c}^{2}}{\gamma^{1/4}\tau^{\ast7/4}% } \int_{-\infty}^{\infty}\frac{dx_{{}}\text { } } { \left ( 1+x^{4}+\zeta x^{2}\right ) ^{2}}%\end{aligned}\ ] ] where @xmath170 the dimensionless parameter @xmath171 controls the relative importance of quartic and quadratic terms yielding a crossover field @xmath172 @xmath173 .
a similar crossover between quartic and quadratic momentum dependence was already encountered at the borderlines of the region iii , see eq.([eq_crossgposi ] ) . here
the crossover is tuned by the magnetic field along the line @xmath133 .
similarly the conductivity @xmath174 behaves as @xmath175 for fields below @xmath176 @xmath177 and @xmath178 for larger fields .
now let us address the issue of the validity of the gaussian approximation .
it is well known that gaussian approximation breaks down in the critical regime wherein the interactions between different fluctuation modes of the order parameter become sizeable . in the isotropic model , brazovski demonstrated that due to the critical fluctuations the transition to the non - uniform state becomes of the first order @xcite .
recently the same conclusion has been obtained in the anisotropic model with isolated minima of @xmath179 @xcite . however performing a simple evaluation of the @xmath180 and @xmath181 interaction terms contained in the nonlinear mgl functional @xcite
, it can be proved that the relative width of the critical region ( in temperature or magnetic field ) is given by the usual ginzburg - levanyuk parameter @xmath182 , @xmath183 being the fermi energy @xcite .
more precisely and using our notations , the width @xmath184 of the fflo critical regime is given by @xmath185 when the fluctuations propagate quadratically , namely when @xmath186 ( regions i and ii of fig.[fig_phase_diagram ] ) . here
@xmath187 is the value of the coefficient @xmath16 far from the tricritical point .
conversely when @xmath188 ( region iii of fig.[fig_phase_diagram ] ) the critical region is somewhat broadened as @xmath189 owing to the presence of quartic soft modes ( region iii of fig.[fig_phase_diagram ] ) .
however even the largest value of the critical region at @xmath133 is still very small , @xmath190 .
therefore the gaussian approximation provides an excellent description of the fluctuation phenomena which may be observed at the superconducting fflo transition . on the other hand
the analysis @xcite may be relevant to the case of fermionic cold atoms where the ratio @xmath191 is not so small .
we have demonstrated that the transition towards the fflo state reveals very unusual and rich fluctuation regimes which may serve as a smoking gun of the fflo state .
atomic fermi gases provide a unique realization of an isotropic system while any real superconductor always presents some degree of anisotropy owing either to crystalline effects and/or @xmath13-wave pairing .
the fluctuation contribution vary non - monotonously near the tricritical point when we go from the uniform to fflo state . in the vicinity of the tricritical point ( region iii )
the very special exponents are expected for specific heat and paraconductivity divergencies owing to the presence of soft modes with unusual quartic dispersion .
the presence of the tricritical point is signaled by very pronounced crossovers regimes which might be observed in experiments , by varying temperature or magnetic field . finally , the fluctuation behavior of the magnetic susceptibility is also very peculiar at the fflo transition and warrants detailed analysis @xcite .
larkin a. and varlamov a. _ the physics of superconductors , vol .
i : conventional and high-@xmath18 superconductors _ ( k.h . bennemann and j.b .
ketterson , springer - verlag , 2003 ) , chapter 3 , fluctuation phenomena in superconductors . | recently some experimental evidences have been obtained in favour of the existence of the inhomogeneous fulde - ferrell - larkin - ovchinnikov ( fflo ) superconducting state in heavy - fermion superconductor cecoin@xmath0 and organic superconductor @xmath1-(bets)@xmath2fecl@xmath3 .
however the unambiguous identification of fflo state remains very difficult .
we present the theoretical studies of the gaussian fluctuations near the tricritical point ( where the fflo modulation appears ) and demonstrate that the behavior of the fluctuational specific heat , paraconductivity and diamagnetism is qualitatively different from the usual superconducting transition .
special values of the critical exponent and the crossovers between different fluctuational regimes may provide a unique test for the fflo state appearance . |
the hermitian quantum mechanics is a well - developed framework because a hermitian hamiltonian leads to a real spectrum and unitary time evolution , which preserves the probability normalization .
however , a decade ago it was observed that a large class of non - hermitian hamiltonians possess real spectra @xcite and a pseudo - hermitian hamiltonian connects with its equivalent hermitian hamiltonian via a similarity transformation am43,am37 , quantum theory based on non - hermitian hamiltonian was established @xcite . in additional , such
a framework also indicates the preservation of probability normalization if a positive - definite inner product is employed to replace the dirac inner product .
nevertheless , to date the interpretation and the measurement in experiment of such an inner product are not clear . while the dirac probability ( dirac inner product ) can be measured in a universal manner . therefore being an acceptable theory of quantum mechanics , the dirac probability is of central importance to most practical physical problems .
parity and time - reversal symmetric ( @xmath0-symmetric ) system has attracted much attention due to recent progresses on experimental investigations in @xmath0-symmetric optical systems , observation of passive @xmath0-symmetric breaking in passive optical double - well structure @xcite and observation of spontaneous symmetry breaking together with power oscillations in optical coupled system @xcite were carried out .
optics offers the rather unique advantage in detection the wave function evolution and seem to be the most readily applicable @xcite . in the past two decades , general issues of quantum effects in quantum systems have proven to be successfully investigated in the framework of quantum optical analogy based on the fact that paraxial propagation of light in optical guided structures is governed by a schrdinger - like equation @xcite . actually , the intensity observed in optical experiment corresponds to the dirac probability of electric field envelope .
it is not guaranteed for generic systems that the dirac probability is preserved even when the non - hermitian hamiltonian is time independent .
nevertheless , the violation of the conservation of dirac probability in non - hermitian system does not contradict the copenhagen interpretation .
the implications of pseudo - hermitian system are still under consideration , peculiar featureswere exhibited such as double refraction , power oscillations , _ etc .
_ @xcite following by the suggestion of realization of @xmath1-symmetric structure in the realm of optics @xcite , while nonreciprocal bloch oscillation with no classical correspondence was also shown in @xmath0-symmetric complex crystal @xcite .
we propose a class of non - hermitian lattice systems in this work , the system is composed by a set of isomorphic pseudo - hermitian clusters , which connected with each other in a hermitian way .
we show that in such non - hermitian systems , hermitian like dynamics can be observed , including the property that the time evolution is dirac probability preserving .
this is made possible by considering the dynamics of a state involving the superposition of an identical eigenmode of each isomorphic cluster in general case . in the case of having additional orthogonal modes
, it still holds when multiple eigenmodes are involved .
this hermitian dynamics , as well as the quasi - hermitian behavior , are specifically demonstrated for the case of an exactly solvable pseudo - hermitian system .
this paper is organized as follows . in section [ sec_h_basic ] ,
we present the model we focus on and its basic properties .
section sec_pseudo_h_ladder consists of an exactly solvable example to illustrate our main idea .
section [ sec_conclusion ] is the summary and discussion .
a general tight - binding network is constructed topologically by the sites and the various connections between them . as a simplified model
, it captures the essential features of many discrete systems .
also it is a nice testing ground for the study of the non - hermitian quantum mechanics due to its analytical and numerical tractability .
much effort has been devoted in recent years to the pseudo - hermitian lattice system mznojil , bendix , longhi , joglekar , joglekar83,weston , fring , ghosh , giorgi , zli , bousmina , ozlem , fabio .
the hamiltonian of the concerned tight - binding network reads as follows @xmath2 which consists of @xmath3 isomorphic clusters @xmath4 , with each cluster has a dimension @xmath5 .
the label @xmath6 denotes the @xmath6th subgraph of @xmath3 clusters , and @xmath7 ( @xmath8 ) is the boson or fermion creation ( annihilation ) operator at the @xmath9th site in the @xmath6th cluster .
the cluster @xmath4 is defined by the distribution of the hopping integrals @xmath10 where @xmath11 is real .
the set of clusters are isomorphic due the fact that they have the same eigenfunctions and spectral structures .
note that terms @xmath12 is self - adjoint since @xmath13 , which describes the hermitian connection between clusters . and
such kind of couplings are the type of _ similarity mapping _ , which is crucial for the conclusion of this paper .
the total hamiltonian @xmath14 is not hermitian when the matrix @xmath15 is not hermitian .
figure fig_illus(a ) shows a schematic example . fig_illus in this paper , we consider the case of @xmath4 being pseudo - hermitian , i.e. , @xmath4 is non - hermitian but has entirely real spectrum .
then @xmath14 is also pseudo - hermitian in the case of real @xmath11 , hence possessing the common exceptional point as @xmath4 . in general
, a pseudo - hermitian hamiltonian does not guarantee the dirac probability preserving .
it has been shown that the dirac norm of an evolved wavepacket ceases preserving as long as it touches the region of on - site imaginary potentials @xcite . in the following
we will show due to the pseudo - hermitian clusters combined together in a hermitian way , that there exist quantum states obeying dirac probability preserving , even if their profiles cover the imaginary potentials .
we start with the eigen problem of the hamiltonian @xmath4 . in single - particle invariant subspace , following the well established pseudo - hermitian quantum mechanics @xcite , we always have @xmath16 and @xmath17 where @xmath18 $ ] and @xmath19 $ ] , the operators @xmath20 and @xmath21 have the form@xmath22where @xmath23note that @xmath24 , @xmath25and @xmath26 are independent of @xmath6
. then the operators @xmath20 and @xmath21 are canonical conjugate pairs , satisfying@xmath27_{\pm } & = & \delta _ { \alpha \alpha ^{\prime } } \delta _ { \sigma \sigma ^{\prime } } , \label{can cr } \\
\lbrack a_{\alpha , \sigma } , a_{\alpha ^{\prime } , \sigma ^{\prime } } ] _ { \pm } & = & [ \bar{a}_{\alpha , \sigma } , \bar{a}_{\alpha ^{\prime } , \sigma ^{\prime } } ] _ { \pm } = 0,\end{aligned}\]]where @xmath28_{\pm } $ ] denotes the the commutator and anti - commutator . and
accordingly , the original hamiltonian can be rewritten as the form @xmath29 which has the following subtle features : ( i ) the matrix representation of @xmath14 with respect to the biorthogonal basis \{@xmath30 } is hermitian , i.e. , @xmath31 @xmath32 ; ( ii ) although it is a non - hermitian operator , i.e. , @xmath33 , straightforward algebra shows that @xmath34_{\pm } \propto \delta _ { \alpha \alpha ^{\prime } } , \text { } [ a_{\alpha , \sigma } , a_{\alpha ^{\prime } , \sigma ^{\prime } } ] _ { \pm } = 0 , \label{quasi cr}\ ] ] which indicates that although with nonorthogonality of the eigenstates as inherent feature of non - hermitian system , @xmath21 and @xmath35 obey quasi - canonical commutation relations due to the hermitian connection structure between clusters .
this results in a new type of particle statistics , that is rarely observed in hermitian systems , thus becomes highly relevant in the presence of non - hermitian terms .
considering an arbitrary state in the form@xmath36as the initial state , where @xmath37 and in which only the eigenmode @xmath38 of each cluster is involved . at instant @xmath39 , we have @xmath40 in the framework of metric operator theory , @xmath14 acts as a hermitian system , obeying unitary time evolution in the positive - definite inner product am37 .
however to date the physical meaning of the positive - definite inner product is unclear , while the dirac probability can be measured in a universal manner , e.g. dirac probability of wave electric field corresponds to the light intensity in optics and is simple to detect in experiment longhirev , therefore dirac norm is of central importance .
the aim of this paper is to show that contrary to the nonclassical dynamical behavior prl103904,longhiprl , the unitary dirac probability dynamics can also be observed in the pseudo - hermitian system . actually inserting @xmath41 into eq .
( [ phi_t ] ) , we have @xmath42 where@xmath43is the propagator in the framework of biorthogonal basis and satisfies@xmath44due to the above mentioned feature ( i ) of @xmath14 . accordingly , the dirac norm has the form@xmath45where the relation eq .
( [ unitary ] ) is applied and the @xmath6-independent factor @xmath46 can be obtained from@xmath47it follows that although @xmath48 is not the eigenstate of the entire network system , the time evolution is dirac norm - conserving , this is a direct consequence of the quasi - canonical commutation relations .
the result presented here for the evolution of an arbitrary state involving an identical isomorphic - cluster - eigenmode provides a new way for connecting the pseudo - hermitian and hermitian systems .
it is worth to mention that this probability preserving evolution can also occur for a state involving multiple eigenmodes .
this due to the fact that there always exist states , which parts belong to different eigenmodes are orthogonal in terms of dirac inner product , hence preserve the dirac probability .
for instance , a state involves two eigenmodes @xmath49 and @xmath50 , its parts on @xmath49 and @xmath50 are spatially separated local states with respect to the coordinate space @xmath6 , then the two parts of the state are orthogonal in terms of dirac inner product and the evolution of such a state is probability preserving since the quasi - canonical commutation relations .
we will demonstrate this point explicitly via the following illustrative example .
now we investigate a concrete example to demonstrate the application of the previous result .
we consider a system of a two - leg ladder [ fig . [ fig_illus](b ) ] , consisting of @xmath3 dimers as pseudo - hermitian clusters .
the hamiltonian reads@xmath51where @xmath52 is the particle number operator and the operators obey the periodic boundary condition @xmath53 with @xmath54 , @xmath55 .
@xmath56 ( @xmath57 ) is the hopping integral along legs ( rungs ) and @xmath58 denotes the norm of the imaginary on - site potential .
note that the ladder is a @xmath0-symmetric hamiltonian , where @xmath59 is the parity and @xmath60 denotes time - reversal .
the simple structure of this model makes it an ideal testing ground for a more profound understanding of the hermitian dynamics in a pseudo - hermitian system .
taking the transformations@xmath61 where @xmath62 \text { , } \sigma = \pm \right ) $ ] , which are obtained from the solution of the dimer ( a general solution of @xmath63-dimension cluster is shown in ref .
@xcite . ) , the ladder hamiltonian can be written as @xmath64which is illustrated in fig .
[ fig_illus](c ) , here @xmath65 and @xmath66 , @xmath67 $ ] .
the biorthogonal structure of the solution for a dimer admits the following canonical commutation relations eq .
( [ can cr ] ) and @xmath68 _
{ \pm } & = & \left [ a_{\alpha , \sigma } , a_{\alpha , \sigma } ^{\dagger } \right ] _ { \pm } = \sec \theta \delta _ { \alpha \alpha ^{\prime } } , \\ \lbrack \bar{a}_{\alpha , -\sigma } ^{\dagger } , \bar{a}_{\alpha ^{\prime } , \sigma } ] _ { \pm } & = & \lbrack a_{\alpha , \sigma } , a_{\alpha , -\sigma } ^{\dagger } ] _ { \pm } = i\sigma \tan \theta \delta _ { \alpha \alpha ^{\prime } } .\end{aligned}\]]obviously , hamiltonian eq .
( [ h_ladd_2band ] ) represents a two - band model , which has an interesting feature comparing to a hermitian two - band model : although there are no interband transitions , the two bands are not independent .
it is due to the pseudo - hermiticity of the clusters , which allows @xmath69 _ { \pm } = 0 $ ] but @xmath70_{\pm } \neq 0 $ ] .
this characteristic will be further demonstrated through the following quasi - canonical commutation relations eq .
( [ commu_rela ] ) and the time evolution for various gaussian wavepackets . figure [ fig_illus](c ) schematically illustrates such an equivalent two - band structure
. nevertheless , hamiltonian eq .
( [ h_ladd_2band ] ) can be diagonalized as a hermitian one , i.e. , we have@xmath71by using the linear transformations@xmath72where @xmath73 , @xmath74 $ ] .
the linearity of the transformations allows @xmath75 _ { \pm } & = & \delta _
{ kk^{\prime } } \delta _ { \sigma \sigma ^{\prime } } , \label{can cr k } \\ \left [ \bar{a}_{k,\sigma } , \bar{a}_{k^{\prime } , \sigma ^{\prime } } \right ] _ { \pm } & = & \left [ a_{k,\sigma } , a_{k^{\prime } , \sigma ^{\prime } } \right ] _ { \pm } = 0.\end{aligned}\ ] ] however , when dealing with the dirac inner product , the quasi - canonical commutation relations [ commu_rela ] @xmath76_{\pm } & = & [ a_{k,\sigma } , a_{k^{\prime } , \sigma } ^{\dagger } ] _ { \pm } = \sec \theta \delta _ { kk^{\prime } } , \label{commu_rela_a } \\ \lbrack \bar{a}_{k,-\sigma } ^{\dagger } , \bar{a}_{k^{\prime } , \sigma } ] _ { \pm } & = & [ a_{k,\sigma } , a_{k^{\prime } , -\sigma } ^{\dagger } ] _ { \pm } = i\sigma \tan \theta \delta _ { kk^{\prime } } , \label{commu_rela_b}\end{aligned}\ ] ] will be taken into account .
such quasi - canonical commutation relations reflect the subtle features of the system : when dealing with different @xmath77 , @xmath78 and @xmath79 act as canonical conjugate pairs and the system displays hermitian behavior .
we can gain some insight regarding the role of the quasi - canonical statistics .
we will see shortly that such a model displays the similar dynamics as a hermitian ladder .
we start our investigation from the quantum dynamics of various initial wavepackets . in the situation of a hermitian ladder ,
any two wavepackets propagate independently and the total probability is preserving . considering an arbitrary state involving both upper and lower bands @xmath80 with @xmath81 , we have@xmath82_{\pm } \left\vert \text{vac}\right\rangle & & \label{phi_t } \\ + \sum_{k,\sigma } f_{k,-\sigma } ^{\ast } f_{k,\sigma } e^{-i2\sigma \delta t}\left\langle \text{vac}\right\vert [ \bar{a}_{k,-\sigma } ^{\dag } , \bar{a}% _ { k,\sigma } ] _ { \pm } \left\vert \text{vac}\right\rangle & & \notag \\ = \sec \theta + i\tan \theta \sum_{k,\sigma } \sigma f_{k,-\sigma } ^{\ast } f_{k,\sigma } e^{-i\sigma 2\pi ( t / t_{\text{d } } ) } , & & \notag\end{aligned}\]]where @xmath83 denotes the period of the oscillation .
the first term gives the contribution from single band , while the second term captures the influence of the non - hermiticity .
for vanishing @xmath84 we recover the unitary evolution in hermitian system . evidently , @xmath85 for a state with @xmath86 , which involves only a single mode .
note , however , that mathematically speaking the time dependent terms can vanish even in the case of @xmath87 , e.g. additional orthogonality of the wavepacket with multiple eigenmodes . to demonstrate this , we study the evolution of initial wavepackets of the form @xmath88
\left\vert \text{vac}\right\rangle , \notag\end{aligned}\]]which is the superposition of wavepackets @xmath89 and @xmath90 , where @xmath91 @xmath92 .
the time evolution of wavepacket is a powerful tool for understanding the dynamical property of hermitian quantum systems @xcite .
recently , the propagation of wavepacket in discrete systems has been utilized as flying qubit for quantum state transfer @xcite . in the hermitian case
an initially gaussian state stays gaussian as it propagates for a long time , especially for the case of @xmath93 @xcite . for a sufficient broad wavepacket ( @xmath94 )
, we have @xmath95 .
equation ( [ gwp_k_space ] ) can also be expressed in the coordinate space spanned by \{@xmath96 , @xmath97 } as @xmath98 \left\vert \text{vac}\right\rangle , \notag\end{aligned}\ ] ] which involves both eigenmodes ( @xmath99 ) and actually composed of four wavepackets with centers at @xmath100th and @xmath101th sites of the legs @xmath102 and @xmath55 , and with the velocities @xmath103 and @xmath104 , respectively . to investigate the dynamics of the dirac norm , substituting @xmath105 into eq .
( [ phi_t ] ) , we have @xmath106 where @xmath107 @xmath108 .
we note that if the two wavepackets of eq .
( gwp_k_space ) are well separate in @xmath77 or @xmath6 space initially ( wavepackets orthogonal in @xmath77 or @xmath6 space ) , the weighted exponential factor becomes zero , then the probability is always conserved in the evolution even they meet each other in the coordinate space @xmath6 .
this indicates that for states having additional orthogonal modes , hermitian like behavior still holds even multiple eigenmodes are involved . to show more detailed propagation behavior , we study the profile of @xmath109 ( @xmath110 ) , where@xmath111it is a convenient way to investigate the dynamical properties from two typical cases : ( a ) @xmath112 , @xmath113 and ( b ) @xmath114 , @xmath115 . in case ( a )
, the situation corresponds to two counter - propagating wavepackets , with the evolved wave function @xmath116 \left\vert \text{vac}\right\rangle \notag \\ & = & \left\vert \psi ^{\prime } \left ( n_{a}+2\kappa t , n_{b}-2\kappa t,\pi /2,-\pi /2,0\right ) \right\rangle , \notag\end{aligned}\ ] ] where the approximation of taylor expansions for @xmath117 around @xmath118 are used for two wavepackets and @xmath119 represents the superposition of two wavepackets as state @xmath120 but with different overall phases .
it shows that the evolved state is still the independent nonspreading wavepackets .
similarly , the evolved wave function for case ( b ) has the form @xmath121 \left\vert \text{vac}\right\rangle .\end{aligned}\ ] ] it has more clear profile in the coordinate space @xmath122 , i.e.@xmath123 ^{2}/2}e^{ij\pi /2}a_{j,\ell } ^{\dagger } \left\vert \text{vac}\right\rangle , \notag\end{aligned}\]]where@xmath124obviously , it represents two breathing shape - invariant wavepackets propagating along two legs of the ladder with the breathing period @xmath125 .
furthermore , the dirac norm @xmath126 ( @xmath110 ) and @xmath127 can be obtained as the form@xmath128as mentioned in the introduction , the profile of the evolved wave function @xmath109 can be observed in experiment . in practice ,
the quantum - optical analogy has been employed to visualize the dynamics in the non - hermitian system @xcite . in this context
, the light intensity corresponds to @xmath129 ( for a review , see @xcite ) and the profile corresponds to the light intensity distribution along its propagation direction .
it follows that a manifestation of the non - hermitian nature of @xmath130 is represented by the relative phase @xmath84 between the breathing oscillations of the two legs , which also leads to the time - dependent dirac probability .
the profiles of the evolved wave functions and the dirac norms are plotted in figs .
[ fig_evol ] and [ fig_pro ] .
we can see that in case ( a ) the evolved wavepackets propagate independently and the dirac norms are preserving .
it indicates that although the hamiltonian is non - hermitian , due to the quasi - canonical commutation relations which is a direct consequence of the hermitian connection structure between clusters , it acts as a hermitian ladder for some initial state .
in contrast , the dynamics of case ( b ) differs drastically from the hermitian case and the dirac norm is no long preserved .
further , the phase difference between the breathing oscillations on the two legs can also be observed in case ( b ) .
in summary , we show in this paper within the context of a class of non - hermitian lattice systems , which consist of a set of isomorphic pseudo - hermitian clusters combined in a hermitian manner , that hermitian like dynamics could be observed in such non - hermitian systems , including the property that the time evolution is dirac probability preserving . as an application
, we investigate a concrete network , a @xmath1-symmetric ladder , composed of many pseudo - hermitian dimers .
it is shown that it acts as a hermitian system in the following sense : besides the reality of the spectrum and probability preserving , the propagation of certain wavepackets exhibit the same behavior as that in a hermitian ladder .
our finding indicates that the reality of the spectrum as well as the dirac probability preserving dynamics can occur in a system that violating the axiom of hermiticity .
this will pave the way for the development of descriptions of quantum system and provide a topic of considerable interest in a wide range of subjects .
a. mostafazadeh , int .
mod . phys .
* 7 * , 1191 ( 2010 ) .
a. guo , g. j. salamo , d. duchesne , r. morandotti , m. volatier - ravat , v. aimez , g. a. siviloglou , and d. n. christodoulides , phys .
lett . * 103 * , 093902 ( 2009 ) . | we investigate a pseudo - hermitian lattice system , which consists of a set of isomorphic pseudo - hermitian clusters coupled together in a hermitian manner .
we show that such non - hermitian systems can act as hermitian systems .
this is made possible by considering the dynamics of a state involving an identical eigenmode of each isomorphic cluster .
it still holds when multiple eigenmodes are involved when additional restriction on the state is imposed .
this hermitian dynamics is demonstrated for the case of an exactly solvable @xmath0-symmetric ladder system . |
consider a graph @xmath3 and a straight - line embedding of @xmath4 in the euclidean plane . for simplicity
, we identify each vertex with the corresponding point in @xmath5 .
an embedding of @xmath4 is _ greedy _ if for every pair @xmath6 , vertex @xmath0 has a neighbor @xmath2 in @xmath4 , for which it is @xmath7 , where @xmath8 denotes the euclidean distance between points @xmath9 and @xmath10 .
equivalently , every pair @xmath6 is joined by a distance - decreasing , or _ greedy _ , path .
greedy embeddings are motivated by geometric routing in wireless sensor networks .
given such an embedding , we can use vertex coordinates as addresses . to route a message to a destination
, a vertex can simply forward the message to a neighbor that is closer to the destination , and a successful delivery is guaranteed .
the existence of greedy embeddings has been studied for various graph classes .
papadimitriou and ratajczak @xcite conjectured that every 3-connected planar graph has a greedy embedding in the euclidean plane .
this conjecture has been proved independently by leighton and moitra @xcite and angelini et al .
both proofs use the fact that 3-connected planar graphs have a spanning _
christmas cactus _ subgraph .
a christmas cactus is a connected graph in which every two simple cycles have at most one vertex in common and in which every cutvertex is part of at most two biconnected blocks .
the authors show that every christmas cactus has a greedy embedding .
however , both constructions produce embeddings of exponential size in the worst case . in order for the greedy embedding to be practical for geometric routing
, it must be possible to represent vertex coordinates using only few bits , otherwise , message headers containing the destination address would be too big @xcite .
goodrich and strash @xcite showed how to construct an euclidean greedy embedding of a christmas cactus , in which the coordinates of every vertex can be encoded using only @xmath11 bits . in the presented encoding scheme for
the vertex coordinates , their positions in the euclidean plane are not stored explicitly , and the drawings might still have exponential size .
angelini et al .
@xcite proved that some trees require exponential aspect ratio of the edge lengths .
it was open whether this bound also holds for christmas cactuses . in this note
we prove moitra s conjecture that euclidean greedy embeddings of christmas cactuses require exponential area in the worst case @xcite .
we now present a family of christmas cactuses that requires exponential aspect ratio of edge lengths in every greedy embedding . for an integer @xmath12 , consider the christmas cactus @xmath13 with root @xmath14 in fig .
[ fig : gk ] .
we then construct the cactus @xmath15 by attaching the roots of 30 copies of @xmath13 to a cycle of size 31 ; see fig .
[ fig : fk ] .
we shall prove that the aspect ratio of edge lengths in every greedy embedding of @xmath15 is at least @xmath16 .
the following fact follows from lemma 3 in @xcite .
we rename the vertices for brevity : @xmath21 , @xmath22 , @xmath23 , @xmath24 , @xmath25 ; see fig .
[ fig : lengths - decrease ] .
note that every greedy @xmath26-@xmath27 path as well as every greedy @xmath27-@xmath26 path must contain @xmath28 and @xmath29 .
therefore , the path @xmath30 is greedy in both directions .
thus , the ray with origin @xmath28 and direction @xmath31 and the ray with origin @xmath29 and direction @xmath32 diverge @xcite .
the paths @xmath33 and @xmath34 are also greedy in both directions , therefore , @xmath35 and @xmath36 .
it is @xmath43 . similarly , @xmath44 .
also , @xmath45 .
thus , by considering the triangle @xmath46 it follows : @xmath47 and @xmath48 .
since it is @xmath49 , it is @xmath50 .
analogously , it is @xmath51 .
it follows : next , recall that it is @xmath55 , for @xmath56 and @xmath57 . therefore , @xmath58 . also , since the path @xmath59 is greedy in both directions , it is @xmath60 .
now consider @xmath61 . since @xmath48 , it is @xmath62 , and @xmath63
. therefore , cactus @xmath13 has @xmath70 vertices .
thus , cactus @xmath15 has @xmath71 vertices . by lemma [ lem : lengths - decrease ]
, every greedy embedding of @xmath15 contains an embedding of @xmath13 , such that it is @xmath72 .
therefore , the ratio of the longest and shortest edge in every greedy embedding of @xmath15 is at least @xmath73 .
m. t. goodrich and d. strash .
succinct greedy geometric routing in the euclidean plane . in y.
dong , d .- z .
du , and o. ibarra , editors , _ algorithms and computation ( isaac09 ) _ , volume 5878 of _ lncs _ , pages 781791 .
springer , 2009 . | an euclidean greedy embedding of a graph is a straight - line embedding in the plane , such that for every pair of vertices @xmath0 and @xmath1 , the vertex @xmath0 has a neighbor @xmath2 with smaller distance to @xmath1 than @xmath0 .
this drawing style is motivated by greedy geometric routing in wireless sensor networks .
a christmas cactus is a connected graph in which every two simple cycles have at most one vertex in common and in which every cutvertex is part of at most two biconnected blocks .
it has been proved that christmas cactus graphs have an euclidean greedy embedding .
this fact has played a crucial role in proving that every 3-connected planar graph has an euclidean greedy embedding .
the proofs construct greedy embeddings of christmas cactuses of exponential size , and it has been an open question whether exponential area is necessary in the worst case for greedy embeddings of christmas cactuses .
we prove that this is indeed the case . |
the state of the art of classical computer concepts is rapidly approaching the physical limits as fabrication technology of metal oxide semiconductor logic is minimized to scales where quantum effects determine device properties . while these natural limitations of classical electronics are the dead end for the development of conventional electronics , they open up possibilities for new alternative concepts such as spintronics and quantum computing ( qc ) @xcite . a concept for a silicon based solid state spin quantum computer as outlined by kane @xcite , combines the advantages of conventional semiconductor technology with regard to the high degree to which this technology has been developed and the fundamental concepts of qc , the massive parallel processing of information by coherent quantum states .
kane s concept takes advantage of the two nuclear spin energy eigenstates of @xmath0p - donor nuclei which can be used as well isolated ( long relaxation times ) quantum bits ( qbits ) if they are embedded in a nuclear spin
free crystalline @xmath3si matrix .
interaction between these nuclear spin qbits can be controlled by electric fields from charged metal gates above and between the donor atoms which can selectively increase the hyperfine interaction between the localized electron donor states as well as the exchange interaction between electron donor states of different @xmath0p - atoms @xcite . before an implementation of the silicon based spin
qc is possible , many technological challenges have to be overcome among which the problem of a single spin readout is particularly difficult . in the original proposal
@xcite , the readout of nuclear spin states is done by charge measurements of the qbits electronic shell which can contain one or two donor electrons from adjacent @xmath0p - atoms .
recently , other proposals for the measurement of a single nuclear spin state have been made utilizing single electron transistors @xcite or spin transport in combination with spin refrigeration / spin readout devices @xcite . in this study ,
a concept is presented , which utilizes spin dependent charge carrier recombination in silicon for nuclear spin measurements .
electronic transitions between localized , singly occupied paramagnetic band gap states in silicon have known to be spin
dependent since lepine @xcite discovered that electron spin resonant ( esr ) changes of such defect states can change recombination rates and therefore photoconductivity .
the spin selection rules on these recombination transitions arise from the weakness of spin orbit coupling in silicon which imposes spin conservation . thus , before a transition into a singly occupied state can occur , the two electrons participating have to be in a spin state with singlet content . a model of spin dependent recombination has been developed by kaplan et al .
@xcite , who suggested , that spin
dependent recombination processes are always preceeded by a selective pair formation of the recombining electrons and holes .
figure [ fi1 ] illustrates an example of such a process , which was observed by kanschat et al .
@xcite in hydrogenated microcrystalline silicon ( @xmath4c - si : h ) . due to the disorder of this material ,
shallow trap states exist close to the conduction band which can be singly occupied with excess conduction electrons ( ce ) at low temperatures . beside these ce centers , additional dangling bond defect states ( db )
exist which are excellent recombination centers . a formation of a spin pair as described by kaplan et al .
@xcite takes place , when a ce state in proximity of a db is occupied .
similar to this example , spin dependent recombination mechanisms exist in other silicon based materials such as dangling
bond recombination in amorphous silicon @xcite , various donor acceptor and donor radiation
defect recombination pathes in crystalline silicon ( c - si ) @xcite , in silicon devices @xcite and even at c - si / silicon dioxide interfaces @xcite where spin dependent recombination occurs between stress induced traps and deep interface states . with regard to silicon qc
, the selectiveness of these spin dependent recombination mechanisms raises the question , whether they could be utilized for a readout of electronic and hence also nuclear spin states , especially since @xmath0p donors are known to show nuclear spin resonance influence on recombination @xcite .
precondition for the information readout of coherent spin states with recombination processes is the ability of a given transition to reflect the coherence of the spin states involved , which means the spin
pair states that determine whether a recombination transition takes place or not must not fall into one of the four eigenstates , before a second transition , the actual electronic transfer takes place .
this however has never been proven in the past since the experiment used for the detection of all the processes mentioned above ( often refered to as electrically detected magnetic resonance , edmr ) is a pure incoherent steady state magnetic field
sweep experiment , where a constant current of light or electrically injected excess charge carriers is measured while a constant microwave radiation is imposed on the sample . when the magnetic field reaches the resonance of one of the involved centers , a change of the steady state recombination and thus a current change takes place .
while edmr shows that certain magnetic centers have an influence on recombination , it fails completely to reveal any information about coherence times of the systems involved which are not only the spin
relaxation times of the respective spin centers but also electronic transition times such as recombination or dissociation of the spin pair . in the following an experiment
is presented that shows how coherent spin precession of electrons in localized band gap states can govern recombination rates .
based on this observation we propose an architecture for a recombination based readout of @xmath0p - qbits .
time domain measurements of spin dependent recombination ( tsr ) were carried out on the ce - db mechanism in @xmath4c - si : h , that is mentioned above .
tsr is the electrical detection of pulsed esr and therefore the time resolved equivalent of r85 mm edmr @xcite .
it combines the advantages of edmr with regard to the selective detection of distinct recombination processes and the advantages of pulsed esr with regard to its ability to detect coherent spin - motion effects .
the idea of the experiment carried out is to manipulate the steady state of the ce - db pair ensemble with pulsed esr on a nanosecond time scale and to observe the transient photocurrent response that is determined by the recombination rate through these centers .
excess charge carriers are generated by a continuous light source .
since singlet states have only short lifetimes , a high density of triplet states is present in the steady state when resonant microwave radiation is absent @xcite . the orientation of the magnetic moments of @xmath5pairs is illustrated in fig .
[ fi2 ] in the rotating frame bloch sphere representation @xcite of the two respective spins contained in the charge carrier pairs .
if a microwave that is in resonance with the dangling bond and polarized along the @xmath6axis is switched on , rabi oscillations of the db spins take place .
the influence of the microwave pulses on the ce centers is small due to their different resonance frequencies .
the rabi oscillation of the db spin would lead to an oscillating recombination rate which could be detected as an oscillation of the photocurrent , if the inhomogeneities present in @xmath4c - si : h @xcite did not cause a rapid dephasing of the db ensemble within less than one oscillation period ( illustrated in fig .
[ fi2 ] ) .
this dephasing is still a coherent process , however , its effect on the recombination rate ( strong attenuation of the rabi induced oscillation ) makes it indistinguishable from incoherence .
thus , a rephasing of the ensemble has to be induced , so that incoherence and dephasing becomes distinguishable . as illustrated in fig .
[ fi2 ] , a reversal of the direction of spin precession caused by a reversed @xmath7-microwave field causes such a rephasing .
the @xmath7field reversal , done by a microwave phase change that is introduced at an arbitrary time @xmath8 , causes a maximum recovery of the @xmath5density at time @xmath9 and hence , a minimum of the recombination takes place .
this minimum , which we call a recombination echo @xcite will only be detectable , if incoherent processes are not faster than the time scale of the pulse sequence .
the detection is a proof of the coherence of the charge carrier pair s spin motion .
experimental details with regard to the @xmath4c - si : h sample and the tsr setup are identical to those of ref .
@xcite execept that the laser intensity was 300mw , the sample temperature @xmath10 = 30k and the intensity of the coherent microwave radiation 500w .
the experimental results are displayed in figs .
[ fi3 ] and [ fi4 ] .
note that the detection of fast photocurrent changes in the pa - range with constant offsets in @xmath4a - range is extremely difficult due to noise limitations .
the inital decrease of the real time photocurrent displayed in fig.[fi3 ] is therefore due to the rise time of the detection setup .
as explained elsewhere @xcite , the amplitude of the exponential photocurrent relaxation back to the steady state is determined by the coherent spin state prepared during the ns - pulse sequence .
hence , a measurement of the real time signal at a certain time after the rise time versus the pulse length reveals the spin motion during the pulse sequence on a ns - scale .
this pulse length dependence , measured at the photocurrent minimum ( @xmath11s ) after the pulses , is displayed in fig.[fi4 ] .
it clearly displays the photocurrent increase after a sequence of two 100ns long microwave pulses with equal @xmath7-field strength but opposite phase orientations .
the width of the echo displayed in fig .
[ fi4 ] turns out to be antiproportional to the @xmath7-field strength which proves the involvement of rabi oscillation in the recombination echo effect ( not shown here ) .
the two real time transients displayed in fig.[fi3 ] are recorded after resonant excitation of the db center . in both cases , the total length and the power of the pulses were identical and only a phase change of 180@xmath12 was introduced in one case .
the smaller amplitude of one transient is solely due to this phase change after 100ns which led to the rephasing of triplet states .
an identification of singlet and triplet states with the two binary digits ( labeled `` 1 '' and `` 0 '' for instance ) used for information processing allows a direct readout with current measurements .
based on the observation above , a setup for a recombination readout of kanes silicon based quantum computer is proposed as illustrated in fig .
the device proposed consists of a point defect that introduces a deep level paramagnetic state in proximity to the @xmath0p - donor which is to be read out , a readout gate ( r - gate ) , and an additional a - gate above the defect .
the latter is only necessary if the defect is induced by a deep level donor with nuclear spin ( @xmath13 ) .
when the nature and the implementation of the paramagnetic deep level state will be discussed in the following section , it will be refered to as db in this section in consistency with the dangling bond used for the experimental demonstration discussed above . the idea behind this setup is to take advantage of pauli s principle , similar to the charge readout proposed by kane @xcite .
as long as the r - gate is charged negatively , the wave function overlap between the donor electron of the @xmath0p and the db is small which minimizes a transition probability between the two states and keeps the spin
exchange negligibly small as well .
when the r - gate is charged positively , such that the wave function overlap between the @xmath0p and the db is sufficiently large , exchange interaction increases .
if this increase is introduced slowly , an adiabatic change of the spin - pairs energy eigenstates can take place from an uncoupled product base into a set of singlet and triplet states . at low temperatures ( @xmath14mk ) ,
the uncoupled pair is polarized in a @xmath1-state as long as the coupling is absent . when the exchange is increased slowly , this @xmath1-state remains either unchanged or shifts into a @xmath2-state , depending on the orientation of the @xmath0p - nucleus ( illustrated in fig .
[ fi6 ] ) . note that in presence of a second nuclear spin at the db - site , the a - gate above this site would have to be charge positively in order to minimize hyperfine coupling .
after the r - gate has been `` opened '' and the electronic pair is in singlet or triplet state , an electronic transition can take place which charges the db ; this transition however is possible only when the electronic pair has singlet content in advance and thus , the charging of the db state depends on the nuclear state of the @xmath0p qbit . the timing and a scetch of the transitions during a readout sequence
are displayed in fig .
as long as quantum operations on the qbit take place , no light is imposed on the sample and the r - gate is charge negatively .
when the readout sequence begins at a time @xmath15 , the bias of the r - gate is slowly inverted towards a positive voltage .
once exchange coupling is established after a time @xmath16 , it remains unchanged until the actual recombination transition has taken place .
the time necessary for this transition is of the order of the electronic pair states s lifetime @xmath17 , after which the exchange interaction is switched off again .
the nuclear spin state of the @xmath0p donor is then coded in the donor and db charge state . a short ( @xmath18 , ns - range ) and weak ( nw - range ) laser pulse imposed on the sample will then increase conductivity by the generation of a few pairs of excess electrons and holes . if the donor and the db state are not charged (
no transition ) , a slow decrease of the photoconductivity will follow , which is determined by slow band band recombination in the ultra pure @xmath3si bulk . if the two states are charged , a fast decay will take place since charge carriers will be trapped in the charged states .
thus , the level of the photoconductivity a time @xmath19 after the end of the laser pulse will reveal the result of the readout process .
note , that the readout process itself automatically neutralizes the two states such that a new series of operations can take place after its completion .
the experiments presented above are the motivation for the concept proposed in this study and many other questions will have to be answered experimentally before an actual proof of this concept can be given .
the implementation of the singly occupied deep level state will mostly depend on whether it is possible to control its location and whether its charge carrier capture cross sections and the given transition times of the system are fast enough .
the dangling bond present in amorphous and microcrystalline silicon would be an ideal system with regard to the latter properties .
however , the high disorder in the microcrystalline morphology of silicon makes it a bad choice for qc .
since no process has been established which allows the creation of a single db with @xmath20-accuracy at an arbitrary site , a different way of deep level implementation must be chosen .
various impurities and defects provide singly occupied , paramagnetic deep level donor states in c - si , some of which are gold , potassium , strontium or chromium @xcite .
an additional possibility could be interface recombination .
recombination at c - si / sio@xmath21interfaces has shown to have spin - dependent paths @xcite .
hence , the very same processes which limited the original pure c - si qc - concept and led to the proposal of si / sige - heterostructures @xcite could actually be benificial for a recombination readout mechanism . beside the questions for a proper implementation of the deep level center ,
several other issues remain to be investigated : due to the necessity of error correction , one of the preconditions for qc is that readout does not destroy the state of the qbit .
therefore , only p - deep level transitions are possible which leave the @xmath0p relaxation unchanged .
important for the feasibility of the recombination readout will also be the question of the overall readout time which should not exceed a lower microsecond range .
when the p - deep level transition takes place , the slow relaxation of non equilibrium polarization at low temperatures could pose a problem due to charged p - deep level pairs which are neutralized by excess charge carriers produced in excited spin states .
the time needed for the relaxation of these states could be very long .
however , this problem may be solved by injection of excess carriers by spin polarized electronic injection instead of light injection .
finally , the question for the one qbit sensitivity has to be asked : time domain measurements of spin dependent recombination have shown to be sensitive enough to detect the influence of as little as 50 charge carrier pairs , even though the experiment was carried out in the presence of a strong constant photocurrent offset . for the readout as proposed ,
a measurement without offset could be made and the detection would be even more sensitive .
a 2ev laser pulse of 1ns length and 3.2nw intensity produces 10 electron hole pairs if the internal quantum efficiency is assumed to be 100% .
the detection of these 10 charge carrier pairs would require a pa
current measurement on a microsecond scale which does not pose a problem however , whether the 10% difference between the two readout results caused by the one recombination center are detectable can only be proven when an implementation of the proposed setup is available .
the demonstration that coherent spin states of electrons and holes trapped in ce and db states can govern their recombination rate has motivated the idea of a readout mechanism for @xmath0p qbits in c - si .
a concept , based on a sequence of exchange gating , light pulse and photocurrent measurements was proposed , various implementations suggested and possible limitations and drawbacks discussed . | it is shown that coherent spin motion of electron hole pairs localized in band gap states of silicon can influence charge carrier recombination .
based on this effect , a readout concept for silicon based solid
state spin quantum computers as proposed by kane is suggested .
the @xmath0p quantum bit ( qbit ) is connected via hyperfine coupling to the spin of the localized donor electron .
when a second localized and singly occupied electronic state with an energy level deep within the band gap or close to the valence edge is in proximity , a gate controlled exchange between the @xmath0p nucleus and the two electronic states can be activated that leaves the donor deep level pair either unchanged in a @xmath1-state or shifts it into a singlet state @xmath2 . since the donor deep level transition is spin
dependent , the deep level becomes charged or not , depending on the nuclear spin orientation of the donor nucleus .
thus , the state of the qbit can be read with a sequence of light pulses and photo conductivity measurements . |
there is a long history of studying exactly soluble one - dimensional models in statistical physics .
while sometimes these models are interesting in their own right , the main motive for their study is usually to cast some light upon behavior in higher , more physical dimensions @xmath7 , such as @xmath8 and @xmath9 , where exact solutions can not usually be found . in this paper
we study a quasi - one - dimensional system of @xmath10 hard disks of diameter @xmath0 confined by impenetrable walls separated by distance @xmath11 , where @xmath12 is the width of the region containing the centers of the disks ( see fig .
[ fig : notation ] ) . for @xmath13 , a disk may make contact with its nearest- and next - nearest - neighbor ( nnn ) disks , but neighboring disks can not pass each other .
the ordering of the disks is still preserved : @xmath14 , where @xmath15 is the position of the center of disk @xmath16 , measured along the channel , and @xmath17 is the total length available to the disk centers .
the transverse coordinates of the disk centers are denoted by @xmath18 , where @xmath19 .
we have in an earlier paper studied the case when @xmath20 , which permits only nearest - neighbor ( nn ) contacts @xcite .
all our results and figures for the nnn model are for @xmath21 .
the numbering 1 , 2 , 3 , , @xmath10 of the disks follows the ordering of their @xmath22 coordinates . in fig .
[ fig : buckled](a ) , the first disk interacts with its nearest neighbor , the second disk , but it also interacts with another disk , the third disk , giving rise to a nnn coupling . if instead we had @xmath20 , such nnn couplings would not be possible : we call this case the _ nn model_. thus , the nn and nnn models describe disks in narrow channels of different widths . for spin glasses
, nnn interactions between the spins in a one - dimensional system can introduce frustration .
suppose we have spins 1 , 3 , 5 , in the lower of two rows of spins , and spins 2 , 4 , 6 , in the upper row , as shown below : @xmath23 with nnn interactions , spin 3 interacts with both of its nearest neighbors 2 and 4 , but also with its next - nearest neighbors 1 and 5 .
this spin system has the same topology of interactions as will be found for some arrangements of the disks in the narrow channel system .
while the spin system is effectively one dimensional , the spins in the triangles 123 and 234 could be frustrated with appropriate spin couplings .
frustration is regarded as a key feature for producing glassy behavior , so perhaps it is not surprising that the presence of nnn contacts introduces features that are absent in the narrower channel which has only nn contacts . in the nn model ,
the dynamics becomes activated around a packing fraction @xmath24 @xcite .
this is due to the onset of caging , a feature connected with the growth of a particular structural feature , zigzag order @xcite .
zigzag order ( see eq . for its definition )
can be regarded as a form of bond - orientational order . accompanying this activated dynamics are many of the other features which one associates with glasses , such as the appearance of a plateau in time dependent correlation functions , dynamic heterogeneities , etc .
we expect there to be very similar behavior in the nnn model , but that is not the focus of this paper .
instead our interest is in a feature which occurs _ only _ in the nnn model , at a packing fraction @xmath25 .
( the maximum packing fraction for @xmath26 is @xmath27 ; see fig .
[ fig : buckled](a ) . )
the phenomenology of this feature turns out to be very similar to that of the ideal glass transition @xcite .
there are standard arguments against there being any genuine phase transition in one dimension @xcite which we expect will apply to our narrow channel problem , and will rule out any genuine phase transition at finite pressure .
the ideal glass transition of hard spheres in three dimensions is estimated to occur at a @xmath2 of around @xmath28 @xcite ; the analogue of the dynamical transition at @xmath29 is around @xmath30 . in two dimensions ,
activated dynamics has been observed to set in at around @xmath31 in binary @xcite and weakly polydisperse @xcite mixtures of hard disks and the ideal glass transition is estimated to occur at a packing fraction @xmath32 @xcite .
it is usually believed that the dynamical transition at @xmath29 is avoided in any finite dimension and it is uncertain whether the transition found at @xmath2 in an approximate treatment , valid for dimension @xmath33 , will survive to lower dimensions .
these approximate calculations involve the analogue of one - step replica symmetry breaking , which may be destroyed by fluctuation effects in any physically interesting dimension @xcite .
thus while the absence of a genuine phase transition in our channel system is not surprising , it is possible that glass transitions in two or three dimensions might also be avoided .
a piece of evidence that the feature at @xmath2 might be the remnant of the ideal glass transition is from the growth of a particular correlation length , which grows to its maximum value ( around @xmath5 ) at @xmath2 and then falls .
we shall explain in sec .
[ sec : conclusion ] that this length , which we call the shear - penetration length @xmath34 , is the distance over which a shear displacement of the two rows of disks decreases with distance from an amorphous boundary .
it is very similar in nature to the penetration length for amorphous order calculated by molecular dynamics in refs .
the penetration length can be defined in equilibrium ( like our @xmath34 ) and is conceptually simpler than the point - to - set length scale @xcite , which is the length scale below which ergodicity is broken in a region of fluid with amorphous boundaries ; but both of these length scales are expected to diverge at the glass transition @xcite . in three dimensions , the largest values found for the penetration length to date are smaller than those we can find in our system , so in this sense we have in our model better evidence for the ideal glass transition than has been found in three dimensions .
the second aspect of our feature at @xmath2 which makes us identify it as an avoided ideal glass transition is that the @xmath6-relaxation time @xmath35 , which we estimate using the ideas in ref .
@xcite , will appear to diverge in a vogel fulcher manner , @xmath36 which is of the form expected for behavior near the ideal glass transition . because there is no real singularity at @xmath2 as the transition is avoided , the time scales will not truly diverge there and will in fact diverge only for @xmath37 .
there is a long tradition in the theory of glasses of attributing glass behavior to the growth of particular types of structural order , such as icosahedral order in three dimensions @xcite .
recent work by cubuk et al .
@xcite on the flow of jammed and glassy systems under stress has shown that regions that are susceptible to rearrangement can be discovered by machine - learning methods that combine information derived from several features of the local structure , such as the radial distribution of particles and the bond angles .
since our narrow - channel system with nnn contacts is relatively simple , we are able to identify the collective motions of the disks that lead to the apparent ideal glass behaviors .
thus the onset of slow dynamics near @xmath29 we can identify with the growth of bond - orientational order .
the feature at @xmath2 which seems related to the ideal glass transition is connected with the growth of crystal - like order .
we shall discuss this matter in detail in sec .
[ sec : conclusion ] . in the limit of very wide channels ,
when a real transition can arise rather than just an avoided transition , the feature at @xmath29 becomes , we suspect , the fluid hexatic transition , and the feature at @xmath2 evolves to be the hexatic crystal transition .
the length scale @xmath34 appears to be similar to the penetration length for amorphous order studied in the glass community . for a system of disks in a channel of any length
, work must be done to displace one disk along the channel , relative to its nearest neighbors .
but in a short channel with length @xmath38 , displacing one disk will cause the displacement of all of the disks of one row relative to the other row , as shown in fig .
[ fig : buckled](c ) .
this rigidity for @xmath38 is like that of an amorphous solid .
another example of the utility of the nnn channel problem for the study of glasses is the work of ashwin and bowles @xcite .
they determined exactly the number @xmath39 of jammed states at packing fraction @xmath40 and calculated from it the complexity @xmath41/n$ ] .
they found an apparent kink in @xmath42 at a packing fraction @xmath43 , a density which we shall call @xmath44 , above which @xmath42 decreases rapidly to zero .
the jammed states which exist above this density contain increasing amounts of the buckled - crystal ordering seen in fig .
[ fig : buckled](a ) .
it is this feature which makes it natural to identify the kink in @xmath42 with random close packing , which for hard spheres in three dimensions occurs at a packing fraction close to @xmath45 @xcite : jammed states with @xmath46 are known to contain increasing amounts of face - centered cubic crystal ordering @xcite .
ashwin and bowles also found that using the lubachevsky
stillinger algorithm @xcite to find jammed states always led to a jammed state with a packing fraction close to @xmath47 if one started from a high enough initial density .
observe that in our nnn channel system @xmath48 , @xmath25 , and @xmath49 , which are rather close to each other , compared to their corresponding values in three dimensions .
these numerical values are for @xmath50 .
the general formulas are @xmath51 and @xmath52 , where @xmath53^{1/2}\ ] ] is the length of one unit cell of the buckled crystal ; @xmath54 has been estimated only for @xmath26 @xcite .
notice that @xmath3 coincides with @xmath2 when @xmath55 and that the difference between them increases as @xmath56 increases . from the results of simulations in three dimensions ,
it has been noted @xcite that the rate at which distinct , disordered states disappear with increasing density has a sharp maximum at a particular density . this maximum becomes narrower for larger system sizes , perhaps becoming infinitely sharp and leading to a well - defined @xmath47 in the thermodynamic limit @xcite .
kamien and liu @xcite argue that such a feature , which corresponds to a discontinuity in @xmath42 , will be accompanied by a divergence of the pressure on a metastable branch of the equation of state .
they suggest that the position of this singularity may be obtained by extrapolation from the low - density portion of the equilibrium equation of state .
we suspect that in our equilibrium calculation , the apparent divergence of the force @xmath57 , when extrapolated to @xmath2 , might have some connection with their argument
. it should be noted , however , that the equilibrium state at @xmath2 is certainly not jammed and that a snapshot of it would show much more disorder than in fig .
[ fig : buckled](b ) , with disks at random distances from the walls .
( this can be deduced from the density profile discussed later in sec .
[ sec : eqofstate ] and illustrated in fig .
[ fig : density ] . )
using such a configuration as the starting point for the lubachevsky stillinger algorithm @xcite will always result in a slightly denser state , quite possibly one with packing fraction close to @xmath47 .
an interesting discussion of possible singularities in the metastable fluid branch of the hard sphere fluid is given in ref .
@xcite .
the plan of the paper is as follows . in sec .
[ sec : model ] we set up the transfer matrix formalism for the nnn channel problem .
it is more intricate than that for the nn case and is harder to solve numerically , especially for large packing fractions . in sec .
[ sec : eqofstate ] we study the equation of state and show that the pressure apparently diverges when @xmath40 is extrapolated to @xmath2 .
we also obtain the local density variation across the channel .
as the packing fraction increases , the disks are pushed more and more against the channel walls , but for @xmath58 some of the disks lift away from the walls in order to achieve the buckled crystalline state of fig . [
fig : buckled](a ) . in sec .
[ sec : corrlengths ] we obtain from the eigenvalues of the transfer matrix three of the correlation lengths : the length scales for zigzag and buckled - crystal order ( @xmath59 and @xmath60 ) , and the shear penetration length @xmath34 . in sec .
[ sec : probdensities ] we describe the distribution functions for the longitudinal separation of nearest- and next - nearest - neighbor disks . from these we can obtain the variances of these separations , which are used in sec . [ sec : lengthscalesexplained ] to understand the variation of @xmath34 with @xmath57 , the longitudinal force applied to the system . in sec .
[ sec : highdensity ] we use the transfer matrix formalism to describe how the system evolves towards the crystalline state as @xmath61 . finally in sec .
[ sec : conclusion ] we further discuss the shear penetration length @xmath34 and explain why the @xmath6-relaxation time might be expected to show an apparent divergence at @xmath2 .
in this section we set up the transfer matrix formalism for disks in a channel that is wide enough to allow contact between next - nearest neighbors . because a disk can not overlap its nearest neighbors ,
it is convenient to reparametrize the configuration not in terms of @xmath62 but rather in terms of horizontal separations @xmath63 , defined by @xmath64 where @xmath65 ^ 2\right){}^{1/2}.\ ] ] notice that when the disks touch , @xmath66 . in terms of the variables
@xmath67 , the configuration integral can be written @xmath68 where @xmath69 is the heaviside step function and @xmath70 in eq . , the variables @xmath18 are integrated over the range @xmath71\,$ ] ; the same limits will be assumed in later expressions that involve @xmath72-integrations . note that the first step function in imposes the constraint @xmath73 and the remaining step functions exclude configurations in which next - nearest - neighbor disks overlap .
a variant of our model which is periodic in the @xmath72-direction has been studied in ref .
@xcite .
it is inconvenient to work in an ensemble in which the length @xmath17 is constant , though this was the approach taken by barker @xcite . instead we follow kofke and post @xcite in transforming to an ensemble in which the longitudinal force , @xmath57 , is constant .
this amounts to taking the laplace transform of with respect to @xmath17 , @xmath74 where @xmath75 .
equation agrees with the results of kofke and post @xcite and varga et al .
@xcite who considered only the case @xmath76 in developing their transfer matrix formalism : in that special case , the step functions are all unity , because @xmath77 ; the @xmath63-integrations can be completed analytically , giving @xmath78 in this paper , we shall be concerned mainly with the case @xmath79 , so that we can not make use of the simplification . in
, the integrals over @xmath67 can be performed one at a time .
we write @xmath80 where the function @xmath81 can be calculated iteratively , starting from @xmath82 subsequent functions @xmath83 ( with @xmath84 to @xmath10 ) are given by @xmath85 for @xmath86 , the form of the function @xmath83 approaches that of the nodeless eigenfunction corresponding to the largest eigenvalue @xmath87 of the integral equation @xmath88 so that @xmath89 where @xmath90 is a constant , independent of @xmath91 . accordingly , the free energy @xmath92 derived from the partition function @xmath93 is given by @xmath94 this free energy is essentially @xcite the gibbs free energy with respect to the longitudinal force @xmath57 , so that @xmath95 but it is also ( via the dependence on the channel width ) of the helmholtz type with respect the transverse force @xmath96 , @xmath97 we are aware of no simple , general relationship between @xmath57 and @xmath96 ; in particular , the stress tensor can not be assumed to be isotropic .
it is of interest to see how the integral equation of ref .
@xcite arises from when @xmath98 .
for that case , the step function @xmath99 in is always unity , so that the dependence on @xmath100 reduces to @xmath101\,u_n(y_2,y_1,0)$ ] . by using this form in and integrating both sides with respect to @xmath102 , we obtain @xmath103 where @xmath104 .
the one - dimensional integral equation is of the form derived by kofke and post @xcite , with eigenvalue @xmath105 .
the integral equation has the same form when all the @xmath72-coordinates are reversed in sign .
each solution consequently has either even or odd parity , @xmath106 , corresponding to the parity of the solutions of kofke and post s integral equation for the case of nearest - neighbor interactions . by using the symmetry of the function @xmath99 under interchange of @xmath107 with @xmath108
, it is straightforwardly shown that real eigenfunctions @xmath109 and @xmath110 corresponding to different real eigenvalues @xmath111 and @xmath112 are orthogonal , in the sense that the bilinear form @xmath113 & { } \equiv { \int_0^\infty}ds{\int}dy_1 { \int}dy_2\ , e^{\beta f(s+\sigma_{1,2})}\nonumber\\ & \qquad{}\times u_n(y_1,y_2,s)\,u_m(y_2,y_1,s ) \label{eq : bfunctional}\end{aligned}\ ] ] equals zero for @xmath114 .
this integral always converges because , from , both eigenfunctions decrease as @xmath115 $ ] for large @xmath116 .
equation can also have complex eigenvalues , which occur in complex conjugate pairs : the corresponding eigenfunctions @xmath109 and @xmath117 satisfy @xmath118=0 $ ] and are also orthogonal to the real eigenfunctions .
these complex solutions are used later in sec .
[ sec : highdensity ] , in a calculation of the correlation length at high density . the orthogonality relation can be used to determine the coefficient @xmath90 in eq . .
assuming that the function @xmath119 can be expanded in terms of the eigenfunctions @xmath109 , @xmath120 projection of both sides onto the eigenfunction @xmath121 gives the exact relation @xmath122 \nonumber\\ & = \frac1{\beta f}{\int_0^\infty}ds{\int}dy_1 { \int}dy_2 \,u_1(y_1,y_2,s ) \nonumber\\ & = a{\lambda_1 ^ 2}\,b[u_1,u_1]\ , , \label{eq : a}\end{aligned}\ ] ] which will be of use later , in sec .
[ sec : distribution ] .
the eigenvalue problem can also be formulated variationally .
the eigenvalues are stationary points of the functional @xmath123 = \frac{k[u , u]}{b[u , u]}\,,\ ] ] where @xmath124 = & { \int_0^\infty}ds_1 { \int_0^\infty}ds_2 { \int}dy_0 { \int}dy_1 { \int}dy_2\,\nonumber\\ & \quad\times u(y_1,y_2,s_2)\ , \theta_1\ , v(y_1,y_0,s_1)\ , .
\label{eq : kfunctional}\end{aligned}\ ] ] this stationary property of the functional @xmath125 can be verified by evaluating the functional derivative @xmath126 at @xmath127 , making use of the eigenvalue equation satisfied by @xmath109 .
the variational formulation of the eigenvalue problem can be used to obtain the length of the system , by using in the form @xmath128\ , .
\label{eq : lambdalen}\ ] ] the functional @xmath125 depends on @xmath57 explicitly via the factor @xmath129 $ ] and implicitly via the eigenfunction @xmath121 .
terms in that arise from the @xmath57-dependence of @xmath121 vanish , owing to the variational property of @xmath125 , leaving @xmath130\ , , \label{eq : lbyn}\ ] ] which is analogous to eq .
( 2.12 ) of ref .
the form of the last expression strongly suggests that @xmath131 \label{eq : rho}\end{aligned}\ ] ] is the equilibrium distribution function for the horizontal separation and @xmath72-coordinates of a neighboring pair of disks .
this identification will be verified directly in the following section . in a similar way
, the variational principle can be used to derive a wall contact theorem @xcite for the transverse force , @xmath132 = \rho_1(h/2)\ , , \label{eq : wallcontact}\ ] ] where @xmath133 is the probability density for a disk to have its center at @xmath72 : @xmath134 the probability distribution @xmath135 for neighboring disks @xmath136 and @xmath91 to be found at separation @xmath137 with @xmath72-coordinates @xmath138 and @xmath139 can be derived systematically by performing all of the integrals in except those for @xmath137 , @xmath138 and @xmath139 .
the integrals over variables @xmath140 and @xmath141 for @xmath142 give a factor of @xmath143 and those for @xmath140 and @xmath144 for @xmath145 give a factor of @xmath146 .
the normalized result for @xmath147 is @xmath148 for the special case @xmath149 , use of shows that reduces correctly to @xmath150 , which is normalized to unity . in the thermodynamic limit @xmath151 ,
most of the disks @xmath91 are far from both ends of the system , so that @xmath86 and @xmath152 .
we can make the approximation for @xmath83 , @xmath153 , and @xmath81 , giving @xmath154 which , by using the result , is equivalent to .
we solve the integral equation by discretization , approximating the integrals by sums .
this converts into a matrix eigenvalue problem @xmath155 in which the matrix @xmath156 is dense and nonsymmetric . for large values of @xmath57 , the eigenfunctions are large when @xmath102 and @xmath157 lie within a distance @xmath158 of the walls @xmath159 , where @xmath160 it is important to treat these regions accurately without increasing the dimension of @xmath156 unnecessarily . to achieve this ,
we make a change of variables @xmath161\ , , \label{eq : ytransform}\ ] ] where @xmath162 and the values of @xmath163 are uniformly spaced in the interval @xmath164 $ ] .
the positive parameters @xmath165 , @xmath166 , and @xmath167 are chosen so that approximately two thirds of the values of @xmath72 given by lie within a distance @xmath158 of the walls and are nearly uniformly spaced within these narrow regions . for simplicity ,
the eigenfunctions @xmath168 are tabulated at equal intervals in @xmath116 . for any specified values of @xmath157 and @xmath102 ,
there is a value of @xmath116 ( call it @xmath169 ) beyond which the eigenfunctions @xmath168 decrease as @xmath115 $ ] .
the portion of the @xmath116 integration with @xmath170 can be completed analytically , so that there is no need to tabulate @xmath109 beyond this point ; this helps to reduce the dimension of the eigenvalue problem .
elsewhere , the integrands are linearly interpolated from their tabulated values . even with the @xmath72 transformation and the treatment of the @xmath116 integration discussed above , the matrix @xmath156 in eq .
is still too large to be stored in computer memory , so that the eigenvalue problem must be solved by an iterative method .
( with 100 points for each of @xmath102 , @xmath157 , and @xmath116 , @xmath156 is a @xmath171 matrix . with fewer points
we have found it hard to capture accurately the rapid variation of the eigenfunctions near the wall , as is visible in fig .
[ fig : density ] . ) simple iteration of is adequate for finding the first two real eigenfunctions , provided the process starts with a function of the appropriate parity .
when more eigenfunctions are required , we find it convenient to use the arpack subroutine library @xcite , which implements the iterative method due to arnoldi @xcite .
to test our numerical methods , we have used them to calculate the equation of state and the correlation length @xmath172 for a system of disks in a narrow channel with @xmath55 .
for this case in which nnn disks do not interact , these quantities are more easily found by solving the one - dimensional integral equation @xcite , which leads to a much smaller ( and symmetric ) transfer matrix .
results from the two approaches are in excellent agreement and give us some confidence in the reliability of our later results for the case @xmath173 .
the equation of state is central to the study of fluids . from it
one can determine the phase diagram .
equation gives @xmath174 in terms of the @xmath57-dependent eigenfunction @xmath175 , and so determines the equation of state ; that is , the relation between @xmath57 and the packing fraction @xmath40 , where @xmath176 the equation of state is shown in fig .
[ fig : eos ] , for moderate values of @xmath57 and @xmath173 .
notice the shoulder near @xmath177 : this is due to the onset of zigzag order @xcite . if one could study progressively wider channels we believe that this shoulder would evolve into the feature seen in the equation of state of a two - dimensional fluid for @xmath178 , which is the transition between the fluid and hexatic and crystal transitions @xcite .
we expect that if one could study the compressibility for a range of channel widths it would have a peak at the shoulder which would evolve into the singularity seen in the infinite system in ref .
this is reminiscent of what happens in ( say ) strips @xmath179 of the two - dimensional ising ferromagnet : the peak in the specific heat for finite @xmath180 evolves into a divergence at criticality as @xmath181 @xcite . in figs .
[ fig : eosnn ] and [ fig : eos0.95 ] we compare the longitudinal and transverse pressures @xmath182 and @xmath183 calculated for the nn and nnn models with the equation of state of a system of disks in two dimensions , simulated by molecular dynamics @xcite . notice that at very low densities or packing fractions , figs .
[ fig : eosnn ] and [ fig : eos0.95 ] show that the equation of state approaches that of the two - dimensional system , which is isotropic : @xmath184 .
however , as the packing fraction increases beyond about @xmath185 , the system becomes sensitive to its channel structure . for wider channels , the approximate equality of @xmath186 and @xmath187 for small @xmath40 should extend to higher packing fractions , but probably not in any smooth way because of layering effects .
it can be observed also that for the wider channel studied in fig .
[ fig : eos0.95 ] , the results are closer overall to those of the full two - dimensional system than those for the narrower system displayed in fig . [
fig : eosnn ] .
the force @xmath57 becomes large for @xmath188 , the density of the most closely packed jammed state , which is the buckled crystal identified in ref .
@xcite and shown in fig . [ fig : buckled](a ) . from the general arguments of salsburg and wood @xcite
, we might expect to find @xmath189 as @xmath40 approaches @xmath3 , with @xmath190 .
this form for the equation of state can also be obtained from the high - density limit of the integral equation , as discussed in sec .
[ sec : highdensity ] .
it is therefore of interest to consider @xmath191 as a function of @xmath40 , as shown in fig .
[ fig : onebyf ] .
this shows a linear dependence on @xmath40 over a range of @xmath40 , extrapolating to zero at @xmath192 , which is _ not _ the maximum density of the buckled crystal .
instead , the data in fig .
[ fig : onebyf ] is consistent with the equation @xmath193 with an effective dimensionality @xmath194 , at least for the data that lies close to the straight line .
there is no actual divergence of @xmath57 as @xmath195 , as the data points in the vicinity of @xmath2 do not lie on the straight line .
the only true divergence in @xmath57 is at the largest possible value of @xmath40 , @xmath48 , which is the packing fraction of the buckled crystal .
our finding @xmath196 has a counterpart in the case of three - dimensional hard spheres @xcite , where @xmath197 rather than
3 . notice that our @xmath198 is close to @xmath199 .
this may be because in fig .
[ fig : buckled](b ) the top row and the bottom row are partially decoupled so that each behaves like a one - dimensional hard rod gas of maximum density @xmath2 .
each row contains @xmath200 disks which as @xmath195 will exert a force @xmath201 as for hard rods : @xmath202 $ ] .
the combined force exerted by both rows is @xmath203 .
the packing fraction @xmath204 at which the force appears to diverge corresponds to the configuration shown in fig .
[ fig : buckled](b ) .
this is not a jammed state as it can be deformed into fig .
[ fig : buckled](c ) by sliding the top row of disks to the left . as the density increases above @xmath2 ,
disks have to move off the wall to fit into the system and the buckled crystal state starts to develop .
at all but the highest forces , however , the shaded disks in fig . [ fig : buckled](a ) are only lightly pinched by their neighbors , so that the entropy gained by allowing these disks to be delocalized near the walls overcomes the work that must be done to lengthen the system .
if the apparent singularity in the force at @xmath2 is avoided by an evolution to the buckled state , we should expect to see evidence for this in @xmath133 , the probability density for finding a disk at @xmath72 , defined in eq . .
in fig . [
fig : density ] , @xmath205 is plotted as a function of distance from one wall of the channel containing the disk centers . in all cases ,
the distribution has its maximum at the wall , but a secondary maximum starts to appear for @xmath206 , and is well developed for @xmath207 .
this secondary maximum clearly corresponds to the trapping of the shaded disks in fig .
[ fig : buckled](a ) : in the jammed state , the centers of these disks are at a small distance ( roughly @xmath208 for @xmath173 ) from the walls @xmath159 that confine the centers of the disks .
it is possible to obtain a great many correlation lengths from the ratios of eigenvalues of the transfer integral equation . in our work
we focus on three whose interpretation is particularly simple .
the longest correlation length is that associated with the growth of zigzag order or , more precisely , the decay of the correlation between the @xmath72-coordinates of well - separated disks @xmath16 and @xmath209 , @xmath210 where @xmath211 the eigenfunction @xmath121 is of even parity ( it is nodeless ) and @xmath212 has odd parity , so that @xmath72 has a nonzero matrix element between @xmath121 and @xmath212 ; it may also be noted that the eigenvalues @xmath87 and @xmath213 have opposite sign .
these two features are consistent with the zigzag correlations described by eq . .
the @xmath57-dependence of the length @xmath59 is given in fig .
[ fig : zigzagxi ] and is discussed in sec . [
sec : lengthscalesexplained ] .
the character of the eigenvalues @xmath111 for @xmath214 depends on @xmath215 . as @xmath37
the next four eigenvalues are two different complex conjugate pairs , as we shall explain in sec .
[ sec : highdensity ] . outside this limit
( and when @xmath216 is not small ) the first few eigenvalues are real and occur in parity - related doublets of opposite sign , just like @xmath87 and @xmath213 . a second length scale with a simple interpretation is @xmath217 , calculated from the eigenvalues of the first two even - parity eigenfunctions , @xmath121 and @xmath218 .
the functions @xmath121 and @xmath218 are depicted in figs .
[ fig : mode1 ] and [ fig : mode3 ] for the case @xmath219 , for which the correlation length @xmath220 .
the function @xmath218 has an approximately planar nodal surface near @xmath221 , so it is clear that the nn separation @xmath222 will have a large matrix element between @xmath218 and the nodeless function @xmath121 .
accordingly , @xmath34 is the distance over which the nearest - neighbor separations @xmath223 remain correlated .
it is shown in fig .
[ fig : corrlens ] that the correlation length @xmath34 peaks at a value of around @xmath224 for the packing fraction @xmath225 .
note that our @xmath59 and @xmath34 are actually _ dimensionless _ quantities , as may be seen from eq .
, for example . to obtain the physical length scales one has simply to multiply them by @xmath174 .
thus @xmath226 where @xmath4 is the physical length . when @xmath26 and @xmath227 , @xmath228 . the maximum value of @xmath34 shown in fig .
[ fig : corrlens ] therefore corresponds to a physical length scale @xmath229 .
a correlation length that increases to a maximum and then decreases is indicative of an avoided transition .
the value of @xmath40 that corresponds to the peak in @xmath34 is a little less than the value , @xmath230 , at which @xmath231 appears to approach zero ( see fig .
[ fig : onebyf ] ) .
given that there is no true phase transition at @xmath2 , it is perhaps not surprising that different measures for locating the underlying avoided transition should disagree slightly .
the third length scale , @xmath60 , that we have studied is a measure of the extent of buckled - crystal order . for the larger values of @xmath232
, it can be determined from the ratio @xmath233 , where @xmath234 is the magnitude of the largest complex conjugate pair of eigenvalues ( see sec .
[ sec : highdensity ] ) .
figure [ fig : corrlens ] shows that @xmath60 grows with increasing @xmath216 , whilst remaining small in the range of @xmath235 that we can study numerically .
its expected behavior for @xmath188 is discussed later , in sec .
[ sec : highdensity ] .
we have studied the probability density @xmath236 for the centers of neighboring disks to be separated by a distance @xmath237 using @xmath238 this probability density changes in a striking manner as the packing fraction or force @xmath57 is increased .
figure [ fig : denx ] shows @xmath239 for @xmath240 , which corresponds to a packing fraction at which zigzag order is starting to develop . for small values of @xmath215 ,
the main feature is a cusp at @xmath241 ; but , as zigzag order grows , a broad hump develops at a smaller value of @xmath237 : both features may be seen in fig .
[ fig : denx ] . results for larger values of @xmath216 are shown in fig .
[ fig : moredenx ] . when @xmath242 the cusp at @xmath243 has nearly disappeared and
would be undetectable on the scale used in fig .
[ fig : moredenx ] . as @xmath244 increases further , the position of the new peak moves towards the value @xmath245 , as would be expected from fig .
[ fig : buckled](b ) . for @xmath246 ,
the peak is nearly gaussian in shape and we have directly determined its width by obtaining the variance @xmath247 .
the standard deviation of the nearest - neighbor separation is plotted in fig .
[ fig : sd - xnn ] : it varies as @xmath248 as @xmath215 increases , which is a much slower decrease than would have been naively expected and reflects the ease of sliding the top row with respect to the bottom row in fig .
[ fig : buckled](b ) .
the separation of next - nearest neighbors has much smaller fluctuations .
the variance of @xmath249 is given by @xmath250+[x_{k+1}-x_k])\nonumber\\ & = 2{\operatorname{var}}x_{\rm nn}+2{\operatorname{cov}}(x_{k+2}-x_{k+1},x_{k+1}-x_k)\ , .
\label{eq : covar}\end{aligned}\ ] ] the covariance term ( which is negative ) can be obtained from @xmath121 . by following a procedure similar to that described in sec .
[ sec : distribution ] , it can be shown that @xmath251}{k[u_1,u_1 ] } \nonumber\\ & = \frac{k[e_1,e_1]}{\lambda_1b[u_1,u_1]}\ , , \label{eq : covexpr}\end{aligned}\ ] ] where @xmath252 . the calculation of @xmath253 from and thus requires only the distribution of @xmath237 ( obtained from @xmath121 ) and a single application of the transfer matrix to @xmath254 . in fig .
[ fig : sd - xnnn ] the standard deviation of @xmath255 is plotted for a range of @xmath57 up to @xmath256 .
it is close to the expected @xmath191 form , but with some flattening off at the largest values of @xmath57 .
eventually , the standard deviation must tend to a constant , independent of @xmath57 , because , as the state of highest density is approached , the nnn separation can have either of two distinct values ( see fig .
[ fig : buckled](a ) ) , with fluctuations of order @xmath191 around them .
there are two important length scales in our problem : the length scale associated with the growth of zigzag or bond - orientational order , @xmath59 ( see eq . ) , and the length scale @xmath34 , which describes how the correlations between the nearest - neighbor separations of pairs of disks at numbers @xmath91 and @xmath257 decay with @xmath116 , @xmath258 where @xmath167 denotes the cumulant of the correlation .
@xmath34 is a measure of the distance along the channel over which the system with @xmath259 comes to resemble fig .
[ fig : buckled](b ) .
we will show later that @xmath34 can also be regarded as the size of a localized thermal fluctuation in which disks of one row are displaced relative to the other row .
we have previously obtained @xmath59 and @xmath34 from the eigenvalues of the transfer matrix , but in this section we give simple arguments to explain their dependence on @xmath57 . in ref .
@xcite we were able to give an account of the dependence of @xmath59 for the nn model by relating it to the typical distance between defects .
for the nnn model , there are many types of defects @xcite , but we believe the following argument captures the physics of what is going on
. for moderate values of @xmath57 , typical configurations of the system resemble fig .
[ fig : buckled](b ) .
a pair of defects in this zig - zag pattern can be created by removing one disk from the pattern and reinserting it elsewhere , between two disks that lie on the same side of the channel .
if the two rows of disks maintain their relative position between the two defects , the increase in length of the system will be @xmath0 @xcite , giving an additional length per defect @xmath260 ; the work done to create it will be @xmath261 .
we therefore expect the density of defects to vary as @xmath262 $ ] and the correlation length @xmath59 to vary as @xmath263 $ ] , which is the typical spacing of the defects .
the dashed line in fig . [ fig : zigzagxi ] shows that this simple argument gives a reasonably good account of the numerical data for @xmath264 , despite its neglect of any relaxation of the system near the defects .
we also note that agreement can not be maintained indefinitely , as our argument ignores the eventual appearance of buckled crystalline order .
the length scale @xmath34 has no obvious connection with defects , but instead can be understood as an effect due to the accumulation of the random variations in the spacing of nnn disks . to estimate @xmath34 , we assume that the coupling of the two rows of disks is weak and that the gaps between next - nearest neighbors are approximately independent random variables , like the gaps between neighbors in a one - dimensional gas of hard rods . with the notation
@xmath265 we can write @xmath266 the left - hand side of this identity is an alternating sum of @xmath267 next - nearest - neighbor separations : to the extent that these separations are independent random variables , the variance of the left - hand side will grow as @xmath268 as @xmath267 increases . at the same time , the variance of the right - hand side will increase towards @xmath269 as the correlation between @xmath270 and @xmath271 decreases .
we therefore expect the correlation to be small when @xmath272 . by setting @xmath273
we obtain the estimate @xmath274 fig .
[ fig : corrlens ] shows how well this simple formula works .
it accounts for both the increase of @xmath34 to its maximum as a function of @xmath57 as well as its decrease .
the correlation length @xmath34 can also be understood as the typical number of disks that participate in a thermal fluctuation in which a portion of one row moves relative to the other row .
the configuration shown in fig .
[ fig : buckled](c ) could be regarded as the most extreme fluctuation of this kind : it involves the correlated motion of many disks , and so should be expected to be long - lived on the scale of the collision time . in a localized fluctuation in which @xmath267 disks of one row are displaced relative to the other , the distances between the disks in one row ( averaged over a time that is short compared with the duration of the fluctuation ) will change by small amounts of order @xmath275 , with @xmath276 .
the total strain energy , which is on the order of the thermal energy @xmath277 , will be approximately @xmath278 where the effective spring constant @xmath279 for the force conjugate to the next - nearest neighbor separation can be estimated from the fluctuation response relation @xmath280 with our estimate for @xmath281 , eqs . and lead to @xmath282 , where @xmath34 is given by eq . .
we have been unable to obtain numerical solutions of the integral equation for very large values of the force , @xmath283 . to help fill this gap left by our numerical work , in this section we present an analytical method of solution which is expected to become exact in the limit @xmath284 .
we relate the period - six correlations of the buckled crystal to the asymptotic form of the eigenvalues and we calculate the correlation length for crystalline order .
we also indicate what features of the eigenfunctions will lead to the high - density form of the equation of state given in eq . . for @xmath285
, the system of disks adopts the high - density buckled crystal configuration shown in fig .
[ fig : buckled](a ) , but for any finite @xmath57 the crystalline order will be disrupted by the presence of mobile defects .
the concentration of these defects will vary as @xmath286 , where @xmath287 is the additional length needed to accommodate a defect in the crystal .
we therefore expect the mean spacing of defects to increase with @xmath57 according to @xmath288 our notation for this length scale reflects the fact that @xmath60 will also be the distance over which crystalline order persists in the presence of defects .
the shortest ( and hence most abundant ) defects are of two kinds , illustrated in fig .
[ fig : defects ] . each kind of defect can be created by inserting a pair of disks into the buckled crystal .
if six disks , comprising one complete unit cell of length @xmath165 in the buckled crystal , are removed from a region of perfect crystal , these disks can be reinserted elsewhere in the crystal to create three defects . the net increase in length
is @xmath289 where @xmath290 and @xmath291^{1/2}$ ] , so that @xmath292 we believe it is worth understanding how the correlation length @xmath60 given by and can arise from the integral equation and also how the periodicity of the buckled crystal can be explained .
these two matters are very closely connected : the periodicity must appear in the correlation functions , and the dominant contributions to the large - distance correlation functions can be constructed from the eigenfunctions whose eigenvalues are closest in magnitude to the largest eigenvalue @xmath87 .
it follows that there should , for sufficiently large @xmath57 , be _ complex _ eigenvalues of whose magnitudes are very similar to @xmath87 and whose phases can be related to the periodicity of the buckled crystal . in what follows we show that complex eigenvalues with these properties can be inferred from the form of the integral equation . for sufficiently large @xmath57 , the solutions of will be sharply peaked near @xmath159 and @xmath293 , with peak widths proportional to @xmath294 . as seen in fig .
[ fig : density ] , this transverse localization is starting to appear above @xmath219 and it is well developed when @xmath295 .
but in addition to this localization in @xmath102 and @xmath157 , the eigenfunctions will also eventually show a similar degree of localization in @xmath100 . in the buckled crystal
( see fig .
[ fig : buckled](a ) ) , adjacent disks are either touching , with @xmath296 , or non - touching , with @xmath297 , where @xmath298 .
in addition to these two separations , the shaded disks in the second of the configurations shown in fig . [
fig : defects ] are separated by a third distance , @xmath299 .
this local information regarding pairs of disks can be used to contruct an approximate solution of the integral equation . from the preceding discussion
, we expect a solution @xmath300 of to have eight narrow peaks with widths @xmath301 .
if @xmath302 has even parity , there will be only four distinct peak values of @xmath302 , which we denote by @xmath303 where @xmath158 is given by .
approximate equations satisfied by the quantities @xmath304 to @xmath305 can be read off from eq . ; for example , @xmath306 in eq . , the factors @xmath307 and @xmath119 arise from the integrations over @xmath308 and @xmath309 in eq . ,
so they will be of order @xmath310 for sufficiently large @xmath57 .
the exponential factor , which corresponds to @xmath311)$ ] in , can be transferred to the left - hand side of . by proceeding in this way for the remaining quantities
@xmath312 , we obtain a matrix eigenvalue equation @xmath313 where @xmath314 is a vector with components @xmath312 and the matrices @xmath315 and @xmath316 are given by @xmath317 and @xmath318{$0 $ } } & { \makebox[\bwidth]{$0 $ } } \\ 0 & k / k_\epsilon & 0 & 0 \\ 0 & 0 & 0 & 1 \\ 0 & 0 & 1 & 0 \end{pmatrix } , \ ] ] where @xmath319 and @xmath320 .
the symmetry of the matrix @xmath315 is not obvious from the intuitive derivation given here . to refine the argument
, it can be supposed that the eigenfunction @xmath302 is approximated by a linear combination @xmath321 of localized functions @xmath322 .
projection of eq . onto the basis functions
then leads to an equation of the form , in which the symmetry of the matrices @xmath316 and @xmath315 follows from the symmetry of the functionals @xmath323 $ ] and @xmath324 $ ] defined in eqs . and . it might therefore be an interesting task to construct a localized basis adapted to the special case @xmath188 , but we do not attempt this here . instead we proceed to the qualitative results that can be derived from the matrix eigenvalue equation .
the equation for @xmath325 , obtained from @xmath326 , is @xmath327 to determine the form of the solutions of this equation , it is helpful first to introduce a scaled , dimensionless variable @xmath328
. written in terms of @xmath329 , becomes @xmath330 where @xmath331 is small and @xmath332 and @xmath333 are expected to be independent of @xmath57 for large @xmath57 . in the limit @xmath334 ( or @xmath335 ) ,
the solutions of are @xmath336 , 1 , @xmath337 , and @xmath338 , where @xmath337 is a complex cube root of unity . for small @xmath339 ,
the solutions can be expanded to first order in @xmath339 , giving @xmath340 for the three roots with modulus close to unity ; the fourth root is of order @xmath341 .
so far we have considered only the even - parity solutions of .
the same method can be applied to the odd - parity solutions , which leads to a further three roots @xmath329 with modulus close to unity .
these roots are given by the same expressions as in , but with the opposite signs .
this leads to a two - fold degeneracy in @xmath342 for the real solutions and a four - fold degeneracy for the complex solutions . in our approximate treatment
, the cause of the degeneracy is our selective use of configurations in which neighboring disks lie at ( or near ) opposite sides of the channel , so that perfect zigzag order is assumed .
this assumption about the configurations is reasonable , given our knowledge that @xmath343 for large @xmath57 , but it does prevent us from estimating the correlation length @xmath59 from the ratio of the first two real eigenvalues . the result that emerges from our analysis is that , for sufficiently large @xmath57 , the six eigenvalues of largest magnitude are approximately @xmath344 , with @xmath345 to 5 .
an observable such as @xmath72 will have nonzero matrix elements between @xmath121 and the complex conjugate pair of eigenfunctions with @xmath346 and @xmath347 , which are the primitive sixth roots of unity .
the correlation function @xmath348 will therefore contain a contribution that oscillates with period six in @xmath116 . to calculate the correlation length @xmath60
we must go beyond the zeroth order in @xmath339 and use the expressions given in eq . .
we find @xmath349/|\omega + \tfrac13\,p_1\omega^2\,r| ) \nonumber\\ & \simeq \frac2{p_1}\exp(\beta f\delta_d)\,,\end{aligned}\ ] ] which is of the form anticipated in , though we have not determined the quantity @xmath332 , which will be a function only of the ratio @xmath350 .
finally , we use the large-@xmath57 approximation @xmath351 to derive the equation of state for @xmath188 . the free energy per disk
is given by @xmath352 in which the @xmath57-dependence of @xmath119 and @xmath353 has been used to obtain the coefficient of the logarithmic term ; the contribution written as @xmath354 contains only terms that do not increase with @xmath57 .
differentiation of @xmath355 with respect to @xmath57 ( cf .
eq . ) gives the equation of state @xmath356 expected on the general grounds discussed in ref .
the factor @xmath357 in the expression for the force is the same coefficient 2 that multiplies @xmath358 in eq . .
in the context of our derivation , the source of this factor is the two - fold integration in eq . , where the eigenfunction is localized within regions whose widths are proportional to @xmath294 .
we have solved the transfer integral equation appropriate to disks moving in a channel of a width which permits nnn contacts and we have determined its thermodynamic properties and some of its correlation lengths numerically exactly .
the most striking feature of our results are those connected with the features at @xmath29 and at @xmath2 .
the packing fraction @xmath29 marks the point at which bond - orientational order ( i.e. zigzag order ) sets in , and will be where the dynamics of the system will start to become activated .
we believe that if the channel width were increased this feature would evolve into the fluid
hexatic phase transition of hard disks @xcite .
the new feature , which is absent for narrower channels with only nn contacts , is that at @xmath2 .
we believe it would evolve for wider channels into the hexatic
crystal phase transition @xcite .
it has a clear connection with the development of structural order . despite being of structural origin , the behavior near
@xmath2 mimics that normally associated with the ideal glass transition . at the ideal glass transition ,
a divergence of the point - to - set length scale is expected @xcite and this will be accompanied by a divergence of the penetration length for amorphous order @xcite .
a review of some of the length scales which are perhaps of relevance to amorphous systems can be found in ref .
this penetration length is the distance over which the effect of a boundary condition ( imposed by freezing the positions of a subset of spheres ) extends into the liquid ; for the case of three dimensions it has been determined from molecular dynamics simulations by gradenigo et al .
@xcite and by berthier and kob @xcite . for our system of disks in a channel ,
freezing the positions of the disks in the region @xmath359 supplies a boundary condition for the motion of the mobile disks in the region @xmath360 : in particular , it fixes the nearest - neighbor separation near @xmath361 .
following the discussion in secs .
[ sec : corrlengths ] and [ sec : lengthscalesexplained ] , the time - averaged nearest - neighbor separations @xmath362 will approach the bulk value @xmath174 as @xmath363 for large @xmath91 , so that we can identify the penetration length with @xmath34 . at the ideal glass transition
it is expected that the relaxation time @xmath35 should diverge @xcite .
it is argued in refs .
@xcite that @xmath364 is the time at which a disk can escape its cage by crossing from one side of the channel to the other . within transition - state theory ,
whose accuracy for a system of hard disks has been tested in ref .
@xcite , @xmath364 is given by @xmath365 where the length @xmath366 is @xmath367 , as discussed in ref .
@xcite for the case of nearest - neighbor interactions . on using eq . for @xmath57 , the relaxation time will _ appear _ to diverge in a vogel fulcher manner of eq
. for a range of @xmath40 less than @xmath2 .
note that at @xmath2 it will be a very challenging problem to study the properties of the equilibrated system by molecular dynamics : the densities are so close to @xmath3 that the time scales in the system will be extremely long .
any real divergence of @xmath364 will , of course , be avoided in the narrow channel systems , and we suspect that it might also be avoided for hard spheres in three dimensions @xcite . above the ideal glass transition discussed in the infinite dimensional limit , i.e. when
@xmath368 , particles are expected to be caged near their initial positions forever , whereas for @xmath369 , the particles can escape to infinity . in our narrow channel system , the particles are in one sense caged for all time as they can not pass each other ; nevertheless , we have seen in sec . [
sec : probdensities ] that there is a qualitative difference in the nature of the confinement above and below @xmath2 .
below @xmath2 , the nearest - neighbor separations in the @xmath22-direction fluctuate as @xmath248 , whereas above @xmath2 the fluctuations are smaller and will decrease as @xmath191 as @xmath188 .
similarly , there is a change in the nature of the fluctuations in the @xmath72-direction on passing through @xmath2 .
below @xmath2 the fluctuations in @xmath72 are small but of order @xmath0 ; above @xmath2 , they will shrink at large @xmath57 like those of the @xmath22-component , as @xmath191 .
thus many properties of our system are similar to the properties of real glasses .
one notable difference between our channel system and the three dimensional hard sphere system is that in the latter system , glass behavior occurs in the supercooled metastable region at densities above that of the fluid - crystal transition .
our own work is a study of equilibrium behavior , in which the long - lived metastable states are simply included in the thermal average .
but there are parallels nonetheless between the two types of system above @xmath2 : as the packing fraction increases towards its maximum value , both systems develop a jammed crystal ordering .
one of our observations is that the structural feature which is behind the onset of activated dynamics at @xmath29 , which is bond - orientational order in the channel problem , is not the same as the structural feature which produces our @xmath2 .
it is possible that something similar may occur in three dimensional glasses , but simulations of them at @xmath2 are difficult , so that the situation remains unclear , like much else in the study of glasses .
@xmath355 differs from the gibbs free energy by an additive term that depends only on @xmath10 and @xmath371 .
this is because the canonical partition function differs from the configuration integral by @xmath372 factors of the thermal wavelength , @xmath373 .
the expression for @xmath374 is different in the nearest - neighbor case discussed in ref . @xcite . for @xmath20 , removing a disk from one side of the channel leaves a gap of length @xmath375 .
after this gap has been closed , the net increase in length per defect is @xmath376 . | we use an exact transfer - matrix approach to compute the equilibrium properties of a system of hard disks of diameter @xmath0 confined to a two - dimensional channel of width @xmath1 at constant longitudinal applied force . at this channel width , which is sufficient for next - nearest - neighbor disks to interact ,
the system is known to have a great many jammed states .
our calculations show that the longitudinal force ( pressure ) extrapolates to infinity at a well - defined packing fraction @xmath2 that is less than the maximum possible @xmath3 , the latter corresponding to a buckled crystal . in this quasi - one - dimensional problem
there is no question of there being any _ real _ divergence of the pressure at @xmath2 .
we give arguments that this avoided phase transition is a structural feature
the remnant in our narrow channel system of the hexatic to crystal transition but that it has the phenomenology of the ( avoided ) ideal glass transition .
we identify a length scale @xmath4 as our equivalent of the penetration length for amorphous order : in the channel system , it reaches a maximum value of around @xmath5 at @xmath2 , which is larger than the penetration lengths that have been reported for three dimensional systems .
it is argued that the @xmath6-relaxation time would appear on extrapolation to diverge in a vogel fulcher manner as the packing fraction approaches @xmath2 . |
the interstellar medium ( ism ) is a highly compressible turbulent , magnetized fluid , exhibiting density fluctuations on all observable scales .
it has been long realized by many researchers that incompressible hydrodynamic , i.e. kolmogorov , description is inadequate for such a medium ( see * ? ? ?
* for review ) .
scaling relations , if they were obtained for the interstellar gas , would be very helpful for addressing many problems , including the evolution of molecular clouds and star formation .
attempts to include effects of compressibility into the interstellar turbulence description can be dated as far back as the work by @xcite .
there a simple model based on a hierarchy of clouds was presented . according to this picture every large cloud
consists a certain number of smaller clouds , which contain even smaller clouds .
for such a model @xcite proposed a relation between subsequent levels of hierarchy @xmath5 where @xmath6 is the average density inside a cloud at level @xmath7 , @xmath8 is the mean size of that cloud , 3 is the number of dimensions , and @xmath9 is constant that reflects the degree of compression at each level @xmath7 .
the kolmogorov energy spectrum ( @xmath10 ) follows from the assumption of a constant specific energy transfer rate @xmath11 .
lighthill @xcite pointed out that , in a compressible fluid , the volume energy transfer rate is constant in a statistical steady state @xmath12 in an important , but not sufficiently appreciated work , fleck @xcite ( henceforth , f96 ) incorporated above hierarchical model with energy transfer in compressible fluid to obtain the scaling relations for compressible turbulence . by combining the equations ( [ eqn : hierar ] ) and ( [ eqn : enden ] ) fleck @xcite presented the following set of scaling relations in terms of the degree of compression @xmath9 : @xmath13 where @xmath14 and @xmath15 are , respectively , the column density of the fluctuation with the scale @xmath16 and the mass of the cloud of size @xmath16 .
the fluctuations of velocities in f96 model entail the spectrum of velocities @xmath17 . in the spirit of f96 model , @xcite proposed to use the density - weighted velocity @xmath18 as a new quantity , for which the kolmogorov scaling for second order structure functions ( sfs ) can be restored in compressible hydrodynamic turbulence .
their hydrodynamic simulations provided the spectrum for @xmath19 close to -5/3 and they showed that in supersonic hydrodynamic turbulence the structure functions of @xmath19 scale linearly with separation .
will the f96 model be valid for compressible _ strongly magnetized _ turbulence ?
this is the major question that we address in this paper .
we used an second - order - accurate essentially nonoscillatory ( eno ) scheme ( see * ? ? ? * ; * ? ? ? * for details ) to solve the ideal isothermal magnetohydrodynamic ( mhd ) equations in a periodic box with maintaining the @xmath20 constraint numerically .
we drove the turbulence at wave scale @xmath21 ( 2.5 times smaller than the size of the box ) using a random solenoidal large - scale driving acceleration .
this scale defines the injection scale in our models .
the rms velocity @xmath22 is maintained to be approximately unity , so that @xmath4 can be viewed as the velocity measured in units of the rms velocity of the system and @xmath23 as the alfvn velocity in the same units .
the time @xmath24 is in units of the large eddy turnover time ( @xmath25 ) and the length in units of @xmath26 , the scale of the energy injection .
the magnetic field consists of the uniform background field and a fluctuating field : @xmath27 . initially @xmath28 and @xmath29 .
we use units in which the alfvn speed @xmath30 and @xmath31 initially . structures of density , velocity and magnetic field develop completely due to the forcing from uniform initial conditions . the values of @xmath32 have been chosen to be similar to those observed in the ism turbulence . for our calculations we assumed that @xmath33 . in this case
, the sound speed is the controlling parameter , and basically two regimes can exist : supersonic and subsonic .
note that within our model , supersonic means low @xmath34 ( @xmath35 ) , i.e. the magnetic pressure dominates , and subsonic means high @xmath34 , i.e. the gas pressure dominates .
we present results for selected 3d numerical experiments of compressible mhd turbulence with a strong magnetic field for sonic mach numbers @xmath36 between @xmath37 and @xmath38 .
the alfvnic mach number @xmath39 .
mach numbers are defined as @xmath40 and @xmath41 for the sonic and alfvnic mach number , respectively . to study effects of magnetization we also performed superalfvnic experiments with @xmath42 .
all models were calculated with the resolution 512@xmath0 up to 6 dynamical times .
in figure [ fig : spectra ] we present the spectra for velocity @xmath4 and density - weighted velocity @xmath1 for two strongly magnetized models : subsonic ( @xmath43 ) and supersonic ( @xmath44 ) .
naturally , for subsonic model the differences between spectra for @xmath4 and @xmath3 are marginal and both spectra correspond to kolmogorov s @xmath45 scaling ( see subplot in fig . [
fig : spectra ] ) .
however , we can see that for the supersonic case , the velocity spectrum gets steeper .
the steepening corresponds to @xmath46 ( from @xmath47 ) . at the same time
, the spectrum of @xmath3 matches well the kolmogorov slope . .
[ fig : spectra],scaledwidth=80.0% ] in the original kolmogorov theory ( * ? ? ?
* hereafter k41 ) it was shown that the exponent of the third order structure function ( sf ) , e.g. structure function of @xmath4 , @xmath48 , should be equal 1 , i.e. @xmath49 . in figure
[ fig : sf3rd ] we show the sfs of the third order for velocity and the density - weighted velocity for supersonic model .
we checked , that for the subsonic case for both @xmath4 and @xmath3 the index @xmath50 is indeed close to unity .
for the supersonic case , @xmath50 increases with @xmath36 for @xmath4 , but stays the unity for @xmath3 ( see fig .
[ fig : sf3rd ] ) .
this suggests that the kolmogorov universality is preserved for supersonic mhd turbulence when density weighting is applied .
( dashed line ) and @xmath3 ( solid line ) compensated by @xmath51 for supersonic mhd turbulence model .
dotted lines correspond to best fitting within the inertial range .
two dotted vertical lines bound the intertial range .
the sub - panel shows the scaling exponents @xmath50 for @xmath4 ( crosses ) and @xmath3 ( diamonds ) as a function of @xmath36 .
[ fig : sf3rd],scaledwidth=80.0% ] a proper description of turbulence requires higher moments ( see * ? ? ?
* for review ) .
those characterize intermittency , which is in the original k41 model is not accounted for .
a substantial progress in understanding turbulence intermittency is related to a discovery by she & lvque ( @xcite , hereafter sl94 ) , who found a simple form for the scaling of exponents @xmath52 of higher order longitudinal correlations @xmath53 \cdot \hat{\boldsymbol{l } } |^p \rangle \sim l^{\zeta_p}$ ] . while in k41 model @xmath54 , sl94 provides ( after modifications by mller & biskamp @xcite to more general form ) @xmath55,\ ] ] where @xmath56 is related to the scaling of the velocity @xmath57 , @xmath58 is related to the energy cascade rate @xmath59 and @xmath60 is the dimension of the dissipative structures . in hydrodynamic incompressible turbulence , we have @xmath61 and @xmath62 . for mhd turbulence
the dissipation happens in current sheets , which are two - dimensional dissipative structures , corresponding to @xmath63 @xcite .
thus , for subsonic mhd turbulence we expect @xmath64 for both velocity and the density - weighted velocity .
this is what we actually observe in figure [ fig : scalexp ] ( see subpanel ) .
the same scaling , however , is preserved for @xmath3 for _ supersonic _ magnetized turbulence . and @xmath3 for subsonic ( subplot ) and supersonic models .
the plots present unnormalized values of the scaling exponents obtained directly from sfs by fitting the relation @xmath65 within the intertial range , i.e. without using the extended self - similarity @xcite .
[ fig : scalexp],scaledwidth=80.0% ] magnetic field is known to induce anisotropies of compressible mhd turbulence ( see * ? ? ?
anisotropy increasing with the decrease of scale was predicted for alfvnic motions by goldreich & sridhar ( @xcite , henceforth gs95 , see also @xcite ) and confirmed numerically for compressible mhd in @xcite . and @xmath3 in the local reference frame for the supersonic experiment .
sfs for velocity scale as @xmath66 and @xmath67 for parallel and perpendicular directions to the local magnetic field , respectively .
sfs for @xmath3 scale as @xmath68 and @xmath69 for @xmath70 and @xmath71 directions , respectively .
[ fig : local_frame],scaledwidth=80.0% ] for supersonic motions figure [ fig : local_frame ] shows that the scalings for @xmath4 are much steeper in both directions than those predicted by gs95 model ( @xmath72 and @xmath73 for @xmath70 and @xmath71 directions to the local magnetic field , respectively ) .
however , those slopes still give a close to gs95 anisotropy ( @xmath74 ) , which is indicative of the dominance of the alfvnic ( `` incompressible '' ) motions .
note , that in figure [ fig : local_frame ] the sfs are obtained in the system of reference of the _ local _ magnetic field , i.e. the field on the scales of the fluctuations under study . @xmath75 and
@xmath76 denote second order sfs parallel and perpendicular to local magnetic field , respectively . for @xmath3
the scalings are significantly shallower ( @xmath77 and @xmath78 for @xmath70 and @xmath71 directions to the local magnetic field , respectively ) .
the sf in perpendicular direction scales more like incompressible motions , i.e. @xmath79 . the slope of @xmath80 for @xmath3 is smaller than the corresponding one for @xmath81 resulting in the reduced degree of anisotropy ( @xmath82 ) .
this is well visible for small - scale structures as presented in the figure [ fig : contours ] .
intuitively , this can be understood in terms of dense clamps strongly distorting magnetic field as they move in respect to magnetized fluid . and @xmath3 ( left and right plots , respectively ) in the local reference frame for the supersonic experiment .
the small - scale structures show the reduced degree of anisotropy .
[ fig : contours],title="fig:",scaledwidth=50.0% ] and @xmath3 ( left and right plots , respectively ) in the local reference frame for the supersonic experiment .
the small - scale structures show the reduced degree of anisotropy .
[ fig : contours],title="fig:",scaledwidth=50.0% ] our results for velocity show that our simulations of strongly magnetized turbulence provide @xmath83 for @xmath84 .
the spectrum of density fluctuations @xmath85 follows from the scaling relation of density ( see eq .
[ eqn : relations ] ) according to the f96 model .
this suggests the existence of a rising spectrum of density fluctuations within the hierarchy of density clumps when @xmath86 .
( long dashed , short dashed and dot - dashed lines , respectively ) .
solid line correspond to the spectrum of density without filtering .
the embedded table shows values of spectral indices and percent of the total mass and total volume occupied by the filtered density for subsequent thresholds .
[ fig : densspec],scaledwidth=80.0% ] the figure [ fig : densspec ] shows spectra of the low - value filtered densities for supersonic model .
spectra for densities above a threshold of 5% of the maximum value start to be rising .
the figure [ fig : densspec ] contains also a table of mass and volume filling factors , which shows that with increasing threashold , the same amount of matter occupies less space .
we try to make our study more related to _ observations _ which usually measure densities integrated along the line of sight , i.e. column densities , or alternatively study the hierarchy of observed clump masses ( see eq .
[ eqn : relations ] ) .
f96 model assumes the existence of an infinitely extended hierarchy . in our computations
the structures are generated by turbulence at scales less than the scale of the computational box .
therefore the f96 scaling relations ( in eq .
[ eqn : relations ] ) should be modified as follows @xmath87 our procedure of obtaining the scaling relation from column density maps is similar to that in @xcite , with the difference that they dealt with 3d data , while we deal with 2d data .
first , we seek for a local maximum of column density .
then we calculate the average column density within concentric boxes with gradually increasing the size @xmath16 . in case of determining the relation for @xmath88 , instead of averaging we
apply the integration over the boxes .
naturally , the results should correspond to each other .
( left plot ) and for density ( right plot ) for three models of turbulence : subsonic ( @xmath89 ) and supersonic ( @xmath90 2 and 7 , see legend ) .
[ fig : densmass],title="fig:",scaledwidth=50.0% ] ( left plot ) and for density ( right plot ) for three models of turbulence : subsonic ( @xmath89 ) and supersonic ( @xmath90 2 and 7 , see legend ) .
[ fig : densmass],title="fig:",scaledwidth=50.0% ] in figure [ fig : densmass ] we present an example of scaling relation for column density and density for three models of turbulence with @xmath91 0.7 , 2 , and 7 .
one can note , that the relation becomes more steep with the sonic mach number within the intertial range .
the fractal dimensions can be calculated from the relation @xmath92 ( see * ? ? ?
* ) , where @xmath93 is a slope estimated from the plot within the intertial range . for our models the fractal dimension ranges from @xmath94 for the highly supersonic models to @xmath95 for subsonic model .
respectively , the compressibility coefficients for presented models @xmath96 for @xmath97 to @xmath98 for @xmath84 .
the latter roughly consistent with @xmath9 obtained for the velocity sf measurement in section `` kolmogorov scalings for supersonic flows '' .
the differences are probably due to insufficient statistics of rather rare high density peaks . in general , the filling factor of a peak decreases with the maximum density of this peak , which means that the higher maximum of density the peak has the smaller space it occupies . what is the effect of magnetic field on the @xmath3-scaling ? the spectra , third and higher moments of correlations obtained for our superalfvnic simulations with @xmath99 happen to be very similar to the case of strongly magnetized turbulence .
our results indicate that , unlike velocity , @xmath3 is much less affected by magnetic field .
naturally , in super - alfvnic turbulence the anisotropies induced by magnetic field are not observed at larger scales within the intertial range ( cf .
the last paragraph of section `` astrophysical implications '' .
[ [ dependence - of - alpha - on - the - extend - of - inertial - range ] ] dependence of @xmath9 on the extend of inertial range + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + if we combine several facts together , namely , ( a ) that @xmath9 is a function of mach number , ( b ) that the maximum of density correspond to the dissipation scale , e.g. shock thickness scale @xmath100 , ( c ) that the amplitude of density in peaks scales as the mean density times @xmath101 , we have to conclude that as the inertial range from the injection scale @xmath102 to @xmath100 increases , for a given mach number , @xmath9 should decrease .
connecting these facts we get the following relations , @xmath103 , which gives the dependence of @xmath9 on @xmath104 and @xmath36 , namely , @xmath105
. an interesting consequence of this would be a prediction of kolmogorov scaling for supersonic _ velocities _ when the injection and dissipation ranges are infinitely separated .
consequently , the steeper velocity spectra reported in @xcite can be interpreted as an indication of a limited inertial range .
further research justifying such conclusions is required , however . [
[ sins - of - supersonic - turbulence ] ] sins of supersonic turbulence + + + + + + + + + + + + + + + + + + + + + + + + + + + + + ubiquitous small ionized and neutral structures ( sins ) are observed in interstellar medium ( see * ? ? ?
their nature is extremely puzzling if one thinks in terms of kolmogorov scalings for density fluctuations .
the fact that the spectrum of fluctuations of density in supersonic turbulence is shallower that the kolmogorov one is well - known ( see * ? ? ? * and references therein ) .
however , just the difference in slope can not explain the really dramatic variations in column densities observed .
the present paper provides a different outlook at the problem of sins .
we see that , while low amplitude density fluctuations exhibit kolmogorov scaling @xcite , high peaks of density correspond to a rising spectrum of fluctuations .
thus , observing supersonic turbulence at small scales , we shall most frequently observe small amplitude fluctuations corresponding to kolmogorov - like spectrum of density fluctuations . occasionally , but inevitably , one will encounter isolated high density peaks .
an alternative mechanism for getting infrequent large density fluctuations over small scales is presented in @xcite and is related to current sheets in viscosity - damped regime of mhd turbulence .
[ [ clumps - and - star - formation ] ] clumps and star formation + + + + + + + + + + + + + + + + + + + + + + + + + interstellar medium is known to be clumpy .
frequently clumps in molecular clouds are associated with the action of gravity .
our study shows that supersonic turbulence tend to produce small very dense clumps .
if such clumps happen to get jean s mass , they can form stars . therefore , star formation is inevitable in supersonic turbulence .
however , the efficiency of star formation is expected to be low , as the filling factor of peaks decreases with the increase of the peak height . inhibiting of star formation via shearing may dominate in terms of influencing of star - formation efficiencies .
in paper above we have studied the scaling of supersonic mhd turbulence .
we found that : * fleck 1996 model is applicable to strongly magnetized compressible turbulence .
* spectra and structure functions of density - weighted velocities are consistent with predictions of the kolmogorov theory .
* intermittency of density - weighted velocity can be well described by the she - lvque model with the dimension of dissipative structures equal 2 @xcite . *
strongly magnetized supersonic turbulence demonstrate lower degree of anisotropy if described using the density - weighted velocity .
* the high peaks of column densities exhibit increase of the mean values of column densities with the decrease of scale , which may be relevant to the explanation of sins .
* acknowledgments * the research of grzegorz kowal and alex lazarian is supported by the nsf grant ast0307869 and the center for magnetic self - organization in astrophysical and laboratory plasmas .
we thank alexei g. kritsuk for valuable discussion .
r. benzi , s. ciliberto , r. tripiccione , c. baudet , f. massaioli and s. succi , _ phys .
, 48 , 29 ( 1993 ) a. beresnyak , a. lazarian and j. cho , _ astrophysical journal _ , 624 , l93 ( 2005 ) , blc05 j. cho and a. lazarian , _ phys .
rev . letters _ , 88 , 245001 ( 2002 ) j. cho and a. lazarian , _ mnras _ , 345 , 325 ( 2003 ) b. g. elmegreen and j. scalo , _ ara&a _ , 42 , 211 ( 2004 ) r. c. jr .
astrophysical journal _
, 458 , 739 ( 1996 ) , f96 p. goldreich , s. sridhar _ astrophysical journal _ , 438 , 763 ( 1995 ) , gs95 c. heiles & d. stinebring , _ `` small ionized and neutral structures in the diffuse interstellar medium '' _ , eds . m. haverkorn and w.m .
goss , astro - ph/0701656 , ( 2007 ) j. c. higdon , _ astrophysical journal _ , 285 , 109 ( 1984 ) a. n. kolmogorov , _ dokl .
nauk sssr _ , 30 , 301 ( 1941 ) ,
k41 g. kowal , a. lazarian and a. beresnyak , _ astrophysical journal _ , 658 , 423 ( 2007 ) a. g. kritsuk , m. l. norman , p. padoan and r. wagner , _ astrophysical journal _ , 665 , 416 , ( 2007 ) a. lazarian , _ international journal of modern physics d _ , 15 , 1099 ( 2006 ) a. lazarian , astro - ph/0608046 ( 2006 ) , in press m. j. lighthill , _ iaus _ , 2 , 121 ( 1955 ) y. lithwick and p. goldreich , _ astrophysical journal _ , 562 , 279 ( 2001 ) w. c. mller and d. biskamp , _ phys .
, 84 , 3 ( 2000 ) p. padoan , a. nordlund , a. g. kritsuk , m. l. norman and p. s. li , _ astrophysical journal _ , ( 2007 ) , in press z. she and e. lvque , _ phys .
rev . letters _ , 72 , 3 ( 1994 ) c. f. von weizscker , _ astrophysical journal _ , 114 , 165 ( 1951 ) | we study scaling relations of compressible isothermal strongly magnetized turbulence using numerical simulations with resolution 512@xmath0 .
we find a good correspondence of our results with the fleck ( 1996 ) model of compressible hydrodynamic turbulence .
in particular , we find that the density - weighted velocity , i.e. @xmath1 , proposed in kritsuk et al .
( 2007 ) obeys the kolmogorov scaling , i.e. @xmath2 for the high mach number turbulence .
similarly , we find that the exponents of the third order structure functions for @xmath3 stay equal to unity for all mach numbers studied .
the scaling of higher order correlations obeys the she - lvque ( 1994 ) scalings corresponding to the two - dimensional dissipative structures , and this result does not change with the mach number either .
in contrast to velocity @xmath4 which exhibits different scaling parallel and perpendicular to the local magnetic field , the scaling of @xmath3 is similar in both directions .
in addition , we find that the peaks of density create a hierarchy in which both physical and column densities decrease with the scale in accordance to the fleck ( 1996 ) predictions .
this hierarchy can be related ubiquitous small ionized and neutral structures ( sins ) in the interstellar gas .
we believe that studies of statistics of the column density peaks can provide both consistency check for the turbulence velocity studies and insight into supersonic turbulence , when the velocity information is not available .
address = department of astronomy , university of wisconsin , madison , wi , [email protected] address = department of astronomy , university of wisconsin , madison , wi , [email protected] |
in an elegant paper @xcite takayanagi has recently proposed an extension of the ads / cft conjecture @xcite to deal with the situation , when the cft is defined on a space with a boundary .
these so called boundary cft s , or bcft s , naturally appear in various applications to condensed matter phenomena .
the main idea behind the takayanagi s construction is to appropriately extend the boundary of the cft inside the bulk of the @xmath6 space in order to cut out a physically relevant piece .
the ads / bcft conjecture finds its natural roots in the holographic derivation of entanglement entropy @xcite and in the randall - sundrum model @xcite .
indeed , the extension of the cft s boundary inside the bulk of the @xmath6-space may be regarded as a modification of a `` thin '' randall - sundrum brane , intersecting the @xmath6 boundary .
for this brane to be a dynamical object one must impose neumann boundary conditions on the metric ; then , the discontinuity in the bulk extrinsic curvature across the `` defect '' will be compensated by the brane s tension .
thus , neumann boundary conditions constitute the dynamical principle defining the additional boundaries introduced by takayanagi . in the following
we will often refer to these boundaries as to randall - sundrum ( rs ) branes . from the point of view of the ads / cft applications
it is interesting to see how the addition of the boundaries affects physical behavior . in particular
, it would be interesting to construct a solution describing a physical system at finite temperature and charge density .
the most common playground is provided by the charged @xmath7 black hole since this background has been already shown to encode many interesting condensed - matter - like phenomena such as superconductivity / superfluidity @xcite and strange metallic behavior @xcite . here , we shall analyze the charged @xmath7 black hole in the presence of boundaries .
a model of gauge fields in the @xmath7 background with boundary rs branes has been already considered by fujita , kaminski and karch in @xcite , who showed that the additional boundary conditions impose relevant constraints on the gauge field parameters and that , as a consequence of those constraints , there is the possibility of quantum - hall - like behavior of the conductivity in the dual field theory .
however , their approach @xcite does not account for the backreaction of the gauge fields on the geometry , restricting to the geometry of the empty @xmath6 space
. a natural generalization of their work would be to extend their analysis to the case of a charged black hole . in this paper
we make a first step in this direction .
the physical system analyzed in this paper is based on the model proposed by fujita _
indeed , we start from the same lagrangian for an einstein - maxwell system describing the gravity dual of a field theory on a half - plane .
we find that , for the simple plane - symmetric black hole ansatz , only tensionless rs branes are allowed and that the background solution does not allow to model the situation with external electric fields as in @xcite .
furthermore , as a result of the neumann boundary conditions for the gauge fields , we find that the charge density @xmath8 in the dual field theory must be supported by an external magnetic field @xmath9 . the ratio @xmath10 , which is equal to the hall conductivity , is a constant inversely proportional to the sum of the coefficients of the topological terms present in the gravity action : namely , a @xmath11-term in the bulk action and a chern - simons term in the boundary action on the rs branes .
in addition , we get zero longitudinal conductivity .
all these behaviors are expected for a quantum hall system tuned to a quantized value of the conductivity . in the classical hall effect
@xmath8 and @xmath9 are independent quantities , _
e.g. _ their ratio is a function of the density of conductance electrons @xmath8 . in the quantum hall effect ( qhe )
the transverse conductivity @xmath12 , exhibits plateaus independent of either @xmath8 or @xmath9 .
the plateaus are generally attributed to disorder @xcite , which is responsible for localization ( _ i.e. _ existence of localized electron states ) .
the localized states fill the gaps between the landau levels , but do not participate in the hall conductivity . in other words , even if one varies the density of conductance electrons , so that the fermi energy shifts in the gap between the landau levels , the conductivity does not change . for the same reason the longitudinal components of the conductivity vanish .
an important aspect of the qhe is the quantization of the conductivity plateaus in integer ( integer qhe ) or fractional ( fractional qhe ) numbers of flux quanta @xmath13 .
the mechanisms of the integer and fractional quantization are , however , different : while the fractional quantization is due to the interaction between electrons @xcite , the integer qhe ( iqhe ) exists even for a non - interacting electron gas with disorder and is a consequence of gauge invariance @xcite . in the widely accepted composite fermion model the fractional qhe ( fqhe )
may be regarded as an iqhe for composite quasiparticles of fractional charge @xcite .
it is by now well understood that the fqhe is effectively described by chern - simons gauge theories @xcite .
furthermore , as explicitly seen in computations of the gauge effective action of planar electrodynamics @xcite , the hall relation @xmath14 can be derived from the chern - simons lagrangian with @xmath12 being a constant inversely proportional to the chern - simons level @xmath15 .
if , for some reason , @xmath15 is constrained to take integer values ( for details , see @xcite ) , the conductivity would be also quantized . the general philosophy of the ads / bcft correspondence leads naturally to a generalization of the chern - simons description of the qhe . indeed , the neumann boundary condition for the gauge fields , derived by fujita _ et .
_ , turns out to be a covariant `` holographic '' form of the hall relation .
furthermore , in the ads / bcft formulation , the chern - simons theory lives on the rs branes cutting the bulk of the @xmath6 space and the holographic hall relation potentially leads to a more interesting dependence of @xmath12 on various topological coefficients @xcite .
an interesting question is to ascertain whether and how models using the ads / bcft correspondence may be made consistent with the physics underlying the integer or the fractional qhe .
since the classical gravity limits of the holographic correspondence enable to describe strongly coupled gauge theories one may naively expect that the ads / bcft correspondence may provide us with an appropriate description of the fqhe .
this naive expectation is supported by the study in @xcite , which revealed a non - fermi liquid behavior of the charged @xmath7 black hole .
in addition , the gravity solution describes a `` clean '' system , where disorder has still to be introduced . in this paper
we provide a further analysis of the physical consequences stemming from the @xmath6 black hole solution with the aim of elucidating its relationship with realistic quantum hall systems . one way to test
how fractional is the qhe is to look at the behavior of the heat conductivities . for a non - interacting electron gas
the heat conductivity is related to the electric conductivity through the wiedemann - franz law . for interacting electrons , _
i.e. _ for the fqhe , the wiedemann - franz law is violated @xcite and the ratio of the transverse heat and electric conductivities becomes a non - trivial function of the filling fraction , which may yield zero , or negative , heat conductivities .
our analysis of the low temperature behavior of the transverse thermal conductivities shows that the @xmath7 black hole is perfectly consistent with the wiedemann - franz law .
in particular , the @xmath16 coefficient of the ratio , the lorenz number , is precisely @xmath17 ; a well known result for non - interacting electrons .
another characteristic feature of quantum hall systems is the emergence of current - carrying edge states , extending along the perimeter of the sample @xcite .
their existence is due to the fact that , at the edges , there is a confining potential , preventing the electrons from leaving the sample and locally raising the energy of the landau levels . as a result , while the system remains gapped ( incompressible ) far from the edges , there is no gap at the edges .
the edge states may then be regarded as the massless modes arising at the intersection of the fermi energy with the filled landau levels .
below , we show that the edge currents do not appear for a black hole solution with tensionless rs branes .
in addition , we argue that there is an edge current whenever the tension of the rs branes is different from zero .
this observation is consistent with the fact that a non - vanishing tension should represent a system with a geometry - induced gap in the `` bulk '' of the sample .
our analysis hints to the fact that a black hole solution with tensionless rs branes may only describe the situation where the difference between the fermi energy and the nearest landau level is smaller than the temperature and , thus , it is expected to be rather unstable against thermal fluctuations .
although the behavior found for the conductivities and the absence of a gap seem to be in contrast with our expectations for a quantum hall system , we believe that this only hints to the instability of the gravity solution with tensionless rs branes . we feel that a more appropriate solution would be the one with non - vanishing tension of the rs branes ; unfortunately , an explicit form of such a solution is still lacking .
we expect that this solution will modify the result for conductivities and may lead to a violation of the wiedemann - franz law ; a preliminary analysis indicates , indeed , that rs branes with tension enable to describe situations where the electric conductivity is quantized and the tension is related to the fermi energy , rather than being an independent parameter .
alternatively , it is interesting to understand how disorder may be introduced in a ads / bcft model describing the iqhe .
there already exists an extensive literature on the application of holographic correspondence to quantum hall physics .
here we will take advantage of the computation of the transport coefficients carried by hartnoll and kovtun for the dyonic @xmath7 black hole @xcite and further elaborated in @xcite .
we also stress that the ads / bcft model analyzed in this paper is reminiscent of some recent top - down stringy constructions @xcite . for more details
we refer the reader to these papers .
the presentation of our results is organized as follows . in section [ sec : setup ]
we review the ads / bcft construction in the particular setup studied in @xcite . in section [ sec : model ]
we investigate the black hole solution constrained by the additional boundary conditions required by the ads / bcft construction .
section [ sec : conductivities ] is devoted to the discussion of the electric and thermal conductivities .
the computation of the edge currents is the main topic of section [ sec : edgecurrents ] . finally , in section
[ sec : conclusions ] , we provide a critical summary of our results .
we aim at studying a gravity dual of a @xmath0-dimensional system with finite temperature and charge density living on a space @xmath18 with a boundary which , for the sake of simplicity , will be taken in the following sections as a half plane or a strip . since systems at finite temperature and charge density may be dually described by charged black holes , we can describe the bulk using the abelian einstein - maxwell action .
namely , one may start with [ bulkaction ] s_n = _ n^4 x ( r-2 ) - _
n^4 xf_f^+_n ff . following @xcite
we denote with @xmath19 the bulk space , with @xmath20 the determinant of the bulk metric @xmath21 and with @xmath22 the cosmological constant . in the following we set the gravitational constant @xmath23 and the @xmath24 coupling @xmath25 ; the only role of @xmath26
will be then to define the probe limit @xmath27 , in which the backreaction of the matter fields on the gravity solution can be ignored .
since the last term in the bulk action is topological , it does not affect the equations of motion ; however , it is relevant for the analysis of the boundary conditions .
the bulk equations of motion may then be rewritten as [ einstein ] r _ - g _ - g _
= f_f_^ - g_f_f^ , [ maxwell ] ^f _ = 0 . the dual theory lives on a @xmath0 dimensional space @xmath18 with a boundary which , following @xcite , we denote with @xmath28 . for a proper definition of the holographic model it is natural to require that this @xmath29 dimensional boundary is extended to the bulk to make a @xmath0 dimensional hypersurface @xmath30 @xcite , so that @xmath31 and @xmath32 .
according to takayanagi , for this hypersurface @xmath30 to be dynamical , one needs to introduce the following boundary action on @xmath30 [ qaction ] s_q = _ q ^3 x(k - ) + s_q[matter ] , where @xmath33 is the determinant of the induced metric @xmath34 on the boundary @xmath30 and @xmath35 is the extrinsic curvature on @xmath30 .
we denote by @xmath36 the corresponding cosmological constant , which may also be regarded as the tension of the brane @xmath30 . in the following
we shall use @xmath37 only to account for the temperature of the dual field theory . as pointed out in the introduction ,
the boundary @xmath30 plays a role analogous to the cutoff branes in the randall - sundrum model ; in the following , we shall refer to the additional boundaries required by the ads / bcft construction as rs branes . to have a well - defined variational principle on a space with boundaries one needs to specify the boundary conditions .
variation of the bulk ( [ bulkaction ] ) plus boundary ( [ qaction ] ) action in response to a variation of the metric yields the following boundary term [ gvariation ] ( s_n+s_q)[g _ ] = _ q ^3 x ( k_ab-(k-)h_ab)h^ab - _ q ^3 x t_abh^ab , where @xmath38 is the boundary stress - energy tensor : t_ab = -s_q[matter ] .
for the variation to vanish one could impose a dirichlet boundary condition on the metric @xmath39 .
however , if @xmath30 has to be determined dynamically , one should employ neumann boundary conditions .
namely , one should require that [ neumann - metric ] k_ab-(k-)h_ab= t_ab . as a result the shape of @xmath30 and
the induced metric arise as a solution of dynamical equation ( [ neumann - metric ] ) , rather than as an explicit boundary condition . to study quantum hall systems we chose the boundary matter action as in @xcite : s_q[matter]=
q af . notice that , for this choice of the action , @xmath40 and , as in equation ( [ gvariation ] ) , the variation of the gauge fields induces a boundary term [ avariation ] ( s_n+s_q)[a _ ] = 2_q ( 12f + ( + ) f)a ; the neumann boundary conditions then imply that [ neumann - gauge0 ] c_1f + c_2f|_q=0 , where , for simplicity , we have set [ topologicalcoeffs ] c_1= 12 , c_2 = + . in the next section
we shall analyze the solutions to the equations of motion ( [ einstein ] ) and ( [ maxwell ] ) constrained by the neumann boundary conditions ( [ neumann - metric ] ) and ( [ neumann - gauge0 ] ) . here
, we would like to complete the analysis of the gravity action needed by the ads / bcft construction .
namely , the variation of the action gets contributions also from the boundary @xmath18 where , in accordance with the general lore of the ads / cft correspondence , one must impose dirichlet boundary conditions . following the standard procedure @xcite we introduce the following counter terms to render the action finite , [ counterterms ] s_m = _
m^3x(k- ) . here
@xmath41 is the determinant of the induced metric on @xmath18 , which is fixed by the dirichlet boundary conditions to be the minkowski metric @xmath42 of the dual gauge theory .
finally , one needs to introduce additional boundary terms at the boundary @xmath28 : [ paction ] s_p = _
p ^2x ( - ) - _
p^2x a_t a_x .
the first term was already introduced in @xcite .
it accounts for the singularity of the boundary @xmath28 .
indeed in ( [ gvariation ] ) we omitted a term , which is a total derivative on @xmath30 , and thus lives only on the boundary @xmath28 .
a similar term comes from the variation of the counter terms ( [ counterterms ] ) on @xmath18 . in equation ( [ paction ] )
@xmath43 ( @xmath44 ) is the angle between the two outward pointing unit normal vectors @xmath45 and @xmath46 at the boundary @xmath28 ( _ e.g. _ @xcite ) .
when @xmath47 the two normal vectors coincide and the boundary becomes smooth .
the second term in ( [ paction ] ) is required by gauge invariance . to ensure that the chern - simons action on @xmath30 is gauge invariant we choose the gauge condition as @xmath48 . with this choice
the boundary term protects the residual gauge transformations .
notice that , in ( [ paction ] ) , @xmath49 and @xmath50 are generic ( _ i.e. _ not yet specified ) coordinates parameterizing @xmath28 .
collecting all the pieces together the total action proposed to describe the ads / bcft holographic model of a quantum hall system is given by [ totalaction ] s = s_n + s_q + s_m+ s_p .
in this section we describe a solution to the equations ( [ einstein ] ) and ( [ maxwell ] ) , satisfying the additional boundary conditions imposed by the ads / bcft construction . for simplicity , we take the @xmath0-dimensional manifold @xmath18 to be half of an infinite plane ; our subsequent analysis may , however , be readily generalized to include the case of an infinite strip .
we make the following simple black hole ansatz for the metric [ ansatz - metric ] s^2 = ( - f(z)t^2 + x^2 + y^2 + ) , and for the gauge fields [ ansatz - gauge ] a_t = - z , a_x = - by + cz , a_y , z = 0 . here
@xmath49 , @xmath51 and @xmath50 are coordinates on @xmath18 such that @xmath52 ( see figure [ fig : geometry ] ) . .
@xmath19 denotes the part of the bulk space bounded by @xmath18 , horizon of the black hole at @xmath53 and rs brane @xmath30 .
( b ) dyonic black hole solution ( @xmath54).,title="fig : " ] ( a ) a general solution to the equations ( [ einstein])-([maxwell ] ) with the ansatz ( [ ansatz - metric])-([ansatz - gauge ] ) is given by [ metric - soln ] c=0 , f(z ) = 1 + ( ^2+b^2)z^4 - cz^3 , where @xmath55 is an integration constant related to the mass of the black hole . the other parameters of the solution have the following interpretation in terms of the dual theory : @xmath56 is the chemical potential , @xmath8 is the charge density and @xmath9 is the external magnetic field , normal to the plane @xmath18 .
the largest of the positive roots @xmath53 of the function @xmath57 the so called blackening factor determines the horizon of the black hole .
it is sometimes convenient to switch from a description in terms of the mass of the black hole to a description in terms of @xmath53 . upon defining q^2=(^2+b^2)z_h^4
, @xmath57 takes the form f(z)=1-(1+q^2)()^3 + q^2()^4 , where both @xmath30 and @xmath53 are now functions of the mass and the charge of the black hole . in terms of @xmath30 and @xmath53 the black hole temperature reads t=. to have a finite energy density in the dual theory one must require that @xmath58 ; as a result , @xmath8 and @xmath56 are related as : = z_h .
the thermodynamics of this black hole was already discussed in @xcite and @xcite .
one can describe the dual system by four thermodynamic variables : volume , temperature @xmath37 ( or entropy @xmath59 ) , charge density @xmath8 and magnetic field @xmath9 .
the entropy density is given by = .
from @xcite the energy density @xmath60 in the dual system may be written as a function of @xmath8 , @xmath9 and the entropy density @xmath61 as [ energydensity ] ( s,)= ( ) ^3/2(1++ ) . in the ads
/ bcft correspondence the parameters @xmath8 , @xmath61 and @xmath9 will not be independent due to the boundary conditions at the rs brane @xmath30 .
let us start with the boundary condition ( [ neumann - metric ] ) .
we assume that @xmath30 is parameterized by the equation @xmath62 and , to find the extrinsic curvature , we define an outward unit vector normal to @xmath30 [ normalq ] ( n^t , n^x , n^y , n^z)=(0,0,,- ) ; then , the pullback of equation ( [ neumann - metric ] ) to @xmath19 reads : [ neumann - metric2 ] k_-(k-)h_=0 , where the extrinsic curvature is given by k = h^k _ , k_=h_^^_n _ , and @xmath63 is the projector onto @xmath30 : ^= _ ^- n_n^. the simplest path to solve equation ( [ neumann - metric2 ] ) is to set @xmath64 ; this enables to recover the result of @xcite .
the metric is pure @xmath7 and @xmath30 is given by the solution to [ qtension ] y(z)= .
setting @xmath65 one gets [ qprofilet ] y= - z , _
i.e. _ @xmath30 can be described as a plane intersecting the boundary @xmath18 at an angle @xmath66 , as shown in figure [ fig : geometry](a ) .
more generally , the only possibility to get an rs brane with non - zero tension from ( [ ansatz - metric ] ) is to set @xmath67 ; this means that no black hole with finite tension of @xmath30 can be constructed from our simple ansatz for the black hole . for @xmath68 , however , a black hole ( [ metric - soln ] ) solution is allowed . in the latter case
the rs brane @xmath30 is just perpendicular to the boundary @xmath18 ( figure [ fig : geometry](b ) ) : [ qzerotension ] q : y = const , = .
the existence of non - trivial gravity solutions with non - zero tension of the rs branes was recently addressed in @xcite .
it was shown that such solutions do exist ; it is very interesting , indeed , to ascertain whether such solutions may be adapted to describe charged black holes .
let us now turn to the discussion of the restrictions imposed by the boundary condition ( [ neumann - gauge0 ] ) for the gauge fields .
although we already know that , in our simple setup , the background gauge fields are only allowed for zero tension , it will be instructive to keep the discussion more general using ( [ qprofilet ] ) and setting @xmath54 only at the end of our derivation . in component notation , the neumann boundary condition ( [ neumann - gauge0 ] )
may be written as ._f^ + c_2f_|_q=0 . to restrict the bulk equation on @xmath30 one
may project it on a 3-form orthogonal to the normal vector @xmath69 so that [ neumann - gauge ] - c_1n_f^ + n_^ f_=0 . equation ( [ neumann - gauge ] ) then leads to [ neumann - gauge2 ] ccc 0 & = & c_1- c_2(c+b ) , + 0 & = & c_1(b - c ) + c_2 . for @xmath70 and tensionless @xmath30 , @xmath54 ,
one gets [ rhobratio ] = - .
we see that the density and the magnetic field are no longer two independent parameters .
since their ratio is just the hall conductivity this is very reminiscent of the quantum hall effect , where this ratio is independent of both @xmath8 and @xmath9 and is inversely proportional to the topological coefficients .
notice however that the @xmath66 dependence of equation ( [ neumann - gauge2 ] ) shows that , in the limit of maximal tension of the rs brane , ( @xmath71 ) , the ratio @xmath10 becomes proportional to @xmath72 .
these results are particular cases of a more general one obtained by fujita _
_ but unlike in @xcite , here we consider a full backreacted black hole solution , which further reduces the freedom in the choice of the parameters . in particular , the black hole ansatz ( [ ansatz - metric ] ) does not allow to include additional background fields such as the electric field @xmath73 in @xcite .
thus , it would be interesting to look for generalizations of the simple gravity solution analyzed in this paper
. equations ( [ neumann - gauge2 ] ) are invariant under the shift @xmath74 , which makes explicit the invariance of ( [ neumann - gauge ] ) under the change of sign of @xmath69 . as a result
, the same solution may live on an infinite strip of finite width @xmath75 .
in fact , since the background fields ( the charge density and the magnetic field ) do not depend on the spatial coordinates , _
i.e. _ they do not fall off at infinity , one must worry about boundary conditions at the boundary @xmath76 even when dealing with a model living on the half - infinite plane .
for an infinite plane one can always assume periodicity .
thus , @xmath18 is always topologically equivalent to a compact space ; as a result it is more appropriate to picture the spatial part of @xmath18 as being a closed strip rather than an infinite plane .
to further test the ads / bcft construction against the behavior of a realistic quantum hall system we compute , in this section , the transport properties emerging from the ads / bcft construction : for this purpose , one has to analyze the linear response of the system to a small perturbation of the external sources .
in particular , one has to compute the electric and heat conductivities as a response to a small applied electric field or to a temperature gradient .
one can not consider the two kinds of conductivities separately , since the dual fluctuations of the gravity fields couple them to each other .
the transport properties of the dyonic black hole were originally studied in @xcite . in our ads
/ bcft model the action used in @xcite is modified by various boundary terms and one can , in principle , expect that the boundary terms may lead to a modification of previously known holographic results for conductivities . in the following
we shall show that , for a gravity solution with tensionless rs branes , the kubo formulae for electrical and thermal conductivities do not get contributions from these additional boundary terms .
the derivation goes as follows .
the transport coefficients are encoded in the 2-point correlation functions .
the holographic prescription @xcite relates the 2-point functions to the second derivatives of the total renormalized on - shell action ( [ totalaction ] ) with respect to the sources the @xmath77 asymptotic values of the gravity fields . since , in the linear response theory , only linearized equations are relevant one should consider only the quadratic part of the action . to compute the charge transport
one has , then , to turn on small electric fields @xmath78 and @xmath79 on the boundary and derive the linearized equations for the corresponding fluctuations of the bulk gauge fields ; at finite temperature , the electric currents will be accompanied by the heat currents and one needs to turn on the bulk fluctuations of the metric as well : [ variations ] a_x , y = a_x , y(z ) e^-it , g_tx , ty = g_tx , ty(z ) e^-it .
the linearization of the system of equations ( [ einstein])-([maxwell ] ) with respect to the fluctuations over the black hole solution may be reduced to the following pair of complex equations written in terms of complex fluctuations @xmath80 and @xmath81 : [ complex1 ] a
+ a + a - g - 2 g + g = 0 , [ complex2 ] ( g+g - 4l^2a ) - 4z^2b g + 4l^2b f a = 0 , where @xmath82 is the blackening factor given in equation ( [ metric - soln ] ) .
the general solutions for the metric and the gauge field fluctuations in the background ( [ ansatz - metric])-([metric - soln ] ) take the following asymptotic form at the boundary @xmath77 .
namely , one has that [ gboundary ] g_tj= + g_tj^(1 ) z+ o(z^2 ) , [ aboundary ] a_j= a_j^(0 ) + a_j^(1 ) z + o(z^2 ) .
holography then prescribes to identify the coefficients of the leading terms with the sources in the dual field theory , that is with the electric field and the temperature gradient , a_j^(0)= - , g_tj^(0)= - , while the subleading coefficients should be proportional to the expectation values of the dual operators , _
i.e. _ to the charge and to the momentum density currents @xmath83 and @xmath84 , respectively .
notice that the second derivative of the metric fluctuation is absent from the equations ( [ complex1 ] ) , ( [ complex2 ] ) .
therefore , @xmath85 may be written in terms of @xmath86 and the sources as : [ density - current ] g_tj^(1)=-a_j^(1 ) + g_tj^(0 ) + l^2 a_j^(0 ) .
computing the on - shell quadratic action yields @xmath87 = \\
= \int \d x\,\d y\,\d t~\left[-\frac{b}{2\omega l^2}\ , g_{tj}^{(0)}a_j^{(1)}- \frac{\left(1+q^2\right)\omega - 2 z_h^3b\rho}{4\omega z_h^3l^4}\,g_{tj}^{(0)}g_{tj}^{(0)}+\frac{1}{2}\,a_j^{(0)}a_j^{(1)}\right].\end{gathered}\ ] ] the boundary terms in the action ( [ totalaction ] ) do not give any finite contribution to equation ( [ quadaction ] ) for the gravity solution with tensionless rs branes .
as a result the quadratic action is the same as the one considered by hartnoll and kovtun @xcite , who computed the retarded green s functions needed for the evaluation of @xmath88 and @xmath84 and derived the electric conductivity .
the thermal conductivity was later derived by hartnoll _
_ in @xcite . in the following we simply adapt their results to our ads / bcft model .
the electric conductivity is found from the retarded green s function @xmath89 of 2 electric currents , through the kubo formula _
jk = -_0 from @xcite one then finds _
xx=_yy=0,_xy=-_yx= = - .
that is , the hall conductivity is inversely proportional to the sum of the coefficients in front of the topological terms appearing in the ads / bcft gravity action .
the computation of the thermal conductivity must be done with a greater care . as we learn from @xcite , in the presence of a magnetic field ,
the kubo formula must be modified in order to subtract the `` magnetization currents '' .
the modified kubo formula reads @xmath90 where @xmath91 and @xmath92 are retarded green s functions of the electric and heat @xmath93 currents .
the magnetization density @xmath94 and energy magnetization density @xmath95 were computed in @xcite as : = -bz_h , ^e = . a low temperature expansion of the result obtained in @xcite for @xmath96 and @xmath97 yields the thermoelectric conductivity _ xx=_yy=0,_xy=-_yx= = + o(t ) , where we used @xmath98 .
for the longitudinal thermal conductivity one gets _
xx=_yy= = t+o(t^2 ) .
as expected , the conductivity is linear with temperature .
the transverse conductivity behaves as _ xy=-_yx= = _
ht+o(t^2 ) .
the last equation shows that the wiedemann - franz law is satisfied for the transverse conductivities .
in fact , [ wflaw ] = t , = ( ) ^2 . remarkably , at low temperature , the @xmath7 black hole yields for the lorenz number @xmath99 the well known result valid for non - interacting electrons
. it would be interesting to see whether this has anything to do with the @xmath100 near - horizon structure of the extremal @xmath7 black hole ; indeed , the latter feature is believed to underly the non - fermi - liquid - like behavior found in @xcite .
a quantum hall system is characterized by the presence of edge currents @xcite . in this section
we determine the conditions under which the existence of edge currents may be accounted for by the ads / bcft construction analyzed in this paper .
our analysis shows that no edge current can emerge from geometries with tensionless rs branes since they may be associated only to a system with no gap in the bulk .
we then argue that , for geometries with tensionful rs branes , one may induce a gap in the bulk of the sample by relating the brane tension to the fermi energy and to the scale of the closest approach of the brane @xmath30 to the horizon ; in this setting edge currents may emerge . to see how edge currents may be accounted for in the ads / bcft construction developed in the previous sections we look at the gauge part of the action on the rs brane @xmath30 : s_k = _ q af - _
p ^2 x a_x a_t .
the @xmath49 component of the current density can be found as the variation of the on - shell action with respect to the asymptotic value of the gauge field component @xmath101 , j^x(x^i)= . on shell ,
the action @xmath102 becomes s_k = _ q ^3 x a_x_y a_t - _ p ^2x
a_t . choosing the background geometry ( [ qprofilet ] ) with non - zero tension ( @xmath103 ) of the rs brane yields for the current the following expression : [ edgecurrent ] j^x ( x^i)= ._y a_t|_q(y+z_h ) - .
a_t|_p(y ) = - ( ( y ) - ) . here
@xmath104 is the step function with the support @xmath105 $ ] .
its appearance is not surprising since one may expect that the rs brane will be cut by the horizon of the black hole as it happens for the geometry with zero tension of the rs brane analyzed in the previous sections . in the above derivation
we have also used the relation between the charge density and the chemical potential . albeit in the opposite direction . ]
let us now analyze the dependence of the current on @xmath66 or , equivalently , on the tension @xmath36 .
although we did not find a charged black hole solution with non - zero tension , we can address its expected features in the probe limit @xmath27 in ( [ bulkaction ] ) , in which the backreaction of the gauge field on the metric can be neglected .
when @xmath106 , the rs brane may be regarded as an ir cutoff , just as the original rs branes ( figure [ fig : gap](c ) ) . in this limit @xmath107 and the last term in ( [ edgecurrent ] ) vanishes ; it is safe , then , to associate the cutoff scale with a geometry - induced gap in the dual hall system . in a realistic quantum hall system
the gap is the difference between the fermi energy and the energy of the closest upper landau level _ i.e. _ just the amount of energy needed to excite a charge current . due to the potential barrier confining the electrons to the interior of the sample the landau levels shift in the vicinity of the edge and the fermi energy level
is crossed by the lower landau levels .
`` edge modes '' at the intersection carry then the edge current along the perimeter of the system .
the edge current is given as _
h = e , which is consistent with our previous results if one sets _
indeed this is the solution to equations ( [ neumann - gauge2 ] ) with @xmath71 .
our conventions for @xmath108 and @xmath109 yield the proper quantization of the filling fraction . )
( a ) , finite negative , @xmath110 ( b ) and minimal tension , @xmath71 ( c ) .
( b ) and ( c ) can be interpreted as a position dependent and constant gap in the dual theory respectively .
more realistic configurations are expected to have shape ( d).,title="fig : " ] ( a ) ) ( a ) , finite negative , @xmath110 ( b ) and minimal tension , @xmath71 ( c ) .
( b ) and ( c ) can be interpreted as a position dependent and constant gap in the dual theory respectively .
more realistic configurations are expected to have shape ( d).,title="fig : " ] ( c ) when @xmath54 , _
i.e. _ for the gravity solution with zero tension of the rs brane analyzed in this paper ( figure [ fig : gap](a ) ) , there is no geometric gap in the bulk " of the conductor , but rather an infinite gap at the edges ; of course , the edge current is zero in this situation . when @xmath110 , the gap is position dependent and the total edge current diffuses over a characteristic region @xmath111 ( figure [ fig : gap](b ) ) set by the cutoff scale @xmath53 .
the absence of edge currents when the rs brane is tensionless indicates that the ads / bcft black hole solution analyzed in this paper does not give a full description of a realistic quantum hall system since there is no scale in the geometric description that can be associated with the energy gap needed to describe the bulk of a quantum hall system .
however , even when the rs brane is tensionless , the hall conductivity still exhibits a plateau .
the `` unphysical '' behavior the ads / bcft model with tensionless rs brane is likely to indicate its instability . indeed a more general profile of the rs brane should have the shape illustrated by figure [ fig : gap](d )
. it would be interesting to find such tensionful solutions and analyze their stability . for solutions of the type depicted on figure [ fig : gap](d )
we expect that the tension is not a free parameter , but is dynamically related to the fermi energy and to the scale @xmath112 ( position of the closest approach of the brane @xmath30 to the horizon ) .
the variation of the chemical potential will then correspond to the variation of @xmath112 , with @xmath113 as @xmath56 increases . at some value of the chemical potential
one should have @xmath114 and the gap will close . in this respect
the solution in figure [ fig : gap](a ) looks like a metastable branch of the profile in figure [ fig : gap](d ) for @xmath115 . since the `` bulk '' of the sample in the configuration depicted in figure [ fig : gap](d ) is gapped just as it happens for the background geometry ( [ qprofilet ] ) with non - zero tension ( @xmath103 ) of the rs brane , we expect to observe currents at the edges
. it would be interesting to further understand the edge effects in such a configuration .
naively , one should expect that the conductivity is a function of @xmath66 near the edges , varying between the @xmath71 value ( @xmath116 ) and @xmath54 value ( @xmath117 )
. it would be interesting to ascertain whether this feature has any relationship with quantum hall samples , where stripes of different filling fraction may appear close to the edges @xcite .
transport at the edges may exhibit other very interesting phenomena .
in particular , the existence of `` upstream '' modes was conjectured in @xcite ; these modes should carry charge in the direction opposite to the classical skipping orbits typical of electrons in a magnetic field .
although no charge - carrying upstream modes has been detected experimentally , the heat transport in the upstream direction has been already observed @xcite .
this effect leads to a violation of the wiedemann - franz law ( [ wflaw ] ) .
the violation is a result of the electron interaction : in a state with a fractional value of the filling fraction @xmath118 ( fqhe ) the modified wiedemann - franz law can be written as [ wflaw2 ] = _ qt , where @xmath119 is a function of @xmath118 , which equals unity on integer values . for fractional values of @xmath118 the function @xmath120 and may vanish or become negative @xcite . in the previous section we obtained @xmath121 for a background geometry with zero tension of the rs brane ; this points to the fact that this geometry may enable to describe a non - interacting system ( iqhe ) once disorder is included in the ads / bcft construction .
nevertheless , since we expect that solutions with non - vanishing tension of the rs brane will modify the computations of the conductivities ( in the most naive form we can expect that the ratio in ( [ wflaw ] ) will be @xmath66-dependent , ) our ads / bcft approach may yield other solutions violating the wiedmann - franz law .
we analyzed an ads / bcft model of a condensed matter system at finite temperature and charge density living on a @xmath0-dimensional space with a boundary .
it turns out that a straightforward generalization of a known ads / cft solution , such as the plane - symmetric charged @xmath7 black hole , only allows for tensionless rs branes in the ads / bcft construction and requires that the static uniform charge density is supported by a magnetic field .
specifically , we found that @xmath10 is a constant proportional to a ratio of the coefficients appearing in the gravity action .
such a property indicates that a pertinent generalization of the @xmath7 black hole may describe a quantum hall system at a plateau of the transverse conductivity .
we provided further tests of the ads / bcft holographic model to see how accurately it can account for the physical behaviors expected in a quantum hall system . the ads / bcft construction yields that the hall conductivity is inversely proportional to a sum of the coefficients of the topological terms appearing in the gravity lagrangian .
namely , we got that [ sigmahall ] _
h==- , c_2= + . here
@xmath15 is the level of the chern - simons term on the boundary and @xmath11 is the bulk @xmath66-angle . in the qhe
the conductivity is related to the number of filled landau levels ( filling fraction ) by = _
-2 , where @xmath122 is the magnetic flux quantum .
thus , the holographic description seems to provide results similar to the chern - simons description of the qhe . as in the chern - simons description @xcite , in the ads / bcft model
the filling is inversely proportional to the level of the chern - simons theory .
indeed equation ( [ neumann - gauge0 ] ) comes from the variation of the chern - simons lagrangian , which lives on the rs brane @xmath30 .
it can be seen as an extension of the covariant form of the hall relation @xmath14 .
however , in the holographic ads / bcft model , the conductivity receives a contribution also from the bulk @xmath66-term , resulting in a renormalization of the chern - simons coefficient . in the qhe ,
the filling fraction is quantized .
it may take either integer ( iqhe ) or special fractional ( fqhe ) values . from the point of view of the chern - simons description , quantization of the filling fraction
signifies quantization of the chern - simons coefficient . unlike the non - abelian case
, there is no _ a priori _ quantization of the abelian chern - simons coefficient . nevertheless , in some cases , like compact manifolds , or finite temperature ,
it can be argued to take integer or rational values , _
e.g. _ see @xcite . as pointed out in section [ sec : model ]
the boundary conditions effectively make the manifold @xmath18 topologically equivalent to a compact one .
thus , the ads / bcft holographic model rather easily yields the integer quantization of the chern - simons level @xmath15 which , in turn , leads to fractional values of the filling fraction and conductivity in the appropriate units .
the computation of the electrical and thermal conductivities in the ads / bcft model with tensionless rs branes goes along the same lines as in @xcite .
equipped with the result of @xcite we observed here that , at low temperatures , the coefficient of the leading @xmath16 term yields precisely one quantum of the transverse heat conductance per quantum of the transverse electric conductance .
that is the conductances satisfy the wiedemann - franz law for a non - interacting electron gas ( [ wflaw ] )
. quantum hall systems also exhibit currents running along the edges of the sample .
ln section [ sec : edgecurrents ] we derived an expression for the edge current , which is consistent with conventional wisdom : namely , the current is proportional to the chemical potential and to the conductivity , while its existence is associated to the existence of a gap in the `` bulk '' of the sample .
since the charged black hole solution with a tensionless rs brane analyzed in this paper can not describe a gapped system , we find no edge current .
we conclude that such a black hole does not describe completely a realistic quantum hall system ; we conjecture indeed that such black hole is much likely to be an unstable solution of the ads / bcft holographic model .
our analysis points to the fact that , from a charged black hole with tensionless rs brane , it can not emerge a complete picture of a quantum hall system . to find a more realistic ads / bcft description of the qhe one needs to construct solutions with non - vanishing tension of the rs brane .
such solutions should enable us to describe the configurations shown in figure [ fig : gap](d ) .
we expect that the new solutions should have the following properties : namely , we expect that the dual system exhibits a gap in the bulk " , an edge current running along its perimeter and that the tension of the rs branes should be somewhat related to the fermi energy of the dual system . as a result
the conductivities will be modified ; in particular , the transverse electric conductivity in the bulk " and near the edges may differ , the ratio of the transverse heat and electric conductivities may satisfy a wiedemann - franz law ( [ wflaw2 ] ) modified by the interaction of the charge carriers .
we would like to thank e. akhmedov , a. dymarsky , a. ferraz , r. jackiw , h. johannesson , u. kol , g. semenoff and especially m. portnoi for useful discussions .
dm is grateful to the particle physics group at tel aviv university , where this work was initiated , for the warm hospitality .
this work was supported by the brazilian ministry of science and technology , capes foundation
brazil , and the iip - mit exchange program .
the work of dm was also supported through the contract # 8206 with the ministry of education and science of the russian federation , grant rfbr 11 - 02 - 01220 and grant of the president of the russian federation for support of scientific schools nsh-3349.2012.2 .
the work of ps was also supported by the `` iniziativa specifica infn '' fi11 and by the ec seventh framework programme ( fp7/2007 - 2013 ) under grant # 295234-fp7-people-2011-irses ( project qicft ) .
t. takayanagi , `` holographic dual of bcft , '' phys .
* 107 * ( 2011 ) 101602 [ arxiv:1105.5165 [ hep - th ] ]
. j. m. maldacena , `` the large n limit of superconformal field theories and supergravity , '' adv .
* 2 * ( 1998 ) 231 [ hep - th/9711200 ] .
s. ryu and t. takayanagi , `` holographic derivation of entanglement entropy from ads / cft , '' phys . rev .
lett . * 96 * ( 2006 ) 181602 [ hep - th/0603001 ] .
l. randall and r. sundrum , `` an alternative to compactification , '' phys .
rev . lett .
* 83 * ( 1999 ) 4690 [ hep - th/9906064 ] .
s. s. gubser , `` breaking an abelian gauge symmetry near a black hole horizon , '' phys .
d * 78 * ( 2008 ) 065034 [ arxiv:0801.2977 [ hep - th ] ] .
s. a. hartnoll , c. p. herzog and g. t. horowitz , `` building a holographic superconductor , '' phys .
* 101 * ( 2008 )
031601 [ arxiv:0803.3295 [ hep - th ] ] . h. liu , j. mcgreevy and d. vegh , `` non - fermi liquids from holography , '' phys .
d * 83 * ( 2011 ) 065029 [ arxiv:0903.2477 [ hep - th ] ] .
m. fujita , m. kaminski and a. karch , `` sl(2,z ) action on ads / bcft and hall conductivities , '' jhep * 1207 * ( 2012 ) 150 [ arxiv:1204.0012 [ hep - th ] ] .
r. b. laughlin , `` quantized hall conductivity in two - dimensions , '' phys .
b * 23 * ( 1981 ) 5632 .
r. b. laughlin , `` anomalous quantum hall effect : an incompressible quantum fluid with fractionally charged excitations , '' phys .
* 50 * ( 1983 ) 1395 .
j. k. jain , `` composite fermion approach for the fractional quantum hall effect , '' phys .
* 63 * ( 1989 ) 199 .
j. k. jain , `` theory of the fractional quantum hall effect , '' phys .
b * 41 * ( 1990 ) 7653 . the quantum hall effect , 2nd ed .
r. e. prange and s. m. girvin , springer - verlag , berlin ( 1990 ) .
r. jackiw , `` fractional charge and zero modes for planar systems in a magnetic field , '' phys .
d * 29 * ( 1984 ) 2375 [ erratum - ibid .
d * 33 * ( 1986 ) 2500 ] .
j. avron , r. seiler and b. simon , quantization of the hall conductance for general multiparticle schrodinger operators , " phys .
* 54 * ( 1985 ) 259 .
g. w. semenoff and p. sodano , `` nonabelian adiabatic phases and the fractional quantum hall effect , '' phys .
* 57 * ( 1986 ) 1195 .
g. w. semenoff , p. sodano and y. -s .
wu , `` renormalization of the statistics parameter in three - dimensional electrodynamics , '' phys .
* 62 * ( 1989 ) 715 .
a. p. polychronakos , `` topological mass quantization and parity violation in ( 2 + 1)-dimensional qed , '' nucl .
b * 281 * ( 1987 ) 241 .
a. p. polychronakos , `` on the quantization of the coefficient of the abelian chern - simons term , '' phys .
b * 241 * ( 1990 ) 37 .
n. bralic , c. d. fosco and f. a. schaposnik , `` on the quantization of the abelian chern - simons coefficient at finite temperature , '' phys .
b * 383 * ( 1996 ) 199 [ hep - th/9509110 ] . c. i. kane , m. p. a. fisher , `` quantized thermal transport in the fractional quantum hall effect , '' phys .
b * 55 * ( 1997 ) 15832 [ cond - mat/9603118 ] .
b. i. halperin , `` quantized hall conductance , current carrying edge states , and the existence of extended states in a two - dimensional disordered potential , '' phys .
b * 25 * ( 1982 ) 2185 .
s. a. hartnoll and p. kovtun , `` hall conductivity from dyonic black holes , '' phys .
d * 76 * ( 2007 ) 066001 [ arxiv:0704.1160 [ hep - th ] ] .
s. a. hartnoll , p. k. kovtun , m. muller and s. sachdev , `` theory of the nernst effect near quantum phase transitions in condensed matter , and in dyonic black holes , '' phys .
b * 76 * ( 2007 ) 144502 [ arxiv:0706.3215 [ cond-mat.str-el ] ] .
j. l. davis , p. kraus and a. shah , `` gravity dual of a quantum hall plateau transition , '' jhep * 0811 * ( 2008 ) 020 [ arxiv:0809.1876 [ hep - th ] ] .
m. fujita , w. li , s. ryu and t. takayanagi , `` fractional quantum hall effect via holography : chern - simons , edge states , and hierarchy , '' jhep * 0906 * ( 2009 ) 066 [ arxiv:0901.0924 [ hep - th ] ] . o. bergman , n. jokela , g. lifschytz and m. lippert , `` quantum hall effect in a holographic model , '' jhep * 1010 * ( 2010 ) 063 [ arxiv:1003.4965 [ hep - th ] ] .
m. nozaki , t. takayanagi and t. ugajin , `` central charges for bcfts and holography , '' jhep * 1206 * ( 2012 ) 066 [ arxiv:1205.1573 [ hep - th ] ] .
m. fujita , t. takayanagi and e. tonni , `` aspects of ads / bcft , '' jhep * 1111 * ( 2011 ) 043 [ arxiv:1108.5152 [ hep - th ] ] .
m. henningson and k. skenderis , `` the holographic weyl anomaly , '' jhep * 9807 * ( 1998 ) 023 [ hep - th/9806087 ]
. g. hayward , `` gravitational action for space - times with nonsmooth boundaries , '' phys .
d * 47 * ( 1993 ) 3275 .
s. s. gubser , i. r. klebanov and a. m. polyakov , `` gauge theory correlators from noncritical string theory , '' phys .
b * 428 * ( 1998 ) 105 [ hep - th/9802109 ] .
e. witten , `` anti - de sitter space and holography , '' adv .
theor . math .
* 2 * ( 1998 ) 253 [ hep - th/9802150 ] .
d. t. son and a. o. starinets , `` minkowski space correlators in ads / cft correspondence : recipe and applications , '' jhep * 0209 * ( 2002 ) 042 [ hep - th/0205051 ] .
c. p. herzog and d. t. son , `` schwinger - keldysh propagators from ads / cft correspondence , '' jhep * 0303 * ( 2003 ) 046 [ hep - th/0212072 ] .
n. r. cooper , b. i. halperin , i. m. ruzin , `` thermoelectric response of an interacting two - dimensional electron gas in a quantizing magnetic field , '' phys .
b * 55 * ( 1997 ) 2344 .
t. faulkner , h. liu , j. mcgreevy and d. vegh , `` emergent quantum criticality , fermi surfaces , and ads(2 ) , '' phys .
d * 83 * ( 2011 ) 125002 [ arxiv:0907.2694 [ hep - th ] ] .
d. b. chklovskii , b. i. shklovskii , l. i. glazman , `` electrostatics of edge channels , '' phys .
b * 46 * ( 1992 ) 4026 .
a. h. macdonald , `` edge states in the fractional - quantum - hall - effect regime , '' phys .
* 64 * ( 1990 ) 220 .
x. g. wen , `` gapless boundary excitations in the quantum hall states and in the chiral spin states , '' phys .
b * 43 * ( 1991 ) 11025 .
x. g. wen , `` electrodynamical properties of gapless edge excitations in the fractional quantum hall states , '' phys .
* 64 * ( 1990 ) 2206 .
x. g. wen , `` edge transport properties of the fractional quantum hall states and weak - impurity scattering of a one - dimensional charge - density wave , '' phys .
b * 44 * ( 1991 ) 5708 . | in this paper we study a simple gravity model dual to a @xmath0-dimensional system with a boundary at finite charge density and temperature . in our naive @xmath1 extension of a well known ads / cft system a non - zero charge density must be supported by a magnetic field . as a result ,
the hall conductivity is a constant inversely proportional to the coefficients of pertinent topological terms .
since the direct conductivity vanishes , such behaviors resemble that of a quantum hall system with fermi energy in the gap between the landau levels .
we further analyze the properties stemming from our holographic approach to a quantum hall system .
we find that at low temperatures the thermal and electric conductivities are related through the wiedemann - franz law , so that every charge conductance mode carries precisely one quantum of the heat conductance . from the computation of the edge currents we learn that the naive holographic model is dual to a gapless system if tensionless rs
branes are used in the ads / bcft construction . to reconcile this result with the expected quantum hall behavior
we conclude that gravity solutions with tensionless rs branes must be unstable , calling for a search of more general solutions . we briefly discuss the expected features of more realistic holographic setups .
itep - th-41/12 + iip - th-16/12 * on the ads / bcft approach to quantum hall systems * + _ @xmath2 international institute of physics , federal university of rio grande do norte , + av .
odilon gomes de lima 1722 , capim macio , natal - rn 59078 - 400 , brazil _ + _ @xmath3 institute for theoretical and experimental physics , + b. cheremushkinskaya 25 , moscow 117218 , russia _ + _ @xmath4 infn - laboratori nazionali di frascati , + via enrico fermi 40 , i-00044 frascati , italy _ + _ @xmath5 infn - sezione di perugia , + via a. pascoli , i-06123 perugia , italy _ + |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.