article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
searching for new modes in charm decays is of great interest .
it not only investigates possible decay mechanism and finds its contribution to the total decay width , but is also useful to simulate accurately cascade decays of bottom mesons and to eliminate backgrounds of charm modes in studying bottom decays . in this paper
, the semileptonic and hadronic @xmath9 decays in which the final state particles contain a @xmath10 meson are studied . whenever a specific state or decay mode is mentioned in this work
, the charge - conjugate state or decay mode is always implied .
the besii detector upgraded from the bes @xcite is a large solid - angle magnetic spectrometer described in detail elsewhere @xcite . a 12-layer vertex chamber ( vc ) surrounding the beryllium beam pipe
provides the trigger and coordinate informations . a forty - layer main drift chamber ( mdc ) , located outside the vc , yields precise measurements of charged particle trajectories with a solid angle coverage of @xmath11 of @xmath12
; it also provides ionization energy loss ( @xmath13 ) measurements used for particle identification .
momentum resolution of @xmath14 ( @xmath15 in gev / c ) and @xmath13 resolution of @xmath16 for bhabha scattering are obtained for data taken at @xmath17 = 3.773 gev .
an array of 48 scintillation counters surrounding the mdc measures the time of flight ( tof ) of charged particles with a resolution of about 180 ps for electrons . outside
the tof is a 12 radiation length barrel shower counter ( bsc ) comprised of gas tubes interleaved with lead sheets .
the bsc measures the energies of electrons and photons over @xmath18 of the total solid angle with an energy resolution of @xmath19 ( @xmath20 in gev ) and spatial resolution of @xmath21 mrad and @xmath22 cm for electrons . a solenoidal magnet outside the bsc provides a 0.4 t magnetic field in the central tracking region of the detector .
the magnet flux return is instrumented with three double layers of counters , that are used to identify muons with momentum greater than 500 mev / c and cover @xmath23 of the total solid angle .
the data used for this analysis were collected around the center - of - mass energy of 3.773 gev with the besii detector operated at the beijing electron positron collider ( bepc ) .
the total integrated luminosity of the data set is about 33 pb@xmath0 . at the center - of - mass energy 3.773 gev
, the @xmath24 resonance is produced in electron - positron ( @xmath25 ) annihilation .
the @xmath24 decays predominately into @xmath26 pairs .
if one @xmath27 meson is fully reconstructed , the @xmath9 meson must exist in the system recoiling against the fully reconstructed @xmath27 meson ( called singly tagged @xmath27 ) .
using the singly tagged @xmath28 sample , the semileptonic decays @xmath29 and @xmath30 are searched in the recoiling system .
the hadronic candidates @xmath3 and @xmath31 are reconstructed directly from the data sample of 33 pb@xmath0 .
events which contain at least three charged tracks with good helix fits are selected . to ensure good momentum resolution and reliable charged particle identification ,
every charged track is required to satisfy @xmath32cos@xmath33 , where @xmath34 is the polar angle .
all tracks , save those from @xmath35 decays , must originate from the interaction region by requiring that the closest approach of a charged track is less than 2.0 cm in the @xmath36 plane and 20 cm in the @xmath37 direction .
pions and kaons are identified by means of the combined particle confidence level which is calculated with information from the @xmath13 and tof measurements @xcite .
pion identification requires a consistency with the pion hypothesis at a confidence level ( @xmath38 ) greater than @xmath39 . in order to reduce misidentification ,
a kaon candidate is required to have a larger confidence level ( @xmath40 ) for a kaon hypothesis than that for a pion hypothesis . for electron or muon identification , the combined particle confidence level ( @xmath41 or @xmath42 ) , calculated for the @xmath43 or @xmath44 hypothesis using the @xmath13 , tof and bsc measurements ,
is required to be greater than @xmath39 .
the @xmath45 is reconstructed in the decay of @xmath46 . to select good photons from the @xmath45 decay , the energy deposited in the bsc is required to be greater than 0.07 gev , and the electromagnetic shower
is required to start in the first 5 readout layers . in order to reduce backgrounds ,
the angle between the photon and the nearest charged track is required to be greater than @xmath47 , and the angle between the cluster development direction and the photon emission direction to be less than @xmath48 @xcite .
the singly tagged @xmath28 sample used in the analysis was selected previously @xcite .
the singly tagged @xmath28 mesons were reconstructed in the nine hadronic modes of @xmath49 , @xmath50 , @xmath51 , @xmath52 , @xmath53 , @xmath54 , @xmath55 , @xmath56 and @xmath57 .
the distributions of the fitted invariant masses of the @xmath58 combinations are shown in fig .
[ dptags_9modes ] .
the number of the singly tagged @xmath28 mesons is @xmath59 @xcite , where the first error is statistical and the second systematic . , ( b ) @xmath60 , ( c ) @xmath51 , ( d ) @xmath61 , ( e ) @xmath62 , ( f ) @xmath63 , ( g ) @xmath64 , ( h ) @xmath65 , ( i ) @xmath57 combinations ; ( j ) is the fitted masses of the @xmath66 combinations for the nine modes combined together.,title="fig:",width=340,height=264 ] ( -210,175)*(a ) * ( -90,175)*(b ) * ( -210,145)*(c ) * ( -90,145)*(d ) * ( -210,115)*(e ) * ( -90,100)*(f ) * ( -210,65)*(g ) * ( -90,62)*(h ) * ( -210,35)*(i ) * ( -90,45)*(j ) * ( -150,0)invariant mass ( gev/@xmath67 ( -250,50 ) candidates for @xmath70 and @xmath71 are selected from the surviving tracks in the system recoiling against the tagged @xmath28 . to select these candidates
, it is required that there are three charged tracks , one of which is identified as an electron or a muon with the charge opposite to the charge of the tagged @xmath28 meson , the other two are identified as a @xmath72 and a @xmath73 mesons .
the @xmath10 meson is selected by requiring that the difference between the invariant mass of @xmath74 and the nominal @xmath10 mass should be less than 20 mev/@xmath75 . in the semileptonic
decays , one neutrino is undetected . a kinematic quantity @xmath76 is used to obtain the information of the missing neutrino , where @xmath77 and @xmath78 are the total energy and the total momentum of all missing particles respectively .
monte carlo study shows that the background modes for @xmath79 come primarily from @xmath80 , @xmath81 and @xmath3 .
the requirement @xmath82 is imposed to reduce these backgrounds , where @xmath83 is the standard deviation of the @xmath84 distribution obtained from monte carlo simulation . in fig .
[ dp_phimunu_phipi_invm_umiss ] , the solid and dashed histograms show respectively the distributions of the invariant masses of @xmath85 and @xmath86 for @xmath2 and @xmath3 from monte carlo events .
the candidate mode @xmath2 has a potential hadronic background of @xmath3 due to the misidentification of a charged pion as a muon .
these backgrounds can be suppressed by requiring the invariant masses of the @xmath85 combinations to be less than 1.8 gev/@xmath75 .
similarly the invariant masses of the @xmath87 combinations are also required to be less than 1.8 gev/@xmath75 in order to remove the @xmath3 background for the @xmath68 decay . to suppress backgrounds from decays with neutral pion , for example @xmath81 , the number of the isolated photons is required to equal two if the singly tagged @xmath28 mode contains a @xmath45 meson , otherwise zero .
the events with isolated photons those energies are larger than 0.1 gev excluded in the tagged @xmath28 are not kept .
( solid ) and @xmath86 ( dashed ) for @xmath88 and @xmath3 from monte carlo events.,title="fig:",width=302,height=226 ] ( -137,13)@xmath89 ( gev/@xmath75 ) ( -225,45 ) fig .
[ dp_phienu_2d ] and fig .
[ dp_phimunu_2d ] show respectively the scatter plots of the invariant masses of the @xmath74 combination versus the @xmath66 combination for @xmath29 and @xmath30 before and after applying both @xmath84 and @xmath90 cuts .
no event for the @xmath29 and @xmath30 decays satisfies the selection criteria .
invariant masses versus the @xmath66 invariant masses for the @xmath29 candidates : ( a ) before and ( b ) after applying both @xmath84 and @xmath91 cuts , where the rectangle represents the combined signal region of @xmath10 and @xmath92.,title="fig:",width=264,height=226 ] ( -159,130)*(a ) * ( -159,65)*(b ) * ( -130,-3)@xmath93 ( gev/@xmath67 ( -208,50 ) invariant masses versus the @xmath66 invariant masses for the @xmath30 candidates : ( a ) before and ( b ) after applying both @xmath84 and @xmath94 cuts , where the rectangle represents the combined signal region of @xmath10 and @xmath92.,title="fig:",width=264,height=226 ] ( -159,130)*(a ) * ( -159,65)*(b ) * ( -130,-3)@xmath93 ( gev/@xmath67 ( -208,50 ) the candidate for @xmath3 is reconstructed from @xmath95 . the distribution of the fitted invariant mass of the @xmath96 combination is shown in fig .
[ dptags_9modes](d ) . as mentioned in previous subsection , the @xmath10 meson
is selected though its decay to @xmath74 .
[ dp_phipi_k2p_data](a ) shows the fitted invariant mass spectrum of the @xmath74 pairs from the @xmath92 signal region for the @xmath97 candidate events .
fitting the mass spectrum with a gaussian function convoluted breit - wigner gives @xmath98 @xmath10 signal events . to select the decay @xmath3
, we require the invariant mass of the @xmath74 combination to be within 20 mev/@xmath75 of the nominal @xmath10 mass , and the invariant mass distribution of @xmath96 combination is shown in fig .
[ dp_phipi_k2p_data](b ) , a clear signal of the decay @xmath3 is observed .
fitting the mass spectrum of the @xmath96 combination with a gaussian function as the signal and a special function @xcite for the background yields @xmath99 signal events , where the mass resolution is fixed at 0.0022 gev/@xmath75 determined from the monte carlo simulation
. there may be the @xmath74 combinatorial background in the observed @xmath86 events .
the background events are estimated to be @xmath100 by the @xmath10 sideband @xmath101 0.05 gev/@xmath75 . for the @xmath102 candidates , the fitted invariant mass spectrum is shown in fig .
[ dptags_9modes](a ) , and @xmath103 events are yielded from the fit .
combination in the @xmath92 signal region , ( b ) @xmath96 combination in the @xmath10 signal region.,title="fig:",width=302,height=264 ] ( -115,157)*(a ) * ( -115,70)*(b ) * ( -114,95)@xmath104 ( gev/@xmath67 ( -114,6)@xmath105 ( gev/@xmath67 ( -226,55 ) in the fitted yields , there are still some background contaminations which shape a peak under the @xmath92 peak . these backgrounds are estimated by analyzing the monte carlo sample which is about 14 times larger than the data .
the monte carlo events are generated as @xmath106 , and both @xmath9 and @xmath107 mesons decay to all possible modes according to the decay modes and the branching fractions quoted from pdg @xcite excluding decay modes under study .
the number of events satisfying the selection criteria is then normalized to the data . for the @xmath3 decay ,
the dominant background modes are @xmath2 , @xmath68 and @xmath81 , and the number of the background events is @xmath108 . for the @xmath31 decay ,
the number of the background events is @xmath109 , after subtracting the numbers of the background events , @xmath110 and @xmath111 signal events for the @xmath3 and @xmath6 decays are remained .
the reconstruction efficiencies of the semileptonic decays @xmath68 , @xmath2 and the hadronic decays @xmath3 , @xmath31 are estimated by monte carlo simulation with the geant3 based monte carlo simulation package @xcite .
a detailed monte carlo study shows that the efficiencies are @xmath112 , @xmath113 , @xmath114 and @xmath115 .
a @xmath92 signal is observed neither in the @xmath68 decay nor in the @xmath2 decay .
the upper limit of the branching fraction can be calculated using @xmath116 where @xmath117 is the upper limit of the signal yield given with the feldman - cousins prescription @xcite , which is 2.44 for zero observed event in the absence of background for the confidence level of 90% , @xmath118 is the detection efficiency , @xmath119 is the number of the singly tagged @xmath28 mesons and @xmath120 is the systematic error .
the upper limits of the branching fractions are summarized in the second column of table [ uplbr ] .
comparisons with those obtained by markiii @xcite and besi @xcite experiments are also listed in the third and fourth columns of table [ uplbr ] .
.upper limits for the branching fractions ( % ) are given at the 90% confidence level . [ cols="^,^,^,^",options="header " , ] in addition , the branching fraction for @xmath3 is obtained to be @xmath121 by using the branching fraction of the @xmath6 decay quoted from pdg @xcite , where the first error is statistical and the second systematic .
the systematic error of the branching fraction for @xmath3 includes the systematic error of the ratio @xmath122 ( 5.3% ) and the uncertainty of the branching fraction for @xmath6 ( 6.5% ) .
the total systematic error is obtained to be 8.4% by adding these uncorrelated errors in quadrature .
using a data sample of integrated luminosity of 33 @xmath123 collected around 3.773 gev with the besii detector at the bepc , the semileptonic decays @xmath1 , @xmath2 and the hadronic decay @xmath3 are studied .
the upper limits of the branching fractions are set to be @xmath4 2.01% and @xmath5 2.04% at the 90% confidence level .
the ratio of the branching fractions for @xmath3 relative to @xmath6 is measured to be @xmath7 .
in addition , the branching fraction for @xmath3 is obtained to be @xmath8 .
the bes collaboration thanks the staff of bepc for their hard efforts .
this work is supported in part by the national natural science foundation of china under contracts nos .
10491300 , 10225524 , 10225525 , 10425523 , the chinese academy of sciences under contract no .
kj 95t-03 , the 100 talents program of cas under contract nos .
u-11 , u-24 , u-25 , the knowledge innovation project of cas under contract nos .
u-602 , u-34 ( ihep ) , the national natural science foundation of china under contract no .
10225522 ( tsinghua university ) . | using a data sample of integrated luminosity of about 33 pb@xmath0 collected around 3.773 gev with the besii detector at the bepc collider , the semileptonic decays @xmath1 , @xmath2 and the hadronic decay @xmath3 are studied .
the upper limits of the branching fractions are set to be @xmath4 2.01% and @xmath5 2.04% at the 90% confidence level .
the ratio of the branching fractions for @xmath3 relative to @xmath6 is measured to be @xmath7 .
in addition , the branching fraction for @xmath3 is obtained to be @xmath8 . |
there has been increasing interest in developing high efficiency , high power thermoelectric devices , constructed from the bottom - up using nanoscale designs .
the primary applications driving interest in this area are energy - harvesting , the collection and conversion of waste heat to electrical power , produced from sources ranging from hand - held electronics to industrial sources of heat , and refrigeration , actively cooling a spatial region via electrons to evacuate heat out of an area .
the use of nanoscale architecture instead of bulk materials is motivated by the low figure of merit - or poor thermoelectric conversion efficiency - of bulk materials , whereas conduction through nanoscale electronics can reach carnot efficiency .
one way to produce high thermodynamic efficiency in the conversion of heat to power is the use of structures with sharp spectral features , such as quantum dots @xcite .
the use of quantum dots in thermoelectric transport has been extensively researched in the past several years @xcite .
see ref .
@xcite for recent reviews of these and related activities .
in particular , if the dot is transporting electrons via resonant tunneling , the quantum dot acts as an energy filter - permitting the tight - coupling " of heat and charge transport which can lead to carnot efficiency .
other structures from mesoscopic physics , including the quantum point contact and electron cavity @xcite , quantum wells @xcite , quantum hall bar @xcite , superconducting leads @xcite , and coulomb blockaded quantum dots(s ) @xcite have also been investigated for their multi - terminal thermoelectric properties .
the late prof .
markus bttiker , for whom this special issue is in memory of , was highly influential in the theoretical development of these ideas , as can be seen in the above list of references .
several experiments have begun exploring this physics .
et al_. @xcite performed experiments on a cavity connected to resonant tunneling quantum dots acting as an electronic refrigerator , based on the proposal of edwards
_ @xcite . they demonstrated that applying bias to the system results in cooling a large @xmath0 cavity from @xmath1 to below @xmath2 . very recently ,
_ @xcite and f. hartmann _
@xcite showed rectification of electrical current of the nano amp scale and power production on a pico watt scale from a capacitively coupled source of fluctuations .
this was based on the theoretical proposal of sothmann _ et al .
_ @xcite .
while a nanoscale thermoelectric generator can power nanoscale devices , it is of great interest to find practical ways to scale up these nanoengines .
one way is to simply add them in electrical series while being able to couple to a common source of heat . in commercial thermoelectric generators ,
this is usually done by alternating the semiconductor type , of either p - type or n - type to be able to apply the heat difference in parallel because the heat and electrical transport are in opposite directions in a p - type semiconductor @xcite .
this permits the generated voltage to grow with the number of elements , while keeping the current fixed .
various other ways of scaling the devices have been proposed @xcite . in jordan _ et al .
_ , a layered structure was proposed by alternating layers of semiconductor and self - assembled quantum dots , so as to create a large - scale device where heat and electrical transport are separated , while keeping the high thermodynamic efficiency @xcite .
this is a parallel strategy of scaling , so the generated current grows with the number of dots , while the voltage difference is fixed .
considered a technically simpler method of creating quantum wells that permit resonant tunneling @xcite .
the physics there is somewhat different because energy may be distributed into the transverse degrees of the electron motion .
nevertheless , reasonable thermodynamic efficiency was found , with increased power production .
one of the outstanding challenges to creating high - efficiency thermoelectric devices is phonon transport .
phonons give a way for the hot and cold side of the device to exchange energy directly , without converting it to power via the electrons .
therefore , any possible way to reduce the phonon transport while still allowing electron transport will aid in the overall thermodynamic efficiency .
interface - based devices , such as described above can help with this , because the interface helps to reflect the phonons @xcite .
ideally , there will be additional material layers that act as thermal insulators .
the purpose of the present article is to build on these accomplishments , and make an analysis of a thermoelectric device based on semiconductor super - lattices .
these structures are fabricated by making a periodic layered structure of alternating materials , such as gaas / algaas .
the effect on the electronic transport is to form a series of mini - bands of allowed and forbidden energies where conduction electrons can transport @xcite .
this structure can be considered as a generalization of the resonant tunneling quantum wells .
the mini - band gives a top - hat profile of variable width for the energy - filtering .
such a top - hat profile has been argued by whitney to offer the highest efficiency for a given power extraction @xcite .
however , we note the transverse degrees of freedom make the system somewhat different . at a small band width
, our system will be similar to a quantum well , but can be extended to allow a fixed width longitudinal energy window .
we make a first - principles analysis of the heat and charge transport in a three - terminal geometry , where two terminals carry charge , and a third carries heat ( see fig . [ fig_gemoetry_energydiagram ] ) .
the offset of the miniband centers and their respective widths determine the power produced and efficiency of the heat conversion given fixed temperature differences .
we next make an analysis of the coefficient of performance of this device for the purposes of refrigeration of the central region .
the final purpose of the present work is to also make a systematic calculation of the heat transport due to phonons .
we make a detailed investigate the heat current through the device from phonon transport using a kronig - penney model , and consider different ways to stop it .
the paper is organized as follows . in section [ heat_engine_electron ] ,
we introduce a model of the superlattice heat engine in its dual roles : the energy harvester in section [ energy_harvesting ] and the refrigerator in section [ refrigeration ] .
we discuss our results for the generated power and the efficiency of the engine in section [ result ] , and show the cooling power and the coefficient of performance in section [ refrigeration ] .
the second part , section [ heat_engine_phonon ] focuses on phonon heat current generated by the heat engine and discusses the effect on the efficiency of the heat engine .
we finish with our conclusions in section [ conclusion ] .
we consider a setup shown schematically in fig . [ fig_gemoetry_energydiagram ] .
it consists of a center cavity connected to two electronic reservoirs @xmath3 via a superlattice .
the electronic reservoirs are characterized by the occupation of the states given by the fermi function , @xmath4 + 1]^{-1}$ ] with temperature @xmath5 and chemical potential @xmath6 , and the underlying assumption is that inelastic scattering processes restore the local thermal equilibrium on a fast time scale .
the center cavity is also assumed to be in thermal equilibrium with a heat bath of temperature @xmath7 .
we assume the structure is translationally symmetric within the @xmath8 and @xmath9 directions , perpendicular to the growth direction @xmath10 .
the superlattices are designed as periodic structures with lattice constant @xmath11 ( sum of the width of well and barrier thickness ) .
the periodicity of the structure in the @xmath10 direction implies that the eigenstates of the hamiltonian can be written as bloch states with the bloch vector @xmath12 $ ] , so the simple solutions with the kronig - penney model resemble the standard superlattices @xcite .
the corresponding eigenvalues of the hamiltonian form a miniband .
the allowed energies of the miniband can be written , @xmath13 where this is the result of a standard tight - binding calculation with band indices @xmath14 .
@xmath15 is a center of miniband @xmath14 and its width is @xmath16 .
our discussion is restricted to the electron transport through the lowest miniband , thus the miniband index @xmath14 is neglected from here on . to find the electric and heat currents through the superlattices expressed in terms of an integral over energy
, we first find the density of states of the superlattice .
for the given miniband , the energy is that of a two dimensional electron gas with the bottom of the band at @xmath17 in eq .
( [ superlattice_energy ] ) . therefore the three dimensional density of states is given by @xcite @xmath18 the factor @xmath19 is to avoid double counting the spin and @xmath20 is a two dimensional density of states per unit area .
the density of states of the superlattice can be factorized into longitudinal and transverse parts , and the one dimensional superlattice density of states @xmath21 is @xmath22 where we use a heaviside step function @xmath23 to show the range of the energy @xmath24 with the maximum / miminum value @xmath25 . therefore
, the electrons only in selected values of energy @xmath26 will transport and generate current .
we assume the central cavity region is strongly coupled to the external source of energy , so the occupation is described as a fermi function with local temperature @xmath27 determined by the thermal reservoir .
the electric and energy currents for simplified miniband transport in @xmath10 direction emitted by the reservoir @xmath28 into an cavity @xmath29 can be evaluated within a landauer - bttiker approach as @xmath30 , \label{current_r}\\ j_r&=&\frac{\nu_{2d}\mathcal{a}}{2\pi\hbar}\int_{e_{z}^{r-}}^{e_{z}^{r+}}de_{z}de_{\perp}\;e\;[f_r(e)-f_c(e)],\label{heatcurrent_r}\end{aligned}\ ] ] where @xmath31 is the surface area of the superlattice , @xmath32 is the maximum / minimum energy of the reservoir @xmath28 , and we denote @xmath33 as the energy carried in the transverse degrees of freedom , and @xmath34 as the energy carried in the longitudinal degree of freedom , so that @xmath35 . here , the square root in eq .
( [ density_of_state ] ) cancels the velocity of the electron , @xmath36 , in the current @xmath37 $ ] @xcite .
we see that the range of the integral comes from the density of states which gives a transmission function of flat box form , @xmath38 . to rewrite the above equations
, we introduce the integrals @xmath39 and @xmath40 with the dilogarithm @xmath41 .
then the simplified analytic expressions of the electric and energy currents are @xmath42 , \label{electric_current_general}\\ j_r&=&\frac{\nu_{2d}\mathcal{a}k_b}{2\pi\hbar } \left[\int_{e_{z}^{r-}}^{e_{z}^{r+}}de_{z}e_z \left [ t_rk_1(\tilde{e_z^r})-t_ck_1(\tilde{e_z^c } ) \right]\right . \nonumber\\ & & \left.+k_b^2\int_{e_{z}^{r-}}^{e_{z}^{r+}}de_{z } \left [ t_r^2k_2(\tilde{e_z^r } ) -t_c^2k_2(\tilde{e_z^c})\right ] \right],\label{heat_current_general}\end{aligned}\ ] ] where @xmath43 and @xmath44 .
we now analyze the system by focusing on linear response and later turn to the nonlinear regime . to simplify the analysis of the system ,
we introduce the average temperature @xmath45 and the temperature difference @xmath46 . for energy harvesting ,
the temperature difference is considered to be @xmath47 .
( if we consider refrigeration , the temperature difference is @xmath48 . )
we introduce the bias @xmath49 to the system by applying @xmath50 and @xmath51 .
we also rewrite the width of each miniband , @xmath52 and @xmath53 where @xmath54 , with @xmath55 , determines the asymmetry between the left and right energy width , so the relative thickness of the left and right miniband width is determined by @xmath54 .
the chemical potential of the cavity @xmath56 as well as the temperature @xmath27 are determined by imposing conservation of charge and energy , @xmath57 and @xmath58 where @xmath59 is the heat current entering from the heat source . from these conservation laws , we can obtain the electric and heat currents through the system , as well as @xmath59 and @xmath56 .
to linear order in the temperature difference @xmath60 and the bias voltage @xmath61 , the net current flowing through the system , @xmath62 , is given by @xmath63 the electrical conductance @xmath64 and thermopower ( or seebeck coefficient ) @xmath65 of the system are @xmath66,\label{thermopower}\end{aligned}\ ] ] where we have introduced the auxiliary functions @xmath67 the electrical conductance shows that @xmath68(@xmath69 ) is proportional to the electrical conductance of the left(right ) superlattice , so the net conductance @xmath64 is simply the series combination of the two conductors .
the three - terminal thermopower @xmath65 is determined by the difference between left and right two - terminal thermopower of each superlattice , and vanishes when they are identical @xcite . therefore , depending on the magnitude of the left and right properties , @xmath65 shows whether the system is analogous to a p - type or n - type semiconductor .
when @xmath65 is positive for @xmath70 , the system acts as if the mobile charge carrier is positive and behaves like a p - type semiconductor , and vice versa .
the bias voltage @xmath61 applied against heat driven charge current generates a finite output power @xmath71 .
the power vanishes either when no bias voltage is applied or at the stopping voltage , @xmath72 .
the output power takes its maximum value at half of the stopping voltage , @xmath73 in order to evaluate the efficiency @xmath74 given by the ratio of output power to input heat current , we need to find the heat current @xmath75 injected from the heat bath , @xmath76 which shows the peltier effect . here
the coefficients @xmath77 , @xmath78 , and @xmath79 are defined as @xmath80,\label{peltier_coeff}\\ h_{t1}&=&\frac{\nu_{2d}\mathcal{a}}{2\pi\hbar}\:k_b^2t\frac{g_{l1}-g_{r1}}{g_{l1}+g_{r1 } } \left[\frac{(g_{r2}+g_{r3})^2}{g_{r1 } } \right.\nonumber\\ & & \left.-\frac{(g_{l2}+g_{l3})^2}{g_{l1}}-4(g_{r2}+g_{r3})(g_{l2}+g_{l3 } ) \right],\nonumber\\ \label{h_t1}\\ h_{t2}&=&\frac{\nu_{2d}\mathcal{a}}{2\pi\hbar}\:k_b^2 t \left [ ( g_{l4}+2g_{l5}-g_{l6})\right.\nonumber\\ & & \left.+(g_{r4}+2g_{r5}-g_{r6 } ) \right],\label{h_t2}\end{aligned}\ ] ] together with the auxiliary functions @xmath81 rewriting the heat current in eq . ( [ heatcurrent_linear ] ) as @xmath82 , we see the meaning of the coefficients more clearly . the first part is simply proportional to the charge current , and @xmath77 shows the presence of heating from the peltier contribution .
the relation @xmath83 from eq .
( [ thermopower ] ) and eq .
( [ peltier_coeff ] ) shows the coefficient @xmath77 can be considered as the back action counterpart to @xmath65 , and the minus sign of the relation comes because @xmath59 is the heat current injected from the heat bath into the cavity .
this relation shows our system satisfies onsager symmetry resulting from the time reversibility , which relates the seeback and peltier coefficients @xcite .
the second part shows the heat generated by the temperature difference and the thermal conductance is @xmath84 . the first one , @xmath78 , contributes to the heat flow from the asymmetric superlattices and disappears when the left and right superlattices are symmetric . the heat current at maximum power takes the form @xmath85\delta t.\end{aligned}\ ] ] therefore , the efficiency at maximum power is given by @xmath86 where the second approximation comes from the carnot efficiency for the small temperature difference , @xmath87 .
therefore , the combination of the coefficients , inside of the vertical bars in eq .
( [ efficiency_maxp ] ) determines the efficiency ratio to carnot efficiency .
[ fig_linear_maxpower ] shows the maximum output power and the corresponding efficiency on equal size bands ( @xmath88 ) as a function of the centers of minibands of the two superlattices , @xmath89 .
both the power and efficiency are symmetric with respect to an exchange of @xmath90 and @xmath91 . on the one hand ,
the maximum power arises when one of the two center value stays around twice of @xmath92 and the other center is deep below the equilibrium chemical potential , below about @xmath93 .
the center around @xmath94 is the position that makes the bottom of the miniband stay around the equilibrium chemical potential , and the other center which is below @xmath93 makes the top of the miniband deep below the chemical potential , because the miniband width is @xmath95 . on the other hand ,
the efficiency acts symmetrically depending on the position of band centers as well as the power .
however , maximum efficiency comes when one of the center of miniband is bigger than @xmath94 with an appropriate value of the other center , and it comes with suppressed output power .
therefore , depending on what we want to optimize , the output power or efficiency , we can chose the position of band centers .
we also show the maximum power for the different miniband widths from the top panels of fig .
[ fig_linear_maxpower ] . as the electrons transport only within the miniband , a wider miniband allows more electrons to transport and generate more power
. however , as the miniband width goes too far above @xmath96 , the power increase stops .
the reason is that the energy window of the order @xmath97 will be a more effective energy guard for the carriers .
different from the power , efficiency is reduced as the miniband width increases , eq .
( [ fig_linear_maxpower_asym](c),(d ) ) . as the width increases
, the energy filtering by the superlattices is lesser efficient so the efficiency decreases .
however , as the width increases continuously , the efficiency will saturate at some point with the same reason for the power saturation .
these results show that the superlattice heat engine has more power output than the quantum well engine @xcite .
the maximal output power of the superlattice heat engine with units @xmath98 is @xmath99 with the efficiency @xmath100 ( for the miniband width , @xmath101 ) , while the quantum well heat engine is @xmath102 with the efficiency @xmath103 . from eq .
( [ gr1]-[gr3 ] ) , we can give a simple reason for this : when the miniband width is suppressed ( longitudinal energy window becomes one value and the electrons transport only with the certain longitudinal energy ) , our results approach the quantum well case .
therefore , a large miniband width permits more electrons to transport in the longitudinal degree of freedom .
while the superlattice engine is more powerful , it is less efficient an energy filter than the quantum well case : this is from the difference between the miniband of the superlattice and the sub - band threshold of the quantum well .
we now turn to an asymmetric system , @xmath104 . to find the optimized combination of the centers of miniband and the asymmetric parameter @xmath54 for the maximum output power , we consider the output power and the efficiency for @xmath105 . fig .
[ fig_linear_maxpower_asym ] shows plots of the power as a function of @xmath54 and @xmath91 . when @xmath101 , the largest output power @xmath106 ( around a @xmath107 increases from the symmetric case ) arises for @xmath108 with @xmath109 . in this case
, the efficiency goes to @xmath110 ( a @xmath111 increase from the symmetric case ) .
meanwhile , the maximum efficiency @xmath112 comes with @xmath113 and @xmath114 for @xmath101 . when @xmath113 , the miniband widths become @xmath115 and @xmath116 , and the output power is strongly suppressed .
we can understand the reason for this last fact because when the width of the left miniband vanishes , transport from the left is cut off , giving no power , but very good energy filtering , which increases efficiency . to explain other features of the plots , we note that for the right superlattice , the center of miniband stays slightly above the equilibrium chemical potential , and the width of miniband @xmath117 is larger than the width of the distribution function @xmath118 .
consequently , making the miniband wider has no effect on the electron transport because the number of electrons with energies around minimum / maximum of the energy window is exponentially small .
simultaneously , for the smaller miniband of the left barrier , centering it around the energy region where the left reservoir is occupied but the cavity not , gives very good current production .
as the miniband width @xmath92 increases , fig .
( [ fig_linear_maxpower_asym]b ) , @xmath54 moves to a positive value while the center @xmath91 stays around the value that maximizes the power for the symmetric case .
therefore , depending on the size of the miniband width relative to the width of the occupation function of the reservoir and the cavity , @xmath54 can be fixed to give the optimized results .
it is interesting next to consider the output power and the efficiency in the nonlinear regime , where qualitatively new physics can appear @xcite .
we numerically calculated the stopping voltage @xmath119 , the centers of miniband @xmath89 , and asymmetry parameter @xmath54 in order to maximize the output power . for the case @xmath101
, we show the maximum output power and the efficiency in fig .
[ fig_non ] .
this is obtained for the miniband centers @xmath120 and @xmath121 with the asymmetric parameter @xmath122 .
the output power increases quadratically in the temperature difference for the fixed average temperature @xmath123 as in the linear case , eq .
( [ maxpower_linear ] ) , but it also depends on the average temperature , @xmath124
. therefore higher average temperature as well as the temperature difference gives bigger output power . in table
[ table_comparison ] , we compare the maximum output power and the corresponding efficiency for three systems : the quantum dot @xcite , quantum well @xcite , and superlattice based three terminal heat engines for a realistic device parameter @xmath125 with the room temperature @xmath126 and the temperature difference @xmath127 . we see that @xmath128 of the superlattice heat engine generates a larger power about @xmath129 with a small reduction for the efficiency @xmath130 .
therefore , the superlattice heat engine is the more powerful heat engine .
.we compare our result with other three terminal geometries : ( power - optimized resonant width ) quantum dot and quantum well heat engines for @xmath126 and @xmath127 . [
cols="^,^,^,^",options="header " , ] now , we consider the same geometry but for a different purpose , a cooling system . if proper bias voltage @xmath131 is applied over the junction and the minibands of superlattices are suitably arranged , a current flows from right to left as hot electrons tunnel through the miniband of the left superlattice from the cavity to left reservoir and cold electrons from the right reservoir tunnel through the right miniband to the cavity .
this leads to decrease in the average energy of electrons in the central cavity , that is cooling utilizing the peltier effect : in the peltier effect , the junction is electrically biased and a produced heat current flow given an electric current . for the refrigerator , @xmath60 is negative in our notation , and the temperature of the center cavity cooled down to the amount of @xmath132 .
the purpose of refrigeration is to achieve a large temperature reduction .
the base temperature of the refrigerator is defined as the temperature for equilibrium where the evacuated heat current balances any external heat leaks . since we are now considering no external heat leaks , such as electron - phonon coupling
, the base temperature is the temperature for which @xmath133 in eq .
( [ heatcurrent_linear ] ) with the temperature reduction @xmath134 in the linear regime .
the refrigeration works only when the heat current is emitted by the cavity into the reservoirs , @xmath135 , therefore the temperature reduction of the cavity should be in the range of @xmath136 .
the applied bias voltage @xmath61 and the absorbed heat from the cold cavity let the heat flow @xmath59 rejected to the left and right reservoirs .
therefore , the cooling power is @xmath59 , and an efficiency of the cooling is normally characterized by the coefficient of performance ( cop ) , defined as the ratio of the cooling power to the total input power @xmath137 , @xmath138 similar to the efficiency of the energy harvester , cop is also bounded by the carnot value , @xmath139 .
moving beyond the linear regime is important to find the nonlinear behavior of the refrigerator , which determines the lowest temperature it can reach .
[ fig_rfg](a ) shows a plot of the cooling power vs applied bias and temperature differences , when other variables , @xmath140 , @xmath141 , @xmath101 and @xmath142 , are optimized to give the largest cooling power for the positive applied bias @xmath143 .
here , @xmath144 means that left miniband width is wider than the right miniband , and the reason can be understood as the energy harvesting that we discussed in section [ result ] .
note that , the position of the left / right miniband of the refrigerator is opposite from the energy harvesting , so this makes the conduction electrons take energy from the cavity and continue to cool it . the black curve in panel ( a ) of fig . ( [ fig_rfg ] )
represents when the cooling power becomes zero , and the right hand side of the line ( within the parabola ) is for the positive cooling power which is the region that works for the refrigeration .
therefore , we can see the optimal region for the temperature difference and the applied voltage , as shown in the plot .
when the temperature difference is anywhere between the zero and @xmath145 , as an example the vertical line in the figure , the applied voltage should stay in between @xmath146 and @xmath147 to make the engine work as the refrigerator .
specially , when the applied voltage is around @xmath148 , the refrigerator gives maximum cooling power for a given temperature difference , and we also can have the maxumum temperature reduction @xmath149 of the system . when @xmath150 , the heat current increases linearly in panel ( b ) .
so , the maximum temperature reduction for this bias voltage is @xmath151 .
for example , at room temperature , the maximum temperature reduction is @xmath152 by applying the bias around @xmath153 .
if the temperature reduction is @xmath154 , the corresponding cooling power is approximately @xmath155 .
the cop at @xmath156 as a function of the temperature difference is also plotted in fig .
( [ fig_rfg](b ) ) .
the maximum cop @xmath157 comes with when @xmath158 .
therefore , the refrigerator can be operated at the optimal regime for the cooling power or cop by reasonably choosing the parameters .
one important consideration in thermoelectric heat engine is the heat flow carried not only by the conduction electrons , but by the phonons as well . here
, we calculate this effect for our system to see how it will affect the efficiency in a more realistic modeling .
the major heat flow from hot to cold reservoirs is carried by excitations such as phonons .
for this three terminal heat engine , the phonon heat flow @xmath159 is in parallel with electronic flow @xmath160 , and the total heat flow for a given generated power @xmath161 is the sum of them , @xmath162 .
then we can rewrite the efficiency in the present of the phonons @xmath163 the phonon heat current in the superlattice differs from the heat current based on the bulk material properties due to the new periodicity of the structure .
the presence of interfaces can alter the phonon spectra result from wave interference scattered at the interfaces .
the formation of miniband gaps in superlattices due to phonon interference leads to a reduction of the phonon velocity @xmath164 which gives a reduction of the thermal conductivity . to illustrate this effect in a simple model
, we assume the complete separation of longitudinal and transverse vibrations for phonons with the wave vector to parallel to @xmath10 axis . also assuming perfect interfaces , the transverse momentum @xmath165 is conserved . when we consider phonons propagating in the growth direction , this can be treated from a kronig - penney type approach for one dimensional atomic chain @xcite .
we suppose the same monolayer spacings @xmath166 and the magnitude of atomic constants @xmath167 in between all the atoms . the layer one(two ) has @xmath168 atoms with mass @xmath169 , and the thickness of sublattice one ( two ) is @xmath170 ( @xmath171 ) which gives the length of unit period @xmath172 .
then the characteristic equation is @xmath173 where @xmath174 is the phonon wave vector of each layer @xmath175 with @xmath176 , and @xmath177 is a longitudinal superlattice wave vector .
this model gives us the dispersion of the longitudinally polarized phonons for the cross plane transport with zero transverse momentum .
the dispersions for different periods are shown in fig .
[ fig_phonon_disp ] for @xmath178 with the mass ratio @xmath179 .
increasing the superlattice periods ( @xmath180 and @xmath181 ) gives more band folding and decreases the average group velocity .
at nonzero transverse momentum , the longitudinally polarized phonon mode for the cross plane dynamics is described as @xcite @xmath182 where @xmath183 is a solution from eq .
( [ characteristic equation ] ) and @xmath184 is a characteristic frequency of the material .
therefore , the dispersion of non zero transverse momentum can be used to calculate the group velocity in the superlattices , @xmath185 .
since we are dealing with the heat transfer only to the cross plane direction , the group velocity that we need is only the @xmath10 direction , @xmath186 where we use the one dimensional group velocity @xmath187 .
now , we can write the phonon heat current from the reservoir @xmath28 into the cavity through the superlattice , @xmath188 where @xmath189 is the occupation difference between the reservoir and cavity with the bose - einstein distribution function @xmath190 and @xmath191 represents the transmission function which depends only on the longitudinal momentum @xmath177 . in our analysis , we assume the ballistic case with @xmath192 . when we put the longitudinal group velocity eq.([group_vel ] ) into the first equation of eq .
( [ phonon_heat_current_mom_gen ] ) , we can factorize the heat current to the longitudinal @xmath193 and transverse @xmath194 directions @xmath195 with a modified occupation difference @xmath196 this function gives the effective phonon occupation difference per unit area between the cavity and reservoir due to the transverse momentum
. the transverse momentum of phonons contributes only through the distribution function , and the energy and velocity dependencies of the heat current are only from the longitudinal momentum @xmath177 .
this analysis suggests that the heat current can be treated as a one dimensional calculation with an effective occupation function .
we rewrite the new expression in terms of an integral over longitudinal frequency instead of the momentum space @xmath197 where @xmath198 is the minimum ( maximum ) frequency of the @xmath199 miniband of longitudinal momentum from eq .
( [ characteristic equation ] ) .
the conversion to frequency space introduces the phonon density of state of the superlattice , and we have @xmath200 .
therefore , the group velocity and the density of states cancel out and we have a simple equation of the phonon heat current in eq .
( [ phonon_heat_current_freq ] ) . when we consider for the electron transport of the superlattice , note that only the lowest miniband was considered , because the energy gap of the first and second miniband is larger than the thermal energy @xmath96 , for the typical materials of the superlattice .
so , when the chemical potentials of the reservoir and the cavity stay around the first miniband , energy window @xmath201 of the occupation will guide the transport only through the first miniband . therefore , it is a reasonable approximation to ignore the higher minibands .
in contrast , the bose - einstein distribution function in the frequency domain is broader than the maximum frequency of the acoustic dispersion , and so all of the acoustic phonons contribute to the heat transport .
the total phonon heat current from the heat engine is also obtained from the heat conservation @xmath202 . however ,
different from the heat current by the electrons , the difference of the phonon heat current through the left / right reservoir is only determined by the temperature difference between the cavity and reservoirs which is contained only in the modified occupation difference @xmath203 .
the temperatures of the left / right reservoir are the same in our case , therefore , the phonon heat current conservation leads to the total phonon heat current such as @xmath204 only if there is symmetric heat conductance . in order to calculate the effect of the phonons on the thermodynamic efficiency of this engine
, we consider a low temperature regime , where an analytic investigation can be made , followed by a numerical investigation of the high temperature regime . in the low temperature case
, we assume the temperature only excites low energy modes .
up to a small numerical factor , this permits us to approximate the bose - einstein as a boltzmann distribution , as well as only consider the long wavelength modes , so the dispersion relation may be approximated as , @xmath205 .
computing the effective distribution by integrating over @xmath206 , we find , @xmath207 this allows us to approximate the heat current per unit area from the first mini - band as , @xmath208 \\ & \approx & \frac{16}{(\pi a \hbar \omega_t)^2 \hbar } ( k_b t)^3 \delta t,\end{aligned}\ ] ] where the last limit is in the linear response limit .
this is consistent with a debye treatment of the phonon transport .
this result shows that for low temperatures , the phonons freeze out , and the energy is predominately carried by the electrons . moving on to the high temperature limit , we consider a numerical investigation .
this limit is quite important for room temperature applications in mind .
[ fig_phonon_heatcurrent ] shows the heat currents for gaas / alas superlattices when @xmath127 , as an example .
the phonon heat currents of superlattices and bulk material in panel ( a ) increase with temperature increasing , but saturate at high temperature . moreover , as the superlattice atomic layer increases , the phonon heat current decreases and saturates when the atomic layers are @xmath209 @xcite .
the practical superlattices usually have the atomic layers @xmath209 , so , the practical phonon heat current of the superlattices has about @xmath210 reduction of the heat current from the bulk material around the room temperature .
panel ( b ) compares the electron and phonon heat currents for the optimized parameters to give the maximum output power .
as the temperature increases , the electron heat current increases quadratically while the phonon heat current saturates , so the electron heat current reaches the phonon heat current at high temperature . however , at room temperature , the phonon heat current is still an order of magnitude higher than the electron heat current .
we compare the efficiency by the electron heat current and the total heat current in fig .
[ fig_total_efficiency ] .
the phonon heat current dominates over the electronic one , @xmath211 , when the temperature is around room temperature .
in this situation , we can rewrite the total efficiency @xmath212 the efficiency depends only on the phonon heat current and the generated power .
therefore , we have the maximal efficiency when the power is maximal , and the power and efficiency reach maximal value together . because the phonon heat current is an order of magnitude higher than the electron heat current , the total efficiency decreases about an order of magnitude .
practically , there could be a way to reduce the phonon heat current more than we showed based on the simple theory and example . based on the one dimensional atomic chain model
, we have calculated the phonon heat current in the growth direction and estimated their contributions to the efficiency in the heat engine based on gaas / alas superlattices .
our approach shows that the reduction of the phonon heat current comes from the reduction of the group velocity near the folded brillouin zone edges .
this calculation also assumes the ballistic transport @xmath213 .
however , as the constituent layers become thicker than the phonon mean free path , the phonon transmission function needs to be modified as @xmath214 . in this case
, we also need to treat the phonons as particles and use the theory such as the boltzmann transport equation @xcite .
moreover , other mechanisms such as phonon spectra mismatch and scattering arising from the roughness of the layer interfaces also play an important role to understand the reduction of the experimental results @xcite .
some of high figure of merit thermoelectric materials show reduced lattice thermal conductivity in the superlattice structures , for example si / ge or @xmath215 .
the reduction of phonon heat current can be maximized by the proper choice of superlattice period compared to the mean free path of the phonons : a theory predicts the thermal conductivity minimum as a function of layer spacing @xcite , and some works show the minimum thermal conductivity depending on the superlattice period and a ratio of the layer thickness for the materials @xcite .
therefore , these materials , instead of our example , with optimized superlattice period will give more reduction of phonon heat current of the system .
in this paper , we investigated a heat engine by thermoelectric effects in superlattice structures in three terminal geometry .
first , our work considers the engine as a energy harvester , and shows the advantages of superlattice heat engine in the large output power compared to the other similar geometry due to the box shaped transmission function which comes from the electron energy miniband .
our theory predicts that the maximum power under optimized conditions can be larger than similar resonant tunneling devices , with comparable efficiency at maximum power .
second , a different regime of the system parameters makes the engine works as a refrigerator , and we shows the optimized regime of the parameters for the maximum cooling power and coefficient of performance . in addition , we analyzed the phonon heat current to find the total efficiency by the performance of electrons and phonons together .
the reduction of phonon heat current in the superlattice compared to the corresponding bulk material offers higher total efficiency .
the trade off between power and efficiency is overcome by the reduction of the phonon heat current which can be achieved either by operating at low temperatures , or by engineering the system to have low phonon conductivity , while keeping high electron conductivity .
easy fabrication of these devices with advantages for both the power and efficiency show this heat engine is a promising device for next - generation thermoelectrics .
we would like to thank antonio badolato for suggesting this line of research and for discussions .
we thank jian - hua jiang , and paul ampadu for discussions , and bjrn sothmann and rafa snchez for helpful comments on the manuscript .
this work is dedicated to the memory of markus bttiker , a mentor , colleague , and friend .
99 l. d. hicks and m. s. dresselhaus , phys .
b * 47 * , 12727 ( 1993 ) .
l. d. hicks and m. s. dresselhaus , phys .
b * 47 * , 16631 ( 1993 ) . g. d. mahan and j. o. sofo , proc
. natl acad .
usa * 93 * , 7436 ( 1996 ) . c. w. j. beenakker and a.
a. m. staring , phys .
b * 46 * , 9667 ( 1992 )
. t. e. humphrey , r. newbury , r. p. taylor , and h. linke , phys .
lett . * 89 * , 116801 ( 2002 ) .
b. kubala , j. knig , and j. pekola , phys .
100 * , 066801 ( 2008 ) .
g. billings , a. d. stone , and y. alhassid , phys .
b * 81 * , 205303 ( 2010 ) .
n. nakpathomkun , h. q. xu , and h. linke , phys .
b * 82 * , 235428 ( 2010 )
. a. n. jordan , b. sothmann , r. snchez , and m. bttiker , phys .
b * 87 * , 075312 ( 2013 ) .
d. m. kennes and v. meden , phys .
b * 87 * , 075130 ( 2013 ) .
i. weymann and j. barna , phys .
b * 88 * , 085313 ( 2013 ) .
b. sothmann , r. snchez , and a. n jordan , nanotechnology * 26 * , 032001 ( 2015 ) .
g. benenti , g. casati , t. prosen , and k. saito , arxiv:1311.4430 .
b. sothmann , r. snchez , a. n. jordan , and m. bttiker , phys .
b * 85 * , 205301 ( 2012 ) .
b. sothmann , r. snchez , a. n. jordan , and m. bttiker , new j. phys .
* 15 * , 095021 ( 2013 ) .
a. agarwal and b. muralidharan , appl .
* 105 * , 013104 ( 2014 ) .
b. sothmann , r. snchez , and a. n. jordan , europhys . lett . * 107 * , 47003 ( 2014 ) .
r. snchez , b. sothmann , a. n. jordan , phys .
lett . * 114 * , 146801 ( 2015 ) .
f. mazza , s. valentini , r. bosisio , g. benenti , v. giovannetti , r. fazio , f. taddei , arxiv:1503.01601 .
r. snchez and m. bttiker , phys .
b * 83 * , 085428 ( 2011 ) .
r. snchez , b. sothmann , a. n. jordan , and m. bttiker , new j. phys .
* 15 * , 125001 ( 2013 ) .
jiang , j. appl .
phys . * 116 * , 194303 ( 2014 ) .
f. mazza , r. bosisio , g. benenti , v. giovannetti , r. fazio , and f. taddei , new j. phys .
* 16 * , 085001 ( 2014 ) .
l. henriet , a. n. jordan , and k. le hur , arxiv:1504.02073 ( 2015 ) .
j. r. prance , c. g. smith , j. p. griffiths , s. j. chorley , d. anderson , g. a. c. jones , i. farrer , and d. a. ritchie , phys .
. lett . * 102 * , 146602 ( 2009 ) . h. l. edwards , q. niu , g. a. georgakis , and a. l. de lozanne , phys .
b * 52 * , 5714 ( 1995 ) .
b. roche , p. roulleau , t. jullien , y. jompol , i. farrer , d.a .
ritchie , and d. c. glattli , nat .
commun , * 6 * , 6738 ( 2015 ) . f. hartmann , p. pfeffer , s. hfling , m. kamp , and l. worschech , phys .
lett . * 114 * , 146805 ( 2015 ) .
f. j. disalvo , science * 285 * , 703 ( 1999 ) .
t. c. harman , p. j. taylor , m. p. walsh , and b. e. laforge , science * 297 * , 2229 ( 2002 ) .
l. chena , j. lia , f. suna , and c. wub , appl .
energy * 82 * , 300 ( 2005 ) .
j. yu and h. zhao , j. power sources * 172 * , 428 ( 2007 ) .
l. e. bell , science * 321 * , 1457 ( 2008 ) . c. colvard , t. a. gant , m. v. klein , r. merlin , r. fischer , h. morkoc , and a. c. gossard , phys . rev .
b * 31 * , 2080 ( 1985 ) .
s. tamura , d. c. hurley , and j. p. wolfe , phys .
b * 38 * , 1427 ( 1988 ) . p. hyldgaard and g. d. mahan , phys .
b * 56 * , 10754 ( 1997 ) .
g. chen , phys .
b * 57 * , 14958 ( 1998 ) .
s. tamura , y. tanaka , and h. j. maris , phys .
b * 60 * , 2627 ( 1999 ) .
m. v. simkin and g. d. mahan , phys .
* 84 * , 927 ( 2000 ) .
r. venkatasubramanian , phys .
b * 61 * , 3091 ( 2000 ) .
r. venkatasubramanian , e. siivola , t. colpitts , and b. oquinn , nature * 413 * , 597 ( 2001 ) .
p. a. lebwohl and r. tsu , j. appl .
phys . * 41 * , 2664 ( 1970 ) .
l. esaki and l. l. chang , phys .
* 33 * , 495 ( 1974 ) .
d. l. smith and c. mailhiot , rev .
phys . * 62 * , 173 ( 1990 ) .
a. wacker , phys .
reports * 357 * , 1 ( 2002 ) .
g. chen , ieee trans .
. technol .
* 29 * , 238 ( 2006 ) .
d. g. cahill , w. k. ford , k. e. goodson , g. d. mahan , a. majumdar , h. j. maris , r. merlin , and s. r. phillpot , j. appl .
phys . * 93 * , 793 ( 2003 ) .
m. j. huang , w. y. chong , and t. m. chang , j. appl .
phys . * 99 * , 114318 ( 2006 ) . | we propose a three terminal heat engine based on semiconductor superlattices for energy harvesting .
the periodicity of the superlattice structure creates an energy miniband , giving an energy window for allowed electron transport .
we find that this device delivers a large power , nearly twice than the heat engine based on quantum wells , with a small reduction of efficiency .
this engine also works as a refrigerator in a different regime of the system s parameters .
the thermoelectric performance of the refrigerator is analyzed , including the cooling power and coefficient of performance in the optimized condition .
we also calculate phonon heat current through the system , and explore the reduction of phonon heat current compared to the bulk material .
the direct phonon heat current is negligible at low temperatures , but dominates over the electronic at room temperature and we discuss ways to reduce it . |
transiting extrasolar planets ( teps ) provide unique opportunities to study the physical properties of planetary mass objects outside of the solar system . by combining time - series photometric observations taken during transit with radial velocity ( rv ) measurements of the star ,
it is possible to precisely measure the mass and radius of the planet , if the stellar mass and radius can be determined by other means . the bulk density of the planet
may then be compared with the predictions of theoretical planetary structure models ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) to infer the structure of the planet .
discoveries of planets that fall outside the predicted mass - radius range ( e.g. inflated hot jupiters such as tres-4b ; * ? ? ? * ) lead in turn to refinements of these models .
teps also provide unique opportunities to study planetary atmospheres , including their composition ( e.g. * ? ? ?
* ) , and their thermal profiles ( e.g. * ? ? ?
it is also possible to measure the projection of the angle between the orbital axis and the stellar spin axis for these planets ( e.g. * ? ? ?
* ) , which may be used to test theories of planetary migration @xcite .
to date more than 40 tep discoveries have been published , with the majority coming from dedicated photometric surveys .
these planets span a range covering more than two orders of magnitude in mass from a super - earth tep ( corot-7b ; * ? ? ?
* ) and super - neptunes ( gj 436b and hat - p-11b ; * ? ? ?
* ; * ? ? ?
* ) to brown dwarf size objects ( corot-3b , xo-3b ; * ? ? ?
* ; * ? ? ?
focusing on the low - mass end , we note that the three least massive teps with well determined masses ( gj 436b , hat - p-11b and hd 149026b , * ? ? ?
* ) , are also the three teps with the smallest radii ( excluding corot-3b ) and highest inferred core mass fractions . above this , we begin to see planets with a wide range of radii . the planets wasp-11/hat - p-10b @xcite , wasp-6b @xcite , hat - p-1b @xcite , ogle - tr-111b @xcite , wasp-15b @xcite , and xo-2b @xcite all have radii larger than or comparable to that of jupiter , whereas hat - p-3b @xcite has a radius that is only slightly larger than that of saturn . given the small number of teps known with @xmath17
, we can not yet say what is the empirical minimum mass of core - less , or envelope dominated , gas giant planets . to explore the possible transition from envelope dominated to core dominated planets , it is necessary to find more low - mass teps .
the hungarian - made automated telescope network ( hatnet * ? ? ? * ) survey has been a major contributor to the discovery of teps .
operational since 2003 , it has covered approximately 10% of the northern sky , searching for teps around bright stars ( @xmath18mag ) .
hatnet operates six wide field instruments : four at the fred lawrence whipple observatory ( flwo ) in arizona , and two on the roof of the submillimeter array hangar ( sma ) of sao in hawaii . since 2006
, hatnet has announced and published 11 teps . in this work
we report on the 12th such discovery .
this planet is only the fourth sub - saturn mass tep announced to date , but unlike the other planets , it is of low density , and appears to be h / he dominated .
the structure of the paper is as follows . in
we summarize the observations , including the photometric detection , and follow - up observations . in
we describe the analysis of the data , such as the stellar parameter determination ( ) , blend modeling ( ) , and global modeling of the data ( ) .
we discuss our findings in .
the transits of hat - p-12b were detected with the hat-5 telescope in arizona .
the region around gsc 03033 - 00706 , a field internally labeled as 145 , was observed on a nightly basis between january and july of 2006 , whenever weather conditions permitted .
we gathered 4205 exposures of 5 minutes at a 5.5-minute cadence , of which 2927 images were used in the final light curve of hat - p-12 .
each image contained approximately 10,000 stars down to @xmath19 .
for the brightest stars in the field we achieved a per - image photometric precision of 3mmag .
the calibration of the hatnet frames was done utilizing standard procedures .
the calibrated frames were then subjected to star detection and astrometry , as described in @xcite .
aperture photometry was performed on each image at the stellar centroids derived from the 2mass catalog @xcite and the individual astrometrical solutions .
the resulting light curves were decorrelated against trends using the external parameter decorrelation technique in `` constant '' mode ( epd , see * ? ? ? * ) and the trend filtering algorithm ( tfa , see * ? ? ?
the light curves were searched for periodic box - like signals using the box least squares method ( bls , see * ? ? ? * ) .
we detected a significant signal in the light curve of gsc 03033 - 00706 ( also known as 2mass 13573347 + 4329367 ; @xmath20 , @xmath21 ; j2000 ; @xmath22 , @xcite ) , with a depth of @xmath23mmag , and a period of @xmath24days .
the dip had a relative duration ( first to last contact ) of @xmath25 , corresponding to a total duration of @xmath26 hours ( see fig . [
fig : hatnet ] ) .
all hatnet candidates are subjected to thorough investigation before using more precious time on large telescopes .
one of the important tools for establishing whether the transit - like feature in the light curve of a candidate is due to astrophysical phenomena other than a planet transiting a star is the cfa digital speedometer ( ds ; * ? ? ?
* ) , mounted on the telescope .
this yields high - resolution ( @xmath27 ) spectra with low signal - to - noise ratio sufficient to derive radial velocities with moderate precision ( roughly 0.5 - 1@xmath28 ) , and to determine the effective temperature and surface gravity of the host star . with this facility
we are able to reject many types of false positives , such as f dwarfs orbited by m dwarfs , grazing eclipsing binaries , triple and quadruple star systems , or giant stars where the transit signal can not be due to a planet .
we obtained @xmath29 observations of hat - p-12 with the ds .
the rv measurements of hat - p-12 showed an rms residual of 0.43@xmath28 , consistent with no detectable rv variation .
the spectra were single - lined , showing no detectable evidence for more than one star in the system .
atmospheric parameters for the star , including the effective temperature @xmath30 , surface gravity @xmath31 ( log cgs ) , and projected rotational velocity @xmath32 consistent with zero with an asymmetric error of about 1@xmath28 , were derived as described by @xcite .
the effective temperature and surface gravity correspond to a mid - k dwarf .
the mean heliocentric radial velocity of hat - p-12 is @xmath33@xmath28 .
given the significant transit detection by hatnet , and the positive ds results that exclude obvious false positives , we proceeded with the follow - up of this candidate by obtaining high - resolution and high s / n spectra to characterize the radial velocity variations and to determine the stellar parameters with higher precision . using the hires instrument @xcite on the keck i telescope located on mauna kea , hawaii , we obtained 22 exposures with an iodine cell , plus one iodine - free template .
the observations were made on 16 nights during a number of observing runs between 2007 march 27 and 2008 september 17 .
[ ht ] the width of the spectrometer slit used on hires was @xmath34 , resulting in a resolving power of @xmath35 , with a wavelength coverage of @xmath36 .
the iodine gas absorption cell was used to superimpose a dense forest of @xmath37 lines on the stellar spectrum and establish an accurate wavelength fiducial ( see * ? ? ?
relative rvs in the solar system barycentric frame were derived as described by @xcite , incorporating full modeling of the spatial and temporal variations of the instrumental profile .
the final rv data and their errors are listed in tab .
[ tab : rvs ] .
the folded data , with our best fit ( see ) superimposed , are plotted in fig .
[ fig : rvbis ] . in fig .
[ fig : rvbis ] we also plot the relative s index .
this index is computed following the prescription given by @xcite after matching each spectrum to a reference spectrum using a transformation that includes a wavelength shift and a flux scaling that is a polynomial as a function of wavelength .
the transformation is determined on regions of the spectra that are not used in computing the s index .
note that the relative s index has not been calibrated to the scale of @xcite .
the relative s index does not show any significant variation correlated with the orbital phase ; such a correlation might have indicated that the rv variations are due to stellar activity .
@xmath38 & @xmath39 & @xmath40 & @xmath41 & @xmath42 & @xmath43 & @xmath44 + @xmath45 & @xmath46 & @xmath47 & @xmath48 & @xmath49 & @xmath50 & @xmath51 + @xmath52 & & & @xmath53 & @xmath54 & @xmath55 & @xmath51 + @xmath56 & @xmath57 & @xmath58 & @xmath59 & @xmath60 & @xmath50 & @xmath51 + @xmath61 & @xmath62 & @xmath63 & @xmath64 & @xmath65 & @xmath66 & @xmath67 + @xmath68 & @xmath69 & @xmath70 & @xmath71 & @xmath60 & @xmath72 & @xmath51 + @xmath73 & @xmath74 & @xmath75 & @xmath76 & @xmath77 & @xmath50 & @xmath44 + @xmath78 & @xmath79 & @xmath80 & @xmath81 & @xmath82 & @xmath83 & @xmath84 + @xmath85 & @xmath86 & @xmath87 & @xmath88 & @xmath89 & @xmath72 & @xmath90 + @xmath91 & @xmath92 & @xmath93 & @xmath94 & @xmath95 & @xmath96 & @xmath97 + @xmath98 & @xmath99 & @xmath100 & @xmath101 & @xmath102 & @xmath103 & @xmath104 + @xmath105 & @xmath106 & @xmath107 & @xmath108 & @xmath109 & @xmath50 & @xmath97 + @xmath110 & @xmath111 & @xmath75 & @xmath112 & @xmath113 & @xmath72 & @xmath44 + @xmath114 & @xmath115 & @xmath116 & @xmath117 & @xmath118 & @xmath66 & @xmath51 + @xmath119 & @xmath120 & @xmath121 & @xmath122 & @xmath123 & @xmath124 & @xmath125 + @xmath126 & @xmath127 & @xmath128 & @xmath129 & @xmath130 & @xmath131 & @xmath44 + @xmath132 & @xmath133 & @xmath134 & @xmath135 & @xmath136 & @xmath103 & @xmath125 + @xmath137 & @xmath138 & @xmath139 & @xmath140 & @xmath141 & @xmath103 & @xmath142 + @xmath143 & @xmath144 & @xmath145 & @xmath146 & @xmath147 & @xmath131 & @xmath148 + @xmath149 & @xmath150 & @xmath139 & @xmath151 & @xmath152 & @xmath96 & @xmath90 + @xmath153 & @xmath154 & @xmath155 & @xmath156 & @xmath157 & @xmath124 & @xmath90 + @xmath158 & @xmath159 & @xmath160 & @xmath161 & @xmath162 & @xmath72 & @xmath44 + to confirm the transit signal and obtain high - precision light curves for modeling the system we conducted photometric follow - up observations with the keplercam ccd on the telescope .
we observed four transit events of hat - p-12b on the nights of 2007 march 27 , 2007 april 25 , 2009 february 5 and 2009 march 6 ( fig .
[ fig : lc ] ) . on 2007
march 27 , 151 frames were acquired with a cadence of 90 seconds ( 75 seconds of exposure time ) in the sloan @xmath163-band ; the observations were interrupted at mid - transit due to clouds . on 25 april 2007 ,
372 frames were acquired with a cadence of 45 seconds ( 30 seconds of exposure time ) in the sloan @xmath164-band .
on 5 february 2009 , 218 frames were obtained with a cadence of 70 seconds ( 30 seconds exposure time ) in @xmath164-band .
finally , on 6 march 2009 , 213 frames were acquired with a cadence of 70 seconds ( 60 seconds exposure time ) in the sloan @xmath165-band .
this follow - up @xmath165-band light curve was obtained to further constrain possible blend scenarios ( ) .
the reduction of the images was performed as follows .
after bias and flat - field calibration , we derived an initial second - order astrometric transformation between the @xmath166 brightest stars and the 2mass catalog , as described in @xcite , yielding a residual of @xmath167 pixels .
the primary reason for precise astrometry is to minimize the photometric errors that would originate from the centroid errors for the individual stars on each frame .
aperture photometry was then performed on the resulting fixed positions , using a series of apertures .
the instrumental magnitude transformation was done in two steps : first , all magnitude values were transformed to the photometric reference frame ( selected to be the sharpest image ) , using the individual poisson noise error estimates as weights . in the second step
, the magnitude fit was repeated using the mean individual light curve magnitudes as reference and the rms of these light curves as weights . in both of the magnitude transformations , we excluded from the fit the target star itself and the @xmath168@xmath169 outliers .
we performed epd and tfa against trends simultaneously with the light curve modeling ( for more details , see and @xcite ) . from the series of apertures , for each night , we chose the one yielding the smallest fit rms for the light curve .
this aperture conveniently fell in the middle of the aperture series .
the final light curves are shown in the upper plots of fig .
[ fig : lc ] , with our best fit transit light curve models superimposed ( see also ) .
lrrrr @xmath170 & @xmath171 & @xmath172 & @xmath173 & i + @xmath174 & @xmath172 & @xmath172 & @xmath175 & i + @xmath176 & @xmath177 & @xmath178 & @xmath179 & i + @xmath180 & @xmath181 & @xmath182 & @xmath183 & i + @xmath184 & @xmath185 & @xmath172 & @xmath186 & i + @xmath187 & @xmath188 & @xmath178 & @xmath189 & i + @xmath190 & @xmath191 & @xmath172 & @xmath192 & i + @xmath193 & @xmath194 & @xmath172 & @xmath195 & i + @xmath196 & @xmath197 & @xmath182 & @xmath198 & i + @xmath199 & @xmath200 & @xmath182 & @xmath201 & i +
we derived the initial stellar atmospheric parameters by using the template spectrum obtained with the keck / hires instrument .
we used the sme package of @xcite along with the atomic - line database of @xcite , which yielded the following _ initial _ values and uncertainties ( which we have conservatively increased to include our estimates of the systematic errors ) : effective temperature @xmath202k , stellar surface gravity @xmath203(cgs ) , metallicity @xmath204}}={\ensuremath{-0.29\pm0.05}}$]dex , and projected rotational velocity @xmath205 . at this stage
we could use the effective temperature and the surface gravity as a luminosity indicator , and determine the stellar parameters based on these two constraints using a set of isochrones .
however , the effect of @xmath206 on the spectral line shapes is typically subtle and as a result it is generally a rather poor luminosity indicator in practice . for planetary transits , the @xmath207 normalized semi - major axis and related @xmath208 mean
stellar density typically impose a stronger constraint on possible stellar models @xcite .
the validity of our assumption , namely that the adequate physical model describing our data is a planetary transit ( as opposed to a blend ) , is shown later in .
using the values of @xmath209 , @xmath210 $ ] , and @xmath206 from the sme analysis , and corresponding quadratic limb darkening coefficients ( @xmath211 , @xmath212 , etc . ) from @xcite , we performed a global modeling of the data ( ) , yielding a full monte - carlo distribution of @xmath213 .
this was complemented by a monte - carlo distribution of @xmath209 and @xmath210 $ ] , obtained by assuming gaussian uncertainties based on the 1-@xmath169 error bars of the initial sme analysis .
for each combination within the large ( @xmath214 ) set of @xmath213 , @xmath209 , and @xmath210 $ ] values , we searched the stellar isochrones of the @xcite models for the best fit stellar model parameters ( such as @xmath215 , @xmath216 , @xmath206 , etc ) .
we interpolated these isochrones to the sme - based stellar metallicity of @xmath204}}= { \ifthenelse{\equal{{i}}{i}}{{\ensuremath{-0.29\pm0.05}}}{{\ensuremath{-0.36\pm0.04}}}}$ ] .
the majority of the parameter combinations in the monte carlo search did not match any isochrone .
in such cases ( @xmath166% of all trials ) we skipped to the next randomly drawn parameter set . at the end we derived the mean values and uncertainties of the physical parameters based on their _ a posteriori _ distribution .
we note that the spread of the input stellar parameters ( based on the gaussian uncertainties ) was large compared to what the isochrones cover as a function of age , due to the very slow evolution of k dwarfs ( see fig .
[ fig : iso ] ) .
this is partly the reason for the 40% match ratio .
we also note that the match ratio is very sensitive to changing fundamental parameters of the isochrones , such as the mixing length or the metallicity .
we then repeated the sme analysis by fixing @xmath206 to the refined value of @xmath217 based on the isochrone search , and only adjusting @xmath209 , @xmath210 $ ] and @xmath218 .
this second iteration yielded @xmath219k , @xmath204}}= { \ensuremath{-0.36\pm0.04}}$ ] and @xmath220@xmath28 .
curiously , the new @xmath209 and @xmath210 $ ] values from this second iteration provide a somewhat inferior match with the @xcite isochrones , as compared to the initial match .
possible reasons for this include i ) systematic errors in the sme analysis due to the relatively low snr of our keck spectra , ii ) increasing uncertainty in the sme analysis due to the late stellar type of the host star ( note that hat - p-12 has a temperature that is below the 4700 k cut - off for stars included in the analysis of @xcite ) , iii ) general uncertainty in the isochrones for mid - k dwarfs ( there is a well - known discrepancy between the observed and predicted mass - radius relation for k and m dwarf stars in double - lined eclipsing binaries such that the observed radii are larger than the predicted radii , though there is some evidence that this discrepancy only holds for rapidly rotating , active stars , e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
thus we accepted the initial values of @xmath209 , @xmath210 $ ] and @xmath218 as the final atmospheric parameters for this star , along with the isochrone based stellar parameters , yielding @xmath215=@xmath221@xmath5 , @xmath216=@xmath222@xmath7 and @xmath223=@xmath224@xmath225 .
along with other stellar parameters , these are summarized in tab .
[ tab : stellar ] .
the stellar evolutionary isochrones from @xcite for metallicity @xmath210$]=@xmath226 are plotted in fig .
[ fig : iso ] , with the final choice of effective temperature @xmath227 and @xmath213 marked , and encircled by the 1-@xmath169 and 2-@xmath169 confidence ellipsoids .
the stellar evolution modeling also yields the absolute magnitudes and colors in various photometric passbands .
we used the apparent magnitudes from the 2mass catalogue @xcite to determine the distance of the system .
the magnitudes reported in the 2mass catalogue have to be converted to the cit system @xcite , in which the stellar evolution models specify the colors .
the reported magnitudes for this star are @xmath228 , @xmath229 and @xmath230 ; which are equivalent to @xmath231 , @xmath232 and @xmath233 ; in the cit photometric system ( see * ? ? ?
thus , the converted 2mass magnitudes yield a color of @xmath234 that is within 1-@xmath169 of the expected , isochrone - based @xmath235 .
we thus relied on the 2mass @xmath236 apparent magnitude and the @xmath237 absolute magnitude derived from the above - mentioned modelling to determine the distance : @xmath238pc .
the @xmath236-band was chosen because it is the longest wavelength band - pass with the smallest expected discrepancies due to molecular lines in the spectrum of this k4 dwarf .
lcl @xmath227 ( k ) & & sme + @xmath204}}$ ] & & sme + @xmath32 ( @xmath28 ) & & sme + @xmath239 ( @xmath28 ) & & sme + @xmath240 ( @xmath28 ) & & sme + @xmath241 ( @xmath28 ) & @xmath33 & ds + @xmath211 & @xmath242 & sme+claret + @xmath212 & @xmath243 & sme+claret + @xmath244 & @xmath245 & sme+claret + @xmath246 & @xmath247 & sme+claret + @xmath248 & @xmath249 & sme+claret + @xmath250 & @xmath251 & sme+claret + @xmath252 ( @xmath253 ) & @xmath221 & baraffe+@xmath213+sme + @xmath254 ( @xmath255 ) & @xmath222 & baraffe+@xmath213+sme + @xmath256 ( cgs ) & @xmath257 & baraffe+@xmath213+sme + @xmath258 ( @xmath259 ) & @xmath224 & baraffe+@xmath213+sme + @xmath260 ( mag ) & 12.84 & tass + @xmath261 ( mag ) & @xmath262 & baraffe+@xmath213+sme + @xmath236 ( mag , cit ) & @xmath263 & 2mass+carpenter + @xmath264 ( mag , cit ) & @xmath265 & baraffe+@xmath213+sme + age ( gyr ) & @xmath266 & baraffe+@xmath213+sme + distance ( pc ) & @xmath238 & baraffe+@xmath213+sme + following @xcite , we explored the possibility that the measured radial velocities are not real , but are instead caused by distortions in the spectral line profiles due to contamination from a nearby unresolved eclipsing binary . a bisector analysis based on the keck spectra
was done as described in 5 of @xcite .
[ fig : rvbis ] shows the bisector spans ( bs ) phased with the orbital period of the planet . in calculating the bs we use the convention @xmath267 where @xmath268 is the velocity of the bisector of the cross - correlation profile at a low cross - correlation value , and
@xmath269 is the velocity at a high cross - correlation value .
while the bs do not show significant variations with an amplitude that is comparable to or greater than the rv variations , there does appear to be a correlation between the bs and rv measurements . applying a spearman rank - order correlation test
, we find that the two variables are correlated with 98% confidence ( i.e. there is a @xmath270 false alarm probability ) .
since such a correlation might indicate a blend scenario , we consider below , and rule out , the possibility that the system is a blend between a bright foreground k star and a background eclipsing binary ( ) or a hierarchical triple system ( ) .
the scenarios that we consider are summarized in table [ tab : blendtypes ] .
we then consider the possibility that the correlation is not astrophysical , but rather results from variations in the sky contamination of the spectra ( ) ; we conclude that this is the most likely explanation for the correlation .
the high proper motion of hat - p-12 allows us to rule out one possible scenario that could potentially fit the available observations , namely a background eclipsing binary that is aligned , by chance , with the foreground k4 dwarf hat - p-12 ( we refer to this as the h , b(s - s ) model , where h stands for the foreground star hat - p-12 , and the b(s - s ) is a background eclipsing binary star system ; here b refers to the fact that the eclipsing binary is in the background rather than being associated with the star h ) . to reproduce the observed @xmath271 deep transit the background object can not be more than 4 mag fainter than hat - p-12 ( objects fainter than this would contribute less than 2.5% of the total combined light and so could not cause the transit even if they were to be completely eclipsed by an object that emits no light ) . because hat - p-12 has a high proper motion ( @xmath272@xmath273 ; *
* ) it is possible to use the palomar observatory sky survey plates from 1955 ( poss - i , red and blue plates ) to view the sky at the current position of hat - p-12 ( this same technique was used for hat - p-11 , * ? ? ?
* ) . between 1955 and the follow - up observations in 2009 , hat - p-12 has moved @xmath274 .
[ fig : poss ] shows an image stamp from the poss - i plate compared with a recent observation from the .
we can rule out a background object down to @xmath275 mag within @xmath276 of the current position of hat - p-12 .
any background object must be @xmath277 mag fainter than hat - p-12 and thus could not be responsible for the observed transit .
following @xcite we consider the possibility that hat - p-12 is a hierarchical triple system , consisting of two eclipsing bodies that are diluted by a third star . in the following we refer to the bright star , with properties determined from the sme analysis , as hat - p-12 .
we consider three scenarios . in the first scenario
we assume that the bright star hat - p-12 is uneclipsed , and that the two eclipsing components are stars with parameters constrained by common origin to fall on the same age / metallicity isochrone as hat - p-12 ( we refer to this model as the h , s - s model , where h denotes the bright star hat - p-12 and s - s denotes a physically associated eclipsing binary consisting of a brighter star s and a fainter star s ) . in the second scenario
we assume that hat - p-12 is uneclipsed , that one of the eclipsing components is a fainter star and that the other component is a planet with negligible mass and luminosity compared to the star ( the h , s - p model , where s stands for the fainter star , and p is the planet ) . in the third scenario
we assume that hat - p-12 is a star that is transited by a planet and that there is a fainter star diluting the observed transit ( the h - p , s model ) .
these models will be compared to the fiducial model of a single star orbited by a planet ( the h - p model ) .
h , b(s - s ) & eclipsing binary star diluted by unresolved , unrelated star & proper motion ( ) + h , s - s & hierarchical triple star system , two components are eclipsing & light curve fit ( ) + h , s - p & binary star system , fainter star has a transiting planet & light curve fit ( ) + h - p , s & binary star system , brighter star has a transiting planet & light curve fit ( ) + h - p & single star with a transiting planet & not excluded + for the h , s - s and h , s - p scenarios we fit the follow - up @xmath164- , @xmath163- , and @xmath165-band light curves together with the hatnet @xmath278-band light curve following the procedure described by @xcite .
we include the hatnet light curve to constrain the possibility of a secondary eclipse ; to exclude points that do not contribute to the fit , we only include points that are within 1 transit duration of the start of transit ingress or end of transit egress , or within 1 transit duration of the start of secondary ingress or end of secondary egress .
we use the tfa hatnet light curve and apply epd on the out - of - transit portion of the follow - up light curves .
we scale the formal photometric errors on each light curve so that @xmath279 for the out - of - transit portion of the light curve .
we take the magnitudes and radii of the stars from the @xcite isochrones , transforming the @xmath280 magnitudes to the sloan system using the relations from @xcite .
to make a fair comparison between the blend models and the h - p model we also fit the h - p model to the light curves using the same procedure used to fit the blend - models ( see for a more detailed analysis of the h - p model used for the final parameter determinations ) .
[ fig : blend - ht - lcs ] compares the best fit h , s - s , h , s - p and h - p models .
the best fit h , s - s model , consisting of an eclipsing pair with masses @xmath281 and @xmath282 that is diluted by the star hat - p-12 ( @xmath283 ) , has @xmath284 with 1333 degrees of freedom .
we compare this to the best fit h - p model which has @xmath285 with 1334 degrees of freedom . because the photometric noise appears to be temporally correlated ( fig .
[ fig : blend - ht - lcs ] ) , formal estimates for the significance of @xmath286 between two models will overestimate the confidence with which one model can be rejected in favor of another .
we therefore conduct monte carlo simulations to estimate the expected distribution of @xmath286 values under the assumption that the h - s , s model is correct , and accounting for temporal correlations in the noise .
to generate light curves for the monte carlo simulations that have similar time - correlated noise as the real light curves we fourier transform the residual of each light curve from the best - fit h , s - s model , randomize the phases , inverse fourier transform it , and then add in the h , s - s model .
this method forces the simulated light curves to have the same noise power - spectrum ( and hence auto - correlation function ) as the actual light curve residuals .
we scale the errors of each simulated light curve to have @xmath287 in the out - of - transit portion of the light curve .
the fourier transforms are carried out assuming uniform time - sampling ( a good approximation for the follow - up light curves ) ; to use the fast - fourier transform algorithm , we cyclically repeat each light curve so that the total number of points in the light curve is a power of 2 .
we then fit the h - p and h , s - s models to the simulated sets of light curves and record @xmath288 for each simulation . from 1000 simulations
we find a median value of @xmath289 with a standard deviation of 12.4 ; we find no instances where @xmath290 , the minimum value attained is @xmath291 . we conclude that , based on the light curves , the h , s - s model can be rejected in favor of the h - p model at the @xmath292 confidence level . for the h , s - p scenario
, we find that the best - fit model consists of a star with mass @xmath293 transited by a planet with @xmath294 and diluted by the star hat - p-12 ( @xmath283 ) .
this model has @xmath295 with 1333 degrees of freedom . to determine the significance of @xmath296 we repeat the monte carlo simulations , this time adopting the best - fit h , s - p model as the fiducial model . from 1000 simulations
we find a median value of @xmath297 with a standard deviation of 13.9 .
there are 4 simulations with @xmath298 , so we conclude that the h - p model is preferred over the h , s - p model with @xmath299 ( @xmath300 ) confidence .
as described in it is possible to correct for systematic errors in the photometry by simultaneously applying epd and tfa to the light curves while fitting a physical model to them . by using a more sophisticated model of this form we are able to rule out the h , s - p model with higher confidence , and
also rule out the h - p , s model .
we perform the global modeling as described in incorporating three additional parameters that allow for dilution in the @xmath165- , @xmath163- and @xmath164-bands .
this model effectively encompasses both the h - p , s and h , s - p models because the only h , s - p models that provide a reasonable fit to the light curve are models where the planet - bearing star has a mass that is nearly equal to the mass of the diluting star hat - p-12 .
we allow the dilution factors to vary independently in the fit .
we find that models with no dilution are strongly preferred , and place 1@xmath169 upper limits on the light contribution in each filter from an uneclipsed star ( `` third light '' ) of @xmath301 , @xmath302 and @xmath303 .
any additional star thus makes a negligible contribution to the total light of the system .
this test thus rules out both the h - p , s and h , s - p models . as shown in sections [ sec : hitrip ] and
[ sec : bgeb ] , blend scenarios involving an eclipsing binary star system are inconsistent with other observations of the system .
we therefore look for an explanation of the apparent bs - rv correlation shown in that does not invoke a blend .
one possibility is that it is due to varying contamination from the sky spectrum . because hat - p-12 is relatively faint ,
the flux from the sky is non - negligible compared to the flux from the source .
scattered light from the moon illuminating the sky near hat - p-12 has a solar - like spectrum which yields a peak near @xmath304 in the cross - correlation profile of the observed spectrum . the degree to which this second peak contaminates the peak from the star varies with the sky brightness and the radial velocity difference between the moon and the star .
because the observer - centric velocity of hat - p-12 is always less than the velocity of the moon , an increase in the sky brightness or a decrease in the velocity difference will lead to a positive bs variation for our adopted bs sign convention . while the sky brightness is not directly measureable from the available data , in order to quantify this effect we may introduce a sky contamination factor ( scf ) given by @xmath305 where @xmath278 is the ratio of the flux in the spectrum due to the moon to the flux due to the star , @xmath306 is the observer - centric radial velocity difference between the moon and the star , @xmath307 is the width of the lorentzian function that best fits the mean cross - correlation profile , and the form for the denominator is chosen because the cross - correlation profile is well fit by a lorentzian function .
we estimate @xmath278 via the relation @xmath308 where @xmath309 is the total flux received in the region of the spectrum used to compute the bs , @xmath310 is the exposure time , @xmath311 and @xmath312 are the values for the spectrum with the highest count - rate ( these are used to account for changes in the flux received from the star due to variations in the seeing or transparency ) , @xmath313 is the @xmath314-band magnitude of the star ( we take @xmath315 assuming @xmath316 for a dwarf star with @xmath317k ) , and @xmath318 is the effective magnitude of the sky due to the moon at the position of the star ( in an area of @xmath319 square arcseconds ) . to estimate @xmath318 we use the model for the sky brightness due to moonlight given by @xcite ,
extending it to the @xmath314 band by taking @xmath320 for the moon ( e.g. * ? ? ?
* ) and @xmath321 for the extinction coefficient ( this is a typical value at the summit of mauna kea for the johnson @xmath314-band which is roughly the region of the spectrum used to calculate the bs , * ? ? ?
the values for @xmath318 range from @xmath322 to @xmath323 .
when the moon is below the horizon we take @xmath324 .
[ fig : scf ] compares the scf to the bs values and to the orbital phase .
note that we normalize the scf to have a mean value of 1.0 .
there is a positive correlation between the scf and the bs . by chance , spectra taken between orbital phases 0.5 and 1 had higher sky contamination on average than those taken between orbital phases 0 and 0.5 . when points with @xmath325 are removed , the correlation between the remaining bs and rv values is no longer significant ( the correlation significance is 37% ) . we conclude that this a plausible explanation for the apparent bs - rv correlation . as a further test on this hypothesis we simulate sky contaminated spectra and measure the bs values using the same procedure as for the actual spectra . to simulate a spectrum we take @xmath326 where @xmath310 is the iodine - free template spectrum of hat - p-12 ,
@xmath327 is the iodine - free template spectrum of hat - p-13 scaled to have the same total flux though the @xmath314-band as @xmath310 ( hat - p-13 has @xmath328 k , and is thus a better approximation to a solar spectrum than hat - p-12 ; * ? ? ?
* ) , @xmath329 and @xmath330 are the barycentric velocity corrections for the templates , @xmath331 is the barycentric velocity correction for spectrum @xmath163 , @xmath332 is the average radial velocity of hat - p-13 , @xmath278 is given by eq .
[ eqn : moonbrightness ] , and @xmath333 is a function that redshifts the spectrum @xmath334 by velocity @xmath335 . figure [ fig : scfsim ] compares the scf to the bs for the simulated spectra .
the simulations show a correlation between the scf and bs that is comparable to that seen in fig . [ fig : scf](a ) .
this confirms that sky contamination may affect the bs values at the level that is observed . our model for the follow - up light curves used analytic formulae based on @xcite for the eclipse of a star by a planet , where the stellar flux is described by
quadratic limb - darkening .
the limb darkening coefficients were derived from the sme results ( ) , using the tables provided by @xcite for @xmath164- , @xmath163- , and @xmath165-bands .
the transit shape was parametrized by the normalized planetary radius @xmath336 , the square of the impact parameter @xmath337 , and the reciprocal of the half duration of the transit @xmath338 .
we chose these parameters because of their simple geometric meanings and the fact that these show negligible correlations ( see * ? ? ? * ) .
our model for the hatnet data was the simplified `` p1p3 '' version of the @xcite analytic functions , for the reasons described in @xcite . following the formalism presented by @xcite ,
the rv curve was parametrized by an eccentric keplerian orbit with semi - amplitude @xmath236 , and lagrangian orbital elements @xmath339 .
we assumed that there is a strict periodicity in the individual transit times . in practice
, we assigned the transit number @xmath340 to the first high quality follow - up light curve gathered on 2007 march 27 .
the adjusted parameters in the fit were the first transit center observed by hatnet , @xmath341 , and the last transit center observed by the telescope , @xmath342 , covering all of our measurements with the hatnet telescopes , and the telescope .
we prefer using @xmath341 and @xmath342 as adjusted parameters rather than the period and epoch for the reasons discussed by @xcite and @xcite .
the transit center times for the intermediate transits were interpolated using these two epochs and the @xmath343 transit number of the actual event .
the model for the rv data contains the ephemeris information through the @xmath341 and @xmath342 variables @xcite .
altogether , the 11 parameters describing the physical model were @xmath341 , @xmath342 , @xmath344 , @xmath337 , @xmath338 , @xmath236 , @xmath345 , @xmath346 , and three additional ones related to the instrumental configuration .
these are the instrumental blend factor @xmath347 of hatnet which accounts for possible dilution of the transit in the hatnet light curve , the hatnet out - of - transit magnitude , @xmath348 , and the relative rv zero - point @xmath349 .
we extended our physical model with an instrumental model that describes the systematic variations of the data .
this was done in a similar fashion to the analysis presented in @xcite .
basically , the hatnet photometry has been already epd- and tfa - corrected before the global modeling , so we only considered systematic corrections to the follow - up light curves .
we chose the `` eltg '' method , i.e. epd was performed in `` local '' mode with epd coefficients defined for each night , and tfa was performed in `` global '' mode using the same set of stars and tfa coefficients for all nights .
the underlying physical model was based on the @xcite analytic formulae , as described earlier .
the five epd parameters were the hour angle ( characterizing a monotonic trend that changes linearly over time ) , the square of the hour angle , and the stellar profile parameters ( equivalent to fwhm , elongation , position angle ) .
the exact functional form of the above parameters contained 6 coefficients , including the auxiliary out - of - transit magnitude of the individual events .
the epd parameters were independent for all 4 nights , implying 24 additional coefficients in the global fit .
for the global tfa analysis we chose 18 template stars that had good quality measurements for all nights and on all frames , implying an additional 18 parameters in the fit .
we apply epd to the template star light curves using the same set of parameters as used for the hat - p-12 light curves before incoporating them in the analysis .
thus , the total number of fitted parameters is 11 ( physical model ) + 24 ( local epd ) + 18 ( global tfa ) = 53 , i.e. much smaller than the number of data - points ( @xmath350 ) .
the joint fit was performed as described in @xcite .
we minimized @xmath351 in the parameter space by using a hybrid algorithm , combining the downhill simplex method ( amoeba , see * ? ? ?
* ) with the classical linear least squares algorithm .
uncertainties on the parameters were derived using the markov chain monte - carlo method ( mcmc , see * ? ? ?
* ) using `` hyperplane - clls '' chains @xcite .
the _ a priori _ distributions of the parameters for these chains were chosen from a generic gaussian distribution , with eigenvalues and eigenvectors derived from the fisher covariance matrix for the best fit value .
the fisher covariance matrix is calculated analytically using the partial derivatives given by @xcite and @xcite . since the eccentricity of the system appeared as insignificant ( @xmath352 , @xmath353 ) , we repeated the global fit by fixing these to 0 .
the best fit results for the relevant physical parameters are summarized in tab .
[ tab : planetparam ] . tab .
[ tab : planetparam ] also lists the rv `` jitter '' , which is a component of assumed astrophysical noise intrinsic to the star that we add in quadrature to the rv measurement uncertainties in order to have @xmath354 from the rv data for the global fit .
in addition , some auxiliary parameters ( not listed in the table ) were : @xmath355 ( bjd ) , @xmath356 ( bjd ) , @xmath357@xmath358(for the keck rvs , note that this does _ not _ correspond to the true center of mass radial velocity of the system , but is only a relative offset ) .
the planetary parameters and their uncertainties can be derived by the direct combination of the _ a posteriori _ distributions of the light curve , radial velocity and stellar parameters .
we found that the mass of the planet is @xmath359 , the radius is @xmath360 and its density is @xmath361@xmath14 .
the final planetary parameters are summarized at the bottom of table [ tab : planetparam ] .
lc @xmath362 ( days ) & @xmath363 + @xmath364 ( @xmath365 ) & @xmath366 + @xmath367 ( days ) & @xmath368 + @xmath369 ( days ) & @xmath370 + @xmath207 & @xmath371 + @xmath338 & @xmath372 + @xmath344 & @xmath373 + @xmath337 & @xmath374 + @xmath375 & @xmath376 + @xmath163 ( deg ) & @xmath377 + @xmath236 ( @xmath358 ) & @xmath378 + @xmath379 & @xmath380 + @xmath381 & @xmath380 + @xmath382 & @xmath380 + rv jitter ( @xmath358 ) & @xmath383 + @xmath384 ( @xmath385 ) & @xmath386 + @xmath387 ( @xmath388 ) & @xmath389 + @xmath390 & @xmath391 + @xmath392 ( @xmath14 ) & @xmath393 + @xmath394 ( au ) & @xmath395 + @xmath396 ( cgs ) & @xmath397 + @xmath398 ( k ) & @xmath399 + @xmath400 & @xmath401 + @xmath402 ( @xmath403 ) & @xmath404 +
comparing hat - p-12b to the theoretical models of @xcite , we find that the mass and radius of the planet are consistent with the 1.0 gyr , @xmath405 non - irradiated model , or with a 1.0 - 5.0 gyr , @xmath406 irradiated model ( fig .
[ fig : exomr ] ) . with an equilibrium temperature of @xmath407k ,
hat - p-12b has an equivalent solar semi - major axis of @xmath408 au , so the irradiation received by the planet , while not insignificant , is less than what is used to calculate the irradiated model ( @xmath409 au ) .
the inferred metal fraction is expected to be closer to @xmath410 than @xmath411 for the @xcite models if the correct irradiation were used .
we have also compared hat - p-12b to the theoretical models of @xcite . in fig .
[ fig : exomr ] we have interpolated these models to @xmath412 au , and find that the mass and radius of hat - p-12b are consistent with a 10@xmath413 core , 1 gyr model , and lie between the core - less and 10@xmath413 core , 4.5 gyr models .
we conclude , therefore , that hat - p-12b is most likely a h / he dominated planet with a core of perhaps @xmath414 , and a total metal fraction of @xmath415% .
this makes hat - p-12b the least massive h / he dominated gas giant planet found to date ; the previous record holder was saturn .
it is interesting to compare the properties of hat - p-12b to those of saturn and hd 149026b , the two planets with known radii that have masses closest to that of hat - p-12b .
measurements of the mass of hd 149026b range from @xmath416 to @xmath417 , while determinations of its radius range from @xmath418 to @xmath419 @xcite . the planet appears to have a significant core , with estimates ranging from @xmath420 to @xmath421 ( see * ? ? ?
* and references therein ) , implying a high metal fraction of @xmath422 .
saturn has a mass of @xmath423 @xcite , equatorial radius of @xmath424 @xcite , an estimated core mass of @xmath425 , and a total heavy element fraction of @xmath426 @xcite .
although hat - p-12b is less massive than both hd 149026b and saturn , it has a larger radius than both planets .
note that hat - p-12b does not have a detectable eccentricity , so its large radius may not be due to tidal heating ( in the models by * ? ?
* however , close - in planets may have tidally inflated radii even with eccentricities @xmath427 ) .
the large radius in comparison with saturn may be due in part to the enhanced irradiation received by hat - p-12b , and to hat - p-12b potentially having a smaller core mass than saturn .
hd 149026b , on the other hand , receives more irradiation than hat - p-12b ( @xmath428 au using the parameters from * ? ? ?
* ) , so the difference in radii suggests that hat - p-12b has a substantially smaller core mass and metal enhancement than hd 149026b .
it is interesting to note that the inferred core mass of the three planets appears to correlate with the host star metallicity ( hat - p-12 has @xmath204}}= { \ensuremath{-0.29\pm0.05}}$ ] , the sun has @xmath204}}= 0 $ ] , and hd 149026 has @xmath204}}= 0.36 \pm 0.05 $ ] from * ? ? ?
this correlation has been previously noted by @xcite and by @xcite , and is perhaps suggestive evidence for the core accretion model of planet formation ( e.g. * ? ? ? * and references therein ) .
further discoveries of teps with masses comparable to or less than that of saturn are needed to determine whether or not this correlation holds .
note from fig .
[ fig : exomr ] that the radii of planets in this mass regime are more sensitive to the core mass than are the radii of more massive planets for which a given core mass is a smaller fraction of the total planet mass .
finally , one might wonder why other planets like hat - p-12b have not been found to date ( see fig . [
fig : exomr ] ) .
the significant @xmath271 transit depth of hat - p-12b is well within the range that is easily detectable for many transit surveys , and the @xmath429@xmath28rv semi - amplitude , while small , is still easily measured with high - precision rv spectrometers ( though more observations may be needed for a robust confirmation , which may slow the rate at which these planets are announced ) .
we conclude that of hot gaseous planets with radii similar to jupiter , only a small fraction have masses similar to saturn such as hat - p-12b ; the majority have masses similar to jupiter . with the discovery of hat - p-12b
, we estimate that the fraction is @xmath430 , with considerable uncertainty .
we would like to thank the referee , peter mccullough , for several suggestions that improved the quality of this paper , and scott gaudi for a helpful discussion .
hatnet operations have been funded by nasa grants nng04gn74 g , nnx08af23 g and sao ir&d grants .
work of g..b . and j. johnson were supported by the postdoctoral fellowship of the nsf astronomy and astrophysics program ( ast-0702843 and ast-0702821 , respectively ) .
we acknowledge partial support also from the kepler mission under nasa cooperative agreement ncc2 - 1390 ( d.w.l . ,
thanks the hungarian scientific research foundation ( otka ) for support through grant k-60750 .
this research has made use of keck telescope time granted through noao and nasa .
the digitized sky surveys were produced at the space telescope science institute under u.s .
government grant nag w-2166 .
the images of these surveys are based on photographic data obtained using the oschin schmidt telescope on palomar mountain and the uk schmidt telescope .
the plates were processed into the present compressed digital form with the permission of these institutions .
this research has made use of the simbad database , operated at cds , strasbourg , france . | we report on the discovery of hat - p-12b , a transiting extrasolar planet orbiting the moderately bright @xmath0 k4 dwarf gsc 03033 - 00706 , with a period @xmath1d , transit epoch @xmath2 ( bjd ) and transit duration @xmath3d .
the host star has a mass of @xmath4@xmath5 , radius of @xmath6@xmath7 , effective temperature k and metallicity @xmath8 = { \ifthenelse{\equal{{i}}{i}}{{\ensuremath{-0.29\pm0.05}}}{{\ensuremath{-0.36\pm0.04}}}}$ ] .
we find a slight correlation between the observed spectral line bisector spans and the radial velocity , so we consider , and rule out , various blend configurations including a blend with a background eclipsing binary , and hierarchical triple systems where the eclipsing body is a star or a planet .
we conclude that a model consisting of a single star with a transiting planet best fits the observations , and show that a likely explanation for the apparent correlation is contamination from scattered moonlight . based on this model
, the planetary companion has a mass of @xmath9@xmath10 , and radius of @xmath11@xmath12 yielding a mean density of @xmath13@xmath14 . comparing these observations with recent theoretical models
we find that hat - p-12b is consistent with a @xmath15 gyr , mildly irradiated , h / he dominated planet with a core mass @xmath16 .
hat - p-12b is thus the least massive h / he dominated gas giant planet found to date .
this record was previously held by saturn . |
one of the most fundamental applications of non - classical light field states is the improvement of measurement precision beyond the standard quantum limits for classical light sources . of particular interest
is the possible enhancement of phase sensitivity in interferometry @xcite , which could be useful in a wide range of fields , from quantum lithography @xcite to atomic clocks @xcite .
it is well known that the optimal phase resolution @xmath3 that can be achieved using a non - classical @xmath4-photon state is given by the heisenberg limit of @xmath5 .
recently , few photon interferometry at this limit has been accomplished by new methods of generating @xmath4-photon path entangled states using parametric downconversion and post - selection @xcite .
such path entangled states are an equal superposition of the two @xmath4-photon states where all photons are located in the same optical mode , @xmath6 .
they are therefore ideally suited to obtain @xmath4-photon interference fringes with a period of @xmath7 in the optical phase shift between the two paths . in principle
, the generation of path entangled states can be extended to higher photon numbers using the methods proposed and realized in @xcite . in practice , however , the statistical bottlenecks in the post - selection ( or heralding ) used to generate the path entangled states rapidly reduce the probabilities of generating an appropriate output as photon number increases .
it may therefore be useful to consider alternative few photon states that can be generated more efficiently from a given number of downconverted photon pairs . in this paper
, it is shown that a highly phase sensitive state can be generated by subtracting a single diagonally polarized photon from the @xmath8-photon state generated in collinear type ii downconversion . since
single photon subtraction can be performed with equal efficiency for any number of input photons , this method could be very helpful in achieving phase resolutions at the heisenberg limit for higher photon numbers .
moreover , the coherence induced between two adjacent photon number states ensures that the narrowness of the photon number distribution is maintained under phase shifts of up to @xmath2 .
phase shifts at the heisenberg limit can therefore be observed as nearly deterministic transfers of a single photon between the output modes .
the proposed experimental setup is shown schematically in fig .
[ setup ] .
the initial state generated by collinear type ii parametric downconversion is a superposition of photon number states with equal photon number in the horizontal and vertical polarizations , @xmath9 if it can be assumed that all of the emitted photons will eventually be detected , it is possible to isolate a single @xmath1-photon component by post - selecting only outputs where a total of @xmath1 photons are detected @xcite . effectively , the input state is then given by @xmath10 .
this @xmath1-photon input component is reflected at a beam splitter with a reflectivity of @xmath11 close to one , and one photon is detected in the transmitted light .
the components of the @xmath1-photon states in the beam splitter output with exactly one transmitted photon are given by @xmath12 & & \sqrt{n(1-r)r^{2n-1}}\left ( \mid \ !
n ; n\!-\!1 \rangle_{hv } \otimes \mid 0 ; 1 \rangle_{hv } + \mid \ !
n \rangle_{hv } \otimes \mid 1 ; 0 \rangle_{hv } \right ) + \ldots,\end{aligned}\ ] ] the beam splitter thus entangles the polarization of the transmitted one photon component and the polarization of the reflected @xmath13-photon component .
it is now possible to measure the diagonal polarization of the transmitted photon using a @xmath14-plate set at @xmath15 and a polarization beam splitter . this measurement projects the state of the transmitted photon onto an equal superposition of horizontal and vertical polarization , resulting in a conditional output state of @xmath16 in the reflected light .
( 400,220 ) ( 20,122)(60,36)pdc ( 80,137)(1,0)150 ( 80,143)(1,0)150 ( 100,115)(60,20)@xmath17 ( 178,137)(0,-1)62 ( 172,143)(0,-1)68 ( 175,72)(1,1)10 ( 175,72)(-1,1)10 ( 120,45)(80,20)@xmath18 ( 150,25)(80,20)@xmath19 ( 235,137)(1,0)35 ( 235,143)(1,0)35 ( 300,137)(1,0)30 ( 300,143)(1,0)30 ( 333,140)(-1,1)10 ( 333,140)(-1,-1)10 ( 288,125)(0,-1)30 ( 282,125)(0,-1)30 ( 285,92)(1,1)10 ( 285,92)(-1,1)10 ( 150,175)(50,15)beam ( 150,160)(50,15)splitter ( 160,155)(1,-1)30 ( 190,110)(15,15)@xmath11 ( 220,175)(25,15)@xmath14- ( 220,160)(25,15)plate ( 230,125)(5,30 ) ( 220,105)(25,15)@xmath20 ( 270,160)(30,15)pbs ( 270,155)(1,-1)30 ( 270,125)(30,30 ) ( 290,100)(80,15)photon ( 290,85)(80,15)detectors ( 340,130)(0,1)20 ( 340,130)(350,130)(350,140 ) ( 340,150)(350,150)(350,140 ) ( 350,140)(360,140)(360,130 ) ( 350,110)(20,20)1 " ( 275,85)(1,0)20 ( 275,85)(275,75)(285,75 ) ( 295,85)(295,75)(285,75 ) ( 285,75)(285,65)(295,65 ) ( 298,55)(20,20)0 " equation ( [ eq : bs ] ) shows that the probability of successfully subtracting exactly one photon from the @xmath8-photon input is given by @xmath21 .
this value can be optimized independently for any desired photon number by varying the reflectivity @xmath11 .
the maximal efficiency of photon subtraction is obtained at @xmath22 .
the probability of successful photon subtraction is then equal to @xmath23 .
interestingly , this maximal probability decreases only slightly with photon number , from an initial value of 25% at @xmath24 towards a value of @xmath25 for extremely high photon numbers . by selecting an optimized reflectivity of @xmath22 ,
it is thus possible to achieve post - selection probabilities greater than 18% for any number of input photons .
the efficiency of photon subtraction is therefore almost independent of photon number .
it should be noted that this is quite different from the photon bottleneck used to generate the path entangled state @xmath6 , where the corresponding post - selection probability drops rapidly with increasing photon number as more and more beam splitters become necessary to `` bunch up '' the photons in the single mode bottleneck . in the basic scheme introduced in @xcite , the bottleneck efficiency is @xmath26 for an @xmath4 photon state . at 5 photons , this is an efficiency of only 0.24 % , almost a hundred times less than the optimal efficiencies of photon subtraction .
it should therefore be much easier to increase the output photon number of wedge states than to achieve the same photon number for path entangled states .
the complete polarization statistics of @xmath4-photon quantum states can be expressed by the three stokes parameters describing the photon number differences between horizontal ( h ) and vertical ( v ) , plus ( p ) and minus ( m ) diagonal , and right ( r ) and left ( l ) circular polarization , @xmath28 as can be seen from eq .
( [ eq : wedge ] ) , the stokes parameter @xmath29 describing the @xmath30-polarization takes on values of @xmath31 or @xmath32 , with a 50% probability each .
the average of @xmath29 is therefore zero , and its uncertainty is @xmath33 .
the low uncertainty in @xmath29 is a direct consequence of the quantum correlations in parametric downconversion .
in fact , the original @xmath34-state is an @xmath29 eigenstate with an uncertainty of zero , which already provides phase sensitivities at the heisenberg limit in the photon statistics @xcite .
photon subtraction actually increases the @xmath29-uncertainty by one , thus increasing the observed photon number noise in the @xmath30-basis .
however , the essential effect of the photon subtraction on the polarization statistics of the output state is the generation of coherence between the horizontal and vertical polarization components .
this effect can be observed in the statistics of the stokes parameter @xmath35 , describing the photon number difference between the diagonal polarizations @xmath36 and @xmath37 . due to the coherence between @xmath38 and @xmath39 , the expectation value of this stokes
parameter is @xmath40 since the total photon number is @xmath41 , this means that on average , more than 3/4 of all output photons are polarized along the same diagonal as the transmitted photon @xcite . since the wedge state polarization of @xmath42 originates from complete coherence between two adjacent eigenstates of @xmath29 , it is the maximal diagonal polarization possible at an uncertainty of only @xmath33 in the difference between horizontally and vertically polarized photons . as shown in fig .
[ wedge ] , the polarization distribution of the output state thus resembles a quantum mechanically narrow `` wedge '' inserted between the @xmath29 eigenstates from the positive side of the diagonal polarization @xmath35 - somewhat like the slice of an orange , with a quantum limited thickness of @xmath43 .
( 120,120 ) ( 20,60)(20,76)(32,88 ) ( 32,88)(44,100)(60,100 ) ( 60,100)(76,100)(88,88 ) ( 88,88)(100,76)(100,60 ) ( 100,60)(100,44)(88,32 ) ( 88,32)(76,20)(60,20 ) ( 60,20)(44,20)(32,32 ) ( 32,32)(20,44)(20,60 ) ( 53,21)(0,1)78 ( 53.5,21)(0,1)78 ( 67,21)(0,1)78 ( 67.5,21)(0,1)78 ( 60,60 ) ( 60,60)(0,-1)21 ( 60,60)(53,41)(46,22 ) ( 60,60)(67,41)(74,22 ) ( 60,15)(-1,0)14 ( 60,15)(1,0)14 ( 60,15 ) ( 30,0)(60,12)@xmath44 due to the narrowness of its @xmath29 distribution and due to its comparatively high expectation value @xmath45 , the wedge state is very suitable for measurements of small phase shifts between the right and left circular polarizations , which result in a rotation of the stokes parameters around the @xmath46 axis . as indicated in fig .
[ wedge ] , the phase uncertainty of the polarization wedge can then be estimated by @xmath47 this phase uncertainty is close to the heisenberg limit and corresponds to the phase uncertainties of the various phase squeezed states proposed for optimized phase estimation in quantum interferometry @xcite .
the wedge state is therefore an almost ideal phase squeezed @xmath4-photon state .
( 380,200 ) ( 70,130)(20,20)@xmath48 ( 90,140)(1,0)150 ( 110,170)(20,20)@xmath49 ( 120,170)(0,-1)110 ( 170,120)(20,20)@xmath50 ( 122,90)(20,20)@xmath51 ( 240,140)(0,-1)28 ( 240,100 ) ( 255,100)(26,12)phase ( 255,90)(30,12)shift ( 225,85)(30,30)@xmath52 ( 240,88)(0,-1)58 ( 239,30)(-1,2)5 ( 239,30)(1,2)5 ( 120,60)(1,0)150 ( 270,60)(-2,1)10 ( 270,60)(-2,-1)10 ( 270,60)(70,20)@xmath53 ( 295,42)(70,20)@xmath54 ( 180,5)(120,20)@xmath55 ( 70,165)(36,12)beam ( 70,155)(40,12)splitter ( 105,155)(1,-1)30 ( 80,50)(40,12)mirror ( 105,75)(1,-1)30 ( 104,74)(1,-1)30 ( 240,140)(40,12)mirror ( 225,155)(1,-1)30 ( 226,156)(1,-1)30 ( 190,85)(36,12)beam ( 190,75)(40,12)splitter ( 225,75)(1,-1)30 to clarify the application of this phase sensitivity in interferometry , it may be useful to consider the possibility of converting the polarization modes @xmath48 and @xmath49 to spatial input modes with equal polarization . as shown in fig .
[ mz ] , the two paths inside the interferometer then correspond to the modes @xmath56 , and the effect of a phase shift @xmath52 between the two paths is to rotate the original @xmath30 basis towards the @xmath57 basis . in terms of the photon number difference @xmath58 observed in the two output ports , this effect
can then be expressed as the rotation of the stokes vector around the @xmath46 axis mentioned above , with @xmath59\ ; \hat{s}_1 ( \mbox{in } ) + \sin[\phi]\ ; \hat{s}_2 ( \mbox{in}).\ ] ] the phase shift @xmath52 between the arms of the interferometer thus corresponds directly to the rotation of the linear polarization components obtained e.g. by a half wave plate set at @xmath60 .
the phase sensitivity of the @xmath13-photon wedge state can then be described by the average @xmath61 and the variance @xmath62 of the measurement result @xmath63 , @xmath64 \nonumber \\[0.2 cm ] \delta s_1 ^ 2(\mbox{out } ) & = & 1+(n^2 - 2 ) \sin^2[\phi].\end{aligned}\ ] ] for phase shifts @xmath52 smaller than @xmath2 , @xmath65 and @xmath66 corresponds to a phase resolution of @xmath67 , as given by eq.([eq : delta ] ) above .
( @xmath68)-photon wedge states can thus achieve phase resolutions close to the heisenberg limit for arbitrarily high photon numbers .
moreover , the variance of the output photon number distribution at phase shifts @xmath52 with @xmath69=\pm 1 $ ] is still smaller than two . even at phase shifts that change the average output photon number difference by one ,
the photon number distribution is therefore sharper than the difference of two between two adjacent measurement outcomes of @xmath29 .
this result indicates that phase shifts at the heisenberg limit are observed as nearly deterministic transfers of a single photon between the horizontal and vertical polarizations , with measurement probabilities greater than 50% of finding a measurement outcome equal to the expectation value of @xmath70 at @xmath69=\pm 1 $ ] .
it is interesting to note that the above argument only relies on averages and variances of the photon number differences @xmath71 .
it is therefore straightforward to estimate the effect of basic photon counting errors . in particular , a photon loss error may occur when the downconversion actually generates @xmath72 pairs , and two photons
are subsequently lost due to limited detector efficiencies . at low detector efficiencies ,
the probability of such a photon loss error is of the order of @xmath73 , corresponding to the ratio of the @xmath72 pair probability to the @xmath74 pair probability in eq.([eq : pdc ] ) .
since random ( = unpolarized ) photon losses do not change the average polarization of the photons , the strong coherence between the @xmath75 and @xmath76 polarization given by @xmath45 is unchanged by this error .
however , the polarization fluctuations increase due to the possibility that both of the photons lost had the same polarization . specifically , the increase in @xmath77 is equal to two times the probability of the photon loss error .
for reasonably low error probabilities ( e.g. for @xmath78 ) , this additional uncertainty is much smaller than the pure state uncertainty of @xmath33 , and the effect on the phase resolution will be negligible .
these considerations indicate that the high phase resolution of wedge states is rather robust against photon loss errors .
again , this is an important difference to path entangled states , where the loss of a single photon completely destroys the quantum coherence responsible for the high phase resolution . due to this robustness against photon loss errors
, it may be interesting to investigate wedge state generation at high pump powers even if high detector efficiencies can not be achieved @xcite .
( 360,220 ) ( 80,195)(200,20)probability @xmath79 ( 0,0)(320,220 ) [ 1.2 ] probability of finding two photons in the horizontally polarized mode and three photons in the vertically polarized mode as a function of phase shift @xmath52 between the circular polarizations for the five photon wedge state .
the dotted line shows the same probability for a corresponding five photon path - entangled state .
, title="fig : " ] ( 130,5)(100,20)phase shift @xmath52
to illustrate the full implications of the phase sensitivity of wedge states at the level of precise photon counting , it may be useful to take a closer look at a specific example . here
, a compromise is necessary between the experimental difficulties and the increasing phase resolution permitted by higher photon numbers .
a good choice may be the five photon wedge state , generated by subtracting one photon from @xmath80 pairs of downconverted photons , since it should be just within reach of present technological possibilities .
a phase shift of @xmath52 transforms this state according to @xmath81 + \cos[\frac{3}{2}\phi + \frac{\pi}{4 } ] - 2 \cos[\frac{1}{2}\phi -\frac{\pi}{4}]\right ) \nonumber \\
\langle 4;1 \mid \hat{u}_\phi \mid \mbox{wedge } \rangle & = & -\frac{\sqrt{2}}{16 } \left ( 5 \cos[\frac{5}{2}\phi + \frac{\pi}{4 } ] - 3 \cos[\frac{3}{2}\phi -\frac{\pi}{4 } ] - 2 \cos[\frac{1}{2}\phi + \frac{\pi}{4}]\right ) \nonumber \\
\langle 3;2 \mid \hat{u}_\phi \mid \mbox{wedge } \rangle & = & \frac{1}{8 } \left ( 5 \cos[\frac{5}{2}\phi -\frac{\pi}{4 } ] + \cos[\frac{3}{2}\phi + \frac{\pi}{4 } ] + 2 \cos[\frac{1}{2}\phi -\frac{\pi}{4}]\right ) \nonumber \\
\langle 2;3 \mid \hat{u}_\phi \mid \mbox{wedge}\rangle & = & \frac{1}{8 } \left ( 5 \cos[\frac{5}{2}\phi + \frac{\pi}{4 } ] + \cos[\frac{3}{2}\phi -\frac{\pi}{4 } ] + 2 \cos[\frac{1}{2}\phi + \frac{\pi}{4}]\right ) \nonumber \\
\langle 1;4 \mid \hat{u}_\phi \mid \mbox{wedge } \rangle & = & -\frac{\sqrt{2}}{16 } \left ( 5 \cos[\frac{5}{2}\phi -\frac{\pi}{4 } ] - 3 \cos[\frac{3}{2}\phi + \frac{\pi}{4 } ] - 2 \cos[\frac{1}{2}\phi -\frac{\pi}{4}]\right ) \nonumber \\
\mid \hat{u}_\phi \mid \mbox{wedge } \rangle & = & -\frac{\sqrt{10}}{16 } \left ( \cos[\frac{5}{2}\phi + \frac{\pi}{4 } ] + \cos[\frac{3}{2}\phi -\frac{\pi}{4 } ] - 2 \cos[\frac{1}{2}\phi + \frac{\pi}{4}]\right).\end{aligned}\ ] ] each of these six amplitudes includes a five photon interference component oscillating at a rate of @xmath82 .
this five photon interference effect is particularly strong in the @xmath83 and the @xmath84 components .
[ fringes ] shows the interference fringes observed in the @xmath84 component . for comparison , the dashed line shows the corresponding fringes of a five photon path - entangled state .
the main difference between the two fringes are the different peak heights of the wedge state indicating the average diagonal polarization of the five photon state .
the wedge state thus combines features of the maximal five photon interference of path - entangled states with the well defined polarization direction of a phase squeezed state .
( 240,220 ) ( 0,180)(120,20)(a ) @xmath85 ( @xmath86 ) ( 10,40)(1,0)220 ( 35,40.5)(1,0)20 ( 35,20)(20,20)-5 ( 65,40)(0,1)9 ( 85,40)(0,1)9 ( 65,49)(1,0)20 ( 65,20)(20,20)-3 ( 95,40)(0,1)19 ( 115,40)(0,1)19 ( 95,59)(1,0)20 ( 95,20)(20,20)-1 ( 125,40)(0,1)150 ( 145,40)(0,1)150 ( 125,190)(1,0)20 ( 125,20)(20,20)+1 ( 155,40)(0,1)21 ( 175,40)(0,1)21 ( 155,61)(1,0)20 ( 155,20)(20,20)+3 ( 185,40)(0,1)1 ( 205,40)(0,1)1 ( 185,41)(1,0)20 ( 185,20)(20,20)+5 ( 80,0)(80,20)@xmath87 ( 20,120)(20,136)(32,148 ) ( 32,148)(44,160)(60,160 ) ( 60,160)(76,160)(88,148 ) ( 88,148)(100,136)(100,120 ) ( 100,120)(100,104)(88,92 ) ( 88,92)(76,80)(60,80 ) ( 60,80)(44,80)(32,92 ) ( 32,92)(20,104)(20,120 ) ( 66.5,81)(0,1)78 ( 67.5,81)(0,1)78 ( 60,120 ) ( 60,120)(1,-3)6 ( 160,170)(60,20)@xmath88 ( 160,140)(60,20)@xmath89 ( 240,220 ) ( 0,180)(120,20)(b ) @xmath90 ( @xmath91 ) ( 10,40)(1,0)220 ( 35,40)(0,1)3 ( 55,40)(0,1)3 ( 35,43)(1,0)20 ( 35,20)(20,20)-5 ( 65,40)(0,1)8 ( 85,40)(0,1)8 ( 65,48)(1,0)20 ( 65,20)(20,20)-3 ( 95,40)(0,1)8 ( 115,40)(0,1)8 ( 95,48)(1,0)20 ( 95,20)(20,20)-1 ( 125,40)(0,1)85 ( 145,40)(0,1)85 ( 125,125)(1,0)20 ( 125,20)(20,20)+1 ( 155,40)(0,1)85 ( 175,40)(0,1)85 ( 155,125)(1,0)20 ( 155,20)(20,20)+3 ( 185,40)(0,1)11 ( 205,40)(0,1)11 ( 185,51)(1,0)20 ( 185,20)(20,20)+5 ( 80,0)(80,20)@xmath87 ( 20,120)(20,136)(32,148 ) ( 32,148)(44,160)(60,160 ) ( 60,160)(76,160)(88,148 ) ( 88,148)(100,136)(100,120 ) ( 100,120)(100,104)(88,92 ) ( 88,92)(76,80)(60,80 ) ( 60,80)(44,80)(32,92 ) ( 32,92)(20,104)(20,120 ) ( 67,81)(0,1)78 ( 81,86)(0,1)68 ( 60,120 ) ( 60,120)(3,-4)12 ( 160,170)(60,20)@xmath92 ( 160,140)(60,20)@xmath93 the well defined polarization direction is particularly visible at phase angles of @xmath85 ( or @xmath86 ) , where the probability of measuring @xmath83 has its maximal value of 75.2% , and at @xmath90 ( or @xmath91 ) , where the probabilities of measuring @xmath84 and @xmath94 are both equal to 42.5% . fig .
[ prob ] shows these two probability distributions , along with a schematic illustration of the corresponding quantized levels on the poincare sphere . as indicated by fig .
[ prob ] ( a ) , the output photon number distribution can be `` switched '' between 75.2 % @xmath95 ( @xmath84 ) at @xmath96 and 75.2 % @xmath97 ( @xmath83 ) at @xmath98 .
likewise , fig .
[ prob ] ( b ) indicates that the photon number distribution at @xmath99 can be shifted by exactly one photon with only 15% of the outcomes scattered to different photon numbers .
considering the fact that the measurement outcomes are discrete , this shift in the probability distribution is surprisingly smooth . specifically , the high fidelity of the @xmath83 component at @xmath98 shown in fig . [ prob ]
( a ) suggests a `` polarization wedge '' that is much sharper than the photon number distribution at @xmath99 @xcite .
the quantum coherence between the adjacent photon number states induced by the post - selected photon subtraction thus permits a surprisingly high level of control at the single photon level .
although the detailed calculations presented here only apply to the specific case of five photons , it should be remembered that the basic features of the statistics for a general @xmath13-photon wedge state are defined by the stokes parameter statistics given in section [ sec : stokes ] .
since the uncertainty of @xmath29 is always one , the photon number distributions will be limited to only a few possible measurement outcomes close to @xmath100 for any number of photons .
specifically , the photon number distribution at @xmath101 will always be similar to fig .
[ prob ] ( a ) , and the distribution at @xmath102 will be similar to fig . [ prob ]
( b ) . for high @xmath74 , we can therefore expect a maximal probability of @xmath97 at a phase angle of @xmath103 , where , according to eq.([eq : stats ] ) , @xmath101 and @xmath104 . the `` switch '' from @xmath95 to @xmath97
is therefore also observable at higher photon numbers .
in conclusion , it has been shown that it is possible to obtain highly phase sensitive @xmath0-photon wedge states by photon subtraction from @xmath74 photon pairs generated in collinear type - ii parametric downconversion . since the post - selection condition for the generation of this state can be higher than 18% regardless of photon number , the only limitation in extending this scheme to high photon numbers is the efficiency of the parametric downconversion
. the generation of five photon wedge states should therefore be well within reach of present technological capabilities .
quantitatively , the phase sensitivity of wedge states is comparable to that of the recently realized path entangled states , with the advantage that a phase shift can be related directly to a shift in the photon number distribution observed in the output .
specifically , phase shifts at the heisenberg limit appear as nearly deterministic transfers of one photon between the two output ports .
@xmath0-photon wedge states should therefore be highly suitable for the determination of phase shifts @xmath52 of about @xmath2 .
the generation of polarization wedge states by downconversion and photon subtraction thus provides a simple and effective experimental approach to phase measurements at the heisenberg limit with non - classical @xmath4-photon inputs .
it should be noted that precise photon counting would require special detectors such as the ones described in s.kim , y.yamamoto , and h.h .
hogue , appl .
74 * , 902 ( 1999 ) . without such special technology , the effects of photon losses
are usually avoided by using downconverted light with @xmath105 much smaller than one .
the effects of higher photon number components can then be neglected even if the detectors have limited quantum efficiencies and can not resolve photon number .
however , this method limits the output intensity and makes it difficult to achieve high photon numbers .
it may seem a bit counterintuitive that the subtraction of a photon causes the remaining photons to have the same polarization as the photon removed from the state .
however , the effect can be understood quite clearly if one remembers that photon bunching implies that the photons preferably have the same diagonal polarization .
a similar robustness of bipartite entanglement against the photon losses caused by limited detector efficiencies has recently been investigated in h.s .
eisenberg , g. khoury , g.a .
durkin , c. simon , and d. bouwmeester , phys .
* 93 * , 193901 ( 2004 ) .
this paper also gives an interesting example of parametric downconversion with @xmath106 .
it may be worth noting that the apparent localization of the `` wedge '' between the quantized measurement results is somewhat similar to the correlation between coherence and half - integer photon number results in observable in finite resolution measurements , as discussed in h.f .
hofmann , phys .
a * 61 * , 033815 ( 2000 ) . | it is shown that a highly phase sensitive polarization squeezed @xmath0-photon state can be generated by subtracting a diagonally polarized photon from the @xmath1 photon component generated in collinear type ii downconversion .
this polarization wedge state has the interesting property that its photon number distribution in the horizontal and vertical polarizations remains sharply defined for phase shifts of up to @xmath2 between the circularly polarized components .
phase shifts at the heisenberg limit are therefore observed as nearly deterministic transfers of a single photon between the horizontal and vertical polarization components . |
increasingly , many complex systems are being modeled as networks since the framework of nodes representing the basic elements of the system and the interconnections of the network representing the interaction between the elements fits well for a theoretical study .
when the complex systems are large - dimensional dynamical systems , the network framework comprises many interacting subsystems of smaller dimension each of which constitutes a node . as a particular example , the whole or a portion of the entire neuronal activity in the human brain can be regarded as the consequential dynamics of interacting subsystems , where the dynamics of a subsystem is generated by a group of neurons .
the enormously interconnected subsystems in the brain generate a wide variety of dynamical patterns , synchronised activities and rhythms .
epilepsy is a disorder that affects the nerve cell activity which in turn intermittently causes seizures . during such seizures
, patients could experience abnormal sensations including loss of consciousness .
clinical and theoretical research have shown that underpinning the cause of this disorder has not proved to be easy , in particular the predicament of resolving the question as to whether the disorder manifests due to the nature of the interconnections in the brain or the pathologicity of a portion of the brain tissue itself remains .
since epilepsy is one of the most common neurological disorders with an estimate of more than 50 million individuals being affected @xcite , there is a strong need both for curative treatment and as well for the analysis of the underlying structure and dynamics that could bring about seizures . in this paper , our attention is on focal epilepsy where the origin of the seizures are circumscribed to certain regions of the brain called the seizure focus @xcite , and the aim of this paper is to foster theoretical investigation into the connection strengths of the underlying nodes in such a seizure focus in comparison to the other nodes in the network .
different models have been proposed to understand different aspects of focal epilepsy @xcite .
mathematically speaking , dynamical system models of focal epilepsy studied in the literature are mainly of two broad categories : ( i ) .
models that consider noise to be inducing seizures in a node @xcite ( ii ) .
models that consider seizures to be caused or terminated by a bifurcation , a change in the intrinsic feature of the ( autonomous ) dynamics at the node @xcite .
while modeling the neuronal activity of the human brain , there are far too many numerous parameters which also dynamically evolve on separate spaces and scales , and it is unrealistic to observe or encapsulate all these aspects in a static network or an autonomous dynamical system ( see appendix [ appendixaa ] ) . since the brain evolution does not depend on its own internal states as it responds to external stimuli and the interconnections in the brain are likely to change in response to stimuli , the human brain can be safely said to be a nonautonomous system and in the language of network dynamics , this is effectively an evolving network ( see section [ sec_sfn ] ) . besides evidence that network connections play a role in initiating seizures ( e.g. , @xcite ) , some authors also provide evidence that stronger interconnections play a role in epileptic seizures ( e.g. , @xcite ) .
also , a commonly found feature of biological systems is adaptation : an increase in exogenous disturbance beyond a threshold activates a change in the physiological or biochemical state , succeeded by an adaptation of the system that facilitates the gradual relaxation of these states toward a basal , pre - disturbance level .
based on all these , we present a novel phenomenological model where seizures could be activated in a node , if certain interconnections change to have larger weights resulting in a larger exogenous drive or disturbance or energy transmitted from other nodes into it , and upon experiencing seizure , adjustment of interconnections takes place eventually to subside the seizures . since the interconnections are allowed to change , the model dynamics emerges from an evolving network . in this paper
, we propose a patient and data - specific , macroscopic , functional network model that can act as a surrogate " representation of the electroencephalography ( eeg ) or electrocorticography ( ecog ) time - series .
the network model is new in the sense that it satisfies the following simultaneously : the network connections are directed and weighted ; the network connections are time - varying ; there is a discrete - time dynamical system at each node whose dynamics is constantly influenced by the rest of the network s structure and dynamics ; the ability of the network to generate a time - series that sharply resembles the original data is based on a notion of robustness of the dynamics of the network to its initial conditions what we call as the state - forgetting " property .
the essence of the state - forgetting property , roughly put is that the current state of the brain does not have an
distinct imprint " of the brain s past right since from its genesis " or origin " ( rigorous definitions are provided later in the paper ) .
effectively , this means that the brain tends to forget its past history as time progresses .
such a notion has been adapted from the previously proposed notions for studying static ( artificial neural ) networks with an exogenous input in the field of machine learning viz .
, echo state property @xcite , liquid state machines , @xcite , backpropagation - decorrelation learning,@xcite and others .
however , the network model that we represent is very different from that of the artificial neural networks and moreover the nodes in our model correspond to different regions of the brain .
our methodology is as follows : we present an evolving network model which exhibits state - forgetting property for all possible inter - connections , and a particular sequence of interconnection strengths ( matrices ) is determined from a given eeg / ecog time - series so that the model can almost regenerate the time - series .
the weights of the interconnections are chosen from a set of possibilities based on a heuristic algorithm that infers a causal relationship between nodes a ( directed ) synchrony measure between two nodes @xmath0 and @xmath1 represents the conductance ( ease at which energy flows ) from node @xmath1 into node @xmath0 , and weight of the connection from node @xmath1 into node @xmath0 has some degree of proportionality with the synchrony measure .
the state forgetting property ensures the synthesized evolving network has in some sense captured the original time series as an attractor " ( see @xcite and proposition [ prop_pullback ] ) , and hence the network acts as surrogate representation of the original time series .
in other words , since any initial condition presented to a state - forgetting network has no effect on its asymptotic dynamics , and that the asymptotic dynamics resembles the time - series , we assume that the evolving network model has captured the essence of the time - series data .
to indicate the potential utilities of the model , we analyse the constructed evolving network from a patient s clinical data . by comparing the relative strength of the interconnections ,
we identify nodes that are outgoing hubs of the network as a theoretically determined seizure focus .
the nodes in such a focus have a strong location - wise correlation with the set of nodes in a clinically determined seizure focus .
we also , discuss the efficacy of surgical removal of the determined focus in subsiding seizures .
the remainder of the paper is organised as follows .
we introduce the notion of state - forgetting evolving networks in section [ sec_sfn ] ; formal definitions are made in appendix [ appendixa ] . in section [ sec_model ] , we present the evolving network model in detail and state a result concerning the state - forgetting property ; the complete mathematical proof of this result is presented in appendix [ appendixb ] . in section [ sec_weights ] ,
we present a method to obtain the weight matrices to obtain a time - varying network from a clinical time series . in section [ sec_results ]
, we present some results on the network inference followed by a brief discussion and conclusion in section [ sec_conclusion ] .
the aim of this section is to give an intuitive description of the state forgetting property of evolving networks ; the formal and rigorous definitions of this property are found in appendix [ appendixa ] .
we employ discrete - time systems @xcite since they are more amenable to computer simulations as the computed orbits or solutions unlike in the case of differential equations do not depend on numerical methods for computing the solutions and their related accuracy .
also , discrete - time systems can exhibit complex dynamical behaviour ( e.g. , @xcite ) even in one - dimension whereas a continuous time - system needs a dimension of at least three . using fig .
[ fig_3nodes ] , we present a less formal and simplistic description of an evolving network with three nodes @xmath2 .
let @xmath3 be a @xmath4 matrix defined for each integer @xmath5 with @xmath6 being the strength of the incoming connection from node @xmath1 to node @xmath0 ; here @xmath6 is the element in the @xmath7 row and @xmath8 column of @xmath3 .
there is a discrete - time dynamical system at each node in the evolving network represented in fig .
[ fig_3nodes ] , and the @xmath7 node dynamics is given by @xmath9 where @xmath10 is some function ( not dependent on @xmath5 ) that calculates @xmath11 using information from the dynamical state in the three nodes at time @xmath5 and the respective incoming strengths into node @xmath0 . by specifying @xmath12 , @xmath13 and @xmath14 together with the sequence of weight matrices @xmath3 ,
the evolving network is specified completely . by denoting @xmath15
, we can represent the evolving network dynamics also as a nonautonomous system @xmath16 , where the variation of @xmath17 with @xmath5 represents the inherent plasticity of the system ( change in @xmath3 above ) .
we can also allow @xmath17 s to represent external stimuli to the network and other obfuscated factors . from node @xmath18
since an evolving network is essentially a nonautonomous system , we recall some relevant results of a nonautonomous system . of interest to us
is the notion of an entire - solution of a nonautonomous system .
an entire solution ( or orbit ) of a nonautonomous system @xmath19 is a sequence @xmath20 that satisfies @xmath21 for all @xmath22 .
it is possible to show under natural conditions that there exist at least one entire solution @xcite . in the particular case , where the evolving network has a unique entire solution
, we say that the network ( or the nonautonomous system ) has the state - forgetting property ( formally defined in definition [ def_sf ] ) .
now , we comment on the rather intriguing consequence of an evolving network @xmath19 having multiple entire - solutions .
suppose @xmath20 and @xmath23 are two distinct entire - solutions of a nonautonomous dynamical system , then for some time @xmath24 , we have @xmath25 i.e , the system could have evolved into two possible states at time @xmath26 . since @xmath27 and @xmath28 , it is necessary to have @xmath29 .
similarly , @xmath29 would necessitate @xmath30 and so on .
thus , in general , at any time - instant @xmath26 , @xmath25 implies that there are two entire - solutions that have had distinct infinitely long pasts . in other words , the dynamics of such systems with distinct entire - solutions depends " not only on the inherent plasticity or
the exogenous stimuli represented by the maps @xmath19 , the distinct solutions carry a distinct and non - vanishing imprint of their past . from the arguments in the preceding paragraph , for nonautonomous systems like the human brain , having more than one entire solution translates to saying that the current state of the brain is not just the result of the changing interconnections in the brain or the exogenous input but to also say that its past states have a non - vanishing effect on the asymptotic dynamics . since it is hard to envisage or believe that seizures in current time had something to do with the state of the brain right from its genesis " ( genesis is used as a metaphor to refer to the earliest factual time instant from which the current brain state has evolved from , and this time could be matter of debate ) , we assume the more plausible seizures at a certain time are a consequence of how the network structure has evolved / changed till that time or a result of a stream of exogenous inputs to the brain , but certainly not a consequence of the state at which the brain was in during its genesis .
this also forms the basis for employing echo - state networks in artificial neural network training and machine learning@xcite .
whenever a nonautonomous system ( technically , defined on a compact state space ) has the state - forgetting property , it also has state - space contraction towards its entire - solution in the backward direction ( see proposition [ prop_pullback ] ) .
in fact , a stronger result @xcite holds : if @xmath31 is the unique entire solution , then for every @xmath32 the successive iterates @xmath33 would satisfy @xmath34 as @xmath35 .
in other words , @xmath31 attracts " all other state - space iterates towards it ( component - wise ) as time tends to infinity .
this convergence due to the attractor property of the state - forgetting property is what makes our evolving network model generate a time - series that actually sharply resembles ( mathematically , converges to ) the data from which the network was conceived .
further discussion on attractivity is beyond the scope of this paper .
the larger point is that for an evolving network to be of utility as a surrogate representation of the time - series data , we need the state - forgetting property .
before , we actually describe the evolving network model in , we list and describe its features : * ( i ) . * the evolving network model has @xmath36-nodes for all time . *
( ii ) . * placed at each node is a certain bi - parametric one - dimensional discrete - time dynamical system which is autonomous when there are no incoming connections from other nodes to it , and nonautonomous otherwise .
we refer to the dynamics of the system at a node , as the dynamics of the node or nodal dynamics . * ( iii ) . * among the two parameters that specify this one - dimensional dynamical system is a stability parameter .
the nodal dynamics in autonomous mode has a globally stable fixed point at @xmath37 whenever the stability parameter is in a certain range ( stable - regime ) ; a globally stable fixed point attracts all solutions , i.e , the time series values measured in the node are close to @xmath37 and tend towards to it . * ( iv ) .
* when the stability parameter is chosen not to lie in the stable regime , the autonomous nodal dynamics can render a range of different behaviours including oscillatory behavior , complex behavior like chaos @xcite depending upon the value of the stability parameter . *
* when the nodal dynamics is nonautonomous , the dynamics of the node is perturbed at each time - instant from the dynamical states of other nodes . we call the net perturbation from other nodes as the drive to the node under consideration .
the exogenous drive from other nodes has its signature on the nodal dynamics , and if sufficiently strong , is also capable of setting the nodal dynamics far away from the fixed point @xmath37 . *
* we term a seizure in the node to be a dynamical excursion away from the fixed point , and that the seizures in a node are brought about ( and terminated ) in two distinct scenarios : ( via ) . changing the stability parameter of the node in the autonomous mode of operation ( vib ) . by the nature of the exogenous drive from the other nodes in the network in the nonautonomous mode . * ( vii ) . * in this paper , we consider the case of nodal dynamics slipping into seizures as in scenario ( vib ) above , i.e , we consider each node to have its stability parameter to lie in the stable regime , and the exogenous drive to cause seizures .
the drive is time - varying on two accounts : first , being a function of the states of the other nodes it is time - varying as other nodes have a dynamical system and not static ; second , the drive is time - varying since it is a function of the interconnection strengths of the nodes .
our model is also motivated on the lines of previous research @xcite showing that the hyper - interconnected regions in the brain are prone to seizures ( see section [ sec_model ] ) .
hence , we expect the drive to an individual node is likely to be having large magnitude if it has large weighted incoming connections . to describe a family of evolving networks ( as in definition [ def_en ] ) , we first sketchily describe the nonautonomous dynamics at each node @xmath0 ( the map @xmath10 described in section [ sec_sfn ] or in definition [ def_en ] ) via the update equation : @xmath38 where the various entities are listed and briefly explained : * the iteration of the map @xmath39 on an initial condition gives the autonomous dynamics of the node . some neuron models like that of fitzhugh - nagumo ( e.g. , @xcite )
permit an individual neuron to have a regenerative self - excitation via a feedback term .
based on this principle , we consider a highly simplistic discrete time model of the node dynamics where feedback from the node dynamics can cause self - excitation , and set @xmath40 where parameters @xmath41 and @xmath42 ( for a graph of @xmath39 , see fig .
[ figa ] ) .
explanation on how a parameter can cause self - excitation dynamics is made in section [ sub_sec_isolated ] . * adopting the principle that the individual drive from a connecting node has a larger magnitude if the incoming connection weight and the magnitude of the state value of the connecting node are together larger , we set the total sum of the drives from the other @xmath43 nodes into the node @xmath0 to be @xmath44 @xmath45 where @xmath6 is the weight of the incoming connection from the @xmath8 node to the @xmath7 node ; @xmath46 is an offset parameter to enable a positive argument for the @xmath47 function ; @xmath48 is an increasing function and is defined by @xmath49 the function @xmath50 although increasing , has a monotonically decreasing slope ( see fig . [ figc ] ) .
the net effect of the function @xmath50 can be thought of as the node s mechanism to have a relatively stronger inhibition to a stronger incoming connection
node s internal mechanism to resist seizures by inhibiting stronger exogenous drive .
lastly , the negative sign in is only for mathematical convenience and has no physical significance .
when the drive is zero , the dynamics of the node @xmath0 is in autonomous mode .
* the saturation function henceforth would be the mapping @xmath51 which is linear in a certain adjustable range , and its slope becomes zero asymptotically ( see fig .
[ figb ] ) defined by : @xmath52 $ } \\
\frac{1}{r } \log(\frac{1}{r } + ( y - p ) ) + k & \text { : $ y >
p$}\\ -\frac{1}{r } \log(\frac{1}{r } - ( y+p ) ) - k & \text { : $ y < p$ } , \end{cases}\ ] ] where @xmath53 and @xmath54 is the identify function on @xmath55 $ ] ; @xmath56 determines the asymptotic rate at which the slope of @xmath54 decreases and @xmath57 .
( see fig .
[ figb ] ) ; @xmath54 is linear in the region @xmath58 with a slope @xmath59 , and the slope of @xmath60 approaches @xmath37 as @xmath61 monotonically according to @xmath62 .
putting together the above into , the update equation at the @xmath7 node is modeled by : @xmath63 note that in the model in , the weights @xmath6 are unspecified , but they actually can be solved for given a ecog / eeg time - series @xmath31 ( topic of section [ sec_weights ] ) . in section
[ sub_sec_isolated ] and [ sub_sec_nonauto ] , we describe the dynamics of a node when it operates in autonomous node and in the nonautonomous mode . in section [ sub_sec_thm ] , we state sufficient conditions on the parameters @xmath64,@xmath65 and @xmath66 , to ensure state - forgetting property of any evolving network arising out of .
however , to understand the model , a few remarks are in order concerning the role of the offset parameter @xmath66 and the saturation function @xmath54 in : from , it follows that the individual drive from the node @xmath1 into node @xmath0 , at time @xmath5 is given by @xmath67 . to make the role of the weights meaningful
, a smaller weight @xmath6 should ensure smaller individual drive from node @xmath1 , i.e. , we must have @xmath68 close to zero whenever @xmath6 is close to zero .
we note that this would be the case whenever the offset parameter @xmath66 is sufficiently large . to observe this , we remark on the behavior of @xmath68 as the weight @xmath6 gets close to zero . by definition of @xmath50 in ,
when @xmath6 is close to zero , it follows that @xmath69 is close to @xmath70 .
when @xmath69 is close to @xmath70 , @xmath68 is close to @xmath71 . for practical purposes we restrict @xmath72 $ ] ( see section [ sec_weights ] ) , and
hence if @xmath66 is much larger than @xmath73 , then @xmath74 is close to @xmath59 , and @xmath71 is close to zero , and thus @xmath68 is very close to zero .
however , it is also to be noted that if @xmath66 is made unreasonably large , the signature of @xmath73 on @xmath68 gets weaker , so @xmath66 is chosen in an appropriate range , although with a large value .
physiologically , the parameter @xmath66 has no direct relation , but serves as a mathematical plug to overcome the difficulty in modeling the drive directly from the dynamical states of a connecting node that take both positive and negative values .
the offset @xmath66 has another role : its value could determine the state - forgetting property of an evolving network ( see theorem [ thm_main ] ) . the particular choice of the saturation function @xmath54 is not critical to the model ; for the convenience of mathematical proofs , we need @xmath54 to be differentiable and unbounded and hence its form . on the other hand , for practical reasons ,
choosing such a @xmath54 , helps in the simulation stage to troubleshoot if each of the parameters @xmath64,@xmath65 and @xmath66 are chosen in an appropriate range lest the network lose the state - forgetting property . while the network does not have the state - forgetting property
, the dynamics can be unbounded in the absence of a saturation function since the range of @xmath39 is unbounded and the numerical values on a computer could indicate an error .
we present a short summary of the dynamics of the family of maps @xmath39 described in .
the detailed analysis of this family of maps is not the subject matter of this paper , and the discussion here is only intended to convey the point that the dynamics of @xmath75 adds notable features to the evolving network model .
recall that the parameters @xmath76 are restricted to @xmath41 and @xmath77 . with @xmath78
, it can be easily verified that there is an invariant closed interval @xmath79 symmetrically extending on either side of the real number @xmath37 given by @xmath80 $ ] , i.e. , @xmath81 .
the dynamics of @xmath39 on @xmath79 is relevant to us since the dynamics outside @xmath79 escapes to @xmath82 . with @xmath64
fixed , and increasing @xmath65 through @xmath37 to @xmath83 , the dynamics of this family of maps on @xmath79 takes a route to chaos qualitatively similar to the variation in @xmath65 from @xmath37 to @xmath84 in the well studied logistic family of maps @xmath85 on @xmath86 $ ] ( for e.g. , @xcite ) . elaborating this further , when @xmath87 , the fixed point is globally stable , in the sense that every orbit asymptotically approaches the fixed point as @xmath5 tends to @xmath88 ; when @xmath89 , the fixed point loses stability and undergoes a period doubling bifurcation . as @xmath65 is increased beyond @xmath90 , period doubling cascades occur alike in the logistic family of maps @xcite , and eventually there is a transition to chaos . thus each node has the potential to have self - excitatory dynamics ( oscillate or exhibit complicated behaviour ) if the stability parameter for the node @xmath65 is set between @xmath91 . adding a time varying drive @xmath92 to the update equation of @xmath39 by @xmath93 makes the mapping
depend on @xmath5 , and clearly the graph of such a mapping at any @xmath5 is a vertically shifted version of the graph of fig .
[ figa ] with the shift determined by @xmath92 . since
, we expect @xmath92 to be changing with every @xmath5 , the stable fixed point dynamics is not a feature of nonautonomous dynamics . as mentioned above ,
increasing @xmath65 beyond @xmath59 , makes the dynamics move away from the fixed point at @xmath37 .
we make use of the fact that to move the nonautonomous dynamics away from the fixed point , it is not necessarily required to increase @xmath65 beyond @xmath59 , but a sufficiently strong drive is capable of moving the dynamics . also , some background observations and computer simulations suggest that larger the magnitude of drive , further away is the likelihood of the dynamics away from @xmath37 . having made some points on the nodal dynamics , we now state a result on the evolving network .
next , we state a theorem which says that when @xmath66 is sufficiently large , then any evolving network obtained from is state - forgetting .
the complete proof of this result is provided in appendix [ appendixb ] .
[ thm_main ] fix non - zero , positive reals @xmath95 such that the following hold @xmath96 also , fix a saturation function @xmath54 as in equation with @xmath97 , and a time series @xmath98 such that @xmath99 $ ] for all @xmath100 and all @xmath101 .
there exists a real number @xmath102 such that for all @xmath103 , if holds for all @xmath100 and @xmath22 for some @xmath104 , then the resultant evolving network is state - forgetting .
henceforth , for ease of jargon we call the @xmath7 component of a @xmath36-dimensional time series data @xmath105 , i.e. , @xmath106 the data in the @xmath7 channel . given @xmath105 , we present a method for solving the weights @xmath6 in based on inferring a causal relationship between the nodes , which we call a synchrony measure . to explain the methodology of solving weights from , we rewrite by rearranging as @xmath107 where @xmath108 is the inverse of the saturation function @xmath54 .
clearly , given a @xmath36-dimensional time - series @xmath105 of eeg or ecog clinical data , and having fixed @xmath64 , @xmath65 in the model , the right hand side of can be computed easily for each @xmath5 from the @xmath7 channel data .
let @xmath109 denote the right hand side term in . for a fixed @xmath0
, we have a single equation @xmath110 where there are @xmath111 unknown quantities @xmath112 as @xmath1 is varied from @xmath59 to @xmath36 with @xmath113 .
hence , multiple solutions of @xmath6 could exist since we have more unknowns than equations .
we regard seizure spread in the network due to sustained energy transmitted from a node that is seizing to the ones which is not . with this principle
, we hypothesize @xmath6 to be an increasing parametric function of a certain synchrony measure @xmath114 between the two distinct nodes @xmath0 and @xmath1 .
physiologically , the synchrony measure between two nodes @xmath0 and @xmath1 represents the ease at which energy dissipates ( a measure of conductance ) from node @xmath1 into node @xmath0 , and hence is directed nodes @xmath0 and @xmath1 could have different interconnections , and energy dissipation between nodes is not symmetric . to get a measure of such conductance , we let the synchrony measure @xmath114 depend on ( i ) a sustained influence from node @xmath1 into node @xmath0 measured by a leaky integrator placed between the two nodes ( ii ) the similarity between the average amplitude or power in the two channels up to time @xmath5 . the algorithm for computation of directed synchrony @xmath114 between two channels is described in section [ sub_sec_alg ] .
we now explain the idea / reasoning behind some of the steps of the algorithm .
the discussion can be read in conjunction with the various steps of the algorithm .
the very first step of the algorithm involves normalisation of the amplitude of data for setting the amplitude of the time - series in any channel lies between @xmath115 $ ] ( see ( s1 ) of section [ sub_sec_alg ] ) . in a succeeding stage of the algorithm
, we compute the average power at each instant in a window of length @xmath116 in each channel ( see definition of @xmath117 in ( s6 ) of section [ sub_sec_alg ] ) .
it turns out that the time interval in which there is sustained influence of one node on the other by the synchrony measure is determined by @xmath116 .
we comment on the choice of @xmath116 in section [ sub_sec_practical ] the directed synchrony @xmath114 depends on the product of two factors ( see ( s7 ) of section [ sub_sec_alg ] ) .
one of the factors is the similarity between the powers in the @xmath7 and @xmath8 channels which we quantify by a similarity index @xmath118 .
note that due to normalisation in ( s1 ) , the average power @xmath117 in any window is between @xmath37 and @xmath59 , and hence @xmath119 , and the similarity index attains @xmath59 if and only if @xmath117 and @xmath120 are identical .
the other factor that contributes to @xmath114 is @xmath121 , the output of a ( directed ) nonlinear leaky integrator placed between two distinct nodes and whose states are evaluated recursively ( for details see ( s8 ) to ( s12 ) of section [ sub_sec_alg ] ) ) .
the purpose of such a leaky integrator is to model the sustained influence of the signal strength in the node @xmath1 on node @xmath0 the word sustained points to the fact that the integrator leaks , and unless the influence is not sustained over a period of time , the influence diminishes .
any such leaky integrator in ( s11 ) of section [ sub_sec_alg ] is a nonautonomous dynamical system in its own right , and hence its future states are obtained by iteration on the current state .
in essence , a leaky integrator sums up ( or integrates ) its previous state with what is fed to it typically as an input , but also leaks a small fraction of the sum while integrating . here , we set the leaky integrator dynamics for the connection from node @xmath1 to @xmath0 as @xmath122 where the cumulative average power @xmath123 in the @xmath8 channel is calculated as in ( s9 ) of the algorithm .
this cumulative average fractional power is determined by first calculating the fractional power in the @xmath8 channel ( see ( s8 ) of the algorithm ) .
the cumulative average of this fractional power is calculated as in ( s9 ) of the algorithm ; this cumulative average is calculated iteratively using the elementary idea : to calculate the average of a collection of @xmath124 numbers @xmath125 , if the average of @xmath126 numbers is given to be @xmath127 , then the average of @xmath124 numbers @xmath128 can be updated by @xmath129 .
once the synchrony @xmath114 from the time series from the algorithm in section [ sub_sec_alg ] is computed , then for convenience , we proceed to deduce the intermediate quantity @xmath130 from which the actual weight @xmath6 can be readily obtained ( see ) .
we hypothesise @xmath131 to be an increasing parametric function of @xmath114 and is in the form @xmath132 where the parameter @xmath133 can be solved as explained below . substituting @xmath134 in eq .
, we have the following straightforward deduction to get an expression for @xmath133 : @xmath135 once @xmath131 is found , the weight @xmath6 can be easily found since @xmath50 is invertible . from , we get @xmath136 the algorithm
is described in the following ordered steps ( s1 ) to ( s12 ) . 1 .
normalize or scale the time - series data @xmath105 , by dividing each component in each channel by @xmath137 so that upon normalisation the time - series satisfies : ( i ) every value of the time series in each channel lies between @xmath115 $ ] ( ii ) the maximum of absolute value of the normalised time - series @xmath105 ( retaining the same notation as the original time - series ) is @xmath59 , i.e. , @xmath138 .
2 . fix the parameters @xmath64 , @xmath65 and @xmath66 of the network , and a large positive integer @xmath116 .
it is suggestive that the parameters @xmath64 and @xmath65 are chosen to satisfy @xmath139 , @xmath140 and @xmath141 .
once these values are fixed , ( by theorem [ thm_main ] ) the value of the parameter @xmath66 determines the state - forgetting property of the evolving network .
the state - forgetting property can be cross - verified after simulations ( see figure [ fig : network_inf1 ] ) ; the minimum value of @xmath66 required increases with the number of nodes in the network . for instance , in our simulation results in section [ sec_results ] , with @xmath142 , the value of @xmath143 is found to be sufficient for the network to satisfy the state - forgetting property when @xmath144 , @xmath145 and @xmath146 .
3 . for all time @xmath147 , initialise the cumulative average power in each of the @xmath8 channel to be @xmath148 .
4 . at time @xmath149 ,
initialize , the leaky integrator state @xmath121 a random number @xmath150 for all @xmath151 .
5 . set a counter to an initial state , @xmath152 ( to aid a calculation in ( s9 ) ) .
obtain the average power in every channel @xmath1 of the time series at the time instant @xmath5 using the data in the window @xmath153 $ ] as @xmath154 ( note : clearly , since the data is normalized in ( s1 ) , @xmath155 ) 7 .
define the directed synchrony @xmath156 for all @xmath0 , @xmath1 running from @xmath157 with @xmath158 .
find the fraction of the power in each channel @xmath1 , using @xmath159 9 . for each @xmath1
, compute the cumulative average @xmath123 of the fractional power from @xmath160 and @xmath161 using @xmath162 10 .
denote @xmath163 , the quantity that is to be fed into the leaky integrator to obtain its next state .
the @xmath164 function when shifted up and to the right is a positive increasing map on @xmath165 with a large slope in a very short interval ; we use such a function to update the leaky integrator state according to @xmath166 12 .
if @xmath167 needs to be computed , set @xmath168 ; @xmath169 , and return to ( s6 ) .
@xmath170 we recapitulate from section [ sec_sfn ] , that only a state - forgetting network can attract all other iterates towards a unique entire solution . as a numerical verification of this ,
in section [ sec_results ] we show the ability of the evolving network model to generate a time series from an arbitrary initial condition that begins to resemble the original one in a very few time steps .
we infer cortical connectivity from a 58 year old male patient diagnosed at mayo clinic .
the patient had intractable focal seizures ( seizures that fail to come under control due to treatment ) with secondary generalization ( seizures that start in one area and then spread to both sides of the brain ) .
the patient underwent surgical treatment and pre - surgical evaluations were performed using a combination of history , physical exam , neuroimaging , and electrographic recordings . for delineating the epileptogenic tissues , electrocorticography ( ecog ) recording was performed at a sampling frequency of 500 hz for four days continuously with grid , strip , and depth electrodes implanted intracranially on the frontal , temporal and inter - hemispheric cortical regions .
a total of nine seizures were captured during the recording period and complete seizure freedom was achieved ( ilae class i ) following the resection of pathological tissues in the left medial frontal lobe and medial temporal structures .
all recordings were performed conforming to the ethical guidelines and under the protocols monitored by the institutional review board ( irb ) according to national institutes of health ( nih ) guidelines .
the clinical data and ecog recordings for this patient are publicly available on the ieeg portal ( https://www.ieeg.org ) with dataset i d ` study 038 ' . in our analysis , we consider only the electrodes in the @xmath171 grid placed on the frontal lobe where both the seizure onset and its spread were captured .
the electrodes which circumscribed the epileptogenic tissues are shaded in black on the schematic shown in fig .
[ fig : network_inf1](a ) .
a small segment of ecog recording is sufficient to demonstrate the efficacy of our modelling framework .
therefore , we select an ecog segment comprising of both pre - ictal and ictal activities to explore how the network evolves during seizures .
we remove the noise artefacts and the power - line interference from the raw ecog signals by applying a bandpass filter between @xmath172hz and a notch filter at @xmath173hz .
the ecog signals from a few exemplary electrodes ( labeled with a prefix lg ) are shown in fig .
[ fig : network_inf1](b ) . next , we infer an evolving network synthesized from the ecog data .
we consider the cortical area under each ecog electrode as a network node and apply algorithm detailed in section [ sub_sec_alg ] to get the directed and weighted edges or interconnections between the nodes so that is satisfied .
in particular with parameters @xmath174 , @xmath145 , @xmath143 , @xmath146 we apply the algorithm to obtain the weights of the interconnections between the nodes at each time sample of the data . to cross
validate that the weights obtained eventually render a state - forgetting evolving network , we adopt the following procedure . with an arbitrary initial condition @xmath175 and the obtained weights at time @xmath26 , we use to obtain @xmath176 . repeating this ,
we obtain @xmath177 , @xmath178 and so on .
the time - series in a few channels obtained from this procedure is shown in fig .
[ fig : network_inf1](d ) . apparently , the time - series in fig . [ fig : network_inf1](b ) and fig . [ fig : network_inf1](d ) are similar . actually , the similarity is true barring a few hundred samples at the beginning ( see fig .
[ fig : network_inf1](c ) ) .
this convergence of the model output from an arbitrary initial condition to the given time - series verifies state - forgetting property and validates our modelling framework for the study of network evolution during seizure onset and seizures . as a passing remark to the reader , the model s ability to sharply reconstruct the given data as indicated in fig .
[ fig : network_inf1](c ) ( plots of the first 2000 samples of an ecog channel data , corresponding model output and their difference ) is robust even for a different choice of parameters in the algorithm .
in fact , theorem [ thm_main ] tells us that such a reconstruction is possible as long as the parameters are in a certain range .
grid on the brain schematic of a patient electrodes within clinical onset zone are shaded in black ; ( b ) shows the time series of ecog recordings from a few exemplary electrodes ( channels ) labeled with a prefix lg ; ( d ) shows the model generated time - series with an arbitrary initial condition in those channels shown in ( b ) ; ( c ) zooms on one of the channels in ( d ) and over the first 2000 samples to indicate the convergence of the model output to the ecog recording from the model . ] with an objective of identifying the nodes crucial for the seizure genesis , we study the network properties before and after the seizure onset . consequently , we divide the ecog recordings into two segments : pre - ictal ( before seizure onset ) and ictal ( after seizure onset ) . for each segment of this data , we compute an ensemble average of node strengths as explained next
. it may be recalled from section [ sec_model ] that the drive from node @xmath1 to @xmath0 at time @xmath5 is given by @xmath67 .
for each node @xmath0 , we compute its mean incoming drive strength of node over a time period @xmath179 as @xmath180 where @xmath36 is the number of nodes , and @xmath181 is the number of sample points in @xmath179 . in a similar vein , we compute the mean outgoing drive strength of node @xmath0 over a time period @xmath179 as @xmath182 figure [ fig : segmentation_eeg ] shows the mean incoming and outgoing node strengths for each node during the pre - ictal and ictal periods .
[ fig : segmentation_eeg ] ( a ) indicates a strong correlation of the mean incoming node strengths with the clinical seizure onset zone or seizure focus ( depicted by the electrodes shaded in black ) during the pre - ictal period , while fig .
[ fig : segmentation_eeg ] ( b ) indicates a more uniform spread of mean incoming node strengths across all nodes during the ictal period .
the correlation of the mean incoming node strengths with the clinical seizure onset zone during the pre - ictal period can be understood as that the external drives from other nodes play a supportive role in initiating seizures ( a hypothesis of our model as well ) . since during the ictal period all channels seem to have similar average power due to seizures , all nodes are likely to receive a similar quantum of external drive .
hence , the mean incoming drive strength appears to be uniform .
in contrast , fig . [
fig : segmentation_eeg ] ( c , d ) shows that the distinctively higher mean outgoing node strength ( during both ictal and pre - ictal period ) is much strongly correlated " with the clinical seizure onset zone than that shown by the corresponding mean incoming node strengths .
therefore , from fig . [
fig : segmentation_eeg ] ( c , d ) , it is evident that the nodes whose node strengths are shown as columns in red ( i.e. , nodes @xmath183 ) are the outgoing hubs .
we refer to this set of nodes as the theoretical seizure onset zone .
incidentally in this experiment there are 12 nodes in the theoretical seizure onset zone , similar in cardinality to the set of the nodes at the site of clinical resection ( i.e. , nodes @xmath184 ) .
we refer to the nodes at the site of clinical resection as the clinical seizure onset zone .
it is evident that the theoretical seizure onset zone and the clinical seizure onset zone point to very similar regions in the brain .
thus , the model has the potential utility of identifying an approximate seizure focus zone prior to surgery by analysing some pre - ictal and ictal data . although the theoretical and clinical onset zones differ slightly , from a theoretical standpoint of view , we can hypothesise the outcome of the surgical removal of these two zones and compare these outcomes if we have a suitable objective measure . to explain this ,
let @xmath185 be the indices of the subset of the nodes that are omitted from a network of @xmath36 nodes in a simulation experiment , i.e , the nodes in @xmath185 are made non - functional by setting @xmath186 for all @xmath5 if either @xmath0 or @xmath1 belongs to @xmath185 .
using such modified weight matrices in we can obtain a time - series @xmath187 of the depleted network can be calculated by starting with any initial condition that satisfies @xmath188 whenever @xmath189 . in other words ,
the signal has to have zero amplitude in all the channels corresponding to the nodes in @xmath185 since they are assumed to not participate in the network dynamics .
we quantify the efficacy of the removal of the nodes that belong to @xmath185 by calculating the ratio of the average power in those channels outside @xmath185 prior to removal of nodes and after it by : @xmath190 where @xmath181 denotes the number of samples over a time - interval @xmath179 and @xmath191 denotes the cardinality of the set @xmath185 .
similarly the average power in the signal generated from the depleted network can be calculated by @xmath192 and we calculate the efficacy of removing the nodes @xmath185 by the ratio @xmath193 we tabulate the values of @xmath194 below , where @xmath185 is determined by the following choices : ( i ) the clinical seizure onset zone ( ii ) theoretical seizure onset zone ( outgoing hubs ) ( iii ) an arbitrary set of 12 nodes ( iv ) 12 nodes randomly drawn ( average of @xmath194 is tabulated in this case ) .
now , we discuss the potential application of the model in a clinical study when surgery is part of the treatment
. the aim of a clinical surgery would be to a minimise the removal of tissue from the brain along with subsiding seizures post surgery . in our study
above , the removal of a set of nodes @xmath185 from the evolving network is intended to simulate a surgical removal of the tissue beneath the nodes in @xmath185 . a time series @xmath187 from the depleted network is intended to simulate the resultant clinical ecog time series observed after surgery , and the quantity @xmath194 is meant to measure the efficacy of the surgery . thus prior to an actual surgery ,
the theoretical efficacy of surgical removal of nodes can be studied by using the corresponding depleted network . from the tabulated values of @xmath194
, we observe that the removal of nodes from the outgoing hubs that form the theoretical onset zone maximises the efficacy of a surgery when ( the tissue beneath ) a set with 12 nodes is assumed to be removed in each simulation .
in fact the large value of @xmath194 obtained after the removal of the nodes in the theoretical onset zone translates to complete subsiding of seizures in the resultant time - series ( not shown here ) generated from the depleted network .
when a small subset of the outgoing hubs were retained in the network , @xmath194 was still observed to be smaller and the seizures continued to persist from the resultant time - series ( not shown here ) generated from the depleted networks . from this ability of outgoing hubs to subside seizures in the rest of the network upon their omission , we also refer the outgoing hubs to as spreaders . for a reader interested in finding an optimal set of nodes for subsiding seizures , we note that one does not have to necessarily aim to get a very large value of @xmath194 , but any @xmath194 that yields a seizure free time - series is adequate . [ cols="<,<,<",options="header " , ] we remark that in our simulations , the mean incoming ( outgoing ) node strengths and the mean incoming ( outgoing ) weights of a node exhibit similar variation across nodes , but we employ the former as it accounts for more accurate interaction between two nodes since it subsumes both the effect of the weight of the interconnection and the signal strength of the node that connects .
the parameter @xmath116 determines the length of the windows in which average powers are compared ( in the algorithm in section [ sub_sec_alg ] ) while computing the synchrony measure .
since the power during the inter - ictal period ( period between two distinct seizures ) is negligible in all channels and seizures are relatively short epochs , if @xmath116 is made extremely large then the synchrony measures would not show distinct variation across nodes , and we could end up with no useful network information as the weights lose physical significance .
it is required to set @xmath116 in a range so that some noticeable pre - ictal activity or larger power in the data that typically occurs ( e.g.,@xcite ) just prior to seizures is captured in the synchrony measure . in this paper , we set @xmath195 so that with a sampling rate of @xmath196 hz we use about 10 seconds of data for such a power comparison . however , it turns out that the mean node strengths show similar variation even if @xmath116 is increased in a range of up to @xmath197 , but a large @xmath116 increases computational complexity . in conventional machine learning problems ,
complicated models are avoided fearing an over - fitted model .
an over - fitted network model is usually a static network with a very large number of nodes and it tends to model the random noise or anomalies in the data rather than the underlying relationship .
we evaluate the sensitivity of the weight computation algorithm or the model by analysing the change in computed weights with additive artefact noise to the data ; we use weights rather than node strengths since we are analysing the model sensitivity and not evaluating a seizure focus . by adding realizations of zero mean iid noise having uniform distribution to the above analysed patient s clinical ecog data ( taking values in @xmath115 $ ] ) , we plot in fig .
[ figic ] and fig .
[ figoc ] the normalised average incoming and outgoing connections for different noise variances the average weights are normalised to have a maximum average value of @xmath59 for the purpose of comparison across different noise variances .
the pattern in the variation of the average weights has many similar features even when the noise variance is high ( fourth panel in fig .
[ figic ] and fig .
[ figoc ] ) .
with such a high variance of noise , we observe from fig .
[ figoc ] that all the nodes in the theoretically determined focus are still outgoing hubs .
similar robustness of the pattern in normalised weights were observed when the data was quantised or when the parameters ( @xmath64 , @xmath65 and @xmath66 ) in our algorithm were varied .
[ figicoc ]
lastly , we remark that the model is flexible to changes and can be modified to work as long as the state - forgetting property is satisfied . in particular , the dynamical system at each node in and inferring causality in the algorithm in [ sub_sec_alg ] can be made more sophisticated or changed to be based on a deeper or alternate physiological understanding of how epileptic seizures are generated .
in this paper , we have presented a modeling approach to transform data into an evolving network that has the state forgetting property or robustness to initial conditions .
although , this principle of transformation can be applied to data originating from any source , the particular evolving network model concerned an epileptic network .
although time - series can be modeled with reasonable accuracy by static networks of very large dimension such as recurrent neural networks , the individual nodes in such case have no physical significance and they do not correspond to the regions in the brain .
the interconnection - strengths between the nodes were solved via a causal relationship inferred from the sustained power similarity between two nodes .
since the model can produce a sharp reconstruction of the time - series , there is flexibility to study the interconnections in the evolving network in conjunction with the value of the time series at any time point , or over a time - interval .
we have observed from modeling clinical data of an epileptic patient that certain nodes turn into outgoing hubs during the pre - ictal period and persist through the ictal period .
these outgoing hubs also have a location in the brain similar to that of the clinically identified seizure onset zone ( seizure focus ) .
this hints that the ( pathological ) seizure causing areas in the brain could be having strong outgoing connections during seizure onset and during seizures .
further , to quantify the influence of such outgoing hubs on the network dynamics , we have presented a result quantifying the theoretical efficacy of removal of such nodes in subsiding seizures .
of course , the result is based on data of a single patient , but then a comprehensive study of the effect of a removal of any node in an evolving network has to be backed up with a justification since altering even a single interconnection could alter the entire network dynamics , a reliable model should show small changes in the network dynamics for small changes in the weights . in a sequel paper , we present a mathematical justification for inferring network information from a depleted network where we establish that changing the interconnections results in a continuous change in the time series whenever the evolving network is state - forgetting . such a result is needed to justify a more detailed study of the theoretical efficacy of removal of nodes .
in fact , it is possible to prove a very strong result showing that the effect of the removal of a node can be studied reliably if and only if a ` nontrivial ' evolving network has state - forgetting property .
the author would like to thank nishant sinha at nanyang technological university for assistance in presenting plots in section [ sec_results ] and some discussions there in .
the author thanks the team of the ieeg portal ( https://www.ieeg.org ) for providing access to the ecog data .
access to the image(s ) and/or data was supported by award number u24ns06930 from the u.s .
national institute of neurological disorders and stroke .
the content of this publication / presentation is solely the responsibility of the author , and does not necessarily represent the official views of the u.s .
national institute for neurological disorders and stroke or the u.s .
national institutes of health .
the author thanks an anonymous referee for pointing out several typographical errors .
benjamin , o. , fitzgerald , t. , ashwin , p. , tsaneva - atanasova , k. , chowdhury , f. , richardson , m. p. , and terry , j. r. ( 2012 ) .
a phenomenological model of seizure initiation suggests network structure may explain seizure frequency in idiopathic generalised epilepsy .
, 2(1):1 .
bertram , e. h. , mangan , p. , fountain , n. , rempe , d. , et al .
functional anatomy of limbic epilepsy : a proposal for central synchronization of a diffusely hyperexcitable network .
, 32(1):194205 .
bonilha , l. , nesland , t. , martz , g. u. , joseph , j. e. , spampinato , m. v. , edwards , j. c. , and tabesh , a. ( 2012 ) .
medial temporal lobe epilepsy is associated with neuronal fibre loss and paradoxical increase in structural connectivity of limbic structures . , pages jnnp2012 .
bragin , a. , wilson , c. , and engel , j. ( 2000 ) .
chronic epileptogenesis requires development of a network of pathologically interconnected neuron clusters : a hypothesis .
, 41(s6):s144s152 .
breakspear , m. , roberts , j. , terry , j. r. , rodrigues , s. , mahant , n. , and robinson , p. ( 2006 ) . a unifying explanation of primary generalized seizures through nonlinear brain modeling and bifurcation analysis .
, 16(9):12961313 .
devaney , r. l. ( 1989 ) . .
addison - wesley .
diessen , e. , diederen , s. j. , braun , k. p. , jansen , f. e. , and stam , c. j. ( 2013 ) .
functional and structural brain networks in epilepsy : what have we learned ?
, 54(11):18551865 .
elaydi , s. ( 2005 ) . .
springer science & business media .
fitzhugh , r. ( 1969 ) .
mathematical models of excitation and propagation in nerve . in _ biological engineering _ , pages 185 . mcgraw - hill , new york .
furi , m. and martelli , m. ( 1991 ) . on the mean value theorem , inequality , and inclusion .
, pages 840846 .
hutchings , f. , han , c. e. , keller , s. s. , weber , b. , taylor , p. n. , and kaiser , m. ( 2015 ) . predicting surgery targets in temporal lobe epilepsy through structural connectome based simulations . , 11(12):e1004642 .
jaeger , h. ( 2001 ) .
the `` echo state '' approach to analysing and training recurrent neural networks - with an erratum note .
, 148:34 .
jansen , b. h. and rit , v. g. ( 1995 ) .
electroencephalogram and visual evoked potential generation in a mathematical model of coupled cortical columns .
, 73(4):357366 .
jirsa , v. k. , stacey , w. c. , quilichini , p. p. , ivanov , a. i. , and bernard , c. ( 2014 ) . on the nature of seizure dynamics . , 137(8):22102230 .
kalitzin , s. n. , velis , d. n. , and da silva , f. h. l. ( 2010 ) .
stimulation - based anticipation and control of state transitions in the epileptic brain .
, 17(3):310323 .
khosravani , h. , mehrotra , n. , rigby , m. , hader , w. j. , pinnegar , c. r. , pillay , n. , wiebe , s. , and federico , p. ( 2009 ) . spatial localization and time - dependant changes of electrographic high frequency oscillations in human temporal lobe epilepsy . , 50(4):605616 .
kloeden , p. , ptzsche , c. , and rasmussen , m. ( 2013 ) .
discrete - time nonautonomous dynamical systems . in _ stability and bifurcation theory for non - autonomous differential equations _ , pages 35102 . springer .
kloeden , p. e. and rasmussen , m. ( 2011 ) . .
number 176 .
american mathematical soc .
, providence .
kramer , m. a. and cash , s. s. ( 2012 ) .
epilepsy as a disorder of cortical network organization .
, 18(4):360372 .
lemieux , l. , daunizeau , j. , and walker , m. c. ( 2011 ) .
concepts of connectivity and human epileptic activity .
, 5:113 .
maass , w. , natschlger , t. , and markram , h. ( 2002 ) .
real - time computing without stable states : a new framework for neural computation based on perturbations .
, 14(11):25312560 .
manjunath , g. and jaeger , h. ( 2013 ) .
echo state property linked to an input : exploring a fundamental characteristic of recurrent neural networks .
, 25(3):671696 .
manjunath , g. and jaeger , h. ( 2014 ) .
the dynamics of random difference equations is remodeled by closed relations .
, 46(1):459483 .
marten , f. , rodrigues , s. , suffczynski , p. , richardson , m. p. , and terry , j. r. ( 2009 ) .
derivation and analysis of an ordinary differential equation mean - field model for studying clinically recorded epilepsy dynamics .
, 79(2):021911 .
morgan , r. j. and soltesz , i. ( 2008 ) .
nonrandom connectivity of the epileptic dentate gyrus predicts a major role for neuronal hubs in seizures .
, 105(16):61796184 .
ngugi , a. k. , bottomley , c. , kleinschmidt , i. , sander , j. w. , and newton , c. r. ( 2010 ) .
estimation of the burden of active and life - time epilepsy : a meta - analytic approach .
, 51(5):883890 .
richardson , m. p. ( 2012 ) .
large scale brain models of epilepsy : dynamics meets connectomics . , 83(12):12381248 .
robinson , p. , rennie , c. , and rowe , d. ( 2002 ) .
dynamics of large - scale brain activity in normal arousal states and epileptic seizures . , 65(4):041924 .
rocsoreanu , c. , georgescu , a. , and giurgiteanu , n. ( 2012 ) . ,
volume 10 .
springer science & business media .
schmidt , h. , petkov , g. , richardson , m. p. , and terry , j. r. ( 2014 ) .
dynamics on networks : the role of local dynamics and global networks on the emergence of hypersynchronous neural activity .
, 10(11):e1003947 .
silva , f. h. , blanes , w. , kalitzin , s. n. , parra , j. , suffczynski , p. , and velis , d. n. ( 2003 ) .
dynamical diseases of brain systems : different routes to epileptic seizures .
, 50(5):540548 .
spencer , s. s. ( 2002 ) .
neural networks in human epilepsy : evidence of and implications for treatment . , 43(3):219227 .
steil , j. j. ( 2004 ) .
backpropagation - decorrelation : online recurrent learning with o ( n ) complexity . in _
neural networks , 2004 . proceedings .
2004 ieee international joint conference on _ , volume 2 , pages 843848 .
suffczynski , p. , kalitzin , s. , and da silva , f. l. ( 2004 ) .
dynamics of non - convulsive epileptic phenomena modeled by a bistable neuronal network . , 126(2):467484 .
taylor , p. n. , goodfellow , m. , wang , y. , and baier , g. ( 2013 ) . towards a large - scale model of patient - specific epileptic spike - wave discharges .
, 107(1):8394 .
terry , j. r. , benjamin , o. , and richardson , m. p. ( 2012 ) .
seizure generation : the role of nodes and networks .
, 53(9):e166e169 .
wendling , f. , bartolomei , f. , bellanger , j. , and chauvel , p. ( 2002 ) .
epileptic fast activity can be explained by a model of impaired gabaergic dendritic inhibition .
, 15(9):14991508 .
wendling , f. , hernandez , a. , bellanger , j .- j . , chauvel , p. , and bartolomei , f. ( 2005 ) .
interictal to ictal transition in human temporal lobe epilepsy : insights from a computational model of intracerebral eeg .
, 22(5):343 .
when interconnections change with time in a network of dynamical systems , the overall dynamics of the entire network is nonautonomous
autonomous dynamics is rendered when systems evolve according to a rule fixed for all time , and when it is not fixed ( here , due to the time - varying interconnections ) , nonautonomous dynamics emerges ( see section [ sec_sfn ] for definitions ) .
the evolution of an autonomous system depends on only the initial condition presented to it , whereas in a nonautonomous system the evolution depends both on the initial condition and the time at which it is presented .
the evolution of an autonomous system in discrete - time can be described by an update equation @xmath198 , where @xmath199 is a function governing the rule for evolution .
the evolution of a nonautonomous system can be described by an update equation @xmath200 , where a family of maps @xmath201 govern the rule of evolution .
we explain the advantage of considering a nonautonomous model .
consider a two - dimensional autonomous dynamical system which updates its dynamics according to the rule @xmath202 ( @xmath199 and @xmath203 are any functions whose domain and range are @xmath204 and @xmath165 ) to generate a @xmath90-dimensional time series @xmath205 .
suppose in reality , only the first coordinate or dimension of the time series , i.e , @xmath31 is observable .
then , in general it is not possible to model the time - series @xmath31 as the dynamics of a one - dimensional dynamical system , in other words it is not possible to find a function ( map ) @xmath206 so that @xmath207 holds for all @xmath5 since @xmath208 inherently depends on @xmath209 .
however , it is possible to find a nonautonomous system , i.e. , a family of maps @xmath210 so that @xmath211 . in general , whenever we are unable to completely observe all the variables from a dynamical system ( including nonautonomous systems )
it is possible to synthesise a nonautonomous system that models the observed variables . in real world systems ,
the non - autonomous system can account for obfuscated factors such as the unfeasibility of observing all variables , exogenous influence etc . by subsuming their net - effect by a collection or family of maps that vary with time .
mathematically , viewing an evolving network as a static network being acted upon by an exogenous stimulus that influences the network connections at each time step gives us the advantage of employing the formal framework of the echo - state static networks theory in machine learning @xcite .
we recall some notions @xcite and adapt to our setup .
we begin now with a formal description . here , and throughout , only the euclidean distance is employed as a metric / distance on any two elements of @xmath212 .
also @xmath213 represents a nondegenerate compact interval of @xmath165 , and @xmath214 denotes the @xmath36-times product @xmath215 . following @xcite and several others ( e.g. , @xcite ) , we define a _ discrete - time nonautonomous system _ on a metric space @xmath216 as a sequence of maps @xmath19 , where each @xmath217 is a continuous map .
any sequence @xmath218 that satisfies @xmath219 for all @xmath101 is called an entire - solution of @xmath19 . for the discussion in this paper it is sufficient to consider an evolving network with each of the nodes having a one - dimensional dynamical system on an interval @xmath213 , i.e. , for an isolated node
, there is some function mapping @xmath213 to @xmath213 as it is an autonomous system .
we formally define an evolving network below : [ def_en ] let @xmath213 be a compact interval of @xmath165 , and let @xmath220 be an integer .
suppose a function @xmath221 has a real - valued @xmath222 matrix @xmath3 , and a value in @xmath214 as its arguments and maps it to a value in @xmath214 for each @xmath101 , i.e. , @xmath223 is such that 1 .
@xmath221 is continuous and 2 .
@xmath221 is defined by a set of @xmath36 ( continuous ) coordinate maps @xmath224 such that @xmath225 , where @xmath226 represents the @xmath7 row of a matrix @xmath3 , and @xmath227 the @xmath7 component of @xmath228 ; here @xmath10 generates the node dynamics at the @xmath7 node , then such a map @xmath221 is called a _ family of evolving network_.
if in particular , a bi - infinite sequence of matrices @xmath229 is fixed , then the sequence of mappings @xmath230 is referred to as an evolving network . clearly , any evolving network @xmath230 is also a nonautonomous system @xmath19 on @xmath214 if we set @xmath231 for all @xmath5 .
we call @xmath3 the connectivity weight matrix or just the weight matrix .
we next define the state - forgetting property of a family of evolving networks and also for a particular evolving network .
[ def_sf ] let @xmath232 be a family of evolving networks .
if for each choice of @xmath233 , the evolving network @xmath230 has exactly one entire - solution , then the family of evolving networks @xmath221 is said to be state - forgetting . in particular , for any fixed @xmath233 , the corresponding evolving network ( or the nonautonomous system ) @xmath234
is said to be _ state - forgetting _ if it has exactly one entire - solution .
we first recall some preliminaries and known results that help us in presenting the proof of theorem [ thm_main ] .
it is convenient to have the following alternate representation of a nonautonomous system .
let @xmath235 the collection of all integer tuples @xmath236 so that @xmath237 , i.e. , @xmath238 . given a nonautonomous system @xmath19 on @xmath216 , it is convenient ( see @xcite ) to denote the nonautonomous system through a function @xmath239 called a _ process _ on @xmath216 that satisfies : @xmath240 and @xmath241 we say @xmath50 has the state - forgetting property if @xmath19 has the state - forgetting property .
[ lemma_backward_contraction ] let @xmath242 be a process on a compact metric space @xmath216 metrized by @xmath243 .
suppose that , for all @xmath22 , there exists a sequence of positive reals @xmath244 converging to @xmath37 such that @xmath245 for all @xmath246 and for all @xmath247
. then @xmath50 has the state - forgetting property , i.e. , there is exactly one entire solution of the process @xmath242 .
the essence of the following result is that if @xmath50 has the state - forgetting property then all initial conditions converge to the unique entire - solution in proposition [ prop_pullback ] . for a detailed explanation
see @xcite and references therein .
[ prop_pullback ] let @xmath242 be a process on a compact metric space @xmath216 metrized by @xmath243 .
suppose @xmath20 is the unique entire - solution of @xmath242 , then @xmath248 for all @xmath249 and @xmath5 . a generalization of the mean - value theorem in one - dimensional calculus to higher dimensions results in the following mean - value inequality ( e.g. , @xcite ) .
this result can be stated as : if @xmath255 is a @xmath256-function where @xmath257 is an open subset of @xmath258 then for any @xmath259 @xmath260 where @xmath261 is the @xmath262-norm , and @xmath263 is the induced norm of the jacobian of @xmath264 at the point @xmath265 . to simplify the notations in the proof of theorem [ thm_main ] , whenever @xmath266 , we adopt the notation @xmath267 and + @xmath268 , where @xmath269 and @xmath270 are column vectors and the superscript @xmath271 represents the transpose of the vector @xmath272 .
* proof of theorem [ thm_main ] . *
fix a @xmath104 that satisfies for all @xmath273 and @xmath101 .
let @xmath274 be defined by @xmath275 , where @xmath50 is as in .
since fixing a @xmath233 or a @xmath274 ( @xmath131 related to @xmath6 by ) amounts to fixing an evolving network , we have fixed a nonautonomous system .
@xmath276 where @xmath277^n$ ] , @xmath278 and @xmath279 + @xmath280 , with @xmath281 as the column vector @xmath282 .
since the function @xmath54 in , ensures @xmath283 , we have @xmath284 since the choice of @xmath233 was made arbitrarily , if we show that @xmath242 has exactly one - entire solution then we would have shown that the evolving network is state - forgetting by definition [ def_sf ] . since @xmath50 is a process on @xmath115^n$ ] , a compact subset of @xmath285 , we make use of lemma [ lemma_backward_contraction ] to show @xmath242 has exactly one - entire solution .
we employ the metric induced by the norm @xmath286 for getting an inequality of the form in lemma [ lemma_backward_contraction ] .
let @xmath287 be any real number in @xmath288 .
we first aim to show that there exists a @xmath289 ( that depends on @xmath287 ) such that for all @xmath103 , the following inequality is satisfied .
@xmath290^n.\end{aligned}\ ] ] we next proceed to prove @xmath293 for all @xmath294^n$ ] .
recall that @xmath295 .
we first consider the first term on the right hand side , i.e. , @xmath296 and prove an upper bound for @xmath297 using the mean value theorem .
since @xmath39 is differentiable and has a continuous derivative , by applying the mean value theorem on the domain @xmath115 $ ] we infer : @xmath298\ ] ] by hypotheses , @xmath299 and @xmath300 . if @xmath301 , it is straightforward to deduce the following equivalent statements : @xmath299 @xmath302 @xmath303 for all @xmath304 @xmath302 @xmath305 for all @xmath304 ( since @xmath306 ) @xmath302 @xmath307 for all @xmath304 @xmath302 @xmath308 for all @xmath304 since the derivative @xmath309 .
since @xmath310 is continuous , @xmath308 for all @xmath304 implies @xmath311 for all @xmath312 $ ] . using this in
, it follows that @xmath313.\ ] ] now , @xmath314 let @xmath315 be such that @xmath316 .
since @xmath317 , the jacobian @xmath318 evaluated at @xmath319 is a diagonal matrix and the @xmath7 diagonal element of this matrix is found to be @xmath320 the following deductions give an upper bound for @xmath321 : @xmath322 finally , and yields the desired bound in . applying iteratively , we get for any @xmath324 , @xmath325 . denoting @xmath326 in the inequality in the statement of lemma [ lemma_backward_contraction ]
it follows that there is exactly one entire - solution for @xmath242 .
this proves the theorem . | in many realistic networks , the edges representing the interactions between the nodes are time - varying .
there is growing evidence that the complex network that models the dynamics of the human brain has time - varying interconnections , i.e. , the network is evolving . based on this evidence ,
we construct a patient and data specific evolving network model ( comprising discrete - time dynamical systems ) in which epileptic seizures or their terminations in the brain are also determined by the nature of the time - varying interconnections between the nodes .
a novel and unique feature of our methodology is that the evolving network model remembers the data from which it was conceived from , in the sense that it evolves to almost regenerate the patient data even upon presenting an arbitrary initial condition to it .
we illustrate a potential utility of our methodology by constructing an evolving network from clinical data that aids in identifying an approximate seizure focus nodes in such a theoretically determined seizure focus are outgoing hubs that apparently act as spreaders of seizures .
we also point out the efficacy of removal of such spreaders in limiting seizures .
* keywords : mathematical modeling , time - varying network , state forgetting property , focal epilepsy , discrete - time dynamical system , nonautonomous dynamical system * |
the vast majority of galactic globular clusters ( gcs ) host multiple stellar populations ( mps ) characterized by different abundance ratios of selected light elements ( see , e.g. , * ? ? ? * for a review ) : some stars share the same light element abundance ratios measured in galactic field stars with similar metallicity , but a large fraction of the cluster population has enhanced n , na , and al and depleted c and o abundances .
the patterns are not random , but anticorrelated variations of the pairs c - n and o - na are commonly observed .
these are generally considered to arise from hot hydrogen burning in a previous generation of more massive stars , as asymptotic giant branch ( agb ) stars @xcite , fast - rotating massive stars @xcite , interacting massive binary stars @xcite , and/or super - massive stars @xcite .
objects with standard composition are commonly denoted as first generation ( fg ) stars , and those with modified chemistry as second generation ( sg ) stars , although the assumption that they are formed in subsequent star formation episodes is sometimes questioned ( see , e.g. , * ? ? ?
* ) . in a few gcs
the sg / fg star ratio measured along the red giant branch ( rgb ) is observed to differ from that measured along the agb , with a substantial deficiency of sg stars within the agb population , compared to the rgb @xcite . in principle , this can be explained by taking into account that stars with evolving masses below @xmath0 are expected to fail reaching the agb phase ( the so - called _ agb - manqu _ stars ; see , e.g. , @xcite ) and sg stars are indeed expected to have a lower mass along the hb with respect to fg stars . in fact , since they are typically he - enhanced , they originate from rgb stars with a lower mass and end up , on average , with a lower mass along the hb , if the rgb mass loss is approximately the same for fg and sg sub - populations ( see e.g. * ? ? ? * ) .
one therefore expects that the agb of gcs with a blue hb should lack at least part of the sg component , compared to what is seen along the rgb .
this is consistent with the findings of @xcite , who empirically showed that the number ratio between agb and hb stars ( the @xmath1 parameter ) correlates with the hb morphology , in the sense that clusters with the bluest hb morphology have lower @xmath1 values .
ngc 6752 is a metal - intermediate gc with an extreme blue hb morphology and a low @xmath1 value , and it is claimed to be the most extreme case of a gc lacking sg stars along the agb .
in fact , ( * ? ? ?
* hereafter c13 ) measured the na abundance of 20 agb stars in this cluster and from the derived [ na / fe ] distribution , they concluded that all objects belong to the fg population . in their interpretation ,
the sg stars fail to reach the agb phase because their hb progenitors are all located at effective temperatures ( @xmath2 ) hotter than the grundahl jump ( at @xmath311 500 k ) and experience a very strong mass loss ( a factor of 20 larger than that suffered along the rgb ) .
an alternative solution has been proposed by @xcite , who argued that the lack of sg agb stars can be explained within the fast - rotating massive stars scenario by assuming very high he abundances ( up to @xmath40.7 ) for the sg objects , that therefore become _ agb - manqu _ stars . on the other hand , by using detailed synthetic hb simulations , @xcite were able to reproduce the star distribution along the hb of ngc 6752 and its observed @xmath1 value assuming the initial he - abundance distribution derived from the cluster main sequence ( @xmath5 between @xmath30.25 [email protected] ; see * ? ? ?
* ) without invoking any extreme hb mass loss or initial he enhancement .
however , these simulations show that @xmath6 of the agb population should be composed of sg stars , at odds with the claim by c13 . with the aim of solving this intriguing issue , here we present the chemical abundances of iron and several light elements that we recently determined from high - resolution spectra for the same sample of agb stars discussed in c13 .
the 20 agb stars in ngc 6752 previously studied by c13 have been re - observed ( program 095.d-0320(a ) , pi : mucciarelli ) with the uves spectrograph @xcite mounted at the eso - very large telescope .
we used the dichroic1 mode adopting the gratings 390 blue arm cd#2 and 580 red arm cd#3 with the 1 arcsec slit ( r= 40000 ) .
exposure times range from @xmath310 min for the brightest targets to @xmath325 min for the faintest ones , to obtain pixel signal - to - noise ratios higher than 100 .
the data reduction was performed by using the dedicated eso pipeline , including bias subtraction , flat fielding , wavelength calibration , spectral extraction and order merging .
the chemical analysis has been performed following the same procedure described in @xcite .
the stellar atmospheric parameters have been derived as follows : + ( 1 ) @xmath2 have been derived spectroscopically by requiring no trend between iron abundances and excitation potentials ; + ( 2 ) surface gravities ( log@xmath7 ) have been obtained through the stefan - boltzmann relation , adopting the spectroscopic @xmath2 , the distance modulus ( @xmath8)@xmath9 = 13.13 and color excess e(@xmath10 ) = 0.04 @xcite , and a mass of 0.61 m@xmath11 , according to the median value of the hb mass range estimated by @xcite0.55 m@xmath11 ) median mass .
the adoption of this value decreases log g by @xmath30.04 , with a negligible impact on the abundances , @xmath30.02 for [ feii / h ] and smaller than 0.01 dex for the other species . ] .
stellar luminosities have been calculated using the bolometric corrections by @xcite and the @xmath12-band magnitudes from the ground - based photometric catalog reduced and calibrated following the procedures described in @xcite ; + ( 3 ) microturbulent velocities ( v@xmath13 ) have been obtained by requiring no trend between iron abundances and line strengths .
the derived values of @xmath2 and v@xmath13 well agree with those by c13 , with average differences @xmath14 = @xmath15 k and @xmath16 = @xmath17 km s@xmath18 .
for log@xmath7 there is a systematic difference @xmath19 = @xmath20 dex , probably due to the different distance modulus and the larger stellar mass adopted by c13 .
the abundances of fe , na , mg , and al have been derived using the classical method of the equivalent widths ( ew ) with the package gala @xcite .
ews have been measured by means of the daospec package @xcite , iteratively launched with the code 4dao@xcite .
the linelist was built using a synthetic reference spectrum calculated at the uves resolution and selecting only transitions predicted to be unblended .
we adopted the atomic data of the last release of the kurucz / castelli compilation for all species except for feii lines , which have been taken from @xcite .
the adopted model atmospheres have been computed with the atlas9 code adopting a global metallicity of [ m / h ] = @xmath21 dex .
the abundances of na have been corrected for nlte effects according to @xcite and consistently with the analysis of c13 . for 7 stars
the al lines at 6696 - 6698@xmath22 are too weak to be detected and only upper limits can be obtained .
the abundances of c , n and o have been measured through the spectral synthesis technique , using the forbidden oxygen line at 6300@xmath22 , and the ch and cn molecular bands at 4300@xmath22 and 3880@xmath22 , respectively . to derive the abundance of n we have taken into account the abundance of carbon measured from the ch band , while for the o abundance we adopted the average c and n abundances thus obtained , together with the measured abundance of ni .
this was done to take into account the close blending of the o line at 6300@xmath22 with a ni transition .
we also checked that the o transition is free from telluric contamination in 19 out of 20 agb stars .
for the star 1620 the contamination is severe and we did not derive the o abundance . as reference solar abundances we assumed those of @xcite except for c , n and o , for which we assumed the values of @xcite .
the computation of the final abundance uncertainties adds in quadrature two terms .
the first is the error arising from spectral features measurements .
for the abundances derived from ews , this term is obtained for each star by dividing the line - to - line dispersion by the square root of the number of lines used . for the elements analyzed with spectral synthesis
, the fitting procedure is repeated for a sample of 500 synthetic spectra where poissonian noise has been injected to reproduce the noise conditions ( see * ? ? ?
the second term is the abundance error arising from atmospheric parameters .
this has been computed by varying each parameter by its 1@xmath23 uncertainty obtained in the analysis . due to the quality of the spectra we found that the typical internal uncertainties for @xmath2 are lower than @xmath335k , while for log @xmath24 and v@xmath13 we found values lower than 0.1 dex and 0.05 km s@xmath18 , respectively .
the derived iron abundance ratios are listed in table [ tab1 ] together with the stellar atmospheric parameters .
we obtain average [ fei / h ] = @xmath25 dex ( @xmath23 = 0.05 dex ) and [ feii / h ] = @xmath26 dex ( @xmath23 = 0.02 dex ) .
the average [ feii / h ] abundance is consistent with the values measured in rgb stars by @xcite , while [ fei / h ] is 0.22 dex lower than the metallicity inferred from fe ii lines .
such a discrepancy between [ fei / h ] and [ feii / h ] among agb stars is too large to be explained within internal uncertainties and has been observed previously in other gcs @xcite .
the same [ fei / h]-[feii / h ] discrepancy remains also if we adopt the atmospheric parameters quoted in c13 .
note that c13 do not measure directly the fe abundance , but assume the average rgb [ fe / h ] by @xcite for all the targets . with their atmospheric parameters
we derive [ fei / h][email protected]@xmath280.01 dex ( @xmath23= 0.05 dex ) and [ feii / h][email protected]@xmath280.01 dex ( @xmath23= 0.02 dex ) .
even if a complete explanation of this effect is still lacking , this _ iron discrepancy _ seems to be a general feature of agb stars in gcs .
significant inhomogeneities in the light element abundances of the studied agb stars are immediately apparent already from the visual inspection of the acquired spectra .
this can be appreciated in figure [ spectra ] , where the ch and cn molecular bands , and o , na , mg and al lines of star 44 and star 65 ( having very similar atmospheric parameters ; see table [ tab1 ] ) are compared .
apart from mg , notable differences in the line strength are well visible for all the other elements .
moreover , the strength of the c and o features appears to anticorrelate with the strength of the n and na lines .
this clearly shows that the two stars are highly inhomogeneous in their light element content .
the abundance ratios obtained for the entire agb sample are listed in table [ tab1 ] .
following the approach discussed in @xcite and @xcite , the abundance ratios have been computed by adopting [ fei / h ] as reference but for o , for which we used [ feii / h ] .
this method provides the best agreement between the abundance ratios in agb and rgb stars of the same cluster .
however , because the origin of the fei - feii discrepancy is still unclear , we will discuss the abundances of agb stars with respect to both hydrogen and iron , to ensure that our results are independent of the adopted normalization of the abundance ratios . with the only exception of mg , for which we find values confined within a narrow range , the abundances of all the other light elements show dispersions well exceeding the internal errors ( see table [ tab1 ] ) .
this is true not only for the abundance ratios referred to iron , but it also holds for normalizations to hydrogen .
in particular , the measured sodium abundances span a range @xmath29[na / fe]@xmath30[na / h]@xmath31 , for nitrogen we find @xmath29[n / fe]@xmath30[n / h]@xmath32 , and for oxygen we obtain @xmath29[o / fe]@xmath30[o / h]@xmath33 .
the detected inhomogeneities also appear to be mutually correlated .
in fact , figure [ el_corr ] shows clear c - n and o - na anticorrelations , and n - na and na - al correlations , both if we consider the abundance ratios referred to fe , and if we normalize to h. in all cases , the statistical significance , as measured by the spearman rank coefficients @xmath34 , is very high ( values of @xmath34 larger than 0.74 corresponds to non - correlation probabilities lower than @xmath35 ) . in these diagrams , star 44 and star 65 ( see figure [ spectra ] ) reside at two opposite ends , the former being c and o - rich and n , na , al - poor , while star 65 showing a specular pattern .
the existence of such well - defined correlations , by itself , indicates the presence of multiple sub - populations along the agb of ngc 6752 . by definition , in fact , a sample composed exclusively of fg stars ( as suggested by c13 ) would display homogeneous abundances and produce no correlations .
indeed , the detected correlations are perfectly in agreement with those commonly ascribed to fg and sg sub - populations in gcs ( see , e.g. , @xcite ) .
figure [ nao ] shows the agb population ( solid blue circles ) of ngc 6752 in the `` standard '' [ na / fe]-[o / fe ] plane . for reference
, we also plot the results obtained for the rgb population of ngc 6752 ( empty red squares , from @xcite ) and several rgb samples in 19 gcs ( gray dots , from @xcite ) .
the agb population of ngc 6752 clearly outlines and follows the anti - correlation stream defined by the rgb samples , thus confirming the existence of sg agb stars in ngc 6752 . to better characterize the cluster sub - populations , in figure [ nao ] we also plot three ellipses corresponding to the values of [ na / fe ] and [ o / fe ] that @xcite , on the basis of their photometric study and the chemical abundances measured by @xcite , associate to the fg , sg and extreme - sg sub - samples in ngc 6752 ( `` populations a , b , and c '' in their nomenclature
; see figure 15 in @xcite ) .
notably , the abundances here determined for the agb population nicely match the fg and sg loci , thus demonstrating that , at odds with the claim by c13 , also sg stars do experience the agb phase in ngc 6752 .
in particular , based on table 1 and figure [ nao ] , we count 13 ( out of 20 ) sg stars , corresponding to @xmath36 of the total agb sample here investigated .
following @xcite , the fg stars in ngc 6752 have standard chemical mixture ( @xmath37 ) , sg stars have moderate enhancement of [ na / fe ] and he ( @xmath38 ) , and a mild depletion of [ o / fe ] , and extreme - sg stars have high [ na / fe ] and he ( @xmath39 ) , and low [ o / fe ] .
our results show that the agb sample is composed of the first two populations only , while the extreme - sg stars are not observed .
nicely , the sg/(fg+sg ) fraction estimated photometrically by @xcite is @xmath40 , in very good agreement with the value ( 65% ) found here .
the lack of the extreme - sg stars along the agb is exactly what the synthetic hb simulations by @xcite predict .
these simulations consider three stellar populations with initial he abundances equal to those quoted by @xcite , and they are able to well reproduce the observed value of the @xmath1 parameter and the hb morphology , magnitude and color distribution , without invoking exceptional mass loss during the hb phase .
they predict that the extreme - sg is the most he - rich population , which populates the bluer end of the hb and produces _ agb - manqu _ objects .
also the observed fraction of fg agb stars ( 35% of the total ) is consistent , within the statistical fluctuations due to the small size of the sample , with the predictions ( @xmath6 ) of @xcite .
therefore , the observed fraction of failed agb stars in ngc 6752 can be easily explained within the standard stellar evolution framework , with no need of invoking exceptional mass loss for hb stars hotter than the grundahl jump ( c13 ) , or extremely high ( and inconsistent with the photometric constraints from the main sequence ) initial he abundances for the sg population @xcite .
the results presented here provide firm evidence that the agb population of ngc 6752 includes both fg and sg stars .
while this is in line with similar results obtained from both spectroscopic @xcite and photometric @xcite observations in various gcs , it is in contrast with the findings of c13 .
although a detailed discussion of the origin of the discrepancy with the result by c13 is beyond the scope of this letter , a preliminary comparison between the two samples demonstrates that there is a systematic offset ( @xmath29[na / fe ] @xmath3 0.25 dex ) between the values of [ na / fe ] measured here and those measured by c13 .
this is the main reason why while we count at least 11 stars above [ na / fe ] = 0.18 dex ( the threshold adopted by c13 ) , no stars were found by c13 . on the other hand the clear anticorrelations shown in fig .
[ el_corr ] for the entire set of light elements , clearly indicated the existence of sg stars along the agb of ngc6752 , thus demonstrating that the chemical analysis of a single light element does not allow to draw reliable conclusions about the presence or lack of sg populations .
indeed , the adoption of the classical scheme based on the analysis of light element ( anti)correlations , appears to be the most appropriate spectroscopic way to detect and distinguish fg and sg stars . as a function of [ o / fe ] for the agb ( filled blue circles , this work ) and rgb stars ( open red squares *
* ) of ngc 6752 .
the results obtained for rgb stars in other gcs @xcite , rescaled to the solar values adopted in this work , are shown as gray dots for reference .
the regions corresponding to the three populations identified by ( * ? ? ?
* see their figure 15 ) are encircled . ]
lccclclccccc 22 & 4554 & 1.18 & 1.85 & 1.77 @xmath28 0.01 & 1.61 @xmath28 0.01 & 0.29 @xmath28 0.07 & + 1.20 @xmath28 0.12 & + 0.41 @xmath28 0.01 & + 0.29 @xmath28 0.04 & + 0.37 @xmath28 0.01 & + 0.52 @xmath28 0.02 + 25 & 4351 & 0.92 & 1.80 & 1.79 @xmath28 0.01 & 1.60 @xmath28 0.01 & 0.25 @xmath28 0.06 & + 0.68 @xmath28 0.12 & + 0.54 @xmath28 0.06 & 0.00 @xmath28 0.03 & + 0.44 @xmath28 0.01 & @xmath41 0.11 + 31 & 4413 & 1.06 & 1.85 & 1.76 @xmath28 0.01 & 1.56 @xmath28 0.01 & 0.17 @xmath28 0.07 & + 0.76 @xmath28 0.12 & + 0.53 @xmath28 0.04 & 0.03 @xmath28 0.01 & + 0.38 @xmath28 0.03 & @xmath41 0.22 + 44 & 4585 & 1.31 & 1.65 & 1.79 @xmath28 0.01 & 1.58 @xmath28 0.01 & 0.17 @xmath28 0.07 & + 0.73 @xmath28 0.12 & + 0.49 @xmath28 0.05 & 0.01 @xmath28 0.03 & + 0.40 @xmath28 0.01 & @xmath41 0.21 + 52 & 4752 & 1.55 & 1.65 & 1.81 @xmath28 0.01 & 1.61 @xmath28 0.02 & 0.19 @xmath28 0.08 & + 0.99 @xmath28 0.13 & + 0.44 @xmath28 0.04 & + 0.14 @xmath28 0.02 & + 0.41 @xmath28 0.01 & + 0.20 @xmath28 0.03 + 53 & 4712 & 1.44 & 1.70 & 1.78 @xmath28 0.01 & 1.59 @xmath28 0.01 & 0.34 @xmath28 0.08 & + 1.29 @xmath28 0.13 & + 0.23 @xmath28 0.02 & + 0.22 @xmath28 0.02 & + 0.38 @xmath28 0.01 & + 0.40 @xmath28 0.03 + 59 & 4724 & 1.50 & 1.65 & 1.81 @xmath28 0.01 & 1.58 @xmath28 0.01 & 0.34 @xmath28 0.08 & + 1.29 @xmath28 0.13 & + 0.21 @xmath28 0.01 & + 0.25 @xmath28 0.01 & + 0.38 @xmath28 0.01 & + 0.34 @xmath28 0.05 + 60 & 4690 & 1.44 & 1.65 & 1.68 @xmath28 0.01 & 1.55 @xmath28 0.01 & 0.09 @xmath28 0.07 & + 0.72 @xmath28 0.13 & + 0.56 @xmath28 0.05 & 0.10 @xmath28 0.03 & + 0.36 @xmath28 0.02 & @xmath41 0.21 + 61 & 4722 & 1.50 & 1.70 & 1.77 @xmath28 0.01 & 1.58 @xmath28 0.01 & 0.19 @xmath28 0.08 & + 0.99 @xmath28 0.13 & + 0.44 @xmath28 0.05 & + 0.17 @xmath28 0.02 & + 0.42 @xmath28 0.01 & + 0.36 @xmath28 0.03 + 65 & 4622 & 1.31 & 1.80 & 1.81 @xmath28 0.01 & 1.59 @xmath28 0.01 & 0.41 @xmath28 0.07 & + 1.32 @xmath28 0.13 & + 0.21 @xmath28 0.01 & + 0.35 @xmath28 0.03 & + 0.40 @xmath28 0.01 & + 0.52 @xmath28 0.03 + 75 & 4724 & 1.55 & 1.65 & 1.84 @xmath28 0.01 & 1.59 @xmath28 0.01 & 0.10 @xmath28 0.08 & + 0.68 @xmath28 0.13 & + 0.48 @xmath28 0.03 & 0.05 @xmath28 0.01 & + 0.38 @xmath28 0.02 & @xmath41 0.21 + 76 & 4862 & 1.64 & 1.70 & 1.84 @xmath28 0.01 & 1.61 @xmath28 0.01 & 0.42 @xmath28 0.08 & + 1.30 @xmath28 0.13 & + 0.25 @xmath28 0.05 & + 0.30 @xmath28 0.04 & + 0.42 @xmath28 0.02 & + 0.54 @xmath28 0.05 + 78 & 4877 & 1.65 & 1.75 & 1.82 @xmath28 0.01 & 1.58 @xmath28 0.01 & 0.46 @xmath28 0.08 & + 1.40 @xmath28 0.13 & + 0.16 @xmath28 0.03 & + 0.32 @xmath28 0.03 & + 0.38 @xmath28 0.01 & + 0.53 @xmath28 0.04 + 80 & 4804 & 1.63 & 1.70 & 1.80 @xmath28 0.01 & 1.58 @xmath28 0.01 & 0.18 @xmath28 0.08 & + 0.91 @xmath28 0.13 & + 0.39 @xmath28 0.02 & + 0.05 @xmath28 0.02 & + 0.43 @xmath28 0.01 & @xmath41 0.28 + 83 & 4817 & 1.63 & 1.60 & 1.85 @xmath28 0.01 & 1.59 @xmath28 0.01 & 0.16 @xmath28 0.08 & + 1.08 @xmath28 0.13 & + 0.38 @xmath28 0.02 & + 0.19 @xmath28 0.02 & + 0.36 @xmath28 0.01 & + 0.33 @xmath28 0.05 + 89 & 4798 & 1.63 & 1.65 & 1.86 @xmath28 0.01 & 1.58 @xmath28 0.01 & 0.34 @xmath28 0.08 & + 1.15 @xmath28 0.13 & + 0.28 @xmath28 0.02 & + 0.30 @xmath28 0.02 & + 0.38 @xmath28 0.01 & + 94 & 4864 & 1.71 & 1.65 & 1.88 @xmath28 0.01 & 1.56 @xmath28 0.02 & 0.37 @xmath28 0.08 & + 1.26 @xmath28 0.13 & + 0.17 @xmath28 0.03 & + 0.24 @xmath28 0.04 & + 0.39 @xmath28 0.01 & + 0.20 @xmath28 0.05 + 97 & 4884 & 1.75 & 1.70 & 1.89 @xmath28 0.01 & 1.58 @xmath28 0.01 & 0.33 @xmath28 0.08 & + 1.14 @xmath28 0.13 & + 0.22 @xmath28 0.06 & + 0.25 @xmath28 0.01 & + 0.40 @xmath28 0.01 & + 0.15 @xmath28 0.06 + 104 & 4753 & 1.66 & 1.60 & 1.87 @xmath28 0.01 & 1.54 @xmath28 0.01 & 0.16 @xmath28 0.08 & + 0.73 @xmath28 0.13 & + 0.40 @xmath28 0.07 & + 0.01 @xmath28 0.03 & + 0.42 @xmath28 0.01 & @xmath41 0.03 + 1620 & 4902 & 1.71 & 1.70 & 1.80 @xmath28 0.01 & 1.56 @xmath28 0.01 & 0.26 @xmath28 0.08 & + 1.19 @xmath28 0.14 & & + 0.23 @xmath28 0.01 & + 0.38 @xmath28 0.02 & + 0.18 @xmath28 0.04 + [ tab1 ] el acknowledges the financial support from prin - inaf 2014 .
cl gratefully acknowledges financial support from the european research council ( erc - cog-646928 , multi - pop ) .
dm has been supported by the firb 2013 ( miur grant rbfr13j716 ) .
sc thanks for financial support from prin - inaf 2014 ( pi : s. cassisi ) .
we warmly thank the anonymous referee for his / her useful comments , which improved the paper . | we derived chemical abundances for c , n , o , na , mg and al in 20 asymptotic giant branch ( agb ) stars in the globular cluster ngc 6752 . all these elements ( but mg )
show intrinsic star - to - star variations and statistically significant correlations or anticorrelations analogous to those commonly observed in red giant stars of globular clusters hosting multiple populations .
this demonstrates that , at odds with previous findings , both first and second generation stars populate the agb of ngc 6752 .
the comparison with the na abundances of red giant branch stars in the same cluster reveals that second generation stars ( with mild na and he enrichment ) do reach the agb phase .
the only objects that are not observed along the agb of ngc 6752 are stars with extreme na enhancement .
this is also consistent with standard stellar evolution models , showing that highly na and he enriched stars populate the bluest portion of the horizontal branch and , because of their low stellar masses , evolve directly to the white dwarf cooling sequence , skipping the agb phase . |
with the launch of nasa s stereo mission in october 2006 , a new dimension of solar coronal observations has been opened . for the first time , objects above the solar surface can be perceived in three dimensions by analysing the stereo image pairs observed with the secchi instruments onboard the stereo spacecraft and without making a - priori assumptions about their shape .
the two stereo spacecraft orbit the sun at approximately 1 au near the ecliptic plane with a slowly increasing angle of about 45 degrees / year between stereo a and stereo b. each spacecraft is equipped with , among other instruments , an euv telescope ( secchi / euvi ) . for the objectives of the mission and
more details about the euvi telescopes see @xcite and @xcite .
the major building blocks of the solar corona are loops of magnetic flux which are outlined by emissions at , e.g. , euv wavelengths . in principle , the magnetic field in the lower corona can be derived from surface magnetograms by way of extrapolations ( e.g. * ? ? ?
. however , missing boundary values and measurement errors may introduce considerable uncertainties in the extrapolation results so that there is an obvious need for an alternative three - dimensional determination of the coronal magnetic field geometry . among other goals of the mission ,
this requirement has been one of the drivers for stereo .
attempts for a three - dimensional reconstruction of the coronal magnetic field from euv observations have started long before stereo data was available and date back more than a decade @xcite . here ,
we for the first time use two simultaneously observed euvi images observed by the two stereo probes and rigourously reconstruct loop shapes without any further assumption about their temporal or spatial behaviour from which earlier reconstructions employing consecutive images from a single spacecraft suffered @xcite .
we compare the reconstruction results with field lines derived from linear force - free magnetic field models with variable @xmath0 , the ratio of field - aligned current density to field strength @xcite .
.stereo spacecraft coordinates at the time of the observations .
spacecraft longitude and latitude are given in the heliocentric earth ecliptic ( hee ) coordinate system . [ cols="<,>,>",options="header " , ] the loop reconstruction is also prone to errors , however .
these may occur whenever a projected loop section in the images are directed tangentially to an epipolar line @xcite . for the viewing geometry of our observations , epipolar lines are nearly horizontal in the images and
the critical part for closed , e - w orientated loops therefore lies more or less near their apex .
also the open loop structures 16 - 19 in image b and 17 - 20 in image a ( see figure [ fig : loopab ] ) suffer from this problem as they are orientated almost entirely horizontally in the images .
we have therefore not attempted to reconstruct them even though a correspondence could well be identified . in figure [ fig:3dloop5a3b_e ]
we display the reconstruction of loop ( 5,3 ) ( yellow curve ) which shows by far the largest deviation to its best fit linear force - free field line ( red curve ) . for most other loops ,
this discrepancy is much less although the agreement is rarely perfect .
for some points along the loop ( 5,3 ) , we also show error bars which represent the geometrical reconstruction error when the uncertainty for the loop projection in the images is assumed to be 1.5 pixels . in this case
, the height of the loop top turns out to be @xmath1 1.5 times above that of the corresponding field line .
this field line ( the first entry in table [ tab : loopparam ] ) again shows a relatively small value @xmath2 .
since this @xmath0 value gave the best fit of linear force - free field lines to the loop projection in the images , we conclude that the linear force - free assumption is often not adequate ( cf .
we demonstrated that euv data from the new stereo spacecraft allows for the first time to make a reliable stereoscopic reconstruction of the spatial distribution of hot , magnetically confined coronal plasma and , by inference , provide a full three dimensional view of the arrangement of coronal field lines .
we found that linear force - free field models are helpful to establish correspondences between the loops observed in the stereo image pairs .
the field lines from these linear force - free models need not be physical but only serve as a first order approximation to the final loops .
realistic magnetic field models of the corona will have to be judged by their capability to yield field lines in agreement with the stereoscopically reconstructed loops .
our scheme to determine correspondences will become even more valuable when the stereo base angle grows and loop structures become more difficult to be identified in the image pairs .
the reconstructions will also allow more precise analyses of emissions from loops .
the observed brightness of euv loops is , e.g. , strongly modified by the inverse cosine of the angle between the line of sight and the loop s local tangent .
this may , besides other effects , contribute to the enhanced euv brightness of the lower loop segments commonly observed on the solar disk : these loop segments close to the loop s foot points are more aligned with the radial direction and they make a small angle with the view direction .
this may cause them to appear brighter than the loop top which is viewed at more or less right angles .
other applications have been proposed @xcite .
e.g. , the amount of twist of a reconstructed loop indicates how close the flux tube is to a kink instability .
@xcite found a threshold of about @xmath3 in numerical simulations for the twist @xmath4 . here
@xmath5 is the length of the flux tube , @xmath6 the toroidal field along its axis and @xmath7 the poloidal field at a radius @xmath8 from the flux tube centre . in some cases it may be possible to resolve the number of turns @xmath9 which a field line makes about the flux tube centre from stereoscopic reconstruction and thus to determine the twist from @xmath10
likewise , the twist is also related to @xmath0 and @xmath5 by @xmath11 . for the active region
observed here , table [ tab : loopparam ] gives values of @xmath12 well below the kink instability threshold .
another perspective for stereoscopic loop reconstruction is the analysis of loop oscillations from a series of image pairs .
the reconstructed loops will allow us to determine the transverse polarisation of these oscillations @xcite . since the coronal magnetic field has a complicated geometry without symmetries , the frequency of these oscillations will significantly depend on this polarisation .
note that these phenomena are invisible in the magnetic surface data and therefore can not be retrieved from field extrapolations , which in addition require a stationary magnetic field .
the authors thank the mdi / soho and the secchi / stereo consortia for the supply of their data .
stereo is a project of nasa , soho a joint esa / nasa project .
the secchi data used here were produced by an international consortium of the naval research laboratory ( usa ) , lockheed martin solar and astrophysics lab ( usa ) , nasa goddard space flight center ( usa ) , rutherford appleton laboratory ( uk ) , university of birmingham ( uk ) , max - planck - institut for solar system research ( germany ) , centre spatiale de lige ( belgium ) , institut doptique thorique et applique ( france ) , institut dastrophysique spatiale ( france ) . | we present the first reconstruction of the three - dimensional shape of magnetic loops in an active region from two different vantage points based on simultaneously recorded images .
the images were taken by the two euvi telescopes of the secchi instrument onboard the recently launched stereo spacecraft when the heliocentric separation of the two space probes was 12 degrees .
we demostrate that these data allow to obtain a reliable three - dimensional reconstruction of sufficiently bright loops .
the result is compared with field lines derived from a coronal magnetic field model extrapolated from a photospheric magnetogram recorded nearly simultaneously by soho / mdi .
we attribute discrepancies between reconstructed loops and extrapolated field lines to the inadequacy of the linear force - free field model used for the extrapolation . |
since the discovery of the so called magnetic superconductors rusr@xmath2rcu@xmath16o@xmath17 ( ru-1212 ) and rusr@xmath2(r , ce)@xmath2cu@xmath2o@xmath5 ( ru-1222 ) , with r = gd , eu,@xcite considerable effort has been devoted to the understanding of the interplay between the ferromagnetic ( fm ) component , emerging from the long - range order of the ru moments , and the onset of the superconducting ( sc ) state.@xcite among several important topics , the possibility of @xmath18-phase formation across the ruo@xmath2 layers,@xcite the itinerant or localized character of the magnetism of the ru moments , mccrone01,tallon02,mccrone02 the magnetic phase separation scenario of nanoscale fm clusters with superconductivity nucleating only in the surrounding antiferromagnetic matrix,@xcite and the possibility of triplet paring,@xcite have been considered to explain how this puzzling coexistence may occur . on the other hand ,
some important superconducting parameters have been less thoroughly investigated .
reports include the determination of the coherence length @xmath19 and the higher critical field @xmath20,@xcite and the intragrain london penetration length @xmath21,@xcite and rough estimations of the lower critical field of the grains , @xmath0.@xcite in relation to the determination of @xmath0 , a diamagnetic signal has been observed in a few cases at the low field range of the @xmath22 magnetization loops in ru-1222 , with a negative minimum at about 25 oeawana and 50 oe.@xcite there are also fewer studies on relevant intrinsic superconducting properties , such as the intragrain critical current density , @xmath1,felner ijmp mainly because the strong fm contribution to the magnetization from the ru sublattice makes impracticable the use of the magnetic hysteresis loops to determine both @xmath0 and @xmath1 . in the present study
we overcome this intrinsic difficulty and present a reliable method to determine these magnitudes and their temperature dependencies in ru-1222(gd ) , through a systematic study of the hysteresis in the isotemperature magnetoresistance @xmath23 curves .
since no single crystals are available for this compound , polycrystalline materials were used in the present investigation .
two different ru-1222(gd ) samples were studied , obtained under different partial oxygen pressures . at variance with the behavior of the high - t@xmath13 cuprates , a monotonic increase without saturation in both @xmath0 and @xmath1 on cooling
was observed , reaching values as high as @xmath0 @xmath24 1000 oe and @xmath1 @xmath24 10@xmath11 a/@xmath25 , at @xmath26 = 7.5 k. a comparison with yba@xmath2cu@xmath27o@xmath28 ( ybco ) and with the results reported for modeling the magnetic properties of ru-1212 on the basis of the theory of the sc / fm multilayers is presented .
polycrystalline rusr@xmath2gd@xmath3ce@xmath4cu@xmath2o@xmath5 was prepared by conventional solid - state reaction using an oxygen flow in the final heat treatment .
the room temperature x - ray diffraction pattern corresponds to ru-1222(gd ) , with no spurious lines .
scanning electron microscopy revealed a dense grain packing , with an average grain size d @xmath29 0.5 - 1 @xmath30 m .
more details on sample preparation and microstructure can be found elsewhere.@xcite after characterization , the as - prepared ( _ asp _ ) sample was annealed for 24 hours at 600@xmath31c under 50 atm of pure oxygen [ high oxygen pressure ( _ hop _ ) sample ] .
magnetotransport and ac magnetic susceptibility measurements@xcite reproduce the behavior of good quality samples.@xcite bars of @xmath29 10 mm in length and 0.6 mm@xmath32 cross sectional area were cut from the sintered pellet .
the resistance was measured with a standard four contacts probe using a quantum design ppms system , at @xmath26 = 7.5 , 8 , 9 , 10 , 11.25 , 12.5 , and 15 k for the _ asp _ sample ; the _ hop _ sample was also measured at 8.5 k. a large number of @xmath23 curves were collected for each temperature to accurately follow the different characteristic regimes found in the magnetoresistance response ( as described below ) , and to warrant a reliable quantitative determination of the fields at which the transition from one regime to another occur . the most relevant experimental parameter varied experimentally is the maximum applied field within the @xmath33 curves , @xmath34 , which ranged from a few tens of oe up to 60 koe .
in order to characterize the transport behavior of the studied samples we initially measured the temperature dependence of the resistivity .
it is clear from the data shown in fig .
1 that oxygenation under pressure strongly reduces the absolute resistivity values , enlarges the linear behavior of the normal - state region , and reduces the width of the sc transition .
in addition , the superconducting transition temperature , @xmath35 , as determined from the peaks in the derivative of the resistivity for the as prepared ( _ asp _ ) sample , increases by approximately 3 k in the _ hop _ sample .
these features agree with previous reports in good quality materials.@xcite the present + study concentrates in results of resistance as a function of field , @xmath23 , measured at fixed temperature with different maximum fields , @xmath34 , in each sweep .
four different behaviors were identified in the @xmath23 curves as @xmath34 is increased : a ) a zero resistance region , typically for @xmath36 @xmath37 100 oe ; b ) an interval of reversible dissipation up to an irreversible field @xmath38 , from about 150 oe to 1500 oe for the _ asp _ sample and up to 3000 oe for the _ hop _ sample ; c ) an @xmath34-dependent hysteretic behavior , and d ) an hysteretic response independent of @xmath34 .
figure 2 shows selected magnetoresistance hysteresis loops for the _ hop _ sample with @xmath34 = 5000 oe , measured at @xmath26 = 7.5 and 10 k , normalized to the @xmath39 values .
for each temperature , loops similar in shape but broader were obtained for the _ asp _ sample . in the hysteretic region
the resistance curves measured with decreasing external field , @xmath40 , are always below the corresponding virgin curves , @xmath41 , measured with increasing field .
when the difference @xmath42 is plotted , a peak is always obtained at a certain intermediary value of the applied field , @xmath43 . with the rise in @xmath34 , the peak gradually increases its amplitude , @xmath44 , and shifts to higher @xmath43 values until a certain field @xmath34 @xmath45 @xmath46 is reached , that we denote as a saturation field . for @xmath47 @xmath48
@xmath46 the peak in @xmath49remains unchanged . to illustrate this behavior , fig .
3 plots @xmath50 as a function of @xmath43 for different values of @xmath34 , for the _ hop _ sample at @xmath26 = 12.5 k. we focus our attention on a careful determination of @xmath38 and @xmath46 .
since @xmath40 diverge smoothly from @xmath51 , it is necessary to implement a method for this evaluation .
we proceeded as follows : a ) the error @xmath52 in the determination of @xmath23 was taken as the average of the standard deviations of each measured experimental point in a given loop ; we found that successive @xmath23 curves measured at each temperature for different @xmath34 ( typically 15 - 20 curves ) always fell inside this @xmath52 criterion ; b ) for each temperature , @xmath53 was plotted as a function of @xmath34 , as shown in fig .
4 for the _ hop _ sample + at @xmath26 = 12.5 k , and the data fitted using a polynomial function ; the error in the determination of @xmath53 is @xmath54 ; c ) @xmath40 is considered reversible if @xmath53(@xmath34 ) does not exceed the @xmath52 threshold , as indicated by a dotted line above the zero resistance level in the inset of fig .
4 ( low @xmath34 region ) ; the irreversibility field , @xmath38 , is determined from the intersection of the fitted polynomial function with the @xmath52 level ( fig .
4 , inset ) ; d ) the plateau observed for @xmath44 for the high @xmath34 range was fitted by a straight line parallel to the @xmath34 axis ( continuous line in fig .
4 , main panel ) ; e ) the saturation field , @xmath46 , was determined as the field for which the polynomial function diverges in @xmath54 from the fitted straight line , ( intersection with the dashed line , parallel and below the plateau level in fig .
4 , main panel ) . as we discuss below , @xmath55 is a good estimation of the lower critical field of the grains , @xmath0 .
the temperature dependencies of @xmath0 and @xmath46 for the _ asp _ and _ hop _ samples are plotted , in fig . 5 and in the inset of fig . 6 , respectively .
the behavior of @xmath0 for ybco ( with the field parallel to the sample s _ c _ axis ) is also shown for comparison in the inset of fig .
the @xmath14 and @xmath56 curves for both samples monotonically decrease with the increase in temperature and smoothly merge at about 15 k. the values of @xmath1 at different temperatures , determined by using the bean critical state model,@xcite as described below , are shown in fig . 6 .
a systematic increase of @xmath1 on cooling is found , with slightly higher values for the _ hop _ sample .
a key point to understand the features of the @xmath23 curves is to recognize that the magnetization in the grains contributes to the effective local field at the intergranular junctions .
hysteresis in the magnetoresistance was recently observed in ceramic ru-1222 samples,@xcite and a preliminary analysis evidenced that the granular character of the material is relevant . unlike ru-1212 , in ru-1222 the coercive field and the remanent magnetization vanish at the sc transition , @xmath57 @xmath7 @xmath58 @xmath59 , for both gd@xcite and eu@xcite based compounds , even though the magnetic transition temperatures are much higher . since the gd ions are decoupled from the ru moments,@xcite no changes in the irreversibility behavior are observed when gd is replaced in ru-1222 by eu .
the essential feature governing the irreversibility response in ru-1222 is the thick ( r , ce)@xmath2o@xmath2-fluorite - type blocks separating the ruo@xmath2 planes .
the large separation distance strongly weakens the superexchange interactions between the magnetic layers , which are weakly coupled through dipole - dipole interaction.@xcite therefore , the hysteretic behavior of the @xmath23 curves has a sc origin in ru-1222 . for a low applied magnetic field , @xmath43 ,
the net local field at the intergranular junctions is not strong enough to destroy all the sc percolative paths across the weak - linked network , leading to the observed zero resistance region in @xmath23 . for @xmath43 @xmath48 @xmath36 ,
the connectivity of the network is affected in such extent that the current across the sample can not flow without dissipation , and a sizable resistance appears .
the existence of a reversible + dissipation section in the @xmath23 curves evidence that the grains have not yet been penetrated at this field range , and there is no intergranular pinning .
irreversible @xmath23 loops appear when the local field at the junctions penetrates the grains . for polycrystalline high - t@xmath13 cuprates
this local field is higher than @xmath60 due to flux compression associated to diamagnetic shielding , leading to an underestimation of @xmath0 . for ru-1222
, the contribution of the fm magnetization to the local field at the intergranular network , @xmath61 , is oriented in opposition to @xmath43 , diminishing the flux compression effect .
it was recently determined that @xmath62 is approximately 15 oe at @xmath63= 0 k for ru-1222.@xcite since @xmath64 varies from 100 to 3000 oe , any possible difference between the local penetration field of the grains and @xmath43 due to the contribution of @xmath62 is approximately 10% or less .
therefore , by determining @xmath55 one obtain a good estimation of @xmath0 .
let us consider the effects of the flux trapped in the grains on the @xmath23 loops .
for a given @xmath34 @xmath48 @xmath0 , a fraction of the grain volume is penetrated .
when @xmath43 is decreased different field profiles are created inside the grains due to pinning , gradually evolving as @xmath43 diminishes.@xcite each field profile represents a spatial circulation of currents , i.e. , a certain sc magnetization , contributing with a local field @xmath65 at the intergranular junctions .
the lowest value of @xmath60 at which full reversal is obtained is @xmath66.@xcite until the fully reversed profile is not attained , the exact value of @xmath65 for a given @xmath43 in the returning curve changes as @xmath47 increases , because a different field profile is obtained for that external field .
the extent of the compensation between @xmath43 and @xmath65 at the junctions as the returning curves are collected is clearly evidenced by the changes in the maxima of @xmath67 the amplitude of the peak in @xmath68 , as shown in fig .
the peaks correspond to a maximum average compensation of the local fields in the polycrystal , a relative maximum in the connectivity of the weak - linked network , as @xmath43 is decreased .
the increase in amplitude of the peaks for higher @xmath34 is associated to a narrowing of the distribution of local fields in the sample as the sc magnetization of the grains rises .
the position of the peaks is shifted to higher fields with the rise in @xmath34 up to @xmath46 , because the field profile inside the grains becomes reversed in a larger extent , gradually leading to a higher @xmath65 in opposition to @xmath43 ; the maximum compensation occurs at higher @xmath43 .
once the external field is high enough to reach the fully reversed profile there will be no further increase in @xmath65 with the rise in @xmath34 , and the @xmath44 maxima remain unchanged . therefore @xmath69 . since the fm contribution to the local field is not hysteretic
, it has no effect on @xmath46 .
let us compare the absolute values and temperature dependence of @xmath0 in ru-1222 with ybco .
due to the wide sc transition width in the former , the higher temperature for which @xmath0 was determined is @xmath26 = 15 k ( @xmath70 @xmath7 0.3 ) . for this @xmath6 value , the lower critical field in ybco
has already reached its saturation value , @xmath0(ybco ) @xmath29 850 oe@xcite ( with @xmath71 parallel to the _ c_-axis ) , while it is only 150 oe for both ru-1222 samples . however , due to its monotonic increase on cooling , @xmath0 reaches a value twice larger than ybco for the ru-1222 _ asp _ sample , and even higher for the _ hop _ one ( 3000 oe ) at 7.5 k ( @xmath6 @xmath7 0.2 ) . at this point
it is important to consider whether there is a possible contribution from surface barriers in the determination of @xmath0 .
it should be noted that ru-1222 exhibits a monotonic increase of @xmath72 on cooling without saturation down to @xmath26 = 5 k,@xcite at variance with the high - t@xmath13 cuprates .
this implies a reduced efficiency of any possible barrier shielding with the decrease in temperature , since a decrease in @xmath73 favours the flux penetration through weak spots at the grain surface .
another important point is the strong decrease in @xmath74 in samples with higher partial oxygen pressure during the final heat treatment.@xcite therefore , flux penetration through the weak spots must be easier for the _ hop _ sample in comparison to the _ asp _ one , and the higher @xmath0 values obtained for the former can not be explained as due to surface barriers .
in addition , ru-1212 and ru-1222 present a very distinctive structural feature : the existence of sharp antiphase boundaries with local distortions separating structural domains of coherently rotated ruo@xmath75 octahedra .
a high density of quite sharp boundaries , of several tens of unit cells in length , have been observed in high resolution transmission electron microscopy images.mclaughlin in ceramic superconductors , long - range distortions , as twinning planes@xcite and columnar defects,@xcite are known to act as channels for flux penetration , dramatically suppressing surface barriers shielding .
antiphase boundaries should work very similarly .
these considerations strongly indicate that surface barriers effects are not the source of the observed @xmath0 dependence in polycrystalline ru-1222 .
the large @xmath0 values obtained at low temperatures are somewhat unexpected since , as mentioned above , previous reports based on @xmath22 curves inferred @xmath0 values as low as 25 - 50 oe in ru-1222.awana,felner01,felner02 nevertheless , clear evidence that these reports greatly underestimate the magnitude of @xmath0 comes from measurements in samples with partial substitution of ru by non - magnetic elements .
the negative diamagnetic minimum in @xmath76is shifted to @xmath24 300 oe for 30% substitution of ru by sn,@xcite and 40% substitution by nb,@xcite even though @xmath57 is decreased in both cases .
moreover , the zero net magnetization point in the virgin @xmath22 curves ( for which a significant diamagnetic signal is still present ) is observed for @xmath43 @xmath10 800 oe in both systems .
these results indicate that the masking effect of the fm background is strong , and that the actual magnitude of @xmath77 is in fact considerably higher .
further evidence of @xmath78values of the order of several hundreds oe in ruthenate - cuprates comes from the magnitude of the internal ru magnetization and the local fields at different points of the cell .
it has been determined that the in - plane magnetization at the ruo@xmath2 layers is @xmath79 kg , yielding a macroscopic ( cell volume average ) dipolar internal field @xmath80
@xmath81@xmath37@xmath82@xmath48@xmath83 oe.@xcite the latter value is confirmed by gd@xmath84-electron spin resonance@xcite and muon spin rotationbernhard01 measurements , giving internal fields of @xmath24 600 - 700 oe at the gd site and near to the apical oxygen of the cuo@xmath85 pyramid , respectively .
flux expulsion from the grains can occur only if , at a certain sufficiently low temperature , @xmath14 exceeds the magnetization value of @xmath86 .
this internal meissner effect has been previously observed in ru-1222 at @xmath26 = 16 k , which is about 30 k below @xmath57 , in samples made using an oxygen flow.@xcite the temperature dependence of @xmath0 can be understood when the magnetism of the ru sublattice is properly considered . a model for the magnetic properties of the ru-1212 system that incorporates the theory of the superconducting / ferromagnetic multilayers predicts an increase in @xmath0 with the decrease in temperature.@xcite this response was obtained assuming @xmath87 @xmath88 @xmath57 , which is the case for both ru-1222 and ru-1212 compounds , and ascribed to the decrease of the effective josephson coupling between the sc planes due to the exchange field at the fm layers .
in addition , the mentioned @xmath89 dependence for ru-1222 supports the increase of @xmath0 as the temperature diminishes , within the framework of a bcs - type behavior . under this assumption ,
the strong increase in @xmath89 with the rise of the oxygen pressure supports a significant increase of @xmath0 for the _ hop _ sample .
the order of magnitude of @xmath1 was determined using the expression @xmath90 , corresponding to the bean critical state model for an infinite sc slab,@xcite with @xmath91 @xmath92 @xmath93 and @xmath94 .
the expression for an infinite slab gives the highest possible values for @xmath1 according to the geometry of the sc grain . for a cylindrical geometry
there will be a decrease by one order of magnitude .
therefore , the calculated values @xmath1
@xmath24 10@xmath11 a/@xmath12 should be considered as an upper limit .
this large order of magnitude can be explained in terms of the magnetic phase separation scenario.@xcite ferromagnetic nanoclusters in the normal state are embedded in an antiferromagnetic matrix in which superconductivity nucleates . since the sc coherence length @xmath19 for ru-1222 is about the same size as the magnetic clusters,@xcite they can effectively act as pinning centers .
the monotonic increase of @xmath1 with the decrease in temperature can also be qualitatively explained in terms of the segregation of fm clusters .
it has been demonstrated for ru-1222(eu ) that the saturation magnetization of this magnetic species monotonically increases on cooling down to 10 k,@xcite well inside the sc region .
this result _ _ _ _ is due to an increase of the density of fm clusters and/or to variation of their sizes with the decrease in temperature .
as the number of magnetic clusters increases at lower temperatures , the average matching between the abrikosov lattice and the spatial arrangement of pinning - cluster centers is gradually improved .
finally , it is worth noticing that while oxygenation under pressure increases the values of @xmath0 by nearly a factor of two at low temperatures , the increase in @xmath1 is moderate .
these results suggest that the oxygen treatment has a larger effect in the region near the grain boundaries .
flux penetration begins at lower fields for the _ asp _ sample because a larger deviation from the oxygen stoichiometry at the grain borders locally depresses superconductivity .
since @xmath1 is influenced by the properties of the grain as a whole it is less affected by the oxygenation process .
the significant decrease of the absolute resistivity values and the reduction in the width of the sc transition for the _ hop _ sample in comparison to the _ asp _ one , while the intragrain sc transition temperature increases only by @xmath24 3 k , confirm that the intergranular connectivity is greatly improved with oxygenation under pressure , while the core of the grains change in a much less extent .
in summary , fundamental characteristic properties of ru-1222 , the lower critical field of the grains and the intragrain critical current density , were determined after a detailed processing of magnetoresistance loops , @xmath33 .
the @xmath0 values at low temperatures in well oxygenated samples are about fifty times larger than the naive estimation obtained from magnetization measurements at low magnetic fields ; the magnitude of @xmath1 is similar to the high - t@xmath13 cuprates .
the possible segregation of normal - state ferromagnetic nanoclusters provides a natural scenario to explain the monotonic increase of both @xmath14 and @xmath15 on cooling .
we thank s. smansunaga and r. jardim for assistance with sample oxygenation under high pressures , and e. altshuler for useful comments .
this work was partially financed by cnpq and faperj .
s.g . was supported by capes . for a recent review see t. nachtrab , c. bernhard , c.t .
lin , d. koelle , r. kleiner , to be published in journal comptes rendus de lacademie des sciences ( comptes rendus physique ) , special issue on `` magnetism and superconductivity coexistence '' , available at cond - mat/0508044 ( unpublished ) . | the lower critical field of the grains , @xmath0 , and the intragrain critical current density , @xmath1 , were determined for the superconducting ruthenate - cuprate rusr@xmath2gd@xmath3ce@xmath4cu@xmath2o@xmath5 [ ru-1222(gd ) ] through a systematic study of the hysteresis in magnetoresistance loops .
a reliable method , based on the effects of the magnetization of the grains on the net local field at the intergranular junctions is provided , circumventing the problem of the strong masking of the superconducting diamagnetic signal by the ferromagnetic background .
the temperature dependency of @xmath0 and @xmath1 both exhibit a smooth increase on cooling without saturation down to @xmath6 @xmath7 0.2 .
the obtained @xmath0 values vary between 150 and 1500 oe in the 0.2 @xmath8 @xmath9 @xmath8 0.4 interval , for samples annealed in an oxygen flow ; oxygenation under high pressure ( 50 atm ) leads to a further increase .
these values are much larger than the previously reported rough assessments ( 25 - 50 oe ) , using conventional magnetization measurements .
high @xmath1 values of @xmath10 10@xmath11 a/@xmath12 , comparable to the high - t@xmath13 cuprates , were obtained .
the @xmath14 and @xmath15 dependencies are explained in the context of a magnetic phase separation scenario . |
meson light cone distribution functions ( lcdfs ) play important roles in high energy hadronic exclusive processes @xcite .
the same lcdfs contribute in many processes relevant to measuring fundamental parameters of the standard model @xcite , such as @xmath8 which give the cabibbo - kobayashi - maskawa ( ckm ) quark - mixing matrix element @xmath9 , @xmath10 used for tagging , and @xmath11 which are important for measuring cp violation . with the increasing accuracy in data from the @xmath12 factories ,
the flavor dependence in lcdfs becomes important to understand the flavor symmetry breaking in processes like @xmath13 and @xmath14 , where @xmath15 and @xmath16 are pseudoscalar and vector mesons , respectively . in ref .
@xcite , chiral perturbation theory ( chpt ) was first applied to study the leading su(3 ) symmetry breaking effects in twist-2 lcdfs @xmath17 .
it was shown that terms non - analytic in the quark masses do not affect the shape , and only appear in the normalization constants . furthermore , with the symmetry violating corrections useful model - independent relations between @xmath18 , @xmath19 , @xmath20 , and @xmath21 were derived .
recently chpt has also been applied to the computation of hadronic twist-2 matrix elements @xcite .
many applications have been worked out , e.g. , chiral extrapolations of lattice data dmnrt , qtwist2,beane:2002vq , detmold:2005pt , generalized parton distributions @xcite , large @xmath22 relations among nucleon and @xmath23-resonance distributions@xcite , soft pion productions in deeply virtual compton scattering @xcite , pion - photon transition distributions @xcite and exclusive semileptonic b decays @xcite .
the method is also generalized to the multi - nucleon case @xcite . in this paper
, we apply chpt to higher twist matrix elements .
we study the leading su(3 ) symmetry breaking in the complete set of quark twist-3 lcdfs , including the two - parton distributions @xmath0 , @xmath24 , and the three - parton distribution @xmath25 although those twist-3 contributions are parametrically suppressed by inverse powers of large scales , they are numerically important in @xmath13 , @xmath14 @xcite and the meson electromagnetic form factor pi - ff . in the following sections , we will first summarize our main results on the leading su(3 ) symmetry breaking of lcdfs , then study the three twist-3 lcdfs sequentially . finally , we will comment on calculations of those quantities using lattice qcd @xcite and light cone sum rules @xcite .
the two - parton lcdfs for pseudoscalar meson @xmath15 are defined by the matrix element of the quark bilinear operator @xcite @xmath26\lambda ^{a}q_{\alpha } ( -\frac{y}{2}n)|0\rangle & = & \frac{i}{4}\delta ^{ab}\int_{0}^{1}\!dx\,e^{i\left ( x-1/2\right ) \,yp\cdot n}\big\{f_{m}^{p}p\!\!\slash\gamma _ { 5}\,\phi _ { m}^{p}(x,\mu ) \notag \\ & & -\gamma _ { 5}\left ( f_{m}^{p}\phi _ { m}^{p}(x,\mu ) -\frac{y}{6}f_{m}^{\sigma } \sigma _ { \mu \nu } \,p^{\mu } n^{\nu } \,\phi _ { m}^{\sigma } ( x,\mu ) \right ) \big\}_{\alpha \beta }
.\quad \label{eq:2}\end{aligned}\]]where @xmath27 is a constant light - like vector , @xmath28 and our octet matrices are normalized so that @xmath29=\delta ^{ab}$ ] .
@xmath30 $ ] denotes the wilson line connecting the quark bilinear located at different space - time points on a light cone .
@xmath31 is a twist-2 lcdf while @xmath0 and @xmath24 are twist-3 lcdfs for pseudoscalar meson @xmath15 .
@xmath32 is the quark momentum fraction and @xmath33 is the perturbative qcd renormalization scale . for simplicity we work in the isospin limit and the @xmath34 scheme , and
normalize the distributions so that @xmath35 with @xmath36 , @xmath37 , @xmath38 .
generically from chiral symmetry the leading order su(3 ) violation for @xmath39 takes the form @xmath40 $ ] @xmath41\ .
\label{para}\end{aligned}\]]the functions @xmath42 , @xmath43 , and @xmath44 are independent of @xmath3 , and are only functions of @xmath45 , @xmath33 , and @xmath32 .
@xmath44 also depends on the chpt dimensional regularization parameter @xmath46 which cancels the @xmath47 dependence , so by construction @xmath48 is @xmath46 independent . throughout the text
, @xmath49 denotes the purely octet meson . using chpt
, we found very similar results to the twist-2 case @xcite at leading order in @xmath50 violation [ @xmath51 : \1 ) the twist-2 and twist-3 lcdfs are analytic in @xmath3 , meaning that @xmath52the leading logarithmic corrections can all be absorbed by the normalization constants @xmath53 .
\2 ) by charge conjugation and isospin symmetry , @xmath54 and by isospin symmetry @xmath55 these two equations are true to all orders in @xmath3 .
3)@xmath56where @xmath57 is @xmath3 independent .
\4 ) gell - mann - okubo - like relations exist among the octet mesons @xmath58\ .
\label{d}\ ] ] \5 ) the three - parton lcdfs also have relations similar to 1)-4 ) [ see eqs.(ee-[eee ] ) ] .
\6 ) statements 1)-5 ) are still true in quenched and partially quenched simulations , and with the leading finite volume and finite lattice spacing corrections .
\7 ) the light cone sum rule results @xcite for twist-3 gegenbauer moments are consistent with the chpt prediction @xmath59 $ ] @xmath60the analogous chpt relation for twist-2 moments puts a tight constraint on the numerical values of @xmath61 .
the operator product expansion of the non - local quark bilinear operator in eq.([eq:2 ] ) gives rise to the twist-2 operator @xmath62 and twist-3 operators @xmath63 and @xmath64 : @xmath65where @xmath66 and @xmath67 here having the vector indices dotted into @xmath68 has automatically projected onto the symmetric and traceless part .
the matrix elements of these operators yield@xmath69where @xmath70 and the moments are defined as @xmath71 following similar procedures as in the twist-2 case @xcite , we analyze the twist-3 matrix elements . it is convenient to write @xmath73where @xmath74^{k}\psi _ { r},\]]and similarly for @xmath75 .
@xmath76\psi $ ] is the left(right)-handed quark field .
the distinction between @xmath77 and @xmath78 is only for bookkeeping purposes .
we will set @xmath79 at the end .
when @xmath80 or @xmath81 , @xmath63 transforms simply under charge conjugation ( @xmath82 ) , being even when @xmath83 is even , and odd when @xmath83 is odd
. the meson states @xmath84 and @xmath49 ( i.e. , @xmath85 ) are @xmath82 even . thus from eq .
( [ eq : p - mat ] ) , @xmath86 vanishes when @xmath83 is odd due to @xmath82 ( and using isospin the same applies for @xmath87 ) .
for all @xmath88 s the operator would transform as @xmath89if we demanded that , under the @xmath82 transformation @xmath90 to construct the corresponding hadronic chpt operators , we define @xmath91 and @xmath92 . under a chiral @xmath93 transformation , we have @xmath94under charge conjugation , @xmath95 , @xmath96 , while @xmath97 transform according to eq .
( [ lambda ] ) . at next to leading order ( nlo ) in the @xmath98 and @xmath99 chiral expansion , @xmath100where the @xmath101 s are the leading order ( lo ) and the @xmath102 s are nlo .
the sum on @xmath103 runs over hadronic operators which have the same transformation properties as @xmath63 .
the chpt wilson coefficients @xmath104 and @xmath105 encode physics at the scale @xmath106 and the operators encode @xmath107 . at lo in the chiral expansion only one operator contributes in our analysis : @xmath108 \mathrm{tr}\left [ \lambda _ { lr}^{a}\box ^{k}\sigma ^{\dag } -\lambda _ { rl}^{a}\box ^{k}\sigma \right ] , \label{eq : p - lo}\]]where @xmath109 , @xmath110 in the chiral limit .
@xmath111 vanishes when @xmath83 is odd due to charge conjugation and su(3 ) symmetry .
the normalization of @xmath112 is chosen such that @xmath113 .
note that the operator @xmath114 \mathrm{tr}\left [ \lambda _ { lr}^{a}\sigma ^{\dagger } \left ( \box ^{k}\sigma \right ) \sigma ^{\dagger } -\lambda _ { rl}^{a}\sigma \left ( \box ^{k}\sigma ^{\dagger } \right ) \sigma \right]\]]is also lo but not independent of @xmath111 . since @xmath115 , @xmath116where the ellipse denotes @xmath117 terms that only contribute for matrix elements with more than one meson .
thus , eq . ( [ eq : p - movebox ] ) allows us to move factors of @xmath118 onto a neighboring @xmath119 and reduce @xmath120 to @xmath111 . at nlo
there are two independent operators : @xmath121 \mathrm{tr}\left [ m_{q}\sigma ^{\dag } + \sigma m_{q}^{\dag } \right ] \mathrm{tr}\left [ \lambda _ { lr}^{a}\box ^{k}\sigma ^{\dag } -\lambda _ { rl}^{a}\box ^{k}\sigma \right ] , \label{eq : p - nlo-1 } \\
\overline{\mathcal{o}}_{k,2}^{p , a } & = & \frac{f^{p}f}{8}\mathrm{tr}\big[\lambda _ { lr}^{a}\left\ { m_{q}^{\dag } \sigma \box ^{k}\sigma ^{\dag } + ( -1)^{k}\left ( \box ^{k}\sigma ^{\dag } \right ) \sigma m_{q}^{\dag } \right\ } \notag \\ & & -\lambda _ { rl}^{a}\left\ { m_{q}\sigma ^{\dag } \box ^{k}\sigma + ( -1)^{k}\left ( \box ^{k}\sigma \right ) \sigma ^{\dag } m_{q}\right\ } \big ] . \label{eq : p - nlo-2}\end{aligned}\]]all other nlo operators have derivatives on more than one @xmath119 , or can be reduced to @xmath122 and @xmath123 using the equations of motion .
for instance , consider @xmath124.\end{aligned}\]]the sum and difference @xmath125 contain factors of @xmath126 and @xmath127 .
we can trade @xmath128 for operators with derivatives on more than one @xmath119 by using the equation of motion for @xmath119 , @xmath129together with the analogous equation for @xmath130 .
these operators with derivatives on more than one @xmath119 do not generate one - meson matrix elements at tree level and can be omitted from our analysis .
thus only @xmath131 contributes and for simplicity we trade this for @xmath123 .
we can also insert more @xmath119 or @xmath132 fields into @xmath123 and get @xmath133 $ ] .
but this operator can be reduced to @xmath123 using eq .
( [ eq : p - movebox ] ) . finally , we can consider operators where the power suppression is generated by derivatives rather than a factor of @xmath3 .
boost invariance requires that these operators still have @xmath83 factors of @xmath134 , so they will involve @xmath118 just like the operators we have been considering .
to get power suppression with derivatives we can either use @xmath135 which has derivatives on more than one @xmath119 , or @xmath136 which can be traded for operators with @xmath3 s using eq .
( [ eq : p - eom-1 ] ) .
therefore , the operators with @xmath3 s in eqs .
( eq : p - nlo-1),([eq : p - nlo-2 ] ) suffice .
0.cm -0.1
cm at nlo chiral logarithms can be obtained from loop diagrams involving the lo operators as shown in fig.[fig1 ] . for @xmath137
the operator @xmath138 is the pseudoscalar current whose fig.[fig1 ] graphs give the one - loop corrections to @xmath139 . for odd @xmath83
the one - loop graphs vanish due to the @xmath140 $ ] factor originated from @xmath82 . for any even @xmath141
the diagrams have a term where all derivatives act on the outgoing meson line , and this gives the same corrections as for @xmath139 .
the first diagram could have additional contributions from derivatives acting inside the loop but it is straightforward to show that these diagrams vanish identically since @xmath142 , and that the same holds true for lo operators with derivatives on more than one @xmath119 @xcite .
thus , we have shown that all possible non - analytic corrections are contained in @xmath139 at nlo .
this is true for every moment , and so we conclude that the leading order su(3 ) violation of @xmath143 is analytic in @xmath144 . at nlo the @xmath122 , @xmath123 and
the wave function renormalization counterterms all contribute : @xmath145for @xmath146 and @xmath147 , @xmath139 can be related to @xmath148 by relating @xmath149 and the derivative of @xmath150 in the operator level .
acting @xmath151 on both sides of @xmath152then by the equations of motion and eq.([eq : p - mat ] ) with @xmath153 one obtains @xmath154these relations should be reproduced to all orders in chpt . indeed
, direct computations of the leading chiral logarithmic corrections of @xmath139 , @xmath148 and @xmath155 confirm these results .
there is no relation between @xmath156 and @xmath157 , however , because @xmath158which is not proportional to @xmath159 away from the su(3 ) limit . by comparing the @xmath137 and @xmath160 cases in eq.([mat-1 ] )
, we obtain @xmath161\delta ^{ab}\left ( b_{k,1}-b_{0,1}c_{k,0}\right ) \notag \\ & & + \mathrm{tr}\left [ m_{q}\left ( \lambda ^{b}\lambda ^{a}+(-1)^{k}\lambda ^{a}\lambda ^{b}\right ) \right ] \left ( b_{k,2}-b_{0,2}c_{k,0}\left ( 1+(-1)^{k}\right ) /2\right ) \big\ } , \label{trm}\end{aligned}\]]where @xmath162 in the chiral limit and @xmath163 is used . for @xmath164 ( odd moments ) , the @xmath165 structure yields @xmath166for @xmath167 ( even moments ) , the @xmath165 structure yields @xmath168where @xmath169 and @xmath170 . by isospin symmetry and charge conjugation ,
the even moments of different pion states ( or kaon states ) are equal .
eq.([eq : p - even - moments ] ) implies a gell - mann - okubo - like relation : @xmath171 the moment relations imply the lcdf relations in eqs.([a]),([d ] ) .
they also imply useful relations among the frequently used gegenbauer moments , defined by @xmath172with @xmath173 . here
@xmath174 denote the gegenbauer polynomials which are even ( odd ) functions of @xmath175 when @xmath27 is even ( odd ) .
@xmath176eqs.([eq : p - odd - moments])-([eq : p - g - o-1 ] ) imply that @xmath177 following the analogous procedures , the twist-3 operator @xmath179 can be decomposed into @xmath180with @xmath181^{k+1}\psi _ { r}\]]and similarly for @xmath182 . under charge conjugation ,
@xmath183if we demand @xmath97 transform according to eq .
( [ lambda ] ) . at nlo in the chiral expansion , @xmath184 is matched to lo and nlo hadronic operators @xmath185 and
@xmath186 : @xmath187 again , at lo there is only one hadronic operator @xmath188 \mathrm{tr}\left [ \lambda _ { lr}^{a}\left ( n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma ^{\dag } \right ) -\lambda _ { rl}^{a}\left ( n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma \right ) \right ] , \label{eq : t - lo}\]]where @xmath189}=(n^{\nu } \partial ^{\mu } -n^{\mu } \partial ^{\nu } ) $ ] . at nlo , there are two independent operators , @xmath190 \mathrm{tr}\left [ m_{q}\sigma ^{\dag } + \sigma m_{q}^{\dag } \right ] \mathrm{tr}\left [ \lambda _ { lr}^{a}n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma ^{\dag } -\lambda _ { rl}^{a}n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma \right ] , \label{eq : t - nlo-1 } \\ \left ( \overline{\mathcal{o}}_{k,2}^{\sigma , a}\right ) ^{\mu \nu } & = & \frac{f^{\sigma } f}{24}\mathrm{tr}\big[\lambda _ { lr}^{a}\left\ { m_{q}^{\dag } \sigma n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma ^{\dag } + ( -1)^{k}\left ( n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma ^{\dag } \right ) \sigma m_{q}^{\dag } \right\ } \notag \\ & & -\lambda _ { rl}^{a}\left\ { m_{q}\sigma ^{\dag } n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma + ( -1)^{k}\left ( n^{[\nu } \partial ^{\mu ] } \box ^{k}\sigma \right ) \sigma ^{\dag } m_{q}\right\ } \big ] .
\label{eq : t - nlo-2}\end{aligned}\]]these yield the following relations . for @xmath164 ( odd moments ) , @xmath191 for @xmath167 ( even moments ) , @xmath192where @xmath193 and @xmath194 .
these imply and@xmath195 in terms of the gegenbauer polynomials@xmath196such that the gegenbauer moments are@xmath197with @xmath198 . here
@xmath199 is an even ( odd ) function of @xmath175 when @xmath27 is even ( odd ) , thus we have @xmath200 the normalization @xmath201 is related to @xmath148 for @xmath146 and @xmath147 . by contracting @xmath202with @xmath203 and making use of @xmath204 , one obtains@xmath205 \ , \notag \\ f_{k}^{\sigma } & = & f_{k}^{p}\left ( \frac{m_{k}^{2}}{\overline{m}+m_{s}}\right ) \left [ 1-\left ( \frac{\overline{m}+m_{s}}{m_{k}}\right ) ^{2}\right ] \ .
\label{ff}\end{aligned}\]]direct computations of the chiral logarithms of @xmath206 and @xmath207 also confirm these results .
again , @xmath208 and @xmath157 are not related unless in the su(3 ) limit .
the three - parton lcdf @xmath209 for meson @xmath15 is defined by braun1 : @xmath210\sigma _ { \mu \nu } \gamma _ { 5}gg_{\sigma \rho } ( vyn)\lambda ^{a}[vyn ,-
yn]q(-yn)|m^{b}(p)\rangle \notag \\ & = & if_{m}^{3}\delta ^{ab}\left [ p_{\sigma } p_{[\mu } g_{\nu ] \rho } -\left ( \sigma \leftrightarrow \rho \right ) \right ] t_{m}(v , yp\cdot n)\ , \end{aligned}\ ] ] where @xmath211 is the strong coupling constant and @xmath212 is the gluon field strength tensor , @xmath213and@xmath214and where @xmath215 , @xmath216 and @xmath217 are momentum fractions carried by @xmath218 , @xmath219 and the gluon . for simplicity
, we use the gauge @xmath220 , such that the wilson lines @xmath221=$ ] @xmath222=1 $ ] . after the operator product expansion ,
the resulting twist-3 operator is @xmath223 \lambda ^{a}\left ( in\cdot \overrightarrow{\partial } \right ) ^{k - i}q\ , \]]where we have suppressed the @xmath33 , @xmath224 , @xmath38 , and @xmath225 indexes on the left hand side .
note that @xmath226 transforms in the same way as @xmath227 under chiral transformation and @xmath228thus the matrix element @xmath229\rho } -\left ( \sigma \leftrightarrow \rho \right ) \right ] \left ( n\cdot p\right ) ^{k+j}\int d\underline{\alpha } \left ( \alpha ^{\prime } -\alpha
\right ) ^{k}\alpha _ { g}^{j}\mathcal{t}(\alpha ^{\prime } , \alpha , \alpha _ { g})\]]implies@xmath230@xmath231@xmath232 = \mathcal{t}_{\pi } ( \alpha ^{\prime } , \alpha , \alpha _ { g})+3\mathcal{t}_{\eta } ( \alpha ^{\prime } , \alpha , \alpha _ { g})\ , \label{eee}\]]together with @xmath233 and @xmath234 by isospin .
eq.([ee ] ) was also obtained in refs .
@xcite .
the fact that @xmath235 , @xmath236 and @xmath237 transform in the same way under chiral transformation implies that the normalization constants @xmath238 , @xmath201 and @xmath139 receive the same leading logarithmic corrections .
this is true between @xmath139 and @xmath201 in eqs .
( [ f ] ) and ( [ ff ] ) due to the similar structures of the corresponding lo operators in eqs .
( eq : p - lo ) and ( [ eq : t - lo ] ) . also ,
at nlo the ratio@xmath239satisfies@xmath240and@xmath241analogous relations also exist for the ratio @xmath242 at nlo . as pointed out in ref .
@xcite ( see also @xcite ) , the three twist-3 lcdfs are related by the equations of motion .
this implies in chpt , the counterterms associated with the three lcdfs are also related .
it can be shown straightforwardly that the su(3 ) relations presented in this work are consistent with the constraints imposed by the equations of motion .
the first few moments of the twist-2 and twist-3 lcdfs can be calculated using lattice qcd .
currently only the second twist-2 moment of pion @xmath243 is computed lattice0,lattice1,lattice2 .
the latest results on @xmath244 have converged to @xmath245 gev@xmath246 @xcite and @xmath247 gev@xmath248 lattice2 with the lightest @xmath249 to be @xmath250 mev and @xmath251 mev , respectively . in these two calculations , and many others , chiral extrapolations to the physical point
@xmath252 mev were performed .
thus the model - independent results on the @xmath3 dependence of lcdfs obtained in ref .
@xcite and this work are valuable to reduce the systematic errors in the extrapolations .
a very interesting result we found both in the twist-2 and twist-3 lcdfs is that , in the continuum , all the moments of those distributions are analytic in @xmath3 at @xmath253 .
remarkably , this nice feature is retained in a lattice theory with ( partially ) quenched , finite volume and the leading finite lattice spacing effects included .
it is easy to see why . in all the cases ,
the leading logarithms are generated by one loop diagrams with insertion of the lo @xmath254 operator [ with the trace replaced by super - trace when ( partially ) quenched ] .
these diagrams correct @xmath255 by a universal ( @xmath83 independent ) logarithm which can then be absorbed into the normalization constant .
thus the moments are analytic in @xmath3 .
this feature , however , is only true at @xmath253 . at @xmath256 ,
the @xmath257 corrections to @xmath255 are generated by two loop diagrams with insertion of the lo @xmath254 operator .
these corrections are @xmath83independent and can be absorbed into the normalization factors .
the @xmath258 corrections , however , can arise from one - loop diagrams with insertion of the nlo @xmath259and @xmath260 operators which depend on @xmath261 and @xmath262 and correct @xmath255 in a @xmath83 dependent way , thus the @xmath258 corrections can not be absorbed by the normalization constants . generalizing the above argument to @xmath263 ,
the @xmath264 corrections can be absorbed by the normalization constants , thus asymptotically , the lcdfs has the quark mass dependence contribution to the twist-2 matrix element @xmath265 , where @xmath266 . if the diagram has a @xmath267 pionic operator insertion , @xmath268 internal propagators , @xmath269 loops , and @xmath270 vertex of order @xmath271 , then @xmath272 .
after using the topological identity @xmath273 and removing the @xmath274 kinematical factor , this diagram gives a @xmath275 contribution to the moment @xmath276 with @xmath277 . since @xmath278 and @xmath279 , @xmath280 .
in general , the diagram has the contribution @xmath281 , with the power of logarithm not bigger than @xmath282 and @xmath283 .
the @xmath284 term can only come from the diagram with an insertion of the lo pionic operator ( @xmath285 ) and can be absorbed into the normalization .
the arguments can be easily generalized to the twist-3 cases . ]
@xmath286where @xmath287 and @xmath288 are independent of @xmath3 . for different types of lattice simulations , the leading chiral corrections from the one loop diagrams can still be absorbed into the normalization constants , such that the lcdfs remain analytic at nlo .
different types of simulations give different normalization constants and different analytic contributions to the lcdfs .
for example , the @xmath289 term in eq.([eq : p - even - moments ] ) is originated from @xmath290 $ ] in eq.([trm ] ) . in the ( partially ) quenched theory , @xmath290 $ ]
is replaced by @xmath291 $ ] which is zero in the quenched theory quenched and is @xmath292 in the partially quenched @xcite theory , with @xmath293 denoting sea quark masses .
it is easy to check that eq.([eq : p - g - o-1 ] ) still holds in spite of ( partially ) quenching . as for finite volume effects , the counterterm structures are the same as those in the infinite volume calculations since the effects are infrared in origin . thus in simulations with twisted luscher , paolo , twistedx or partially twisted @xciteboundary conditions for fermions , the leading finite volume effects are absorbed by the normalization constant and the moments are all identical to those in the infinite volume calculations . for lattice fermions
whose chiral symmetry is broken by finite lattice spacing @xmath88 , new counterterms in powers of @xmath88 need to be added to account for the leading finite @xmath88 effects .
since these counterterms are su(3 ) symmetric , gell - mann - okubo - like relations like eq.([eq : p - g - o-1 ] ) will still be true .
the same is true for mixed action simulations with chiral valance quarks but wilson or staggered sea quarks .
the first few gegenbauer moments for the octet mesons have been computed using light cone sum rules lcsr0,lcsr0.1,lcsr0.2,lcsr1,ball:1998tj , lcsr2 .
the twist-2 moments @xmath61 ( defined in the same way as @xmath294 in eq.(g - moment ) ) were determined in different references .
quite different values for @xmath295 gev@xmath296 , @xmath297 @xcite and @xmath298-@xmath299 @xcite , are obtained using light cone sum rules . after combining these values with various constraints from experimental data data , datax , a reasonable range
is assigned @xcite : @xmath300for moments of @xmath147 and @xmath49 , light cone sum rules of refs . @xcite and
@xcite yield @xmath301respectively .
since the range of @xmath302 is big , the results in eqs.([bbb]),([bb ] ) are still consistent with the chpt relation found in ref .
@xcite ; @xmath303however , this equation could provide a tight constraint among @xmath61 since one expects the @xmath256 correction to be 10% at most .
as for @xmath304 , recent results yield @xmath305 @xcite , @xmath306 @xcite and @xmath307 @xcite .
the positive sign corresponds to the @xmath308-quark in a @xmath309 meson carrying a bigger momentum fraction than the @xmath310-quark . in ref .
@xcite , twist-3 gegenbauer moments are computed ( see also huang ) : @xmath311the numerical values of @xmath312 and @xmath313 are independent of @xmath314 while @xmath315 and @xmath316 with @xmath317 in the chiral limit .
eq.([cc ] ) satisfies the relations eqs.([a1]),([a2 ] ) and , hence , is consistent with chiral symmetry .
using chiral symmetry we have investigated the leading su(3 ) violation in the complete set of twist-3 light - cone distribution functions of the pion , kaon , and eta , including the two - parton distributions @xmath0 , @xmath1 and the three - parton distribution @xmath2 .
it has been shown that terms non - analytic in the quark masses do not affect the shape , and only appear in the normalization constants .
predictive power is retained including the leading analytic @xmath3 operators . with the symmetry violating corrections
we derive useful model - independent relations between @xmath318 , @xmath319 , @xmath320 , and @xmath321 .
we have also commented on the calculation of the moments of these distributions using lattice method and light - cone sum rules .
99 g.lepage and s.brodsky , phys.rev.*d22 * , 2157 ( 1980 ) ; a.v .
efremov et al . , phys .
b * 94 * , 245 ( 1980 ) .
chernyak and a.r .
zhitnitsky , phys .
rept . * 112 * , 173 ( 1984 ) .
m. beneke , g. buchalla , m. neubert , and c.t .
sachrajda , phys .
lett . * 83 * , 1914 ( 1999 ) ; nucl . phys . *
b591 * , 313 ( 2000);nucl . phys .
b * 606 * , 245 ( 2001 ) ; m. beneke and m. neubert , nucl .
b * 675 * , 333 ( 2003 ) .
y. h. chen , h. y. cheng , b. tseng and k. c. yang , phys .
d * 60 * , 094014 ( 1999 ) ; h. y. cheng , c. k. chua and a. soni , phys .
d * 71 * , 014030 ( 2005 ) .
keum , h - n .
li and a.i .
sanda , phys .
b * 504 * , 6 ( 2001 ) ; phys .
d * 63 * , 054008 ( 2001 ) ; y. y. keum and h. n. li , phys .
d * 63 * , 074006 ( 2001 ) .
bauer , s. fleming and m. luke , phys .
d * 63 * , 014006 ( 2001 ) ; c.w .
bauer , s. fleming , d. pirjol and i.w .
stewart , phys .
d * 63 * , 114020 ( 2001 ) ; c.w .
bauer and i.w .
stewart , phys .
b * 516 * , 134 ( 2001 ) ; c.w .
bauer , d. pirjol and i.w .
stewart , phys .
d * 65 * , 054022 ( 2002);c .
w. bauer , d. pirjol , i. z. rothstein and i. w. stewart , phys .
d * 70 * , 054015 ( 2004 ) ; c. w. bauer , i. z. rothstein and i. w. stewart , arxiv : hep - ph/0510241 .
see table 1 in i.w .
stewart , arxiv : hep - ph/0308185 .
j. w. chen and i. w. stewart , phys .
lett . * 92 * , 202001 ( 2004 ) .
d. arndt and m. j. savage , nucl .
a697 * , 429 ( 2002 ) .
j. w. chen and x. ji , phys .
b * 523 * , 107 ( 2001 ) ; phys .
lett . * 87 * , 152002 ( 2001 ) .
w. detmold _ et al .
_ , phys .
lett . * 87 * , 172001 ( 2001 ) ; phys .
d * 66 * , 054501 ( 2002 ) ; _ ibid _ * 68 * , 034025 ( 2003 ) .
j. w. chen and m. j. savage , nucl . phys .
a * 707 * , 452 ( 2002);phys . rev .
d * 65 * , 094001 ( 2002 ) ; j. w. chen and m. j. savage , phys .
d * 66 * , 074509 ( 2002 ) .
s. r. beane and m. j. savage , nucl .
a * 709 * , 319 ( 2002 ) .
w. detmold and c .- j .
d. lin , phys .
d * 71 * , 054510 ( 2005 ) .
j. w. chen and x. ji , phys .
* 88 * , 052003 ( 2002 ) ; a. v. belitsky and x. ji , phys .
b * 538 * , 289 ( 2002 ) .
n. kivel and m. v. polyakov , arxiv : hep - ph/0203264 .
n. kivel , m. v. polyakov and s. stratmann , arxiv : nucl - th/0407052 ; m. c. birse , j. phys .
g * 31 * , b7 ( 2005 ) .
s. r. beane and m. j. savage , nucl .
a761 , 259 ( 2005 ) .
b. c. tiburzi , arxiv : hep - ph/0508112 . b. grinstein and d. pirjol , phys .
b * 615 * , 213 ( 2005 ) ; v. cirigliano and d. pirjol , arxiv : hep - ph/0508095 .
j. w. chen and w. detmold , phys .
b * 625 * , 165 ( 2005 ) .
t. huang , x. h. wu and m. z. zhou , phys .
d * 70 * , 014013 ( 2004 ) .
a. s. kronfeld and d. m. photiadis , phys .
d * 31 * , 2939 ( 1985 ) ; g. martinelli and c. t. sachrajda , phys .
b * 190 * , 151 ( 1987 ) ; d. daniel , r. gupta and d. g. richards , phys . rev .
d * 43 * , 3715 ( 1991 ) ; l. del debbio , m. di
pierro , a. dougall and c. t. sachrajda [ ukqcd collaboration ] , nucl .
* 83 * , 235 ( 2000 ) .
v. m. braun and i. e. filyanov , z. phys .
c * 44 * , 157 ( 1989 ) [ sov .
j. nucl .
phys.*50 * , 511 ( 1989 ) ; yad . fiz . *
50 * , 818 ( 1989 ) ] .
s. v. mikhailov and a. v. radyushkin , phys .
d * 45 * , 1754 ( 1992 ) . t. huang , b. q. ma and q. x. shen , phys .
d * 49 * , 1490 ( 1994 ) .
p. ball , jhep * 9901 * , 010 ( 1999 ) .
p. ball , jhep * 9809 * , 005 ( 1998 ) .
p. ball and m. boglione , phys .
d * 68 * , 094006 ( 2003 ) .
a. hardmeier , e. lunghi , d. pirjol and d. wyler , nucl .
b * 682 * , 150 ( 2004 ) .
v. m. braun and i. e. filyanov , z. phys .
c * 48 * , 239 ( 1990 ) [ sov .
j. nucl .
phys . * 52 * , 126 ( 1990 ) ] ; v. m. braun , g. p. korchemsky and d. mller , prog . part .
phys . * 51 * , 311 ( 2003 ) . c. w. bernard and m. f. l. golterman , phys .
d * 46 * , 853 ( 1992 ) ; s. r. sharpe , phys .
d * 46 * , 3146 ( 1992 ) ; j. n. labrenz and s. r. sharpe , phys . rev .
d * 54 * , 4595 ( 1996 ) . c. w. bernard and m. f. l. golterman , phys .
d * 49 * , 486 ( 1994 ) ; s. r. sharpe , phys .
d * 56 * , 7052 ( 1997 ) [ erratum - ibid .
d * 62 * , 099901 ( 2000 ) ] ; s. r. sharpe and n. shoresh , phys . rev .
d * 64 * , 114510 ( 2001 ) .
m. luscher , s. sint , r. sommer and p. weisz , nucl .
b * 478 * , 365 ( 1996 ) .
p. f. bedaque , phys .
b * 593 * , 82 ( 2004 ) .
g. m. de divitiis , r. petronzio and n. tantalo , phys .
b * 595 * , 408 ( 2004 ) .
c. t. sachrajda and g. villadoro , phys .
b * 609 * , 73 ( 2005 ) .
p. f. bedaque and j. w. chen , phys .
b * 616 * , 208 ( 2005 ) .
a. schmedding and o. i. yakovlev , phys .
d * 62 * , 116002 ( 2000 ) ; m. diehl , p. kroll and c. vogt , eur .
j. c * 22 * , 439 ( 2001 ) ; a. p. bakulev , s. v. mikhailov and n. g. stefanis , phys .
b * 508 * , 279 ( 2001 ) [ erratum - ibid .
b * 590 * , 309 ( 2004 ) ] ; phys . rev .
d * 67 * , 074012 ( 2003 ) ; a. p.
bakulev , k. passek - kumericki , w. schroers and n. g. stefanis , phys .
d * 70 * , 033014 ( 2004 ) [ erratum - ibid .
d * 70 * , 079906 ( 2004 ) ] .
p. ball and r. zwicky , phys .
b * 625 * , 225 ( 2005 ) .
a. khodjamirian , t. mannel and m. melcher , phys . rev .
d * 70 * , 094002 ( 2004 ) .
v. m. braun and a. lenz , phys .
d * 70 * , 074020 ( 2004 ) .
p. ball and r. zwicky , arxiv : hep - ph/0510338 . t. huang , x. h. wu and m. z. zhou , phys . rev .
d * 70 * , 014013 ( 2004 ) ; t.
huang , m. z. zhou and x. h. wu , eur .
j. c * 42 * , 271 ( 2005 ) . | using chiral symmetry we investigate the leading su(3 ) violation in the complete set of quark twist-3 light - cone distribution functions of the pion , kaon , and eta , including the two - parton distributions @xmath0 , @xmath1 , and the three - parton distribution @xmath2 .
it is shown that terms non - analytic in the quark masses do not affect the shape , and only appear in the normalization constants .
predictive power is retained including the leading analytic @xmath3 operators . with the symmetry violating corrections
we derive useful model - independent relations between @xmath4 , @xmath5 , @xmath6 , and @xmath7
. we also comment on the calculations of the moments of these distributions using lattice qcd and light - cone sum rules . |
unlike the equilibrium thermodynamics and statistical mechanics , which are well developed after the pioneering works of boltzmann and gibbs , our understanding of non - equilibrium thermodynamics is restricted to some special models and cases .
stochastic lattice gases have provided a fertile testing ground for studying non - equilibrium stationary states in driven systems @xcite .
these models exhibit a variety of phase transition arising from a diffusive ( collisional ) relaxation .
for some of these models local equilibrium and hydrodynamic equations have been derived rigorously @xcite .
there are , however , other physical systems for which the approach to final stationary state is through a process of collisionless relaxation @xcite . gravitational systems and
confined one component plasmas are just two such examples . for these systems
the collision duration time diverges and the relaxation is governed by the collisionless boltzmann ( vlasov ) equation @xcite . in the thermodynamic limit , the collisionless relaxation process leads to non - maxwell - boltzmann velocity distributions , even for stationary states without macroscopic currents . unlike normal thermodynamic equilibrium , the stationary state which follows
the collisionless relaxation depends explicitly on the initial distribution of particle positions and velocities . in spite of this complication
, it was recently shown that it is possible to construct a statistical theory that quantitatively describes these states @xcite .
beams of electrons driven by accelerating vacuum devices , like the thermionic valves , diodes , and magnetrons , also do not relax to the maxwell - boltzmann distribution @xcite . unlike the driven stochastic lattice gases , these systems , however , are intrinsically collisionless . an important practical question concerns the kinetic temperature distribution in thermionic devices in which the directed velocity produced by the electric field is comparable to the thermal velocity @xcite .
this is particularly the case for the transitional region between child - langmuir and no - cutoff regimes in magnetrons , where the electric potential becomes comparable to the thermal energy @xcite . even when the final directed velocity is larger than the thermal velocity , there is a region near the emitting cathode where thermal effects are important .
it is of great practical interest to determine the extent of these regions @xcite .
furthermore , since in these systems the collision duration time diverges , there is no local equilibrium , and one can not _ a priori _ postulate an equation of state relating the beam density and the beam temperature , as for adiabatic or isothermal processes
. instead , given the properties of thermionic filaments such as say the velocity distribution of the emitted electrons one should solve the boundary value problem posed by the vlasov equation .
the purpose of this letter is to develop a theoretical framework which will allow us to study the relaxation dynamics and the stationary states of collisionless driven systems . as a prototype of a collisionless driven system
, we consider a beam of electrons , accelerated by an external electric field , traveling from an emitting ( planar ) cathode to a collecting ( planar ) anode across the device gap .
the cathode , located at position @xmath0 , is kept at electrostatic potential @xmath1 and is heated to temperature @xmath2 , resulting in the emission of electrons . after traversing the device gap ,
these electrons are collected at the cold anode ( @xmath3 ) located at @xmath4 and kept at potential @xmath5 . during the steady state operation ,
the region between the cathode and anode contains a total of @xmath6 electrons , resulting in a current density @xmath7 .
our goal is to relate @xmath7 , to the potential difference @xmath8 , the number of electrons @xmath6 , the device width @xmath9 , and the cathode temperature @xmath2 . for planar electrodes , particle distribution transverse to the @xmath10 axis
can be taken to be uniform .
furthermore , the one particle distribution function for a collisionless system in a steady state must satisfy the stationary vlasov equation , @xmath11 where @xmath12 is the elementary charge , @xmath13 is the electron mass , and @xmath14 is the static distribution function . in the thermodynamic limit ,
vlasov equation becomes exact for particles interacting by long range potentials @xcite .
it can be readily seen that the distribution functions of the form @xmath15 $ ] , where @xmath16 is the mean particle energy , @xmath17 , satisfy eq .
( [ equa1 ] ) . therefore ,
if @xmath18 is specified at @xmath0 , @xmath19 is then also determined for any other position , provided that the electrostatic potential @xmath20 is known .
this potential can , in turn , be calculated self - consistently from the solution of the poisson equation @xmath21 where the particle density @xmath22 is given by @xmath23 , the total particle number is @xmath24 , and the transverse cross sectional area of the essentially 1d device is @xmath25 . to represent both the thermal distribution near the cathode , and the fact that only particles with positive velocities actually move into the device gap , we choose at @xmath0 a unidirectional maxwellian distribution of the form @xmath26 where @xmath27 is the boltzmann constant , @xmath2 is the cathode temperature , and @xmath28 is the beam density at the cathode after the stationary state is achieved .
the value of @xmath28 can only be obtained once the full problem has been resolved .
the distribution function over the length of the whole diode is then integrating the distribution function @xmath31 $ ] over the possible values of velocity , we arrive at a nonlinear integro - differential equation for the electrostatic potential , @xmath32 it is important to note the difference between this equation and the poisson - boltzmann equation obtained for usual collisional plasmas and electrolytes in the mean - field limit @xcite . equation ( [ equa4 ] ) can be solved numerically , to yield the electrostatic potential and the distribution function for the electron beam in the stationary state . for systems with long range interactions ,
vlasov equation should become exact in the thermodynamic limit . to confirm this for our system
, we have performed molecular dynamics simulation of an equivalent one dimensional model .
the simulated system consists of @xmath33 mutually interacting charged sheets of area @xmath25 each containing @xmath34 electrons of the same velocity moving along the @xmath10 axis , under the action of the external electric field produced by the grounded cathode @xmath35 and an anode kept at a fixed potential @xmath36 .
the interaction potential between the two sheets @xmath37 is the green s function @xcite of the laplace equation , @xmath38 with the boundary conditions @xmath39 . solving this equation
we obtain @xmath40 , where @xmath41 and @xmath42 are the smaller and the larger of the two particle coordinates @xmath43 and @xmath44 .
the effective hamiltonian for the sheet dynamics is then @xmath45 where @xmath46 and @xmath47 are the charge and mass of each sheet respectively .
the acceleration of each simulated sheet then follows from the canonical equations of motion , where @xmath49}$ ] is the number of sheets to the left(right ) of the one considered , and @xmath50 denotes the positional average @xmath51 . since @xmath52 , from eq .
( [ equa5 ] ) one sees that the electron acceleration at the device entrance where @xmath53 } \rightarrow 0 $ ] and @xmath54 } \rightarrow n_s$ ] satisfies @xmath55 , which reveals that in _ space - charge dominated _ devices where @xmath56 , acceleration at beam entrance may be zero or even negative @xcite .
when the acceleration vanishes , the associated current is denoted as the _ limiting _ one .
since we wish to describe a hot cathode and a cycling current inside the device , we adopt the following strategy .
we advance the simulation in small time steps , always obeying eq .
( [ equa5 ] ) . whenever a particle crosses the anode and exits the system ,
it is re - injected at the cathode position . at this point
all the particles in a small region @xmath57 around the cathode are re - thermalized , so as to ensure that the distribution there keeps its original form of a truncated maxwellian .
the width @xmath57 must be sufficiently small , @xmath58 , but apart from this condition its precise value is arbitrary .
the simulations were performed with @xmath59 and @xmath60 . in all cases
we start with a uniform distribution of sheets and compute the observables only after the system reaches its final stationary state .
to compare the predictions of the theory with the results of the simulations , we consider the density and the temperature distributions inside the diode .
the kinetic temperature is defined as @xmath61 where the over - bar denotes the velocity average at a given position @xmath10 .
the theoretical averages are calculated using the distribution function @xmath31 $ ] , while in the simulations , the averages are performed over the particle velocities within narrow bins along the @xmath10 axis . note that because of the asymmetry of the velocity distribution at @xmath0 , @xmath62 .
it is convenient to scale space and time with the diode length @xmath9 and the plasma frequency @xmath63 , respectively .
dimensionless coordinate and velocity can then be defined as @xmath64 and @xmath65 .
in addition , eqs .
( [ equa5 ] ) and ( [ equa3 ] ) show that adimensional voltage and adimensional temperature can be defined as @xmath66 and @xmath67 , respectively , and serve as the control parameters for the system . in fig .
[ fig1.eps](a ) the scaled temperature @xmath68 is plotted against the scaled coordinate @xmath69 .
we consider @xmath70 and also consider a device operating at its limiting current , @xmath71 .
a striking feature of this plot is that the temperature drops rapidly as one moves away from cathode towards anode .
we next study the dependence of scaled density @xmath72 along the length of the diode .
the density is very high near the cathode , where the average velocity is small .
it then drops rapidly towards the anode , where particles are accelerated up to high speeds , see fig .
[ fig1.eps](b ) .
agreement between the simulations and the theory for both the kinetic temperature and density is excellent .
we now study the current - voltage phase diagram of the device .
in general , current is a function of the voltage drop , the temperature , the gap length , and the total charge of the device .
however , by measuring the time in units of one over the plasma frequency @xmath73 , and the length in units of the gap length @xmath9 , we can scale away two of these variables . the current density can be calculated using @xmath74 since in the steady state the current does not depend on either time or coordinate , integration along the @xmath10 axis and over the cross sectional area @xmath25 yields @xmath75 furthermore , since the current density is measured in units @xmath76 \sim [ ne v / a l]$ ] , rescaling it in terms of the gap length and the plasma frequency , we can write the reduced current density as @xmath77 , which then satisfies @xmath78 where @xmath79 is the reduced velocity averaged over _ all _ the particles . the reduced average velocity , in turn , must be a function of the two previously introduced control parameters : the reduced voltage and temperature .
eq.([sim ] ) is in fact a similarity transformation relating systems with different charge , length , temperature , and potential difference . in fig .
[ fig2.eps ] we plot @xmath80 vs. @xmath81 for various @xmath68 .
the phase diagram provides all the information about the current - voltage characteristics for all possible planar diodes .
the first feature to note is that all the different curves emanate from the limiting current backbone , which traces a temperature dependent path in the @xmath82 plane . to the left of the limiting current border , indicated by the solid line in fig .
[ fig2.eps ] , the distribution function can no longer be described by a unidirectional maxwellian , such as the expression ( [ equa3 ] ) .
the transition resembles bose - einstein condensation ( bec ) . in the case of bec
below the critical temperature a macroscopically populated ground state appears , and only a fraction of particles remains in the excited states .
similarly , in the case of our diode , to the left of the limiting curve , part of the charge must be expelled from the system before a stationary state can be achieved .
as @xmath83 , the voltage effects become negligible compared to the thermal ones , and the beam density becomes uniform across the gap . in this limit ,
it is possible to show that the dimensionless backbone curve asymptotes to a vertical line , @xmath84 , fig .
[ fig2.eps ] . vs. @xmath85 .
the thick solid line represents the limiting current and the thick dashed line , the zero temperature limit . to the left of the solid curve ,
charge must be expelled from the system before a stationary state can be achieved .
dotted lines represent the theoretical results for the indicated temperatures .
the circles are the results of the simulations at the same temperatures.[fig2.eps],width=302,height=264 ] to conclude , we have studied the dynamics of collisionless driven systems . unlike the stochastic lattice gasses which are significantly abstracted from reality , the models studied in this paper are very similar to real electronic devises , such as the thermionic valves , diodes , and magnetrons . furthermore ,
differently from the lattice gases whose dynamics is diffusive , the distribution function of collisionless systems satisfies the vlasov equation . for the class of driven systems introduced in this letter
, the stationary state vlasov equation can be solved exactly . the theory developed in this paper should
, therefore , be relevant to the design and operation of real electronic devises .
it is important to stress that in the absence of collisions , a charged beam does not relax to an equilibrium with a known equation of state .
in fact , the thermodynamic temperature is defined only in the vicinity of the hot emitting cathode .
away from the cathode , dynamics is controlled by the collisionless vlasov equation , which has to be solved as a boundary value problem .
once the solution is obtained , all the macroscopic quantities can be determined via appropriate averages .
the kinetic temperature is found to vary strongly across the device gap , precluding the use of conventional isothermal or adiabatic assumptions and of the hydrodynamic formalisms . | a statistical theory is presented which allows to calculate the stationary state achieved by a driven system after a process of collisionless relaxation .
the theory is applied to study an electron beam driven by an external electric field .
the vlasov equation with appropriate boundary conditions is solved analytically and compared with the molecular dynamics simulation .
a perfect agreement is found between the theory and the simulations . the full current - voltage phase diagram is constructed . |
the demand for new materials with interesting and useful physical properties has led to a fast development in material science .
properties such as superconductivity , magnetic ordering , nearly ferromagnetic fermi - liquid and heavy fermion behavior have been observed in many materials , including the family of complex intermetallic compounds @xmath15zn@xmath1 ( @xmath16 = rare earth , @xmath2 = transition metal).@xcite this family , first reported two decades ago by nasch @xmath17.@xcite , has been extensively used as a model system due to its rather unique structure which features a complex but well ordered crystal structure .
the @xmath15zn@xmath1 compounds have a cubic structure with @xmath18 ( no .
227 ) space group in which the @xmath16 and @xmath2 ions occupy the crystallographic sites 8@xmath19 and 16@xmath14 , respectively . moreover , these ions are each encapsulated in quasi - spherical cages formed exclusively by zn ions , which occupy three inequivalent wyckoff positions given by 16@xmath20 , 48@xmath21 , and 96@xmath22 as can be viewed in fig . [
fig : fig0 ] .
two particular properties observed in these materials have attracted great attention : a remarkably high magnetic ordering temperature observed in the @xmath16fe@xmath3zn@xmath1 series ( although it contains less than @xmath23 of @xmath16 ion ) and a nearly ferromagnetic fermi - liquid behavior in yfe@xmath3zn@xmath1.@xcite several works have argued , based on macroscopic measurements and band structure calculations , that the elevated magnetic ordering temperatures and the type of magnetic ordering are attributed to a high density of states ( dos ) at the fermi level.@xcite for instance , the co based compounds gdco@xmath3zn@xmath1 and tbco@xmath3zn@xmath1 present an antiferromagnetic ( afm ) ordering below @xmath24 5.7 k and 2.5 k , respectively , in which the larger @xmath16-@xmath16 distance ( @xmath24 6 ) supports an indirect interaction and a low transition temperature .
when the co ions are replaced by fe ions ( gdfe@xmath3zn@xmath1 and tbfe@xmath3zn@xmath1 ) the compounds exhibit ferromagnetic ( fm ) ordering with the transition temperatures drastically raised to 86 k and 66 k , respectively .
the relatively long distance between rare earth ions in the structure weakens the ruderman - kittel - kasuya - yosida ( rkky ) exchange interaction , while the high density of fe 3@xmath14 bands at the fermi level directly affects the conduction electrons .
recently , a detailed study of the magnetic structure of tbco@xmath3zn@xmath1 and tbfe@xmath3zn@xmath1 through magnetic neutron scattering at low temperature was reported.@xcite although the tb - based compounds present similar magnetic properties as compared to gdco@xmath3zn@xmath1 and gdfe@xmath3zn@xmath1 , the absence of crystalline electric field ( cef ) at first order and a strong rkky interaction in the gd - compounds affect the electronic and magnetic interactions between the rare earth ions and the surrounding matrix .
in particular , the investigation of the compounds with half - filled 4@xmath25 shell ( gd - based materials ) at low temperature can provide information about the physical properties and are very important as reference compounds due to their lack of orbital momentum , i.e. , @xmath26 and @xmath27 , which leads to magnetic properties that are unaffected by spin - orbit coupling . in order to understand the implications of such interactions we have used spectroscopy and magnetic scattering techniques to probe in detail the electronic and magnetic properties of the gd@xmath0zn@xmath1 family .
the large gd neutron absorption cross section leaves x - ray measurements as the ideal option to probe of the magnetic and electronic properties in these materials .
furthermore , the incoming beam energy can be tuned to the absorption edge , thus providing chemical and atomic selectivity , i.e. , the magnetic response of each element can be probed separately . here
we report the magnetic and electronic structure at low temperature of the gd@xmath0zn@xmath1 ( @xmath2 = fe , co ) compounds using x - ray resonant magnetic scattering ( xrms ) , x - ray absorption near - edge structure ( xanes ) and x - ray magnetic circular dichroism ( xmcd ) techniques .
the xrms measurements performed on gdco@xmath3zn@xmath1 reveal a commensurate antiferromagnetic ordering with a magnetic propagation vector @xmath28 below @xmath29 k in which the gd magnetic moments are aligned following the magnetic representation @xmath7 .
the xmcd measurements performed below the curie temperature ( @xmath30 k ) in gdfe@xmath3zn@xmath1 display a dichroic signal of 12.5 @xmath8 and 9.7 @xmath8 of the absorption jump for gd @xmath31 and @xmath11 edges , respectively .
surprisingly , a magnetic signal of about 0.06 @xmath8 is detected at the zn @xmath12-edge which suggests that the zn ions are spin polarized .
this magnetic signal might originate from the hybridization between the extended gd 5@xmath14 bands with the empty zn 4@xmath32 states .
absorption measurements performed at the fe @xmath12-edge do not reveal any magnetic contribution coming from the iron ions above the background level .
high quality single crystals of gdfe@xmath3zn@xmath1 and gdco@xmath3zn@xmath1 were grown at ufabc by zn self flux method@xcite similar to that reported in previous studies on the family.@xcite in order to perform the absorption measurements at the gd @xmath33 , fe and zn @xmath12 edge , selected single crystals of gdfe@xmath3zn@xmath1 were ground and sieved , resulting in fine powders with grain sizes around @xmath34-@xmath35 .
the magnetic diffraction measurements were done on a high quality gdco@xmath3zn@xmath1 single crystal cut to dimensions of approximately 2 x 2 x 0.5 mm@xmath36 .
the crystalline piece was carefully polished to achieve a flat surface perpendicular to the @xmath37 $ ] direction , yielding a mosaic width of approximately 0.02@xmath38 .
the phase purity of the samples was confirmed by powder diffraction using conventional laboratory x - ray sources .
temperature dependent magnetic susceptibility measurements ( not shown here ) were performed using a commercial superconducting quantum interference device ( quantum design mpms - squid ) to verify the magnetic ordering temperatures , the effective magnetic moments and the curie - weiss constants .
the resonant diffraction measurements were performed at beamline 6-id - b at the advanced photon source ( aps ) , argonne national laboratory ( argonne , il / usa ) , whereas the absorption measurements were conducted at 4-id - d ( aps)@xcite and at beamline p09 at petra iii ( desy , hamburg / germany).@xcite xanes and xmcd spectra obtained at low temperature for the gd @xmath33 , fe and zn @xmath12-absorption edges were performed in transmission geometry on powdered gdfe@xmath3zn@xmath1 samples .
the samples were cooled down by a displex cryostat with base temperature around 7 k. xmcd spectra were performed in helicity switching mode in which the left and right circular polarization was obtained by means of diamond phase plates.@xcite the degree of circularly polarized beam was higher than 95 @xmath8 for both beamlines ( p09 and 4-id - d).@xcite an external magnetic field of h = 2.0 t ( at aps ) and 0.8 t ( at desy ) was applied in the gdfe@xmath3zn@xmath1 samples along and opposite to the incident beam wave vector @xmath39 to align the ferromagnetic domains and to correct for non - magnetic artifacts in the xmcd data .
those external magnetic fields were enough to reach the saturation magnetization according to the macroscopic measurements .
xrms measurements were performed at t = 4.5 k on gdco@xmath3zn@xmath1 single crystal , mounted inside the closed - cycle displex cryostat in a six - circle diffractometer at the 6-id - b beamline .
the single crystal was oriented with the @xmath37 $ ] direction parallel to the vertical diffraction plane .
several magnetic superlattice reflections of the type @xmath40 with @xmath41 were measured and their integrated intensities were compared to the simulated intensities to determine the magnetic structure below @xmath42 . to enhance the magnetic bragg peak intensities ,
the energy of the incident beam was tuned near the gd @xmath10 or @xmath11 absorption edges .
in addition , in order to investigate any magnetic contribution from co and zn ions , the energy of the incident beam was also tuned to the co ( 7709 ev ) and zn ( 9659 ev ) @xmath12-edges and a search for superlattice reflections was performed .
the charge and magnetic contributions present in the scattered beam were separated by a pyrolytic graphite [ c@xmath43 analyzer crystal installed on the 2@xmath44 arm of the diffractometer . since the incident beam presents the polarization perpendicular to diffraction plane ( @xmath45 polarization ) , by rotation of the analyzer crystal around the scattered beam wave vector @xmath46 we were able to select the two polarization channels ( @xmath47 and @xmath48 ) in this experimental geometry.@xcite
the experimental results are organized into two sections : the first part is dedicated to the absorption measurements on powdered gdfe@xmath3zn@xmath1 in its fm state .
the subsequent section shows the results obtained by the xrms technique on the gdco@xmath3zn@xmath1 single crystal in its afm state .
xanes and xmcd measurements performed at the gd @xmath33 edges in gdfe@xmath3zn@xmath1 are shown in fig . [
fig : fig1 ] .
dipolar selection rules make the dichroic signal at the @xmath33 absorption edges sensitive to the spin polarization of the intermediate 5@xmath14 level .
the gd xanes reported in fig .
[ fig : fig1 ] are normalized to one at the @xmath11 and half at the @xmath10 edge to reflect the 2:1 ratio of the initial state at these edges ( 2@xmath49 and 2@xmath50 , respectively ) .
figure [ fig : fig1 ] also shows the xmcd spectra at the gd @xmath33 edges in which each spectrum is normalized to the corresponding edge jump of the absorption spectrum .
the xmcd signal obtained across the two edges show different intensities with a strong dichroic magnetic signal around 12.5 @xmath8 at the @xmath10 and 9.7 @xmath8 at the @xmath11 absorption edges , which is consistent with gd - based compounds.@xcite the size and the shape of the magnetic contribution obtained by fitting the xmcd signals with lorentzian function can describe additional properties of this system .
the widths of the dipolar contributions ( e1 ) contributions observed at the @xmath10 and @xmath11 absorption edges are 4.3(2 ) ev and 4.5(2 ) ev , respectively , which reflects a short 2@xmath13 core hole lifetime . using the integrated intensities ,
the @xmath51 ratio ( or branching ratio value - br)@xcite obtained experimentally is @xmath52 .
absorption measurements at the fe and zn @xmath12 edges were also carried out on powdered samples .
the absorption measurement near the @xmath12 edge , in which the dipolar transition is probed ( 1@xmath53 4@xmath13 ) , is crucial towards understanding the magnetic and electronic properties due to the delocalized character of the @xmath13 states.@xcite . since the probed @xmath13 states are very delocalized
, a strong influence of the surrounding matrix can be expected due to the hybridization between the rare earth and the transition metal ions .
as shown in figure [ fig : fig1_n ] , the measurements performed near the fe @xmath12 edge do not reveal any magnetic contribution from the fe ions higher than the background level ( @xmath24 0.07 @xmath8 ) .
the inset in fig .
[ fig : fig1_n ] exhibits the xmcd measurements obtained for a 5 @xmath54 m fe - foil in the same experimental conditions : a clear dichroic signal can be observed near the edge .
however , the spectroscopy measurement at the zn @xmath12 edge manifests an interesting behavior .
figure [ fig : fig2 ] shows the absorption and dichroism results at the zn @xmath12 edge in which an induced magnetic signal around 0.06 @xmath8 is detected .
this magnetic signal is due to a hybridization with the rare earth 5@xmath14 orbitals .
the xmcd spectrum exhibits the main positive feature located at 9665 ev with a width around 2.5 ev , surrounded by two negative peaks 6.5 ev away .
the broad feature localized around 9680 ev ( @xmath24 20 ev above the edge ) is likely due to magnetic exafs.@xcite a clear evidence of the zn 4@xmath13 states polarization due to the gd ions can be found in the temperature and field dependence reported in panels [ fig : fig2tdep](a ) and [ fig : fig2tdep](b ) , respectively .
the magnetic intensities for both gd and zn ions follow the same temperature evolution and disappear around the critical temperature ( t@xmath55 85 k ) .
in addition , the two hysteresis loops obtained at the maximum xmcd intensity shows clearly the zn magnetism dependence in relation to the gd ions and therefore it suggests a spin polarization of the zn 4@xmath13 bands by the gd sub - lattice .
figure [ fig : fig3 ] shows the evolution of the integrated intensity for the magnetic bragg reflection @xmath56 as a function of temperature for the gdco@xmath3zn@xmath1 compound fitted by a lorentzian - squared function .
the magnetic peak intensity decreases smoothly to zero as the temperature approaches @xmath42 , indicating a standard second order phase transition from an afm to a paramagnetic state . a dashed red line in fig .
[ fig : fig3 ] displays a fitting using a critical power - law expression , @xmath57 , above 5.0 k. the fitting around the nel temperature yields a @xmath42 = 5.72(6 ) k and a critical exponent @xmath58 = 0.36(3 ) .
the value of @xmath42 is in good agreement with bulk magnetic susceptibility measurements and previous works.@xcite the critical exponent @xmath58 close to 0.367 suggests a three - dimensional ( 3d ) heisenberg magnetic model .
@xcite in blue symbol ( fig .
[ fig : fig3 ] ) is also reported the full width at half maximum ( fwhm ) of the magnetic superlattice peak @xmath56 as a function of temperature .
this figure clearly shows a peak broadening and a decrease in intensity near the phase transition temperature characteristic of a loss of long - range order .
energy dependences across the gd-@xmath10 and @xmath11 edges performed at 4.5 k are displayed in fig .
[ fig : fig4 ] .
the top panels [ fig . [ fig : fig4](a ) and [ fig : fig4](c ) ] show the normalized absorption coefficients ( @xmath54 ) obtained from the fluorescence yield , while the bottom panels [ fig . [ fig : fig4](b ) and [ fig : fig4](d ) ] exhibit the energy profile obtained at the magnetic superlattice position @xmath59 with the analyzer crystal set to the @xmath48 polarization channel .
a resonant enhancement of over two orders of magnitude at both absorption edges can be seen .
in addition , the maximum intensities are observed about 2 - 3 ev above the absorption edge ( defined by the vertical dashed lines ) , which is a characteristic signature of a dipole electronic transition .
the same energy dependence was performed in the @xmath47 polarization channel , and no significant contribution was observed .
the strong resonant enhancement in the spectra [ fig .
[ fig : fig4](b ) and [ fig : fig4](d ) ] indicates a significant overlap between the initial 2@xmath13 and 5@xmath14 states , and a strong exchange interaction between the 4@xmath21 - 5@xmath14 orbitals .
this magnetic polarization of the 5@xmath14 bands via 4@xmath21 states helps shed light on the magnetic structures of these rare earth based materials using the @xmath60 absorption edge measurements , i.e. , 2@xmath61 5@xmath14 transitions .
moreover , the asymmetric peak shape expressed as a long tail below the absorption edges arises from the interference between the resonant and non - resonant magnetic scattering contributions.@xcite the normalized energy line shape dependence after absorption correction for selected magnetic bragg peaks @xmath62 with @xmath41 performed at gd-@xmath10 and @xmath11 edges are displayed in fig .
[ fig : fig5 ] .
the magnetic reflections show a narrow resonant line shape at the @xmath10 ( @xmath63 4.9 ev ) and @xmath11 ( @xmath63 5.9 ev ) edges .
the smaller energy broadening for the @xmath10 edge is associated with a relatively short lifetime compared with the @xmath11 edge .
the magnetic reflections were only observed in the @xmath48 polarization channel ; in the @xmath47 channel , no intensity was detected .
the integrated intensities of the magnetic reflections were used to extract the @xmath51 ratio and to determine the direction of magnetic moment .
the br values show an unusual behavior in which the values varies between @xmath64 - @xmath65 for different magnetic reflections . for xrms ,
the branching ratio is expected to be equal to 1 .
the exact value for the br ratio is difficult to obtain due to the various corrections that must be applied to the experimental data , such as self - absorption and angular corrections , and therefore , we can not affirm that @xmath66 for this afm compound .
the magnetic structure of the gd spins are determined comparing the experimental integrated intensities in fig .
[ fig : fig5 ] with simulated data from selected magnetic reflections .
the program sara_h _
@xcite was run to determine the possible magnetic arrangements that the gd ion can adopt inside the unit cell , i.e. , the magnetic representation ( @xmath67 ) .
in addition , we assumed that only the gd ions carry magnetic moments in this compound .
for this material , whose magnetic propagation vector is @xmath5 , the space group is @xmath18 , and the gd ions occupy the 8@xmath19 crystallographic site , the magnetic representation ( mr ) can be decomposed into four non - zero irreducible representations ( irs ) : two one - dimensional ( 1d - @xmath68 ) and two two - dimensional ( 2d - @xmath69 ) irs .
the four possible magnetic representations for the gdco@xmath3zn@xmath1 compound are summarized in , where the labeling of the propagation vector and the magnetic representation follows the kovalev notation.@xcite .basis vectors ( bv ) for the space group @xmath18:2 with @xmath70 .
the decomposition of the magnetic representation ( mr ) for the gd site can be written like @xmath71 .
the two rare earth atoms positions of the nonprimitive basis are defined according to 1 : @xmath72 and 2 : @xmath73 . [
cols="^,^,^,^,^,^,^,^,^,^,^,^,^,^,^,^,^,^,^ " , ] to determine the magnetic structure , the intensities were calculated assuming only dipole transition ( e1 ) and hence , the x - ray magnetic scattering cross section model can be written as:@xcite @xmath74 where @xmath75 and @xmath76 .
f^{e1}_{n } & = [ ( \hat{\varepsilon}'\cdot \hat{\varepsilon})f^{(0 ) } - i(\hat{\varepsilon}'\times \hat{\varepsilon } ) \cdot \hat{z}_n f^{(1 ) } \\ & + ( \hat{\varepsilon}'\cdot \hat{z}_n ) ( \hat{\varepsilon } \cdot \hat{z}_n)f^{(2 ) } ] .
\end{split}\end{aligned}\ ] ] the term @xmath77 [ eq . ( [ eq : eq3 ] ) ] contains the absorption correction and the lorentz factor .
@xmath78 is the angle between the wave - vector transfer @xmath79 and the @xmath37 $ ] crystal direction , and @xmath44 is half of the 2@xmath44 scattering angle .
( [ eq : eq4 ] ) shows the resonant term .
it carries information about the @xmath80 ( @xmath81 ) and @xmath46 ( @xmath82 ) , i.e. , the incident and scattered wave ( polarization ) vectors , respectively , and the magnetic moment direction at _
n__th site ( @xmath83 ) .
the terms @xmath84 are related to the dipole matrix transition and by atomic properties.@xcite the exponential function in eq .
[ eq : eq2 ] is a function of the wave - vector transfer @xmath85 , and the position @xmath86 of the _
_ n__th gd ion inside the unit cell . for the xrms technique probing afm materials
, we assumed that the magnetic intensity at the first harmonic satellites are due to only dipole contribution comes from the linear term on magnetic moment direction [ @xmath87 displayed in eq .
( [ eq : eq4 ] ) .
the simulated intensities obtained using eq .
( [ eq : eq2])-([eq : eq4 ] ) and the experimental intensities obtained at the gd @xmath10 edge [ fig . [ fig : fig4](g - l ) ] are displayed in fig .
[ fig : fig6 ] . comparing the intensities ,
the best agreement between experimental ( symbol , @xmath88 ) and simulated data ( curves ) is obtained for the magnetic representation @xmath89 . to identify different magnetic propagation vectors , such as @xmath90 and @xmath91 , a systematic search for commensurate and incommensurate magnetic reflections in the reciprocal space was performed below @xmath42 .
however , only magnetic bragg reflections of the type @xmath92 were observed .
in addition , to probe a possible presence of afm magnetic moments in the co and zn ions , the beam energy was tuned to the co and zn @xmath12 edges .
search for magnetic superlattice reflections were performed below @xmath42 , and no measurable magnetic reflection at 4.5 k could be observed .
the nature of the electronic and magnetic properties of the intermetallic @xmath15zn@xmath1 systems depends strongly on the interactions between the rare earth and the transition metals ions . since the gd 5@xmath14 states participate on the conduction bands , the resonant absorption and diffraction measurements at the gd @xmath33 edges provide valuable information . as reported in susceptibility measurements and band structure calculations@xcite
, the replacement of the transition metal affects the electronic density of states at the fermi level ( @xmath93 ) and the conduction band without changing significantly the lattice parameters.@xcite doping studies of gd(fe@xmath94co@xmath95zn@xmath1 compounds shows a monotonic increase of the magnetic ordering temperature when @xmath96 increases to 1 which indicates that there is an increase in the coupling between the rare earth magnetic moments.@xcite absorption measurements performed in powdered samples of fm gdfe@xmath3zn@xmath1 compound below @xmath97 reveal interesting behaviors . as can be seen in fig .
[ fig : fig1]-[fig : fig2tdep ] , only at the gd @xmath60 and zn @xmath12-edges a dichroic signal is observed above the background level whilst an unexpected lack of magnetic intensity is observed at the fe @xmath12-edge .
the intense magnetic signal at the rare earth @xmath60-edges occurs mainly due to the overlap between the gd 2@xmath13 and 5@xmath14 states and a strong energy splitting of the 5@xmath14 sub - bands as a result of a 4@xmath21 - 5@xmath14 exchange interaction.@xcite in addition , the splitting of the @xmath14 state into 5@xmath14 spin - up and spin - down wave function has considerable influence in the magnetism observed at the zn @xmath12-edge .
the zn 3@xmath14 orbitals are completely filled ( 3@xmath98 ) and henceforth , a magnetic moment due to an overlap between the 3@xmath14 and 4@xmath32 orbitals in the zn ion is not expected to occur due to the filled 3@xmath14 orbitals being more contracted .
therefore , the magnetic signal observed in the 4@xmath13 states is due to hybridization with the extended gd 5@xmath14 orbitals and not from the exchange interaction with the zn 3@xmath98 orbitals.@xcite following hund s rule , the gd ion has the 4@xmath21 state filled by seven spin - up electrons in which it pulls the 5@xmath14 sub band spin - up function towards the inner core due to a positive exchange interaction .
the short distances between the first gd - zn ions ( @xmath24 3 ) drives a small hybridization between the broad zn 4@xmath13 and gd 5@xmath14 states , inducing a small amount of magnetic moment in 4@xmath13 states .
theoretical works have suggested that the orbital moment should be almost zero for the 5@xmath14 band ( @xmath99 ) , i.e. , a quenching of the angular momentum , so the 5@xmath100 and 5@xmath101 sub - bands should display the same polarization and thereby the dichroism at the @xmath11 and @xmath10 edges should have equal magnetic intensity.@xcite for the gdco@xmath3zn@xmath1 compound , the br ratio vary between 0.9 - 1.2 , i.e. , close to the theoretical value and therefore we can not suggest any orbital moment for this compound .
however , as showed in fig .
[ fig : fig1 ] for the fm compound , the intensity recorder at the @xmath10 is higher than the @xmath11 edge , where we observe a @xmath51 ratio of approximately @xmath52.@xcite this slightly different value from the theoretical branching ratio expected for xmcd ( br @xmath102 ) suggests that the gd ions may carry a small orbital moment at the 5@xmath14 orbitals in the fm compound .
mssbauer spectroscopy measurements@xcite and simulations@xcite for the dyfe@xmath3zn@xmath1 compound reported that the fe ions align afm with the dy magnetic moments in which the iron ions exhibit a very small magnetic moment @xmath24 @xmath103/fe .
neutron diffraction measurements@xcite performed in tbfe@xmath3zn@xmath1 compound also reported that a small magnetic moment at the fe ions ( @xmath104 ) would improve the refinement .
in addition , recently mssbauer measurement@xcite in gdfe@xmath3zn@xmath1 also reported the presence of a small magnetic contribution at the fe site . in order to verify the quality of our xmcd data at the fe @xmath12 edge , dichroic measurements at low temperature and under magnetic field in a 5 @xmath54 m fe - foil
was also carried out .
a magnetic signal around @xmath105 which , according to the literature@xcite , corresponds to a magnetic moment around @xmath106/fe was observed .
assuming that fe spins order ferromagnetically in gdfe@xmath3zn@xmath1 with a magnetic signal below 0.07 @xmath8 ( noise level ) , it would result in a magnetic moment lower than @xmath107 , i.e. , the same order of magnitude as reported by band structure calculation@xcite and comparable with the dyfe@xmath3zn@xmath1 compound .
therefore , if the fe ions carry magnetic contribution in the gdfe@xmath3zn@xmath1 materials , we can suggest that this magnetic moment must be lower than @xmath108 .
one possible explanation for the difficulty in identifying the magnetic signal at the fe @xmath12-edge ( 7112 ev ) would be due to contamination coming from the gd @xmath11 edge ( 7243 ev ) . despite the fact that the two edges are apart by around 130 ev , the gd @xmath11 pre - edge increases the background around the fe @xmath12-edge which , as a consequence may hide the small magnetic signal . an afm ordering state for the fe ions can not be ruled out , but is less likely to occur
. field dependent magnetization measurements@xcite show that the gdfe@xmath3zn@xmath1 compound reaches a saturation magnetization of 6.5 @xmath109/f.u . at low temperature .
assuming that the gd ions contribute with 7.94 @xmath109/gd for the total magnetic moment ( theoretical value ) , and that the transition metals are coupling antiferromagnetically with the rare earth elements , the transition metals are found to carry a magnetic moment around 1.44 @xmath109/f.u .
opposite to the gd ions .
supposing that the fe ions have a total magnetic moment @xmath107 , consequently the zn ions present in this material would present an induced magnetic moment of approximately 0.05 @xmath109/zn .
therefore , it strongly suggests that the interaction between the rare earth ions affects the environment around the atoms and consequently spin polarizes the transition metal ion .
the rough estimate of magnetic moment for the zn ions based on spectroscopy and macroscopic measurements has to be further investigate by density functional theory calculation . to provide further information about these systems ,
the magnetic properties of the afm compound gdco@xmath3zn@xmath1 were also investigated , using the xrms technique rather than neutron scattering , due to the high neutron absorption cross - section of gd ions .
the transition to the magnetically ordered phase driven by temperature is characterized by the appearance of superlattice magnetic reflections with a magnetic propagation vector @xmath110 .
this @xmath111 magnetic vector indicates that the magnetic unit cell is represented by a doubled chemical unit cell in all three crystallographic directions . as displayed in fig .
[ fig : fig5 ] , the gd spins follow the magnetic representation @xmath89 which is different when compared with the isostructural tbco@xmath3zn@xmath1 compound.@xcite since there is a large separation ( @xmath24 6 ) between the rare earth ions in this system , the magnetic properties will be mainly mediated via the conduction electrons .
the large distance between the first @xmath16-@xmath16 ions explains quite well a weakening of the @xmath112 exchange interaction and thus a very low representative magnetic transition temperature , i.e. , approximately 5.7 k ( gdco@xmath3zn@xmath1 ) and 2.5 k ( tbco@xmath3zn@xmath1 ) . as a consequence of a poor @xmath112 coupling , the surrounding matrix or the zn cages around the @xmath16 ions
are weakly affected and hence it is not possible to induce magnetic moment in the transition metal ions .
the absence of magnetism in the transition metal ions are also observed in the tbco@xmath3zn@xmath1 compound , in which the best refinement of the magnetic structure was obtained with null magnetic moment in the co and zn ions .
although gdco@xmath3zn@xmath1 and tbco@xmath3zn@xmath1 compounds display the same magnetic propagation vector @xmath5 , their magnetic structures are different , which is mainly related to the competition between the rkky and cef interactions . for the tb - based materials the crystalline electric field splitting can affect the @xmath112 constant and therefore influence the magnetic coupling and the temperature transition .
it is well known that an energy level cef splitting induces magnetic anisotropies in the ground state and it may influence the total angular momentum .
several tb - based intermetallics have exhibited distinct magnetic properties in relation to the gd counterparts as shown in the layered family @xmath113 ( @xmath16 = rare earth ; @xmath2 and @xmath114 = transition metal ; @xmath115 = 1,2 and @xmath116 = 0,1 ) .
@xcite although gdco@xmath3zn@xmath1 and tbco@xmath3zn@xmath1 compounds display the same magnetic propagation vector @xmath5 , their magnetic structures are different , which is mainly related to the competition between the rkky and cef interactions .
the modification in the magnetic structure for different rare earth ions but with the same propagation vector has already been extensively investigated and it is mainly related to the rkky and cef interactions .
et al_. [ ] evaluated the cef parameters from the thermodynamic measurements for the entirely @xmath16co@xmath3zn@xmath1 series ( @xmath16 = tb - tm ) and they observe small energy scales ( small @xmath117 parameter ) and a large @xmath118 cef parameter for the complete series .
this finding suggests a small energy level splitting and a strong influence of the zn cage on the rare earth ions , i.e. , guest - framework interaction .
therefore , we can suggest that the rare earth ions located in this large polarized environment are strongly affected by the zn cages , which has direct influence in the electronic and magnetic properties .
this can be seen in the different magnetic structures for the afm compounds and the spin polarization of the zn ions only for the gdfe@xmath3zn@xmath1 compound .
hence , the cef effect has an important role in this class of compound . a detailed investigation for different rare earth elements would allow a better understanding of the @xmath16co@xmath3zn@xmath1 family .
nevertheless macroscopic measurements up to 1.8 k only report magnetic ordering for the compounds with @xmath16 = gd and tb .
we have investigated the intermetallic gd@xmath0zn@xmath1 system with @xmath119 co and fe at low temperature using the xrms and xanes / xmcd techniques , respectively .
the xrms measurements performed in gdco@xmath3zn@xmath1 compound reveal a commensurate antiferromagnetic ordering with a magnetic propagation vector @xmath5 in which only the gd ions carry magnetic moments .
selected magnetic reflections were measured in the polarization channel @xmath48 and we identified that the gd spins follow the magnetic representation @xmath7 , which is different from the isostructural compound tbco@xmath3zn@xmath1 , mainly due to the cef effects in the latter .
the evolution of magnetic signal showed a magnetic phase transition below t@xmath120 = 5.72(6 ) k with a critical exponent @xmath58 = 0.36(3 ) , suggesting a three - dimensional ( 3d ) heisenberg magnetic model .
the xanes and xmcd measurements performed at the gd @xmath33 edges in gdfe@xmath3zn@xmath1 compound reveal a strong magnetic signal ( @xmath24 12.5 @xmath8 - @xmath10 and 9.7 @xmath8 - @xmath11 ) indicating a splitting of the 5@xmath14 orbitals and a strong gd - gd exchange interaction as well as a non zero orbital moment .
in addition , we observed a presence of a small magnetic dichroic signal at the zn @xmath12 edge due to the spin polarization of the gd 5@xmath14 orbitals .
this indicates a large rkky exchange interaction between the gd - gd ions which polarizes the surrounding matrix .
this work was supported by fapesp ( sp - brazil ) under contracts no . 2009/10264 - 0 , 2011/19924 - 2 , 2011/24166 - 0 , 2012/10675 - 2 and 2012/17562 - 9 .
parts of this research were carried out at the light source petra iii at desy , a member of the helmholtz association ( hgf ) .
work at argonne is supported by the u.s .
department of energy , office of science , office of basic energy sciences , under contract no .
de - ac-02 - 06ch11357 .
we would like to thank r. d. reis and c. s.b .
dias for their assistance in macroscopic measurements and j. w. kim for his assistance at beamline 6-id - b .
we are indebted to d. reuther , r. kirchhof and h .- c .
wille for their assistance with the electromagnet at p09/desy .
s. nandi , y. su , y. xiao , s. price , x. f. wang , x. h. chen , j. herrero - martn , c. mazzoli , h. c. walker , l. paolasini , s. francoual , d. k. shukla , j. strempfer , t. chatterji , c. m. n. kumar , r. mittal , h. m. rnnow , ch .
regg , d. f. mcmorrow , and th .
brckel , phys .
b * 84 * , 054419 ( 2011 ) . j.d .
thompson , r. movshovich , z. fisk , f. bouquet , n.j .
curro , r.a .
fisher , p.c .
hammel , h. hegger , m.f .
hundley , m. jaime , p.g .
pagliuso , c. petrovic , n.e .
phillips , j.l .
sarrao j. magn .
magn . mater .
* 226230 * , 5 ( 2001 ) . | we investigate the magnetic and electronic properties of the gd@xmath0zn@xmath1 ( @xmath2 = fe and co ) compounds using x - ray resonant magnetic scattering ( xrms ) , x - ray absorption near - edge structure ( xanes ) and x - ray magnetic circular dichroism ( xmcd ) techniques .
the xrms measurements reveal that the gdco@xmath3zn@xmath1 compound has a commensurate antiferromagnetic spin structure with a magnetic propagation vector @xmath4 = @xmath5 below the nel temperature ( @xmath6 5.7 k ) .
only the gd ions carry a magnetic moment forming an antiferromagnetic structure with magnetic representation @xmath7 . for the ferromagnetic gdfe@xmath3zn@xmath1 compound ,
an extensive investigation was performed at low temperature and under magnetic field using xanes and xmcd techniques .
a strong xmcd signal of about 12.5 @xmath8 and 9.7 @xmath8 is observed below the curie temperature ( @xmath9 85 k ) at the gd-@xmath10 and @xmath11 edges , respectively .
in addition , a small magnetic signal of about 0.06 @xmath8 of the jump is recorded at the zn @xmath12-edge suggesting that the zn 4@xmath13 states are spin polarized by the gd 5@xmath14 extended orbitals . |
the delayed onset of solar cycle 24 , after one of the deepest solar minimum experienced in the past 100 years has had significant solar and heliospheric consequences .
solar cycle 23 has been characterized by a steady decline in solar activity @xcite , a continuous weakening of polar fields @xcite and a decline in micro - turbulence levels in the inner heliosphere since @xmath21995 @xcite .
investigations of the boundary of polar coronal holes during the declining phase of solar cycle 23 using images from the extreme ultraviolet imaging telescope ( eit ) on board the solar heliospheric observatory ( soho ) , have found a decrease in coronal hole area by @xmath215@xmath3 in comparison to that at the beginning of solar cycle 23 @xcite . using both ground - based and space - borne observations of photospheric magnetic fields @xcite and theoretical modeling @xcite ,
there have been continuous efforts to investigate and understand the behavior of solar photospheric fields and their correlation with meridional flows , to try and explain the delayed onset of cycle 24 and the cause of the deep minimum at the end of cycle 23 .
magnetic field measurements using data from the national solar observatory , kitt peak ( nso / kp ) synoptic magnetogram database , have shown that a decline in solar photospheric fields in the latitude range 45@xmath1 to 78@xmath1 began around the minimum of solar cycle 22 , in @xmath21995@xmath41996 @xcite .
however , the dipole field of the sun behaved differently .
it was at its strongest in 1995 , weakened at solar maximum around 2000 , and then increased between @xmath22000 and 2004 @xcite .
signatures of this decline in solar fields above latitude @xmath045@xmath1 have also been observed in the inner heliosphere as a corresponding decline in micro - turbulence levels which in turn , tie into small scale interplanetary magnetic fields @xcite .
the decline in micro - turbulence levels was inferred from extensive interplanetary scintillation ( ips ) observations at 327 mhz in the period 1983@xmath42009 @xcite .
using this fact _ viz . _
that both solar magnetic fields and micro - turbulence levels in the inner - heliosphere have been declining since mid 1990 s , @xcite have argued that the build - up to the deep and extended solar minimum at the end of cycle 23 was actually initiated as early as the mid-1990 s
. one method of gaining insights into underlying basic processes in the solar interior , that govern the nature and evolution of solar photospheric magnetic fields , is by studying periodicities produced by various surface activity features at different times in the solar cycle .
for example , the 158 days ( d ) rieger periodicity reported in the solar flare occurrence rates @xcite has been linked to the periodic emergence of magnetic flux through the photosphere which in turn gives rise to a periodic variation of the total sunspot area on the solar surface .
similarly , a strong 1.3 years ( yr ) periodicity detected at the base of the solar convection zone , through a helioseismic study @xcite , has also been detected in sunspot areas and sunspot number time series studied using wavelet transforms @xcite .
these authors have in fact proposed that the 154@xmath5158 d rieger period is a harmonic of the 1.3 yr periodicity ( 3@xmath6156 d = 1.28 yr ) and that variations in the rotation rate have a strong influence on the workings of the solar dynamo @xcite . in this paper , in order to understand the role of periodic changes , if any , in the solar photospheric fields leading to the build - up of the recent deep minimum at the end of cycle 23 , data from the nso / kp synoptic magnetic database during the period 1975.14 @xmath4 2009.86 were subjected to detailed spectral analyses using both wavelet and fourier techniques .
it has been shown that a significant north - south asymmetry in the quasi - periodic variations of the photospheric magnetic fields exist both prior to and after 1996 ( with 1996 being a clear transition period ) almost 12 yr before the deep minimum period in 2008@xmath42009 .
prominent periodicities related to the solar cycle , apart from the well known synodic rotation period of @xmath227 d and the sunspot activity cycle of @xmath211 yr , have been a topic of interest for over two decades now .
using data from the gamma ray spectrometer @xcite onboard the solar maximum mission , a clear 154 d periodicity in the occurrence rate of high energy ( 0.3 - 100 mev ) flares has been reported @xcite .
this 154 d periodicity in the solar flare occurrence was also confirmed at other wavelengths @xcite , and for solar energetic particle events @xcite .
in addition , other intermediate periodicities with periods ( @xmath71 yr ) have also been reported _
viz_. a 51 d periodicity in the comprehensive flare index @xcite , a 127 d periodicity in the 10 cm radio flux @xcite , periodicities of 33.5 d , 51 d , 84 d , 129 d and 153 d in the solar flare occurrence during different phases of solar cycles 19@xmath423 @xcite , periodicities of 100@xmath4103 d , 124@xmath4129 d , 151@xmath4158 d , 177@xmath02 d , 209@xmath4222 d , 232@xmath4249 d , 282@xmath04 d , 307@xmath4349 d in the unsigned photospheric fluxes @xcite and periodicities of 87 - 106 d , 159 - 175 d , 194 - 219 d , 292@xmath4318 d , and 69@xmath495 d , 113@xmath4133 d , 160@xmath4187 d , 245@xmath4321 d , 348@xmath4406 d in sunspot areas at different phases of cycle 22 and 23 respectively @xcite . for periodicities , between 1@xmath45 yr , a 1.3 yr periodicity in sunspot areas and numbers @xcite , 1 yr and 1.7 yr periodicity in solar total and open fluxes @xcite , 1.3 yr , 1.43 yr , 1.5 yr , 1.8 yr , 2.4 yr , 2.6 yr , 3.5 yr , 3.6 yr and 4.1 yr in unsigned photospheric fluxes @xcite have been reported .
studies of all these periodicities have shown the existence of an asymmetry in solar activity phenomena between the northern and southern hemispheres of the sun .
this hemispheric asymmetry , on different time scales and at different phases of different solar cycles , have been observed in various types of solar activity @xcite and have been providing invaluable information about basic underlying physical processes on the sun . since processes like differential rotation of the sun and emergence of magnetic flux
determine the strength and distribution of solar magnetic activity , investigation of solar cycle related periodic or quasi - periodic variations is crucial in understanding the behavior and nature of solar magnetic fields .
figure [ f - high - lat ] shows temporal variation in the unsigned or absolute solar photospheric magnetic fields in the latitude range , 45@xmath1@xmath478@xmath1 in both solar hemispheres , for the period from february 1975 to november 2009 ( panel a ) covering solar cycles 21 , 22 , and 23 .
the vertically oriented dotted lines demarcate the period of solar minimum of these three solar cycles . the small , black filled dots and the open circles are the actual measurements for the northern and southern hemispheres respectively while the solid red ( northern hemisphere ) and blue ( southern hemisphere ) lines are smoothed curves derived using a robust savitzky - golay ( sg ) algorithm @xcite that has the ability to preserve features of the input distribution like maxima and minima while effectively suppressing noise .
such features are generally flattened by other smoothing techniques . the decline in photospheric magnetic fields in both hemispheres , starting around the early to mid-1990 s
can be clearly seen in fig .
[ f - high - lat ] .
while the decline in photospheric fields had started in @xmath21991 for the northern hemisphere ( solid red curve ) , for the southern hemisphere the decline started in @xmath21995 ( solid blue curve ) .
beyond @xmath21996 , a steady decline in the photospheric fields for both hemispheres can be clearly seen . a similar decline in the sunspot umbral magnetic field strength since @xmath21998
has also been reported @xcite .
also shown in the bottom two panels of fig .
[ f - high - lat ] are the magnetic residuals , obtained by subtracting the sg fit from the actual measurements , for both the hemispheres in the latitude range 45@xmath1@xmath478@xmath1 ( panel b ) and 0@xmath1@xmath445@xmath1 ( panel c ) respectively .
see the upper two panels of fig . 1 in @xcite for measurements of the magnetic field in the range @xmath045@xmath1 which show a strong solar cycle modulation in the data as expected . to 78@xmath1 .
the filled dots and the open circles respectively represent the actual measurements for the northern and southern hemispheres while the fit obtained using the sg algorithm is shown by the solid red ( north ) and blue ( south ) lines .
the vertically oriented dotted lines demarcate the periods of solar cycles 21 , 22 and 23 .
b ) the residuals obtained after subtracting the sg smoothed values from the magnetic fields in the latitude range , 45@xmath1 to 78@xmath1 , are shown for the north and south .
c ) show the residual fields for both the northern and southern hemispheres obtained from the magnetic fields at latitudes below 45@xmath1.,width=604,height=604 ] brief details of the method by which the magnetic fields were derived are given below .
we have used measurements of magnetic field strengths obtained from 466 individual carrington rotations ( cr ) of nso / kp synoptic magnetograms from cr1625 through cr2090 in the period from 19 february 1975 to 09 november 2009 ( [email protected] ) .
each magnetogram , generated from daily ground - based full - disk magnetograms spanning over a carrington rotation period corresponding to 27.27 days , was longitudinally averaged to a 1@xmath1 wide longitudinal strip covering the latitude range from -90@xmath1 to + 90@xmath1 .
surface magnetic fields in both equatorial ( -45@xmath1 to + 45@xmath1 ) and high - latitude zones ( @xmath845@xmath1 in both hemispheres ) of the sun were then derived by averaging over appropriate latitude regions .
full details of the method used are described in @xcite .
the sg smoothing is done in such a way as to make the resulting residual time series stationary _ i.e. _ the mean of time series is or approaches zero .
such a detrending of the data removes large periodic variations ( @xmath211 years ) .
the residuals were then subjected to both a wavelet analysis and a fourier time series analysis to study temporal periodic variations in photospheric magnetic fields . in the rest of the text the fields obtained in the equatorial belt of -45@xmath1 to
+ 45@xmath1 are referred to as low - latitude fields while the fields obtained at latitudes @xmath845@xmath1 in both hemispheres are referred to as high - latitude fields .
a wavelet transform is basically the convolution of a time series with the scaled and translated version of a chosen @xmath9 " wavelet function .
the wavelet analysis now finds frequent use in the analysis of time series data since it yields information in both time and frequency domains @xcite . using a morlet wavelet as the mother wavelet and based on the algorithm by @xcite ,
the magnetic residuals obtained for both the high - latitude and low - latitude fields in the period from 19 february 1975 to 09 november 2009 ( [email protected] ) were subjected to a wavelet analysis .
scattered data gaps amounting to @xmath2 2.5% of the data were replaced with values obtained using a cubic spline interpolation .
the morlet wavelet function @xmath10 , represented in equation [ eq1 ] below , is a plane wave modulated by a gaussian .
@xmath11 here , @xmath12 is a non - dimensional frequency that determines the number of oscillations in the wavelet and @xmath13 is a non - dimensional time parameter .
the choice of different non - dimensional frequencies defines the frequency resolution .
however , by varying the frequency resolution , one has to compromise with time resolution . in this case , we have chosen @xmath12 to be 6 for obtaining a better frequency resolution .
this value is left unchanged and is not tuned to different values as our interest is to look for changes in the frequency components rather than to find precisely the low and high frequency components . however , in the following sections , we have employed a fourier analysis technique to find the various frequency components .
figure [ f - wavelet - plots ] shows the wavelet spectrum for the high - latitude fields in the northern ( upper panel ) and southern hemisphere ( lower panel ) while figure [ f - wavelet - plots - tor ] shows the wavelet spectrum for the low - latitude fields in the northern ( upper panel ) and southern hemispheres ( lower panel ) .
the green cross - hatched regions in both figures represent the cone of influence ( coi ) where the spectral distribution is not reliable .
the white contours are drawn at a significance level of 95% .
the power levels are indicated by a color coded bar on the right of each panel and the dashed vertical line in each panel is drawn at 1996 in both figures [ f - wavelet - plots ] and [ f - wavelet - plots - tor ] .
for the high - latitude fields ( fig .
[ f - wavelet - plots ] ) , in both the north and south , significant wavelet power is only seen at lower periods ( @xmath21@xmath43 yr ) prior to 1985 while significant wavelet power is seen both at lower and higher periods ( @xmath23@xmath45 yr ) for the interval between 1985 and 1996 .
for the low - latitude fields ( fig .
[ f - wavelet - plots - tor ] ) , we do nt see significant changes in the power levels between 1975 and 1996 . to 78@xmath1 .
the green cross hatched regions are the cone of influence ( coi ) while the white contours indicate a significance level at 95% .
the dashed vertical lines are marked at 1996 when a clear transition has observed in wavelet power of periodicities for both the north and south high - latitude fields.,width=453,height=453 ] it is important to note here that there is a clear transition in both periodicities and power levels in both hemispheres around 1996 for both high - latitude and low - latitude fields .
we stress here that we have taken care to check that the selection of different colors or changes in the color intensity in the wavelet plots makes no difference in the transition noticed in the wavelet power spectra . though the decline in the fields has started in the early to mid 1990 s , as seen from fig .
[ f - high - lat ] , we see from the wavelet spectra that there is an unambiguous transition in the quasi - periodic variations of fields in both the hemispheres at @xmath21996 .
we have therefore chosen , in the rest of the analysis , to divide the residuals into two sets based on this transition seen in the wavelet spectra around @xmath21996 in order to study the changes , if any , in the periodicities before and after the transition seen in the wavelet spectra .
a fourier time series analysis was then carried out separately on the residuals used in the wavelet analysis for the period prior to 1996 and the period after 1996 for both high - latitude and low - latitude fields .
it is important to bear in mind here that the transition seen in the wavelet spectra around 1995@xmath41996 corresponds to the time when both solar high - latitude fields above @xmath045@xmath1 and solar wind turbulence levels in the entire inner - heliosphere began declining @xcite . to 45@xmath1 .
the green cross hatched regions are the cone of influence ( coi ) while the white contours indicate a significance level at 95% .
the dashed vertical line demarcates the period of transition in wavelet power of periodicities for both the hemispheres around 1996.,width=453,height=453 ]
based on the transition seen in the wavelet spectra at @xmath21996 , we have subjected the residuals obtained from the sg fits to a fourier analysis after dividing the time series of the residuals , spanning years 1975.14 to 2009.86 , into two parts , one prior to 1996 and the other after 1996 . since the time series of the residual field has some data gaps amounting to @xmath22.5% of the data , the algorithm used for this purpose was based on the lomb - scargle fourier transform for unevenly spaced data @xcite in combination with the welch - overlapped - segment - averaging ( wosa ) procedure @xcite . the spectral power derived from such
an analysis is generally normalized with respect to the total power contained in all the fourier components taken together .
this procedure makes the relative distribution of the spectral power independent of the spectral windowing used in the algorithm .
such normalized power spectra are referred to as normalized fourier periodograms or simply normalized periodograms .
normalized periodograms obtained from the fourier analysis are shown in figure [ f - asym - fields ] .
the top four panels of fig .
[ f - asym - fields ] labeled a , b , c and d represent the high - latitude fields while the bottom four panels of fig . [
f - asym - fields ] labeled e , f , g , and h represent the low - latitude fields . the upper four panels ( high - latitude fields ) show normalized periodograms before and after 1996 in the north ( panels a and b respectively ) , and before and after 1996 in the south ( panels c and d respectively ) . in a similar fashion , the lower four panels ( low - latitude fields ) show normalized periodograms in the north before and after 1996 ( panels e and f respectively ) and in the south before and after 1996 ( panels g and h respectively ) .
the @xmath14 " periodic components were determined using the siegel test statistics @xcite and are those components having power levels above the black horizontal line drawn in each panel of fig .
[ f - asym - fields ] .
the siegel s test is an extension of the fisher s test @xcite . while fisher s test attempts to spot the single dominant periodicity in a time series that has maximum power in the periodogram and is above the @xmath15
" level defined by fisher s test statistics , siegel s test @xcite relaxes this stringent condition a little and considers two to three dominant periodic components above the @xmath15 " level .
both the aforementioned test statistics and the process of determining the confidence / significance level are discussed in @xcite and also briefly in @xcite . in the present analysis ,
95% confidence level ( or 5% significance level ) is chosen .
this is equivalent to @xmath02@xmath16 level in fft method .
it is to be noted that the choice of this confidence level decides the
@xmath15 " level of the sigel test statistics @xcite .
it may be noted that though the fourier components obtained from the analysis range in periods from 54 d to 12 yr corresponding to 214 nhz to 2.6 nhz in frequency , fig .
[ f - asym - fields ] only shows those components with periods lying in the range 300 d to 12 yr @xmath17 the low - frequency range .
normalized periodograms containing fourier components with periods less than 300 d will be discussed later .
the red vertical lines in each panel of fig . [ f - asym - fields ] are drawn through the peak of the component with the longest period .
a shift in the periodicities can be seen after 1996 when compared to that before 1996 .
this shift is indicated by vertical dashed blue lines and a black arrow showing the direction of the shift in periodicities .
it must be emphasized that we are not interested in the actual periodicities themselves but rather in the manner in which the longest periodicities shift between the time interval prior to and after 1996 .
it can be seen from fig .
[ f - asym - fields ] that in addition to the shift in periodicity of the high - latitude fields in both hemispheres before and after 1996 the spectral power also differs significantly before and after 1996
. the longer periods after 1996 have more spectral power than longer periods prior to 1996 . if we denote the longest periods for the high - latitude fields in the north before 1996 as @xmath18 and that after 1996 as @xmath19 , we find that @xmath19 @xmath8 @xmath18 for high - latitude fields in the north . for the southern hemisphere , the direction of the shift in periodicities , as indicated by the direction of the black arrow in fig .
[ f - asym - fields ] , is in the opposite direction as compared to the northern hemisphere .
if we denote the longest periods for the high - latitude fields in the south before 1996 and after 1996 as @xmath20 and @xmath21 respectively , we find that @xmath21 @xmath7 @xmath20 for the high - latitude fields in the south . as in the case of the northern hemisphere ,
the distribution of spectral power in various periodicities is also significantly different in the periods before and after 1996 .
thus , a north - south asymmetry is seen in the distribution of periodicities for the high - latitude fields along with significant changes in the power of various periodic components .
since the normalized periodograms are generated from measurements of magnetic fields , it is a proxy for the behaviour of the photospheric magnetic fields and a north - south asymmetry would imply a similar asymmetry in photospheric magnetic fields towards higher solar latitudes before and after 1996 .
in contrast to the behavior of the high - latitude fields , the low - latitude fields represented in the bottom four panels of fig .
[ f - asym - fields ] and labeled e , f , g and h show no asymmetry in the manner in which the longest periods shift when one compares the periodograms prior to and after 1996 . the shift in the longest periods in both the north and south low - latitude fields before and after 1996 has been depicted in the lower four panels of fig .
[ f - asym - fields ] in a similar fashion to that of the upper four panels .
the vertical solid red line is drawn through the peak of the component with the longest period and the dotted blue line and a black arrow in each panel indicate the direction of the shift . denoting the longest period for the low - latitude fields in the north before and after 1996 as @xmath22 and @xmath23 respectively
, we find that @xmath22 @xmath8 @xmath23 . in the southern hemisphere , denoting the longest period for the low - latitude fields prior to 1996 and after 1996 as @xmath24 and @xmath25 respectively ,
it is clear that @xmath24 @xmath8 @xmath25 .
hence , the longest periods show a shift in the same direction indicating no asymmetry in the distribution of the low - latitude fields before and after 1996 .
to examine how the fourier spectrum , for the high - latitude fields , varies as a function of latitude , the residuals for the high - latitude fields were further subdivided into 15@xmath1 wide latitude bins and a fourier analysis was carried out for the data in each latitude bin .
this was done to examine the changes , if any , in the spectral distribution with latitude and because the strength and distribution of photospheric magnetic fields change with solar latitudes .
it would therefore be interesting to see if the north - south asymmetry in the high - latitude fields persists across all latitude bins .
figure [ f - prof - high ] shows normalized periodograms in the latitude range 30@xmath1 to 75@xmath1 in latitude steps of 15@xmath1 .
the top four panels of fig .
[ f - prof - high ] labeled a , b , c , d , the middle four panels labeled e , f , g , h , and the bottom four panels labeled i , j , k , l represent respectively , normalized periodograms showing the distribution of fourier periodicities in the latitude ranges 30@xmath1@xmath445@xmath1 , 45@xmath1@xmath460@xmath1 and 60@xmath1@xmath475@xmath1 . the latitude ranges are indicated at the top right hand corner of each panel .
starting from the top , each of the left hand panels of fig.[f - prof - high ] represent , in pairs before 1996 and after 1996 respectively , the power spectra for the northern hemisphere in each latitude bin .
similarly , the right hand panels of fig.[f - prof - high ] represent , in pairs before 1996 and after 1996 respectively , the power spectra obtained for each latitude bin in the southern hemisphere . in a manner similar to that shown in fig .
[ f - asym - fields ] , the red vertical lines in each panel of fig .
[ f - prof - high ] are drawn through the peak of the component with the longest period and the dotted blue line and a black arrow indicate the direction in which the longest periodicities shift .
it can be seen from fig.[f - prof - high ] that the north - south asymmetry is present only in the 45@xmath1@xmath460@xmath1bin . though not shown in fig .
[ f - prof - high ] we have verified that the asymmetry is also not there in the latitude bins 0@xmath1 to 15@xmath1 and 15@xmath1 to 30@xmath1 . in the 60@xmath1@xmath475@xmath1 bin ( bottom four panels of the fig [ f - prof - high ] ) , the shift of the longest periods before and after 1996 shows no north - south asymmetry . however , the spectral powers in the longest periods are significantly different before and after 1996 .
the spectral power in the longest period component increases significantly after 1996 for the northern hemisphere while it decreases for the southern hemisphere .
it is important to note here that the latitude band 45@xmath1@xmath460@xmath1 in both hemispheres is dominated by surges or tongues of magnetic flux being carried polewards by the meridional flow .
these surges are a direct surface manifestation of the meridional flow which is in turn governed by an internal solar dynamo . beyond 60@xmath1 in latitude , the high - latitude flux in both hemispheres saturates and can be best seen along with the magnetic surges in the 45@xmath1@xmath460@xmath1 latitude band in a magnetic butterfly diagram which will be described in the following section .
it is known that the emergence and evolution of magnetic field on the solar photosphere is tied in to a solar dynamo operating in the solar interior .
in particular , the toroidal field manifests itself as solar surface bipolar active regions with sunspots migrating toward the equator as the solar cycle progresses .
it is this migration which gives rise to the well known _ butterfly _ " diagram which is basically a map of longitudinally averaged magnetic fields in time .
solar dynamo models that explain the main features of solar magnetic activity consider the strong toroidal field as being generated by the shearing of a pre - existing weak poloidal field by solar differential rotation ( @xmath26 effect ) .
subsequently , the regeneration and inversion of polar fields , at the maximum of each solar cycle , is caused by the cancellation of sunspot fields at the equator along with a net poleward surface flux that is transported via meridional circulation to reach the poles and reverse the pre - existing polar fields ( babcock - leighton type @xmath27-effect ) .
@xmath445@xmath1 , 45@xmath1@xmath460@xmath1 , and 60@xmath1@xmath475@xmath1 .
the four panels ( the top , mid and bottom ) in the figure correspond to the fourier spectral power for magnetic fields in the north before 1996 , for magnetic fields in the north after 1996 , for magnetic fields in the south before 1996 , and for magnetic field in the south after 1996 .
the solid red vertical line , the dotted blue vertical line and the black arrow in each panel are used to show the shift in periodicity and the spectral power of the fields before and after 1996 both in the northern and southern hemisphere .
the solid horizontal lines depict significance levels as determined by siegel test.,width=529,height=604 ] we have generated a butterfly diagram from the nso / kitt - peak synoptic maps by using longitudinally averaged photospheric fields derived from the synoptic maps described in section [ s - mag - field ] .
figure [ f - butterfly ] shows a butterfly diagram for solar cycles 21 , 22 and 23 covering the period from 19 february 1975 to 09 november 2009 ( 1975.14 @xmath4 2009.86 ) .
the leading and following polarity fluxes in panel - a of fig .
[ f - butterfly ] are shown in red and green respectively for the northern hemisphere in cycle 21 , with the derived flux values indicated in a colour coded bar at the top of the figure .
it must be noted that leading and following polarity fluxes will be reversed in the southern hemisphere and will again flip from cycle to cycle and hemisphere to hemisphere . in order to obtain better contrast , the fluxes in fig .
[ f - butterfly]a have been limited to the range @xmath28 g .
the poleward motion of the trailing polarity fluxes , above the latitudes @xmath2945@xmath1 , known as polar surges are episodic in nature @xcite and can be clearly seen as tongues of red and green bands in the upper panel of fig .
[ f - butterfly ] labeled - a. for a better view of the polar surges , the middle and lower panels of fig .
[ f - butterfly ] show variations of intensity as a function of time in each 1@xmath1 strip of latitude in the latitude range 45@xmath1@xmath460@xmath1 for the northern hemisphere ( middle panel - b ) and southern hemisphere ( lower panel - c ) .
the solar field for each degree of latitude has been shown after smoothing over three carrington rotations .
an example of a surge in the north and south has been highlighted in red .
@xcite reported a periodicity of 1.3 yr in the photospheric fields that has been correlated with these polar surge motions .
we see an approximate periodicity of 1 yr in the polar surge flows in the lower two panels of fig .
[ f - butterfly ] .
this agrees with the strong 1 yr periodicity seen in fig .
[ f - asym - fields ] and [ f - prof - high ] .
further , a variation in the strength and occurrence rate of surges prior to and after 1996 can be seen from a careful inspection of fig .
[ f - butterfly]a , b and c. in the latitude bin @xmath2945@xmath1 @xmath4 @xmath2960@xmath1 , we find that the frequency of these surges are comparatively more during the years 1996@xmath42009 while it is less during the years 1986@xmath41995 .
the harmonic analysis yielded fourier components with periods ranging from 54 d to 12 yr corresponding to a frequency range of 2.6 nhz to 214 nhz .
the significant periodic components , as determined by siegel test , were grouped into low frequency periods in the frequency range , 38.5 nhz ( 300 d ) to 2.6 nhz ( 12 yr ) and high frequency periods in the frequency range , 214 nhz ( 54 d ) to 38.5 nhz ( 300 d ) .
these periods have been listed in table [ t - tab - low ] ( long periods ) and table [ t - tab - high ] ( short periods ) respectively .
many of these periodicities have already been reported by other workers and a few new periodicities have also been seen in our analysis .
these new periodicities have been listed in bold face in tables [ t - tab - low ] and [ t - tab - high ] and boxed in blue in figures [ f - asym - fields ] and [ f - asym - fields - high ] . in the low frequency range ,
38.5 nhz ( 300 d ) to 2.6 nhz ( 12 yr ) , we restrict our analysis to periodicities in the range , 305 d to 5.6 yr .
periodicities greater than 5.6 yr are not taken into account because of their inaccurate determination as a consequence of division of time series .
the periodicities are listed in table [ t - tab - low ] where the upper half lists periodicities for high - latitude fields in the north and south grouped before and after 1996 while the lower half lists periodicities for low - latitude fields in a similar fashion . for photospheric fields in the high - latitude range
we find periodicities of 305 d ( 0.84 yr ) , 325 d ( 0.89 yr ) , 339 d ( 0.92yr ) , 341 d ( 0.93 yr ) , 1 yr , 1.2 yr , 1.4 yr , 1.5 yr , 2 yr , 2.4 yr , 2.6 yr , and 4 yr while for photospheric fields in the low - latitude range we find periods of 312 d ( 0.86 yr ) , 318 d ( 0.87 yr ) , 321 d ( 0.88 yr ) , 327 d ( 0.9 yr ) , and 343 d ( 0.94 yr ) , 1 yr , 1.27 yr , 1.3 yr , 1.44 yr , 1.5 yr , 1.8 yr , 2.6 yr , 3.5 yr , and 4 yr . all of these quasi - periodicities in various types of solar activity have been reported earlier by other researchers @xcite .
@xcite had discussed a north - south asymmetry in quasi - periodic variations in unsigned photospheric flux with periods of 1.3yr , 1.5 yr , and
2.6 yr . 45@xmath1 are seen as tongues of red and green bands .
the lower two panels labeled - b and c show the variation of magnetic field , after smoothing over three carrington rotations , for each degree of latitude in the range 45@xmath1 to 60@xmath1 .
an example of a surge is highlighted in red in the bottom two panels representing the northern and southern hemisphere respectively.,width=453,height=453 ] in addition to these known periodicities , we have found some new periodicities in the high - latitude fields with periods 2.3 yr , 2.9 yr , 3.1 yr , and 5.6 yr and in low - latitude fields with periods 1.1 yr , 1.9 yr , 2.5 yr , and 3.1yr .
these new periodicities have been shown in boldface in tables [ t - tab - low ] and [ t - tab - high ] .
though the periodicity of 1 yr has already been reported , we find this periodicity to be the prominent periodicity observed in both north and south high - latitude fields before and after 1996 .
the fourier power in the 1 yr period is comparatively more both in the north and south prior to 1996 than after 1996 . on the other hand , though the 1 yr period is present in the low - latitude fields both in the north and south , it is only present after 1996 . besides these , periods
[email protected] yr is common to both the north and south low - latitude fields and have significant power . among these periodicities ,
the most discussed and reported periodicity is the 1.3 yr periodicity @xcite which has been linked to polar surges that can be seen most clearly in the latitude band 45@xmath1 to 60@xmath1 @xcite in magnetic butterfly diagrams as depicted in fig .
[ f - butterfly ] .
also , the 1.3 yr periodicity has been linked to the variation of the rotation rate near the base of the convection zone @xcite . in the high frequency range
we have frequencies ranging from 214 nhz ( 54 d ) to 38.5 nhz ( 300 d ) .
these periodicities for both the high - latitude and low - latitude fields are listed in the table [ t - tab - high ] .
the upper half of the table [ t - tab - high ] lists periodicities for the high - latitude fields with periods ranging between 157 d to 294 d. few of these periodicities with their fourier spectral power are shown in the top four panels - a , b , c , d in fig [ f - asym - fields - high ] .
similarly , the lower half of the table [ t - tab - high ] lists significant periodicities for the low - latitude fields with periods ranging in between 134 d to 288 d. the bottom four panels - e , f , g , h in fig [ f - asym - fields - high ] show a few of these periodicities with their spectral power .
as stated earlier the significant periodic components were determined using the siegel test statistics and are those periodic components having power levels above the black horizontal line drawn in each panel of fig .
[ f - asym - fields - high ] .
the known periodicities in the high - latitude zone include those with periods 157 d , 183 d , 184 d , 240 d , 263 d , 278 d , and 288 d and the new periodicities @xmath30 with period 253 d and 294 d , not reported in the literature earlier , are shown in boldface in the upper part of the table [ t - tab - high ] .
the rieger periodicity of 157 d is a well known fundamental periodicity that we find only in the high - latitude fields in the northern hemisphere prior to 1996 .
in addition , the periodicity of 182@xmath4184 d is seen to be always present and the fourier power corresponding to this periodicity does not vary much in both the north and south fields at high latitudes both before and after 1996 .
these high - frequency periods , in the high - latitude fields , do nt show much variation in their normalized fourier power and we do not see any north - south asymmetry in these high - latitude fields .
on the other hand , in the low - latitude zone , shown in the lower part of the table [ t - tab - high ] , we report periodicities of 145 d , 147 d , 151 d , 157 d , 172 d , 228 d , 248 d , 261 d , 278 d , 283 d and 288 d , which have already been reported in earlier work @xcite .
the new periodicity from our analysis was 257 d and is shown in boldface in the lower half of the table [ t - tab - high ] .
the rieger periodicity of 157 d is only seen in the south low - latitude field before 1996 .
also , the semi - annual variation of period 181 d is observed only in north low - latitude fields .
our wavelet analysis shows a transition occurring around 1996 in the distribution of power and periodicities of photospheric fields .
when the data were partitioned into periods prior to and after 1996 a hemispheric asymmetry is seen in the derived fourier periodicities of solar magnetic activity above latitudes of @xmath045@xmath1 .
a more detailed analysis shows that this asymmetry is confined to the latitude band 45@xmath1@xmath460@xmath1 in both hemispheres .
this latitude band is primarily dominated by strong , episodic poleward surges or tongues of magnetic flux in both hemispheres which , as stated earlier , are a direct surface manifestation of the meridional flow and the internal solar dynamo . in the 60@xmath1@xmath475@xmath1 latitude band
we find an asymmetry in the distribution of spectral power in the longer periods since the photospheric surface fields have saturated in this band .
this observed and localized asymmetry , confined to the 45@xmath1@xmath460@xmath1 latitude zone in both hemispheres , when coupled with the fact that both solar fields and the micro - turbulence levels in the inner - heliosphere have been decreasing since early to mid nineties suggests that active changes occurred around this time in the solar dynamo that governs the underlying basic processes in the sun .
these changes in turn probably initiated the build - up to one of the deepest solar minima , at the end of the cycle 23 , experienced in the past 100 years .
the magnetic time series in both the solar hemispheres exhibit a multitude of periodicities with significant variation in the spectral power of midterm ( 1@xmath42 year ) periodicities before and after 1996 .
these prominent periods in the lower solar latitudes , below @xmath2945@xmath1 , are thought to originate as the result of stochastic processes caused by the periodic emergence of surface magnetic flux @xcite as the solar cycle progresses . as stated earlier ,
@xcite have reported a 1.3 year periodicity at the base of the solar convection zone , which has also been detected in sunspot areas and sunspot number time series studied using wavelet transforms @xcite .
these findings have led to the conclusion that the mid - term fluctuations in the solar fields are surface manifestations of changes in the magnetic fluxes generated deep inside the sun . to 78@xmath1 for solar cycles 22 and 23 .
it is clear that the north solar field has reversed while the south solar field is still to do so .
the time of reversal of the north solar field is indicated by a small blue arrow.,width=453,height=453 ] polar surges above latitudes @xmath2945@xmath1 show significant variations in their occurrence rates and strength in each cycle and will therefore show up as a variation in spectral power and periods . a comparison of polar surges in the last three solar cycles , _ viz .
_ 21 , 22 , and 23 , reveals that they are comparatively more in strength after around solar maximum .
though we have restricted our analysis to data from cycles 21 , 22 and 23 in the period from @xmath21975 @xmath4 2010 , an examination of data beyond 2010 shows an asymmetry in the solar field reversal at high - latitudes .
figure [ f - reversal - assym ] shows the reversal of the northern high latitude field in cycle 24 in march 2011 .
however , the southern hemisphere is yet to undergo a reversal . the filled black dots in fig .
[ f - reversal - assym ] are the actual measurements while the red solid line is a smoothed curve .
the reversal of the solar polar field occurs at times when the red curve passes through zero and is indicated for cycle 24 by a small blue line in the northern hemisphere .
while it is known from earlier cycles that the two hemispheres do not reverse polarities at the same time , the time lag between the reversal of the northern hemisphere and southern hemisphere has , to the best of our knowledge , never been as large .
in addition , @xcite and @xcite , using the theory of meridional circulation , have reported an asymmetry in the latitudinal extent of the sun s meridional flow belt in the cycle 22 and 23 wherein surface meridional flows in cycle 22 extend to latitudes of @xmath2960@xmath1 while in cycle 23 , it went all the way to the poles
. thus , the meridional flow in cycle 23 , took a longer time thereby causing a slower return flow which led in turn to the extended solar minimum in cycle 23 .
recent work , using helioseismic data from the birmingham solar oscillation network ( bison ) , an instrument that is very sensitive in probing the solar interior close to the solar surface , has shown that the behavior of solar oscillation frequencies in the sub - surface magnetic layers during the descending phase of cycle 23 was significantly different from that during cycle 22 @xcite .
these authors went on to state that the peculiar solar minimum , at the end of the cycle 23 , could have been predicted long before it happened .
the present work , showing a north - south asymmetry around mid 1990 s in the quasi - periodic variations of photospheric fields at high - latitudes shows that changes were initiated at this time in the basic underlying solar dynamo processes like the meridional flow rates and the magnetic flux emergence that eventually led to the prolonged and deep minimum that we have just witnessed .
the current understanding of the solar dynamo is that it operates through the babcock - leighton mechanism to produce poloidal fields through the decay of tilted bipolar sunspots @xcite .
the strength of the polar field , at each solar maximum then depends upon the tilt angle in bipolar sunspots which , in turn , is determined by the action of the coriolis force acting on magnetic flux tubes that reach the surface by rising through the turbulent convection zone @xcite .
this process causes a large scatter in the average tilt - angle causing the polar field to be randomly weaker or stronger than in the previous cycle @xcite .
we have observed a decline in fields above 45@xmath1 for the past @xmath215 years implying weak polar fields being generated in two successive solar cycles _ viz .
_ cycles 22 and 23 .
a continuation of this declining trend beyond 22 years would imply a third successive weak polar field imparted by the babcock - leighton mechanism .
this is probably unlikely as the polar field is imparted by a random process and if it happens it would have serious implications on our present understanding of the solar dynamo . also , two of the eight strongest geomagnetic storms in the last 150 years have occurred during solar cycles 13 and 14 which were both relatively weak cycles . therefore , continued observations and measurements of solar magnetic fields are extremely important .
_ the authors would thank the free use data policy of the national solar observatory and acknowledge for providing the data in the public domain via the world wide web .
wavelet software was provided by c. torrence and g. compo , and is available at http://paos.colorado.edu/research/wavelets._ | using both wavelet and fourier analysis , a study has been undertaken of the changes in the quasi - periodic variations in solar photospheric fields in the build - up to one of the deepest solar minima experienced in the past 100 years .
this unusual and deep solar minimum occurred between solar cycles 23 and 24 .
the study , carried out using ground based synoptic magnetograms spanning the period 1975.14 to 2009.86 , covered solar cycles 21 , 22 and 23 . a hemispheric asymmetry in periodicities of the photospheric fields
was seen only at latitudes above @xmath045@xmath1 when the data was divided , based on a wavelet analysis , into two parts : one prior to 1996 and the other after 1996 .
furthermore , the hemispheric asymmetry was observed to be confined to the latitude range 45@xmath1 to 60@xmath1 .
this can be attributed to the variations in polar surges that primarily depend on both the emergence of surface magnetic flux and varying solar surface flows .
the observed asymmetry when coupled with the fact that both solar fields above @xmath045@xmath1 and micro - turbulence levels in the inner - heliosphere have been decreasing since the early to mid nineties @xcite suggests that around this time active changes occurred in the solar dynamo that governs the underlying basic processes in the sun .
these changes in turn probably initiated the build - up to the very deep solar minimum at the end of the cycle 23 .
the decline in fields above @xmath045@xmath1 for well over a solar cycle , would imply that weak polar fields have been generated in the past two successive solar cycles _ viz .
_ cycles 22 and 23 . a continuation of this declining trend beyond 22 years ,
if it occurs , will have serious implications on our current understanding of the solar dynamo . |
the recent compilation of oscillation data from the atmospheric @xcite , reactor @xcite , and long baseline @xcite neutrino experiments has provided solid evidence that neutrinos have small but non - zero masses . a global fit
to current data gives the following @xmath7 limits for the mixing parameters @xcite , @xmath8 since then , the measurements of neutrino oscillation parameters have entered a precision era . on the other hand ,
as no information exists for the value of @xmath5 , the dirac or majorana nature of the neutrinos , the dirac and/or majorana cp phases , and the neutrino mass hierarchy , there are discoveries that are still yet to come . in the standard model ( sm ) , due to the lack of right - handed neutrinos and the conservation of lepton numbers , neutrinos are massless . to generate non - zero neutrino masses thus calls for physics beyond the sm .
there have been many theoretical ideas proposed with an attempt to accommodate the experimentally observed small neutrino masses and the larger mixing angles among them . in ref .
@xcite , we have surveyed 63 models in the literature that are still viable candidates and have reasonably well - defined predictions for @xmath5 . we found that the predictions for @xmath9 of half of the models cover the range from 0.015 to the present upper bound of 0.13 . consequently , half of the models can be eliminated in the next generation of reactor experiments .
one of the implications of the observation of neutrino oscillation is the possibility of measurable branching ratio for charged lepton flavor - violating ( lfv ) decays . while not the case in the sm , predictions of the supersymmetric ( susy ) grand unified theories ( gut ) for these rare decays are much enhanced , as these processes are suppressed by the susy scale , rather than the plank scale @xcite .
furthermore , as different models obtain large neutrino mixing angles through different mechanisms , their predictions for the lfv charged lepton decays can be very distinct .
consequently , lfv charged lepton decays may provide a way to distinguish different susy gut models . among the models aiming to explain the neutrino masses and mixing , a particularly promising class are those based on ( susy ) so(10 )
; for recent reviews of so(10 ) models , see @xcite . in this paper
, we investigate the predictions for various lfv charged lepton decays as well as muon - electron conversion in five of the susy so(10 ) models , assuming constrained minimal susy standard model ( cmssm ) boundary conditions where only five soft - susy breaking parameters are present .
furthermore , we impose the wilkinson microwave anisotropy probe ( wmap ) dark matter constraints in the neutralino , stau and stop coannihilation regions . specifically , we present the allowed values for these soft - susy parameters for various branching ratios of these rare lfv processes . in addition , the lower bounds on the predictions for these rare processes in the five different susy so(10 ) models are given .
we find that the predictions in these models are very distinct .
we note the crucial role of the wmap constraints in deducing the lower bounds on the predictions .
many authors have previously studied the branching ratio predictions for charged lfv decays in the susy gut framework .
rather than study specific models , they generally adopt a generic approach and assume a nearly diagonal cabibbo - kobayashi - maskawa ( ckm)-like or bimaximal pontecorvo - maki - nakagawa - sakata ( pmns)-like mixing matrix to diagonalize the yukawa neutrino matrix @xcite . following the procedure of casas and ibarra @xcite to invert the seesaw formula ,
they have carried out monte carlo studies by scanning the unknown right - handed neutrino mass spectrum and the angles and phases of the inversion matrix in order to present scatter plots of the rare branching ratios . a few exceptions to this procedure can be found in ref .
here we are interested in specific models in order to determine the ability of the lfv experiments to differentiate among and rule out some models .
the models chosen are highly predictive and illustrate a wide variety of textures for the charged lepton and neutrino mass matrices , and predictions for the right - handed neutrino mass spectrum and the reactor neutrino mixing angle , @xmath5 . in sec .
ii we describe the five representative susy so(10 ) models that we have analyzed . in sec .
iii we review lfv charged lepton decays in the sm and susy guts and present the predictions for the five susy so(10 ) models considered . in sec .
iv their expectations for @xmath2 conversion are given .
v concludes this paper .
we begin with a brief discussion of the general formalism on which the supersymmetric @xmath3 grand unified models are based .
for all five models to be illustrated , the seesaw mechanism @xcite is of the conventional type i leading to normal hierarchies for the light majorana neutrinos .
the leptonic sector of the yukawa superpotential at the gut scale can then be written as @xmath10 where @xmath11 represents the left - handed lepton doublet , @xmath12 the left - handed conjugate neutrino singlet , and @xmath13 the left - handed conjugate charged lepton of one of the three @xmath14 dimensional representations of @xmath3 . when the two higgs doublets , @xmath15 and @xmath16 , acquire vacuum expectation values , the charged lepton and neutrino mass matrices are generated and can be written in @xmath17 form + @xmath18 { \cal m}^\nu & = & \left(\begin{array}{cc } \nu^t_l & n^{ct}_l\\ \end{array}\right ) \left(\begin{array}{cc } 0 & m^t_n\\ m_n & m_r\\ \end{array}\right ) \left(\begin{array}{c } \nu_l \\
n^c_l\\ \end{array}\right ) .
\end{array } \label{eq : massmat}\ ] ] with the entries in @xmath19 much larger than those in @xmath20 , the light majorana neutrino mass matrix is given by the well - known type i seesaw formula @xcite , @xmath21 note that the above dirac @xmath22 mass matrix entries , @xmath23 and @xmath20 , are written in right - left order and appear in the @xmath3 flavor basis .
as such , the matrices can be diagonalized by the unitary transformations @xmath24 by forming the hermitian products , calculating the left and right unitary transformations of column eigenvectors , and phase rotating the complex symmetric @xmath19 and @xmath25 matrices , so their mass eigenvalues are real .
the pmns neutrino mixing matrix @xcite is then given by @xmath26 .
on the other hand for later use , it is convenient to transform to the bases where @xmath23 and @xmath19 are diagonal and denoted by primed quantities . in this case @xmath27 where now @xmath28 , since @xmath29 is just the identity matrix .
a comparison of the seesaw diagonalization matrices then reveals that the transformed dirac neutrino matrix @xmath30 in the new basis is given by @xmath31 in terms of the original matrix in the flavor bases . with either basis ,
the left - handed neutrino pmns mixing matrix can be written as @xmath32 , where by convention @xcite
@xmath33 in terms of the three mixing angles , @xmath34 and @xmath5 ; and the dirac @xmath35 phase , @xmath36 , in analogy with the quark mixing matrix . the majorana phase matrix , @xmath37 written in terms of the two majorana phases , @xmath38 and @xmath39 , is required since an arbitrary phase transformation is not possible when one demands real diagonal neutrino mass entries in the transformation of @xmath25 in eq . ( [ eq : diag ] ) or @xmath40 in eq . ( [ eq : diagbas ] ) . with this background in mind
, we now turn to a brief discussion of the five @xmath3 models we have considered for our lepton flavor violation study .
the @xmath3 grand unification symmetry is an economical and attractive one @xcite , for all sixteen left - handed quark and lepton fields and their left - handed conjugates fit neatly into one @xmath14 representation per family .
many models exist in the literature which differ from one another by their higgs representation assignments and type of flavor symmetry imposed , if any . to appreciate this ,
it is of interest to note the following decompositions of the direct product of representations : @xmath41 where in the first product the @xmath42 and @xmath43 matrices are symmetric , while the @xmath44 is antisymmetric under the interchange of family indices . given this group structure for @xmath3 , there are two general classes of models which have been extensively studied . those with higgs in the @xmath45 , and possibly the @xmath44 and/or @xmath46 , dimensional representations are often referred to as the minimal higgs models @xcite and lead to symmetric or antisymmetric matrix elements , or a superposition of the two .
the advantage is that the couplings are renormalizable and preserve r - parity , but the latter representations are of rather high rank and disfavored in the string theory framework .
the other class typically involves the lower rank higgs representations such as @xmath47 , with some non - renormalizable effective operators formed from them , @xcite but r - parity is not conserved .
they can lead to lopsided mass matrices for the charged lepton and down quark mass matrices , due to the @xmath48 structure present in the @xmath14 s .
as such , one may anticipate that they will predict a higher level of lepton flavor violation than the first class .
the @xmath49 triplet components in the higgs representations in all five models considered have no electroweak vev and hence lead to the conventional type i seesaw mechanism with the prediction of normal hierarchy for the light left - handed neutrino spectrum @xcite .
we elaborate on each model in turn but give only the neutrino and charged lepton mass matrices .
all successfully predict the observed quark structure and ckm mixings .
this model , based on @xmath3 with a @xmath50 flavor symmetry @xcite , is of the lopsided variety and has matrices of the following textures : @xmath51 where the parameters have the values @xmath52 .
the right - handed majorana mass matrix is given by @xmath53 where @xmath54 gev .
one finds @xmath55 a value of @xmath56 assures that @xmath57 and that the corresponding lopsided nature of the 23 element of the down quark mass matrix gives a good fit to the results for the quark sector .
the @xmath58 elements of the dirac neutrino matrix were added @xcite to the original model to give a better fit to baryogenesis arising from resonant leptogenesis involving the two lighter right - handed neutrinos @xcite .
evolution downward from the gut scale to @xmath59 has little effect on the mixing angles due to the small @xmath60 , the opposite cp - parity of the lighter two right - handed majorana neutrinos , and their large hierarchy with the heaviest one ; but serves to lower the values of the @xmath61 s and @xmath62 and @xmath63 relative to their gut scale values .
this model , based on @xmath3 with a @xmath64 flavor symmetry @xcite , is of the minimal higgs variety leading to symmetric entries in the mass matrices : @xmath65 where @xmath66 gev and the parameters have the values @xmath67 .
the solar and atmospheric neutrino mass squared differences , @xmath68 and @xmath69 were used as input to determine the @xmath70 and @xmath71 parameters in the effective light left - handed majorana neutrino mass matrix @xmath72 one then finds @xmath73 for this model , a value of @xmath74 is used .
the effect of evolution from the gut scale to @xmath59 is to raise @xmath75 and to lower @xmath76 .
this model is based on @xmath3 with an @xmath77 flavor symmetry @xcite .
the higgs fields appear in six @xmath42 s , three @xmath78 s , three @xmath43 , and one @xmath46 representations of @xmath3 , distinguished by their @xmath77 flavor assignments . of the 14 pairs of higgs doublets at the gut scale ,
all but one pair are assumed to get superheavy .
the charged lepton mass matrix is chosen to be diagonal , while the right - handed majorana neutrino mass matrix is proportional to the identity matrix , so all three heavy neutrinos are degenerate . the dirac neutrino and charged lepton mass matrices are symmetric and given by @xmath79 m_e & = & \left(\begin{array}{ccc } b_0 - 2 b_2 - 3(e_0 - 2 e_2 ) & 0 & 0\\ 0 & b_0 + b_1 + b_2 - 3(e_0 + e_1 + e_2 ) & 0\\ 0 & 0 & b_0 - b_1 + b_2 - 3(e_0 - e_1 + e_2)\\ \end{array}\right),\\ \label{eq : mnmecy } \end{array}\ ] ] where the parameters have the values @xmath80 , all in gev . the common mass of the degenerate right - handed neutrinos is determined to be @xmath81 gev , so as to fit @xmath82 with the aid of the seesaw formula .
the authors find @xmath83 a value of @xmath74 is used to evolve the quark and charged lepton masses , though the neutrino masses and mixings have not been evolved downward to the electroweak scale .
this model is based on @xmath3 with a @xmath84 family symmetry @xcite . the charged lepton and
down quark mass matrices have lopsided textures , but they are not so extreme as in the case of the albright - barr model : @xmath85 where @xmath66 gev and the parameters have the values @xmath86 .
the right - handed majorana mass matrix is diagonal with mass eigenvalues @xmath87 , @xmath88 , @xmath89 gev .
these parameters were chosen by making use of the central experimental values for @xmath90 and @xmath91 at the time of writing .
the authors find @xmath92 the evolution has been carried out with @xmath93 and @xmath94 gev down to @xmath59 and then down to the 1 gev scale for the light neutrino masses .
this model is based on @xmath3 with an @xmath95 flavor symmetry @xcite .
the higgs fields appear in one each of the @xmath42 , @xmath44 , and @xmath78 dimensional representations of @xmath3 .
the fermion mass matrices are generated by renormalizable yukawa couplings of these higgs fields to the three families placed in @xmath14 s .
the dirac neutrino and charged lepton mass matrices are written as linear combinations of the yukawa couplings of the @xmath42 , @xmath44 , and @xmath78 higgs representations , respectively , while the right - handed majorana matrix is proportional only to the latter : @xmath96 where the individual higgs contributions to the mass matrices are given by @xmath97 in the right - left order we have adopted .
the coefficient parameters are taken to be @xmath98 .
a value of @xmath74 is used to evolve the quark and charged lepton masses downward to the electroweak scale .
one finds then that @xmath99 although the results are impressive , the authors do note that 21 parameters have been introduced in order to obtain the results .
hence the model should be viewed as an existence proof that the lepton masses and mixings as well as the quark masses and mixings can be described in the framework of renormalizable couplings with only the three higgs representations contributing to the yukawa couplings .
.[table i ] higgs representations , flavor symmetries , and other noteworthy features of the five @xmath3 susy gut models considered in this work .
[ cols="^,^,^,^,^,^,<",options="header " , ] in table iii we summarize the relevant findings from our study of the five models .
the branching ratio ranges apply for the @xmath100 case and with the stricter wmap dark matter constraints imposed .
it is clear that the five predictive @xmath3 susy gut models considered have very representative right - handed neutrino mass spectra and predictions for @xmath76 .
the cm , dr , and gk models have massive hierarchical spectra with @xmath6 ranging from @xmath101 to @xmath102 gev . the cy model , on the other hand , has a degenerate spectrum with @xmath103 gev , while the ab model has degenerate @xmath104 and @xmath105 which can lead to resonant leptogenesis .
the cm model has a relatively large @xmath76 prediction which should be observable at the upcoming reactor neutrino experiments , double chooz and daya bay .
the ab , cy and dr models have similar predictions for @xmath76 which will make observation of @xmath106 oscillation somewhat marginal at those reactors and in the proposed no@xmath107a and t2k long - baseline experiments @xcite without a superbeam source . for the gk model , the observation of such a low @xmath76 prediction would only take place with a neutrino factory .
but it is clear from table iii and the previous figures that the gk and ab models will be tested first with the meg experiment . from our discussion
it is then clear that the lfv branching ratios are more sensitive to large @xmath6 than to large @xmath5 in the models considered . in previous generic studies of susy gut models , the rare branching ratios were nearly equally sensitive to each of the two parameters @xcite .
it is also of interest to consider lepton flavor violation in the @xmath109 to @xmath110 conversion process in titanium , @xmath111 .
while there is no such ongoing experiment , preliminary discussion is underway to propose one which will lower the present limit .
we shall see that the conversion rate predictions relative to the muon capture process , @xmath112 , are such that all five models considered in this paper can potentially be eliminated , if no signal is observed . while predictions for the conversion process in the sm are infinitesimally small , large enhancements in the cmssm framework again occur by virtue of massive super partners appearing in loop diagrams involving @xmath113 and higgs penguins and boxes .
of these , arganda , herrero and teixeira have shown that the @xmath114 penguin contributions dominate the others by at least two orders of magnitude @xcite .
the slepton - neutralino and sneutrino - chargino loop contributions to the @xmath114 penguins are shown in fig .
the virtual massive @xmath12 and @xmath115 with their yukawa couplings again appear in the slepton loops along with a higgsino or higgs particle , respectively .
examples of feynman diagrams for slepton - neutralino and sneutrino - chargino contributions to @xmath108 conversion in susy models with slepton mass insertions.,title="fig : " ] + the complex formulas for lepton flavor violation in this process were first derived by hisano , moroi , tobe , and yamaguchi @xcite . if we restrict our attention to the @xmath114 penguin contribution as suggested by arganda et al .
@xcite , one finds the @xmath108 conversion rate is given by @xmath116 , \label{eq : mueconvrate}\ ] ] where here @xmath117 are form factors for the vertices connecting leptons of equal chirality , while @xmath118 are combinations of the electric dipole and magnetic dipole transition form factors previously denoted by @xmath119 . for a @xmath120 target , @xmath121 and the nuclear form factor
is @xmath122 @xcite . in the case of the conversion process
, we have explicitly carried out the full evolution running from the gut scale to the @xmath123 scale .
the @xmath108 conversion branching ratio is then obtained from the conversion rate above by scaling it with the @xmath109 capture rate on @xmath124 , which is quoted in @xcite as @xmath125 with the present experimental limit on the conversion branching ratio found to be @xmath126 . branching ratio predictions for @xmath108 conversion _ vs. _ branching ratio predictions for @xmath4 in the five models considered .
only thick line segments are shown reflecting the application of the more restrictive wmap dark matter constraints .
note that the predictions for the cm and cy models overlap.,title="fig : " ] + in fig .
we show a plot of the @xmath108 conversion branching ratio _ vs. _ the @xmath127 branching ratio for each of the five models considered .
we have limited the line segments by applying the wmap dark matter constraints of sec .
it is clear that the gk and ab models would be tested first , followed by the dr , cy and cm models .
in fact , a first generation @xmath108 conversion experiment may be able to reach a branching ratio of @xmath128 , while a second generation experiment may lower the limit from the present value down to @xmath129 @xcite .
if such proves to be the case and no signal is seen , all five models will be eliminated .
hence the conversion experiment is inherently more powerful than the meg experiment looking for @xmath130 which is designed to reach a level of @xmath131 , sufficient only to eliminate the gk and ab models .
the caveat , of course , is that meg is now starting to take data , while no new conversion experiment has been approved to date .
there have been many theoretical models constructed which aim to explain the neutrino masses and mixing patterns . while the predictions for the value of @xmath5 in these models may provide a way to distinguish some of these models , as shown in the survey over 63 models we have carried out in ref .
@xcite , the rare lfv processes , such as @xmath4 and @xmath2 conversion can provide an even more sensitive way to disentangle these models .
we have investigated these rare processes in five susy so(10 ) models that are currently still viable and highly predictive , making use of the allowed parameter space for the soft susy breaking parameters in the cmssm framework . utilizing the wmap dark matter constraints , lower bounds on the branching ratios of these rare processes can be placed , and we find that at least three of the five models considered give rise to prediction for @xmath132 that will be tested at meg .
more interestingly , the next generation @xmath2 conversion experiment should be sensitive to the predictions of all five models , making it an even more robust way to test these models . while generic studies have emphasized the important dependence of the branching ratios on the reactor neutrino angle , @xmath5 , and the mass of the heaviest right - handed neutrino , we find the latter to be by far the more significant in the models tested .
we thank stephen martin for making available to us his evolution program for the yukawa couplings and soft susy breaking parameters . the work of m .- c.c .
is supported , in part , by the national science foundation under grant no .
one of us ( c.h.a . ) thanks the members of the theory group at fermilab for their kind hospitality .
fermilab is operated by the fermi research alliance under contract no .
de - ac02 - 07ch11359 with the u.s .
department of energy .
_ note added._after completion of this work , the authors of ref .
@xcite informed us that a suppression factor should be applied to the lepton flavor violating branching ratios calculated in our paper .
this suppression arises from a qed radiative correction to the effective dipole operators responsible for the rare decays .
the multiplicative factor in question is given by @xmath133 , where @xmath134 , @xmath135 gev is the sparticle mass scale responsible for the lfv , and @xmath136 for the rare radiative muon or tau decays . for @xmath127 , the @xmath36 correction amouts to roughly 0.15 , so the branching ratio is about 0.85 times the rate given in the paper .
the authors thank andrzej czarnecki and ernest jankowski for pointing out their earlier work to them .
( super - kamiokande collaboration ) , phys .
d * 71 * , 112005 ( 2005 ) ; j. hosaka _
( super - kamiokande collaboration ) , phys .
d * 73 * , 112001 ( 2006 ) ; b. aharmin _ et al . _
( sno collaboration ) , phys.rev .
c * 72 * , 055502 ( 2005 ) ; m. apollonio _
( chooz collaboration ) , phys . lett.b * 420 * , 397 ( 1998 ) ; ibid .
j c * 27 * , 331 ( 2003 ) ; f. boehm _
( palo verde collaboration ) , phys.rev .
d * 64 * , 112001(2001 ) ; t. araki _ et al . _
( kamland collaboration ) , phys.rev.lett.*94 * , 081801 ( 2005 ) ; s. abe _
( kamland collaboration ) , arxiv:0801.4589 .
( k2k collaboration ) , phys .
d * 74 * , 072003 ( 2006 ) ; d.g .
( minos collaboration ) , phys .
* 97 * , 191801 ( 2006 ) ; ( minos collaboration ) , arxiv:0708.1495 .
m. maltoni , t. schwetz , m.a .
tortola and j.w.f .
valle , new j. phys . *
6 * , 122 ( 2004 ) .
c.h . albright and m .- c .
chen , phys .
d * 74 * , 113006 ( 2006 ) .
f. borzumati and a. masiero , phys .
lett . * 57 * , 961 ( 1986 ) . for recent reviews ,
see , e.g. m .- c .
chen and k. t. mahanthappa , int .
j. mod .
a * 18 * , 5819 ( 2003 ) ; aip conf .
proc . *
721 * , 269 ( 2004 ) ; g. altarelli and f. feruglio , new j. phys .
* 6 * , 106 ( 2004 ) ; s.f .
king , rept .
. phys . * 67 * , 107 ( 2004 ) ; z.z .
xing , int .
a * 19 * , 1 ( 2004 ) ; r.n.mohapatra , new j. phys . *
6 * , 82 ( 2004 ) ; a. yu .
smirnov , int .
a * 19 * 1180 ( 2004 ) ; r.n . mohapatra and a.y .
smirnov , ann .
rev . of nucl . and part .
, * 56 * , 569 ( 2006 ) ; r.n .
. phys . * 70 * , 1757 ( 2007 ) ; a. strumia and f. vissani , arxiv : hep - ph/0606054 .
a. masiero , s.k .
vempati , and o. vives , nucl . phys .
* b649 * , 189 ( 2003 ) ; new j. phys .
* 6 * , 202 ( 2004 ) ; f. deppisch , h. ps , a. redelbach , r. rckl , and y. shimizu , eur . phys . j. c * 28 * , 365 ( 2003 ) ; f. deppisch , h. ps , r. rckl , and a. redelbach , phys . rev.d * 73 * , 033004 ( 2006 ) ; e. arganda and m.j .
herrero , phys .
d * 73 * , 055003 ( 2006 ) ; l. calibbi , a. faccia , a. masiero , and s.k .
vempati , j. high energy phys .
* 7 * , 012 ( 2007 ) .
casas and a. ibarra , nucl .
* b618 * , 171 ( 2001 ) .
a. kageyama , s. kaneko , n. shimoyama , and m. tanimoto , phys .
b * 527 * , 206 ( 2002 ) ; q. shafi and z. tavartkiladze , nucl .
b772 * , 133 ( 2007 ) ; * b778 * , 216(e ) ( 2007 ) ; s. antusch and s.f . king , phys .
b * 659 * , 640 ( 2008 )
. an earlier test of one of the models considered in this work for the @xmath130 branching ratio was published in e. jankowski and d.w .
maybury , phys .
d * 70 * , 035004 ( 2004 ) .
p. minkowski , phys .
b * 67 * , 421 ( 1977 ) ; t. yanagida , in _ proceedings of the workshop on the unified theory and baryon number in the universe , tsukuba , japan 1979 _ , edited by o. sawada and a. sugamoto ( kek , tsukuba , 1979 ) , p. 95 ; m. gell - mann , p. ramond , and r. slansky , in _ supergravity _ , edited by p. van nieuwenhuizen and d. z. fredman ( north - holland , amsterdam , 1979 ) , p. 315
glashow , in _ proceedings of the 1979 cargese summer institute on quarks and leptons _ , edited by m. levy , j .- l .
basdevant , d. speiser , j. weyers , r. gastmans , and m. jacob ( plenum press , new york , 1980 ) , p. 687 ; r.n .
mohapatra and g. senjanovic , phys .
lett . * 44 * , 912 ( 1980 ) .
b. pontecorvo , zh .
* 33 * , 549 ( 1957 ) [ sov .
jetp * 6 * , 429 ( 1957 ) ] ; z. maki , m. nakagawa and s. sakata , prog .
. phys . * 28 * , 870 ( 1962 ) .
review of particle physics , w .- m .
yao et al . , j. phys .
g * 33 * , 1 ( 2006 ) .
babu and r.n .
mohapatra , phys .
* 70 * , 2845 ( 1993 ) .
babu and s.m .
barr , phys .
b * 381 * , 202 ( 1996 ) ; c.h .
albright , k.s .
babu , and s.m .
barr , phys .
lett . * 81 * , 1167 ( 1998 ) .
albright , phys .
b * 599 * , 285 ( 2004 ) .
c.h . albright and s.m .
barr , phys.rev .
d * 64 * , 073010 ( 2001 ) .
albright , phys .
d * 72 * , 013001 ( 2005 ) ; * 74 * , 039903(e ) ( 2006 ) . a. pilaftsis and t.e.j .
underwood , nucl .
b692 * , 303 ( 2004 ) .
c . chen and k.t .
mahanthappa , phys .
d * 70 * , 113013 ( 2004 ) .
y. cai and h .- b .
yu , phys .
d * 74 * , 115005 ( 2006 ) .
r. dermisek and s. raby , phys.lett .
b * 622 * , 327 ( 2005 ) .
w. grimus and h. khbock , phys .
b * 643 * , 182 ( 2006 ) .
s.a . dazley ( double chooz collaboration ) , in proceedings of nufact05 , nucl .
b , proc . suppl .
* 155 * , 231 ( 2006 ) ; j. cao ( daya bay collaboration ) , ibid .
* 155 * , 229 ( 2006 ) .
m. apollonio _
( chooz collaboration ) , phys .
b * 420 * , 397 ( 1998 ) ; eur .
j. c * 27 * , 331 ( 2003 ) ; cf .
also f. boehm _ et al .
_ ( palo verde collaboration ) , phys.rev .
d * 64 * , 112001(2001 ) .
b.w . lee and r.e .
shrock , phys .
d * 16 * , 1444 ( 1977 ) .
a. baldini , nucl .
b , proc . suppl . * 168 * , 334 ( 2007 ) .
_ ( mega collaboration ) , phys . rev
. lett . * 83 * , 1521 ( 1999 ) . for references to the cmssm , cf .
olive , submitted to the susy07 proceedings arxiv:0709.3303 .
j. hisano , t. moroi , k. tobe , and m. yamaguchi , phys . rev . *
53 * , 2442 ( 1996 ) .
petcov , s. profumo , y. takanishi , and c.e .
yaguna , nucl . phys . *
b676 * , 453 ( 2004 ) .
d.n . spergel _
( wmap collaboration ) , astrophys .
170 * , 377 ( 2007 ) .
p. binetruy , g. girardi , and p. salati , nucl . phys . *
b237 * , 285 ( 1984 ) ; k. griest and d. seckel , phys .
d * 43 * , 3191 ( 1991 ) .
l.s . stark , p. hfliger , a. biland , and f. pauss , j. high energy phys . * 0508 * , 059 ( 2005 ) .
b. aubert _
( babar collaboration ) , phys .
* 96 * , 041801 ( 2006 ) ; * 95 * , 041802 ( 2005 ) ; k. abe _
et al . _ ( belle collaboration ) , in _ proceedings of the ichep06 conference , moscow , 2006_. a.g .
( superkekb physics working group ) , arxiv : hep - ex/0406071 .
v. cirigliano , a. kurylov , m.j .
ramsey - musolf , and p. vogel , phys .
* 93 * , 231802 ( 2004 ) .
y. hayato _
( t2k collaboration ) , nucl .
b , proc . suppl . * 143 * , 269 ( 2005 ) ; d.s .
et al . _ ,
technical design report for the no@xmath107a experiment e929 at fermilab , 2007 .
s. antusch , e. arganda , m.j .
herrero , and a.m. teixeira , j. high energy phys .
11 , 090 ( 2006 ) , cf .
e. arganda , m.j .
herrero , and a.m. teixeira , j. high energy phys .
10 , 104 ( 2007 ) . c. dohmen _
( sindrum ii collaboration ) , phys . lett .
b * 317 * , 631 ( 1993 ) .
w. molzon , in _ fermilab program planning report presented before the p5 committee , january 2008_. a. czarnecki and e. jankowski , phys .
d * 65 * , 113004 ( 2002 ) . | there have been many theoretical models constructed which aim to explain the neutrino masses and mixing patterns .
while many of the models will be eliminated once more accurate determinations of the mixing parameters , especially @xmath0 , are obtained , charged lepton flavor violation experiments are able to differentiate even further among the models . in this paper , we investigate various rare lepton flavor violation processes , such as @xmath1 and @xmath2 conversion , in five predictive supersymmetric ( susy ) @xmath3 models and their allowed soft - susy breaking parameter space in the constrained minimal susy standard model . utilizing the wilkinson microwave anisotropy probe dark matter constraints , we obtain lower bounds on the branching ratios of these rare processes and find that at least three of the five models we consider give rise to predictions for @xmath4 that will be tested by the meg collaboration at psi .
in addition , the next generation @xmath2 conversion experiment has sensitivity to the predictions of all five models , making it an even more robust way to test these models .
while generic studies have emphasized the dependence of the branching ratios of these rare processes on the reactor neutrino angle , @xmath5 , and the mass of the heaviest right - handed neutrino , @xmath6 , we find very massive @xmath6 is more significant than large @xmath5 in leading to branching ratios near to the present upper limits . |
study of temporal variability including signatures of quasi - periodic oscillations ( qpos ) is an important aspect of astrophysics of black holes .
several models in the literature attempt to explain origin of low frequency qpos .
they include perturbation inside a keplerian disk @xcite , global disk oscillation @xcite , oscillation of wrapped disk @xcite , accretion ejection instability at the inner radius of a keplerian disk @xcite .
@xcite , envisages a bounded region surrounding compact objects which is called the transition layer ( tl ) and identifies low frequency qpos as that associated with the viscous magneto - acoustic resonance oscillation of the bounded tl .
chakrabarti and his collaborators @xcite showed that the oscillations of centrifugal pressure supported accretion shocks @xcite could cause observed low frequency qpos . according to the two - component advective flow ( tcaf ) model @xcite
the post - shock region itself is the compton cloud . because the shock is formed due to centrifugal force , where energy is dissipated and angular momentum is redistributed , post - shock region is also known as the centrifugal pressure supported boundary layer ( cenbol ) of the black hole .
this tcaf solution has been proven to be of stable configuration ( giri & chakrabarti , 2013 ) and monte - carlo simulations of spectral and timing properties through a time dependent radiative hydrodynamic code showed the formation of qpos very similar to what is observed @xcite . the compton cloud becomes smaller because of higher viscosity as well as higher cooling .
higher viscosity causes the keplerian disk on the equatorial plane to move in .
this causes the compton cloud to cool down .
this picture is clear from the two component model of @xcite and the outburst source picture given in @xcite based on it . to our knowledge , except tcaf
, no other model is found to be capable of explaining continuous and simultaneous variation of spectral properties and timing properties ( see , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
there are mainly two reasons behind oscillation of shock wave in an accretion flow : i ) resonance oscillation : when cooling time scale of the flow is comparable to the infall time scale @xcite , this type of oscillation occurs .
such cases can be identified by the fact that when accretion of the keplerian disk is steadily increased , qpos may occur in a range of the accretion rates , and the frequency should go up with accretion rate . not all the qpos may be of this type .
some sources ( for example , 2010 gx 339 - 4 outburst ) , show signatures of sporadic qpos during rising soft - intermediate states ( where qpo frequencies of in @xmath36 hz were observed for around 26 days ; @xcite ) , although rise in the accretion rates . in these cases
the shock strength has to change in order that the resonance condition holds good .
ii ) non - steady solution : in this case , the flow has two saddle type sonic points , but rankine - hugoniot conditions which were used to study standing shocks in @xcite are not satisfied .
examples of these oscillations are given in @xcite where no explicit cooling was used .
such type of qpos are possible at all accretion rates , outside the regime of type ( i ) qpos mentioned above .
qpo frequencies depend on viscosity ( higher viscosity will remove angular momentum , bring shocks closer to the black hole , and produce higher frequency qpos ) , but not explicitly on accretion rate . in any case , observed qpo frequency is inversely proportional to the infall time ( @xmath4 ) in the post - shock region .
so , when low frequency ( e.g. , mhz to few hz ) qpos are observed , generally during very early phase of an outburst or very late phase of an outburst of transient black hole candidates ( bhcs ) , shocks are located very far away from black holes and size of the cenbol is large . as a result ,
amount of cooling by photons from keplerian disk @xcite is high ( * ? ? ?
* ; * ? ? ?
* hereafter paper - i ) and cenbol pressure drops , moving the shock closer towards black hole ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* paper - i ) until pressure ( including centrifugal ) is strong enough to balance the inward pull .
lower shock location increases the qpo frequency .
different bhcs show different oscillation frequencies during their evolution ( both in rising and declining ) phases .
using propagating oscillatory shock ( pos ) model by chakrabarti and his collaborators @xcite one can satisfactorily explain origin and day - wise evolution of qpo frequencies during rising and declining phases of outbursting bhcs . during rising phase shock moves towards black holes increasing qpo frequencies monotonically with time and opposite scenario
is observed during declining phase , mainly in hard and hard - intermediate spectral states of the outbursts ( see , * ? ? ?
recently @xcite showed that observed qpo frequencies can be predicted from detailed spectral analysis using two component advective flow ( tcaf ) model as a local additive table model in xspec . @xcite and @xcite also showed physical reason behind spectral state transitions from spectral model fitted parameters of tcaf model for two different galactic bhcs h 1743 - 322 , and gx 339 - 4 during their outbursts .
basically , the same shock location is obtained by fitting the spectra produces qpos through oscillations .
so spectral properties are interlinked with timing properties as far as tcaf solution is concerned . in this
_ paper _ , our goal is to explain origin of observed qpo evolution from pure analytical point of view using compton cooling .
biggest uncertainly being that of viscosity parameter , we would like to have an idea of how viscosity usually vary with distance in a known source .
we consider a transient bhc h 1743 - 322 during its 2010 outburst .
we hope that in future , this behavior could be used to better predict qpo evolutions . in august 2010 ,
h 1743 - 322 was found to be active in x - rays @xcite with a characteristics of temporal and spectral evolutions @xcite similar to those observed in other transient bhcs ( see for a review , * ? ? ?
detailed source description is already in literature ( * ? ? ?
* ; * ? ? ?
* and references therein ) .
the _ paper _ is organized in the following way : in the next section , we discuss governing equations of modified rankine - hugoniot ( r - h ) shock conditions in presence of compton cooling . in 3 , observed qpo evolution and what this tells us about viscosity variation in the disk as a function of radial distance .
we also present phase space diagram of the flow in progressive days .
finally , in 4 , we briefly discuss our results and make our concluding remarks .
we assume the accreting flow to be thin , axisymmetric and rotating around vertical axis . to avoid integrating in a direction transverse to flow motion , we consider that the flow is in hydrostatic equilibrium in vertical direction as in @xcite . in tcaf , cenbol is basically the post - shock region of a low angular momentum , sub - keplerian flow .
it is comparatively hotter , puffed up , and much like an ion supported torus @xcite . due to inverse compton cooling effect of intercepted low energy photons from a keplerian disk , energy of cenbol decreases and is radiated away .
energy equation at the shock is modified by , @xmath5 where , @xmath6 is the energy loss due to comptonization .
baryon number conservation equation at the shock is , @xmath7 since the gas is puffed up , r - h conditions @xcite have to be modified , so that only vertically integrated pressure and density parameters are important .
this modification was first carried out in @xcite , where pressure balance condition was modified using vertically integrated values : @xmath8 here , @xmath9 and @xmath10 are pressure and density , integrated in the vertical direction @xcite . in our solution , we use the eq .
( 8a ) of paper - i as an invariant quantity at the shock , which is given by : @xmath11 ^ 2}{2+(\gamma-1)m_{+}^2}= \frac{[m_{-}(3\gamma-1)+(\frac{2}{m_{-}})]^2}{2+(\gamma-1)m_{-}^2-\zeta } , \eqno{(2)}\ ] ] where , @xmath12 , @xmath13 and @xmath14 are the mach number , radial velocity and adiabatic index of flow respectively , @xmath15 . here , @xmath16 is adiabatic sound speed .
we follow the same mathematical procedure and methodology as in paper - i , to find shock location for a given cooling rate . in the standard theory of thin
accretion flows around black holes @xcite viscosity plays a major role .
@xcite showed formation of keplerian disk for super - critical @xmath0 parameter @xcite .
inflow angular momentum is transported outward by viscosity and allow it to fall into the black holes .
as the shock moves closer , the angular momentum must be adjusted by viscosity so that the shock formation is theoretically allowed . for our viscosity calculation
, we use the relation @xcite : @xmath17 where , @xmath18 , is the viscous stress , @xmath0 being @xcite viscosity parameter .
angular momentum variation from eq .
( 3 ) can be written as @xmath19 where , @xmath20 is the change in angular momentum @xmath21 due to viscous transport .
we analyze archival data of @xmath22 observational ids of rxte / pca instrument ( only pcu2 , all layers ) starting from 2010 august 9 ( modified julian day , i.e. , mjd = 55417.2 ) to 2010 august 16 ( mjd = 55424.1 ) , selected from rising phase of 2010 outburst of h 1743 - 322 .
we carry out data analysis using ftools software package heasoft version headas 6.14 and xspec version 12.8 . for generation of source and background ` .pha ' files and spectral fitting ( in @xmath23 kev energy range ) using combined disk blackbody and power - law models , we use same method as described in @xcite . after achieving best fit based on reduced chi - square value ( @xmath24 ) ,
we integrate only power - law component of the spectrum .
this can be written as : @xmath25 where , @xmath26 and @xmath27 are the lower and the upper limits of energy .
for interstellar photo - electric absorption correction , we follow the prescription of @xcite . to calculate cooling time of the compton cloud ( cenbol ) from observed spectrum , we consider distance correction in following way
: we multiply the integrated spectrum by the model normalization value ( @xmath28 ) of @xmath29 , where ` @xmath30 ' is source distance in @xmath31 kpc unit and ` @xmath32 ' is the disk inclination angle . in case of h 1743 - 322 , we use source distance @xmath33 = @xmath34 kpc , and @xmath35 @xcite .
we keep hydrogen column density ( n@xmath36 ) frozen at 1.6@xmath37 10@xmath38 atoms @xmath39 for absorption model _ wabs _ and assume a @xmath40% systematic error for all observations @xcite .
in this _ paper _ , we study origin and evolution of qpos in outbursting bhc h 1743 - 322 , from a purely analytical point of view . in @xcite and @xcite , it was shown that matter from the companion is heated up due to compression and puffed up due to centrifugal barrier to form cenbol .
low energy photons from a @xcite disk with an accretion rate of @xmath41 are intercepted by cenbol and are emitted as high energy photons after inverse comptonization . in fig .
1a , we show that the rate of cooling of the cenbol in progressive days during the rising phase of the outburst .
as day progresses , amount of cooling increases and shock moves towards the black hole ( msc96 ) , which is shown in fig .
1b . on first observed day of the outburst ( mjd = 55417.2 ) , location of the shock ( @xmath42 in schwarzschild radius @xmath43 ) was at @xmath44 and at the end of our observation ( mjd = 55424.1 )
, it reaches at @xmath45 . in fig .
1c , we show mach number variation of the flow in @xmath46 day ( solid curve , shock was at @xmath44 ) , @xmath47 day ( dashed curve , shock was at @xmath48 ) and @xmath49 day ( dotted curve , shock was at @xmath50 ) of the outburst .
we calculate the velocity of movement of the shock to be @xmath51 , which roughly matches with the final velocity of shock wave calculated from the pos model fit of the qpo frequency evolution ( see , * ? ? ?
* ) . in fig . 2a
, we show variation of observed qpo frequencies with time .
if viscosity parameter @xmath0 were constant throughout the outburst then the variations of theoretically calculated qpo frequencies would be different .
dotted curve drawn for a viscosity parameter ( @xmath0 ) @xmath52 shows that qpo frequency increases at almost constant rate . the dashed curve of fig .
2a is for the effect of non - linear variation of the viscosity , which is shown in fig .
as the day progresses , viscosity adjusts in such a way that the angular momentum can produce a shock at a suitable place satisfying r - h conditions .
@xcite and @xcite , with their extensive numerical simulations showed that angular momentum distribution depends on viscosity parameter . in our solution , at the beginning of the outburst during the hard state , from mjd=55417.2 to mjd=55420.2 ( debnath et al
2013 ; mdc14 ) , @xmath0 varies from @xmath53-@xmath54 to @xmath55-@xmath54 . during the hard - intermediate state , from mjd=55421.3 to mjd=55424.1 @xcite , @xmath0 varies rapidly from @xmath56-@xmath57 to @xmath58-@xmath59 .
our @xmath0 calculation is for sub - keplerian component only . in fig .
2c , we show the variation of @xmath0 with shock location .
dashed curve shows variation from our analytical solution , whereas dotted curve is a fitted polynomial , which gives a general trend and could be used in other systems .
we see that @xmath60 @xmath61 , where , @xmath62(=@xmath63 , with asymptotic standard error @xmath64 ) and @xmath65(=@xmath66 , with asymptotic standard error @xmath67 ) are constants for this bhc .
it is to be noted that this viscosity parameter is computed for the sub - keplerian flow component only .
qpos in black hole candidates are very stable features .
they are seen day after day , though the frequency may be drifted slowly as the object goes from hard to soft state in the rising phase .
this is generally observed in most of the outbursting bhcs ( * ? ? ?
* ; * ? ? ?
* and references therein ) .
propagatory oscillating shock solution can explain such frequency drifts very well @xcite .
these phenomenological model is found to be justified when we actually compute shock drifts from radiated energy loss from a self - consistent transonic solution .
we find that in order to have rankine - hugoniot conditions satisfied on each day , viscosity parameter must be evolving too .
if the outer boundary condition is kept fixed , increase in viscosity parameter causes shock to drift outward @xcite , but if the inner boundary condition is kept fixed , the shock moves inward @xcite .
we find support of the latter phenomenon in an outbursting source where matter supply is changing and viscosity enhancement steadily brought the shock closer to the black hole .
cooling is found to rise day by day and so is @xmath0 .
such a movement of the shock increases qpo frequency as is observed .
our result establishes consistency in theoretical understanding of the observed data : as cooling increases , observed qpo frequency increases due to drifting of the shock towards black hole in a way that the cooling time scale roughly matches with the infall time scale .
this process brings the object towards the softer states as is observed .
shock locations were found to be located at the right place ( i.e. , r - h conditions are satisfied ) , only if viscosity is not strictly constant , but gradually rises from @xmath1 to @xmath2 from the first day to @xmath3 seventh day .
it is to be noted that there are alternative models @xcite where the corona is supposed to oscillate at its eigen frequency and the viscosity required in this case is around @xmath68 .
this appears to be too high as compared to what we find in the present paper .
the discrepancy could be due to the fact the latter models rely on oscillations of a keplerian disk with high angular momentum and they require higher viscosity to reduce it drastically . in our case , on the contrary , the oscillating cenbol is highly sub - keplerian to begin with .
therefore , a little viscosity is enough the transport requisite angular momentum .
this range of @xmath0 we require is in the same ball park as obtained from numerical simulations @xcite of magnetorotational instability ( mri ) .
s. mondal acknowledges the support of csir - net scholarship .
arlt , r. & rdiger , g. , 2001 , a & a , 374 , 1035 balbus , s. a. & hawley , j. f. , 1991 , apj , 376 , 214 chakrabarti , s. k. , 1989 , mnras , 340 , 7 chakrabarti , s.k . , 1990a , theory of transonic astrophysical flows " , world scientific ( singapore ) chakrabarti , s. k. , 1990b , mnras , 243 , 610 chakrabarti , s.k .
& titarchuk , l.g . , 1995 , apj , 455 , 623 chakrabarti , s. k. & molteni , d.g . 1995 , mnras , 272 , 80 chakrabarti , s. k. , acharyya , k. & molteni , d. , 2004 , astron .
astrophys . , 421 , 1 chakrabarti , s.k . , nandi , a. , debnath , d. , sarkar , r. , & datta , b. g. , 2005 , ijp , 79(8 ) , 841 ( astro - ph/0508024 ) chakrabarti , s.k . ,
debnath , d. , nandi , a. & pal , p.s . , 2008 ,
a@xmath69a , 489 , l41 chakrabarti , s.k . ,
dutta , b.g . & pal , p.s . , 2009 , mnras , 394 , 1463 das , s. & chakrabarti , s. k. 2004 , ijmpd , 13 , 1955 das , s. , chakrabarti , s.k . , & mondal , s. , 2010 , mnras , 401 , 2053 debnath , d. , chakrabarti , s. k. , & nandi , a. , et al .
, 2008 , basi , 36 , 151 debnath , d. , chakrabarti , s. k. , & nandi , a. , 2010 , a@xmath69a , 520 , 98 debnath , d. , chakrabarti , s. k. , & nandi , a. , 2013 , adspr , 52 , 2143 debnath , d. , chakrabarti , s. k. , & mondal , s. , 2014a , mnras letters , 440 , 121 debnath , d. , mondal , s. , & chakrabarti , s. k. , 2014b , mnras ( submitted ) ebisawa , k. , titarchuk , l. , & chakrabarti , s. k. , 1996 , pasj , 48 , 59 garain , s. , ghosh , h. & chakrabarti , s. k. , 2014 , mnras , 437 , 1329 giri , k. , & chakrabarti , s. k. , 2012 , mnras , 421 , 666 giri , k. , & chakrabarti , s. k. , 2013 , mnras , 430 , 2836 landau , l. d. & lifshitz , e. m. , 1959 , fluid mechanics , pergamon press , oxford masada , y. & sano , t. 2009 , iau symposium no . 259 , 121 matsumoto , r. , kato , s. , fukue , j. & okazaki , a.t . , 1984 ,
pasj , 36 , 71 mondal , s. , & chakrabarti , s. k. , 2013 , mnras , 431 , 2716 ( paper - i ) mondal , s. , debnath , d. & chakrabarti , s. k. , 2014a , apj , 786 , 4 mondal , s. , chakrabarti , s. k. & debnath , d. , 2014b , ap&ss , 353 , 223 molteni , d. , sponholz , h. & chakrabarti , s.k . , 1996 ,
apj , 457 , 805 morrison , r & mccammon , d , 1983 , apj , 270 , 119 nandi , a. , debnath , d. , & mandal , s. , et .
al . , 2012 ,
a@xmath69a , 542 , 56 remillard , r. a. & mcclintock , j. e. , 2006 , ara&a , 44 , 49 rees m. , begelman m. , blandford r. & phinney e. , 1982 , nature , 295 , 17 rodriguez , j. , varnire , p. , tagger , m. & durouchoux , ph .
, 2000 , a&a , 387 , 487 ryu , d. , chakrabarti , s. k. & molteni , d. , 1997 , apj , 474 , 378 shakura , n. i. & sunyaev , r. a. , 1973 , a&a , 24 , 337 shirakawa , a. & lai , d. , 2002 , apj , 564 , 361 steiner , j. f. , mcclintock , j. e. , & reid , m. j. , 2012 , apj , 745 , l7 titarchuk , l. , lapidus , i & muslimov , a. , 1998 , apj , 499 , 315 titarchuk , l & osherovich , v. , 2000 , apj , 542 , 111 titarchuk , l & fiorito , r. , 2004 , apj , 612 , 988 trudolyubov , s. , churazov , e. , & gilfanov , m. , 1999 , a&a , 351 , l15 yamaoka , k. , negoro , h. , & sugizaki , m. , et al . , 2010 ,
atel , 2378 , 1 | in outburst sources , quasi - periodic oscillation ( qpo ) frequency is known to evolve in a certain way : in the rising phase , it monotonically goes up till a soft intermediate state is achieved . in the propagating oscillatory shock model , oscillation of the compton cloud
is thought to cause qpos .
thus , in order to increase qpo frequency , compton cloud must collapse steadily in the rising phase . in decline phases ,
exactly opposite should be true .
we investigate cause of this evolution of the compton cloud .
the same viscosity parameter which increases the keplerian disk rate , also moves the inner edge of the keplerian component , thereby reducing the size of the compton cloud and reducing the cooling time scale .
we show that cooling of the compton cloud by inverse comptonization is enough for it to collapse sufficiently so as to explain the qpo evolution .
in the two component advective flow ( tcaf ) configuration of chakrabarti - titarchuk , centrifugal force induced shock represents boundary of the compton cloud .
we take the rising phase of 2010 outburst of galactic black hole candidate h 1743 - 322 and find an estimation of variation of @xmath0 parameter of the sub - keplerian flow to be monotonically rising from @xmath1 to @xmath2 , well within the range suggested by magneto - rotational instability .
we also estimate the inward velocity of the compton cloud to be a few meters / second which is comparable to what is found in several earlier studies of our group by empirically fitting the shock locations with the time of observations . |
let @xmath2 be a linear partial differential operator on @xmath1 with constant coefficients .
then a classical theorem of lions and titchmarsch states that , for every distribution @xmath6 on @xmath1 with compact support , the convex hulls of the supports of @xmath7 and @xmath6 are equal : @xmath8 by regularization , this is equivalent to stating that for every compactly supported smooth function @xmath4 on @xmath1 , the convex hulls of the supports of @xmath9 and @xmath4 are equal : @xmath10 in fact , lions @xcite proved a more general version of the support theorem , namely @xmath11 we refer to @xcite for an elementary proof of this theorem .
the first version ( [ eq : supportconstantpdo ] ) obviously follows from the third one ( [ eq : suppconv ] ) by taking @xmath12 , the delta distribution at the origin .
the comparison of the supports of @xmath4 and @xmath9 plays an important role at several places in analysis .
one typical situation is the study of solvability of differential operators .
recall that a linear partial differential operator @xmath13 with smooth coefficients @xmath14 is said to be _ solvable in @xmath1 _ , provided @xmath15 , that is , if for every @xmath16 the differential equation @xmath17 has a solution @xmath18 .
the following theorem gives a necessary and sufficient condition for the solvability of @xmath2 ( see @xcite , theorem 3.3 ) .
[ thm : solvable ] let @xmath2 be a linear partial differential operator with smooth coefficients in @xmath1 . then @xmath2 is solvable if and only if the following two condition are satisfied : 1 . _
( semi - global solvability ) _ for every compact subset @xmath19 of @xmath1 and for every @xmath16 there is a function @xmath18 so that @xmath17 on @xmath19 .
( @xmath2-convexity of @xmath1 ) _ for every compact subset @xmath19 of @xmath1 there is a compact set @xmath20 so that for every @xmath21 the inclusion @xmath22 implies @xmath23 . here
@xmath24 denotes the formal transpose of @xmath2 .
the support theorem of lions and titchmarsch implies that , for every linear partial differential operator @xmath2 with constant coefficients , @xmath1 is @xmath2-convex , and that we can in fact take @xmath25 .
observe that , in this case , condition ( b ) also implies that the operator @xmath24 is injective on @xmath26 .
the main result of this paper is theorem [ thm : maindistr ] , which provides an extension of the support theorem of lions and titchmarsch to specific ( but very large ) classes of invariant singular linear partial differential operators with real analytic coefficients and to distributions having a compact support with invariant convex hull .
the invariance considered here is with respect to certain finite groups of orthogonal transformations generated by reflections . in the 2-dimensional case ,
examples of such groups are the groups of symmetries of regular @xmath27-agons .
the principal symbols of the examined differential operators are allowed to vanish , but only in a precise way , along the reflecting hyperplanes .
see formula ( [ eq : symbol ] ) below .
some restrictions in generalizing the theorem of titchmarsch and lions are of course needed .
the following easy example shows that the theorem can not hold for arbitrary linear differential operators with variable coefficients , even in the one - dimensional case and with real analytic coefficients .
[ ex : xdx ] consider the differential operator @xmath28 on @xmath29 .
let @xmath30}$ ] be the characteristic function of the interval @xmath31 $ ]
. then @xmath32 , where @xmath33 denotes the delta measure at @xmath34 . therefore @xmath35 and
@xmath36 $ ] are convex and different . in the one - dimensional case , the differential operator @xmath28 of example [ ex : xdx ] belongs to the class of differential operators to which our support theorem applies . observe that @xmath2 is an even differential operator which is singular at @xmath37 . for this very special differential operator ,
our theorem states that @xmath38 if @xmath39 satisfies one of the following conditions : 1 .
@xmath400,+\infty[$ ] , 2 .
@xmath40-\infty,0[$ ] , 3 .
@xmath41 is symmetric with respect to the origin @xmath42 .
of course the distribution @xmath43}$ ] from example [ ex : xdx ] does not fulfill any of these conditions .
the core of the proof of theorem [ thm : maindistr ] is to show that the considered situation allows us to apply holmgren s uniqueness theorem for comparing the size of the supports of @xmath4 and @xmath9 when @xmath4 is a compactly supported smooth function with the property that @xmath5 is invariant .
the employ of holmgren s theorem is the reason for imposing to the coefficients of the considered differential operators to be real analytic .
several other authors employed holmgren s uniqueness theorem for proving @xmath2-convexity properties ( see e.g. @xcite , @xcite , @xcite ) .
remarkable is nevertheless , that the proof of our theorem is very elementary .
it requires only basic facts from convex geometry and an application of equation ( [ eq : suppconv ] ) .
the article consists of two parts .
the first part contains the proof of theorem [ thm : maindistr ] , which does not require any knowledge of symmetric spaces . in the second part
, we discuss several applications of our main results , including hypergeometric differential operators , bessel differential operators , shift operators , hamiltonian systems , and invariant differential operators on symmetric spaces .
the solvability of @xmath44-invariant differential operators is one of the fundamental problems in the analysis on a symmetric space @xmath45 ( see @xcite , p. 275 ) .
recall that theorem [ thm : solvable ] holds more generally if @xmath1 is replaced with a 2nd countable smooth manifold ( see @xcite , p. 14 ) .
since @xmath45 is a second countable smooth manifold , one obtains the equivalence between global and semi - global solvability , provided one can prove that @xmath45 is @xmath2-convex .
the solvability of invariant differential operators on riemannian symmetric space was proved by helgason @xcite . in the general pseudo - riemannian symmetric case ,
van den ban and schlichtkrull @xcite determined a sufficient condition for a @xmath44-invariant differential operator @xmath2 ensuring that @xmath45 is @xmath2-convex .
this condition involves the degree of the polynomial which is the image of @xmath2 under the harish - chandra isomorphism .
it is for instance always satisfied when @xmath45 is split , i.e. , has a vectorial cartan subspace .
the riemannian symmetric spaces of the noncompact type are examples of split symmetric spaces and so are the @xmath46 space of oshima and sekiguchi , @xcite . as a first application of theorem [ thm : maindistr ] , we deduce in section [ section : appl ] the @xmath2-convexity of riemannian symmetric spaces @xmath47 of noncompact type when @xmath2 is a @xmath44-invariant differential operator on @xmath47 .
our method , which is based on taking the radial component of @xmath2 along the cartan subgroup , is different from those of used in @xcite and @xcite .
support theorems play also an important role in harmonic analysis , in particular in paley - wiener type theorems .
these theorems characterize the space of functions which are image , under a suitable generalization of the fourier transform , of the compactly supported smooth functions .
applications in this setting appeared first in the work of van den ban and schlichtkrull on fourier transforms on pseudo - riemannian symmetric spaces @xcite .
the basic idea , which we shall outline more precisely in section [ section : appl ] , is to cancel the singularities appearing in a wave packet @xmath4 by applying a suitable differential operator @xmath2 .
the problem is to compare the size of the support of @xmath9 , which can be easily determined , with the hard to determine size of the support of the original wave packet @xmath4 .
in fact , our need for support properties like those stated in the present paper turned up in the proof of a paley - wiener type theorem for the @xmath48-hypergeometric transform , which is a fourier type transform related to the theory of hypergeometric functions associated with root systems . see @xcite . however , we point out that theorem [ thm : maindistr ] is stated in a very general setting and applies to many different situations .
see example [ ex : diffrankone ] , and the subsections [ ex : hyper ] , [ ex : bessel ] and [ ex : shift ] for several special cases .
our paper is organized as follows . in section [
section : setting ] we introduce the general setting in which our extension of the theorem of lions and titchmarsch will be proved . the main results , theorems [ thm : maindistr ] and [ thm : mainfunction ] , will be stated and proved in section [ section : thmsupport ] .
section [ section : examples ] presents several concrete situations where our support theorem applies .
the presented examples are related to the theory of special functions associated with root systems .
the last section is devoted to applications .
we deduce the @xmath2-convexity of riemannian symmetric spaces of noncompact type when @xmath2 is an invariant differential operator .
moreover , we describe how to employ theorem [ thm : mainfunction ] for proving of paley - wiener type theorems in the harmonic analysis on symmetric spaces and on root systems .
in this section we introduce the notation and set up that will be used in this article .
in particular we introduce the class of differential operators that will be considered in this article and give few examples . in the following @xmath49 stands for an real euclidean vector space of dimension @xmath27 , i.e. , @xmath50 .
furthermore @xmath2 will stand for a differential operator on @xmath49 with analytic coefficients .
we denote by @xmath51 a ( positive definite ) inner product on @xmath52 .
set @xmath53 . for @xmath54
we denote by @xmath55 the closed euclidean ball in @xmath52 with center @xmath42 and radius @xmath56 .
let @xmath57 denote the real dual of @xmath52 .
for each @xmath58 we denote by @xmath59 the unique element of @xmath52 satisfying @xmath60 for all @xmath61 .
we set @xmath62 and notice that @xmath63 is independent of @xmath51 . with each @xmath64
we associate the reflection @xmath65 of @xmath52 across the hyperplane @xmath66 .
thus @xmath67 a finite set @xmath68 is called a _ ( reduced ) root system _ if the following conditions holds for @xmath69 : 1 .
if @xmath70 , then @xmath71 ; 2 . if @xmath72 , then @xmath73 . the elements of @xmath69 are called _ roots_. observe that _ we are not requiring _ that @xmath69 contains a basis of @xmath57 .
in particular , our definition allows @xmath69 to be the empty set . a subset @xmath74 of @xmath69
is called a _ simple system _ if @xmath74 is linearly independent and if any root in @xmath69 can be written as a linear combination of elements in @xmath74 in which all non - zero coefficients are either all positive or all negative . if @xmath75 , then we set @xmath76 .
let a simple system @xmath74 of @xmath69 be fixed .
set @xmath77 , where @xmath78 .
the elements in @xmath79 are said to be positive .
observe that @xmath80 for all @xmath81 .
hence @xmath82 .
let @xmath83 be the group of orthogonal transformations of @xmath52 generated by the reflections @xmath84 .
it coincides with the group generated by @xmath85 .
we set @xmath86 , if @xmath69 is empty .
if @xmath87 and @xmath88 , then @xmath89 .
the group @xmath0 is a _ finite coxeter group_. conversely , every finite coxeter group originates from a root system as above .
see e.g. @xcite , chapter 1 . among the finite coxeter groups we find the _
weyl groups _
( for instance the finite groups of permutations ) and the _ dihedral groups _ ( that is , the groups of symmetries of the regular @xmath27-gons )
. the group @xmath0 acts on functions @xmath90 according to @xmath91 it also acts on compactly supported distributions @xmath92 and on differential operators @xmath2 on @xmath52 by : @xmath93 where we have written the pairing between distributions and functions by @xmath94 .
the function @xmath4 ( resp .
the compactly supported distribution @xmath6 or the differential operator @xmath2 ) is said to be _
@xmath0-invariant _ provided @xmath95 for all @xmath96 ( resp .
@xmath97 or @xmath98 for all @xmath96 ) .
for instance , if @xmath87 and @xmath99 , then @xmath0-invariant means even .
more generally , let @xmath100 be a character of @xmath0 .
then @xmath4 is said _ to transform under @xmath0 according to @xmath100 _ if @xmath101 for all @xmath96 .
this definition extends similarly to distributions and differential operators .
notice that if @xmath4 transforms under @xmath0 according to a character @xmath100 , then the support of @xmath4 is @xmath0-invariant .
we will be studying differential operators with leading symbol of a specific form .
we will therefore need the following definition : let @xmath102 be a non - empty @xmath0-invariant convex open subset of @xmath52 , and @xmath103 a polynomial function .
we say that @xmath104 is a _
homogeneous @xmath102-elliptic polynomial _ if the following holds : 1 .
@xmath104 is a homogeneous polynomial in @xmath105 with real analytic coefficients on @xmath52 , i.e. there is @xmath106 so that @xmath107 where @xmath108 is real analytic , and @xmath109 , if @xmath110 is a multi - index .
2 . if @xmath111 and @xmath112 , then @xmath113 . let @xmath2 be a @xmath0-invariant linear partial differential operator on @xmath52 with real analytic coefficients .
we say that @xmath2 is _
@xmath102-elliptic _ if its principal symbol is of the form @xmath114^{n({\alpha } ) } \right ) p(x,{\lambda } ) , \qquad ( x,{\lambda})\in \frak a \times \frak a^*,\ ] ] where @xmath115 is a homogeneous polynomial , @xmath116 for all @xmath70 , and @xmath117 is a homogeneous @xmath102-elliptic polynomial . in this paper
we consider @xmath102-elliptic linear partial differential operators as operators acting on functions or distributions on @xmath102 .
note that the decomposition of @xmath118 in ( [ eq : symbol ] ) is in general not unique : if @xmath119 for all @xmath120 , then we can replace @xmath104 by @xmath121 , @xmath122 , and replace @xmath123 by @xmath124 .
the class of @xmath102-ellipltic contains all elliptic partial differential operators on @xmath49 by taking @xmath125 and @xmath75 .
but , more generally , the principal symbols of the considered operators are allowed to vanish at the zeros of @xmath115 as well as along the hyperplanes @xmath126 with @xmath70 .
observe that the condition in ( [ eq : symbol ] ) imposes a restriction only on the principal part of the differential operators .
in particular , suppose @xmath127 , @xmath128 are @xmath0-invariant linear partial differential operators with real analytic coefficients so that @xmath129 .
if @xmath127 satisfies condition ( [ eq : symbol ] ) , then the same is true for @xmath130 . any partial differential operator with constant coefficients @xmath131 satisfies ( [ eq : symbol ] ) when we choose @xmath132 and @xmath75 .
indeed , in this case we do not impose any symmetry condition , and @xmath133 is of the form ( [ eq : symbol ] ) .
[ ex : diffrankone ] suppose @xmath134 and @xmath88 .
we shall identify @xmath135 with @xmath29 .
the possible subsets @xmath102 are the open intervals of the form @xmath136-r , r[$ ] with @xmath137 . then the differential operators considered are the even ordinary differential operators with real analytic coefficients and with principal symbol @xmath138 here , @xmath139
, the polynomial @xmath115 is homogeneous , and @xmath117 is homogeneous @xmath102-elliptic . for @xmath140 , examples are @xmath141 and the euler operator @xmath142 where @xmath143 .
other examples , which are relevant in harmonic analysis , are constructed from the jacobi and the bessel differential operators on @xmath1360,+\infty[$ ] .
let @xmath144 and @xmath145 respectively denote the jacobi and the bessel differential operators , i.e. @xmath146 \frac{d}{dx}\ , , \\
l_0&=\frac{d^2}{dx^2}+ \big[a \frac{1}{x}+b \frac{1}{2x}\big ] \frac{d}{dx}\ , . \end{aligned}\ ] ] then , for specific positive integral values of the constants @xmath147 and @xmath148 , the operator @xmath144 agrees with the radial part of the laplace - beltrami operator of a rank - one riemannian symmetric spaces @xmath47 of noncompact type , whereas @xmath145 is the radial part of the laplace - beltrami operator of the corresponding rank - one riemannian symmetric space @xmath149 of euclidean type .
examples of such spaces @xmath47 are the ( real , complex or quaternionic ) hyperboloids ; examples of spaces @xmath149 are given by @xmath1 regarded as homogeneous space @xmath150 .
we refer to @xcite , ch .
ii , propositions 3.9 and 3.13 , for more information on radial parts of the laplace - beltrami operators .
the operators @xmath151 and @xmath152 satisfy our assumptions since they are even operators on @xmath29 with real analytic coefficients , and @xmath153 and @xmath154 . generalizations in more variables of these examples will be treated in section [ section : examples ] .
note that , in the first case , we can take @xmath155 or @xmath156 . [ ex : calogero ] the calogero model is a non - relativistic quantum mechanical system of @xmath157 identical particles on a line interacting pairwise .
such a system is described by the hamiltonian @xmath158 where the positive constant @xmath159 is the coupling coefficient .
the associated schrdinger operator is @xmath160 see e.g. @xcite , ( 3.1.1 ) and ( 3.1.14 ) , i. see also @xcite .
let @xmath161 be the standard basis of @xmath162 and let @xmath51 be the standard inner product on @xmath1
. for @xmath163 with @xmath164 set @xmath165 .
then @xmath166 is a root system of type @xmath167 .
we take @xmath168 as a set of positive roots .
the corresponding system of simple roots is @xmath169 .
the finite coxeter group @xmath0 associated to @xmath167 is the group @xmath170 of permutations of the set @xmath161 .
we can now write @xmath171 where @xmath172 is the euclidean laplace operator on @xmath52 .
it follows that @xmath173 is @xmath0-invariant .
if @xmath174 , then the differential operator @xmath175 satisfies our requirements in ( [ eq : symbol ] ) .
this example will be generalized in section [ section : examples ] .
multiplication by @xmath176 and @xmath177 in example [ ex : diffrankone ] ( resp . by @xmath178 in example [ ex : calogero ] ) , which is used to obtain @xmath0-invariant differential operators with analytic coefficients in @xmath52 , is in fact inessential when dealing with even ( resp .
@xmath0-invariant functions ) .
indeed , despite the singularity of their coefficients , the operators @xmath144 and @xmath145 map even smooth functions on @xmath29 in smooth functions .
likewise , the operator @xmath173 maps smooth @xmath0-invariant functions on @xmath52 into smooth functions .
this remark will be stated , in a more general form , in section [ section : examples ] .
we now introduce the class of functions that will be considered in this paper . as above , let @xmath102 be a fixed @xmath0-invariant open convex subset of @xmath52 .
let @xmath179 denote the space consisting of distributions @xmath6 on @xmath52 so that @xmath41 is a @xmath0-invariant compact subset of @xmath102 , and let @xmath180 be the subspace of @xmath179 consisting of @xmath181 functions .
important subspaces of @xmath180 are the spaces @xmath182 , formed by the smooth compactly supported functions @xmath183 which transform under @xmath0 according to the character @xmath100 of @xmath0 . for instance , if @xmath100 is the trivial character , then we obtain the space @xmath184 of @xmath0-invariant smooth functions on @xmath102 with compact support .
similarly , inside @xmath179 we find the spaces @xmath185 of compactly supported distributions on @xmath102 which transform according to @xmath100 , and the space @xmath186 of @xmath0-invariant distributions on @xmath0 with compact support .
if @xmath87 and @xmath132 , then we can identify @xmath187 with @xmath29 .
suppose first @xmath75 .
then @xmath0 is trivial , and @xmath188 ( resp .
@xmath189 ) reduces to the space of @xmath181 functions ( resp .
distributions ) on @xmath29 with compact support . if @xmath88 , then @xmath89 . in this case ,
@xmath188 ( resp .
@xmath189 ) is the space of smooth functions ( resp .
distributions ) @xmath4 on @xmath29 so that @xmath5 is a bounded interval of the form @xmath190 $ ] for some @xmath191 .
the subspace @xmath192 consists of the compactly supported even smooth functions on @xmath29 ; with @xmath100 equal to the sign character , the subspace @xmath193 consists of the compactly supported odd smooth functions on @xmath29 .
in this section we proof the main result of this article .
this is the following version of the the support theorem of lions and titchmarsch : [ thm : maindistr ] let @xmath194 be open , convex and @xmath0-invariant .
suppose @xmath2 is a @xmath0-invariant linear partial differential operator on @xmath52 with real analytic coefficients and with principal symbol of the form ( [ eq : symbol ] ) .
then @xmath195 for each @xmath196 .
we shall prove below that , by regularization , theorem [ thm : maindistr ] is equivalent to the following smooth version : [ thm : mainfunction ] let @xmath102 and @xmath2 be as in theorem [ thm : maindistr ] .
then @xmath197 for each @xmath198 .
equivalently , for each @xmath198 and for every compact convex @xmath0-invariant subset @xmath199 we have @xmath200 _ proof of the equivalence of theorems [ thm : maindistr ] and [ thm : mainfunction ] .
_ it is clear that theorem [ thm : maindistr ] implies theorem [ thm : mainfunction ] . for the other direction ,
let @xmath196 .
let @xmath201 be an approximate identity with @xmath202 for all @xmath54 .
since @xmath102 is a @xmath0-invariant , we can , if necessary , replace @xmath203 by @xmath204 .
we can therefore suppose that @xmath203 is @xmath0-invariant .
it follows , in particular , that @xmath205 and hence @xmath206 are @xmath0-invariant .
thus , by ( [ eq : suppconv ] ) , @xmath207 is a @xmath0-invariant compact subset of @xmath52 .
since @xmath102 is open , we can choose @xmath208 so that @xmath209 for @xmath210 .
the functions @xmath211 converge to @xmath6 in @xmath212 as @xmath213 , and @xmath214 .
it is easy to check ( without invoking deeper support theorems ) that there exists @xmath215 , converging to @xmath42 as @xmath213 , so that @xmath216 replacing @xmath6 with @xmath7 , we obtain @xmath217 the same inclusions are preserved for the convex hulls . by theorem [ thm : mainfunction ] , we have @xmath218 hence , @xmath219 which implies that @xmath220 . similarly , @xmath221
. this proves theorem [ thm : maindistr ] . before going into the details of the proof of theorem [ thm : mainfunction ] ,
let us briefly explain the main ideas involved .
since @xmath222 , it suffices to show that if @xmath223 is a compact convex @xmath0-invariant subset of @xmath102 , then @xmath224 implies that @xmath225 .
this will be proved by contradiction .
the main tool will be holmgren s uniqueness theorem .
[ thm : holmgren ] let @xmath226 be open , and let @xmath227 be a real valued function in @xmath228 .
let @xmath2 be a linear partial differential operator with analytic coefficients defined in @xmath229 .
let @xmath230 denote the principal symbol of @xmath2 .
suppose that @xmath231 is a point in @xmath229 such that @xmath232 then there exists a neighborhood @xmath233 of @xmath231 with the following property : if the distribution @xmath234 is annihilated by @xmath2 , i.e. , @xmath235 , and @xmath6 vanishes on the set @xmath236 , then @xmath6 must also vanish in @xmath237 .
this is theorem 5.3.1 in @xcite . for non - elliptic differential operators
the delicate matter is to choose points @xmath231 for which condition ( [ eq : nonvanishingsymb ] ) is fulfilled .
this is exactly the kind of difficulty one encounters in the proof of support theorems .
coming back to the streamline ideas of the proof of theorem [ thm : mainfunction ] , let @xmath223 be as above .
. then @xmath239 . to reach a contradiction , we assume that @xmath240 .
let @xmath241 .
as @xmath223 is convex and compact , there exists a hyperplane strictly separating @xmath231 and @xmath223 , i.e. , we can find @xmath242 such that @xmath243 without loss of generality we can also assume that @xmath244 otherwise we translate the hyperplane to the boundary of @xmath245 in the direction opposite to @xmath223 . in this way , the entire set @xmath245 lies inside a closed half - space supported by the hyperplane @xmath246 our plan is to apply holmgren s uniqueness theorem to @xmath247 , @xmath248 and @xmath249 .
note that in this case @xmath250 is constant and non - zero .
observe also that @xmath251 on @xmath229 and @xmath252 on the side of @xmath253 not containing @xmath223 ( which is described by the equation @xmath254 ) .
if the principal symbol of @xmath2 were not zero at @xmath255 , then all assumptions would be satisfied , and we could conclude that @xmath252 in a neighborhood of @xmath231 .
this would yield the required contradiction because @xmath256 . since @xmath257 ,
the condition @xmath258 is equivalent to @xmath259^{n({\alpha})}\neq 0.\ ] ] this might not be satisfied by the chosen pair @xmath255 .
it is even possible to have a situation where there is no choice of @xmath260 for which the above procedure could guarantee that @xmath261 .
figure 1 sketches an example in which this problem arises because of the `` corner '' at the boundary of @xmath245 .
note that the set @xmath245 in this example is also convex and invariant with respect to the group @xmath262 , where @xmath263 denotes the reflection with respect to the @xmath264-axis . in case @xmath265^{n({\alpha})}=0
$ ] , the above procedure must be modified .
the first step is to show that it suffices to consider the case of smooth functions @xmath4 with the property that the convex hull @xmath266 of @xmath238 is @xmath0-invariant and has @xmath267 boundary .
the point is that the @xmath231 , selected as above , will always belong to the boundary @xmath268 of @xmath266 , and that @xmath253 is a supporting hyperplane for @xmath266 . for the modified procedure
we need some preparations
. for @xmath54 let @xmath269 denote the closed ball centered at the origin and with radius @xmath56 .
let @xmath270 be the ( closed ) @xmath56-neighborhood of @xmath223 .
[ lemma : smoothbdry ] let @xmath223 be a proper convex subset of @xmath271 with nonempty interior , and let @xmath54 .
then @xmath272 is an @xmath27-dimensional convex subset of @xmath52 and its boundary @xmath273 is a @xmath267 @xmath274-dimensional submanifold of @xmath52 .
this is satz 17.6 in @xcite .
[ lemma : withsmoothbdry ] let @xmath2 be a differential operator on @xmath52 .
suppose there exists a smooth compactly supported function @xmath275 on @xmath49 and a compact convex subset @xmath276 with nonempty interior such that @xmath277 then there exists a function @xmath278 and a compact convex subset @xmath279 with nonempty interior such that @xmath280 moreover , the boundary @xmath281 of the convex hull of @xmath282 is a @xmath267-manifold .
if @xmath283 , where @xmath102 is a @xmath0-invariant open convex subset of @xmath52 , and @xmath284 is @xmath0-invariant . then we can choose @xmath198 and @xmath285 to be @xmath0-invariant .
let @xmath286 be an approximate identity with @xmath287 for all @xmath56 .
then @xmath288 and @xmath289 notice that ( [ eq : suppconv ] ) applied to compactly supported smooth functions implies that for all @xmath290 we have @xmath291 hence , there exists @xmath208 so that @xmath292 . otherwise ( [ eq : supp ] ) would imply that @xmath293 for all @xmath54 , and hence @xmath294 . as in ( [ eq : supp ] ) , we have @xmath295 therefore , by lemma [ lemma : smoothbdry ] , @xmath296 has @xmath267-boundary
. we can thus select @xmath297 and @xmath298 .
finally , suppose that @xmath299 and @xmath300 are @xmath0-invariants subset of @xmath102 .
since @xmath102 is open , then , by choosing a sufficiently small @xmath208 , so are also @xmath301 and @xmath302 . in the following
we shall suppose that @xmath4 and @xmath223 are chosen as in lemma [ lemma : withsmoothbdry ] .
as before , we set @xmath238 and @xmath303 .
we also fix @xmath304 .
we now proceed to the selection of the pair @xmath255 .
let @xmath305 be fixed , and let @xmath306 then @xmath307 is a nonempty open subset of @xmath57 .
the separation properties of compact convex sets ensure that @xmath308 .
observe that @xmath309 hence @xmath3100,+\infty[)$ ] where @xmath311 .
this proves the lemma because @xmath312 is proper and convex , hence continuous ( see @xcite , theorems 5.3 and 10.1 ) .
since @xmath307 is open and nonempty we can choose @xmath313 with the following properties : 1 .
@xmath314 , 2 .
@xmath315 for all @xmath70 ( i.e. @xmath316 for all @xmath70 ) .
since @xmath317 , [ le : cofx0 ] let @xmath260 be as above
. then there exists @xmath318 such that @xmath319 .
furthermore @xmath320 and @xmath321 .
finally @xmath322 this follow as @xmath266 is the convex hull of @xmath245 . when @xmath88 , the @xmath0-invariance of the situation plays a role because of the following lemma .
[ lemma : tangent ] suppose @xmath266 is a @xmath0-invariant compact convex subset of @xmath52 with nonempty interior and @xmath267-boundary .
let @xmath323 and @xmath324 such that @xmath325 .
let @xmath326 denote the tangent space to @xmath327 at @xmath328 ( regarded as subspace of @xmath52 ) and let @xmath63 be as in ( [ eq : xa ] ) . then @xmath329 .
the reflection @xmath65 of @xmath52 across the hyperplane @xmath330 maps @xmath266 onto itself and fixes @xmath328 .
hence it maps @xmath331 ( resp .
@xmath327 ) onto itself .
the derived involution @xmath65 is therefore an automorphism of @xmath326 .
hence @xmath332 , where @xmath333 is the outer normal vector to @xmath327 at @xmath328 .
since @xmath334 by invariance of @xmath266 under @xmath65 , we conclude that @xmath335 , i.e. @xmath336 .
thus @xmath337 .
[ prop : nonzerosym ] let @xmath231 and @xmath260 be as in lemma [ le : cofx0 ]
. then @xmath255 satisfies ( [ eq : nonzerosym ] ) . the element @xmath260 has been chosen so that @xmath314 .
this suffices to prove ( [ eq : symbol ] ) when @xmath338 . if @xmath88 , then it remains to show that @xmath339 for all @xmath70 . since @xmath319 , the hyperplane @xmath340 is the supporting hyperplane for @xmath341 at @xmath231 .
the set @xmath266 is @xmath0-invariant and its boundary @xmath327 is @xmath267 .
hence @xmath342 .
if @xmath343 for some @xmath70 , then @xmath344 by lemma [ lemma : tangent ] , i.e. @xmath345 . thus @xmath346 , which implies @xmath347 , against our choice of @xmath260 .
thus @xmath339 for all @xmath81 .
_ proof of theorem [ thm : mainfunction ] .
arguing by contradiction , we assume that there exists @xmath348 and a @xmath0-invariant subset @xmath223 of @xmath102 as in lemma [ lemma : withsmoothbdry ] .
we select @xmath255 as in proposition [ prop : nonzerosym ] . set @xmath247 and @xmath248 .
hence @xmath349 is a supporting hyperplane for @xmath350 , and @xmath250 is constant and non - zero .
moreover , @xmath351 on @xmath229 and @xmath252 on the side of @xmath253 not containing @xmath223 ( which is described by the equation @xmath352 ) .
proposition [ prop : nonzerosym ] ensures that @xmath258 .
holmgren s uniqueness theorem then implies that @xmath252 in a neighborhood of @xmath231 .
this gives the the required contradiction because @xmath353 . before concluding this section we prove some immediate consequences of theorem [ thm : maindistr ] .
recall that the transpose of the partial differential operator @xmath2 given by ( [ eq : d ] ) is @xmath354 hence @xmath24 belongs to the class of invariant differential operators considered in this papers if so does @xmath2 . in fact , the principal symbols of @xmath2 and @xmath24 are linked by the relation @xmath355 [ cor : maindistr ] let @xmath2 be a @xmath0-invariant differential operators @xmath2 on @xmath52 with real analytic coefficients and with principal symbol of the form ( [ eq : symbol ] ) .
then the following properties are true . 1 .
@xmath2 is injective on @xmath180 .
2 . for all @xmath198 we have @xmath356
in this section we present some examples linked to the theory of special functions associated with root systems . in these examples
the group @xmath0 is a parabolic subgroup of a fixed coxeter group @xmath357 acting on @xmath358 , and the differential operator @xmath2 is in fact invariant under the larger finite coxeter group @xmath357 .
the general situation corresponds to a ( not necessarily reduced ) root systems @xmath359 .
a _ root system _ is a finite set @xmath360 satisfying condition ( r2 ) of section [ section : setting ] . in this section
we will also assume that @xmath359 satisfies the following additional conditions : 1 .
@xmath359 spans @xmath57 ; 2 .
@xmath359 is _ crystallographic _ , that is @xmath361 for all @xmath362 .
crystallographic root systems arise naturally in several places in algebra and analysis . in particular , they are relevant in the theory of real lie algebras , lie groups , and symmetric spaces . if @xmath359 is a root system , then @xmath363 is a reduced root system according to the definition of section [ section : setting ] .
the finite coxeter group associated to @xmath359 is the finite coxeter group @xmath357 associated to @xmath69 .
it is also called the _ weyl group _ of @xmath359 .
multiplicity function _ is a @xmath357-invariant function @xmath364 . for @xmath365
, we adopt the common notation @xmath366 to denote @xmath367 .
the set @xmath368 of positive roots and the set @xmath74 of positive simple roots can be defined as in the case of reduced root systems . because of ( r0 ) , @xmath74 is a basis of @xmath57 .
fix a @xmath74 set of simple roots in @xmath368 . for each subset @xmath48 of @xmath74
we define @xmath369 as the set of elements of @xmath359 which are linear combinations of elements from @xmath48 , i.e. , @xmath370 .
it is itself a root system , but in general it does not satisfy ( r0 ) .
we denote the corresponding finite coxeter group by @xmath371 .
note that @xmath372 is generated by the reflections @xmath65 with @xmath373 .
for instance , @xmath374 and @xmath375 .
we also set @xmath376 for the set of positive roots inside @xmath369 .
the coxeter group @xmath371 will play the role of the group @xmath0 of the previous sections . recall that a subgroup @xmath0 of @xmath357 is called a _
parabolic subgroup _ if it is of the form @xmath371 for some @xmath377 .
the parabolic subgroups can also be characterized as those subgroups of @xmath357 that stabilize a subspace of @xmath52 . thus @xmath378 is a parabolic subgroup if and only if there exists a subspace @xmath379 such that @xmath380 a _ chamber _ in @xmath52 a connected component of @xmath381 ( see @xcite , ch .
> from now on we fix the chamber @xmath382 .
it is an open simplicial cone with vertex @xmath42 ( _ loc .
v , 3.9 , proposition 7 ( iii ) ) , and its closure @xmath383 is a fundamental domain for the action of @xmath357 on @xmath52 ( _ loc .
v , 3.3 , theorem 2 ) .
we define @xmath384 where @xmath385 denotes the interior . for instance , @xmath386 and @xmath387 . in the following examples , the sets @xmath388 will play the role of the set @xmath102 appearing in theorem [ thm : maindistr ] .
the set @xmath389 is the smallest cone in @xmath52 which is closed , @xmath371-invariant and contains @xmath390 .
it is a union of closed chambers .
the open polyhedral cone in @xmath57 @xmath391 has the dual cone @xmath392 @xmath393 is a closed convex cone .
it is the intersection of the closed hyperplanes defined by roots , and hence a union of closed chambers in @xmath52 .
[ lemma : cone ] we have @xmath394 consequently , @xmath395 is a closed @xmath371-invariant convex cone in @xmath52 and also its interior @xmath388 is convex .
moreover , @xmath396 this was proven in @xcite , lemma 3.4 , when @xmath0 is a weyl group .
the same proof applies also to the more general case of finite coxeter groups .
specializing the notation of section [ section : setting ] to this context , we consider the space @xmath397 of smooth functions @xmath398 with the property that @xmath5 is a @xmath371-invariant compact subset of @xmath388 .
its subspace of @xmath371-invariant functions on @xmath388 with compact support is @xmath399 .
furthermore , @xmath400 is the space consisting of distributions @xmath6 on @xmath52 so that @xmath41 is a @xmath371-invariant compact subset of @xmath388 . as before let @xmath358 be a finite dimensional euclidean space and let @xmath359 be a ( non - necessarily reduced ) root system in @xmath57 .
further , let @xmath357 the be the corresponding weyl group and let @xmath401 be a multiplicity function on @xmath359 . heckman and opdam associated with such a triple @xmath402 a commutative family of @xmath357-invariant differential operators on @xmath52 , the _ hypergeometric differential operators _ , having meromorphic coefficients of a specific type . more precisely , their coefficients are meromorphic functions on the complexification @xmath403 of @xmath52 and their singularities are cancelled by multiplication by a suitable power of the weyl denominator @xmath404 we refer to @xcite , @xcite and @xcite for more information on heckman - opdam s theory of hypergeometric differential operators . for special values of @xmath401 , the triple @xmath402 arises from a riemannian symmetric space @xmath47 of the noncompact type , i.e. , @xmath44 is a noncompact connected semisimple lie group with finite center and @xmath405 is a maximal compact subgroup , see section [ section : appl ] . in this case
, the hypergeometric differential operators coincide with the radial parts ( with respect to the @xmath19-action ) on @xmath406 of the @xmath44-invariant differential operators on @xmath47 . here
we identify @xmath406 with its diffeomorphic image in @xmath47 under the exponential map ( usually denoted @xmath407 in the literature on analysis on symmetric spaces ) .
for instance , the hypergeometric differential operator @xmath408 coincides with the radial part of the laplace - beltrami operator of @xmath47 .
notice that @xmath144 is singular on the hyperplanes @xmath409 , @xmath410 .
furthermore , @xmath144 is the multivariable analog of the jacobi differential operator of example [ ex : diffrankone ] . in ( [ eq : l ] ) , @xmath411 denotes the directional derivative in the direction of @xmath412 , and @xmath172 is the laplace operator on the euclidean vector space @xmath52 , that is , @xmath413 where @xmath414 is an orthonormal basis of @xmath52 . set @xmath415 with the usual convention that empty products are equal to @xmath34 .
we write @xmath416 instead of @xmath417 .
finally , define @xmath418 .
the principal symbol of @xmath2 is @xmath419 where @xmath420 ^ 2 \big[\frac{{{\delta}_{\scriptscriptstyle{\theta}}}(x)}{{{\pi}_{\scriptscriptstyle{\theta}}}(x)}\big]^2.\ ] ] because of lemma [ lemma : cone ] , each @xmath421 is positive on @xmath388 .
therefore @xmath422 is a homogeneous @xmath388-elliptic polynomial on @xmath423 ( of degree @xmath42 in @xmath424 ) .
more generally , for each hypergeometric differential operator @xmath425 there is @xmath426 so that the linear partial differential operator @xmath427 is a @xmath357-invariant differential operator with real analytic coefficients and with principal symbol of the form ( [ eq : symbol ] ) where @xmath422 is a homogeneous @xmath388-elliptic polynomial . in this case , we say that @xmath428 is a _ regularization _ of @xmath425 .
the nature of the principal symbol of @xmath2 can in fact be deduced by the explicit representation of the hypergeometric differential operators in terms of cherednik operators ( see e.g. @xcite ) . according to this representation ,
each hypergeometric differential operator can be constructed as the `` differential part '' of the differential - reflection operator associated with a @xmath357-invariant polynomial in the symmetric algebra over @xmath403 by means of cherednik operators .
if @xmath425 is associated with the @xmath357-invariant polynomial @xmath429 , then the principal part of the regularized operator @xmath430 is @xmath431^{2k}\,\\ & = p_h({\lambda } ) { { \pi}_{\scriptscriptstyle{\theta}}}(x)^{2k } p({\lambda},x)\end{aligned}\ ] ] where @xmath432 is the highest homogeneous part of @xmath115 and @xmath433^{2k } \big[\frac{{{\delta}_{\scriptscriptstyle{\theta}}}(x)}{{{\pi}_{\scriptscriptstyle{\theta}}}(x)}\big]^{2k}.\ ] ] in this setting , theorems [ thm : maindistr ] and [ thm : mainfunction ] yield the following result .
[ cor : mainhyp ] let the notation be as above .
let @xmath434 .
for a hypergeometric differential operator @xmath425 let @xmath430 be a regularization of @xmath425 . then for every @xmath435 we have @xmath436 equivalently , for every @xmath437
, we have @xmath438 in the case of @xmath371-invariant functions , corollary [ cor : mainhyp ] can be stated directly for the hypergeometric operators .
[ cor : mainhypw ] let the notation be as above .
let @xmath425 be a hypergeometric differential operator and @xmath439 .
then @xmath440 for all @xmath441 and @xmath442 as @xmath443 and @xmath444 is compact and @xmath371-invariant , it follows that there exists a unique @xmath445 so that @xmath446 .
let @xmath447 be a regularization of @xmath425 .
then @xmath448 suppose @xmath425 is the differential part of the cherednik operator @xmath449 ( see e.g. @xcite ) .
then @xmath450 by @xmath357-invariance
. cherednik operators map @xmath357-invariant smooth functions into @xmath357-invariant smooth functions .
it follows that @xmath451 is smooth and @xmath357-invariant .
> from ( [ eq : restrictions ] ) we therefore deduce that @xmath452 extends to be smooth and @xmath357-invariant on @xmath388 .
if @xmath453 is a continuous function , then @xmath454 and @xmath453 have the same support .
the theorem therefore follows .
let @xmath44 be a connected noncompact semisimple lie group with fintie center and @xmath405 a maximal compact subgroup .
then @xmath455 for some cartan involution @xmath456 .
the lie algebra @xmath457 decomposes into eigenspaces of the derived homomorphism @xmath458 : @xmath459 where @xmath460 is the @xmath461 eigenspace and @xmath462 is the @xmath463 eigenspace . consider @xmath462 as a abelian lie group .
the group @xmath19 acts linearly on @xmath462 by @xmath464 .
we can therefore consider the semi - direct product @xmath465 let @xmath402 be as in example [ ex : hyper ] .
the _ bessel differential operators _ associated with @xmath402 are the `` rational '' analogs of the hypergeometric differential operators .
they are @xmath0-invariant and the singularities of their meromorphic coefficients are cancelled by multiplication be a power of the polynomial @xmath466 when the triple @xmath402 arises from the structure of riemannian symmetric space @xmath47 of the noncompact type , the bessel differential operators coincide with the radial parts on @xmath406 of the @xmath467-invariant differential operators on the corresponding riemannian symmetric space @xmath149 of euclidean type .
for instance , the bessel differential operator @xmath468 coincides with the radial part of the laplace - beltrami operator on @xmath149 .
the operator @xmath145 is the multivariable analog of the bessel differential operator of example [ ex : diffrankone ] .
the bessel differential operators can be constructed as `` differential parts '' of differential - reflection operators associated with @xmath0-invariant polynomials functions @xmath429 on @xmath469 by means of the dunkl operators @xcite .
suppose @xmath425 is the bessel differential operator associated with @xmath429 .
let @xmath426 be chosen so that @xmath470 has real analytic coefficients .
then the principal symbol of @xmath2 is @xmath471 where @xmath432 is the highest homogeneous part of @xmath115 and @xmath472 is a homogeneous @xmath388-elliptic polynomial of degree @xmath42 in @xmath473 . as for
the hypergeometric differential operators , we obtain the following corollary of theorems [ thm : maindistr ] and [ thm : mainfunction ] . [ cor : mainbessel ] let the notation be as above .
let @xmath48 be a fixed set of positive simple roots in @xmath359 .
for a bessel differential operator @xmath425 let @xmath474 be a regularization of @xmath425 . then ,
for @xmath435 we have @xmath436 equivalently , for every @xmath437 , we have @xmath438 if @xmath441 , then @xmath440 and @xmath442 for the study of hypergeometric differential operators corresponding to different multiplicity functions ( see example [ ex : hyper ] ) , opdam introduced certain @xmath0-invariant differential operators , called _
shift operators_. for simplicity , we treat here only the case of a reduced root system @xmath359 and we refer the reader to @xcite , part i , ch . 3 for the general case . as in the case of hypergeometric differential operators , opdam s shift operators
have meromorphic coefficients with singularities along the root hyperplanes , and their singularities are cancelled by multiplication by powers of @xmath69 .
a _ shift _ is a multiplicity function @xmath475 which is even ( i.e. such that @xmath476 for all @xmath365 ) .
opdam s shift operators exists for any shift @xmath475 .
they can be considered as generalizations of the hypergeometric differential operators , which form the space @xmath477 of shift operators of shift @xmath42 . for an arbitrary nonzero shift @xmath475 ,
the space @xmath478 of shift operators of shift @xmath475 is a free rank - one ( right ) @xmath477-module generated by a certain shift operator @xmath479 .
see _ loc .
_ , theorem 3.3.7 .
suppose furthermore that either @xmath480 for all @xmath481 or @xmath482 for all @xmath481 .
then @xmath479 is obtained by composition of _ fundamental shift operators _ @xmath483 , shifting of @xmath484 the multiplicities along each weyl group orbit ( _ loc .
cit . _ , p. 42 ) .
the principal symbol of each fundamental shift operator is @xmath485 where @xmath486 and @xmath487 .
see _ loc .
_ , remark 3.3.8 .
the analysis done in example [ ex : hyper ] for the hypergeometric differential operators , in particular corollary [ cor : mainhyp ] , easily extends also to the fundamental shift operators , and hence to all shift operators .
a similar argument can be also applied to the shift operators associated with the bessel differential operators . for more information on the latter we refer to @xcite , where they are studied in the general case
in which @xmath0 is an arbitrary finite coxeter group .
a wide class of integrable hamiltonian systems associated with root systems were introduced by olshanetsky and perelemov in @xcite .
let @xmath488 be a triple consisting of a finite dimensional euclidean space @xmath52 , a root system @xmath359 in @xmath57 , and a real - valued multiplicity function @xmath453 on @xmath359 .
these integrable systems are described by a hamiltonian of the form @xmath489 with potential energy @xmath490 where the function @xmath264 has five possible forms .
here we consider the cases i , ii and v as listed in @xcite , ( 3.1.14 ) and ( 3.8.3 ) : @xmath491 the calogero hamiltonian of example [ ex : calogero ] is a special instance of case i. the associated schrdinger operators are the linear partial differential operators @xmath492 they are invariant with respect to the weyl group @xmath357 of @xmath359 .
the operator @xmath245 can be regularized by multiplication by the polynomial @xmath416 of ( [ eq : pi ] ) in the cases i and v , and by multiplication by @xmath493 as in ( [ eq : delta ] ) .
the possibility of applying theorems [ thm : maindistr ] and [ thm : mainfunction ] to these regularized differential operators can be proven as in examples [ ex : hyper ] and [ ex : bessel ] .
in fact , there is a close relation between the operator @xmath245 of case ii , resp . of case i , and the operator @xmath144 of ( [ eq : l ] ) , resp . to the operator @xmath145 of ( [ eq : lo ] ) .
see e.g. @xcite , theorem 2.1.1 , and @xcite .
for an overview on the role of the hamiltonian systems treated in this example in different areas of theoretical physics and mathematics , we refer the reader to @xcite .
in this section we apply theorem [ thm : mainfunction ] to differential operators on symmetric spaces .
in particular we give a new proof of the @xmath2-convexity of the riemannian symmetric space @xmath47 , c.f .
our proof uses theorem [ thm : mainfunction ] , applied to the radial part of invariant differential operators and is , as far as we know , new .
our notation is the same as in subsection [ ex : bessel ] .
let us start by recalling the notation from subsection [ ex : bessel ] . here
@xmath44 is a connected noncompact semisimple lie group with finite center .
furthermore @xmath494 is a cartan involution and @xmath455 the corresponding maximal compact subgroup .
denote by @xmath495 , the natural projection @xmath496 .
we denote also by @xmath456 the derived involution on @xmath457 .
then @xmath497 , where @xmath460 is the @xmath461-eigenspace of @xmath456 , and @xmath462 the @xmath463-eigenspace .
then @xmath460 is the lie algebra of @xmath19 .
fix a cartan subspace @xmath498 , that is a maximal abelian subspace of @xmath499 .
the killing form on @xmath457 defines an euclidean inner product on @xmath52 . for @xmath500 set @xmath501={\alpha}(x ) y$ for all $ x \in \frak a$}\,\}$ ] .
then the set @xmath359 consisting of all @xmath502 for which @xmath503 is a ( generally non - reduced ) root system as defined in section [ section : examples ] .
it is called the _ ( restricted ) root system _ of @xmath504 .
the multiplicity @xmath366 of @xmath365 is defined as the dimension of @xmath505 .
the map @xmath401 given by @xmath506 is a multiplicity function on @xmath359 .
this construction associates with the riemannian symmetric space @xmath47 a triple @xmath507 . as before we denote by @xmath357 the corresponding weyl group .
let @xmath368 be a choice of positive roots and let @xmath508 be the corresponding positive chamber . denote by @xmath509 the exponential map .
then @xmath510 is an abelian subgroup of @xmath44 diffeomorphic to @xmath52 .
we set @xmath511 .
the map @xmath512 is surjective and the @xmath513-component is unique up to conjugation by an element of @xmath357 .
hence every @xmath19-bi - invariant subset of @xmath44 is of the form @xmath514 where @xmath515 is a @xmath357-invariant subset of @xmath52 .
moreover , @xmath515 is compact if and only if @xmath516 is .
let @xmath517 denote the space of compactly supported smooth functions on @xmath47 .
using the map @xmath518 , @xmath519 , we will often identify smooth functions on @xmath47 with the corresponding right @xmath19-invariant function on @xmath44 .
[ lemma : suppfkak ] let @xmath521 and suppose that @xmath282 is @xmath19-bi - invariant .
then @xmath522 where @xmath523 denotes the restriction of @xmath4 to @xmath513 .
moreover , @xmath524 is a @xmath357-invariant compact subset of @xmath513 .
denote by @xmath525 the ( commutative ) algebra of @xmath44-invariant differential operators on @xmath47 .
we identify @xmath407 can be identified with the submanifold @xmath526 .
then , for every @xmath527 , there is a unique @xmath357-invariant differential operator , @xmath528 on @xmath407 , called the _ radial part of @xmath2 _ , so that for all @xmath529 one has @xmath530 see @xcite , p. 259 .
define a differential operator on @xmath52 , also denoted by @xmath531 , by @xmath532 in this way , we can consider @xmath528 as a ( singular ) @xmath357-invariant differential operator on the euclidean space @xmath52 . to simplify our notation ,
we shall adopt the identification @xmath533 using the exponential map .
then @xmath534 .
we then write @xmath535 instead of @xmath523 , and the above mentioned decomposition of a @xmath19-bi - invariant subset of @xmath44 will be written as @xmath536 instead of @xmath537 . with these identifications ,
the operator @xmath528 turns out to be a hypergeometric differential operator associated with the triple @xmath538 as considered in subsection [ ex : hyper ] .
when @xmath539 is @xmath19-invariant , we can moreover extend ( [ eq : radialpart ] ) by @xmath357-invariance to obtain @xmath540 [ lemma : almostdconvex ] let @xmath47 be a riemannian symmetric space of the noncompact type .
let @xmath527 and let @xmath515 be a compact , convex and @xmath357-invariant subset of @xmath52 .
set @xmath541 .
then for all @xmath521 , we have @xmath542 we need to prove that @xmath543 implies @xmath544 .
the first step is , as in @xcite , to reduce to the case in which the support of @xmath4 is left-@xmath19-invariant .
the function @xmath4 can in fact be expanded as a sum of @xmath19-finite functions . since @xmath545 is left-@xmath19-invariant
, we will obtain @xmath544 if the support of each @xmath19-finite summand is contained in @xmath545 .
we can therefore assume that @xmath4 , and hence @xmath9 , are @xmath19-finite .
as the support of a @xmath19-finite function is left @xmath19-invariant , we obtain from lemma [ lemma : suppfkak ] that @xmath546 and @xmath547 where @xmath548 and @xmath549 are @xmath357-invariant and compact .
since @xmath543 , we then conclude that @xmath550 . since @xmath528 is a hypergeometric differential operator , corollary [ cor : mainhypw ] with @xmath551 yields that @xmath552 .
this proves the required inclusion .
[ thm : dconvex ] let @xmath47 and @xmath527 be as in lemma [ lemma : almostdconvex ] .
then @xmath47 is @xmath2-convex , that is for every compact subset @xmath245 of @xmath47 there is a compact set @xmath553 so that for every @xmath554 the inclusion @xmath555 implies @xmath556 .
here @xmath24 denotes the formal transpose of @xmath2 .
choose @xmath557 so that @xmath558 the set @xmath559 is @xmath19-bi - invariant , hence @xmath560 for some @xmath357-invariant compact subset @xmath515 of @xmath561 .
since @xmath562 we can apply lemma [ lemma : suppfkak ] to it .
so @xmath563 . we can then select @xmath564 .
we conclude this section by a short discussion of the application of the theorem of our support theorem in harmonic analysis corresponding the the @xmath48-hypergeometric transform , c.f .
because of the technical nature of this application , a detailed exposition would require a certain amount of notation and of background information .
instead , we prefer here just to outline the main ideas involved .
we refer the interested reader to @xcite for further information .
as already remarked in the introduction , the support theorem plays a role in the study of the _ paley - wiener space_. this is the set of images under a suitable generalization @xmath565 of the fourier transform , of the compactly supported smooth functions .
the paley - wiener space generally consists of entire or meromorphic functions with exponential growth and possibly satisfying additional symmetry conditions .
the size the support of the original function is linked to the exponential growth of its fourier transform .
the thrust of paley - wiener type theorems is usually to prove that , if a function @xmath453 in the paley - wiener space has a given exponential growth , then the support of the associated `` wave packet '' @xmath566 ( obtained by formal application to @xmath453 of the inverse fourier transform @xmath567 of @xmath565 ) is compact and has the correct size . in the classical situation of the fourier
transform on @xmath1 , this is proven by a suitable `` shift '' of contour of integration by means of cauchy s theorem .
the shift is allowed because @xmath566 is given by integration of an entire function of exponential type and rapidly decreasing .
a suitable generalization of this argument was applied also to the spherical fourier transform on riemannian symmetric spaces .
see @xcite , ch .
iv , 7.2 , or @xcite , 6.6 . for pseudo - riemannian symmetric spaces
the situation is more complicated . here
the wave packages are integrals of meromorphic functions and the required shift of integration would generally require addition of certain `` residues '' . for some kinds of fourier transforms , like the spherical fourier - laplace transform on noncompactly causal symmetric spaces with even multiplicities @xcite ,
the singularities of the integrand are cancelled by multiplication by a certain polynomial function @xmath429 .
hence , for every function @xmath453 in the paley - wiener space the wave - packet @xmath568 is given by integration of an entire function .
furthermore , there is an invariant differential operator @xmath2 so that @xmath569 .
the possibility of shifting the contour of integration in @xmath568 allows us then to determine the size of the support of @xmath570 .
the generalization of the support theorem of lions and titchmarsch presented in this papers allows us finally to recover the size of the support of @xmath571 .
a concrete application of this procedure using theorem [ thm : mainfunction ] can be found in @xcite in the context of the @xmath48-hypergeometric transform .
the latter transform , stated in a setting of transforms associated with root systems , is a common generalization of opdam s hypergeometrical transform ( hence of harish - chandra s spherical transform on riemannian symmetric spaces of noncompact type ) and of the fourier - laplace transform on noncompactly causal symmetric spaces @xcite .
a similar application would also be needed for determining the paley - wiener space for the transform associated with the @xmath48-bessel functions of @xcite .
g. j. heckman , .
in : proceedings of the special session on hypergeometric functions on domains of positivity , jack polynomials and applications at the ams meeting in tampa , fa , march 2223 ( contemp .
* 138 * ( 1992 ) ) , am .
soc . , providence , ri . | let @xmath0 be a finite coxeter group acting linearly on @xmath1 . in this article we study support properties of @xmath0-invariant partial differential operator @xmath2 on @xmath1 with real analytic coefficients .
our assumption is that the principal symbol of @xmath2 has a special form , related to the root system corresponding to @xmath0 . in particular the zeros of the principal symbol
are supposed to be located on hyperplanes fixed by reflections in @xmath0 . we show that @xmath3 holds for all compactly supported smooth functions @xmath4 so that @xmath5 is @xmath0-invariant .
the main tools in the proof are holmgren s uniqueness theorem and some elementary convex geometry .
several examples and applications linked to the theory of special functions associated with root systems are presented . |
@xmath0ne has been known as a typical example of a nucleus which has @xmath1 cluster structure .
there have been numerous works based on the cluster model , which explain the observed doublet rotational band structure .
in addition to the ground @xmath7 band , the negative parity band ( @xmath8 ) starting with the @xmath9 state at @xmath10 mev has been observed , and existence of this `` low - lying '' negative parity band is the strong evidence that simple spherical mean field is broken .
these bands are well explained by the picture that @xmath1 cluster is located at some distance from the @xmath2o core @xcite .
recently `` container picture '' has been proposed to describe the non - localization of the @xmath1 cluster around @xmath2o @xcite . however , according to the shell model , four nucleons perform independent particle motions around the @xmath2o core , which has doubly closed shell of the @xmath11 shell , and the spin - orbit interaction acts attractively to them .
if we apply simple @xmath1 cluster models , we can not take into account this spin - orbit effect . in traditional @xmath1 cluster models ,
@xmath1 cluster is defined as @xmath12 configuration centered at some localized point , and the contributions of non - central interactions vanish .
if we correctly take into account the spin - orbit effect , @xmath1 cluster structure competes with the @xmath3-coupling shell model structure .
previously we have investigated this competition in @xmath0ne based on the antisymmetrized quasi - cluster model ( aqcm ) @xcite .
aqcm is a method that enables us to describe a transition from the @xmath1 cluster wave function to the @xmath3-coupling shell model wave function @xcite . in this model ,
the cluster - shell transition is characterized by only two parameters ; @xmath4 representing the distance between @xmath1 cluster and core nucleus and @xmath5 describing the breaking of the @xmath1 cluster . by introducing @xmath5
, we transform @xmath1 cluster to quasi cluster , and the contribution of the spin - orbit interaction , very important in the @xmath3-coupling shell model , can be taken into account .
it was found that the level structure of the yrast states of @xmath0ne strongly depends on the strength of the spin - orbit interaction in the hamiltonian . in this article
we apply aqcm again to @xmath0ne and introduce @xmath2o plus one quasi cluster model .
particularly we focus on the effect of cluster - shell competition on the @xmath6 transition .
the @xmath6 transition operator has the form of monopole operator , @xmath13 , and this operator changes the nuclear sizes .
however , changing nuclear density uniformly requires quite high excitation energy . on the other hand ,
clusters structures are characterized as weakly interacting states of strongly bound subsystems .
thus it is rather easy for the cluster states to change the sizes without giving high excitation energies ; this is achieved just by changing the relative distances between clusters .
therefore , @xmath6 transitions in low - energy regions are expected to be signatures of the cluster structures , and many works along this line are going on @xcite . in our preceding work for @xmath2o @xcite
, we found that the ground state has a compact four @xmath1 structure and is almost independent of the strength of the spin - orbit interaction ; however the second @xmath14 state , which has been known as a @xmath15c+@xmath1 cluster state , is very much affected by the change of the strength . with increasing the strength ,
the level repulsion and crossing occur , and the @xmath16 cluster part changes from three @xmath1 configuration to the @xmath17 subclosure of the @xmath3-coupling shell model .
the @xmath6 transition matrix elements are strongly dependent on this level repulsion and crossing , and they are sensitive to the persistence of @xmath18 correlation in the excited states . here , `` larger cluster '' part of binary cluster system ( @xmath15c part of @xmath15c+@xmath1 ) has been changed into quasi cluster .
the present study on @xmath0ne is different from the preceding work on @xmath2o in the following two points .
one is that we focus on the change of `` smaller cluster '' part of the binary cluster system , and in this case , we change @xmath1 cluster around the @xmath2o core to quasi cluster .
another difference is that this change influences very much the ground state ( in the case of @xmath2o , the second @xmath14 state with the @xmath15c+@xmath1 configuration is affected by the spin - orbit interaction ) .
since other higher nodal states are determined by the orthogonal condition to the ground state , this change also has influences on the wave functions of the excited states .
naturally @xmath6 transition matrix elements are also affected by this change .
the paper is organized as follows .
the formulation is given in sect .
[ model ] . in sect .
[ results ] , the results for @xmath0ne are shown . finally , in sect . [ summary ] we summarize the results and give the main conclusion .
the wave function of the total system @xmath19 is antisymmetrized product of these single particle wave functions ; @xmath20 the projection onto parity and angular momentum eigen states can be numerically performed .
the number of mesh points for the integral over euler angles is @xmath21 . for the single particle orbits of the @xmath2o part
, we introduce conventional @xmath1 cluster model .
the single particle wave function has a gaussian shape @xcite ; @xmath22 \eta_{i } , \label{brink - wf}\ ] ] where @xmath23 represents the spin - isospin part of the wave function , and @xmath24 is a real parameter representing the center of a gaussian wave function for the @xmath25th particle . for the width parameter
, we use the value of @xmath26 fm , @xmath27 . in this brink - bloch wave function , four nucleons in one @xmath1 cluster share the common @xmath24 value .
hence , the contribution of the spin - orbit interaction vanishes .
we introduce four different kinds of @xmath24 values , and four @xmath1 clusters are forming tetrahedron configuration .
when we take the limit of the relative distances between @xmath1 cluster to zero , the wave function coincide with the closed @xmath11 shell configuration of the shell model @xcite , and this limit is called elliot su(3 ) limit @xcite . in our model , the relative distance is taken to be a small value , 0.1 fm .
we add one quasi cluster around the @xmath2o core based on aqcm . in the aqcm
, @xmath1 clusters are changed into quasi clusters . for nucleons in the quasi cluster ,
the single particle wave function is described by a gaussian wave packet , and the center of this packet @xmath28 is a complex parameter ; @xmath29 \chi_{i } \tau_{i } , \label{aqcm_sp}\ ] ] @xmath30 where @xmath31 and @xmath32 in eq .
represent the intrinsic spin and isospin part of the @xmath25th single particle wave function , respectively . in eq
, @xmath33 is a unit vector for the orientation of the intrinsic spin @xmath31 .
here , @xmath5 is a real control parameter describing the dissolution of the @xmath1 cluster .
the width parameter is the same as nucleons in the @xmath2o cluster ( @xmath26 fm , @xmath27 ) .
as one can see immediately , the @xmath34 aqcm wave function , which has no imaginary part , is the same as the conventional brink - bloch wave function .
the aqcm wave function corresponds to the @xmath3-coupling shell model wave function when @xmath35 and @xmath36 .
the mathematical explanation is summarized in ref .
@xcite .
gaussian center parameters for the four nucleons in the quasi cluster ( @xmath37 ) are given in the following way .
we place quasi cluster on the @xmath38 axis , and the real part of the gaussian center parameters ( @xmath39 ) are given as @xmath40 here @xmath4 is a parameter , which describes the distance between quasi cluster and the @xmath2o cluster , and @xmath41 is the unit vector in the @xmath38 direction .
next , we give the imaginary parts . here
we quantize the spin of the nucleons along the @xmath42 axis , and in order to satisfy the condition of eq .
, we must give the imaginary parts in the @xmath43 direction as , @xmath44 @xmath45 @xmath46 @xmath47 where @xmath48 and @xmath49 are unit vectors in the @xmath38 and @xmath50 direction , respectively .
the gaussian center parameter @xmath51 is for a proton with spin up ( @xmath52 direction ) , @xmath53 is for a proton with spin down ( @xmath54 direction ) , @xmath55 is for a neutron with spin up ( @xmath52 direction ) , and @xmath56 is for a neutron with spin down ( @xmath54 direction ) .
when @xmath5 is set to zero , the wave function consisting the quasi clusters agrees with that of an @xmath1 cluster . if we take the limit of @xmath57 and @xmath58 , four nucleons in the quasi cluster occupy @xmath59 orbits of the @xmath3-coupling shell model . for the hamiltonian , we use volkov no.2 @xcite as an effective interaction for the central part with the majorana exchange parameter of @xmath60 . for the spin - orbit part , g3rs @xcite , which is a realistic interaction originally determined to reproduce the nucleon - nucleon scattering phase shift ,
is adopted ; @xmath61 where @xmath62 @xmath63 @xmath64 , and @xmath65 is a projection operator onto a triplet odd state .
the operator @xmath66 stands for the relative angular momentum and @xmath67 is the spin ( @xmath68 ) . in the present work ,
the strength of the spin - orbit interaction , @xmath69 , is a parameter as in ref .
@xcite and we compare the results by changing the value .
in this section , we apply our aqcm wave function introduced in the previous section to @xmath0ne and discuss the @xmath69 ( strength of the spin - orbit interaction ) dependence of energy levels and @xmath6 transition probabilities .
the @xmath69 value is changed from 0 mev to 3000 mev , and reasonable value of around @xmath70 mev has been suggested in our preceding work @xcite .
we prepare aqcm wave functions with different @xmath4 and @xmath5 values as basis states of generator coordinate method ( gcm ) .
the adopted values are @xmath71 fm and @xmath72 .
the @xmath14 energies of these basis states are presented in table [ basis ] , and here , we show the values for two extreme cases for the strengths of the spin - orbit interaction ; ( a ) @xmath73 mev and ( b ) @xmath74 mev .
the @xmath14 energies of the gcm basis states corresponding to other @xmath69 values can be estimated just by interpolating these values linearly . in table
[ basis ] ( a ) , we find that @xmath34 basis states give lower energies than @xmath5 finite basis states .
this is because of the absence of the spin - orbit interaction ; introducing imaginary part for the gaussian center parameters does not work for the spin - orbit interaction and that simply increases the kinetic energy of four nucleons in the quasi cluster . here
the basis state with @xmath75 fm ( @xmath34 ) gives the lowest energy of @xmath76 mev . on the contrary
, table [ basis ] ( b ) is the case of @xmath74 mev , and basis states with finite @xmath5 values get much lower , since the contribution of the spin - orbit interaction can be taken into account by transforming the @xmath1 cluster to quasi cluster . the basis state which gives the lowest energy has the values of @xmath77 and @xmath78 fm ( @xmath79 mev ) .
this result suggests that when the spin - orbit interaction is switched on , the @xmath4 value of the optimal basis state becomes smaller and the @xmath5 value increases .
this means that not only the @xmath1 cluster dissolutes into quasi cluster , the relative distance between the cluster and the @xmath2o core decreases .
( a ) .the @xmath14 energies of gcm basis states for the cases of different strengths of the spin - orbit interaction ; ( a ) @xmath73 mev and ( b ) @xmath74 mev .
the @xmath14 energies of the gcm basis states with other @xmath69 values can be estimated by linearly interpolating these two . [ cols="^,^,^,^,^",options="header " , ] [ sqol-1- ]
we have applied aqcm , which is a method to describe a transition from the @xmath1-cluster wave function to the @xmath3-coupling shell model wave function , to @xmath0ne .
@xmath0ne has been known as a nucleus which has @xmath2o+@xmath1 structure , and we investigated how the @xmath1 cluster structure competes with independent particle motions of these four nucleons by changing the strength of the spin - orbit interaction ( @xmath69 ) .
we focused on the @xmath6 transition matrix element , which was found to be sensitive to @xmath69 .
based on aqcm , @xmath0ne is characterized by only two parameters ; @xmath4 representing the relative distance between @xmath2o and @xmath1 and @xmath5 describing the breaking of @xmath1 cluster . when the spin - orbit interaction is switched off ( @xmath73 mev )
, the ground @xmath14 state has the squared overlap of 0.92 with the gcm basis state which has @xmath75 fm and @xmath34 .
the second @xmath14 state is a higher nodal state and it has 0.74 with @xmath80 fm and @xmath81 .
the third @xmath14 state also has large @xmath2o-@xmath1 distance , and the fourth @xmath14 state has overlaps with basis states with finite @xmath5 values .
when the spin - orbit interaction is switched on , we found that the decrease of the energy for the fourth @xmath14 state at @xmath73 is much steeper than other states .
eventually the wave function of the fourth @xmath14 state at @xmath73 mev strongly mixes in the ground and second @xmath14 states at @xmath74 mev . on the other hand ,
the wave function of the second @xmath14 state at @xmath73 mev almost stays at this energy .
similar thing can be found for the third @xmath14 state at @xmath73 mev .
these two states correspond to the third and fourth @xmath14 states at @xmath74 mev , and @xmath1 cluster structure becomes important again there .
the @xmath6 transition matrix elements from the ground state to the second @xmath14 state is calculated as 10.0 @xmath82 @xmath83 at @xmath73 mev , which is slightly larger than the experimental value ( 6.914 @xmath82 @xmath83 ) . with increasing @xmath69 value ,
the mixing of basis states with finite @xmath5 values becomes important in both the ground and second @xmath14 states , and the @xmath6 transition matrix decreases .
the value agrees with the experimental one around @xmath84 mev .
this deduced strength is consistent with our preceding work on the level structure of this nucleus . at @xmath84 mev , which is the spin - orbit strength deduced from the present analysis on the @xmath6 transition matrix element , the ground state has the squared overlap of 0.78 with the basis state which has @xmath75 fm and @xmath81 , and the character at @xmath73
mev still remains .
however the largest squared overlap of 0.82 is with the basis state which has @xmath85 fm and @xmath86 .
therefore , the @xmath1 breaking effect due to the spin - orbit interaction is also important in the ground state .
the second @xmath14 states has squared overlap of 0.40 with the basis state which has @xmath87 fm and @xmath34 ; however , it has also components of the basis states with finite @xmath5 values .
the third and fourth @xmath14 states are @xmath1 cluster states and contain the components of basis states with large @xmath4 values .
the presence of low - lying negative parity band starting with the first @xmath9 has been the key evidence for the @xmath1 cluster structure .
we also investigated the @xmath9 states and found that the energy of the first @xmath9 state is almost constant even if the spin - orbit interaction is switched on and @xmath1 breaking basis states are introduced .
the @xmath1 cluster structure is really important in this state .
there have been discussions that the @xmath15c+@xmath1+@xmath1 cluster states appear in this energy region of the third @xmath14 state , and inclusion of this configuration can be done by applying aqcm to the three @xmath1 clusters in the @xmath2o core .
also , here we transformed an @xmath1 cluster to four independent nucleons , in which the spin - orbit interaction acts attractively .
however , in principle it is possible to introduce other shell model configurations , for instance configurations where the spin - orbit interaction acts repulsively , or one of the nucleon is excited from @xmath88-upper orbit to @xmath88-lower orbit .
the analysis aiming at the unified view is going on .
100 hisashi horiuchi and kiyomi ikeda , prog .
( 1968 ) * 40 * 277 .
bo zhou , zhongzhou ren , chang xu , y. funaki , t. yamada , a. tohsaki , h. horiuchi , p. schuck , and g. r pke , phys .
c * 86 * , 014301 ( 2012 ) .
n. itagaki , j. cseh , and m. poszajczak , phys .
c * 83 * , 014302 ( 2011 ) . n. itagaki , h. masui , m. ito , and s. aoyama , phys . rev .
c * 71 * 064307 ( 2005 ) .
h. masui and n. itagaki , phys .
c * 75 * 054309 ( 2007 ) .
t. yoshida , n. itagaki , and t. otsuka , phys .
c * 79 * 034308 ( 2009 ) .
t. suhara , n. itagaki , j. cseh , and m. poszajczak , phys .
c * 87 * , 054334 ( 2013 ) .
n. itagaki , h. matsuno , and t. suhara , arxiv : 1507.02400 . t. kawabata _ et al_. , phys .
b * 646 * , 6 ( 2007 )
. y. sasamoto _
et al_. , mod .
a * 21 * , 2393 ( 2006 ) .
t. yoshida , n. itagaki , and t. otsuka , phys .
c * 79 * , 034308 ( 2009 ) .
taiichi yamada and yasuro funaki , phys .
c * 92 * , 034326 ( 2015 ) .
t. ichikawa , n. itagaki , t. kawabata , tz .
kokalova and w. von oertzen , phys .
c * 83 * , 061301(r ) ( 2011 ) . t. ichikawa , n. itagaki , y. kanada - enyo , tz .
kokalova , and w. von oertzen , phys .
c * 86 * , 031303(r ) ( 2012 ) .
h. matsuno and n. itagaki , arxiv : 1601.05892 .
d. m. brink , in _ proceedings of the international school of physics `` enrico fermi '' course xxxvi _
, edited by c. bloch ( academic , new york , 1966 ) , p. 247 . j. p. elliot , proc .
a * 245 * 128 , 562 ( 1958 ) a. b. volkov , nucl . phys .
* 74 * , 33 ( 1965 ) .
r. tamagaki , prog .
theor . phys . *
39 * , 91 ( 1968 ) . | @xmath0ne has been known as a typical example of a nucleus with @xmath1 cluster structure ( @xmath2o+@xmath1 structure ) . however according to the spherical shell model , the spin - orbit interaction acts attractively for four nucleons outside of the @xmath2o core , and this spin - orbit effect can not be taken into account in the simple @xmath1 cluster models .
we investigate how the @xmath1 cluster structure competes with independent particle motions of these four nucleons . the antisymmetrized quasi - cluster model ( aqcm )
is a method to describe a transition from the @xmath1 cluster wave function to the @xmath3-coupling shell model wave function . in this model ,
the cluster - shell transition is characterized by only two parameters ; @xmath4 representing the distance between clusters and @xmath5 describing the breaking of @xmath1 clusters , and the contribution of the spin - orbit interaction , very important in the @xmath3-coupling shell model , can be taken into account by changing @xmath1 clusters to quasi clusters . in this article ,
based on aqcm , we apply @xmath2o plus one quasi cluster model for @xmath0ne . here
we focus on the @xmath6 transition matrix element , which has been known as the quantity characterizing the cluster structure .
the @xmath6 transition matrix elements are sensitive to the change of the wave functions from @xmath1 cluster to @xmath3-coupling shell model . |
networks of coupled oscillators have been the subject of extensive research in the last decade .
coupled systems can display a huge variety of dynamical phenomena , starting from synchronization phenomena in various types of inhomogeneous or irregular networks , up to complex collective behavior , such as for example various forms of phase transitions , traveling waves @xcite , phase - locked patterns , amplitude death states @xcite , or so called chimera states that display a regular pattern of coherent and incoherent motion @xcite .
of particular interest are situations , where complex spatio - temporal structures can emerge in regular arrays of identical units induced only by the coupling interaction . in many cases ,
the resulting phenomena differ substantially from corresponding situations in continuous media @xcite and depend strongly on the underlying network topology .
our specific interest is in the emergence of spatio - temporal structures in a homogeneous array of identical units that have a stable uniform equilibrium at which the coupling vanishes . as a classical paradigm ,
the turing instability gives an example of a coupling induced instability in such a setting .
this phenomenon has of course a direct counterpart in the discrete setting , but it turns out that there appear some genuinely new phenomena . in @xcite it has been shown that also in a ring of unidirectionally coupled oscillators , i.e. in a purely convective setting , the eckhaus scenario of coexisting diffusive patterns can be observed . in @xcite it has been shown that duffing oscillators coupled in the same way , exhibit a complex transition to spatio - temporal chaos . in this paper
we develop a general theoretical framework for such phenomena in large arrays .
we derive an amplitude equation of ginzburg - landau type that governs the local dynamics close to the destabilization of the uniform steady state .
it resembles several features that are already well known in the context of reaction - diffusion systems @xcite .
but in contrast to these results , it can be applied to much more general coupling mechanisms , including also the case of unidirectional and anti - diffusive interaction and allows also for a mixture of such interactions with several next neighbors .
such an interplay of attractive and repulsive coupling is an essential feature of neural systems . as a specific feature , the convective part will appear in the amplitude equation as a rotation of the coordinates in an intermediate time scale that is faster than the diffusive processes described by the ginzburg - landau equation .
having deduced the amplitude equation and the corresponding scaling laws in terms of the number of oscillators , which is assumed to tend to infinity , we use this theory for the explanation of a specific phenomenon that has been reported in @xcite for a ring of unidirectionally coupled duffing oscillators .
there , it has been shown numerically that for a large number of oscillators @xmath0 , there is an almost immediate transition from homogeneous stationary behavior to high - dimensional chaos .
based on our amplitude equation , we argue that in such systems , one can expect generically that such a transition occurs within a parameter interval of the size @xmath1 .
we consider a generic case , where the control parameter enters already the linear parts of the dynamical equations , e.g. a diffusive coupling strength .
finally , we demonstrate this phenomenon by a numerical example , where we also evaluate the scaling behavior of the parameter interval where the transition to chaos takes place for an increasing number of oscillators .
coupled oscillators . apart from the self - coupling @xmath2 each oscillator @xmath3 is also coupled with @xmath4 ( @xmath5 ) as well as @xmath6 ( @xmath7 ) . see eq .
( [ eq : main ] ) for the equation of motion .
[ fig : ring ] ] we are interested in a system of @xmath0 identical coupled oscillators that has an uniform equilibrium , where the coupling vanishes . the coupling network is organized in a ring structure , where interactions of several next neighbors are possible .
such systems can be written in general form as @xmath8 where @xmath9 describes the state of the @xmath10-th oscillator .
the ring structure is induced by periodic boundary conditions , i.e. all indexes have to be considered modulo @xmath0 .
the linear part of the dynamics is given by the @xmath11 matrices @xmath12 , depending on the bifurcation parameter @xmath13 , which account for the coupling to the m - th neighbor ; in particular @xmath14 describes the linear part of the local dynamics ( self - coupling ) .
the nonlinear part @xmath15 , again including a local dependence and a dependence on the @xmath16-th neighbor , should vanish at the origin @xmath17 and have also zero derivatives there .
note that this system is symmetric ( equivariant ) with respect to index shift .
figure [ fig : ring ] illustrates an example with self coupling and coupling to the neighbor on the left and to the second neighbor on the right .
the specific form of ( [ eq : main ] ) also implies that the coupling vanishes at the equilibrium @xmath18 , which is true e.g. when the coupling is a function of the difference @xmath19 for any two coupled oscillators @xmath20
. the characteristic equation for the linearization at the zero equilibrium of ( [ eq : main ] ) can be factorized as @xmath21=0,\ ] ] where @xmath22 denotes the identity matrix in @xmath23 and the index @xmath24 accounts for the @xmath0-th roots of unity that appear as the eigenvalues of the circular coupling structure @xcite . following the approach in @xcite
, we replace for large @xmath0 the discrete numbers @xmath25 by a continuous parameter @xmath26 , and obtain the _ asymptotic continuous spectrum _
@xmath27=0,\,\varphi\in[0,2\pi)\right\ } , \label{eq : acs}\ ] ] which contains all eigenvalues and which for large @xmath0 is covered densely by the eigenvalues .
since the expression ( [ eq : acs ] ) is periodic in @xmath26 , the asymptotic continuous spectrum @xmath28 has generically the form of one or several closed curves @xmath29 in the complex plane , parametrized by @xmath26 . at the bifurcation value @xmath30 , we assume that the asymptotic continuous spectrum touches the imaginary axis at at some point @xmath31 ( see fig .
[ fig : acs ] ) , i.e. the following conditions are fulfilled @xmath32 the first condition from ( [ eq : bifcond ] ) means that the point @xmath31 belongs to the asymptotic continuous spectrum , while the second condition from ( [ eq : bifcond ] ) guarantees that the real part @xmath33 is tangent to the zero axis at @xmath34 ( see fig . [ fig : acs ] ) .
this tangency condition is quite natural and describes the condition for the destabilization ( or bifurcation ) of the zero solution a large ring of coupled oscillators . indeed ,
if the spectrum @xmath28 is contained in the left half of the complex plane with @xmath35 , then the uniform equilibrium @xmath18 is asymptotically stable .
as soon as @xmath28 crosses the imaginary axis , it becomes unstable for sufficiently large @xmath0 .
asymptotic continuous spectrum @xmath28 at the destabilization ; schematically . ]
before we present our main result , we now introduce some useful notations and formulate a technical lemma , which follows from the bifurcation conditions ( [ eq : bifcond ] ) . with @xmath36 and @xmath37
we denote the eigenvector and the adjoint eigenvector to the critical eigenvalue @xmath38 , which we assumed in ( [ eq : bifcond ] ) to exist for @xmath30 .
moreover , it will be convenient to denote by @xmath39 @xmath40 @xmath41 the `` moments '' of the coupling matrices @xmath42 . in this notation
, the equations for the eigenvectors @xmath36 and @xmath37 read as @xmath43v_{0}=0,\label{eq : v0}\ ] ] @xmath44v_{1}=0.\label{eq : v1}\ ] ] these vectors can be normalized as @xmath45 finally , we expand the matrices @xmath46 with respect to the parameter @xmath13 and write them them as @xmath47 . in this way , we can define @xmath48 assume that @xmath49 is a regular point of the asymptotic continuous spectrum ( [ eq : acs ] ) , such that @xmath50 exists and is locally differentiable in a small neighborhood of @xmath49 .
further , let the bifurcation condition ( [ eq : bifcond ] ) hold
. then @xmath51 the proof of this lemma will be given together with the proof of the main result in section [ sec : proofs ] .
in this section , we present an amplitude equation that describes for system ( [ eq : main ] ) the dynamics close to the destabilization threshold in the limit of large @xmath0 . we perform a limiting procedure , where we assume that a fixed number @xmath52 of neighbors is involved in the coupling while the total number of oscillators @xmath0 tends to infinity .
also the nonlinearity @xmath15 is assumed to depend on @xmath52 next neighbors only .
hence , in the sequel all summation will extend over @xmath53 and the nonlinearity will be written as @xmath54 , independently on @xmath0 . in this way
, the coupling becomes local in the limit @xmath55 , and coupling terms will be approximated by derivatives of the amplitude .
consequently , our results will be valid for large rings , where the coupling range is small compared to the total size .
it will be shown that the amplitude equation has the form of a complex ginzburg - landau equation with periodic boundary conditions .
the derivation is quite technical and we give here in proposition 2 only the main assertions .
the proof is deferred to section [ sec : proofs ] .
[ prop2]assume that the bifurcation condition ( [ eq : bifcond ] ) holds and @xmath49 is a regular point of the asymptotic continuous spectrum @xmath56 .
additionally , let the points @xmath57 not belong to the asymptotic continuous spectrum with @xmath34 ( nonresonance condition ) .
let the nonlinearity @xmath15 be of third order . introduce the small parameter @xmath58 and apply the multiple scale ansatz @xmath59 with the amplitude @xmath60 depending on the rescaled coordinates @xmath61 @xmath62 and @xmath63 ( @xmath64 denotes complex conjugated terms , @xmath65 ) to the following system @xmath66 with the rescaled parameter @xmath67 and @xmath68 with periodic boundary conditions .
then , the solvability conditions up to the order @xmath69 imply the following partial differential equation of ginzburg - landau type @xmath70 with periodic boundary conditions @xmath71 where @xmath72 with @xmath73 $ ] is related to the amplitude @xmath74 by @xmath75 the coefficient @xmath76is given by ( [ eq : bifcond ] ) or ( [ eq : k1 ] ) , and @xmath77 finally , the coefficient of the nonlinearity @xmath78 and the vector @xmath79 have to be determined by the nonlinearity @xmath15 according to ( [ eq : zeta ] ) and ( [ eq : v2 ] ) . according to this proposition , small solutions of a coupled system of the form ( [ eq : perteq ] ) that has a parameter close to criticality , i.e.
@xmath80 can be approximated in the form ( [ eq : multscale ] ) , where the amplitude @xmath74 is related to a solution of the ginzburg - landau equation ( [ eq : gl ] ) via ( [ eq : au ] ) .
note that the relation ( [ eq : au ] ) introduces a rotating frame on the timescale @xmath81 with rotation velocity @xmath82 the time evolution of the ginzburg - landau equation enters only on the slowest time scale @xmath83 .
for the case of a symmetric coupling ( @xmath84 ) the amplitude equation ( [ eq : gl ] ) has all real coefficients @xmath76 , @xmath85 and @xmath86 and the rotation velocity vanishes @xmath87 .
* proof of lemma 1 . *
we have to show that @xmath76 , defined by ( [ eq : bifcond ] ) , can be calculated as given in ( [ eq : k1 ] ) .
to this end , we differentiate the eigenvalue equation ( cf .
( [ eq : v0 ] ) ) @xmath88v\left(\varphi\right)=0,\ ] ] with respect to @xmath26 : @xmath89v(\varphi)+\left[\lambda_{0}(\varphi)\mathrm{id}-\sum_{m}e^{im\varphi}m_{m}(0)\right]\frac{\partial}{\partial\varphi}v(\varphi)=0.\ ] ] evaluating the obtained expression at @xmath90 @xmath91 , and @xmath92 , we obtain @xmath93v_{0}+\left[i\omega_{0}\mathrm{id}-l_{0}\right]\frac{\partial v}{\partial\varphi}(\varphi_{0})=0.\ ] ] the projection onto @xmath36 gives @xmath94\frac{\partial v}{\partial\varphi}(\varphi_{0}),v_{1}\right\rangle = 0,\ ] ] which implies @xmath95v_{1}\right\rangle = 0.\ ] ] taking into account ( [ eq : v1 ] ) , we obtain the relation ( [ eq : k1 ] ) , which proves the lemma .
* proof of proposition 2 . * substituting the multiple scale ansatz ( [ eq : multscale ] ) into ( [ eq : perteq ] ) , we obtain @xmath96 @xmath97 @xmath98 @xmath99 dividing the obtained equation by @xmath100 , expanding necessary arguments of @xmath74 , we obtain up to the terms of the order @xmath101 ( the complex conjugated terms are omitted here for brevity ) @xmath102 @xmath103 @xmath104 @xmath105 where @xmath106 and @xmath107 are determined by the leading terms in the expansion of the nonlinearity .
note that due to our assumption , @xmath15 is of third order and the leading order terms are given by expanding @xmath108 in the homogeneous state @xmath109 with respect to @xmath110 .
the solvability condition requires that different harmonics as well as different orders of @xmath100 up to @xmath101 are equal .
we start with the first harmonic . by equating the terms containing @xmath111
we obtain the following equation @xmath112 @xmath113 since it should be satisfied for all @xmath100 , we first consider the @xmath114 equation @xmath115 which holds according to the spectral condition ( [ eq : v0 ] ) .
the @xmath116 terms result into @xmath117 multiplication with @xmath118 and using ( [ eq : k1 ] ) from lemma 1 , we obtain @xmath119 this will be accounted for by introducing the new amplitude @xmath120 by @xmath121 in a correspondingly rotating coordinate @xmath122 finally , the @xmath101 terms result into @xmath123 note that the dependence on @xmath124 does not show up in this equation . hence , after multiplication with @xmath118 , we can write it in terms of @xmath120 as @xmath125 where @xmath126 finally , it is simple to check that the solvability of the terms for the third harmonic leads to the expression @xmath127^{-1}h_{2}\left(v_{0}\right)\label{eq : v2}\ ] ] here , the existence of a nonzero solution @xmath128 is guaranteed by the nonresonance condition . indeed , since the points @xmath57 do not belong to the asymptotic continuous spectrum with @xmath34 , the matrix @xmath129 $ ] is non - singular .
finally , note that the set of equations should be complemented by periodic boundary conditions in @xmath130 , taking into account the ring structure of system ( [ eq : perteq ] ) .
the proposition is proved .
it is well known that in the complex ginzburg - landau equation ( [ eq : gl ] ) a destabilization transition is possible , where in a series of bifurcations a homogeneous stationary state loses its stability and , via periodic and quasiperiodic motion , a regime of spatio - temporal chaos appears @xcite .
we will now use the result about the amplitude equation in order to trace back this destabilization transition to the original coupled oscillator system of the form ( [ eq : main ] ) .
specific attention will be paid to the scaling behavior of the parameter region where this transition takes place . starting with the scale free amplitude equation @xmath131 with periodic boundary conditions
, we assume that at @xmath132 a destabilization of the homogeneous stationary state takes place and for further increasing parameter @xmath133 a transition to spatio - temporal chaos can be observed .
this transition takes place within some bounded interval @xmath134 considering now the corresponding behavior of an oscillator system that is described by this amplitude equation , we have to introduce the scalings given in proposition [ prop2 ] .
in particular , according to the parameter scaling @xmath135 , the same transition takes place in the coupled oscillator system within a parameter interval @xmath136 of the corresponding parameter @xmath13 of the oscillator system .
therefore , one can expect that a generic transition to chaos ( or hyperchaos ) in the system of @xmath0 coupled oscillators occurs within a parameter interval that scales as @xmath1 . in the following section , we present an example where the assumptions that we made above are satisfied and , consequently , in the oscillator system an almost immediate transition from stationary behavior to chaos can be observed .
in the previous section we discussed the scaling law for the transition to chaos in a ring of coupled oscillators . in this section
we provide a numerical example to illustrate this scaling .
we use the autonomous duffing oscillator described by the following second order ordinary differential equation @xmath137 where @xmath138 and @xmath139 are positive constants . system ( [ eq : duff_single ] ) is a single - well duffing oscillator , which has a single equilibrium point at @xmath140 . due to the presence of damping @xmath141 this equilibrium is an attractor for all initial conditions .
we consider now a ring of @xmath0 such oscillators with a linear unidirectional coupling to the next neighbor .
introducing new coordinates @xmath142 , @xmath143 and the coupling into eq .
( [ eq : duff_single ] ) , the equations of motion have the form @xmath144 where @xmath145 is the coupling coefficient and indices are considered modulo @xmath0 .
a detailed analysis of the behavior of this system was already presented in @xcite . for at least three coupled oscillators and increasing coupling strength @xmath145
, one can observe rich dynamics starting from periodic oscillations to hyperchaos .
here we focus our attention on the transition to chaos ( hyperchaos ) and its dependence on the number of oscillators . in our numerics , we used the fixed parameter values @xmath146 and @xmath147 .
we calculated the maximum lyapunov exponents for increasing @xmath0 and varying coupling coefficient @xmath145 .
based on this , we determined two values : @xmath148 where the uniform stationary state loses its stability in a hopf bifurcation , and @xmath149 where the transition to chaos takes place . a reliable computation of the lyapunov exponents was possible only for moderate values of @xmath0 ( @xmath150 ) . for larger systems , we monitored the behavior in an appropriately chosen poincare section in order to determine @xmath148 and @xmath149 .
[ fig : values - of - coupling](a ) shows that for increasing @xmath0 the distance between @xmath148 and @xmath149 decreases .
finally , for a large enough number of oscillators , the transition to chaos appears practically immediately after the hopf bifurcation . in order to validate our conjecture about the @xmath1 scaling , we plot in fig .
[ fig : values - of - coupling](b ) the scaled transition intervals @xmath151 the numerical results indicate that the scaled transition interval neither tends to zero nor diverges to infinity , which supports the scaling assumption .
( a ) values of the coupling parameter @xmath148 at hopf bifurcation ( crosses ) and @xmath149 at the transition to chaos ( circles ) for increasing number of oscillators @xmath0 .
( b ) rescaled transition interval @xmath152(circles ) versus @xmath0 .
, width=415 ] in fig .
[ fig : lap_sec30 ] we present the lyapunov spectrum for thirty coupled oscillators .
panel ( a ) shows that at @xmath153 two lyapunov exponents become positive practically simultaneously .
a similar behavior has been observed for other large values of @xmath0 .
this shows that the transition to chaos and to hyperchaos occursin a large ring almost simultaneously . with further increasing coupling coefficient @xmath145 ( fig .
[ fig : lap_sec30](b ) ) one can observe that more and more lyapunov exponents become positive which leads to a high dimensional chaos @xcite .
lyapunov exponents for @xmath154 unidirectionally coupled duffing oscillators ( a ) five largest lyapunov exponents , @xmath148 and @xmath149 indicate the hopf bifurcation and the transition to chaos , @xmath155 , ( b ) twenty largest lyapunov exponents for @xmath156 . at @xmath157 , there are fourteen positive lyapunov exponents .
, width=415 ]
the results of this paper are twofold .
the first one is more theoretical : we have derived an amplitude equation for small amplitude solutions of a ring system of coupled oscillators .
this equation has the form of a complex ginzburg - landau equation with periodic boundary conditions .
such an amplitude equation is typical for reaction - diffusion systems @xcite in continuous media and can also be found for systems with large delay @xcite .
obviously , a similar behavior should be expected in discrete systems with diffusive coupling .
however , our results show that in coupled oscillator systems the ginzburg - landau equation can also be used to describe the dynamics in systems with unidirectional , i.e. purely convective coupling or even with anti - diffusive interaction . in this sense , the class of coupled oscillators that we treated in this paper differs substantially from discrete analogs of the classical results for continuous media . for the second result , we applied the amplitude equation to the specific scenario of the transition to spatio - temporal chaos in such systems . in this way
, we provided a general framework for the coupling induced transition to high dimensional chaos in coupled oscillator systems , that has been described for a specific example in @xcite .
in particular , we were able to show that the observed fast transition to chaos is a generic feature and follows a @xmath1 scaling law for systems with a large number @xmath0 of oscillators .
as a result of this scaling , one observes for large systems a practically immediate transition from a uniform stationary state to chaos .
we illustrated this behavior by a numerical example , where the emergence of high dimensional chaos is demonstrated by corresponding lyapunov spectra .
s. yanchuk acknowledges the support of the dfg collaborative research center control of self - organizing nonlinear systems : theoretical methods and concepts of application ( sfb910 ) under the project a3 .
p. perlikowski , a. stefaski and t. kapitaniak acknowledge the support from foundation for polish science - team programme ( project no team/20105/5 ) and p. perlikowski start programme .
25ifxundefined [ 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 @noop * * , ( ) @noop _ _ ( , ) p. @noop _ _ ( , ) p. @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physrevlett.107.228102 [ * * , ( ) ] link:\doibase 10.1103/physreve.69.056217 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.93.174102 [ * * , ( ) ] @noop * * , ( ) link:\doibase
10.1103/physrevlett.100.144102 [ * * , ( ) ] link:\doibase 10.1063/1.3563579 [ * * , ( ) ] link:\doibase 10.1038/nphys1651 [ * * , ( ) ] link:\doibase 10.1103/physreve.77.026212 [ * * , ( ) ] link:\doibase 10.1103/physreve.82.036208 [ * * , ( ) ] link:\doibase 10.1063/1.3665200 [ * * , ( ) ] http://www.matheon.de/research/show_preprint.asp?action=details&serial=667 [ * * , ( ) ] @noop _ _ , , vol .
( , , ) p. @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physreva.42.6238 [ * * , ( ) ] @noop * * , ( ) \doibase doi : 10.1016/0375 - 9601(91)90624-h
[ * * , ( ) ] link:\doibase 10.1103/physreve.47.r2975 [ * * , ( ) ] @noop * * , ( ) | we study the coupling induced destabilization in an array of identical oscillators coupled in a ring structure where the number of oscillators in the ring is large .
the coupling structure includes different types of interactions with several next neighbors .
we derive an amplitude equation of ginzburg - landau type , which describes the destabilization of a uniform stationary state in a ring with a large number of nodes . applying these results to unidirectionally coupled duffing oscillators
, we explain the phenomenon of a fast transition to chaos , which has been numerically observed in such systems .
more specifically , the transition to chaos occurs on an interval of a generic control parameter that scales as the inverse square of the size of the ring , i.e. for sufficiently large system , we observe practically an immediate transition to chaos . 34c15 , 34c23 , 34c28 , 37g35 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ the understanding of the dynamical behavior of networks of coupled oscillators can contribute to the explanation of various collective phenomena that can be observed in coupled systems in biology , economy , or physics @xcite .
discrete media , e.g. in neural systems , can exhibit complex coupling structures including feed - forward loops , large coupling ranges , and interaction mechanisms of different kind .
in such systems , the influence of the coupling can transform the simple dynamics of a single oscillator into complicated spatio - temporal structures of the network dynamics .
amplitude equations of ginzburg - landau type have been a powerful tool for a universal description of spatio - temporal dynamics and pattern formation in continuous media . in this paper
, we apply this technique to a large class of coupled oscillator systems with a ring structure , where we assume that the number of oscillators in the ring is large .
in addition , we use the amplitude equation together with the corresponding scaling laws to explain the coupling induced fast transition from homogeneous stationary behavior to high dimensional chaos , which can be observed in certain coupled oscillator systems . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ |
high quality measurements of anisotropies in the cosmic microwave background ( cmb ) probe the cosmic fluctuations generated during an inflationary epoch in the very early universe @xcite . recently ,
boomerang @xcite and maxima @xcite teams announced the clear detection of a first acoustic peak at an angular scale @xmath0 , which confirms the most important prediction of inflation : the universe seems to be spatially flat @xcite .
another generic prediction of inflation is that the primordial spectra of density perturbations and gravitational waves are _ almost _ scale - invariant .
more cmb precision measurements will be available soon .
we argue @xcite that cmb predictions on the basis of the simplest inflationary model , slow - roll inflation @xcite , are not as precise as could be believed from the accuracy of the power spectra @xcite .
we compare the predictions from the slow - roll approximation @xcite with the exact solutions from the model of power - law inflation @xcite .
we find unacceptable large errors in the predictions of multipole moments .
the reason is as follows : the primordial spectrum is best approximated at some pivot scale @xmath1 .
a small error in the spectral index gives rise to a large error at wavenumbers that differ significantly from @xmath1 , due to a large lever arm .
a natural choice for the pivot scale is the present hubble scale , but leads to extremely large errors for high multipole moments . a shift of the pivot scale to the scale of the first acoustic peak decreases these errors
dramatically ( see figure [ fig1 ] ) . in @xcite
we compare the improved ( optimal pivot scale ) slow - roll predictions with recent cmb data ( see figure 2 ) .
most data analysis so far @xcite was based on a power - law shape of the primordial spectra .
this shape is _ not _ predicted by the slow - roll approximation , only the first two terms in a taylor expansion with respect to the wavenumber coincide .
slow - roll inflation is very simple and is an attractor for many inflationary models .
inflation driven by a single field @xmath2 , may be characterized at a given moment of time @xmath3 by the parameters @xmath4_*$ ] , @xmath5_*$ ] , @xmath6_*$ ] , , where @xmath7 is the hubble rate .
the condition for inflation is @xmath8 , whereas slow - roll inflation is characterized by @xmath9 , and negligible higher derivatives .
based on these approximations the power spectrum of the bardeen potential @xmath10 and of the amplitude of gravitational waves @xmath11 reads @xcite @xmath12 , \\
\label{specsrgw } k^3p_h & = & \frac{16 h_*^2 l_{\rm pl}^2}{\pi } \biggl[1 - 2\epsilon \biggl(c+1+\ln \frac{k}{k_*}\biggr)\biggr],\end{aligned}\ ] ] where @xmath13 , @xmath14 being the euler constant .
the pivot scale is defined as @xmath15 .
fixing @xmath1 corresponds to a choice of the time @xmath3 during inflation .
the spectral indices can be obtained from @xmath16 and @xmath17 .
we call this the next - to - leading order of the slow - roll approximation ( at the leading order strictly scale - invariant spectra are predicted ) .
( 2,0.6 ) ( 0.5,0.05)(0,0)@xmath18 ( 0.22,0.535)(0,0)@xmath19 ( 0.405,0.46)(0,0)@xmath20 ( 0.575,0.36)(0,0)@xmath21 ( 0.63,0.28)(0,0)@xmath22 ( 0.74,0.185)(0,0)@xmath23 ( 0.71,0.53)(0,0)error in % ( 0.45,0.33)(0,0 ) ( 1.4,0.05)(0,0)@xmath18 ( 1.13,0.40)(0,0)@xmath19 ( 1.08,0.325)(0,0)@xmath24 ( 1.08,0.255)(0,0)@xmath25 ( 1.08,0.19)(0,0)@xmath26 ( 1.06,0.14)(3,1)0.3 ( 1.47,0.245)(0,0)@xmath23 ( 1.61,0.53)(0,0)error in % ( 1.35,0.33 ) ( 0,0 ) on the other hand , the power spectra may be calculated exactly for power - law inflation , which is characterized by a power - law behavior of the scale factor , i.e. , @xmath27 . for power - law inflation
we have @xmath28 and @xmath29 during inflation .
we use @xmath30 to parametrize the spectra , i.e. @xmath31 .
the corresponding power spectra then read @xcite @xmath32 where @xmath33^{2/(1-\epsilon ) } \gamma[1/(1-\epsilon ) + 1/2]^2/\pi$ ] , with @xmath34 . for power - law inflation
the spectral indices read : @xmath35 . in the limit
@xmath36 the power spectra ( [ specpl ] ) go to ( [ specsrd ] ) with @xmath28 and to ( [ specsrgw ] ) , respectively .
we can now calculate the multipole moments @xmath37 for the power - law and slow - roll spectra for @xmath28 .
we define the error from the slow - roll approximation as @xmath38 for similar spectra the error ( [ deferr ] ) depends only weakly on the transfer function .
this allows us to neglect the evolution of the transfer function for this purpose and to obtain an analytic result , which is plotted in figure [ fig1 ] .
the values of @xmath39 refer to the exact power - law solution . in the left figure @xmath40
gives the smallest error for the quadrupole and unacceptably large errors at high multipoles . in the right figure
the pivot scale has been chosen to minimize the error around the first acoustic peak , @xmath41 .
the corresponding condition is @xmath42 , where @xmath43 is the comoving distance to the last scattering surface and @xmath44 $ ] with @xmath45 . for @xmath46
this gives @xmath47 , where @xmath48 for @xmath49cdm .
let us now compare @xcite the predictions of slow - roll inflation with recent data from boomerang @xcite and maxima-1 @xcite , supplemented with the cobe / dmr dataset @xcite .
instead of fitting ten cosmological parameters we fix the values of otherwise measured parameters and assume that slow - roll inflation is the correct theory .
in figure [ fig2 ] we present the sum of scalar and tensor cmb band power for a @xmath49cdm model with @xmath50 and @xmath51 .
the boltzmann code used here was developed by one of us ( a.r . ) .
we see without a @xmath52 analysis that qualitatively different inflationary models are consistent with the observations : both models with @xmath53 give reasonable fits , one of these models has a flat scalar spectrum ( with @xmath54 ) , the other one has a negative tilt ( with @xmath55 ) .
both models have an important contribution of gravitational waves ( @xmath56 ) .
we emphasize that the generic slow - roll predictions ( [ specsrd ] ) and ( [ specsrgw ] ) do not have a power - law shape .
this fact induces large differences to multipole moments that are predicted under the assumption that the power - law shape ( [ specpl ] ) is the generic inflationary prediction . besides using the correct primordial spectrum
a clever choice of the pivot scale can hide unavoidable uncertainties of the multipole moments in the cosmic variance on one side and in the instrumental noise on the other side of the spectrum .
a. a. starobinsky , _ jetp lett . _ * 30 * , 682 ( 1979 ) ; v. mukhanov and g. chibisov , _ jetp lett . _
* 33 * , 532 ( 1981 ) ; a. guth and s. y. pi , _ phys .
lett . _ * 49 * , 1110 ( 1982 ) ; s. hawking , _ phys .
* 115b * , 295 ( 1982 ) ; a. a. starobinsky , phys . lett . * 117b * , 175 ( 1982 ) .
de bernadis et al . , _ nature _ * 404 * , 955 ( 2000 ) .
s. hanany et al .
, astro - ph/0005123 ( 2000 ) .
a. e. lange et al . , astro - ph/0005004 ( 2000 ) ;
a. balbi et al . , astro - ph/0005124 ( 2000 ) ; a. jaffe et al . , astro - ph/0007333 ( 2000 ) .
j. martin and d. j. schwarz , _ phys . rev .
d _ , to appear , astro - ph/9911225 ( 1999 ) .
see , e.g. , a. d. linde , _ particle physics and inflationary cosmology _ , harwood ( chur , switzerland , 1990 ) .
i. j. grivell and a. r. liddle , _ phys .
d _ * 54 * , 7191 ( 1996 ) ; _ ibid .
_ * 61 * , 081301 ( 2000 ) .
e. d. stewart and d. h. lyth , _ phys .
* 302b * , 171 ( 1993 ) .
l. f. abbott and m. b. wise , nucl .
b244 * , 541 ( 1984 ) .
j. martin , a. riazuelo , and d. j. schwarz , _ astrophys .
j. letters _ , to appear , astro - ph/0006392 ( 2000 ) .
j. martin and d. j. schwarz , _ phys .
d _ * 57 * , 3302 ( 1998 ) . c. l. bennett et al .
, _ astrophys .
j. _ * 464 * , l1 ( 1996 ) . | inflationary predictions of the cosmic microwave background anisotropy are often based on the slow - roll approximation .
we study the precision of these predictions and compare them with the recent data from boomerang and maxima-1 . |
stars commonly form in groups or clusters within turbulent molecular clouds on timescales which are of about a million years ( hillenbrand 1997 ) .
typical young stellar aggregates have sizes of roughly 1 pc and consist of a few hundred stars .
recent observations have also shown that most stars form in eccentric binary systems and that the binary frequency of young stars is about two times higher than that of main sequence stars in the solar neighbourhood ( ghez et al .
1997 , khler & leinert 1998 ) .
this reflects the fact that secular dynamical processes within newly formed stellar groups tend to reduce their binary fraction over time .
recent numerical modeling ( kroupa 1995 , 1998 ) demonstrates that encounters between binaries can lead to the dissolution of aggregates on timescales of several hundred million years and that stochastic close stellar encounters which are in general very energetic can lead to the dissociation of the widest binaries .
binary dissociation occurs at binary orbital periods greater than about 3000 yrs , corresponding to separations of order a few 100 au .
it is therefore reasonable to expect that most single main sequence stars actually formed as part of a wider binary system which was disrupted through interactions within a young stellar cluster .
even after a proto - star becomes detached from its companion , or if it is born as a single star , encounters by passing stars would occur before the dissolution of the stellar cluster .
the timescale for encounters with pericenter distance @xmath7au may be comparable to the dissolution timescale of the stellar cluster ( laughlin & adams 1998 ) .
thus , if the sun formed in such a clustered environment , it most likely experienced a few close encounters with a transient binary companion or with passing stars at pericenter distances of order 100 au , before the break up of the stellar cluster .
laughlin & adams ( 1998 ) have suggested that the large eccentricities of extrasolar planets associated with cyg b and 14 her could have been pumped up by interactions with passing _
binary systems _ in an open cluster . here
, we will consider interactions of a star ( the proto - sun ) having a protoplanetary system which encounters a passing single star .
in general interactions with a binary system are more disruptive to the protoplanetary system than those with a single star . since we seek to model the solar system the interactions we consider are necessarily much less disruptive to the planetary system than those considered by laughlin & adams .
( more distant encounters with passing binary systems may lead to similar results . )
such an encounter will generally affect the dynamical and material structure of the solar protoplanetary disk and , provided internal conditions allow , the planetesimal disk will remain imprinted with this signature over much of the main sequence lifetime of the star . in this _
letter _ we study the dynamical effects of the stellar encounters on protoplanetary disks and point out that the orbital distribution of edgeworth - kuiper belt ( ekb ) objects may indicate that the solar system has experienced close stellar encounters .
we demonstrate that puzzling kinematical features in the orbital distribution of the ekb objects can be explained naturally if the sun formed as a member of a stellar cluster and experienced a stellar encounter ( or series of encounters ) with @xmath5@xmath6au .
the observed ekb objects observed at multiple oppositions or during relatively long duration are shown in fig.1 ( e.g. see marsden s web site , http://cfa-www.harvard.edu/@xmath8graff/lists/tnos.html ) .
the increasing numbers of ekb objects being revealed by observations presently fall into three distinct groups .
firstly , many objects have semimajor axes close to the 3:2 resonance with neptune s orbit ( located at 39.5au ) , and these display a wide range of eccentricities and inclinations ( each up to @xmath9 ) . secondly ,
outside @xmath10 au , the objects have slightly lower average eccentricity ( @xmath11 ) and inclination ( @xmath11 radian ) . at semimajor axes inside @xmath12au , and between @xmath13au and @xmath10au , there are unpopulated regions ( hereinafter `` gaps '' ) .
the cut - off outside @xmath14au may imply depletion of objects but it could also be due to the present observational sensitivity limit ( jewitt , luu , and trujillo 1998 ; gladman et al .
the third group is comprised of the scattered disk objects ( duncan and levison 1997 ) , which have experienced close approach with neptune .
pericenter for the scattered disk objects is located near neptune s orbit .
an example is tl66 with @xmath15 and @xmath16au , which is outside the range of fig.1 .
secular perturbations by the giant planets can account for the gap between @xmath13au and @xmath10au ( duncan , levison , & budd 1995 ) .
they can not account for the other features ( duncan , levison , & budd 1995 ) . the model of sweeping mean motion resonances due to neptune s outward migration successfully accounts for the concentrated distribution at the 3:2 resonance as well as for the gap inside @xmath12au ( malhotra 1995 ) .
this model also predicts that a large accumulation ought to occur at neptune s 2:1 resonance ( located at @xmath17au ) with a cleared gap interior to the present resonant location .
if the number of objects captured by the 2:1 sweeping resonance is similar to that by the 3:2 resonance , it may be expected that more objects should now be detected near the 2:1 resonance ( jewitt , luu , and trujillo 1998 ) . however , the current population near the 2:1 resonance is still poorly constrained owing to the observational sensitivity limit .
the migration speed of neptune also affects the relative population between the 3:2 and 2:1 resonances ( ida et al .
1999 ) . in summary ,
the good agreement of the theoretical predictions by malhotra ( 1995 ) with the observations for the objects near the 3:2 resonance supports the sweeping of mean motion resonances .
the relatively high eccentricities and inclinations found outside @xmath10au can not be accounted for by long - range secular perturbations of the planets .
the velocity dispersion of these observed objects exceeds their surface escape velocity for most objects , which can not be explained by internal gravitational scattering ( safronov 1969 ) .
the capture probability of the sweeping 3:2 resonance becomes small , and the gap inside 39au can not be created , when the initial eccentricity exceeds @xmath18 ( malhotra 1995 ) . the objects with @xmath19 would not be swept and remain inside 39au , although a clear gap is presently observed inside 39au .
thus , the mechanism to pump up velocity dispersion outside @xmath10 au should satisfy the condition of having occurred in a highly localized manner to keep @xmath20 and @xmath21 small enough inside 39au , although we note that objects with @xmath22 , inside 39au , _ can _ be destabilized by planetary perturbations in the age of the solar system ( duncan , levison , & budd 1995 ) .
some models have been proposed to account for the high @xmath20 and @xmath21 outside @xmath10 au .
the earth - sized bodies that are thought to have once existed in the formation stage and were subsequently ejected might have been able to pump up the velocity dispersion ( stern 1991 ; morbidelli and valsecchi 1997 ; petit , morbidelli , and valsecchi 1999 ) .
partial trapping by sweeping of the 2:1 resonance might have also pumped up the eccentricities outside @xmath10au ( hahn and malhotra 1999 ) . here
we propose another mechanism , stellar encounters , to dynamically heat the planetesimal disk outside @xmath10au .
while the two former mechanisms are associated with processes occurring after the formation of neptune , the stellar encounter model can operate before neptune s formation as well .
although all these mechanisms may be able to account for the dynamical heating of the velocity dispersion between the 3:2 and 2:1 resonances , the predicted velocity dispersion beyond the 2:1 resonance ( which has not been observed up to now ) is expected to be quite different in our model , as discussed below .
we have investigated the possibility that stellar encounters with the young solar nebula could have increased the eccentricity @xmath20 and inclination @xmath21 of ekb objects presently located outside @xmath10au . in our modeling
we assume : 1 .
a single star passes by the proto - sun on a nearly parabolic orbit and perturbs the planetesimal system .
the passing star may be weakly bound to the proto - sun in which case we can consider a series of encounters .
the pericenter distance ( @xmath2 ) of the encounter(s ) is on the order of 100au .
planetesimals with the present ekb object mass ( @xmath23@xmath24 g ) are formed on low-@xmath20 and low-@xmath21 orbits prior to the first encounter that pumps up @xmath20 and @xmath21 significantly . as discussed in the introduction , the assumptions 1 and 2 are consistent with recent observations and numerical modeling .
if we are only concerned with the effects of stellar encounters on the protoplanetary system , the assumption 3 is not necessary .
however , to apply our results to the ekb the assumption 3 is needed in our model because the induced velocity dispersion is larger than the surface escape velocity , which we would expect to halt planetesimal agglomeration ( see below ) . according to conventional models ( e.g. safronov 1969
; goldreich & ward 1973 ; hayashi , nakazawa , & nakagawa 1985 ) , dust grains settle to the equatorial plane of the nebula and subsequent gravitational instability of the dust layer results in planetesimal formation .
the dust grain sedimentation timescale may be only @xmath25@xmath26 yrs ( e.g. , hayashi , nakazawa , & nakagawa 1985 ) and the gravitational instability operates over a timescale that is comparable to the orbital period ( reviewed by papaloizou & lin 1995 ) .
first - born planetesimals have masses of a few @xmath27 g ( e.g. hayashi , nakazawa , & nakagawa 1985 ) , which is already comparable to the masses of the present ekb objects
. however , nebula turbulence may prevent dust grains from settling onto the equatorial plane , so that the gravitational instability does not occur ( e.g. weidenschilling & cuzzi 1993 ) .
if this is the case , planetesimal accretion up to the present size of the ekb objects would require @xmath28@xmath29 yrs ( e.g. stern & colwell 1997 ) , so that assumption 3 may be too restrictive for our model of repeated encounters in an eccentric binary and only the model of flyby stellar encounters ( before dissolution of a stellar aggregate ) would be allowed
. a series of numerical simulations to test the effect of stellar companion encounters in protoplanetary disks has been performed .
we consider collisionless particles ( corresponding to planetesimals ) , orbiting initially on coplanar circles around a primary star ( the proto - sun ) .
this particle disk encounters a hypothetical companion star .
the orbital changes of the test particles are integrated taking into account the gravitational forces of the primary and the companion star using a fourth order predictor - corrector scheme .
many different encounter geometries and companion masses have been examined .
if the scale length is defined by the pericenter distance @xmath2 of the encounter , each encounter is characterized by : the companion mass ( @xmath30 ) , the inclination angle of the companion orbit relative to the initial disk ( @xmath31 ) , and the orbital energy or eccentricity of the perturber ( @xmath32 ) ( ostriker 1994 ) . in the models ,
typically @xmath33 test particles were initially distributed in the region @xmath34
@xmath35 , where @xmath36 denotes semimajor axis .
the initial surface number density @xmath37 is proportional to @xmath38 .
since we consider test particles which do not interact with one another , the particular choice of disk mass or surface number density profile does not affect the generality of the results . the initial eccentricity and inclination ( @xmath39 and @xmath40 ) of the particles
are taken to be @xmath41 .
figure 2 shows the eccentricity and inclination of the particles after the encounter as a function of @xmath42 , in the case with @xmath43 and @xmath44 .
inclination angle @xmath31 is ( a ) 5 degrees , ( b ) 30 degrees , and ( c ) 150 degrees with the line of nodes along the @xmath45-axis . the spatial distribution in the case of @xmath46 degrees is shown in fig.3 .
as shown in fig.2 the encounter leads to a strong increase in @xmath20 and @xmath21 in the outer parts of the disk . in the case of @xmath46 degrees ,
@xmath20 and @xmath21 are pumped up only slightly ( @xmath41 ) at @xmath47 , while they are pumped up highly ( @xmath48 ) at @xmath49 .
note that the former condition is similar to the orbital stability condition of bounded three - body systems ( e.g. black 1982 ) . at @xmath50 a large fraction of particles
are ejected : respectively 70% and 95% of objects with initial @xmath51 and @xmath52 .
the remaining particles at @xmath50 have large eccentricities .
the other @xmath31 cases show similar features .
the spike in @xmath21 at @xmath53 coincides with the 3:1 commensurability of the unperturbed disk orbital frequency and the companion s orbital frequency at pericenter . since the companion s angular velocity at pericenter is @xmath54^{1/2}$ ]
, the 3:1 commensurability is located at @xmath55^{-1/3 } \simeq 0.30 $ ] , for prograde encounters .
thus the highly localized character of the disk response in this region appears to be associated with a corotation resonance occurring near pericenter ( korycansky & papaloizou 1995 ) .
as shown in fig.3 , long - lived features in the spatial distribution are the inclined bar - like envelope and prominent one - armed spiral , both due to close correlations in the longitudes of perihelion and ascending node .
precession of the longitudes due to , for example , the nebula potential would gradually destroy the features .
if the velocity dispersion is greater than the surface escape velocity of a planetesimal , collisions between planetesimals is too destructive ( backman , dasgupta , & stencel , 1995 ) and growth of planetesimals is halted .
the surface escape velocity of a planetesimal with mass @xmath56 and density @xmath57 is @xmath58cm / s .
the velocity dispersion is @xmath59cm / s , where @xmath60 is keplerian velocity . as a result , in the region where
pumped - up @xmath20 or @xmath61 , planetesimal growth would be inhibited .
the steep radial gradient of @xmath20 and @xmath21 seen in fig.2 indicates that there exists a well defined boundary for planetesimal growth at @xmath62@xmath63 : outside this region planetesimal growth is greatly inhibited while it is not affected at all inside . for different encounter parameters the distribution of the pumped - up @xmath20 and @xmath21 is generally very similar to fig .
2 , except for the length scale @xmath2 . in other words , for different encounters the distribution of particles in fig .
2 shifts towards larger or smaller values of @xmath42 , except that @xmath21 is not pumped in the special case of a coplanar encounter . in general more massive companions and
lower inclination encounters yield stronger interactions .
for example , the distribution shifts as @xmath64 .
higher energy ( i.e. more eccentric ) encounters result in stronger effects further into the disk owing to the improved coupling there .
encounters with @xmath31 closer to 90 degrees result in higher @xmath21 relative to @xmath20 .
the @xmath65 degree encounter has the same amplitude of inclined angle as that with @xmath46 degrees .
hence , the pumped - up @xmath21 is similar , but @xmath20 is smaller ( see fig .
2 ) , because the parameter @xmath65 degrees gives a retrograde encounter and the relative velocity between disk particles and the passing star is therefore significantly larger , resulting in only very weak coupling .
as mentioned above , if the proto - sun had a transient binary companion , the proto - sun may have experienced a few close encounters with the companion before the binary system broke up . in this case
, the individual encounters would have similar parameters and @xmath20 and @xmath21 would be pumped up cumulatively with each encounter , so that the perturbed forms of @xmath20 and @xmath21 would be preserved except for shifts towards smaller values of @xmath42 .
we shall now consider an encounter that gives the required @xmath20 and @xmath21 distributions for the inner ekb .
as stated above , such encounters with @xmath2 on the order of @xmath3 au may be reasonable for the protosolar system .
we performed a similar simulation to that presented in fig .
2b ( @xmath46 degrees ) except @xmath66 .
overall features of the pumped - up @xmath20 and @xmath21 are quite similar to fig
. 2b ( see fig .
4a ) . with @xmath67au
, we randomly selected @xmath68 particles in the range @xmath69au@xmath70au from the results , to compare them with the observed numbers of ekb objects .
the selected distribution is shown in fig .
4a . for @xmath71au , @xmath20 and @xmath21
are as large as those of the observed ekb objects .
some objects that originally had larger @xmath36 are scattered to this region with very high @xmath20 and @xmath21 . however , at @xmath72au , @xmath20 is still small enough ( @xmath73 ) to allow the formation of a gap inside @xmath12au via resonance sweeping , without the need for any other processes , e.g. long - term orbital destabilization . in order to study the sweeping mean motion resonances we also performed simulations similar to other authors ( malhotra 1995 , ida et al .
1999 ) , starting from the resultant distribution of particles after the stellar encounter ( fig .
the proto - neptune with a mass of @xmath74 g ( comparable to the present neptunian mass ) was artificially moved from @xmath75au to @xmath69au ( therefore the 3:2 resonance moved from 30au to 39.5au ) , on a circular zero inclination orbit .
we assumed a time dependence for the semimajor axis evolution given by : @xmath76$]au , and a migration timescale @xmath77 yrs .
if we chose a longer migration time , more particles are captured by the 2:1 resonance and a gap is created interior to the resonance while the capture probability of the 3:2 resonance would remain much as before ( ida et al .
1999 ) .
the result after the sweeping is shown in fig.4b .
the objects between @xmath13au and @xmath10au would be destabilized by a long - term secular resonance ( duncan , levison , & budd 1995 ) .
the objects that have high eccentricity and are not trapped by mean - motion resonances may experience close encounters with neptune and go to the scattered disk. sweeping secular resonances , which we do not include in our simulations , may alter the inclination distribution both near the 3:2 resonance and beyond 42 au ( malhotra , duncan , & levison 1999 ) .
thus , our result is consistent with the observed distribution in fig.1 .
in particular , the puzzling high values of @xmath20 at @xmath71au are explained without diminishing the capture probability of the sweeping 3:2 resonance . different geometry of a stellar encounter , multiple encounters , or an encounter with a passing binary system might result in a better match .
the typical damping time of @xmath20 and @xmath21 due to hydrodynamic gas drag at @xmath13au is @xmath78 yrs ( adachi , hayashi , & nakazawa 1976 ) , for a typical minimum mass solar nebular model ( hayashi 1981 ) .
this is much longer than the lifetime of disk gas , inferred from observations , being of order @xmath79@xmath80 yrs ( e.g. zuckerman , forveille , & kastner 1995 ) .
also the two - body relaxation time and collision time for the presently estimated surface density at @xmath13au is longer than the solar system age ( stern 1995 , 1996 ; davis & farinella 1996 ) . hence , the orbital elements of the present ekb objects should not have changed significantly after the orbital perturbation .
it is expected that the orbital distribution in @xmath20 , @xmath21 and @xmath36 after the encounter reflects that observed today .
our simulations show that early stellar encounters would lead to interesting features in the young solar nebula that might explain the structure of the outer part of the ekb .
the stellar encounters would occur on timescales of dissolution of stellar aggregates , which is of the order of @xmath1 yrs .
this may allow the ekb objects to grow to their observed sizes before the encounters .
the objects initially inside 30au would be strongly scattered to form the scattered disk during neptune s migration ( duncan & levison 1997 ) . the objects with initial @xmath36 from 30au to 40au would be captured by the sweeping of the 3:2 resonance with resultant high @xmath20 and @xmath21 .
outside 40au , the stellar perturbations are strong enough to pump up @xmath20 and @xmath21 to @xmath48 .
once their velocity dispersion is pumped up to more than the surface escape velocity , collisions between the ekb objects would produce copious amounts of dust particles which would be removed by gas drag , poynting - robertson drag , and radiation pressure - driven ejection ( stern 1995 ; backman , dasgupta , & stencel 1995 ) .
the initial surface density is therefore eroded by virtue of its dynamical state , the present ekb objects being remnants that have avoided significant erosion ( stern 1996 ; davis & farinella 1996 ) .
this result could explain the fact that the observationally inferred surface density in the ekb is much lower than that extrapolated from a minimum mass solar nebula model ( e.g. stern 1995 ; weissman and levison 1997 ) .
detailed numerical modeling of the subsequent collisional evolution of the perturbed ekb is required to test this hypothesis .
our model predicts that there should be a steep increase in @xmath20 and @xmath21 with semimajor axis .
in contrast , stirring by by earth - sized bodies would predict decrease in @xmath20 and @xmath21 and that by partial trapping by the neptunian sweeping 2:1 resonance predicts a cold disk beyond 50au .
future observations can validate these models by the trend of radial dependences in @xmath20 and @xmath21 .
if @xmath20 and @xmath21 systematically increase beyond 50au , our model is supported .
the high eccentricities and inclinations that follow immediately from such an encounter also have a number of consequences for extrasolar planetary systems .
firstly , as stated previously , the augmented velocity dispersion amongst planetesimals promotes the production of dust particles . this can significantly increase the dust replenishment rates and lead to more prominent circumstellar disks around some main sequence stars ( stern 1995 ; kalas & jewitt 1996 ; holland et al .
the existence of the dust disks may reflect stellar encounters in the formation epoch .
secondly , as stated above , planetesimal growth could be forestalled in the outer region of the disk by a stellar encounter .
this situation could be reflected in the fact that neptune marks the outer boundary of our planetary system at @xmath69au .
thus the existence of substantial planetary bodies outside @xmath81au would be inconsistent with our model . finally , we comment that recent advances in star formation theory and observation suggest that such stellar encounters with disks as those considered here should not be viewed as unique catastrophic events but as an integral part of the star- and planetary - system formation process .
adachi , i. , hayashi , c. , & nakazawa , k. 1976 .
phys . , 56 , 1756 backman , d. e. , dasgupta , a. , & stencel , r. e. 1995 , apj , 450 , l35 black , d. c. 1982 , aj , 87 , 1333 davis , d. r. & farinella , a. p. 1996 , icarus , 125 , 50 davis , d. r. & farinella , a. p. 1998 ,
lpsc abstract duncan , m. j. , levison , h. f. , & budd , s. m. 1995 , , 110 , 3073 ghez , a.m. , mccarthy , d.w . ,
patience , j.l .
, & beck , t.l 1997 apj , 481 , 378 gladman , b. et al .
1998 , aj , 116 , 2042 goldreich , p. & ward , w. r. 1973 , apj , 183 , 1051 hahn , m. j. & malhotra , r. 1999 , , in press hayashi , c. 1981 , prog .
phys . suppl . ,
70 , 35 hayashi , c. , nakazawa , k. , & nakagawa , y. 1985 in protostars and planets ii .
d. c. black & m. s. matthew ( tucson : univ . of arizona press ) , 1100 hillenbrand , l.a .
1997 , aj , 113 , 1733 holland , w. , greaves , j. , zuckerman , b. , webb , r. , mccarthy , c. , coulson , i. , walther , d. , dent , w. , gear , w. , & robson , i. 1998 , nature , 392 , 788 ida , s. , bryden , g. , lin , d. n. c. , & tanaka , h. 1999 , apj , submitted jewitt , d. , luu , j. & trujillo , c. 1998 , aj , 115 , 2125 kalas , p. & jewitt , d. 1996 , aj , 111 , 1374 kroupa , p. 1995 , mnras , 277 , 1507 kroupa , p. 1998 ,
mnras , 298 , 231 khler , r & leinert , c.h .
1998 a&a , 331 , 977 laughlin , g. & adams , f. c. 1998 , apj , 508 , l171 malhotra , r. 1995 , , 110 , 420 malhotra , r. , duncan , m. j. , & levison , h. f. 1999 , in protostars and planets iv eds . v. mannings & a. boss ( tucson : univ . of arizona press ) , in press morbidelli , a. & valsecchi , g. b. 1997 , icarus , 128 , 464 ostriker , e.c .
1994 , apj , 424 , 292 papaloizou , j.c.b . & lin d.n.c .
1995 , ara&a , 33 , 505 petit , j - m , morbidelli , a. & valsecchi , g. b. 1999 , icarus , in press safronov , v. s. 1969 , evolution of the protoplanetary cloud and formation of the earth and planets .
( moscow : nauka press ) stern , s. a. 1991 , icarus , 90 , 271 stern , s. a. 1995 , , 110 , 856 stern , s. a. 1996 , , 112 , 1203 stern , s. a. & colwell , j. e. 1997 , apj , 490 , 879 weidenschilling , s. j. & cuzzi , j. n. 1993 , in protostars and planets iii eds .
levy & j.i .
lunine ( tucson : univ . of arizona press ) , p. 1031
zuckerman , b. , forveille , & kastner , j. h. 1995 , nature , 373 , 494 | we show that early stellar encounters can explain the high eccentricities and inclinations observed in the outer part ( @xmath0au ) of the edgeworth - kuiper belt ( ekb ) .
we consider the proto - sun as a member of a stellar aggregation that undergoes dissolution on a timescale @xmath1 yrs , such that the solar nebula experiences a flyby encounter at pericenter distance ( @xmath2 ) on the order of @xmath3au .
using numerical simulations we show that a stellar encounter pumps the velocity dispersion in the young solar nebula in the outer parts . in the case of a nearly parabolic encounter with a solar - mass companion the velocity dispersion at @xmath4
is pumped up to such an extent that collisions between planetesimals would be expected to become highly disruptive , halting further growth of planetesimals .
this has the consequence that planet formation is forestalled in that region .
we also find that a stellar encounter with pericenter distance @xmath5@xmath6au could have pumped up the velocity dispersion of ekb objects outside 42au to the observed magnitude while preserving that inside neptune s 3:2 mean - motion resonance ( located at 39.5au ) .
this allows for the efficient capture of objects by the resonance during a phase of orbital migration by proto - neptune , which we also test with simulations .
we point out that such a stellar encounter generally affects the dynamical and material structure of a protoplanetary disk and the planetesimal distribution can remain imprinted with this signature over much of the main sequence lifetime of the star . in particular , our results support the notion that an analogous process has operated in some recently observed extrasolar dust disks . |
in this paper , we consider the use of krylov subspace methods for solving large , sparse linear systems @xmath5 , where @xmath6 .
we will focus on the conjugate gradient ( cg ) method @xcite , which is used when @xmath7 is symmetric positive definite .
given an initial approximate solution @xmath8 and corresponding residual @xmath9 , the cg method iteratively updates the approximate solution using a sequence of nested krylov subspaces @xmath10 , where @xmath11 denotes the @xmath12-dimension krylov subspace with matrix @xmath7 and starting vector @xmath13 . in iteration @xmath12 , the updated approximate solution @xmath14 is chosen by imposing the galerkin condition @xmath15 .
thus each iteration of cg requires a matrix - vector product with @xmath7 in order to expand the dimension of the krylov subspace and a number of inner products to perform the orthogonalization . on modern computer architectures ,
the speed with which the matrix - vector products and inner products can be computed is limited by communication ( i.e. , the movement of data ) .
this limits the potential speed of individual iterations attainable by an implementation of cg . to perform a sparse matrix - vector product in parallel ,
each processor must communicate entries of the source vector and/or the destination vector that it owns to neighboring processors .
inner products require a global synchronization , i.e. , the computation can not proceed until all processors have finished their local computation and communicated the result to other processors . for large - scale sparse problems on large - scale machines ,
the cost of synchronization between parallel processors can dominate the run - time ( see , e.g. , the exascale computing report @xcite ) .
research efforts toward removing the performance bottlenecks caused by communication in cg and other krylov subspace methods have produced various approaches .
one such approach are the @xmath0-step krylov subspace methods ( also called `` communication - avoiding '' krylov subspace methods ) ; for a thorough treatment of background , related work , and performance experiments , see , e.g. , the theses @xcite . in @xmath0-step krylov
subspace methods , instead of performing one iteration at a time , the iterations are performed in blocks of @xmath0 ; i.e. , in each iteration , the krylov subspace is expanded by @xmath1 dimensions by computing @xmath1 new basis vectors and then all inner products between the new basis vectors needed for the next @xmath0 iterations are computed in one block operation . in this way , computing the inner products for @xmath0 iterations only requires a single global synchronization , decreasing the synchronization cost per iteration by a factor of @xmath1 .
this approach has been shown to lead to significant speedups for a number of problems and real - world applications ( see , e.g. , @xcite ) . in the remainder of the paper
, we will refer to the matrices whose columns consist of the @xmath1 basis vectors computed in each block as _
@xmath0-step basis matrices_. further details of the @xmath0-step cg method are discussed in section [ sec : sksms ] .
we emphasize that our use of the overloaded term `` @xmath0-step methods '' here differs from other works , e.g. , @xcite and @xcite , in which ` @xmath0-step method ' refers to a type of restarted lanczos procedure . in exact arithmetic
the @xmath0-step cg method produces the exact same iterates as the classical cg method , but their behavior can differ significantly in finite precision . in both @xmath0-step and classical krylov subspace methods ,
rounding errors due to finite precision arithmetic have two basic effects : a decrease in attainable accuracy and a delay of convergence .
it has long been known that for @xmath0-step krylov subspace methods , as @xmath0 is increased ( and so the condition numbers of the @xmath0-step basis matrices increase ) , the attainable accuracy decreases and the convergence delay increases relative to the classical cg method ( see , e.g. , @xcite ) . at the extreme ,
if the parameter @xmath0 is chosen to be too large , the @xmath1-dimensional bases computed for each block can be numerically rank deficient and the @xmath0-step method can fail to converge .
this sensitive numerical behavior poses a practical obstacle to optimizing the performance of @xmath0-step methods , and diminishes their usability and reliability . in a setting where the performance of cg is communication - bound
, we expect that up to some point , increasing @xmath0 will decrease the time per iteration .
if we pick @xmath0 only based on minimizing the time per iteration , however , we can run into problems .
first , the finite precision error may cause a large convergence delay , negating any potential performance gain with respect to the overall runtime . since the number of iterations required for convergence for a given @xmath0 value is not known a priori
, choosing the @xmath0 value that results in the fastest time - to - solution is a difficult problem .
second , the chosen @xmath0 parameter may cause @xmath0-step cg to fail to converge to the user - specified accuracy . in this case , the particular problem is _ unsolvable _ by the @xmath0-step cg method . requiring the user to choose the parameter @xmath0 thus diminishes the practical usability of @xmath0-step krylov subspace methods
it is therefore imperative that we develop a better understanding of the convergence rate and accuracy in finite precision @xmath0-step cg and other @xmath0-step krylov subspace methods .
our hope is that by studying the theoretical properties of methods designed for large - scale computations in finite precision , we can develop methods and techniques that are efficient , capable of meeting application - dependent accuracy constraints , and which do not require that the user have extensive knowledge of numerical linear algebra . toward this goal ,
in this paper we develop a variable @xmath0-step cg method in which the @xmath0 values can vary between blocks and are chosen automatically by the algorithm such that a user - specified accuracy is attained .
we call this the _ adaptive @xmath0-step cg method_. the condition for determining @xmath0 in each block is derived using a bound on the size of the gap between the true residuals @xmath16 and the updated residuals @xmath17 computed in finite precision .
we show that in order to prevent a loss of attainable accuracy , in a given block of iterations , the condition number of the @xmath0-step basis matrix should be no larger than a quantity related to the inverse of the 2-norms of the residuals computed within that block .
in other words , as the method converges , more ill - conditioned bases ( i.e. , larger @xmath0 values ) can be used without a loss of attainable accuracy .
furthermore , we show that the adaptive @xmath0-step cg method can be implemented in a way that does not increase the synchronization cost per block versus ( fixed ) @xmath0-step cg . whereas the fixed @xmath0-step cg method may not attain the same accuracy as classical cg ,
our numerical experiments on a set of example matrices demonstrate that our adaptive @xmath0-step cg method can achieve the same accuracy as classical cg , while still reducing the number of synchronizations required by up to over a factor of @xmath18 . in section [ sec : sksms ] we give a brief background on @xmath0-step krylov subspace methods and detail the @xmath0-step cg method , and in section [ sec : relatedwork ] we discuss related work .
section [ sec : varscg ] presents our main contribution ; in section [ sec : theory ] , we derive a theoretical bound on how large the @xmath0-step basis matrix condition number can be in terms of the norms of the updated residuals without affecting the attainable accuracy , and in section [ sec : algorithm ] , we detail the adaptive @xmath0-step cg algorithm that makes use of this bound in determining the number of iterations to perform in each block .
numerical experiments for a small set of test matrices are presented in section [ sec : experiments ] .
we discuss possible extensions and future work and briefly conclude in section [ sec : conclusion ] .
the basic idea behind @xmath0-step krylov subspace methods is to grow the underlying krylov subspace @xmath1 dimensions at a time and perform the necessary orthogonalization with a single global synchronization . within a block of @xmath0 iterations ,
the vector updates are performed implicitly by updating the coordinates of the iteration vectors in the @xmath1-dimensional basis spanned by the columns of the @xmath0-step basis matrix .
the algorithmic details vary depending on the particular krylov subspace method considered , but the general approach is the same .
there is a wealth of literature related to @xmath0-step krylov subspace methods .
we give a brief summary of early related work below ; a thorough bibliography can be found in ( * ? ? ?
* table 1.1 ) .
the @xmath0-step approach was first used in the context of krylov subspace methods by van rosendale @xcite , who developed a variant of the parallel conjugate gradient method which minimizes inner product data dependencies with the goal of exposing more parallelism .
the term `` @xmath0-step krylov subspace methods '' was first used by chronopoulos and gear , who developed an @xmath0-step cg method @xcite .
this work was followed in subsequent years by the development of other @xmath0-step variants of orthomin and gmres @xcite , arnoldi and symmetric lanczos @xcite , minres , gcr , and orthomin @xcite , nonsymmetric lanczos @xcite , and orthodir @xcite .
walker similarly used @xmath0-step bases , but with the goal of improving stability in gmres by replacing the modified gram - schmidt orthogonalization process with householder qr @xcite .
many of the earliest @xmath0-step krylov subspace methods used monomial bases ( i.e. , @xmath19 $ ] ) for the computed @xmath0-step basis matrices , but it was found that convergence often could not be guaranteed for @xmath20 ( see , e.g. , @xcite ) .
this motivated research into the use of other more well - conditioned bases for the krylov subspaces , including scaled monomial bases @xcite , chebyshev bases @xcite , and newton bases @xcite .
the growing cost of communication in large - scale sparse problems has created a recent resurgence of interest in the implementation , optimization , and development of @xmath0-step krylov subspace methods ; see , e.g. , the recent works @xcite .
in this work , we focus on the @xmath0-step conjugate gradient method ( algorithm [ alg : scg ] ) , which is equivalent to the @xmath0-step cg method ( algorithm [ alg : scg ] ) in exact arithmetic . in the remainder of this section ,
we give a brief explanation of the @xmath0-step cg method ; further details on the derivation , implementation , and performance of the @xmath0-step cg method can be found in , e.g. , @xcite .
@xmath21 @xmath22 @xmath23 @xmath24 @xmath25 @xmath26 the @xmath0-step cg method consists of an outer loop , indexed by @xmath2 , which iterates over the blocks of iterations , and an inner loop , which iterates over iterations @xmath27 within a block . for clarity ,
we globally index iterations by @xmath28 .
it follows from the properties of cg that for @xmath29 , @xmath30 then the cg vectors for the next @xmath0 iterations lie in the sum of the column spaces of the matrices @xmath31 , \text { with } \text{span}(\mathcal{p}_{k , s})={\mathcal{k}}_{s+1}(a , p_{sk+1 } ) \quad\text{and } \nonumber \\ \mathcal{r}_{k , s } & = [ \rho_0(a)r_{sk+1 } , \dots , \rho_{s-1}(a)r_{sk+1 } ] , \text { with } \text{span}(\mathcal{r}_{k , s})={\mathcal{k}}_{s}(a , r_{sk+1 } ) , \label{eq : cg - krylovbasis}\end{aligned}\ ] ] where @xmath32 is a polynomial of degree @xmath33
satisfying the three - term recurrence @xmath34 we define the @xmath0-step basis matrix @xmath35 $ ] and define @xmath36 to be the same as @xmath37 except with all zeros in columns @xmath38 and @xmath39 .
note that the @xmath0 s in the subscripts are unnecessary here , but will be useful in describing the variable @xmath0-step cg method later on . under certain assumptions on the nonzero structure of @xmath7 ,
@xmath37 can be computed in parallel in each outer loop for the same asymptotic latency cost as a single matrix - vector product using the so - called `` matrix powers kernel '' ( see @xcite for details ) . since the columns of @xmath37 satisfy , we can write @xmath40 where @xmath41 is a @xmath42 block tridiagonal matrix defined by the 3-term recurrence coefficients in . by , there exist vectors @xmath43 , @xmath44 , and @xmath45 that represent the coordinates of the cg iterates @xmath46 , @xmath47 , and @xmath48 , respectively , in @xmath37 for @xmath29 .
that is , @xmath49 therefore , in the inner loop of @xmath0-step cg , rather than update the cg vectors explicitly , we instead update their coordinates in @xmath37 , i.e. , for @xmath27 , @xmath50 thus the matrix - vector product with @xmath7 in each iteration becomes a matrix - vector product with the much smaller matrix @xmath51 .
this along with the length-@xmath52 vector updates can be performed locally by each processor in each inner loop iteration .
we can also reduce communication in computing the inner products .
letting @xmath53 , and using and , @xmath54 and @xmath55 can be computed by @xmath56 the matrix @xmath57 can be computed with one global synchronization per outer loop iteration , which , in terms of asymptotic parallel latency , costs the same as a single inner product .
as @xmath41 and @xmath57 are dimension @xmath42 , @xmath54 and @xmath55 can be computed locally by each processor within the inner loop .
[ alg : cacg : akx ] [ alg : cacg : g ] assemble @xmath41 such that holds @xmath58^{t}$ ] , @xmath59^{t}$ ] , @xmath60^t$ ] @xmath61 [ alg:@xmath0-step cg - method : alpha ] @xmath62 @xmath63 [ alg:@xmath0-step cg - method : x ] @xmath64 [ alg:@xmath0-step cg - method : r ] @xmath65 [ alg:@xmath0-step cg - method : beta ] @xmath66 [ alg:@xmath0-step cg - method : p ] [ alg : cacg : recover ]
in this section , we discussed related work in the areas of variable @xmath0-step krylov subspace methods , the analysis of maximum attainable accuracy in finite precision cg , and inexact krylov subspace methods .
williams et al .
@xcite use a variable @xmath0-step bicgstab as the coarse grid solver in a multigrid method .
the technique used here , termed `` telescoping '' , is motivated by the observation than some coarse grid solves converge after only a few iterations , whereas other solves take significantly longer . by starting with a small @xmath0 value and gradually increasing it ,
they ensure that @xmath0-step bicgstab with larger @xmath0 is only used when the solve takes enough iterations to amortize the additional costs associated with @xmath67-step cg .
imberti and erhel @xcite have recently developed an @xmath0-step gmres method that allows variable @xmath0 sizes .
they recommend choosing @xmath3 values according to the fibonacci sequence up to some maximum value @xmath68 , i.e. , starting with @xmath69 and for @xmath70 , @xmath71 . in this way ,
the sequence @xmath72 used by imberti and erhel is predetermined .
in contrast , in our approach , the sequence @xmath72 is dynamically chosen based on the 2-norm of the updated residual ( which is not necessarily monotonically decreasing in cg ) .
in addition , our method is designed such that a user - specified accuracy @xmath73 can be attained when @xmath74 , where @xmath75 is the accuracy attained by classical cg for the same problem . in both @xmath0-step and classical variants of krylov subspace methods , finite precision roundoff error in updates to the approximate solution @xmath76 and the residual @xmath17 in each iteration can cause the updated residual @xmath17 and the true residual @xmath16 to grow further and further apart as the iterations proceed .
if this deviation grows large , it can limit the _ maximum attainable accuracy _ , i.e. , the accuracy with which we can solve @xmath5 on a computer with unit round - off @xmath77 . analyses of maximum attainable accuracy in cg and other classical ksms are given by greenbaum @xcite , van der vorst and ye @xcite , sleijpen , van der vorst , and fokkema @xcite , sleijpen , van der vorst , and modersitzki @xcite , bjrck , elfving , and strako @xcite , and gutknecht and strako @xcite .
one important result of these analyses is the insight that loss of accuracy can be caused at a very early stage of the computation , which can not be corrected in later iterations .
analyses of the maximum attainable accuracy in @xmath0-step cg and the @xmath0-step biconjugate gradient method ( bicg ) can be found in @xcite .
the term `` inexact krylov subspace methods '' refers to krylov subspace methods in which the matrix - vector products @xmath78 are computed inexactly as @xmath79 where @xmath80 represents some error matrix ( due to either finite precision computation or intentional approximation ) .
it is assumed that all other computations ( orthogonalization and vector updates ) are performed exactly . using an analysis of the deviation of the true and updated residuals
, it is shown that under these assumptions the size of the perturbation term @xmath80 can be allowed to grow inversely proportional to the norm of the updated residual without adversely affecting the attainable accuracy .
this theory has potential application in mixed - precision krylov subspace methods , as well as in situations where the operator @xmath7 is expensive to apply and can be approximated in order to gain performance .
much work has been dedicated to inexact krylov subspace methods . in @xcite , simoncini and szyld
provide a general theory for inexact variants applicable to both symmetric and nonsymmetric linear systems and eigenvalue problems and use this to derive criteria for bounding the size of @xmath80 in such a way that does not diminish the attainable accuracy of the method .
a similar analysis was given by van de eschof and sleijpen @xcite .
these analyses confirm and improve upon the earlier empirical results for the inexact gmres method of bouras and fraysse @xcite and the inexact conjugate gradient method of bouras , frayss , and giraud @xcite .
our work is very closely related to the theory of inexact krylov subspace methods in that our analysis results in a somewhat analogous `` relaxation strategy '' , where instead of relaxing the accuracy of the matrix - vector products we are instead relaxing a constraint on the condition numbers of the computed @xmath0-step basis matrices .
in addition , our analysis assumes that all parts of the computation are performed in a fixed precision @xmath77 .
assemble @xmath81 such that holds @xmath82^{t}$ ] , @xmath83^{t}$ ] , @xmath84^t$ ] @xmath85 @xmath86 @xmath63 @xmath87 @xmath88 @xmath89 as discussed , the conditioning of the @xmath0-step basis matrices in @xmath0-step krylov subspace methods affects the rate of convergence and attainable accuracy . in the @xmath0-step cg method ( algorithm [ alg : scg ] ) , the krylov subspace basis is computed in blocks of size @xmath0 in each outer loop .
this method can be generalized to allow the blocks to be of varying size ; in other words , a different @xmath0 value can be used in each outer loop iteration .
we denote the block size in outer iteration @xmath2 by @xmath3 , where @xmath90 for some maximum @xmath0 value @xmath91 .
we call this a _ variable @xmath0-step krylov subspace method_. a general variable @xmath0-step cg method is shown in algorithm [ alg : vscg ] . in this section ,
we derive a variable @xmath0-step cg method which automatically determines @xmath3 in each outer loop such that a user - specified accuracy can be attained , which we call the adaptive @xmath0-step cg method .
our analysis is based on the maximum attainable accuracy analysis for @xmath0-step cg in @xcite , which shows that the attainable accuracy of @xmath0-step cg depends on the condition numbers of the @xmath0-step basis matrices @xmath92 computed in each outer loop . in summary , we show that if in outer loop @xmath2 , @xmath3 is selected such that the condition number of the basis matrix @xmath93 is inversely proportional to the maximum 2-norm of the residual vectors computed within outer loop @xmath2 , then the approximate solution produced by the variable @xmath0-step cg method can be as accurate as the approximate solution produced by the classical cg method .
further , our method allows for the user to specify the desired accuracy , so in the case that a less accurate solution is required , our method automatically adjusts to allow for higher @xmath3 values .
this effectively exploits the tradeoff between accuracy and performance in @xmath0-step krylov subspace methods . in this section
we derive our adaptive @xmath0-step approach , which stems from a bound on the growth in outer loop @xmath2 of the gap between the true and updated residuals in finite precision . in the remainder of the paper ,
hats denote quantities computed in finite precision , @xmath77 denotes the unit round - off , @xmath94 denotes the 2-norm , and @xmath95 denotes the 2-norm condition number @xmath96 . to simplify the analysis
, we drop all @xmath97 terms ( and higher order terms in @xmath77 ) .
writing the true residual @xmath98 , we can write the bound @xmath99 then as @xmath100 ( i.e. , as the method converges ) , we have @xmath101 .
the size of the _ residual gap _ , i.e. , @xmath102 , therefore determines the attainable accuracy of the method ( see the related references in section [ sec : maa ] ) .
we begin by considering the rounding errors made in the variable @xmath0-step cg method ( algorithm [ alg : vscg ] ) . in outer loop @xmath2 and
inner loops @xmath103 of finite precision variable @xmath0-step cg , using standard models of floating point error ( e.g. , @xcite ) we have @xmath104 where @xmath105 denotes the maximum number of nonzeros per row in @xmath7 and @xmath106 .
let @xmath107 denote the residual gap in iteration @xmath12 .
similar to the analysis in @xcite , for variable @xmath0-step cg we can write the growth of the residual gap in iteration @xmath108 in outer loop @xmath2 as @xmath109 our goal is now to manipulate a bound on this quantity in order to derive a method for determining the largest @xmath3 value we can use in outer loop @xmath2 such that the desired accuracy is attained . using standard techniques , we have the norm - wise bound @xmath110 where we have used @xmath111 . to simplify the notation , we let @xmath112 and @xmath113 .
we note that @xmath114 depends on the polynomial basis used and is not expected to be too large ; e.g. , for the monomial basis , @xmath115 .
the above bound can then be written @xmath116 note that assuming @xmath117 is full rank , we have @xmath118 to bound @xmath119 in terms of the size of the residuals , notice that by , @xmath120 so @xmath121 thus together with we can write @xmath122 this allows us to write the bound in the form @xmath123 using @xmath124 , we have @xmath125 and writing @xmath126 , we have @xmath127 then @xmath128 and letting @xmath129 , this bound can be written @xmath130 this gives a norm - wise bound on the growth of the residual gap due to finite precision errors made in outer iteration @xmath2 . letting @xmath131 , notice that we have @xmath132
suppose that at the end of the iterations we want the size of the residual gap to be on the order @xmath73 ( where @xmath133 , as this is the best we could expect ) . then assuming the number of outer iterations @xmath2 is not too high , this can be accomplished by requiring that in each outer loop , for @xmath134 , @xmath135 from , this means that we must have , for all inner iterations @xmath136 , @xmath137 the bound tells us that as the 2-norm of the residual decreases , we can tolerate a more ill - conditioned basis matrix @xmath138 without detriment to the attainable accuracy .
since @xmath139 grows with increasing @xmath3 , this suggests that @xmath3 can be allowed to increase as the method converges at a rate proportional to the inverse of the residual 2-norm .
this naturally gives a relaxation strategy that is somewhat analogous to those derived for inexact krylov subspace methods ( see section [ sec : inexact ] ) .
we discuss the implementation of this strategy in the following subsection . in this section , we propose a variable @xmath0-step cg method that makes use of the constraint in determining @xmath3 in each outer loop @xmath2 , which we call the _ adaptive _ @xmath0-step cg method .
our approach uses the largest @xmath3 value possible in each outer iteration @xmath2 such that is satisfied up to some @xmath91 set by the user ( e.g. , @xmath91 could be selected by auto - tuning to find the @xmath0 value that minimizes the time per iteration in @xmath0-step cg ) . since
a variable @xmath0-step cg method produces the same iterates as classical cg when @xmath140 for all outer loops @xmath2 , we expect that our adaptive @xmath0-step cg method can attain accuracy @xmath73 when @xmath74 , where @xmath75 is the attainable accuracy of classical cg .
when the right - hand - side of is less than one ( due to , e.g. , @xmath73 being set too small ) , the inequality can not be satisfied . in this case , we default to using @xmath140 ( i.e. , we perform an iteration of classical cg ) . in cg ,
it is the @xmath7-norm of the error rather than the @xmath141-norm of the residual that is minimized , which means that in some inner loop iteration @xmath33 , we can have @xmath142 . since @xmath143 for @xmath144 are not known at the start of the outer loop iteration , we must determine @xmath3 online , i.e. , during the iterations .
we therefore first construct a basis matrix that satisfies under the assumption that @xmath145 . within each inner loop
, we can perform an inexpensive check to ensure that is still satisfied after computing the next residual ; if not , we break from the inner loop , discard the current basis matrix , and begin a new outer loop iteration .
we elaborate on this approach in the remainder of this section .
an optimized high - performance implementation of our variable @xmath0-step approach is outside the scope of the present work , although we will briefly discuss how one might implement this strategy efficiently such that each outer loop still requires only a single global synchronization and the number of wasted matrix - vector products is kept to a minimum .
the bound indicates that when @xmath117 is constructed at the beginning of outer loop @xmath2 , we should enforce @xmath146 therefore we first construct an @xmath147-step basis matrix @xmath148 using some @xmath149 , compute the gram matrix @xmath150 , and then use the condition numbers of the leading principle submatrices of the blocks of @xmath151 to compute the condition numbers of the leading columns of the blocks of @xmath148 . to be concrete , for @xmath152 , we have ( using matlab notation ) @xmath153\ ] ] and @xmath154 where @xmath155 .
then since @xmath156 , we can easily compute the condition number of the @xmath12-step basis matrix @xmath157 , i.e. , the basis matrix needed to perform @xmath12 inner loop iterations , for @xmath158 .
we therefore let @xmath159 be the largest value possible such that @xmath160 holds . to handle the case where @xmath161 for some inner iteration @xmath33
, we can add a check within the inner loop iterations : after computing @xmath162 , we check if @xmath163 is still satisfied .
note that since @xmath164 , performing this check does not require any additional communication .
if is satisfied for all @xmath165 , we will perform all @xmath166 inner loop iterations .
if this is not satisfied for some @xmath33 , we break from the current inner loop iteration and begin the next outer loop ; i.e. , we only perform @xmath167 inner loop iterations in outer loop @xmath2 .
although performing the checks for determining when to break from the current outer loop do not incur any additional communication cost in terms of the number of messages sent between processors , there is , however , potential wasted effort in terms of computation and the number of words moved when @xmath168 and/or @xmath169 , since in this case we have computed more basis vectors than we need to perform @xmath3 inner loop updates .
one way to mitigate this is to only allow @xmath147 to grow by at most @xmath170 vectors in each outer loop , i.e. , @xmath171 , where @xmath170 is some small positive integer ( e.g. , @xmath172 or @xmath173 ) .
the approach outlined above is summarized in algorithm [ alg : vscg2 ] .
note that in algorithm [ alg : vscg2 ] , we have specified that the function @xmath174 can be input by the user .
because our bounds are not tight , in many cases using @xmath175 ( from the bound ) overestimates the error , which results in the use of smaller @xmath3 values than necessary .
note that although there is no expectation that @xmath176 decreases between iterations in cg , as long as the method is converging , we still expect the residual norm to trend downward
. therefore one could make the simplifying assumption that @xmath177 for @xmath178 . in this case , the value @xmath166 can be determined at the beginning of outer loop @xmath2 before performing any inner iterations , and so the checks are unnecessary .
this simplification comes at the potential cost of reduced reliability of the adaptive @xmath0-step method , although in our experimenting we have observed that taking @xmath166 usually works well even when the residual norm is nonmonotonic .
because our strategy is based on a loose upper bound on the deviation of residuals , we do not require particularly accurate estimates of @xmath179 .
therefore alternative methods for inexpensively approximating this quantity could also be used in practice .
assemble @xmath180 such that holds @xmath181^{t}$ ] , @xmath182^{t}$ ] , @xmath183^t$ ] @xmath167 @xmath184 @xmath86 @xmath63 @xmath185 @xmath186 @xmath89
in this section , we test the adaptive @xmath0-step cg method ( algorithm [ alg : vscg2 ] ) on a number of small examples from the university of florida sparse matrix collection @xcite using matlab .
the test examples were chosen to exemplify various convergence trajectories of cg , which allows us to demonstrate the validity of our relaxation condition in which @xmath3 depends on the rate of convergence .
the test problems used here are small , so we would not use ( fixed or variable ) @xmath0-step methods here in practice ; nevertheless , they still serve to exhibit the numerical behavior of the methods .
all tests were run in double precision , i.e. , @xmath187 , and use a right - hand side @xmath188 with entries @xmath189 where @xmath190 is the dimension of @xmath7 .
we use matrix equilibration , i.e. , @xmath191 where @xmath192 is a diagonal matrix of the largest entries in each row of @xmath7 .
properties of the test matrices ( post - equilibration ) are shown in table [ mats ] .
our results are plotted in figures [ fig : gr3030]-[fig : ex5 ] . for each test matrix , we ran classical cg ( algorithm [ alg : cg ] ) , as well as both ( fixed ) @xmath0-step cg ( algorithm [ alg : scg ] ) and adaptive @xmath0-step cg ( algorithm [ alg : vscg2 ] ) for @xmath91 values @xmath18 , @xmath193 , and @xmath194 ( for fixed @xmath0-step cg , @xmath195 ) . for each matrix and @xmath91 value , we tested two different @xmath73 values : @xmath196 , the maximum attainable accuracy of classical cg , and @xmath197 .
the monomial basis was used for both variable and fixed @xmath0-step methods in all tests ( note that there is no technical restriction to the monomial basis ; the analysis used in deriving the adaptive @xmath0-step cg method still applies regardless of the polynomials used in constructing @xmath148 ) . in all tests for adaptive @xmath0-step cg , we set @xmath198 ( see algorithm [ alg : vscg2 ] ) , as this gives the optimal choice of @xmath3 in each iteration .
( we also tried using smaller values of @xmath170 , e.g. , @xmath172 and @xmath173 ; this did not significantly change the total number of outer loop iterations performed ) .
we used @xmath199 in our condition for determining the number of inner loop iterations to perform in all experiments except for tests with matrices bcsstk09 ( figure [ fig : bcsstk09 ] ) and ex5 ( figure [ fig : ex5 ] ) ; for these problems , convergence is more irregular , and thus the bound is tighter .
each figure has a corresponding table ( tables [ tab : gr3030]-[tab : ex5 ] ) which lists the number of outer loop iterations ( a proxy for the latency cost or number of global synchronizations ) required for convergence of the true residual @xmath141-norm to level @xmath73 .
for classical cg , the number reported gives the total number of iterations .
dashes in the tables indicate that the true residual diverged or stagnated before reaching level @xmath73 . .test matrix properties . [ cols="<,<,<,<,<",options="header " , ] [ tab : ex5 ] in all experiments , the adaptive @xmath0-step cg method was able to converge to the desired tolerance @xmath73 , whereas for @xmath196 and higher @xmath0 values ( @xmath200 and/or @xmath201 ) , fixed @xmath0-step cg may not achieve convergence to this level .
we point out that even when the matrix is very well - conditioned , choosing @xmath0 too large can very negatively affect the convergence rate in fixed @xmath0-step cg ; see , e.g. , the case with @xmath201 and @xmath197 in figure [ fig : mesh3e1 ] , where @xmath0-step cg requires more than 8 times the number of synchronizations than classical cg . in tests where both fixed @xmath0-step cg and adaptive @xmath0-step cg converge , adaptive @xmath0-step cg takes at most @xmath141 additional outer iterations versus fixed @xmath0-step cg ( although in most cases it takes as many or fewer outer iterations ) .
we note that in all cases , variable @xmath0-step cg uses fewer outer loop iterations ( synchronization points ) than classical cg , so we still expect a performance advantage on latency - bound problems .
the experiments using @xmath202 demonstrate the ability of the adaptive @xmath0-step cg method to automatically adjust to use the largest @xmath3 values possible ( according to our bound ) that will eventually attain the desired accuracy .
for example , in figure [ fig : gr3030 ] for gr_30_30 , for all tests using @xmath202 , adaptive @xmath0-step cg uses @xmath203 in every outer iteration , defaulting to the fixed @xmath0-step method .
in contrast , in plots on the left using @xmath196 , adaptive @xmath0-step cg uses smaller @xmath3 values at the beginning and then increases up to @xmath203 and is able to attain accuracy @xmath75 whereas fixed @xmath0-step cg fails when for @xmath200 and @xmath201 .
below we list the sequence of @xmath3 values chosen in adaptive @xmath0-step cg for tests with @xmath196 and @xmath204 ( plots in the lower left in each figure ) : * gr_30_30 : 1 , 1 , 2 , 2 , 2 , 3 , 3 , 3 , 4 , 5 , 6 , 8 , 10 , 10 * mesh3e1 : 1 , 1 , 2 , 4 , 6 , 9 , 10 * nos6 : 6 , 1 , 2 , 3 , 4 , 5 , 4 , 4 , 1 , 4 , 4 , 5 , 6 , 7 , 8 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 * bcsstk09 : 2 , 2 , 3 , 3 , 3 , 3 , 3 , 1 , 2 , 3 , 3 , 3 , 2 , 2 , 2 , 2 , 3 , 3 , 3 , 3 , 4 , 4 , 4 , 4 , 5 , 5 , 6 , 7 , 1 , 6 , 6 , 7 , 7 , 7 , 7 , 8 , 8 , 8 , 9 , 9 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 , 10 * ex5 : 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 1 , 3 , 1 , 1 , 2 , 1 , 1 , 1 , 3 , 3 , 2 , 1 , 1 , 1 , 2 , 1 , 1 , 2 , 4 , 6 , 2 , 4 , 5 , 7 , 2 , 6 , 1 , 7 , 6 there are a few things to notice . first ,
as expected , the sequence of @xmath3 values is monotonically increasing when the residual 2-norm decreases monotonically or close to it ( matrices gr_30_30 and mesh3e1 ) , but when the residual 2-norm oscillates , we no longer have monotonicity of the @xmath3 values ( matrices nos6 , bcsstk09 , and ex5 ) .
second , it may be that @xmath91 is never reached , as is the case for matrix ex5 .
it is also worth pointing out that the sequence of @xmath3 values need not start with @xmath205 ; for nos6 , @xmath206 and for bcsstk09 , @xmath207 .
it may be that on a particular machine for a particular matrix structure , it is not beneficial in terms of minimizing time per iteration to go above a small value of @xmath91 , say @xmath208 . in this case ,
as long as the matrix is reasonably well - conditioned and the accuracy required is not too high , the fixed @xmath0-step cg method can probably be used without too much trouble .
however , even in this case the adaptive @xmath0-step cg method can provide reliability while still maintaining the reduction in synchronizations of the fixed @xmath0-step method , at the cost of potentially wasting a few matrix - vector multiplications in each outer iteration .
looking at the number of outer loop iterations in tables [ tab : gr3030]-[tab : ex5 ] , we see that in our tests for adaptive @xmath0-step cg it is often the case that increasing the @xmath91 value past a certain point gives diminishing returns in terms of reducing synchronization .
this may represent a fundamental limit of the possible performance using the monomial basis for these problems . in these cases
it may be that better performance ( higher @xmath3 values ) can be achieved only by using a better - conditioned polynomial basis for constructing @xmath148 ( e.g. , newton or chebyshev bases ) .
in this work , we developed the adaptive @xmath0-step cg method , which is a variable @xmath0-step cg method where the parameter @xmath3 is determined automatically in each outer loop such that a user - specified accuracy is attainable .
the method for determining @xmath3 is based on a bound on the residual gap within outer loop @xmath2 , from which we derive a constraint on the condition number of the @xmath0-step basis matrix generated in outer loop @xmath2 .
the computations required for determining @xmath3 can be performed without introducing any additional synchronizations in each outer loop .
our numerical experiments demonstrate that the adaptive @xmath0-step cg method is able to attain up to the same accuracy as classical cg while still significantly reducing the number of outer loop iterations ( i.e. , synchronizations ) performed .
in contrast , the fixed @xmath0-step cg method can fail to converge to this accuracy when @xmath209 .
additionally , the adaptive @xmath0-step cg method adjusts to use larger @xmath3 when an application requires a less accurate solution , automatically optimizing the tradeoff between speed and accuracy . in this way
, a user can use the matrix structure and machine parameters to estimate the value @xmath91 that will minimize the time per iteration ( e.g. , by guessing , or offline auto - tuning ) , input the accuracy they require , and the adaptive @xmath0-step cg method will choose @xmath3 as large as it can ( i.e. , reducing as much synchronization as it can ) while still converging to the desired level .
this takes the burden of parameter selection off the user .
our adaptive @xmath0-step cg method thus makes significantly advances in improving reliability and usability in @xmath0-step krylov subspace methods .
our future work includes a high - performance implementation and performance experiments in order to compare classical cg , fixed @xmath0-step cg , and adaptive @xmath0-step cg on large - scale problems .
another potential improvement is to obtain a tighter bound to replace , which could avoid the requirement of adjusting the function @xmath174 ( which represents a multiplicative factor in the bounds and ) in algorithm [ alg : vscg2 ] .
the adaptive @xmath0-step cg method is designed to maximize @xmath3 subject to a constraint on accuracy , but obviously it would also be useful to optimize @xmath3 to achieve the _ overall _ lowest latency / synchronization cost .
this would require a better understanding of the finite precision convergence behavior of both @xmath0-step and classical krylov subspace methods , which is an active area of research .
we believe that this adaptive approach based on a bound on the residual gap can be extended to other @xmath0-step krylov subspace methods as well .
the same approach outlined here can be used in @xmath0-step bicg ( see , e.g. , @xcite ) , which uses the same recurrences for the solution and updated residual .
we conjecture that a similar approach applies to @xmath0-step variants of other recursively computed residual methods ( see @xcite ) , like @xmath0-step bicgstab @xcite , as well as methods like @xmath0-step gmres . in the case of @xmath0-step gmres
, it may be possible to simplify the adaptive algorithm : since the residual norm monotonically decreases , the sequence of @xmath3 values should monotonically increase .
such a sequence of @xmath3 values was used in the variable @xmath0-step gmres method of imberti and erhel @xcite . as a final comment
, we note that in our analysis in section [ sec : varscg ] , we used equation to determine a bound on @xmath210 that guarantees accuracy on the level @xmath73 .
we could have just as easily used equation to develop a mixed precision @xmath0-step cg method where @xmath0 is fixed , but the precision used in computing in outer loop @xmath2 varies . in other words , rearranging the inequality we see that , given that we ve computed an @xmath0-step basis with condition number @xmath139 , if we want to achieve accuracy @xmath73 , we must have that @xmath211 investigating the potential for mixed precision @xmath0-step krylov subspace methods based on this analysis remains future work .
, _ a relaxation strategy for inner - outer linear solvers in domain decomposition methods _ , tech .
report cerfacs tr / pa/00/17 , european centre for research and advanced training in scientific computation , 2000 . , _ a block lanczos algorithm for computing the q algebraically largest eigenvalues and a corresponding eigenspace of large , sparse , real symmetric matrices _ , in proc . of the 1974 ieee conf . on decision and control , ieee , 1974 , pp . | on modern large - scale parallel computers , the performance of krylov subspace iterative methods is limited by global synchronization .
this has inspired the development of @xmath0-step ( or communication - avoiding ) krylov subspace method variants , in which iterations are computed in blocks of @xmath0 .
this reformulation can reduce the number of global synchronizations per iteration by a factor of @xmath1 , and has been shown to produce speedups in practical settings .
although the @xmath0-step variants are mathematically equivalent to their classical counterparts , they can behave quite differently in finite precision depending on the parameter @xmath0 .
if @xmath0 is chosen too large , the @xmath0-step method can suffer a convergence delay and a decrease in attainable accuracy relative to the classical method .
this makes it difficult for a potential user of such methods - the @xmath0 value that minimizes the time per iteration may not be the best @xmath0 for minimizing the overall time - to - solution , and further may cause an unacceptable decrease in accuracy . towards improving the reliability and usability of @xmath0-step krylov subspace methods , in this work
we derive the _ adaptive @xmath0-step cg method _
, a variable @xmath0-step cg method where in block @xmath2 , the parameter @xmath3 is determined automatically such that a user - specified accuracy is attainable . the method for determining @xmath3
is based on a bound on growth of the residual gap within block @xmath2 , from which we derive a constraint on the condition numbers of the computed @xmath4-dimensional krylov subspace bases .
the computations required for determining the block size @xmath3 can be performed without increasing the number of global synchronizations per block .
our numerical experiments demonstrate that the adaptive @xmath0-step cg method is able to attain up to the same accuracy as classical cg while still significantly reducing the total number of global synchronizations . |
information systems are threatened today as never before .
evolving network technologies have provided powerful information - transfer applications that have increased our reliance on computerized information systems at the same time that public access to the internet has increased the number and sophistication of those who seek to compromise these systems .
the diversity of existing software are hardware devices make the problem of the network the protection very more difficult .
there are many approaches to study the attack tolerance of networks .
these approaches usually include the study of dependance of the network stability on the network complexity and topology ( see , for example @xcite and references therein ) , and signature - based analysis technique and statistical analysis and modelling of network traffic ( see , for example @xcite ) . some methods to study spatial traffic flows traversing the network @xcite and correlation functions of irregular sequences of numbers occurring in the operation of computer networks @xcite have been proposed recently .
in this paper we will not discuss properties related to network structure and topology , but rather concentrate on the information flow on the network .
the main difference between this paper and the existing methods is that we are interested in general properties of information flow on a network .
we suggest a new approach for network monitoring and intrusion detection based on complete network monitoring .
the term `` complete '' will be clarified later .
since , the process of information exchange is extremely complicated for detailed analysis , we apply some methods for analysis of complex systems in physics to the information flow on networks .
this gives us the powerful tools for the analysis and provides a guideline for the application of the obtained result for practical purposes .
we will start from a careful analysis of information exchange on networks to choose the appropriate method of the information flow description .
it will give us the basis for network simulations in terms of mathematical description of information exchange processes .
also , the understanding of information flow on the network is necessary for real time network monitoring and for solutions of difficult problems of intrusion detection@xcite in real time .
to describe the information flow on a network , we work on the level of packets exchanging between computers .
the structure of the packets and their sizes vary from each to another and depend on the process . in general each packet
consists from a header and attached ( encapsulated ) data . since the data part does not affect the packet propagation through the network we consider in this paper only information included in headers .
we recall that the header consists of encapsulated protocols related to different layers of communications : from a link layer to an application layer .
the information contained in the headers controls all network traffic . to extract this information one can use tcpdump utilities , developed with the standard of lbnl s network research group @xcite .
this information is used to analyze network traffic , to find a signature of abnormal network behavior and to detect possible intrusions .
the important difference of the proposed approach from traditionally used methods is the presentation of information contained in headers in terms of well - defined mathematical functions . to do that we have developed software to read binary tcpdump files and to represent all protocol parameters as corresponding time - dependent functions
this gives us the opportunity to analyze complete information ( or a chosen fraction of complete information : a combination of some parameters ) for a chosen time and time window .
the ability to vary the time window for the analysis is important since it gives the possibility to extract different scales in the time dependance of the system .
different scales have different sensitivities for particular modes of the system development and , for an example , could be sensitive to different methods of intrusions .
it should be noted that not all parameters are reliable for the description of the information flow .
the protocol parameters for host - to - host communication could be divided into two separate groups@xcite in respect to the preservation or changing of their values during the packet propagation through the network .
we call these two groups of parameters as dynamic and static .
the dynamic parameters may be changed during the packet propagation through the network ( internet ) .
for example , a `` physical '' address of a computer , which is the mac parameter of the ethernet protocol , is a dynamic parameter because it can be changed if the packet has been re - directed by a router . on the other hand ,
the source ip address is an example of static parameter because it does not change its value during the packet propagation . to describe the information flow
, we use only static parameters since they may carry intrinsic properties of the information flow neglecting the network ( internet ) structure .
now , using packets as a fundamental objects for information exchange on network and being able to describe them in terms of functions of time for static parameters , we can apply methods , developed in physics and applied mathematics to study complex dynamical systems , for the network traffic analysis . to demonstrate the power of these methods we recall an important result for future applications which has been obtained in the framework of the presented approach .
it was shown@xcite that to describe the information flow on network one can use a small number ( 10 - 12 ) of parameters . in other words , the dimension of the information flow space is less or equal to 12 and the information flow properties are independent on network structure , size and topology . to estimate the dimension of the information flow on the network one can apply the algorithm for analysis observed chaotic data in physical systems suggested in paper@xcite(see
also ref.@xciteand references therein ) .
the main idea is related to the fact that any dynamical system with dimension @xmath0 can be described by the set of @xmath0 differential equations of the second order in configuration space or by the set of @xmath1 differential equations of first order in phase space .
assuming that the information flow can be described in terms of ordinary differential equations ( or by discrete - time evolution rules ) for some unknown functions in a ( parametric ) phase space , one can analyze a time dependance of given scalar parameter @xmath2 which is related to the system dynamics ( see , for details ref.@xcite )
. then one can build @xmath3-dimensional vectors from the variable @xmath4 as @xmath5 \label{yvec}\ ] ] at equal - distant time intervals @xmath6 : @xmath7 , where @xmath8 is integer number to numerate @xmath4 values at different time .
now one can calculate a number of nearest neighbors in the vicinity of each point in the vector space and plot the dependance of the number of false nearest neighbors(fnn ) as a function of time .
the false nearest neighbors for the @xmath3-dimensional space are neighbors which move far away when we increase dimension from @xmath3 to @xmath9
. this algorithm could be illustrated by the simple example for a two - dimensional circle .
if we project the circle on one - dimensional space , we get interval with two degenerated points along the projection axis . increasing the dimension by 1 we come to the original two - dimensional circle without the degeneracy .
thus , the degenerated points in 1-dimension which have moved to the opposite sides of the circle in 2-dimensions could be called a false nearest neighbors ( fnn ) . unfolding the space further , to 3-dimension , and further we will not get a false nearest neighbors anymore since a higher dimensional space covers the two - dimensional space to which the circle belongs .
the typical behavior a scalar parameter and corresponding fnn plot are shown on figs .
( [ fig : ip ] ) and ( [ fig : fnn ] ) .
units ) . ] from the last plot one can see that the number of false nearest neighbors rapidly decreases up to about 10 or 12 dimensions . after that it shows a slow , if any dependance at all , on the dimension .
this reflects the fact that by increasing the dimension @xmath3 step - by - step , the number of false nearest neighbors ( fnn ) , which come due to projection of far away parts of the trajectory in higher dimensional space , is decreasing up to the level restricted by the system noise since the noise has infinite dimension .
the analysis of information flow on network has been done in paper@xcite for a wide set of different network configurations and it was shown that it s is about 10 or 12 .
it means one needs not more than 12 independent parameters for a complete description of the information flow and that its dynamics can be described as a trajectory in a phase space with the dimension about 10 - 12 .
it is also important that this dimension does not depend on the network topology , its size , and the operating systems involved in the network .
therefore , this characteristic is an universal and may be applied for any network .
however , we do not know these independent parameters .
moreover , due to the complexity of the system it is natural that these unknown parameters , which are real dynamical degrees of freedom , have very complicated relations with the parameters contained in the network protocols .
fortunately , these technique provide very powerful methods to extract general information about behavior of dynamical complex systems .
for example , the obtained time dependence of only one parameter , the protocol i d , shown on fig.([fig : ip ] ) is enough to reconstruct the trajectory of the information flow in its phase space .
the projection of the trajectory on 3-dimensional space is shown on fig .
( [ fig : atr ] ) , and further three projections of the trajectory on the 2-dimensional spaces are shown on fig.([fig : atr - pr ] ) .
one can see that the complete description of the network information traffic in terms of small number of parameters is possible and the the trajectory of information flow is rather well localized and will be a subject for further analysis and pattern recognition techniques . ) . ]
now we can obtain results for developing an approach to monitor the network in real time and to prevent possible intrusions .
the direct consequences of the proposed approach for the network traffic description provides the possibility of a real time network monitoring and detection of all known network attacks .
this is because the complete information about network traffic at any given point of the network is collected from tcpdump binary output .
all header parameters are converted into time dependant numerical functions .
therefore , each packet for host - to - host exchange corresponds to a point in the multidimensional parametric phase space .
the set of these points ( the trajectory ) completely describes information transfer on the network .
it is clear that this representation provides not only the total description of the network traffic at the given point but rather a powerful tool for analysis in real time .
let us consider some possible scenarios for the analysis .
suppose we are looking for known network intrusions .
the signature of the intrusion is a special set of relations between the header parameters . for example@xcite , the signature for the attempt to identify live hosts by those responding to the ack scan includes a source address , an ack and syn flags from tcp protocol , a target address of internal network , sequences number , and source and destination port numbers .
the lone ack flag set with identical source and destination ports is the signature for the ack scan , since the lone ack flag set should be found only as the final transmission of the three - way handshake , an acknowledgement of receiving data , or data that is transmitted where the entire sending buffer has not been emptied . from this example one can see that the intrusion signature could be easily formulated in terms of logic rules and corresponding equations .
then , collecting the header parameters ( this is the initial phase of network monitoring ) and testing sets of them against the signatures ( functions in terms of the subset of the parameters ) one can filter out all known intrusions . since we can collect any set of the parameters and easily add any signature function , it provides the way for a continuous upgrading of the intrusion detection system ( ids ) built on these principles .
in other words , such i d system is universal and can be used to detect all possible network intrusions by adding new filter functions or macros in the existing testing program .
it is very flexible and easily upgradable .
the flexibility is important , however , and in principle could be achieved even in existing `` traditional '' ids s .
what is out of scope of traditional approaches is the minimization of possible false alarms and maximal optimized and controlled sensitivity to intrusion signals .
these features are an intrinsic feature of our approach .
the important feature of the approach is the presentation of the parameters in terms of time dependant functions .
this gives the opportunity to decrease the false alarm probability of the ids as best possible for the particular network .
this can be done using sophisticated methods already developed for noise reduction in time series .
moreover , representation of the protocol parameters as numerical functions provides the opportunity for detailed mathematical analysis and , as a consequence , for the optimization of the signal - to - noise ratio using not only time series techniques but also numerical methods for analysis of multi - dimensional functions .
the combination of these methods provides the best possible way , in terms of accuracy of the algorithms and reliability of the obtained information , for detection of known intrusions in real time .
also , the description of the information flow in terms of numerical functions gives the opportunity to monitor the network traffic for different time windows without missing information and without overflowing storage facilities .
one can suggest some different ways to do it .
one example is the use of a parallel computer environment ( such as low cost powerful linux clusters ) for the simultaneous analysis of the decoded binary tcpdump output . in this case
the numerical functions of the header parameters , being sent to different nodes of the cluster , will be analyzed by each node using similar algorithms but different scales for time averaging of signals ( or functions ) .
thus each node has separate time window and , therefore , is sensitive to the network behavior in the particular range of time .
choosing time averaging scales for the nodes from microseconds to weeks , for example , one can trace and analyze the network traffic independently and simultaneously in all these time windows .
it is worthwhile to remember that the optimal signal - to - noise ratio is achieved for each time window independently providing the best possible level of information traffic analysis for the whole network .
there are at least three obvious advantages for this approach .
the first one is the possibility to detect intrusions developed on different time scales simultaneously and in real time .
the second one is the automatic decreasing of the noise from a short time fluctuations for long time windows due to time averaging .
this provides detailed information analysis in each time window without lost information and , at the same time , discards all noise related information , drastically reducing the amount of information at the storage facilities .
the third advantage is the possibility to use ( in real time ) output from short time scale analyzed data as an additional information for the long time scale analysis .
let us estimate how many parameters are used to describe signatures of currently known intrusions .
one analyzes the comprehensive ( but probably not a complete ) list of known attacks : smurf @xcite , fraggle @xcite , pingpong @xcite , ping of death @xcite , ip fragment overlap @xcite , brkill @xcite , land attack @xcite , syn flood attack @xcite , tcp session hijacking @xcite , out of band bug @xcite , ip unaligned timestamp @xcite , bonk @xcite , oob data barf @xcite and vulnerability scans ( fin and syn & fin scanning ) @xcite . the frequencies of the parameters involved in signatures for these intrusions are shown on fig.([fig : fr ] ) and the numeration of the parameters is explained in table 1 .
one can see that the small number of parameters that are used for signatures of intrusions .
this fact simplifies the procedure of the analysis .
+ .the parameters involved in intrusion signatures as shown on fig.([fig : fr ] ) . [ cols="^,^,^,^,^",options="header " , ]
the suggested approach could be considered as a powerful and promising method for network monitoring and detection of known network intrusions .
however , there is an even more important feature : the ability to detect previously unknown attacks on a network in a wide range of time scales .
this ability is based on the method of description of information exchange on networks in terms of the exchange of packets provided by the header parameters as described in terms of numerical functions of time ( or a trajectory in multi - dimensional phase space ) as well as on existing methods in theoretical physics for the analysis of dynamics of complex systems .
these methods lead to a very useful result for the dimensionality of the information flow space . since the number of parameters used in packet header is high ( of the order of hundreds )
, the practical search for unknown ( even abnormal ) signal would be a difficult problem if possible at all .
therefore , the reported result @xcite on the small dimension of the parametric space of the information flow is a crucial point for the practical approach for the detection of unknown intrusions . to build a real time intrusion detection system ,
which is capable of detecting unknown attacks , we exploit the fact that one needs to analyze only a small number of parameters . furthermore , as it is known from complex systems
theory , the choice of the parameters is not important unless they are sensitive to the system behavior .
the last statement needs to be explained in more detail .
generally , hundreds different parameters could be encapsulated in the packets headers .
the question is which parameters we need to choose for the right description of the information flow .
one can think , following the discussion in the previous section , that we need to make our choice from the known quoted 17 parameters . it may be a good guess . however , the number 17 is bigger than the dimension of the phase space which we are taking in mind , and it could be a suspicion that hackers will invent new attacks with new signature parameters which are not included in the set presented in the previous section
. the right answer on these remarks follows from complex systems theory : for complete system description one needs only the number of parameters equal to the phase space dimension ( more precisely , the smallest integer number which is larger than fractal dimension of the phase space ) .
it could be set of any parameters which are sensitive to the system dynamics ( and the 17 discussed parameters could be good candidates ) .
we do not know , and do not suppose to know , the real set of parameters until the theory of network information flow is developed or a reliable model for information flow description is suggested .
nevertheless , a method developed to study non - linear complex systems provides tools to extract the essential information about the system from the analysis of even a small partial set of the `` sensitive '' parameters . as an example , one can refer to the fig.([fig : atr ] ) which show the 3-dimensional projection of the reconstructed trajectory from the time dependent behavior of only one parameter ( the protocol i d ) shown on fig.([fig : ip ] ) .
it means that the complete description of the network information flow could be obtained even from one `` sensitive '' parameter .
it should be noted that it is certainly not enough to analyze one parameter for intrusion detection in realistic situation . in the worst case
, some analysis can be done using one parameter , but the reliability , signal - to - noise ratio and time efficiency of the analysis will be rather poor for it to be used as a search for an unknown attack .
however , the above extreme example demonstrates the power of the method . the proper approach is to use as many parameters as possible ( up to the dimension of the phase space ) . in that case
one can reach the optimal sensitivity to an intrusion signal and speed up the analysis procedure .
one of the ways to the realize this approach is to use the multi - window method discussed in the previous section with the proper data analysis for each time scale .
the method of analysis is out of the scope of the current paper and will be reported elsewhere .
we will review only the general idea and the problems related to this analysis . to detect unknown attacks ( unusual network behavior )
we use a deviation of signals from the normal regular network behavior . for these purposes
one can use a pattern recognition technique : as well to establish a normal behavior patterns so as to measure a possible deviation from the normal pattern .
however , the pattern recognition problem is quite difficult for this analysis . according to our knowledge ,
it is technically impossible to achieve reliable efficiency in pattern recognition for rather large , such as a 10 dimensional space . on other hand , for our purposes
, we can not extract reliable information from the analysis of one parameter .
therefore , one can choose the optimal ( compromised ) solution to use pattern recognition technique in the informational flow subspaces with low dimensions .
one can choose these subspaces as cross sections of the total phase space defined by applying appropriate constraints on some header parameters .
in this case , we will have reasonable ratio of signal - to - noise and will simplify and also improve reliability for the pattern recognition technique . for pattern recognition
we suggest useing a 2 - 3 dimensional wavelet analysis chosen on basis of detailed study of the information traffic on the set of networks .
the wavelet approach is promising because it simultaneously drastically reduces the computational time and memory requirements ( which is very important for multidimensional analysis ) and because it can be used for additional effective noise reduction technique .
we suggest a new approach for real time network monitoring which is based on the application of complex systems theory for information flow analysis on networks .
the synthesis of the network traffic description , in terms of numerical time dependant functions , and methods of theoretical physics , for the study of complex systems , provides not only a robust method for network monitoring with detection of known intrusions , but it looks very promising for developing real systems for detection of unknown intrusions . to apply innovative technology approaches that are measurably effective against practical attacks it is necessary to detect and identify the attack on reconnaissance stage . as an element for the future technology we presented the study of the possibility of building an automatic intrusion detector system , based on new methods of data analysis and pattern recognition .
the system will be able to help maintain a high level of confidence in the protection of the networks .
we thank staff of advanced solutions group for technical support .
this work was supported by the darpa information assurance and survivability program and is administered by the usaf air force research laboratory via grant f30602 - 99 - 2 - 0513 , as modified .
j. b. d. cabrera , b. ravichandram and r. k. mehra , `` statistical traffic modeling for network intrusion detection '' , in : proceedings of the international simposium on modeling , ananlysis and simulation of computer and telecommunication systems , ieee ( 2000 ) . | the approach for a network behavior description in terms of numerical time - dependant functions of the protocol parameters is suggested .
this provides a basis for application of methods of mathematical and theoretical physics for information flow analysis on network and for extraction of patterns of typical network behavior .
the information traffic can be described as a trajectory in multi - dimensional parameter - time space with dimension about 10 - 12 .
based on this study some algorithms for the proposed intrusion detection system are discussed .
network monitoring , intrusion detection , information flow |
recent years have brought an increasing realization of the astrophysical importance of black hole spin .
ever since the seminal work of penrose ( 1969 ) and blandford & znajek ( 1977 ) , it has been recognized that black hole spin may be an important source of energy , especially for the powerful relativistic jets seen emerging from many black hole systems .
however , the importance of black hole spin goes beyond its role as a possible power source .
the spins of stellar - mass black holes found in accreting galactic black hole binaries ( gbhbs ) are expected to be natal ( king & kolb 1999 ) and give us a window into the black hole forming core - collapse supernova .
the spin distribution of the supermassive black hole ( smbh ) population ( and its dependence on smbh mass ) encodes the black hole growth history , e.g. , the role of accretion versus mergers ( moderski & sikora 1996 ; volonteri et al . 2005 ) .
thus , across the mass scales , black hole spin gives us a fossil record of how black holes form and grow .
of course , we must have robust observational probes of black hole spin if we are to truly test spin - powered jet models or actually use spin to probe the formation and growth of black holes .
the most direct method imaginable is the characterization of the gravitational radiation emitted as another compact object spirals into the black hole .
however , in the current ( pre- gravitational wave observatory ) era , we must search for signs of black hole spin in the electromagnetic radiation emitted by the surrounding accretion disk .
the natural waveband to use is the x - ray band given the fact that the observed x - rays are thought to be predominately emitted from the inner regions of the accretion flow where the relativistic effects related to black hole spin are strongest .
the most important effect of black hole spin on the accretion disk arises from the spin dependence of the location of the innermost stable circular orbit ( isco ) ; in boyer - lindquist coordinates , @xmath0 shrinks from @xmath1 for a non - rotating black hole down to @xmath2 for a hole that is maximally rotating in the same sense as the accretion disk .
thus , for a given black hole mass , the characteristic temporal frequency , energy release ( and hence temperature ) , and gravitational redshift of the inner accretion flow all increase as one considers progressively higher black hole spin .
these considerations lead to the three most widely discussed techniques for using x - ray data to determine black hole spin . on the timing front ,
the frequency stability of the high - frequency quasi - periodic oscillations ( hfqpos ) seen in the soft intermediate state of gbhbs strongly suggest a connection to the gravitational potential and black hole spin ( strohmayer 2001 ; also see mcclintock & remillard [ 2003 ] for a general review of hfqpo phenomenology ) .
while hfqpos may eventually provide a robust way of measuring spins in gbhbs , the lack of a compelling theoretical framework with which to interpret the frequencies prevents any robust conclusions from being drawn at the present time .
for example , different and contradictory constraints on the mass and spin of the black hole result from assuming that the hfqpos are manifestations of trapped global g - modes ( nowak & wagoner 1992 ; nowak et al .
1997 ) , parametric resonances ( abramowicz et al . 2003 ) , or lens - thirring precession ( merloni et al . 1999 ) .
the second technique involves modeling the thermal continuum spectrum from the disk .
provided that one selects systems that have optically - thick , geometrically - thin inner accretion disks , observations can be compared with model thermal continuum spectra ( which include vertical radiation transfer in the disk atmosphere as well as doppler and gravitational red / blue - shifts ; davis , done & blaes 2006 ) .
spin results from this technique have been reported by mcclintock et al .
( 2006 ) and shafee et al .
( 2006 ) , although the contrary results of middleton et al .
( 2006 ) demonstrate the current fragility of this technique to modeling the non - thermal emission , particularly when applying it to data from the proportional counter array ( pca ) on the _ rossi x - ray timing explorer ( rxte ) _ in which one only sees the wien tail of the thermal disk spectrum . while this is a powerful method for determining the spin of accreting stellar - mass black holes ( especially when applied to broad - band x - ray data that extends into the soft x - ray band ) , the fact that one needs both an estimate of the black hole mass and a high quality measurement of the thermal continuum shape makes it difficult to apply to the smbh in active galactic nuclei ( agn ) .
the thermal continuum of agn disks is in the uv / euv region of the spectrum , making a determination of its shape and normalization extremely susceptible to any errors in the ( large ) correction that must be done for dust extinction and photoelectric absorption .
the third technique for constraining spin uses relativistic broadening and gravitational redshifting of the fluorescent iron emission line seen in many gbhbs and agn ( fabian et al , 1989 ; laor 1991 ; fabian et al .
2000 ; reynolds & nowak 2003 ; fabian & miniutti 2005 ) . as one considers progressively more rapidly rotating black holes , the primary x - ray emission and hence the iron line emission will be dominated by regions with higher gravitational redshift , leading to broader and more highly skewed iron emission lines .
a major advantage of this technique is that the iron line profiles are completely independent of black hole mass and hence one can apply this to an agn in which the mass is often extremely uncertain .
although one actually proceeds via formal fitting of relativistically smeared theoretical disk reflection spectra to data , the black hole spin constraint essentially results from a characterization of the maximum redshift of the iron line .
results using this technique on the seyfert galaxy mcg6 - 30 - 15 have been reported by brenneman & reynolds ( 2006 ) who report rather extreme constraints on the dimensionless black hole spin of @xmath3 ( also see dabrowski et al . 1997 ) . both the thermal continuum and the iron line techniques for determining black hole spin rely on the assumption that the isco defines an inner edge for the thermally or line emitting region of the disk ( see krolik & hawley 2002 for a general discussion of the various `` inner edges '' that one can define for an accretion disk ) .
the robustness of the thermal continuum technique has been explicitly addressed by shafee , narayan & mcclintock ( 2007 ) using a global @xmath4-viscosity model that explicitly tracks the flow properties within the isco .
although mhd effects close to the isco may invalidate the assumptions of such @xmath4-models ( e.g. , reynolds & armitage 2001 ) , the results of shafee et al .
( 2007 ) do indeed suggest that modeling of the thermal continuum can produce an accurate measure of black hole spin .
the subject of the current _ paper _ is to address the robustness of the iron line fitting technique . in section 2
, we present a high resolution 3-dimensional magnetohydrodynamic ( mhd ) simulation of a geometrically - thin ( @xmath5 ) accretion disk that we use to explore the nature of the flow close to the isco . in section 3 , we show that both the midplane density and vertically - integrated surface density of the accretion flow drops precipitously over a small range of radii close to the isco .
we then argue in section 4 that photoionization of the flow within the isco will effectively suppress the iron line emission ( and all other x - ray reflection signatures ) across most of the region within the isco .
we discuss the implications and limitations of our results in section 5 , including an estimate of the uncertainties in the inferred black hole spin that might arise from iron line emission in the region immediately within the isco .
section 6 presents our conclusions .
relativistically broad iron lines are only seen from accretion flows that are relatively cold ( i.e. , there are recombined iron ions in the plasma ) and compton - thick .
thus , we expect systems displaying broad iron lines to have radiatively - efficient , geometrically - thin accretion disks of the type envisaged by pringle & rees ( 1972 ) , shakura & sunyaev ( 1973 ) and nokivov & thorne ( 1974 ) .
our goal is to simulate such an accretion flow in order to assess the nature of the flow around the isco .
of course , such black hole accretion disks are very complex systems .
it is believed that the underlying angular momentum transport that actually allows accretion to proceed is due to sustained mhd turbulence driven by the magneto - rotational instability ( mri ; balbus & hawley 1991 , 1998 ) .
magnetic energy is dissipated and thermalized through magnetic reconnection on small scales within the body of the disk . in the systems under consideration here ,
the accretion disk is optically - thick and the resulting thermal radiation comes to dominate the gas pressure and , indeed , provides the principal vertical pressure support for the disk .
the radiation diffuses vertically out of the disk and is radiated from the disk atmosphere
. finally , this physics is occurring in a region where both special and general relativistic effects are strong and can have qualitative effects .
the most obvious general relativistic effect is the existence of the isco .
however , frame - dragging , light bending , and the strong displacement currents that accompany the relativistic fluid velocities can all have important effects on disk dynamics ( e.g. , see fully relativistic mhd simulations of de villiers et al .
2004 , mckinney & gammie 2004 ) .
global accretion disk simulations that include all of these physical processes are still beyond current computational capabilities . moreover , since we are motivated to model geometrically - thin accretion disks , the need to spatially resolve the vertical structure of the disk while still modeling a significant range of radii imposes severe resolution requirements as will be discussed in more detail below .
clearly , compromises must be made .
these considerations lead us to construct a model of a thin accretion disk employing the simplest possible physics that will enable us to study the structure of the flow around the isco .
we employ 3-dimensional mhd in order to correctly capture angular momentum transport within the flow due to mri - driven turbulence . given that fully relativistic simulation at the required resolution
would prove extremely computational intensive , together with the current lack of an adequate publicly available 3-d grmhd code , we use the pseudo - newtonian potential of pacynski & witta ( 1980 ) , @xmath6 where @xmath7 is the gravitational radius of the black hole which has mass @xmath8 .
this potential has an isco at @xmath9 and , in fact , is a good approximation for the gravitational field of a non - rotating black hole .
although we essentially integrate the equations of non - relativistic mhd within this potential , we use the prescription of miller & stone ( 2000 ) to include some effects of the displacement current , principally forcing the alfvn speed to correctly limit to the speed of light as the magnetic fields becomes strong .
we note that this `` alfvn speed limiter '' only plays a role within the tenuous magnetized atmosphere that forms at large vertical distances above and below the disk ; it never plays a significant role in the body of the accretion disk .
furthermore , gas velocities within our simulation are always very subluminal ( achieving a maximum of @xmath10 at the inner boundary of the simulation domain ) .
thus , we believe that the neglect of the full set of relativistic terms in the mhd equations should not significantly impact our simulation . we also note that de villiers & hawley ( 2003 ) explicitly comment upon the similarity of pseudo - newtonian simulations with full grmhd simulations performed within a schwarzschild metric .
in another major simplification , we neglect radiation processes ( radiative heating , radiative cooling and radiation pressure ) . in place of a full energy equation , the gas is given an adiabatic equation of state with @xmath11 .
our simulation does not , however , capture the magnetic energy lost to numerical ( i.e. grid - scale ) reconnection . while not particularly physical , this does acts as an effective cooling term for the disk and helps maintain a thin , cold structure .
the simulation is performed in cylindrical polar coordinates @xmath12 .
the initial condition for the simulation consists of a disk with a constant mid - plane density @xmath13 for @xmath14 .
the vertical run of initial density and pressure correspond to an isothermal atmosphere in vertical hydrostatic equilibrium , @xmath15 where @xmath16 is the cylindrical radius , @xmath17 is the vertical height above the disk midplane and @xmath18 .
the scale height of the disk as a function of radius is a free function ; we choose to set @xmath19 for all @xmath16 , i.e. , a constant height disk with initial @xmath20 at the isco . the initial density is set to zero for @xmath21 .
the initial velocity field is everywhere set to @xmath22 corresponding to rotation on cylinders and pure keplerian motion for material on the mid - plane .
an initially weak magnetic field is introduced in the form of poloidal field loops specified in terms of their vector potential @xmath23 in order to ensure that the initial field is divergence free .
we choose the explicit form for the vector potential , @xmath24 where @xmath25 is a normalization constant and @xmath26 is an envelope function that is unity in the body of the disk and then smoothly goes to zero at both @xmath27 and at a location three pressure scale heights away from the midplane of the disk .
the use of @xmath26 keeps the initial field configuration well away from either the inner radial boundary of the initial disk configuration or the vertical boundaries of the calculation domain .
the final multiplicative term produces field reversals with a radial wavelength of @xmath28 .
this results in a number of distinct poloidal loops throughout the disk and acts as a much more effective seed for the mri than would a single loop which followed contours of pressure ( in the sense that a steady - state is reached more quickly ) .
the normalization constant @xmath25 is set to give an initial ratio of gas - to - magnetic pressure of @xmath29 in the inner disk . from the body of literature on mri driven turbulence ( e.g. , see balbus 2003 )
, we expect the final turbulent state to be independent of most aspects of the initial field configuration ( e.g. , see the study of hawley 2000 ) .
in general terms , the one aspect of the initial field that can continue to leave imprints on the turbulent state is the net flux threading the simulation domain ( e.g. , see reynolds & armitage 2001 ) .
our initial condition has zero net flux , a choice that is particularly well defined and robust to implement but is expected to minimize the saturation field strength and the turbulent angular momentum transport .
it is beyond the scope of this paper to explore disks that have a net poloidal field resulting , for example , from advection of magnetic flux from the mass reserviour .
apart from the miller & stone ( 2000 ) alfvn speed limiter , grid effects and the usual artificial viscosity , this simulation deviates from strict ideal mhd in one other way .
we impose a floor to the density field of @xmath30 times the initial maximum density in order to prevent the numerical integration from producing negative densities .
this essentially amounts to a subtle distributed mass source .
the density only reaches this floor close to the @xmath17-boundary ( i.e. , many scale heights above and below the disk plane ) . from these initial conditions , the equations of ideal mhd
were evolved using an mpi parallelized version of the zeus code ( stone & norman 1992a , b for a discussion of the basic zeus algorithm ) .
the computational domain was the cylindrical wedge defined by @xmath31 $ ] , @xmath32 $ ] and @xmath33 $ ] .
the computational grid was uniformly spaced in @xmath12 with @xmath34 .
this gridding provides @xmath35 grid cells per vertical pressure scale - height in the body of the disk , adequate to capture the mri and the large - scale properties of the mhd turbulence that it drives , while also giving a radial grid spacing that is only a factor of two larger than the vertical grid spacing .
it is highly desirable to avoid large aspect ratio computational cells due to the anisotropic numerical viscosity and resistivity that it would introduce .
periodic boundary conditions were imposed on the @xmath36-boundaries , and `` zero - gradient '' outflow boundary conditions were imposed at both the inner and outer radial boundaries ( stone & norman 1992a , b ) .
the choice of the @xmath17-boundary condition is less clear .
experiments employing outflow boundaries in the @xmath17-direction showed that tenuous matter generally flows slowly across these boundaries at very sub - sonic and sub - alfvnic speeds ; strictly , this invalidates the use of such boundary conditions ( since the flow on the other side of the boundary should be able to act back on the simulation domain ) .
we also note that the use of this outflow boundary condition together with the imposition of the density floor led to occasional numerical instabilities at the @xmath17-boundaries ( in the form of a small and accelerating magnetized vortex ) that , while having no effect on the body of the accretion disk , would eventually halt the simulation .
instead , we choose to employ periodic boundary conditions in the @xmath17-directions . while this is obviously unphysical in the sense that matter can not leave the simulation domain in the vertical direction , it does guarantee mathematically reasonable behaviour at the boundary ( eliminating numerical instabilities ) and , more importantly , appears to have no effect on the dynamics of the accretion disk itself ( as diagnosed through comparisons with our vertical - outflow runs ) . .
there are 10 contours per decade of density .
the distinct ridges at approximately @xmath37 and @xmath38 are cuts through transient spiral density waves .
the simulation domain has an inner radial boundary at @xmath39 .
however , the figure shows an extension of this domain to @xmath40 along with a representation of the event horizon ( thick solid black line ) .
notice the rapid transition of the high density disk into a geometrically - thin , low density accretion stream close to the isco .
, scaledwidth=90.0% ] the initial state undergoes rapid evolution as radial pressure forces push matter into the ( numerical ) vacuum within the isco .
this drives axisymmetric , outward - radially directed gravity waves which rapidly break to become outward traveling rolls .
these hydrodynamic transients are short lived , lasting for only @xmath41 , where @xmath42 .
concurrently , mri amplifies the initial magnetic field until the volume - averaged @xmath43 at which point the flow becomes turbulent .
the simulated disk is turbulent at all radii after @xmath44 . over the subsequent 7 orbits , the @xmath45 of the disk declines a little due to the effects of magnetic buoyancy .
examination of the mass accretion rate and total magnetic and thermal energy of the disk show that a quasi - steady state ( characterized by approximately constant total magnetic and thermal energies ) is achieved after approximately @xmath46 orbital periods . in this steady state ,
the dense body of the disk ( i.e. , close to the mid - plane ) is characterized by @xmath47 . while this is significantly larger than the @xmath45 found in global simulations of thicker disk , it is in line with what might be expected for thin accretion disks with zero net field as diagnosed through local shearing - box simulations ( hawley , gammie & balbus 1996 ) .
we follow the simulation for a total of 40 isco orbital periods in order to confirm that it is , indeed , in a quasi - steady state . figure [ fig : snapshot ] shows the density of the disk across an azimuthal cross - section at a time @xmath48 .
( solid black line )
. all quantities are determined at @xmath48.,title="fig:",scaledwidth=80.0% ] ( solid black line ) .
all quantities are determined at @xmath48.,title="fig:",scaledwidth=80.0% ] using this simulation , we can examine the nature of the flow close to the isco .
figure [ fig : density ] ( top ) shows the mid - plane density and total vertical column density through the disk as a function of @xmath16 .
we see that both the midplane density and total column density undergo a rapid decrease over a small range of radii as the flow crosses the isco ; the mid - plane density drops by a factor of 20 and the column density drops by a factor of 60 as between @xmath49 and @xmath50 . just within the isco ,
the radial scale - length for the mid - plane density and the column density ( i.e. , the change in radius over which these quantities drop by a factor @xmath51 ) varies from @xmath52 , i.e. , within a factor of two of the vertical scale - height of the disk outside of the isco . in section 4
, we will show that this precipitous drop in density close to the isco effectively truncates the iron line emission ( i.e. , using the terminology of krolik & hawley [ 2002 ] , we argue that the disk has a `` reflection edge '' that is close to the isco ) . as expected from simple continuity
, this density drop accompanies a rapid increase in the ( inward ) radial velocity of the gas ( bottom panel of fig .
[ fig : density ] ) .
the azimuthally - averaged midplane radial velocity is very close to that of a ballistic infall starting from a circular orbit at @xmath53 ( i.e. , inset from the isco by an amount comparable to the vertical pressure scale - height of the disk just beyond the isco ) .
interestingly , the inflow becomes super - alfvnic within @xmath54 , implying that mhd forces can no longer transport energy and angular momentum upstream within this radius .
this is the `` stress edge '' of the disk , as defined by krolik & hawley ( 2002 ) .
the flow is essentially ballistic within this radius .
a comparison of this result with those of hawley & krolik ( 2001 ) supports the conjecture ( armitage , reynolds & chiang 2000 ; afshordi & paczynski 2003 ) that the magnetic extraction of energy and angular momentum from within the isco ( krolik 1999 ; gammie 1999 ) becomes less important as one considers progressively thinner accretion disks , at least for the initially zero net - field case . .
, scaledwidth=80.0% ] another radius of interest is the `` turbulence edge '' , i.e. , the radius at which the flow transitions from being a turbulence dominated flow to a more laminar radial inflow . following krolik & hawley ( 2002 )
, we examine the turbulence edge by comparing the density - weighted azimuthally and vertically - averaged radial velocity @xmath55 to the density - weighted root mean square fluctuations of @xmath56 around that average ( see fig .
[ fig : turb_edge ] ) .
this figure reveals that the disk does not start feeling the effect of the plunging region until @xmath49 , and that the formal turbulence edge at @xmath57 .
again , we see an abrupt transition close to the isco .
we note that the results discussed in this section appear to be mostly insensitive to the numerical resolution and placement of the vertical boundaries .
re - running the simulation with half the radial and vertical resolution leads to 2030% increases in the density field and column density within the isco , with a comparable decrease in the inflow velocity .
a third simulation in which the vertical size of the computational domain is doubled to @xmath58 $ ] ( at the reduced resolution ) shows essentially identical density and velocity fields both within the disk and inside the isco .
as discussed in the introduction , the current methodology for using broad iron lines to constrain black hole spin relies on the assumption that the iron line emission essentially truncates at the isco . here
, we employ our mhd simulation to examine this assumption .
the physical requirements for producing significant iron line emission from within the isco were first addressed by reynolds & begelman ( 1997 ; hereafter rb97 ) using a simple one - zone ballistic model for the accretion flow .
there are three principal considerations .
firstly , significant iron line production requires the flow to be thomson - thick in the vertical direction . for a steadily accreting source with a radiative efficiency @xmath59 , conservation of baryon number gives the electron scattering optical depth of the plunging region to be , @xmath60 where @xmath61 is the radial component of the 4-velocity , and @xmath62 is the eddington ratio of the source . evaluating this at the event horizon of a schwarzschild black hole (
where @xmath63 and @xmath64 ) , we see that the entire flow within the isco is thomson - thick provided that @xmath65 . using the canonical schwarzschild value of @xmath66 , we see that the entire flow is optically - thick provided that it is accreting at 6% or more of its eddington limit , a condition that is very likely satisfied for the active agn and high - state gbhbs in which we find relativistic broad iron lines .
of course , for more rapidly rotating black holes , the radiative efficiency is higher ( achieving a theoretical maximum of @xmath67 for a black hole with @xmath68 ) leading to the conclusion that the accretion flows around rapidly spinning black holes may have an inner optically - thin region unless the eddington ratio is close to unity .
the second requirement for significant iron line emission within the isco is that this region is irradiated by a hard x - ray source .
rb97 adopted the optimal geometry , a hard x - ray source that is located above the black hole , on or close to the symmetry axis of the system .
observational support for just such a geometry has emerged from detailed studies of iron line and reflection bump variability by _ xmm - newton _ ( fabian & vaughan 2003 ; miniutti & fabian 2004 ) and _ suzaku _ ( miniutti et al .
2007 ) .
the third and most constraining requirement is that , despite the irradiation by the hard x - ray source ( which may be significantly enhanced by gravitational light - bending ) and the plummeting density , the surface layers of the plunging flow must maintain a sufficiently low ionization state to still produce iron features .
rb97 found this to be possible only in the case where a very small fraction of the accretion power ( @xmath69 or less ) is channeled into the non - thermal x - ray source .
even in this case , the emission line from within the isco is mostly from h - like and he - like iron .
young , ross & fabian ( 1998 ) showed that a significant contribution of such highly ionized iron line emission from within the isco would be accompanied by a deep k - shell absorption edge of ionized iron that is absent from the data .
of course , even a fully ionized plunge region will still reflect incident x - ray photons in this case , the interaction is dominated by compton scattering and , for x - ray energies under consideration here , the reflected spectrum has the same shape as the incident continuum . when interpreting time - averaged spectra , one could not distinguish between direct primary x - ray emission and primary x - ray emission that has been reflected off the fully ionized flow .
however , spectral variability and time - lags might reveal such a component .
we now reassess the ionization state of the accretion flow within the isco , and its ability to produce significant amounts of iron line emission , in the light of the high - resolution mhd simulation presented in sections 2 and 3 . as a preliminary to this discussion
, we must dimensionalize our simulation . for an assumed eddington ratio @xmath70
, we can apply eqn .
[ eq : tau ] to the flow at the inner radial edge of the simulation to relate the ( dimensionless ) column density in the simulation to the physical column density in the corresponding `` real '' accretion flow .
further , once the mass @xmath8 of the central object is chosen , the length and velocity scale of simulation is set , allowing one to dimensionalize the density and pressure .
the x - ray reflection spectrum is sensitive to the ionization state of the matter at the x - ray `` photosphere '' of the accretion disk , situated approximately at the @xmath71 surface . for the conditions relevant to these irradiated disk atmospheres , photoionization dominates collisional ionization .
furthermore , the photoionization and recombination timescale of the plasma in both the disk and the plunge region is always very short .
hence , the ionization state of the plasma can be described by an ionization parameter , @xmath72 where @xmath73 is the ionizing ( @xmath74ev ) radiation flux and @xmath75 is the electron number density of the plasma at the photosphere .
the density and hence ionization state of the plasma at the photosphere will very much depend on the vertical structure of the accretion disk .
while we expect our simulation to capture well the radial angular momentum transport and hence the radial structure of the disk around the isco , we are forced to neglect physics that is important for determining the vertical structure .
most importantly , we fail to capture the radiative cooling of the disk material , as well as the role of radiation pressure and radiative transport in the vertical structure .
these radiation effects are just recently being included in local ( shearing box ) simulations of accretion disks ( e.g. , see turner 2004 ; blaes , hirose and krolik 2007 ) and it is not yet feasible to perform the corresponding global disk simulations .
thus , we must attempt to diagnose the ionization state of the disk atmosphere despite the knowledge that the vertical structure of the simulated atmosphere is incorrect .
we parameterize this uncertainty via the ratio of the density at the photosphere to the average density , @xmath76 , defined by @xmath77 where @xmath78 is the scale height defined as the @xmath17-value at which the density falls to @xmath79 of its midplane value .
we shall refer to @xmath76 as the _ photospheric density parameter_. we expect the radial dependence of @xmath80 and @xmath78 close to the isco to be determined from the radial dynamics and hence be well described by this simulation .
we discuss constraints on @xmath76 below . in order to examine the radial run of ionization parameter
, we must specify the illuminating ionizing radiation flux as a function of radius . for concreteness ( and following rb97 )
, we shall consider a source of ionizing flux situated at a height @xmath81 on the @xmath17-axis . in the absence of any relativistic effects , the resulting illumination pattern ( assuming a flat thin disk ) would be @xmath82 where @xmath83 is the ionizing luminosity of the source .
this form is modified due to gravitational light - bending , and the gravitational and doppler redshifting / blueshifting of source photons ( including the @xmath84-correction that brings photons into or takes them out of the ionizing regime ) . for @xmath85 ,
the result is to remove the flattening of the illuminating profiles the resulting profile typically _ steepens _ from @xmath86 to approximately @xmath87 as one considers smaller and smaller radii ( fukumura & kazanas 2007 ) . in what follows
, we shall make the conservative assumption that the illumination law maintains a @xmath86 dependence at all radii @xmath88 this is conservative in the sense that it under - illuminates the innermost regions of the disk ( including the region within the isco ) and hence produces an under - estimate of the ionization parameter in the plunge region .
thus , the final expression that we use to estimate an ionization parameter from our simulation is @xmath89 which , when scaled to typical parameters relevant to a luminous seyfert galaxy or a high - state gbhb gives , @xmath90 where @xmath91 is the eddington ratio for just the ionizing luminosity .
note that eqn .
[ eq : tau ] is only strictly meaningful inside the plunge region ( where the directed radial inflow exceeds the turbulent fluctuations ) and hence it would be inappropriate to use eqn .
[ eq : tau ] to simplify eqn .
[ eq : scaledxi ] . to demonstrate the problems encountered
if we naively adopt the vertical structure of the simulated disk , let us examine the ionization in the disk at @xmath92 .
observations tell us that these radii in the accretion disks of moderately luminous seyfert galaxies ( with @xmath93 and @xmath94 ) are capable of producing `` cold '' 6.4kev iron lines , implying that @xmath95 . employing the @xmath48 snapshot ,
the simulation gives @xmath96 and , for this eddington ratio , a total optical depth of @xmath97 .
however , integrating a ray into the disk , we hit the @xmath98 surface at @xmath99 giving a photospheric density parameter @xmath100 .
for @xmath101 , this gives an ionization parameter of @xmath102 .
such a high ionization parameter would render the disk completely incapable of producing iron line features , contrary to observations .
the problem arises due to the low - density , high - temperature atmosphere of gas that envelopes the simulated accretion disk . in real disks ,
we expect radiative cooling ( principally compton cooling ) to collapse this tenuous atmosphere back towards the dense parts of the disk , thereby increasing the plasma density at the photosphere and allowing iron line features to be emitted .
the density at the photosphere would be further increased by the action of radiation pressure in the optically - thick body of the disk which tends to flatten the density profile in the optically - thick body of the disk .
we can see from the numbers above that the photospheric density parameter must be increased to @xmath103 in order to produce the iron line features found in the observations . .
for all curves shown , @xmath101 and @xmath104 . within this choice of parameters , these curves are valid for all accretion rates .
shown here are the results for @xmath105 ( black , solid line ) and the extreme case of @xmath106 ( red , dashed line ) .
the blue dot - dashed line shows an ionization state of @xmath107 , above which the production of any iron line emission becomes extremely inefficient .
see main text for further details.,scaledwidth=80.0% ] in the absence of global simulations that include this physics , we must take a semi - empirical approach .
as is clear from above , we need @xmath108 outside of the isco in order to produce the iron line features that we see in many seyfert galaxies and high - state gbhbs .
we take the approach of assuming that @xmath109 maintains a constant value of order unity into the plunging region .
figure [ fig : xi ] shows the run of ionization parameter for @xmath105 and the extreme case of @xmath106 . since @xmath110 , eqn .
[ eq : scaledxi ] and the curves plotted in fig . [ fig : xi ] depend only on the fraction of the radiation that is ionizing ( @xmath111 ) and not the absolute accretion rate or eddington ratio .
note that the photospheric ionization parameter climbs steeply as one moves inwards into the plunge region .
the ionized reflection models of ross & fabian ( 2005 ) show that distinguishable iron line features are lost when @xmath112 .
in fact , a deep iron absorption edge ( which is not present in the data ) accompanies the iron line once @xmath113 . even for the extreme case of @xmath106 ( i.e.
, the photosphere is characterized by the full average density of the disk ) , the plunge region becomes incapable of producing relevant iron line emission within @xmath114 . for @xmath105 ,
the truncation radius is @xmath115 .
in section 4 , we have demonstrated that the presence of the isco leaves a strong imprint on the reflection spectrum of the accretion disk due to the rapid increase in ionization parameter as the flow plunges inwards .
the increase in ionization parameter is directly associated with the decrease in both the midplane density and column density of the disk as the flow accelerates inwards .
more precisely , for the particular scenario simulated here , we find that the transition in the flow properties occurs on a _ radial _ length scale which is similar to the _ vertical _ scaleheight of the disk .
our simulation shows that the azimuthally - averaged midplane radial velocity is very close to that of a ballistic infall starting from a circular orbit at @xmath53 ( i.e. , inset from the isco by an amount comparable to the vertical pressure scale - height of the disk beyond the isco ) .
indeed , our results explicitly demonstrate that the turbulence edge , stress edge , and reflection edge all occur within 2 scaleheights of the isco .
thus , the basic principle underlying the use of broad iron lines as a spin diagnostic is validated .
however , it should also be apparent that the iron line emission may not truncate _
precisely _ at the isco .
thus , we must expect to introduce a systematic error in the black hole spin when we fit observational data with models that do truncate precisely at the isco .
a detailed quantification of these systematic errors would involve fitting isco - truncated iron line profiles to realistic synthetic data that have been produced from models that include some contribution from within the isco . here
, we address this issue using a simpler ( if somewhat less constraining ) approach based on the maximum observed redshift of the iron line features .
motivated by our simulation , we consider the following ( relativistic ) toy - model for the plunge region of an accretion disk around a rotating black hole .
we consider an accretion disk whose mid - plane lies in the equatorial plane of a kerr spacetime with dimensionless spin parameter @xmath116 .
we suppose that the disk has a vertical scale - height of @xmath78 just outside of the isco ; @xmath78 and @xmath116 are considered parameters of this toy - model .
we assume that the accreting matter follows circular test - particle orbits within the kerr spacetime down to a radius @xmath117 .
motivated by our mhd simulation above , we suppose that @xmath117 is located a proper radial distance @xmath78 inside the isco of the kerr black hole ( as measured by an observer orbiting with the flow at @xmath117 ) .
the flow within @xmath117 is assumed to follow ballistic trajectories with the specific energy and specific angular momentum of the circular flow at @xmath117 .
thus the velocity field of the accretion flow is defined once we specify @xmath78 and @xmath116 .
see the appendix of reynolds et al .
( 1999 ) for a summary of the relativistic expressions used to compute this velocity field .
we assume that there is a source of ionizing radiation situated on the spin axis of the black hole at a height @xmath118 .
given the velocity field for the accretion flow described above , we can combine the conservation of baryon number ( eqn .
[ eq : tau ] ) and the relation between ionization parameter and vertical optical depth ( eqn .
[ eq : xi ] ) in order to deduce that @xmath119 where @xmath120 is the efficiency with which the rest mass of the accretion flow is converted into _ ionizing _ luminosity of the source .
then , for assumed values of @xmath121 , @xmath120 , @xmath109 and @xmath122 , we can estimate the radius in the accretion flow becomes too highly ionized to produce significant iron line emission .
we set this limiting ionization parameter to be @xmath123 .
suppose that one is attempting to constrain the spin of the black hole through the maximum redshift of the iron line .
if the true spin of the black hole is @xmath124 , we use the above toy - model to determine the innermost radius that contributes to the observable iron line emission .
hence , the observed iron line feature will be more highly redshifted than if it were truncated at the isco .
we use the algorithms and subroutines of roland speith ( speith , riffert & ruder 1995 ) to compute the maximum observed redshift of this iron feature .
we then determine the black hole spin @xmath125 for which the same maximum redshift would be obtained if the iron line emitting region is strictly truncated at the isco . as a function of actual spin @xmath124 using the maximum redshift method described in the text
; here we assume @xmath105 , @xmath118 , @xmath126 .
we also assume that the accretion flow has the standard radiative efficiency of a novikov & thorne ( 1974 ) accretion disk , and that 20% of that luminosity is in the form of ionizing radiation .
the _ top panel _ shows the effect of varying the vertical scale height of the disk just outside of the isco ; an inclination of @xmath127 is assumed and lines show the very thin case @xmath128 ( solid black line ) , @xmath129 ( red dashed line ) , @xmath130 ( dot - dashed blue line ) , and @xmath131 ( dot - dot - dot - dashed blue line ) .
the _ bottom panel _ demonstrates that the viewing inclination of the accretion disk has only minor impact on these constraints ; the razor - thin disk limit @xmath132 is assumed and lines show the face - on case @xmath133 ( solid black line ) , @xmath127 ( red dashed line ) , @xmath134 ( dot - dashed blue line ) , and @xmath135 ( dot - dot - dot - dashed magenta line).,title="fig:",scaledwidth=80.0% ] as a function of actual spin @xmath124 using the maximum redshift method described in the text ; here we assume @xmath105 , @xmath118 , @xmath126 .
we also assume that the accretion flow has the standard radiative efficiency of a novikov & thorne ( 1974 ) accretion disk , and that 20% of that luminosity is in the form of ionizing radiation .
the _ top panel _ shows the effect of varying the vertical scale height of the disk just outside of the isco ; an inclination of @xmath127 is assumed and lines show the very thin case @xmath128 ( solid black line ) , @xmath129 ( red dashed line ) , @xmath130 ( dot - dashed blue line ) , and @xmath131 ( dot - dot - dot - dashed blue line ) .
the _ bottom panel _ demonstrates that the viewing inclination of the accretion disk has only minor impact on these constraints ; the razor - thin disk limit @xmath132 is assumed and lines show the face - on case @xmath133 ( solid black line ) , @xmath127 ( red dashed line ) , @xmath134 ( dot - dashed blue line ) , and @xmath135 ( dot - dot - dot - dashed magenta line).,title="fig:",scaledwidth=80.0% ] [ fig : spinerror ] figure [ fig : spinerror ] shows the results of this exercise ; the difference between @xmath125 and @xmath136 in these plots indicates the degree to which the black hole spin can be overestimated due to the effects of unmodelled iron line emission within the plunge region . in generating these figures ,
we have assumed the canonical value of the photospheric density parameter @xmath105 and a source which is at @xmath118 .
motivated by our simulations , we also assume that the flow within the plunge region has collapsed in the vertical direction such that @xmath137 .
finally , we assume that the total radiative efficiency of the flow is that of a standard novikov & thorne ( 1974 ) accretion disk , and that 20% of this luminosity is release in the form of ionizing luminosity from the on - axis source .
this last assumption is probably conservative and motivated by the multiwaveband observation of the prototypical source mcg6 - 30 - 15 which shows that _ at least _ 20% of the accretion luminosity is radiated into the highly - variable x - ray/@xmath138-ray tail ( reynolds et al .
1997 ) .
there are several noteworthy aspects of the results reported in fig .
[ fig : spinerror ] .
firstly , the uncertainty in the black hole spin parameter introduced by bleeding of the iron line emission region within the isco is generally more important for slowly spinning black holes .
this is directly related to the fact that both the spacetime metric and the properties of the accretion disk are slowly varying functions of black hole spin when @xmath139 ; hence a significant change in black hole spin is required if an isco - truncated iron line is to mimick the additional redshift introduced by the bleeding of the line emission region within the isco .
secondly , the systematic error on the black hole spin increases as one considers accretion disks that have greater vertical scale height just outside of the isco . even for @xmath140
, however , one will still clearly distinguish rapidly spinning black holes ( @xmath141 ) from slowly spinning holes .
future global simulations of radiative and radiation - pressure dominated accretion disks will be required in order to assess the realistic range of disk thicknesses that characterize broad iron line sources .
we do note , however , that current observations impose an important constraint on the geometric thickness of the inner disk in broad iron line seyfert galaxies and gbhbs .
for example , spectral modeling of mcg6 - 30 - 15 shows the inner accretion disk to have a ( photospheric ) ionization parameter of @xmath142 ( brenneman & reynolds 2006 ) . using eqn [ eq : scaledxi ] with canonical parameters discussed in section 4.2 ( @xmath143 , @xmath144 , @xmath145 )
, we see that this measured ionization parameter along with the condition @xmath146 demands that @xmath147 at @xmath148 , and @xmath149 at @xmath150 .
of course , a more rigorous exploration of this toy - model , employing relativistically correct treatments of the ionizing irradiation ( beyond the scope of this paper ) , is required to obtain reliable limits on the disk thickness for rapidly rotating black holes .
thirdly , the precise inclination of the accretion disk has a very weak effect on the systematic error on the black hole spin ( as long as the inclination is known from , for example , modeling of the overall broad iron line profile ) .
our results suggest that the true spin and the inferred spin become much closer as one considers more rapidly rotating black holes .
of course , for extreme spin values , the inner emitting part of the disk is very close to the event horizon and well within the ergosphere of the black hole .
interactions of the ( magnetized ) accretion flow with the strong frame - dragging will likely invalidate the assumptions of the toy - model presented here .
significant further work using fully relativistic codes to explore thin accretion disks is required in order to assess whether the picture of a well defined transition in flow properties around the isco makes sense for such rapidly rotating black holes .
it is a fundamental fact that one can , in principle , observe arbitrarily large gravitational redshifts from around any black hole , irrespective of its spin parameter . indeed
this is very closely tied to the definition of a black hole . at first glance
, this fact would appear to render impotent attempts to determine spin using the gravitational redshift of iron emission lines .
however , it is the _ astrophysics _ of black holes accretion and the change in the character of the flow around the isco that gives observations of broad iron lines ( and x - ray reflection in general ) the power to diagnose black hole spin . to understand the behavior of the accretion flow close to the isco , one must capture the angular momentum transport as fully as possible .
models based on parameterized anomalous viscosity ( e.g. , @xmath4-models ) can become acausal within the plunge region and can not capture the non - local mhd stresses that may operate around the isco .
fully 3-d mhd simulations provide the most powerful tool to examine these complex flows .
all previously published global 3-d mhd simulations of black hole accretion disks have revealed rather gradual transitions of flow properties around the isco , but have modeled rather thick ( @xmath151 or greater ) accretion flows .
motivated by the fact that luminous seyfert nuclei and high - state gbhbs ( i.e. , the systems in which we actually observe broad iron lines ) are thought to possess geometrically - thin disks , we have examined the nature of the flow around the isco using a high - resolution pseudo - newtonian mhd simulation of a disk with @xmath5 .
we find a rather abrupt transition of flow properties across the isco ; the radial length - scale of these changes is comparable to the vertical scaleheight of the disk beyond the isco . in the language of krolik & hawley ( 2002 ) ,
the `` stress edge '' and the `` turbulence edge '' of the disk are at @xmath57 compared with @xmath9 .
most importantly for the present discussion , we find that both the midplane density and column density of the disk drop dramatically as the flow accelerates inwards .
although there is no radiation in the simulation , we relate the simulation quantities to the expected ionization state of the accretion flow and find that ( even under conservative assumptions aimed at minimizing the ionization state of the plunging flow ) significant iron line emission can not be produced from inside @xmath50 .
in fact , for canonical parameters , this `` reflection edge '' is at @xmath152 , very close to the isco .
the fact that the iron line emission does not truncate precisely at the isco will , however , introduce a systematic error into iron line based spin measurements that assume strictly isco - truncated line profiles .
we make a first attempt to quantify this systematic error in section 5 , although we note that this approach neglects the spin information which is carried by the full line profile ( as opposed to just the maximum redshift ) . further study is required to assess whether , in practice , the information carried by the full line profile dramatically decreases the systematic error estimates of fig . [
fig : spinerror ] .
we thank laura brenneman , eve ostriker , cole miller , jon miller and john vernaleo for stimulating discussion related to this project .
we also thank the anonymous referee for insightful and courteous comments .
all simulations described in this paper were performed on the beowulf cluster ( `` the borg '' ) supported by the center for theory and computation ( ctc ) in the department of astronomy at the university of maryland college park .
csr gratefully acknowledges support from the national science foundation under grant ast 06 - 07428 , and the university of maryland under the graduate research board semester award program .
acf thanks the royal society for support . | we address the ability of broad iron emission lines from black hole accretion disks to diagnose the spin of the black hole . using a high - resolution 3-dimensional mhd simulation of a geometrically - thin accretion disk in a pseudo - newtonian potential
, we show that both the midplane density and the vertical column density of the accretion flow drop dramatically over a narrow range of radii close to the innermost stable circular orbit ( isco ) .
we argue that this drop of density is accompanied by a sharp increase in the ionization parameter of the x - ray photosphere , and that the resulting imprint of the isco on the x - ray reflection spectrum can be used to constrain spin .
motivated by this simulation , we construct a simplified toy - model of the accretion flow within the isco of a kerr black hole , and use this model to estimate the systematic error on inferred black hole spin that may result from slight bleeding of the iron line emission to the region inside of the isco .
we find that these systematic errors can be significant for slowly spinning black holes but become appreciably smaller as one considers more rapidly rotating black holes . |
let @xmath8 be an algebraically closed field of characteristic 0 , @xmath9\kern - 1.8pt ] } } $ ] , @xmath10 its maximal ideal and @xmath11 its quotient field .
let @xmath5 be a rank - one discrete valuation of @xmath12 , @xmath13 the valuation ring , @xmath14 the maximal ideal and @xmath15 the residue field of @xmath5 .
the center of @xmath5 in @xmath16 is @xmath17 . throughout this paper discrete valuation of @xmath12
" means rank - one discrete valuation of @xmath12 whose center in @xmath16 is the maximal ideal @xmath18 " .
the dimension of @xmath5 is the transcendence degree of @xmath15 over @xmath8 . in order to simplify the redaction
we shall assume , without loss of generality , that the group of @xmath5 is @xmath19 .
let @xmath20 be the completion of @xmath21 with respect to @xmath5 , @xmath22 the extension of @xmath5 to @xmath20 , @xmath23 , @xmath24 and @xmath25 the ring , maximal ideal and the residue field of @xmath22 , respectively ( see @xcite for more details ) .
we know that @xmath15 and @xmath25 are isomorphic ( @xcite ) .
let @xmath26 be a @xmath27section of the natural homomorphism @xmath28 , @xmath29 an element of value 1 and @xmath30 an indeterminate .
we consider the @xmath27isomorphism @xmath31\kern - 1.8pt ] } } } \to{r_{{{\widehat}{v}}}}\ ] ] given by @xmath32 and denote also by @xmath33 its extension to the quotient fields .
we have a @xmath27isomorphism @xmath34 which , when composed with the usual order function on @xmath35 , gives the valuation @xmath22 .
this is the situation we will consider throughout this paper , and we will freely use it without new explicit references . we shall use two basic transformations in order to find an element of value 1 and construct the residue field : 1 .
_ monoidal transformation : _ @xmath36\kern - 1.8pt ] } } & { \longrightarrow } & k{{\rm [ \kern - 1.8pt [ } } { y_1,\ldots , y_n}{{\rm ] \kern - 1.8pt ] } } \\ x_1 & { \longmapsto } & y_1 \\ x_2 & { \longmapsto } & y_1y_2 \\ x_i & { \longmapsto } & y_i,\ i=3,\ldots , n. \end{array}\ ] ] with @xmath37 .
change of coordinates:_@xmath38\kern - 1.8pt ] } } & { \longrightarrow } & l{{\rm [ \kern - 1.8pt [ } } { y_1,\ldots , y_n}{{\rm ] \kern - 1.8pt ] } } \\ x_1 & { \longmapsto } & y_1 \\ x_i & { \longmapsto } & y_i+c_iy_1,\ i=2,\ldots , n , \end{array}\ ] ] where @xmath39 and @xmath40 is an extension field of @xmath8 . for both transformations we have the following facts : 1 .
the transformations are one to one : in the case of the monoidal transformations this property is well known . in the other case
it is a consequence of @xcite ( corollary 2 , page 137 ) .
2 . new variables @xmath41 lie in @xmath23 , so we can put @xmath42 .
3 . let @xmath43 be the restriction of @xmath44 to @xmath21 .
let us denote by @xmath45 the natural extension of @xmath46 to the field @xmath47 .
then @xmath48 , with @xmath49 the usual order function over @xmath50 .
therefore , if @xmath51 is injective we can extend the valuation @xmath5 to the field @xmath47 and the extension is @xmath52 .
from now on transformation will mean monoidal transformation , change of coordinates , variables interchanges or finite compositions of these .
remember that we are assuming that the group of @xmath5 is @xmath19 , so there exists an element @xmath53 such that @xmath54 .
[ lema1 ] let @xmath55 for all @xmath56 . by a finite number of monoidal transformations we can find @xmath57 elements
@xmath58 such that @xmath59 .
we can suppose that @xmath60 and consider the following two steps : * step 1.- * if there exists @xmath61 such that @xmath62 for all @xmath63 , then for each @xmath64 we apply @xmath65 monoidal transformations @xmath36\kern - 1.8pt ] } } & { \longrightarrow } & k{{\rm [ \kern - 1.8pt [ } } { y_1,\ldots , y_n}{{\rm ] \kern - 1.8pt ] } } \\ x_i & { \longmapsto } & y_1y_i\\ x_j & { \longmapsto } & y_j,\ j\ne i. \end{array}\ ] ] trivially @xmath66 for all @xmath67 .
* step 2.- * assume there exists @xmath64 , with @xmath68 , such that @xmath69 does not divide to @xmath70 .
we can suppose that @xmath71 with no loss of generality and then @xmath72 . so we apply @xmath73 times the monoidal transformation @xmath36\kern - 1.8pt ] } } & { \longrightarrow } & k{{\rm [ \kern - 1.8pt [ } } { y_1,\ldots , y_n}{{\rm ] \kern - 1.8pt ] } } \\ x_2 & { \longmapsto } & y_1y_2 \\ x_i & { \longmapsto } & y_i,\ i\ne 2 \end{array}\ ] ] to obtain a new ring @xmath74\kern - 1.8pt ] } } $ ] where @xmath75 and @xmath76 is the element of minimum value .
as the values of the variables are greater than zero , in a finite number of steps 2 we come to the situation of step 1 .
in fact , this algorithm is equivalent to the euclidean algorithm " to compute the greatest common divisor of @xmath77 .
we can construct an element of value 1 applying a finite number of monoidal transformations and changes of coordinates .
we call @xmath78 the elements found after @xmath79 transformations
. we can suppose that we have applied the previous lemma to obtain @xmath80 such that @xmath81 for all @xmath67 .
let us prove that there exists @xmath39 for each @xmath82 such that @xmath83 .
we can take @xmath84 and so it suffices taking @xmath85 and @xmath86 .
the following two steps defines a procedure to obtain an element of value 1 : * step 1.- * we apply the coordinate change @xmath87\kern - 1.8pt ] } } & { \longrightarrow } & l{{\rm [ \kern - 1.8pt [ } } y_{1,2},\ldots , y_{n,2}{{\rm ] \kern - 1.8pt ] } } \\ y_{1,1 } & { \longmapsto } & y_{1,2 } \\
y_{i,1 } & { \longmapsto } & y_{i,2}+ c_iy_{1,2},\ i=2,\ldots , n. \end{array}\ ] ] with this transformation the values of the new variables are not equal to @xmath88 . * step 2.- * we apply lemma [ lema1 ] to equalize the values of elements and go to step 1 .
obviously , the minimum of the values of the elements does not increase , because the greater common divisor of the values does not exceed the minimum of the values .
moreover the first variable does not change .
if we obtain an element of value 1 then we are finished .
we have to show that the procedure produces an element of value 1 in a finite number of transformations . the only way for the process to be infinite is that , in step 2 , the minimum of the values of the elements does not decrease .
this means that , in step 1 , the value of the first variable divides the values of the new variables .
the composition of steps 1 and 2 is the transformation @xmath89\kern - 1.8pt ] } } & { \longrightarrow } & l{{\rm [ \kern - 1.8pt [ } } y_{1,r+1}\ldots , y_{n , r+1}{{\rm ] \kern - 1.8pt ] } } \\ y_{1,r } & { \longmapsto } & y_{1,r+1 } \\ y_{i , r } & { \longmapsto } & y_{i , r+1}+c_iy_{1,r+1}^{m_i},\ i=2,\ldots , n. \end{array}\ ] ] if we use steps 1 and 2 infinitely many times , we have an infinite sequence of transformations @xmath90\kern - 1.8pt ] } } & { \longrightarrow } & l{{\rm [ \kern - 1.8pt [ } } y_{1,j},\ldots , y_{n , j}{{\rm ] \kern - 1.8pt ] } } \\ y_1 & { \longmapsto } & y_{1,j } \\
y_i & { \longmapsto } & y_{i , j}+\sum_{k=1}^j c_{i , k}y_{1,j}^{m_{i , k}},\ i=2,\ldots , n. \end{array}\ ] ] then we can obtain an infinite sequence of variables @xmath91 with @xmath92 for all @xmath93
. so any sequence of partial sums of the series @xmath94 have strictly increasing values .
then these series converge to zero in @xmath23 , so @xmath95 let @xmath96 , then @xmath97 in this situation , the group of @xmath5 is @xmath98 ( see @xcite ) but as the group is assumed to be @xmath19 , @xmath99 .
let us consider the embedding @xmath100\kern - 1.8pt ] } } & { \longrightarrow } & { { \mathbb{c}}}(t_2,t_3){{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } \\ x_1 & { \longmapsto } & t^2 \\ x_2 & { \longmapsto } & t_2t^4+t_2t^6 \\ x_3 & { \longmapsto } & t_2t^2+t_3t^5 \end{array}\ ] ] with @xmath30 , @xmath101 and @xmath102 variables over @xmath103 .
we are going to denote its extension to the quotient fields by @xmath104 as well .
the composition of this injective homomorphism with the order function in @xmath30 gives a discrete valuation of @xmath105 , @xmath106 .
if we apply the procedure given in this section we construct the following element of value 1 : @xmath107 where @xmath108 such that @xmath109 with @xmath110\kern - 1.8pt ] } } $ ] . in this case
we can take @xmath111 we need to know some elements @xmath39 ( or @xmath112 and @xmath26 ) such that @xmath113 for each @xmath82 in order to apply the procedure described in the proof of theorem 2 .
let @xmath114 be a field , if the valuation is given as a composition of an injective homomorphism @xmath115\kern - 1.8pt ] } } & { \longrightarrow } & \delta{{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } \\ x_i & { \longmapsto } & \sum_{j\geq 1}a_{i , j}t^j \end{array}\ ] ] with the usual order function of @xmath116 , @xmath106 , then we can find the @xmath117 s using the coefficients @xmath118 of @xmath119 .
in the following sections we give a procedure to construct the residue field @xmath15 of a discrete valuation of @xmath12 , as a transcendental extension of @xmath8 . before the description of the procedure we have to do the following remark about the @xmath27section @xmath120 . [ remark4 ]
we are going to check all the variables searching those residues which generate the extension @xmath121 .
hence we will have to move between @xmath23 and @xmath25 by the @xmath27section @xmath120 and the natural homomorphism @xmath122
. we can do the following considerations : * 1 ) * let us consider the diagram = 0.75 mm ( 49.00,60.00 ) ( 10.00,4.00)(0,0)[cc]@xmath8 ( 29.50,4.33)(1,0)14.00 ( 29.50,4.33)(-1,0)14.00 ( 28.00,6.33)(0,0)[cc]@xmath123 ( 49.00,4.00)(0,0)[cc]@xmath8 ( 10.00,7.00)(0,1)15.00 ( 49.00,7.00)(0,1)15.00 ( 10.00,27.67)(0,0)[cc]@xmath124 ( 10.00,32.33)(0,1)15.00 ( 49.00,32.33)(0,1)15.00 ( 10.00,52.67)(0,0)[cc]@xmath23 ( 49.00,52.67)(0,0)[cc]@xmath25 ( 16.00,26.00)(1,0)28.00 ( 16.00,51.00)(1,0)28.00 ( 44.00,28.00)(-1,0)28.00 ( 44.00,53.00)(-1,0)28.00 ( 28.00,24.00)(0,0)[cc]@xmath125 ( 28.00,49.00)(0,0)[cc]@xmath125 ( 28.00,30.00)(0,0)[cc]@xmath120 ( 28.00,55.00)(0,0)[cc]@xmath120 ( 49.00,27.67)(0,0)[cc]@xmath126 where @xmath124 and @xmath126 are subfields of @xmath23 and @xmath25 respectively .
let @xmath127 an element such that @xmath128 .
the question is : if @xmath129 is transcendental over @xmath126 , is @xmath130 transcendental over @xmath124 ?
what happens in the algebraic case ?
so we suppose @xmath129 to be transcendental over @xmath126 .
let @xmath131 $ ] be a non - zero polynomial .
let us put @xmath132 then @xmath133 because @xmath129 is transcendental over @xmath126 .
so we have proved that @xmath134 is transcendental over @xmath124 if @xmath129 is transcendental over @xmath126 * 2 ) * in the algebraic case let us consider the next diagram : = 0.75 mm ( 49.00,59.67 ) ( 10.00,4.00)(0,0)[cc]@xmath8 ( 29.50,4.33)(1,0)14.00 ( 29.50,4.33)(-1,0)14.00 ( 28.00,6.33)(0,0)[cc]@xmath123 ( 49.00,4.00)(0,0)[cc]@xmath8 ( 10.00,7.00)(0,1)15.00 ( 49.00,7.00)(0,1)15.00 ( 10.00,27.67)(0,0)[cc]@xmath124 ( 10.00,32.33)(0,1)15.00 ( 49.00,32.33)(0,1)15.00 ( 10.00,52.67)(0,0)[cc]@xmath23 ( 49.00,52.67)(0,0)[cc]@xmath25 ( 16.00,26.00)(1,0)28.00 ( 44.00,28.00)(-1,0)28.00 ( 28.00,24.00)(0,0)[cc]@xmath125 ( 28.00,30.00)(0,0)[cc]@xmath120 ( 49.00,27.67)(0,0)[cc]@xmath126 let @xmath135 be an algebraic element over @xmath126 , with @xmath136 ( i.e. @xmath137 ) .
let @xmath138\ ] ] be its minimal polynomial over @xmath126 .
let us take the polynomial @xmath139,\mbox { with } b_i={\sigma}({\beta}_i).\ ] ] by hensel s lemma ( @xcite , corollary 1 , page 279 ) we know that there exists @xmath140 such that @xmath141 is a simple root of @xmath142 y @xmath143 .
as @xmath144 , @xmath142 is the minimal polynomial of @xmath141 , so we can extend @xmath145\to{{\mathbb{f}}}[a]$ ] .
then we have = 0.75 mm ( 49.00,77.34 ) ( 10.00,4.00)(0,0)[cc]@xmath8 ( 29.50,4.33)(1,0)14.00 ( 29.50,4.33)(-1,0)14.00 ( 28.00,6.33)(0,0)[cc]@xmath123 ( 49.00,4.00)(0,0)[cc]@xmath8 ( 10.00,7.00)(0,1)15.00 ( 49.00,7.00)(0,1)15.00 ( 10.00,27.67)(0,0)[cc]@xmath124 ( 10.00,32.33)(0,1)15.00 ( 49.00,32.33)(0,1)15.00 ( 10.00,51.67)(0,0)[cc]@xmath146 ( 49.00,51.67)(0,0)[cc]@xmath147 ( 10.00,57.00)(0,1)15.00 ( 49.00,57.00)(0,1)15.00 ( 10.00,77.34)(0,0)[cc]@xmath23 ( 49.00,77.34)(0,0)[cc]@xmath25 ( 16.00,26.00)(1,0)28.00 ( 44.00,28.00)(-1,0)28.00 ( 28.00,24.00)(0,0)[cc]@xmath125 ( 28.00,30.00)(0,0)[cc]@xmath120 ( 49.00,27.67)(0,0)[cc]@xmath126 ( 20.00,50.67)(1,0)15.00 ( 35.00,52.67)(-1,0)15.00 ( 28.00,48.67)(0,0)[cc]@xmath125 ( 28.00,54.67)(0,0)[cc]@xmath120 let us consider the set @xmath148 partially ordered by @xmath149 by zorn s lemma there exists a maximal element @xmath150 , and again by hensel s lemma ( @xcite , corollary 2 , page 280 ) we have @xmath151 .
so we can extend @xmath120 to a @xmath27section @xmath152 of @xmath125 in such a way that @xmath153 is an algebraic element over @xmath124 .
* 3 ) * hence we have showed that if @xmath154 , @xmath128 , is a transcendental ( resp .
algebraic ) element over @xmath126 , there exists a @xmath27section of @xmath125 which extends @xmath120 and @xmath130 is transcendental ( resp .
algebraic ) over @xmath124 .
= 0.75 mm ( 49.00,54.00 ) ( 10.00,4.00)(0,0)[cc]@xmath8 ( 29.50,4.33)(1,0)14.00 ( 29.50,4.33)(-1,0)14.00 ( 28.00,6.33)(0,0)[cc]@xmath123 ( 49.00,4.00)(0,0)[cc]@xmath8 ( 10.00,7.00)(0,1)15.00 ( 49.00,7.00)(0,1)15.00 ( 10.00,27.67)(0,0)[cc]@xmath124 ( 10.00,31.33)(0,1)15.00 ( 49.00,31.33)(0,1)15.00 ( 10.00,51.67)(0,0)[cc]@xmath23 ( 49.00,51.67)(0,0)[cc]@xmath25 ( 16.00,26.00)(1,0)28.00 ( 44.00,28.00)(-1,0)28.00 ( 28.00,24.00)(0,0)[cc]@xmath125 ( 28.00,30.50)(0,0)[cc]@xmath120 ( 49.00,27.67)(0,0)[cc]@xmath126 ( 16.00,50.00)(1,0)28.00 ( 44.00,52.00)(-1,0)28.00 ( 28.00,48.00)(0,0)[cc]@xmath125 ( 28.00,54.50)(0,0)[cc]@xmath152
we devote this section to finding a first transcendental residue of @xmath15 over @xmath8 . note that this preliminary transformations construct the residue field in the case @xmath155 .
[ lema6 ] there exists a finite number of monoidal transformations and changes of coordinates that constructs @xmath57 elements @xmath156 such that @xmath157 and the residue @xmath158 is not in @xmath8 .
we can suppose that we have applied lemma [ lema1 ] to obtain @xmath159 such that @xmath160 for all @xmath161 .
in this situation @xmath162 , so @xmath163 . if this residue lies in @xmath8 then there exists @xmath164 such that @xmath165 so @xmath166 and then @xmath167 so we have @xmath168 .
if @xmath169 divides to @xmath170 then @xmath171 with @xmath172 and @xmath173 if the residue of this element lies too in @xmath8 , then exist @xmath174 such that @xmath175 if @xmath169 divides to @xmath176 then @xmath177 with @xmath178 and we can repeat this operation . the above procedure is finite for some pair @xmath179 . we know ( @xcite ) that any discrete valuation of @xmath180 has dimension 1 , so the restriction , @xmath181 , of our valuation @xmath5 to the field @xmath180 is a valuation with dimension 1 , and the dimension of @xmath5 is greater or equal than 1 , because a transcendental residue of @xmath181 over @xmath8 is a transcendental residue of @xmath5 too .
if the procedure never ends for all @xmath179 then all the residues of @xmath5 are in @xmath8 , so the dimension of @xmath5 is 0 and there is a contradiction .
so we can suppose that the above procedure ends for @xmath182 by reordering the variables if necessary .
hence there exists a first transcendental residue .
we can apply the above procedure to the variables @xmath183 , and so we have the transformations : @xmath184 @xmath185 such that one of the following two situation occurs : \a ) @xmath186 divides @xmath187 and the residue of @xmath188 is not in @xmath8 with @xmath189 .
\b ) @xmath186 does not divide @xmath187 . in case
a ) , we make the transformation @xmath190 and apply lemma [ lema1 ] to obtain elements with the same values .
we note these elements by @xmath156 again in order not to complicate the notation .
so , after this procedure , we have a transcendental element @xmath191 over @xmath8 . in case
b ) we make the same transformation and go back to the beginning of the proof .
anyway this procedure stops , because the value of the variables are greater or equal than 1 .
then we can suppose that , after a finite number of transformations , we have @xmath57 elements @xmath156 such that @xmath157 and the residue @xmath158 is not in @xmath8 .
let @xmath106 the discrete valuation of @xmath192 defined by the embedding @xmath193\kern - 1.8pt ] } } & { \longrightarrow } & { { \mathbb{c}}}(u){{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } \\ x_1 & { \longmapsto } & t\\ x_2 & { \longmapsto } & t+t^3+\sum_{i\geq 1}u^it^{i+3 } \end{array}\ ] ] with @xmath194 and @xmath30 independent variables over @xmath103 .
the residue @xmath195 , because @xmath196 .
so we have @xmath197 the residue @xmath198 too , because @xmath199 .
so we have @xmath200 as @xmath201 and @xmath194 is trancendental over @xmath103 , then @xmath202 and this is a first transcendental residue of @xmath15 over @xmath103 .
in this situation we can do the transformation @xmath203\kern - 1.8pt ] } } & { \longrightarrow } & { { \mathbb{c}}}{{\rm [ \kern - 1.8pt [ } } y_1,y_2{{\rm ] \kern - 1.8pt ] } } \\ x_1 & { \longmapsto } & y_1\\ x_2 & { \longmapsto } & y_2y_1 ^ 3+y_1+y_1 ^ 3 \end{array}\ ] ] to obtain elements @xmath204 such that @xmath205 and @xmath206 .
so @xmath207 and the residue @xmath208 is not in @xmath103 . in this example
the extension of the valuation @xmath5 to the field @xmath209 is the usual order function .
theorem [ thm3 ] says that , for @xmath155 , we always have this .
we end up the section with some specific arguments for the case @xmath155 .
the proof of the following lemma is straightforward from ( @xcite , theorem 2.4 ) : let @xmath5 be a discrete valuation of @xmath12 .
if @xmath5 is such that @xmath210 for all forms @xmath211 of degree @xmath79 with respect to the usual degree , then the group of @xmath5 is @xmath212 .
so we have [ thm3 ] in the case @xmath155 , the extension of the valuation @xmath5 to the field @xmath213 is the usual order function . after a finite number of transformations we are in the situation of the end of the previous proof . obviously , if @xmath155 , @xmath214 so @xmath5 can be extended to a valuation @xmath181 over @xmath213 such that @xmath215 .
we denote the extension by @xmath5 for simplifying .
let @xmath26 a @xmath27section of @xmath28 , @xmath191 , @xmath216 a form of degree @xmath79 and @xmath217 . from the construction procedure of @xmath218
we know that @xmath219 ( remember @xmath220 ) .
then @xmath221 where @xmath222 is such that @xmath223 . as @xmath224
, @xmath218 is transcendental over @xmath8 , so @xmath225 and @xmath226 . by the previous lemma ,
the group of @xmath5 is @xmath212 , so @xmath227 and @xmath5 is the usual order function .
let us move to the general case .
assume that @xmath228 and suppose we have applied the procedure of the lemma [ lema6 ] to find @xmath229 such that \a ) the value of these elements are @xmath230 .
\b ) the residue of @xmath231 is transcendental over @xmath8 .
this section and the next one describe the transformations that we have to do in order to construct the residue field of @xmath5 .
let @xmath232 a purely transcendental extension of @xmath8 of transcendence degree 1 .
let @xmath233 defined by @xmath234 we know that there exists a @xmath27section @xmath120 which extends @xmath235 in the sense of the remark [ remark4 ] .
let us suppose that the residue of @xmath236 is algebraic over @xmath237 , and let @xmath238 be its image by @xmath120 .
then @xmath239 . if @xmath169 divides to @xmath170 then there exists @xmath240 and @xmath241 such that @xmath242
let us suppose that @xmath243 is algebraic over @xmath237 too and @xmath169 divides to @xmath176 .
then we can find ourselves in one of the three situations shown in the following items .
* ( situation 1 ) * after a finite number of transformations , we obtain a value @xmath244 such that it is not divided by @xmath169
. then we make the transformation @xmath245 with @xmath246 algebraic over @xmath237 for all @xmath247 .
so we have to apply transformations to find elements with the same values and begin with all the procedure described in this section .
when this occurs , the values of the elements decrease , so we can suppose that after a finite number of transformations we have reached a strictly minimal value .
in fact this value should be 1 , because we are assuming that the values group is @xmath19 .
we shall denote these elements by @xmath156 in order not to complicate the notation .
so we can suppose that this situation will never occur again for any variable .
* ( situation 2 ) * after a finite number of steps , we have a transcendental residue of @xmath237 .
let us denote this residue by @xmath248 .
this means @xmath249 where the elements @xmath250 are algebraic over @xmath237 and @xmath251 is transcendental over @xmath237 .
we shall note @xmath252 .
in this situation , if @xmath253 we can apply monoidal transformations to obtain elements with the same values . we will denote these elements again by @xmath254 .
the extension of the valuation @xmath5 to the field @xmath255 with @xmath256 , is the usual order function , for analogy with the case @xmath155 ( theorem 9 ) .
* ( situation 3 ) * all the residues obtained are algebraic elements .
then we take @xmath257 , an algebraic extension of @xmath237 .
let us suppose that we have repeated the previous construction with each element @xmath258 , so we have a field @xmath259 where each @xmath260 is : - either @xmath261 if @xmath262 are algebraic over @xmath263 and @xmath264 is a transcendental element over @xmath263 ( i.e. situation 2 ) , - or @xmath265 is an algebraic extension ( i.e. situation 3 ) .
so we have two possible situations concerning variable @xmath41 : * 1 ) * there exists a transformation @xmath266 where the elements @xmath267 are algebraic over @xmath268 and @xmath269 is a transcendental element over @xmath268 .
so we have the transcendental extension @xmath270 * 2 ) * all the elements @xmath267 we have constructed are algebraic over @xmath268 , so we have the algebraic extension @xmath271 we have given a procedure to construct elements @xmath272 such that they satisfy these important properties : 1 . after reordering if necessary
, we can suppose that the first @xmath273 elements give us all the transcendental residues over @xmath8 , i.e. the residue of each @xmath274 is transcendental over @xmath268 with @xmath275 .
so the rest of variables @xmath276 are such that we enter in the situation of previous item * 2)*. 2 . with the usual notations
, the extension @xmath277 is algebraic .
[ theorem9 ] the residue field of @xmath5 is @xmath278 and the transcendence degree of @xmath279 over @xmath8 is @xmath280 . in this section
we have given a construction by writing the elements @xmath41 depending on @xmath281 and some transcendental and algebraic residues .
so we have constructed a map @xmath282 this map is not injective in the general case , but we know that @xmath283 .
so the residue field of @xmath5 is equal to the residue field of @xmath49 , i.e. @xmath279 .
a straightforward consequence of this theorem is the following well - known result [ cor8 ] the usual order function over @xmath21 has dimension @xmath284 , i.e. the transcendence degree of its residue field over @xmath8 is @xmath284 .
let @xmath285 be the usual order function over @xmath21 .
all the residues @xmath286 are transcendental over @xmath287 : if this were not the case , there would exist @xmath288 such that @xmath289 and @xmath285 would not be an order function .
so @xmath290 .
in order to compute explicitly the residue field of a valuation we need to construct a section @xmath26 as in remark [ remark4 ] .
this procedure is not constructive in general . as in section 1 ,
if the valuation is given as a composition @xmath106 , where @xmath291\kern - 1.8pt ] } } \to\delta{{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } $ ] is an injective homomorphism and @xmath49 is the order funcion in @xmath292\kern - 1.8pt ] } } $ ] , then we can construct @xmath120 using the coefficients @xmath118 of @xmath293 .
( of course , explicit does not mean effective because we are working with the series @xmath294 and this input is not finite ) .
[ ejemplo ] let us consider the embedding @xmath295\kern - 1.8pt ] } } & { \longrightarrow } & \delta{{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } \\ x_1 & { \longmapsto } & t \\ x_2 & { \longmapsto } & t_2 t \\ x_3 & { \longmapsto } & t_2 ^ 2t+t_2t^2+t_3t^3 \\ x_4 & { \longmapsto } & t_2 ^ 3t+t_2 ^ 2t^2+t_3t^3+t_4t^4\\ x_5 & { \longmapsto } & t_2t\sum_{j\geq 1 } ( t_4^{1/p}t)^j , \end{array}\ ] ] with @xmath30 , @xmath101 , @xmath102 and @xmath296 variables over @xmath103 , @xmath297 prime and @xmath114 is a field such that @xmath298 .
@xmath299 is the algebraic closure of @xmath300 .
we are going to denote its extension to the quotient fields by @xmath104 .
the composition of this injective homomorphism with the order function in @xmath30 gives a discrete valuation of @xmath301 , @xmath106 .
the residues of @xmath302 are not in @xmath103 for @xmath303 .
let us put @xmath304 , a transcendental element over @xmath103 . by remark [ remark4 ] we know how to construct @xmath120 step by step , so let take us @xmath305 and @xmath306 .
the residue @xmath307 is algebraic over @xmath308 , in fact @xmath309 so we can take @xmath310 .
the value of @xmath311 is 2 , therefore we have to see if the residue @xmath312 is algebraic over @xmath308 .
we have that @xmath313 so it is algebraic and we can take @xmath314 . now @xmath315 and we have to check if @xmath316 is algebraic over @xmath317 . in this case , as @xmath318 this residue is transcendental .
so we take @xmath319 let us take @xmath320 .
we have to apply this procedure to @xmath321 .
the residue @xmath322 is algebraic over @xmath323 because @xmath324 so we can take @xmath325 now @xmath326 , and we have to check what happens with the residue @xmath327 as @xmath328 it holds @xmath329 clearly @xmath330 and @xmath331 therefore @xmath332 the following residue is transcendental because @xmath333 and @xmath334 then we can take @xmath335 @xmath336 so @xmath337 . with the variable @xmath338 we obtain the next algebraic residues @xmath339 for all @xmath340 .
so we have @xmath341 , an algebraic extension of @xmath342
. then the residue field of @xmath5 is @xmath343 @xmath344 in this case , by the transformation @xmath345 we can extend the valuation @xmath5 to a discrete valuation @xmath346 of @xmath347 , with the injective homomorphism @xmath348\kern - 1.8pt ] } } & { \longrightarrow } & \delta{{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } \\ y_1 & { \longmapsto } & t \\ y_i & { \longmapsto } & t_it,\ i=2,3,4\\ y_5 & { \longmapsto } & \sum_{j\geq 1 } ( t_4^{1/p}t)^j . \end{array}\ ] ] the restriction @xmath349 is the usual order function .
this is not the general case because @xmath350 may not be injective .
we can summarize the constructions of previous sections in the following theorem wich generalize the results of @xcite 1 .
if the dimension of @xmath5 is @xmath284 , we can embed @xmath1\kern - 1.8pt ] } } $ ] into a ring @xmath351\kern - 1.8pt ] } } $ ] , where @xmath352 and the extended valuation of @xmath5 over the field @xmath7 is the usual order function .
if the dimension of @xmath5 is @xmath353 , we can embed @xmath1\kern - 1.8pt ] } } $ ] into a ring @xmath351\kern - 1.8pt ] } } $ ] , where @xmath352 and the restriction into @xmath354 of the `` extended valuation '' of @xmath5 over @xmath7 is the usual order function . 1 . in the case @xmath356 , @xmath357 for all @xmath358 .
let @xmath49 be the usual order funtion over @xmath359 .
the homomorphism @xmath360 is injective and the valuation @xmath361 of @xmath7 is the usual order fuction over this field .
obviously @xmath181 extends @xmath5 .
2 . if @xmath362 we can consider the elements @xmath363 .
hence we have @xmath364 .
we define the discrete valuation of rank @xmath365 over @xmath366 : @xmath367 @xmath368 the restriction of this valuation to @xmath21 is a rank one discrete valuation , because the value of any element is in @xmath369 .
in fact @xmath370 for all @xmath371 , so @xmath181 `` extends '' @xmath5 in this sense .
obviously @xmath372 is the usual order function .
we want note that this ideal @xmath373 is the _ implicit ideal of @xmath5 _ that appears in some works of m. spivakovsky ( @xcite ) .
let us consider the homomorphism @xmath295\kern - 1.8pt ] } } & { \longrightarrow } & \delta{{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } \\ x_1 & { \longmapsto } & t \\ x_2 & { \longmapsto } & t_2 t \\ x_3 & { \longmapsto } & t_2 ^ 2t+t_2t^2+t_3t^3 \\ x_4 & { \longmapsto } & t_2 ^ 3t+t_2 ^ 2t^2+t_3t^3+t_4t^4\\ x_5 & { \longmapsto } & t_2t\left(\sum_{j\geq 1 } a_j(t_4t)^j\right ) , \end{array}\ ] ] with @xmath374 such that @xmath104 is injective ( we can take @xmath375 ) .
then the residue field of this valuation ( see example [ ejemplo ] ) is @xmath343 @xmath376 by the transformation ( see example [ ejemplo ] ) @xmath377 we obtain a new field @xmath347 , but we can not extend @xmath5 to this field because the homomorphism @xmath348\kern - 1.8pt ] } } & { \longrightarrow } & \delta{{\rm [ \kern - 1.8pt [ } } t{{\rm ] \kern - 1.8pt ] } } \\ y_1 & { \longmapsto } & t \\ y_i & { \longmapsto } & t_it,\ i=2,3,4 \\ y_5 & { \longmapsto } & \sum_{j\geq 1 } a_j(t_4t)^j \end{array}\ ] ] is not injective . then let us take @xmath378 ( because we can consider @xmath379 ) .
then @xmath380 and the discrete valuation of rank 2 defined by @xmath381 for @xmath382 and @xmath383 is such that for all @xmath384 we have @xmath385 and @xmath349 is the usual order function . | in this paper we study the rank one discrete valuations of @xmath0 whose center in @xmath1\kern - 1.8pt ] } } $ ] is the maximal ideal @xmath2 . in sections 2 to 6
we give a construction of a system of parametric equations describing such valuations .
this amounts to finding a parameter and a field of coefficients .
we devote section 2 to finding an element of value 1 , that is , a parameter .
the field of coefficients is the residue field of the valuation , and it is given in section 5 .
the constructions given in these sections are not effective in the general case , because we need either to use the zorn s lemma or to know explicitly a section @xmath3 of the natural homomorphism @xmath4 between the ring and the residue field of the valuation @xmath5 .
however , as a consequence of this construction , in section 7 , we prove that @xmath6 can be embedded into a field @xmath7 , where the _ `` extended valuation '' is as close as possible to the usual order function_. valuation theory , local uniformization , formal power rings , completions .
13f25 , 13f30 , 14b05 , 16w60 |
the effect of inhomogeneity on the critical behaviour of magnetic systems has been considered in various contexts ( e.g. disorder , coupling randomness , quasiperiodic structures ) ; in particular , discrete - spin models defined on fractal topologies possess critical properties significantly different and richer than those found for translationally invariant systems @xcite .
the interest in fractal structures is not purely theoretical : many condensed - matter systems display strong nonuniformity on all length scales and can therefore be characterized as fractal objects ; examples include the backbone of percolation clusters , aggregates obtained from diffusion - limited growth processes , and absorbent surfaces .
one of the most known fractals is the sierpinski gasket ( sg ) , which , due to its exact decimability , allows analytical approaches ; in particular , by means of renormalization group techniques , it was proved that the ising model on the sg exhibits phase transition only at zero temperature , while at any finite temperature the system breaks into domains and loses long - range order @xcite . while this result was found in the thermodynamic limit , at the mesoscopic sizes peculiar and interesting thermodynamic properties arise @xcite .
more precisely , the ising model defined on a finite sg exhibits critical - like features at nonzero , low temperatures and the solution found in the thermodynamic limit turns out to be a poor approximation .
this anomalous behavior has been investigated from a thermodynamic point of view and derives from long - range , slowly decaying correlations at low temperatures @xcite .
the way such a behavior is reflected by the evolution of spin configurations is an item so far overlooked ( even if , for rectangular lattices , the idea of studying cluster dynamics may be traced back to peierls and griffith @xcite ) .
given the importance of the dynamics of ising - like clusters in many research areas , from condensed matter to biological systems @xcite , the definition and the development of proper tools for this kind of analysis would be very useful .
moreover , it would be particularly intriguing for inhomogeneous substrates , due to the emerging non - trivial thermodynamic behavior ; on the other hand , it is just on such structures that the definition of a proper metrization or evolutionary dynamics can be more awkward .
the interest in cluster mobility actually extends to an extremely wide class of models : indeed the ising model on the sg may be seen as a particular realization of cellular automata on graphs , i.e. discrete time dynamical systems assigning to each node of a graph @xmath0 a value chosen in a alphabet @xmath1 , along a rule depending only on a finite neighborhood at previous time @xcite . in this work
we aim to introduce and develop proper algorithms for the quantitative characterization of cluster dynamics on graphs , and we use this approach for the ising model on a finite sg , meant as a prototype of cellular automata on graphs .
the procedure requires a projection of evolving configurations into an appropriate _ partition space _
, where an information - based metrics ( rohlin distance ) and a method measuring the effective emergence of configurational novelty ( reduction process and amplification parameter ) may be naturally defined and worked out in order to focus the changing and the stable components of configurations .
the algorithmic implementations of rohlin distance and related quantities are deeply affected by the topological features of the substrate @xcite .
for instance , in previous implementations designed specifically for automata on regular lattices , the very passage from one to two dimensions yields a much higher order of complexity , and could not be exported on different substrates @xcite . on the contrary
, the algorithm developed here can be directly applied to generic cellular automata for which the metric characterization of cluster dynamics gets feasible ; a brief description is given in the appendix @xmath2 .
our investigations on the ising sg highlight the existence of two `` critical '' temperatures , @xmath3 and @xmath4 demarcating three main regimes which recover , both qualitatively and quantitatively , the results of the previous thermodynamic analysis ; in addition , within the above mentioned regimes , we obtain a more detailed dynamical and geometrical characterization .
in particular , at very low temperatures ( @xmath5 ) , a long - range order is established , and the few small - sized clusters display poor overlap from one time step to the next one : this makes the distance close to zero , and the reduction ineffective . as the spot sizes start to increase , the slowness of the evolution is still such that both non - similarity and overlap between successive configurations rapidly grow . at greater temperatures ( @xmath6 ) ,
large - scale correlations start to decay , clusters of all sizes appear , and overlap gets easier ; then , for @xmath7 , any trace of order has vanished and , even if overlaps are very important , the complexity of magnetic pattern is irreducible .
the progression of such different kinds of non - similarity and dynamical overlap is well described by our parameters ( distance , amplification and intersection , see below ) .
we will also evidence the existence of different scales controlling the disappearance of local and correlated order .
thus , the phenomenology provided by our method , confirms with a deeper geometric insight the peculiar critical - like behavior exhibited by the system we finally notice that , in view of future extensions to non - equilibrium situations , we will adopt a microcanonical dynamics @xcite , which allows implementations even in the presence of temperature gradients . in the next two sections , after recalling basic notations on graphs , we review some facts concerning the thermodynamic properties of the ising model on the sierpinski gasket ( sec . [ sec : sg ] ) , and the microcanonical dynamics working as evolutionary dynamics ( sec . [
sec : micro ] ) .
then , we introduce general procedures for cluster identification and reduction ( sec . [ sec : metrica ] ) and we show our results on the sg ( sec .
[ sec : numerics ] ) .
finally , we present our conclusions and perspectives ( sec . [
sec : con ] ) .
technical remarks about partitions on graphs can be found in the appendices .
a generic graph @xmath8 is mathematically specified by the pair @xmath9 consisting of a non - empty , countable set of points @xmath10 joined pairwise by a set of links @xmath11 .
the cardinality @xmath12 of @xmath10 represents the number of sites making up the graph , i.e. its volume . from an algebraic point of view , a graph is completely described by its adjacency matrix @xmath13 .
every entry of this off - diagonal , symmetric matrix corresponds to a pair of sites , and it equals @xmath14 if and only if this couple is joined by a link , otherwise it is @xmath15 . here
we consider the sierpinski gasket which can be built recursively with the following procedure : the initial state ( @xmath16 ) is a triangle and the @xmath17th stage @xmath18 is obtained joining two of the three external corners of three @xmath19 to form a bigger triangle ( see fig . [
fig : sierp ] ) . in this way , the volume of @xmath20 is @xmath21 .
the gasket is obtained as the limit for @xmath22 of this procedure .
( color on line ) sierpinski gasket of generation @xmath23 ( a ) , @xmath24 ( b ) and @xmath25 ( c ) , with volume @xmath26 , @xmath27 and @xmath28 , respectively . ] the ising model on a generic graph @xmath29 is defined associating the spin variable @xmath30 to every site @xmath31 of the graph , and considering a nearest - neighbours interaction between points @xmath31 and @xmath32 , such that @xmath33 .
the hamiltonian is therefore @xmath34 where @xmath35 denotes the magnetic configuration of the system and the coupling @xmath36 is assumed to be the same for any couple ; in the following we will set @xmath37 .
as it is well known , a magnetic model defined on a finite lattice can not exhibit critical behaviour at nonzero temperatures ; critical features can only emerge when the underlying lattice becomes infinitely large , i.e. in the thermodynamic limit .
another necessary condition in order to have a nonzero critical temperature concerns the topology of the ( infinite ) substrate : for euclidean structures it was rigorously shown that the dimension must be larger than 1 .
analogously , it has been shown that the discrete symmetry ising model on finitely ramified fractals can not have a nonzero critical temperature @xcite . however , the critical behaviour of one - dimensional systems and finitely ramified fractals can be markedly different , since for the latter it is further governed by additional geometric aspects such as ramification , lacunarity , and connectivity @xcite .
the thermodynamic properties of the ising model on the sg were studied in details in @xcite , where it was shown that its scale - invariant , fractal structure leads to highly cooperative correlations and , at sufficiently low temperatures , the correlation length @xmath38 becomes extremely large and slowly decaying ( compared with the one - dimensional case ) , so that any system with size smaller than @xmath38 displays long - range order .
indeed , one can define an apparent magnetic transition temperature @xmath39 as the point where @xmath40 ; for a system of generation @xmath17 ( @xmath41 ) , @xmath42 , for example , for generation @xmath43 and @xmath44 one finds @xmath45 and @xmath46 , respectively .
hence , as the system size is enlarged , @xmath39 diminishes slowly ; more precisely , being @xmath47 the number of nodes for a gasket of generation @xmath17 , one has @xmath48 $ ] .
on the other hand , the specific heat @xmath49 does not display any anomaly associated with long - range order , yet it exhibits a peak at a temperature @xmath50 @xcite .
conversely , the `` reduced '' specific heat @xmath51 , ( basically the derivative of the magnetic energy per link with respect to the inverse temperature @xcite ) , evidences a qualitative difference between the sg and a one - dimensional system , since for the former it exhibits a peak , while for latter it grows monotonically as the temperature is increased .
indeed , for the sg , as @xmath52 is increased from small values , the energy decreases rather slowly as a result of a relatively large cost due to the large ( bulk ) coordination number ; at larger temperatures it becomes progressively easier to reduce the energy and this reflects the fact that the large fluctuations begin to develop ; finally , since the energy must ultimately vanish , the rate of change gets smaller and smaller @xcite . in conclusion , the sg displays a non - trivial thermodynamic behavior which can be summarized as follows . at @xmath53
a long - range order is established ; for @xmath54 a short - range order is still present with large fluctuations on all length scales . at larger temperatures @xmath55
any trace of order has disappeared and a paramagnetic state is approached .
when studying transport properties , microcanonical dynamics are usually chosen as they allow to describe an isolated system , or its isolated bulk , without any assumption on the equilibrium state between the system and the surrounding . here
we adopt a recently introduced microcanonical dynamics @xcite , which features a high degree of flexibility , being ergodic in any temperature range and implementable on a generic structure , even in the presence of disorder . on regular lattices ( e.g. cylinder , torus ) such a dynamics
has already been shown to be able to lead the system to thermalized states compatible with those expected from a canonical dynamics and to allow the study of out - of - equilibrium properties @xcite .
although we will focus only on equilibrium regimes , the reason for choosing this dynamics is twofold : first we test its reliability on an inhomogeneous structure ; second , we pave the ground for the study of transport properties on such a substrate . in the following ,
we briefly resume how it works having in mind as substrate a generic graph @xmath8 described by the adjacency matrix @xmath13 .
for each pair of connected sites , namely each link @xmath56 , such that @xmath33 , besides the magnetic energy @xmath57 , we introduce a local _ kinetic energy _ @xmath58 which is , in principle , unbounded .
now , the dynamical rule proceeds as follows : 1 . start from a ( discrete ) distribution of energies @xmath59 ; 2 . choose randomly a link @xmath60 ; 3 .
extract one over the possible four spin - configurations for the couple of sites @xmath61 , and evaluate the magnetic energy variation @xmath62 induced by the move ; 4 . if @xmath63 , accept the move and increase the link energy @xmath64 of @xmath65 .
when @xmath66 , the move is accepted only if @xmath67 and the link energy is consequently decreased of @xmath65 ; otherwise the move is not accepted and the link energy is not updated .
it is worth remarking that @xmath65 allows for energy variations occurred on the link @xmath60 as well as on those pertaining to links adjacent to sites @xmath31 or @xmath32 : @xmath68 where @xmath69 if the @xmath31th spin has undergone a spin flip , otherwise it is zero .
it is therefore clear that , due to the discreteness of the system and to the fact that @xmath36 is constant over all links , both @xmath70 and @xmath64 are discrete variables : the former can only assume two different values corresponding to the aligned and non aligned configurations of the adjacent spins @xmath31 and @xmath32 ; the latter can only assume integer values deriving by proper combinations of the pertaining @xmath65 .
we also notice that @xmath64 works as an additional degree of freedom and the above dynamics conserves the total energy given by the following hamiltonian function @xmath71 as shown in @xcite , the magnetic and kinetic energies result to be non - correlated : this allows a natural definition of temperature at equilibrium , which , in a very natural way , depends only on the average kinetic energy .
in fact , the link energy satisfies the boltzmann distribution @xmath72 , and the fitted constant @xmath73 just corresponds to the expected inverse temperature of the system .
finally , we stress that the possible coupling with thermostats set at a temperature @xmath74 can be realized straightforwardly by selecting a subset of links @xmath75 ( or , analogously a subset of nodes ) and by extracting the pertaining kinetic energies according to the boltzmann distribution , being @xmath76 @xcite .
as foresaid , an interesting characterization of the ising model on the sg may be accomplished by a configurational analysis defined in the wider context of cellular automata on graphs .
such an analysis can be realized by a particular metrization referring non directly to the configuration space , whose points are the states @xmath77 of the system , but to a peculiar _ partition space _ containing - among other elements - the cluster distributions of the system .
the mathematical framework , which is summarized in the appendix @xmath78 , requires that the graph is endowed with the structure of a probability space . precisely , we consider the triple @xmath79 , where the measure @xmath80 on the subset algebra @xmath81 of the gasket @xmath0 is simply given by the normalized number of nodes in every subset .
a _ configuration _ ( or _ state _ ) on @xmath0 is a function assigning to each node a value in an alphabet @xmath1 .
the set of all possible configurations will be denoted as @xmath82 , the configuration space .
since we consider the ising model on @xmath83 , the alphabet is binary , but all we are going to say is independent of the number @xmath84 .
the adjacency matrix , combined with the list of values on nodes , allows an easy definition of clusters on an arbitrary structure ( the procedure is identical to the recognization of connected subset in graph colouring ) : two homogeneous nodes belong to the same cluster if they are connected through a path of nodes sharing the same value .
thus , every state @xmath85 determines in a natural way a _ partition _ of @xmath83 , i.e. an exhaustive collection @xmath86 of disjoint subsets @xmath87 , each connected and homogeneous , commonly called _ atoms _ of the partition .
the set of all partitions of @xmath0 constitutes the _ partition space _ @xmath88 .
the application @xmath89 is many - to - one , since , for instance , permutations in @xmath1 produce different states but the same partition .
obviously , @xmath90 contains much more partitions than those derived from clusterization , e.g. since , in general , atoms do not require to be connected sets , as clusters are . in the present case , @xmath91 is discrete and finite because @xmath0 is such , but the formalism applies in abstract probability spaces ( see @xcite ) .
clearly , when a dynamics is defined on the graph , this determines configuration orbits @xmath92 starting from any initial state @xmath93 , and the corresponding partition orbits @xmath94 , where @xmath95 basic operations between two arbitrary partitions @xmath96 and @xmath97 are the minimal common multiple @xmath98 , and the maximal common factor ( m.c.f . ) or _ intersection _ @xmath99 ( see fig .
[ fig : example ] and appendix @xmath78 for details ) .
the entropy @xmath100 of the intersection @xmath101 , when calculated between partitions at next steps along an orbit , is an index of the relevance of the ( instantaneous ) non evolving part .
the metrization of the partition spaces is based on the rohlin distance , which describes the non - similarity of two arbitrary partitions @xmath96 and @xmath97 .
this distance , requiring the shannon conditional entropy @xmath102 of measurable partitions ( see eq .
[ condiz ] ) , is given by the functional of eq .
[ rohlin ] , we anticipate here : @xmath103 in order to amplify non similarity , in the appendix @xmath104 we present also a method , referred to as `` reduction process '' and denoted @xmath105 , which acts on couples of partitions and uses both operations @xmath106 and @xmath107 . more precisely , given two partitions , say @xmath96 and @xmath97 , their reduction is obtained by first defining their intersection @xmath99 and by keeping from both partitions only those subfactors @xmath108 and @xmath109 , prime with @xmath101 , namely such that @xmath110 .
then , the reduced partitions are given by , @xmath111 and @xmath112 respectively ( see fig . [
fig : all ] , lower panel ) .
the process @xmath105 gives evidence of the essentially different sub - partitions of any couple in @xmath113 , and therefore amplifies their distance : @xmath114 .
hence , by comparing the distance between reduced and non - reduced couples , it is possible to introduce an _ amplification ratio _ , that is @xmath115)/d_r ( \alpha , \beta)$ ] , which provides further information about the cluster distribution and mobility .
however , all this analysis has to be performed in correlation with other observables . as explained in appendix @xmath116 ,
the reduction process is effective , namely gives rise to a large amplification ratio , whenever one of the two partitions , say @xmath96 , displays at least one cluster which in @xmath97 is exactly decomposed into smaller ones .
now , in the case under study @xmath96 and @xmath97 are partitions defined by cluster configurations at two successive steps , and the existence of a common cluster is furthered by the special topology of the sg : clusters corresponding to subgraphs which are ( combinations of ) gaskets of generation @xmath117 are rather stable ( the border is made by two vertices only ) , nonetheless internal fluctuations may occur and hence decompose the cluster itself .
it is worth underlining that while the @xmath98 operation is rather trivial , the @xmath99 operation and the reduction process @xmath105 are quite tricky .
reduction in particular is the main algorithmical obstacle in handling large graphs ( @xmath118 ) .
of course , the rohlin distance is deeply different from the well known hamming distance @xmath119 in the configuration space , i.e. @xmath120 where @xmath121 and @xmath122 are the values of the @xmath123th node , @xmath124 is a distance functional in the alphabet .
the simplest distance in @xmath125 is @xmath126 , leading to hamming distance 0.5 for purely random configurations .
looking for instance to fig .
[ fig : all ] ( upper panel ) , we would get the maximal hamming distance for the configurations @xmath127 and @xmath128 , and a minor distance for @xmath127 and @xmath129 , while the rohlin metrics on the corresponding partitions gives null distance in the former case and a high distance in the latter .
in other terms , @xmath130 and @xmath119 are deeply different in principle not only because they refer to different objects ( defined in @xmath131 and @xmath82 , respectively ) , but also because the former has do to with mutual distribution of clusters , which could involve geometrical features and long range correlations , while the latter is the sum of strictly local differences .
therefore , even if the correspondence @xmath132 is many - to - one , the loss of information should be compensated by the fact that by @xmath130 we get a global estimate on cluster distributions , instead of a bare sum of uncorrelated differences .
finally , more details concerning the implementation of the algorithm for measure of the above mentioned metric observables can be found in appendix @xmath133 .
we now focus on the ising model on the sg , with the microcanonical dynamics described in sec . [
sec : micro ] .
the equilibrium regime is ensured by coupling the system to thermostats set at the same temperature ; in this particular case the simplest way is to consider , for a given generation @xmath17 , the external triangle , i.e the @xmath134 links defining the perimeter of the gasket ; if then we exclude the six angular links , we get @xmath24 separated thermostats .
this could be useful in the future , as it consistutes a quite simple way of imposing temperature gradients to the system .
however , it should be noticed that , due to the fact that the contact between the thermostats and the system gets vanishingly small with respect to the bulk as @xmath17 is increased , the time for thermalization is expected to grow with the size of the system . anyway , as mentioned , we are now interested only in the equilibrium behavior . before proceeding , we underline that our checks strongly confirm that the thermalized states reached by the microcanonical dynamics are consistent with those expected from a canonical dynamics , e.g. based on the metropolis algorithm . in particular , we verified that , for a given temperature , macroscopic observables like the magnetization and the energy measured with the two kinds of dynamics are indistinguishable . for any given temperature @xmath52 ,
the geometric observables have been calculated as time series starting after a thermalization time , and lasting an observation time @xmath135 , where time is measured in units of monte carlo ( mc ) steps ; i.e. @xmath136 elementary moves .
in particular , we consider finite segments of trajectories @xmath137 in @xmath138 or @xmath139 in @xmath140 , for @xmath141 , as well as the related intersections @xmath142 and the couples of reduced partitions at successive time steps @xmath143 . from such equilibrium trajectories
we obtain segments of time series for the following quantities : 1 .
the entropy @xmath144 , applied either to the orbits @xmath145 or @xmath146 ; 2 .
the rohlin distance @xmath147 , ( see eq .
[ rohlin ] ) ; 3 . the amplified rohlin distance @xmath148)$ ] ; 4 . the amplification ratio @xmath149 ; 5 . the hamming distance @xmath150 , ( see eqs . [ eq : dh ] and [ hamming ] ) . here
we focus to the case of unitary time steps ( @xmath151 equals one mc step ) , leaving the study of the role of the time gap to future investigations . after setting the experimental parameters ( size , thermalization time , confidence length of trajectories @xmath135 ) , we calculate time averages , variances etc . for each of the series above .
results found for different choices of size ( we especially focused on gaskets of generation @xmath152 , @xmath43 and @xmath44 , corresponding to @xmath153 , @xmath154 and @xmath155 sites respectively ) and of @xmath135 are qualitatively in very good agreement . moreover , to approach the thermalized state , the initial configuration is taken ferromagnetic ; this minimizes the likelihood of pinning effects during the evolution @xcite . in the following we report only the essential information , dropping redundant numerical outputs . in general
, the observables analyzed highlight the existence of different regimes , demarcated by remarkable temperatures @xmath156 and @xmath157 .
more precisely , @xmath3 corresponds to the flex in the rohlin distance and to the peak in the amplification ratio , while @xmath158 corresponds to the peak in the variance of rohlin distance and to the crossover in the intersection entropy ( see figs .
@xmath159 ) .
interestingly , @xmath3 and @xmath158 recover the `` critical '' temperatures @xmath39 and @xmath160 evidenced by thermodynamic analysis ( see sec .
@xmath23 and @xcite ) . indeed ,
consistently with thermodynamic results , the highlighted regimes correspond to a long - range order region and to a disordered region with a critical - like transition region in between . more precisely :
[ fig : d_r ] as a function of the temperature ; three different sizes are compared.,title="fig:",scaledwidth=100.0% ] [ fig : d_h ] as a function of the temperature ; three different sizes are compared.,title="fig:",scaledwidth=100.0% ] * for @xmath161 , @xmath162 is close to @xmath14 ( see @xcite ) , while the distances @xmath119 and @xmath130 are approximately @xmath15 ( see figs .
@xmath23 and @xmath24 ) ; in fact , at such small temperatures a ferromagnetic order is established over large length - scales : clusters are constituted by few single spots in the large `` sea '' of equally oriented spins and spin - flips are rather unlikely to happen .
so , configurations - and partitions - at consecutive steps differ for such small spots that distances are extremely small
. moreover , during a mc step , the rare spin flips occurring are yet able to change the atoms in such a way that overlaps between consecutive partitions are quite improbable .
this inhibits the reduction , and the amplification is close to 1 . for @xmath52
approaching @xmath3 , such spots get larger , but , due to the slowness of the evolution at low temperature , their borders can remain sufficiently unchanged for several steps .
hence , the emergence of spots within such clusters allows the reduction , processing couples of consecutive partitions , to get more effective , and the time average of the amplification is manifestly increased .
the growth of the spots can be retrieved by the intersection entropy increase , as shown in the inset of fig .
@xmath25 ( left panel ) . * for @xmath163
, @xmath162 is close to @xmath164 and both @xmath119 and @xmath130 start to be significantly larger than @xmath15 .
this is a consequence of the fact that spin flips are getting more frequent .
more interestingly , the amplification ratio reaches a maximum .
we have seen that , at the middle of the previous regime , two conditions cooperated to start the growth of the amplification ratio : non empty intersection @xmath101 , and the fact that at consecutive steps there exist uniform large clusters which are decomposed internally , yielding an effective reduction ( as explained in appendix @xmath116 ) .
here , the peak in the amplification ratio means that cluster sizes and the speed of the dynamics optimally fit such conditions ; moreover , fixed clusters are more likely .
the exact determination of the peak temperature is tricky , due to the growing complexity of the configuration dynamics , but we argue that @xmath3 ( or @xmath39 ) is a good approximation . this behavior is consistent with the apparent transition occurring at @xmath39 as a long - range order breakdown . * for @xmath6
, @xmath162 is approaching zero .
clusters get more intricate due to the growing temperature , and @xmath100 , which measures the relevance of overlapping between successive configurations , exhibits a rapid growth ( see fig .
@xmath25 , left panel ) .
the coexistence of fragmentation of large clusters , which is a signature of decay for long range correlations , and overlapping may be related to `` critical slowing down '' effects @xcite .
clearly , fragmented clusters have a higher probability to overlap , but are unlikely to include small spots ; as a consequence the amplification ratio is still larger than @xmath14 , though rapidly decreasing .
note that @xmath100 is rather far from saturation ( namely , uncorrelated chaos ) and this suggests the persistence of a local short - length order . * for @xmath165 ,
the amplification ratio is practically 1 : such a breakdown of the reduction , as forementioned , has a completely different meaning than in the case of very low temperatures , where overlapping was insignificant for the smallness of sparse clusters in the `` big sea '' of dominant magnetic orientation .
here , due to the smallness of the @xmath101 atoms , it is unlikely that a sufficiently large cluster is internally decomposed at the next step ( see also the appendix @xmath116 ) .
+ at this temperature the variance of @xmath130 has a maximum ( whose value scales like @xmath166 ) , and this constitutes a signature that , along the trajectory , the fluctuation of the distance is particularly important . indeed , the complexity of clusters shape and their mobility can give rise to wide fluctuations of distances in time .
also , @xmath100 exhibits a crossover : from @xmath158 , the growth is due only to fragmentation , while the `` critical slowing down '' is over .
both the persistent growth of @xmath100 and the value of @xmath167 ( neatly below @xmath164 ) indicate that the complete chaos is far from being established at this temperature and for these sizes .
moreover , the large fluctuations make the system more susceptible to spin - flips , i.e. energy changes , and this is consistent with the peak in the specific heat . * for @xmath168 ,
the rohlin variance decreases roughly as @xmath169 toward an asymptotic value corresponding to the uncorrelated chaos , whose onset may be recognized by @xmath170 .
we argue that the fragmentation and the mobility of clusters stabilize the behavior of the time series for global quantities like distances .
+ [ fig : ampl ] ( left panel ) and amplification ratio @xmath171 ( right panel ) as a function of the temperature . for the latter ,
accurate numerical data are presented only for generation @xmath43 , due to time consuming calculations ; nonetheless we have checked that for other values of @xmath17 the qualitative behavior is robust .
the inset of the left panel shows a magnification of the departure from zero of the entropy ; notice the semilogarithmic scale.,title="fig:",scaledwidth=100.0% ] + we note however that the maximum for the @xmath119-variance occurs at @xmath172 , well above the corresponding maximum of the @xmath130-variance at @xmath173 .
we deduce that in this temperature interval big clusters of all sizes , typical of the previous fractal configurations , have started their fragmentation , so that the global shape diversity between subsequent configurations is stabilizing and this explains why the @xmath130-variance decreases . however , the fragmentation process is still slow enough that the local matching between different or equal spins at next steps is highly unstable in time , allowing for the growth of the @xmath119-variance .
indeed , spins belonging to the inner part of a cluster ( even if of small size ) result from the dynamics to be more stable than those belonging to its periphery . for @xmath174
the fragmentation is such that both local overlaps and medium length correlations tend to stabilize the time behavior .
there is therefore a temperature interval exhibiting a subtle interplay between correlation length and time stability . in other terms , in the way to the chaos we recognize two time scales : one in terms of global similitude ( shape ) of clusters along the orbits , which starts to stabilize just at the end of the `` critical slowing down '' ( @xmath173 ) ; the other in terms of local overlaps , whose stabilization requires a higher temperature ( @xmath172 ) . the direct visual inspection could be awkward for such properties , undetectable also from the mean magnetization .
in this work we have performed a quantitative characterization of cluster dynamics for the ising model defined on a finite sierspinski gasket , whose thermodynamic behavior is known to be non - trivial .
the analysis is based on a set of geometric observables , such as hamming and rohlin distances , and on reduction operations among partitions which allow to detect the evolving and the stable components of clusters .
the phenomenology evidenced by previous thermodynamic analysis is qualitatively and quantitatively confirmed by the present metric approach , which provides in addition a geometric characterization of the anomalous behavior of the system .
indeed , we highlight first the existence of two `` critical '' temperatures : @xmath3 corresponding to a peak in the amplification ratio , meaning that there an abrupt change in the clusters behavior occurs , and @xmath158 , corresponding to a peak in the rohlin variance and other crossovers , evidencing the loss of short range order .
more precisely , the amplification ratio gives detailed information about the development of clusters , starting from little independent spots ( amplification ratio equal to 1 ) to first overlaps ( amplification ratio larger than 1 and growing to the maximum ) while the next phase of decreasing ratio indicates the complex effect of decreasing efficiency of the reduction due to the fragmentation , up to @xmath158 when the fragmentation is such that , notewithstanding the large intersection , the reduction process is inhibited by the extreme improbability of clusters with a decomposable inner part ( see [ ampli ] ) .
another new information we obtain in the high temperature regime is the existence of different scales in the destruction of local and correlated order , as evidenced by the hamming and rohlin variances , with an anticipated peak for the latter quantity .
thus , the analysis above enabled us to distinguish different behaviors and phases demarcated by @xmath3 and @xmath158 as main milestones ; we can not speak of course of `` phase transitions '' , since the onset of these distinct behaviors seems to be a rather smooth process .
however , from such a detailed description of the collective motion we get an insight on the interplay between dynamic and statistical features , especially the decay of spatial and temporal correlations .
this work opens a lot of possible extensions and further insights .
the next step is the metric characterization of clusters dynamics in condition of non - equilibrium ; a kind of analysis allowed by the chosen dynamics even in the presence of a disordered coupling pattern @xcite .
in particular , it could be interesting to deepen the effect of the double scale in the approach to uncorrelated chaos on the conductivity . in the same context ,
another interesting item to explore is the relevance of topology for the cluster diffusion .
indeed , the whole set of operations performed on the sg , from cluster identification to reduction , immediately applies to any automaton on arbitrary , connected graphs : the process depends only on the adjacency matrix , on a generic alphabet and on a dynamics , working as a proper external engine generating a succession of configurations .
therefore , by the bare substitution of the adjacency matrix , the algorithm is ready to fit a great variety of statistical models ( e.g. pott s model on lattices of arbitrary dimension or graphs ) , or other network problems where a dynamics is defined . if necessary , nodes could also be weighted , defining alternative probability measures .
true algorithmic problems and non trivial extensions could only arise from alternative definitions of partitions ( assuming e.g. atoms of a different kind with respect to the clusters ) or from a different factorization , modifying the reduction process .
the formalism and general results for partition spaces and rohlin metrics may be recovered e.g. in @xcite .
let @xmath175 be a probability space , that is an arbitrary set @xmath176 , a @xmath177-algebra @xmath178 of subsets of @xmath176 , and a normalized measure @xmath179 on @xmath178 . in our case
the set @xmath176 is just given by @xmath180 .
a partition of @xmath176 is a finite collection @xmath181 of measurable disjoint subsets covering @xmath176 , i.e. @xmath182 and @xmath183 .
the @xmath184 s are called the `` atoms '' of @xmath185 .
the set of all finite measurable partitions is denoted @xmath186 .
the unit partition @xmath187 consists of the single atom @xmath176 .
a partial order in @xmath188 , i.e. a relation @xmath189 , means that @xmath97 is a refinement of @xmath96 ; equivalently , every @xmath190 is exactly composed with some @xmath191 included in @xmath97 . in such case , @xmath96 is said to be a `` factor '' of @xmath97 . clearly , @xmath192 for every @xmath96 .
such terms as `` unit '' and `` factor '' depend on a commutative and associative pseudo - product , or composition , @xmath193 , denoting the less refined of all partitions greater or equal to both @xmath96 and @xmath97 , whose atoms are the non empty intersections of the @xmath96 and @xmath97 atoms . if not ambiguous , we can also write @xmath194 . obviously , @xmath195 whenever @xmath196 , and in particular @xmath197 for every @xmath96 .
such properties make the result of this operation a kind of `` minimal common multiple '' .
conversely , @xmath198 is the greatest partition such that @xmath199 and @xmath200 . in this case ,
@xmath201 for every @xmath202 , and @xmath203 implies that @xmath202 and @xmath204 are `` relatively prime '' ( i.e. they have no common factor ) .
therefore , the result is a sort of `` maximal common factor '' .
see fig .
[ fig : example ] for an example of product and intersection among partitions .
product ( upper panel ) and intersection ( lower panel ) between two examples of partition realized on a square lattice.,title="fig : " ] product ( upper panel ) and intersection ( lower panel ) between two examples of partition realized on a square lattice.,title="fig : " ] a partition may represent a probabilistic experiment with non overlapping outcomes @xmath205 , where the `` atomic '' event @xmath190 has probability @xmath206 .
a factor is therefore a sub - experiment of the finer experiment , grouping several outcomes as equivalent : for instance , `` odd or even '' is a two - atoms sub - experiment of the @xmath207 dice experiment .
the shannon s entropy @xmath208 defined on every partition as @xmath209 is a measure of the mean information obtained from the experiment .
if @xmath210 is another partition , the conditional entropy of @xmath96 with respect to @xmath97 is @xmath211 where , as usual , one takes @xmath212 for @xmath213 .
this conditional entropy is the mean residual information obtained from @xmath185 when the result of @xmath214 is known .
note that the shannon entropy depends only on the distribution of the atom measures , not on their nature or `` shape '' ( this term coud have no meaning in abstract spaces ) . on the contrary
, the mutual relations among atoms ( and possibly their shapes ) directly influence the conditional entropy ( see figure a.2 , upper row ) . now , the rohlin distance @xmath130 is defined in @xmath215 by @xmath216 and it may be considered as a measure of the overall non - similarity between @xmath217 and @xmath218 .
if @xmath219 is finite , a _ configuration _ or _ state _
@xmath220 on @xmath219 is a function assigning to each point @xmath221 a value @xmath222 in an alphabet @xmath223 .
all possible configurations form a space @xmath224 . in @xmath138
the hamming distance @xmath119 is defined by @xmath225 where @xmath226 is a distance in @xmath223 and @xmath227 a possible normalization coefficient .
to each configuration corresponds an exhaustive collection of clusters , i.e. connected subsets of @xmath228 with homogeneous value in @xmath229 , defining a particular partition in @xmath230 .
this establishes a many - to - one correspondence @xmath231 , making possible the comparison between @xmath232 in @xmath138 and @xmath233 in @xmath140 , where @xmath234 and @xmath235 .
upper row : the three configurations depicted evidence the difference between hamming and rohlin distances : @xmath119 is maximal for configurations @xmath127 and @xmath236 , and minor for @xmath127 and @xmath237 ; @xmath130 is null in the former case and large in the latter . lower row : for the partitions @xmath96 and @xmath97 the common factor is given by @xmath101 , while @xmath238 and @xmath239 are the reduced partitions respectively ] 30.0 pt upper row : @xmath96 and @xmath97 are two partitions of the square , each of four atoms ; @xmath101 is their intersection or m.c.f . of their atoms .
lower row : list of relevant elementary dichotomic factors , where the black specifies the atom and the white the complementary set . ]
the partitions @xmath240 and @xmath241 are the reduced partitions of those appearing in figure [ fig : redaa ] : atoms are now three , individuated by black , white and grey .
the partition @xmath242 is the reduced intersection .
notice that the grey atom is non connected .
] 10.0 pt the essential non - similarity between two partitions could be confused and weakened by the presence of a tight common factor , i.e. a common sub - partition ( see fig . [
fig : all ] , lower panel ) .
therefore , we would eliminate common factors as far as possible , a `` reduction '' which is expected to amplify the distance .
however , this operation ( analogous to the reduction to minimal terms for fractions @xcite ) is not uniquely definite because partitions , differently from integers , do not admit a unique factorization into primes .
the role of primes ( i.e. indecomposable ) factors can be played by dichotomic sub - partitions , which are still extremely redundant ( @xmath243 indeed for a partition with @xmath244 atoms ) .
for @xmath181 we shall define therefore a restricted family @xmath245 of `` elementary '' dichotomic factors @xmath246 such that 1 .
@xmath245 must be well defined for every @xmath247 ; 2 .
@xmath245 does not contain more than @xmath244 ( the number of atoms in @xmath217 ) elementary factors ; 3 .
@xmath248 .
a universal choice consists in taking as dichotomic factors @xmath249 , the partitions formed by single atoms and their complements to @xmath250 .
elementary factors of this form , used throughout in the present paper , will be called `` simple '' . for a couple @xmath217 and @xmath214 ,
once their elementary factors @xmath245 and @xmath251 have been defined , the reduction process consists in the following steps : 1 .
define the maximal common factor @xmath198 ; 2 .
drop from @xmath245 and @xmath251 those factors which are not relatively prime with @xmath177 , and note the surviving factors @xmath252 and @xmath253 respectively ( this means @xmath254 ) ; 3 .
define @xmath255 and @xmath256 .
the reason of step @xmath257 , which seems to be cumbersome with respect to the simple dropping of common factors in @xmath245 and @xmath251 , is that in general two families could have no common factors and , nevertheless , @xmath258 .
this happens , for instance , when @xmath259 with no common elementary factors . then @xmath260 and @xmath261 with the reduction above , while @xmath262 with the dropping of common factors .
it results that @xmath263 , as requested @xcite .
note however that while two relatively prime partitions are already reduced , not necessarily two reduced partitions are relatively prime : see for instance the case of figure a.3 , which could represent small portions of wider partitions deduced from almost chaotic configurations . in the next subsection
we give more precise details on this .
the correspondence @xmath264 is many - to - one and idempotent , i.e. @xmath265 .
it is a sort of projection from @xmath266 on the subset of irreducible pairs .
the reduction process , therefore , essentially depends on the choice of the family @xmath245 of elementary factors .
besides simple factors , other families exist , which in particular cases could prove more convenient for algorithmic reasons or for the observer s attitude in the probabilistic experiment .
while simple dichotomic factors correspond to looking at the `` occurrence or not '' of single atomic outcomes , other attitudes could isolate dichotomic outcomes enjoying supplementary properties which are not universal but depend , typically , on some additional geometrical structure of @xmath250 ( order , connection , orientation etc . ) .
for instance , in previous works on rectangular lattices @xcite elementary factors were identified by external contours of clusters , a choice intended to optimize the simple connection of the factor atoms .
this is not convenient on general graphs , where @xmath267 , because the determination of external surfaces could be cumbersome or impossible .
therefore , in the present work , elementary factors will always be the simple factors @xmath268 .
10.0 pt let @xmath96 and @xmath97 such that @xmath269 , and @xmath270 .
every @xmath271 is the union of some subsets @xmath272 and @xmath273 of atoms in @xmath96 and @xmath97 .
possibly , such subsets may be of one single atom or more : we indicate @xmath274 and @xmath275 the single or multiple atoms cases for @xmath96 , and analogously @xmath276 and @xmath277 for @xmath97 . clearly , at fixed @xmath278 , @xmath279 and @xmath280 , but it is useful to keep a distinct notation in order to remember that such atoms are not only in the partitions but also in their m.c.f .
therefore , @xmath281 may be composed in four forms : @xmath282 , @xmath283 , @xmath284 and @xmath285 .
for instance , in figure [ fig : all ] , the m.c.f .
@xmath101 has one @xmath286 , one @xmath287 and three @xmath288 atoms , while , in figure [ fig : redaa ] , @xmath101 has one @xmath289 and two @xmath288 atoms .
* proposition 1 : * the atoms of @xmath238 are all those of @xmath96 contained in the @xmath290 groups , plus one atom constituted by the intersection of their complementary sets , or equivalently by @xmath291 ( and similarly for @xmath239 ) .
the proof immediately follows from the fact that the @xmath292 elementary factors are dropped in the reduction process if and only if @xmath293 .
this is true in any abstract partition space , whenever simple factors are used ; notice however that , speaking of partitions generated by _ connected _ configurations , as those considered in the present paper , this last atom could be non - connected ( an example in the figure [ fig : redbb ] ) .
* proposition 2 : * the atoms of @xmath294 are : _
i ) _ all the @xmath271 of the @xmath295 form ; _ ii ) _ one more atom ( the complementary part to the union of the previous ones ) if at least one term @xmath282 exists ; or _
iii ) _ two more atoms if , besides the @xmath295 , only mixed terms @xmath283 and @xmath284 exist .
the point _
i ) _ follows from proposition 1 , since the @xmath296 and @xmath297 groups reconstitute common subpartitions in @xmath238 and @xmath239 ; point _
iii ) _ depends on the fact that , in absence of @xmath282 terms , there is an exact correspondence between @xmath298 and @xmath299 ( and equivalently @xmath300 and @xmath301 ) , where the @xmath278 index runs over the mixed - type @xmath281 s only ; the two supplemetary atoms are therefore @xmath302 and @xmath300 ; as to point _
ii ) _ , we observe that the complementary set of the @xmath303 , i.e. @xmath304 , can not be exactly decomposed by the atoms of @xmath305 , and vice versa , just because the presence of the supplementary simple terms .
several easy corollaries follow .
for instance , when simple factors are used , propositions 1 and 2 constitute a direct constructive proof that the reduction @xmath105 is a projection . or that there is no reduction at all when in @xmath101 there are only @xmath285 terms , or when @xmath101 has only two atoms . or else that in order to have @xmath306 , no terms @xmath285 are allowed and at least one term @xmath282 is necessary
moreover , if for one partition , say @xmath96 , no @xmath307 group exists , then @xmath308 and @xmath309 .
an important item is the relation between reduction ( meant as pattern simplification , as stated in proposition 1 ) and the amplification ratio @xmath310 , measuring the metric effectiveness of the reduction .
* proposition 3 : * the necessary and sufficient conditions in order to have @xmath311 , and therefore amplification ratio @xmath312 , are the existence in @xmath101 of at least one atom of mixed form , @xmath283 or @xmath284 , and the existence of at least two distinct group @xmath313 from the same partition ( @xmath314 or @xmath97 ) .
this may be proved using the important equality @xmath315 ( see @xcite ) . in the computation of all entropies appearing in eq .
[ dr ] , one can split the shannon s sums ( see eq . [ shannon ] ) along the atoms @xmath316 of @xmath101 , because the intersections @xmath317 appearing in @xmath318 are surely empty for different @xmath281 s .
therefore , using propositions 1 , for atoms of the @xmath285 form , the partial contributes to entropies in eq .
[ dr ] are the same in @xmath319 and @xmath320 . in absence of mixed terms , the remaining part of both @xmath96 and @xmath97 is constituted by the same atoms @xmath321 , and the contribute to the distance is 0 , exactly as for the supplementary single atom @xmath322 common to @xmath323 and @xmath324 .
assume now that also mixed terms @xmath283 and @xmath284 are present ( and possibly also terms @xmath282 ) in such a way that at least two @xmath325 or two @xmath326 appear .
using again eq .
[ dr ] for these components , @xmath327 and @xmath328 are equal , because the intersection of the multiple atoms in one partition give the same result with the single atoms in the other partition ( before reduction ) or with their union ( after reduction ) . as to the subtracted quantities in eq .
[ dr ] , the contributes from @xmath208 and @xmath329 are strictly greater than @xmath330 and @xmath331 because there are contributes from separate atoms in the former case , and from their union in the latter ( this is why at least two @xmath325 or two @xmath326 are requested ) .
clearly , elementary convexity properties of the @xmath332 function used here , see e.g. @xcite .
therefore @xmath333 is strictly greater than @xmath334 , and after the subtraction in eq .
[ dr ] the distance is increased , i.e. the amplification ratio is greater than one .
proposition 3 clarifies that the metric effectiveness of the reduction , besides the pattern simplification , is due to the difference between @xmath335 and @xmath336 , where @xmath337 is @xmath96 or @xmath97 . in particular ,
if @xmath96 and @xmath97 are partitions defined by cluster configurations at two successive steps as those considered in the present paper , then the amplification requires that there exist big clusters at one time exactly decomposed at the next ( or previous ) time into smaller ones .
10.0 pt the @xmath338 adjacency matrix @xmath13 of the graph @xmath0 and its state @xmath339 , i.e. the @xmath340-valued list of sites labelled @xmath341 , constitute the essential information necessary to work out the partition algebra and the metrization algorithm , which is independent of the way the states are generated ( dynamics ) . by standard `` colouring '' techniques , connected sites
are iteratively recognized testing their value in @xmath340 , and the partition is easily produced as a list of grouped labels , one group for every atom .
for a ten sites graph , for instance , a particular partition could be @xmath342 , with measures @xmath343 for @xmath344 respectively .
a useful representation for the atoms is a binary string of length @xmath136 : e.g. , in the example above , @xmath345 , etc .
note that from now on the topological nature of the graph does not influence the operations , which regards only the label lists .
the `` simple factors '' of the family @xmath346 are immediately defined . in the example , @xmath347 , etc .
( the corresponding binary strings are obviously complementary ) .
clearly , the extreme simplicity of this procedure could be replicated with other choices of dichotomic factors , e.g. by taking the `` internal - close '' and the `` external - open '' parts of jordan contours ( provided that such contours are well defined ) . only in the very special case of one - dimensional chains
the latter choice ( internal - external , i.e. left - right ) proves not only more effective but easier than the former , because contours are left extremities of semi - open intervals .
the @xmath348 operation may be easily implemented by boolean intersections on the atom lists .
this is enough for the rohlin distance computation by formula [ shannon ] , thanks to the equality of eq .
[ dr ] above .
more attention is required for the @xmath107 operation , which is the key step to define the maximal common factor @xmath99 , but the task may equally accomplished by boolean operators : every atom of @xmath101 is built site by site , testing the simultaneous belonging to some atoms of @xmath96 and @xmath97 , up to exhaustion .
once the maximal common factor is defined , most of the computation time is spent in the storage and management of the surviving simple factors @xmath349 and @xmath350 , along the criteria of the subsection [ ampli ] .
however , all this does not imply conceptually new operations .
reduced partitions , and their atoms , which in principle are defined by intersections on the atoms of such surviving factors , can take advantage of proposition 1 in subsection [ ampli ] , completing the @xmath105 process .
for the amplified distance @xmath351 , the functional in eq .
[ dr ] is applied to the couple @xmath352 , and the observations of the previous subsection reveal useful again .
this work is partly supported by the firb grant : @xmath353
y. gefen , a. aharony , b.b .
mandelbrot and s. kirkpatrick , 1981 _ phys .
lett . _ * 47 * 1771 .
y. gefen , a. aharony and b.b .
mandelbrot , 1983 _ j. phys .
a _ * 16 * 1267 .
y. gefen , a. aharony , y. shapir and b.b .
mandelbrot , 1984 j. phys .
a. * 17 * 435 . m. knezevic , j. joksimovic and d. knezevic , 2005 physica a. * 367 * 207 .
r. campari and d. cassi , 2010 phys . rev .
e * 81 * 021108 .
a. vezzani , 2003 j. phys .
a * 36 * 1593 .
r. burioni , d. cassi and l. donetti , j. phys .
a * 32 * , 5017 ( 1999 ) .
liu , phys .
b * 32 * 5804 ( 1985 ) j.h .
luscombe and r. c. desai , phys .
b * 32 * , 1614 ( 1985 ) .
r. peierls , 1936 _ proc .
cambridge phys .
soc _ * 32 * , 477 .
v.i.arnold and a. avez , _ problmes ergodiques de la mcanique classique _ , gauthier - villars , paris , 1967 . n. f. g. martin and j , w. england , _ mathematical theory of entropy _ , addison - wesley , reading ma 1981 . | we develop and implement an algorithm for the quantitative characterization of cluster dynamics occurring on cellular automata defined on an arbitrary structure . as a prototype for such systems we focus on the ising model on a finite sierpsinski gasket , which is known to possess a complex thermodynamic behavior .
our algorithm requires the projection of evolving configurations into an appropriate partition space , where an information - based metrics ( rohlin distance ) can be naturally defined and worked out in order to detect the changing and the stable components of clusters .
the analysis highlights the existence of different temperature regimes according to the size and the rate of change of clusters .
such regimes are , in turn , related to the correlation length and the emerging `` critical '' fluctuations , in agreement with previous thermodynamic analysis , hence providing a non - trivial geometric description of the peculiar critical - like behavior exhibited by the system . moreover , at high temperatures , we highlight the existence of different time scales controlling the evolution towards chaos .
+ _ keywords _ : dynamical processes ( theory ) , classical phase transitions ( theory ) , classical monte carlo simulations |
recently , great theoretical and experimental efforts are taken to studying the fusion of heavy nuclei leading to several new phenomena including the understanding of the formation of neutron -rich and super heavy elements @xcite .
the precise knowledge of the interaction potential between two nuclei is a difficult task and continuing efforts are needed in this direction .
this problem has been of very active research over the last three decades and remains one of the most widely studied subject in low - energy heavy - ion physics @xcite .
the total interaction potential is sum of the long range coulomb repulsive force and short range nuclear attractive force .
the coulomb part of the interaction potential is well - known , whereas nuclear part is not clearly understood .
a large number of efforts have been made to giving simple and accurate forms of the nuclear interaction potentials @xcite . among such efforts ,
proximity potential is well known for its simplicity and numerous applications .
based upon the proximity force theorem @xcite , a simple formula for ion - ion interaction potential as a function of the separation between the surfaces of two approaching nuclei was presented @xcite . as pointed out by many authors @xcite , original form of the proximity potential 1977 overestimates the experimental data by 4% for fusion barrier heights . in a recent study involving the comparison of 16 proximity potentials , one of us and collaborators pointed out that proximity potential 1977 overestimates the experimental data by 6.7% for symmetric colliding nuclei @xcite .
similar results were obtained for asymmetric colliding nuclei @xcite . with the passage of time , several improvement/ modifications were made over the original proximity potential 1977 to remove the gray part of the potential .
it includes either the better form of the surface energy coefficient @xcite or the universal function and/or nuclear radius @xcite .
a careful look reveals that these modifications / improvements are not able to explain the experimental data @xcite .
a deep survey also pointed out that these technical parameters ( i.e. surface energy coefficient , nuclear radius , and universal function ) were chosen quite arbitrarily in the literature . among them , the surface energy coefficient is available in a large variety of forms from time to time @xcite .
it affects the fusion barrier heights and cross sections significantly @xcite . also , nuclear radius is available in large variety of forms @xcite .
these forms varies either in terms of its coefficients or either different mass or isospin dependence .
the third technical parameter i.e , the universal function , is also parametrized in different forms @xcite .
unfortunately , no systematic study is available in the literature , where one can explore the role of these technical parameters in fusion barrier positions , heights , and cross sections .
alternatively , a best set of the above - mentioned parameters is still missing . in the present study ,
our aim is to modify the original proximity potential 1977 by using a suitable set of the above - stated technical parameters available in the literature .
in addition , to compare the final outcome with the huge amount of experimental data available since last three decades .
the choice of the potential and its form to be adopted is one of the most challenging task when one wants to compare the experimental data with theory .
the present systematic study includes the reactions with combine mass between a = 19 and a = 294 units . in total ,
390 experimentally studied reactions with symmetric as well as asymmetric colliding partners are taken into consideration .
section [ model ] describes the model in brief , section [ result ] depicts the results and summary is presented in section [ summary ] .
the total ion - ion interaction potential @xmath0 between two colliding nuclei with charges @xmath1 and @xmath2 , center separation @xmath3 , and density distribution assumed spherical , and frozen , is approximated as @xcite @xmath4 where e is the charge unit .
the above form of the coulomb potential is suitable when two approaching nuclei are well separated .
the nuclear part of the potential @xmath5 is calculated in the framework of the proximity potential 1977 @xcite as @xmath6 where @xmath7 is the reduced radius . here
@xmath8 denotes the matter radius and is calculated using relation @xcite @xmath9 where @xmath10 denotes the half - density radii of the charge distribution and @xmath11 is the neutron skin of the nucleus . to calculate @xmath10
, we used the relation given in ref .
@xcite as @xmath12 here , @xmath13 is the nuclear charge radius read as @xmath14 where @xmath15 and @xmath16 refer to neutron and proton contents of target / projectile nuclei .
this form of radius is taken from the recent work of royer and rousseau @xcite and is obtained by analyzing as many as 2027 masses with n , z @xmath17 8 and a mass uncertainty @xmath18 150 kev . the neutron skin @xmath11 used in eq .
( [ eq:4 ] ) is calculated according to ref .
@xcite .
the surface energy coefficient @xmath19 was taken from the work of myers and wiatecki @xcite and has the form @xmath20 , \label{eq:8}\ ] ] where n and z refer to the total neutrons and protons content .
it is clear from eqs .
( [ eq:6 ] ) and ( [ eq:8 ] ) that both nuclear radius as well as surface energy coefficient depend on the relative neutron excess . in the above formula , @xmath21 is the surface energy constant and @xmath22 is the surface - asymmetry constant .
both constants were first parameterized by myers and wiatecki @xcite by fitting the experimental binding energies .
the first set of these constants yielded values @xmath23 and @xmath24 and 1.79 , respectively . in original proximity version , @xmath23 and @xmath22
were taken to be @xmath25 and 1.7826 @xcite , respectively .
later on , these values were revised in a large variety of forms depending upon the advancement in the theory as well in experiments @xcite . in total , 14 such coefficients are highlighted in ref .
@xcite and the role of extreme 4 sets is analyzed deeply .
out of them , two best sets of surface energy coefficients are stressed . in the present study
, we shall restrict to the latest set of @xmath19 values i.e. @xmath23 = @xmath26 and @xmath22 = @xmath27 presented in ref @xcite .
this particular set of values were obtained directly from a least - squares adjustment to the ground - state masses of 1654 nuclei ranging from @xmath28o to @xmath29106 and fission - barrier heights @xcite
. the universal function @xmath30 used in eq .
( [ eq:1 ] ) has been derived by several authors in different forms @xcite . in original proximity potential ,
@xmath30 was parametrized in the cubic - exponential form @xcite @xmath31 with @xmath32 = @xmath33)/@xmath34 .
the surface width @xmath34 ( i.e. @xmath35 has been evaluated close to unity . we labeled this universal function as @xmath36-@xmath37 . later on , blocki et al .
, @xcite modified the above form as @xmath38 in the present study , we use this form of universal function and marked it as @xmath36-@xmath39 . by using the above stated parameters
, we construct a new proximity potential and labeled as prox 2010 .
along with the above modified form , we shall also use the original proximity potential 1977 @xcite and its recently modified form proximity potential 2000 @xcite .
we labeled them as prox 1977 and prox 2000 , respectively .
by using the above new version of the proximity potential ( prox 2010 ) along with its older versions ( i.e. prox 1977 and prox 2000 ) , fusion barriers are calculated for 390 reactions by using the conditions : @xmath40 the height of the barrier and position is marked , respectively , as @xmath41 and @xmath42 .
as one see from the preceding section , three factors govern the success of proximity potential are ( i ) the surface energy coefficient , ( ii ) the universal function , and ( iii ) nuclear radius .
we analyzed the literature very carefully and found that the latest information on these three factors can shape the new proximity potential .
recently , the role of surface energy coefficient stated above is studied in detail in ref @xcite .
as for as radius is concern , we shall restrict to its latest form given in ref .
however , the role of the third parameter i.e. , the universal function in fusion barriers is analyzed in fig . 1 . here , we display @xmath43 and @xmath44 defined as @xmath45 and @xmath46 as a function of @xmath47 using two sets of above mentioned universal functions [ eqs .
( [ eq:9 ] ) and ( [ eq:10 ] ) ] .
it is clear from the figure that deviations are significantly reduced by using @xmath36-@xmath39 compared to its original form @xmath36-@xmath37 .
the universal function @xmath36-@xmath39 reduces the average deviation over 390 reactions by 1 % for fusion barriers .
the experimental values are taken directly from the literature @xcite .
actually , it is clear from the literature that no experiment can extract information about the fusion barriers directly . all experiments measure the fusion differential cross sections and then with the help of a theoretical model , one can extract the fusion barriers . in fig .
2 , we display the theoretical fusion barrier heights @xmath48 ( mev ) and positions @xmath49 ( fm ) verses the corresponding experimental values .
we note from the figure that prox 2010 potential reproduces the experimental fusion barrier heights within 1.4% .
this result is in close agreement with other recently parametrized potentials presented in ref .
@xcite . however , the original form of the proximity potential 1977 presented in ref .
@xcite overestimates the data by 6.7% for symmetric colliding nuclei .
however , the fusion barrier positions show some scattering from the central line ( marked by shaded area ) . this scattering may be due to the variation in the experimental setups and theoretical method one used to extract these values @xcite .
we quantify our outcome in figs . 3 and 4 .
3 , the percentage deviations between the theoretical and experimental values are presented . the original proximity potential 1977 ( prox 1977 ) along with its recently modified form ( prox 2000 ) are also displayed .
we note from the upper panel of fig .
3 that prox 2010 potential on average gives better results compared to its older versions for fusion barrier heights .
however , slight deviations are visible for fusion barrier positions
. this may be due to the fact that in the proximity potential prox 2010 , we use the value of surface energy coefficient that gives stronger attraction compared to one used in prox 1977 and prox 2000 potentials .
therefore , in prox 2010 potential , the counterbalance between the repulsive coulomb and attractive nuclear part of the interaction potential occurs at larger distances , and hence it pushes the barrier outwards .
the fusion barrier heights are reproduced within @xmath50 on average . on the other hand ,
fusion barrier positions reproduced the experimental values within @xmath51 . especially for the heavier colliding nuclei
, we see that prox 2010 potential reproduces the data much better on the average compared to other versions . for lighter nuclei
, however , small scattering is visible .
this could also be due to the uncertainty in the radius of the lighter colliding nuclei . in figs . 1 - 4
, only 155 reactions are displayed to maintain the clarity .
the average deviation for the fusion barrier heights over 390 reactions is 0.77 % using our modified potential prox 2010 , whereas prox 1977 , and prox 2000 give 3.99 % , and 4.45 % , respectively .
this shows that our modified proximity explains the experimental data nicely . in fig .
4 , we display the difference between the theoretical and experimentally extracted fusion barriers .
we further note that prox 2010 potential gives better results .
the difference especially for the heavy systems is significantly reduced .
this was the problem with original as well as its recently modified form as pointed out by several authors @xcite .
it is clear from figs . 3 and 4 , that prox 2010 potential is able to reproduce the experimental data much better than its older versions .
the small difference is not significant because of the uncertainties in the analysis of the experimental data .
finally , we test our newly modified proximity potential prox 2010 on fusion probabilities . in fig . 5
, we display the fusion cross sections @xmath52 ( in mb ) as a function of the center - of - mass energy @xmath53 ( mev ) for the reactions of @xmath54 @xcite , @xmath55 @xcite , @xmath56 @xcite , @xmath57 @xcite , @xmath58 @xcite , and @xmath59 @xcite .
the fusion cross sections are calculated using well known wong model @xcite .
the older versions of proximity potentials that is , prox 1977 and prox 2000 are also displayed .
it is clearly visible from the figure that prox 2010 potential is in good agreement , whereas , its older forms are far from the experimental data .
we further note that prox 1977 and prox 2000 potentials show similar results .
it means that no improvements is seen in prox 2000 potential as was claimed in ref .
in the present study , we present a best set of the surface energy coefficient , the nuclear radius , and the universal function available in the literature .
we find that these parameters which were used quite arbitrarily in past years affect the fusion barrier heights , positions , and cross sections significantly . by using the above set of parameters ,
a new proximity potential is constructed .
our newly constructed proximity potential prox 2010 reproduces the fusion barriers and cross sections better than its earlier versions .
0 i. dutt and r. k. puri phys . rev .
c * 81 , * 044615 ( 2010 ) ; * 81 , * 064609 ( 2010 ) .
i. dutt and r. k. puri phys .
c * 81 , * 047601 ( 2010);*81 , * 064608 ( 2010 ) .
r. k. puri _
et al . _ ,
j. a * 23 , * 429 ( 2005 ) ; r. arora _
et al . _ ,
* 8 , * 103 ( 2000 ) ; r. k. puri _ et al .
* 3 , * 277 ( 1998 ) ; r. k. puri _ et al .
c * 45 , * 1837 ( 1992 ) ; ibid .
* 43 , * 315 ( 1991 ) .
j. blocki , j. randrup , w. j. wiatecki , and c. f. tsang , ann . phys .
( n.y . ) * 105 , * 427 ( 1977 ) .
j. blocki and w. j. wiatecki , ann .
* 132 , * 53 ( 1981 ) .
w. reisdorf , j. phys .
g : nucl . part
. phys . * 20 * , 1297 ( 1994 ) .
w. d. myers and w. j. wiatecki , phys .
c * 62 , * 044610 ( 2000 ) ; and earlier references therein .
k. siwek - wilczyska and j. wilczyski , phys .
c * 69 , * 024611 ( 2004 ) .
a. dobrowolski , k. pomorski , and j. bartel , nucl .
a * 729 , * 713 ( 2003 ) ; v. i. zagrebaev , nucl .
phys . a * 734 , * 164 ( 2004 ) ; n. wang _ et al .
c * 78 , * 054607 ( 2008 ) . v. y. denisov , phys .
* b526 , * 315 ( 2002 ) .
g. royer and r. rousseau , eur .
j. a * 42 , * 541 ( 2009 ) .
w. d. myers and w. j. wiatecki , nucl .
* 81 , * 1 ( 1966 ) .
w. d. myers and w. j. wiatecki , ark .
* 36 , * 343 ( 1967 ) .
p. mller , j. r. nix , w. d. myers , and w. j. wiatecki , at .
data nucl .
data tables * 59 , * 185 ( 1995 ) .
h. a. aljuwair _
et al . _ ,
c * 30 , * 1223 ( 1984 ) .
m. trotta _
et al . _ , phys .
c * 65 , * 011601(r ) ( 2001 ) . a. morsad , j .
j. kolata , r. j. tighe , x. j. kong , e. f. aguilera , and j. j. vega , phys .
c * 41 * , 988 1990 r. l. neto , j. c. acquadro , p. r. s. gomes , a. s. d. toledo , c. f. tenreiro , e. crema , n. c. filho , and m. m. coimbra , nucl .
* a512 * , 333 ( 1990 ) .
a. m. stefanini , _ et al .
_ , phys . lett .
* b 679 * , 95 ( 2009 ) . j. o. newton , c. r. morton , m. dasgupta , j. r. leigh , j. c. mein , d. j. hinde , h. timmers , and k. hagino , phys .
c * 64 * , 064608 ( 2001 ) .
b. sikora , j. bisplinghoff , w. scobel , m. beckerman , and m. blann , phys .
c * 20 , * 2219 ( 1979 ) .
j. r. leigh _ et al .
c * 52 * , 3151 ( 1995 ) . c. y. wong , phys .
* b42 , * 186 ( 1972 ) ; c. y. wong , phys
* 31 , * 766 ( 1973 ) . | by using a suitable set of the surface energy coefficient , nuclear radius , and universal function , the original proximity potential 1977 is modified .
the overestimate of the data by 4% reported in the literature is significantly reduced .
our modified proximity potential reproduces the experimental data nicely compared to its older versions . |
leptogenesis is perhaps the most attractive model for generating the matter - antimatter asymmetry in our universe @xcite after inflation .
the process generates a lepton asymmetry via the production and subsequent decay of a heavy majorana neutrino .
this lepton asymmetry is partially converted to a baryon asymmetry via sphaleron processes @xcite which break both @xmath0 and @xmath1 , but conserve @xmath2 .
the fraction of @xmath2 that ends up in @xmath0 by the sphaleron processes is given by @xmath3 giving @xmath4 where the @xmath5 is created by leptogenesis which gives a non zero value of @xmath1 ( see for instance @xcite ) .
in its simplest form ( within the context of the see - saw model ) leptogenesis consists of adding three heavy right - handed neutrinos to the standard model . in the hierarchical limit one of these right handed neutrinos , @xmath6 ,
is much lighter than the other two and the leptogenesis mechanism consists essentially of the process @xmath7 where @xmath8 is a light , left - handed neutrino and @xmath9 the higgs .
all light flavours behave identically , so you can think of the other light flavours as included as a factor of @xmath10 in the decay rate . because of loop corrections there can be @xmath11 violation in the decay , usually quantified by the asymmetry parameter @xmath12 . in the following
we denote the heavy neutrino by @xmath13 , the light by @xmath1 , and the higgs by @xmath14 .
this model has been studied extensively in the literature , including deviations from the hierarchical limit , thermal corrections etc @xcite .
however , all studies have used the integrated boltzmann equation to follow the evolution of the heavy neutrino number density and the lepton asymmetry .
this approach assumes maxwell - boltzmann statistics for all particles as well as kinetic equilibrium for the heavy species .
this assumption is normally justified in freeze - out calculations where elastic scattering is assumed to be much faster than inelastic reactions . however , in the present context , kinetic equilibrium in the heavy species would have to be maintained by the decays and inverse decays alone .
therefore it is not obvious that the integrated boltzmann equation is always a good approximation .
furthermore @xmath15 processes are generally inefficient for thermalization compared with @xmath16 processes . for @xmath17 deviations from kinetic equilibrium are always of order 20% or less @xcite , but for @xmath15 processes they can be very large ( see for instance refs .
@xcite for a case where deviation from equilibrium is extremely large ) . in this paper
we investigate how the use of the full boltzmann equation affects the final lepton asymmetry in a simplified model with only decays and inverse decays and resonant scattering .
we will return to the point of resonant scattering in due course .
we find that when @xmath18 the difference can be very large .
however , at small temperature where the inverse decay dominates the difference decreases in magnitude to about 20% .
here we study only the two - body decay of a heavy right - handed neutrino to a light neutrino plus a higgs .
we do not include thermal corrections to the particle masses , so that for instance the process @xmath19 is not kinematically possible .
we assume that the asymmetry , represented by @xmath20 is small , so that when we calculate anything with @xmath13 , we can assume identical distributions of @xmath1 and @xmath21 .
we consider only initial zero abundance of @xmath13 .
we use only single particle distribution functions , in which case the boltzmann equation for the heavy species can be written as @xcite @xmath22,\end{aligned}\ ] ] and for the light neutrino it is @xmath23,\end{aligned}\ ] ] with a similar equation for @xmath21 .
the interesting boltzmann equations for the present purpose are those for @xmath13 and for @xmath2 .
@xmath14 and @xmath24 have gauge interactions which are very fast .
this means that @xmath14 can be described by a distribution in chemical equilibrium , and that @xmath24 can be described as distributions in kinetic equilibrium , @xmath25 with @xmath26 .
using cpt - invariance , following the idea of @xcite , we find @xmath27 , \nonumber \label{eq : r}\end{aligned}\ ] ] where @xmath28 is the total rest frame decay rate . the corresponding equation for @xmath29 to @xmath0 conversion during the decay process because we track only @xmath29 .
including it would have only a modest effect on our numerical results .
] is @xmath30 , \label{eq : l}\end{aligned}\ ] ] to first order in @xmath12 and with @xmath31 since we have so far only included @xmath32 processes , eq .
[ eq : l ] suffers from the well - known problem of lepton asymmetry generation even in equilibrium @xcite . to remedy this problem the resonant part of the @xmath33
must be included .
to lowest order in @xmath12 this amounts to adding the term @xmath34 in eq .
[ eq : l ] @xcite , so that the final form of the equation for the lepton asymmetry is @xmath35 , \nonumber \label{eq : l2}\end{aligned}\ ] ] this equation does not exhibit any lepton asymmetry generating behaviour in thermal equilibrium because @xmath36 = 0 $ ] explicitly .
to first order in @xmath37 equations eq .
[ eq : r ] and [ eq : l2 ] can be easily integrated numerically for given values of @xmath38 and @xmath39 .
if maxwell - boltzmann statistics and kinetic equilibrium for @xmath13 are assumed the equations can be further simplified .
the integrated boltzmann equations are then @xmath40 where @xmath41 and @xmath42 is the thermally averaged total decay rate .
these equations are the ones normally used in leptogenesis calculations for the simplest case of one massive and one light neutrino .
however , compared with eqs .
( [ eq : r]-[eq : l2 ] ) they involve the approximation of assuming maxwell - boltzmann statistics and kinetic equilibrium . particularly for the case
where @xmath43 this approximation is not necessarily good .
when decays and inverse decays are included ( as well as gauge interactions of @xmath1 and @xmath14 which establish kinetic equilibrium for @xmath24 , and @xmath14 individually , but which conserve @xmath29 ) there are essentially only two interesting parameters .
all terms that contribute to the development of densities are proportional to either @xmath14 or @xmath39 .
this competition can be parameterised by the single parameter @xmath44 , the decay parameter .
the other important parameter is the net decay asymmetry @xmath12 . in the present study
we assume that the heavy neutrinos start with zero abundance , i.e. that they are not equilibrated at high temperatures through other interactions .
we refer the reader to @xcite for a thorough discussion of the various possibilities for initial @xmath13 abundance . in figs
. 1 - 3 we show a calculation of @xmath45 and the net asymmetry for different values of @xmath46 . at high temperatures
the equilibration rate of @xmath13 is much higher when the full boltzmann equation is used .
the main reason for this can be seen in fig .
4 . when @xmath47 the low momentum states of @xmath13 are populated very efficiently , while there is little population of high momentum modes . this in turn means that the inverse decay @xmath48 is suppressed relative to the decay process @xmath49 . to quantify this , we want to define a momentum dependant inverse decay rate , @xmath50 the efficiency with which @xmath13 states with momentum @xmath51 are produced from the background ( ignoring decays and pauli blocking ) is then given by @xmath52\nonumber.\end{aligned}\ ] ] this function is sharply peaked at low momentum with the following limiting behaviour @xmath53 this is a steeply increasing function of @xmath54 and at low temperatures when @xmath55 the distribution of @xmath13 is in complete equilibrium and @xmath56 so that the boltzmann approximation is valid . therefore the curves for @xmath57 approach each other in figs . 1 - 3 .
note , however , that the highest @xmath51 modes are always out of equilibrium because of the asymptotic @xmath58 term .
in fact for given values of @xmath54 and @xmath46 there will be a limiting value of @xmath51 above which the distribution will be out of equilibrium . in fig .
5 we show the value of @xmath59 for which @xmath60 as a function of @xmath54 , for the specific case of @xmath61 . at high temperatures no momentum states are in equilibrium , whereas at low temperatures only progressively higher momentum states are out of equilibrium .
this figure corresponds well to what is shown in fig .
4 . for @xmath62 no states are in equilibrium , for @xmath63 states above @xmath64 are not yet in equilibrium , and for @xmath65 almost all states
have been populated to their equilibrium value . as can be seen in figs .
1 - 3 the net asymmetry also grows much more rapidly initially when the full boltzmann equation is used , and the change of sign occurs at higher temperature .
however , at low temperatures the differences are smaller . in fig .
6 we show the ratio of the final asymmetry in the two cases @xmath66 as can be seen from fig . 6 there can be a significant difference in the asymmetry between the two approaches . for very high values of @xmath46
the difference is small because the distribution is kept very close to kinetic equilibrium ( for a fermi - dirac distribution ) throughout the decay .
note , however , that the difference never goes to zero .
the reason is that the thermal decay width is changed relative to the maxwell - boltzmann approximation when pauli blocking and stimulated emission factors are included ( as also noted by @xcite ) . for smaller values of @xmath46
the difference increases and can be as large as 50% .
however , it should be noted that our framework will break down for small values of @xmath46 because we have not included scattering .
another important point to note is that @xmath67 does not depend on @xmath12 because the deviation from kinetic equilibrium is governed completely by the total decay rate ( i.e. by @xmath46 ) , not by the asymmetry , as long as @xmath68 .
we are only interested in the final value of the asymmetry since , in order to maximise the baryon asymmetry for any given value of @xmath12 , we want leptogenesis to end before the conversion through sphalerons end .
we have solved the full boltzmann equation for decays and inverse decays to follow the generation of lepton asymmetry during leptogenesis .
when decays are semi - relativistic the difference between using the full boltzmann equation and the standard integrated boltzmann equation can be very large .
however , at low temperature where washout dominates the difference is relatively modest .
the difference at @xmath69 is less than about 30% between the two approaches . for smaller @xmath46
the difference can be larger . however , this is the regime where @xmath15 processes do not dominate , i.e. outside the regime in which our framework is valid .
the conclusion is that the boltzmann approximation yields results in the strong washout regime which are accurate to at least 30% . for @xmath70
the difference is not larger than 15% .
sh acknowledges support from the alexander von humboldt foundation through a friedrich wilhelm bessel award .
we thank pasquale di bari and alessandro strumia for valuable comments .
99 m. fukugita and t. yanagida , `` baryogenesis without grand unification , '' phys . lett .
b * 174 * , 45 ( 1986 ) .
v. kuzmin , v.a .
rubakov and m.e .
shaposhnikov , `` on anomalous electroweak baryon - number non - conservation in the early uiniverse , '' phys . lett . *
155b * , 36 - 42 ( 1985 ) .
w. buchmuller , p. di bari and m. plumacher , `` leptogenesis for pedestrians , '' annals phys . * 315 * , 305 ( 2005 ) [ arxiv : hep - ph/0401240 ] .
m. a. luty , `` baryogenesis via leptogenesis , '' phys .
d * 45 * , 455 ( 1992 ) .
t. hambye , y. lin , a. notari , m. papucci and a. strumia , `` constraints on neutrino masses from leptogenesis models , '' nucl .
b * 695 * , 169 ( 2004 ) [ arxiv : hep - ph/0312203 ] .
w. buchmuller , p. di bari and m. plumacher , `` the neutrino mass window for baryogenesis , '' nucl .
b * 665 * , 445 ( 2003 ) [ arxiv : hep - ph/0302092 ] .
w. buchmuller and m. plumacher , `` neutrino masses and the baryon asymmetry , '' int .
j. mod .
phys . a * 15 * , 5047 ( 2000 ) [ arxiv : hep - ph/0007176 ]
. g. f. giudice , a. notari , m. raidal , a. riotto and a. strumia , `` towards a complete theory of thermal leptogenesis in the sm and mssm , '' nucl .
b * 685 * , 89 ( 2004 ) [ arxiv : hep - ph/0310123 ] .
g. f. giudice , m. peloso , a. riotto and i. tkachev , `` production of massive fermions at preheating and leptogenesis , '' jhep * 9908 * , 014 ( 1999 ) [ arxiv : hep - ph/9905242 ] . m. hirsch , m. malinsky , j. c. romao , u. sarkar and j. w. f. valle , `` thermal leptogenesis in extended supersymmetric seesaw , '' arxiv : hep - ph/0608006 .
s. blanchet and p. di bari , `` leptogenesis beyond the limit of hierarchical heavy neutrino masses , '' jcap * 0606 * , 023 ( 2006 ) [ arxiv : hep - ph/0603107 ] .
s. blanchet and p. di bari , `` flavor effects on leptogenesis predictions , '' arxiv : hep - ph/0607330 .
a. abada , s. davidson , a. ibarra , f. x. josse - michaux , m. losada and a. riotto , `` flavour matters in leptogenesis , '' arxiv : hep - ph/0605281 .
t. ota and w. rodejohann , `` breaking of l(mu ) - l(tau ) flavor symmetry , lepton flavor violation and leptogenesis , '' phys .
b * 639 * , 322 ( 2006 ) [ arxiv : hep - ph/0605231 ] .
e. nardi , y. nir , e. roulet and j. racker , `` the importance of flavor in leptogenesis , '' jhep * 0601 * , 164 ( 2006 ) [ arxiv : hep - ph/0601084 ] .
a. abada , s. davidson , f. x. josse - michaux , m. losada and a. riotto , `` flavour issues in leptogenesis , '' jcap * 0604 * , 004 ( 2006 ) [ arxiv : hep - ph/0601083 ] .
t. hambye , m. raidal and a. strumia , `` efficiency and maximal cp - asymmetry of scalar triplet leptogenesis , '' phys .
b * 632 * , 667 ( 2006 ) [ arxiv : hep - ph/0510008 ] .
s. hannestad , `` non - equilibrium effects on particle freeze - out in the early universe , '' new astron . * 4 * , 207 ( 1999 ) [ arxiv : astro - ph/9903034 ]
. j. madsen , bose condensates , big bang nucleosynthesis , and cosmological decay of a phys .
lett . * 69 * , 571 ( 1992 ) .
g. d. starkman , n. kaiser and r. a. malaney , `` mixed dark matter from neutrino lasing , '' astrophys . j. * 434 * , 12 ( 1994 ) [ arxiv : astro - ph/9312020 ] .
n. kaiser , r. a. malaney and g. d. starkman , `` neutrino lasing in the early universe , '' phys .
lett . *
71 * , 1128 ( 1993 ) [ arxiv : hep - ph/9302261 ] .
s. hannestad and j. madsen , `` a cosmological three level neutrino laser , '' phys .
d * 55 * , 4571 ( 1997 ) [ arxiv : astro - ph/9702125 ] .
e. w. kolb and s. wolfram , `` baryon number generation in the early universe , '' nucl .
b * 172 * , 224 ( 1980 ) [ erratum - ibid .
b * 195 * , 542 ( 1982 ) ] . | we have studied the two - body production and decay of a heavy , right - handed neutrino to two light states using the full boltzmann equation instead of the usual integrated boltzmann equation which assumes kinetic equilibrium of all species .
decays and inverse decays are inefficient for thermalising the distribution function of the heavy neutrino and in some parameter ranges there can be very large deviations from kinetic equilibrium .
this leads to substantial numerical differences between the two approaches .
furthermore we study the impact of this difference on the lepton asymmetry production during leptogenesis and find that in the strong washout regime the final asymmetry is changed by 15 - 30% when the full boltzmann equation is used . |
the standard scenario of low - mass star formation envisions quasi - static condensation of dense cores out of magnetically subcritical clouds and inside - out core collapse leading to star formation ( shu , adams & lizano 1987 ; mouschovias & ciolek 1999 ) .
an alternative is the turbulence - controlled star formation , with magnetic fields playing a minor role if any ( larson 1981 ; mac low & klessen 2004 ) . in this picture , dense cores arise from compression in the supersonic turbulence observed in molecular clouds ( padoan et al . 2001 ; gammie et al . 2003 ; li et al . 2003 ) .
the cores so formed are transient entities , however . even though some of them may have column density distributions resembling those of static bonnor - ebert spheres , their internal motions are typically dynamic , with transonic or even supersonic speeds ( ballesteros - paredes , klessen , & vzquez - semadeni 2003 ) which are difficult to reconcile with the subsonic internal motions inferred in low - mass starless cores ( lee , myers & tafalla 1999 ) .
a potentially more serious difficulty with this picture is the rate and efficiency of star formation .
numerical simulations have shown that supersonic turbulence decays in one free - fall time or less , with or without a strong magnetization ( mac low et al .
1998 ; stone , ostriker & gammie 1998 ; padoan & nordlund 1999 ) . without additional support , self - gravitating clouds would collapse in one free - fall time , leading to a rate of star formation well above that inferred on the galactic scale ( evans 1999 ) . also , stars in typical low - mass star forming clouds contain less than a few percent of the cloud mass .
such a low star formation efficiency is not naturally explained .
excessive star formation is avoided in the standard scenario by an ordered magnetic field , which is postulated to provide most of the cloud support .
the gradual weakening of magnetic support through ambipolar diffusion leads to core formation .
the cores so formed tend to have subsonic infall speeds consistent with observations ( li 1999 ; ciolek & basu 2000 ) .
however , calculations based on this scenario have so far avoided a direct treatment of turbulence by starting from relatively quiescent regions of moderately high density ( several times @xmath0 @xmath1 or higher ) , as opposed to the average cloud density ( a few times @xmath2 @xmath1 or lower ; blitz 1993 ) .
how the over - dense regions form out of the more turbulent background in the first place was not addressed .
it is unlikely for them to have condensed out quasi - statically through ambipolar diffusion , because the level of ionization in the background is enhanced over the value given by cosmic ray ionization alone ( mckee 1989 ) , as a result of photoionization of interstellar far - ultraviolet radiation field .
the enhancement makes the time scale for quiescent ambipolar diffusion at the average density longer than the cloud lifetime ( myers & khersonsky 1995 ) . in this letter , we show that the supersonic turbulence observed in molecular clouds can speed up ambipolar diffusion in localized regions through shock compression without turning most of the cloud mass into stars in one crossing time .
we consider strongly magnetized , sheet - like clouds , taking advantage of the tendency for cloud material to settle along field lines into a flattened configuration .
the basic formulation we adopt here is the same as that of nakamura & li ( 2002 , 2003 ) and li & nakamura ( 2002 ) , which was originally developed for treating ambipolar diffusion - driven fragmentation of quiescent magnetic clouds in the presence of small perturbations ( see also indebetouw & zweibel 2000 and basu & ciolek 2004 ) .
briefly , we assume that the fluid motions are confined to the plane of the sheet , with force balance maintained in the vertical direction at all times ( fiedler & mouschovias 1993 )
. the cloud evolution is governed by a set of vertically integrated mhd equations that include ambipolar diffusion .
we consider the case where the magnetic field is coupled to the bulk neutral cloud material through ion - neutral collision , with ion density proportional to the square root of neutral density .
a canonical value of @xmath3@xmath4g@xmath5 is adopted for the proportionality constant ( elmegreen 1979 ) .
we solve the hydrodynamic part of the governing equations using roe s tvd method , and determine the gravitational and magnetic potentials outside the sheet using a convolution method based on fft .
the computation timestep is chosen small enough to satisfy both the cfl condition and the stability condition for treating ambipolar diffusion explicitly .
we consider an initially uniform mass distribution in the @xmath6-@xmath7 plane , with column density @xmath8 g @xmath9 ( where @xmath10 is visual extinction for standard grain properties ) .
the cloud is assumed isothermal , with sound speed @xmath11 cm s@xmath12 ( where @xmath13 is the temperature in units of 10 k ) . above a critical column density @xmath14 , the equation of state
is stiffened to enable simulation beyond the formation of the first `` collapsed '' object .
the jeans length @xmath15 pc and gravitational collapse time @xmath16 yr provide two basic scales for our problem .
we adopt a computation box of size @xmath17 ( corresponding to a cloud mass @xmath2 times the jeans mass @xmath18 @xmath19 ) , and impose periodic conditions at the boundaries .
the cloud is threaded vertically by a uniform magnetic field , characterized by the flux - to - mass ratio @xmath20 in units of the critical value @xmath21 . only strongly magnetized clouds with @xmath20 close to or exceeding unity
are considered , in keeping with the adopted thin - sheet approximation . to mimic the turbulent motions observed in molecular clouds
, we stir the cloud with a supersonic velocity field at @xmath22 . following the usual practice ( e.g. , stone et al .
2000 ) , we prescribe the turbulent velocity field in fourier space , with a power spectrum @xmath23 .
other choices yield qualitatively similar results , as long as the power is dominated by large - scale motions .
numerically , we adopt a @xmath24 grid , and consider only discrete values @xmath25 ( where @xmath26 is the box size and @xmath27 , 2 , ... , n-1 , with @xmath28 ) for wavenumber components , @xmath29 and @xmath30 . for each pair of @xmath29 and @xmath30 , we randomly select an amplitude from a gaussian distribution consistent with the power spectrum for the total wavenumber @xmath31 and a phase between 0 and @xmath32 . the resulting field is then transformed into the physical space .
the spatial distributions of @xmath33 and @xmath34 are generated independently , and the final velocity field is scaled such that the rms mach number of the flow @xmath35 has a prescribed value .
the velocity fields so specified tend to be strongly compressive . adopting an incompressible turbulent velocity field does not change the result qualitatively .
two representative clouds will be considered in detail , one magnetically subcritical with @xmath36 and the other supercritical with @xmath37 .
we choose a mach number @xmath38 , which yields a turbulence crossing time @xmath39 .
the results are displayed in figs . 1 and 2 .
1 presents the snapshots of the clouds at various times , and fig .
2 shows , as a function of time , the mass fraction of the regions that have become more than 10 times denser ( in column density ) than the average _ and _ are magnetically supercritical .
these so - called `` dense supercritical '' regions contain both `` collapsed '' objects whose contraction has been arrested by the stiffening of the equation of state and isothermal core material well on its way to star formation .
we will use their mass fraction as a measure of the efficiency of star formation , bearing in mind however that probably only a fraction ( say 1/3 to 1/2 ) of the dense supercritical material would be turned into stars in the presence of powerful protostellar outflows ( matzner & mckee 2000 ; shu , li & allen 2004 ) .
we begin with the subcritical cloud .
its snapshots are displayed in the first five panels of fig . 1 .
panel ( a ) shows the cloud after about one turbulence crossing time ( at @xmath40 , in units of @xmath41 here and below ) , when a large fraction of the cloud mass has been compressed into a network of filaments .
the filaments would collapse quickly were it not for the strong magnetic fields trapped in them .
by @xmath42 ( panel b ) , they are well on their way to re - expansion , leaving behind little dense supercritical material ( which has a mass faction of order @xmath43 according to fig . 2 ) .
although most of the shocked regions do not collapse right away , they are permanently altered .
their flux - to - mass ratios are changed by ambipolar diffusion , whose rate is enhanced by the increase in both density ( which weakens the field - matter coupling ) and gradient of the field strength .
the magnetically altered regions can not return to their pre - compression , uniformly magnetized state after shock passage .
rather , a magnetically differentiated structure develops , with regions of relatively low flux - to - mass ratio embedded within a more strongly magnetized background .
the supercritical `` islands '' are marked by white contours in each panel .
these regions have a head - start over their subcritical surroundings in forming stars , through further turbulent compression and/or ambipolar diffusion .
( we have tested our standard simulation of @xmath36 in the limit of zero ambipolar diffusion , and found that the @xmath44 ratio remains constant to within @xmath45 . ) after the initial phase of strong compression and re - expansion , the cloud settles down to a more relaxed configuration .
an example of this new phase of cloud evolution is shown in panel ( c ) .
prominent at this time ( @xmath46 or @xmath47 myrs , adopting here and below the fiducial combination @xmath48 and @xmath49 , appropriate for taurus clouds according to arce & goodman 1999 ) are several supercritical filaments , the densest parts of which have already begun dynamic contraction and formed `` collapsed '' objects .
note that the dense star - forming regions are well separated , reminiscent of the dispersed mode of star formation observed in taurus - like clouds .
once created by large - scale compression at well separated locations , the dense regions stay apart because their mutual gravitational attraction is cancelled to a large extent by the magnetic repulsion between them .
this mode is illustrated further in panel ( d ) , where @xmath50 ( or @xmath51 myrs ) and @xmath52 of the cloud mass now resides in the dense supercritical regions .
note that the dense `` ridge '' at the lower - left corner has a much lower velocity than its surrounding medium .
this is also true for other high density regions in general , broadly consistent with the observation that low - mass star forming cores are more quiescent than their surroundings .
the mass fraction of dense supercritical regions increases steadily , as a result of continued ambipolar diffusion .
the steady increase is punctuated by bumps at late times ( see fig . 2 ) .
these bumps are produced by interactions of dense regions , which are evident in panel ( e ) of fig . 1 , where @xmath53 ( or 18 myrs ) .
the interactions become more important as more cloud material becomes supercritical , since the gravitational attraction between supercritical regions is not completely offset by magnetic repulsion .
the unbalanced `` effective '' gravity tends to drive the supercritical regions together , which enhances the possibility for further interactions .
the clustering of `` heavier '' supercritical regions against a more strongly magnetized background may , given enough time , drive the cloud from the initial dispersed mode of star formation to a clustered mode of star formation .
in the regions where the dense supercritical material preferentially gathers , the local efficiency of star formation can be much higher than the average .
the magnetically supercritical cloud evolves differently .
its snapshots are displayed in the last three panels of fig . 1 .
panel ( f ) shows the cloud at @xmath54 , when a network of filaments has formed , as in the subcritical case , although the filaments here are thinner , because of a weaker magnetic resistance to shock compression .
these filaments are sufficiently supercritical that , when enough mass is accumulated , they break up gravitationally into a string of dense cores along their length , unlike their more magnetized counterparts . by the time shown in panel ( g ) ( @xmath42 ) ,
some of the dense cores have already merged together , creating several dense supercritical regions .
these regions contain about half of the cloud mass ( see fig . 2 ) .
they dominate the subsequent cloud evolution through mutual gravitational interactions . by the time shown in panel ( h ) ( @xmath55 ) ,
some of the dense regions have merged together , and new blobs have formed through interactions .
most of the dense regions are clustered together gravitationally .
the clustering is retarded only moderately by magnetic repulsion , since the cloud is magnetically supercritical as a whole , and the dense regions are even more so .
nearly @xmath56 of the cloud mass does become subcritical , however , as a result of magnetic flux diffusing from the high - density to low - density regions .
it may explain why the mass fraction of the dense supercritical regions hovers around half , rather than a value closer to unity as one might naively expect for supercritical clouds .
we have explored the effects of the same ( strong ) turbulence on clouds with other degrees of magnetization , including one that is critically magnetized everywhere initially ( @xmath57 ) .
dense supercritical regions are created promptly in this cloud , with a mass fraction intermediate between that of the subcritical ( @xmath36 ) and supercritical ( @xmath58 ) cloud , as shown in fig . 2 .
also shown in fig .
2 is a more subcritical cloud with @xmath59 .
it does not produce a dense supercritical object until @xmath60 ( or @xmath61 myrs ) .
such a long dormant time would argue against star - forming clouds being subcritical by a large factor everywhere , although a stronger turbulence can in principle induce localized cloud collapse and star formation sooner . in practice , a distribution of the flux - to - mass ratio @xmath20 is expected in realistic clouds .
the least magnetized regions are expected to collapse first , either on their own ( if they are supercritical and self - gravitating ) or through turbulent compression .
in such a case , the star formation rate and efficiency would obviously depend on the @xmath20 distribution , which is unknown at present .
we have demonstrated that star formation with a relatively low efficiency can occur over an extended period of time in highly turbulent , somewhat magnetically subcritical clouds . the low efficiency is made possible by the strong magnetic field , which prevents the global cloud collapse in one turbulence crossing time .
the supersonic turbulence , on the other hand , speeds up ambipolar diffusion in localized regions through shock compression .
the acceleration of ambipolar diffusion by turbulence has been examined analytically by kim & diamond ( 2002 ) , zweibel ( 2002 ) and fatuzzo & adams ( 2002 ) , and numerically by heitsch et al .
our investigation differs from these studies in adopting a highly compressive turbulent flow and in that the cloud dynamics are computed rather than prescribed . to further illustrate the role of turbulence in star formation , we have rerun the subcritical ( @xmath36 ) case , varying the turbulent mach number @xmath35 from 0.3 to 10 .
the result is plotted in fig .
3 . for @xmath62
, it takes an order of magnitude longer than the gravitational collapse time to produce the first dense magnetically supercritical object . as @xmath35 increases ,
dense supercritical regions appear earlier . beyond some critical value ( @xmath63 for our particular examples ) ,
such regions form promptly , within one collapse time .
the ability for a subcritical cloud to form stars quickly in the presence of a supersonic turbulence removes a major objection to the standard scenario for low - mass star formation through ambipolar diffusion based on cloud lifetime arguments ( hartmann , ballesteros - paredes , & bergin 2001 ) .
our calculations support the idea of bimodal star formation ( shu et al .
1987 ) . in magnetically subcritical clouds ,
dense supercritical regions are created by large - scale turbulence at well separated locations .
they stay separated until a large enough fraction of the cloud mass becomes supercritical for the gravitational attraction to overwhelm the magnetic repulsion .
we identify this mode of star formation with the mode of relatively inefficient , dispersed star formation exemplified by the taurus clouds .
if the cloud lives long enough , a large fraction of the cloud mass will become supercritical sooner or later , through continued ambipolar diffusion . in regions where the supercritical materials collect gravitationally
, the star formation efficiency could be much higher than the general background .
such regions could be the sites of efficient cluster formation .
efficient star formation occurs much more quickly if the cloud is initially magnetically supercritical , as in our second example .
our simulations suggest that a realistic turbulence with power dominated by large - scale motions does not fundamentally change the bimodal nature of star formation in magnetic clouds .
3d calculations are needed to firm up this conclusion .
we thank c. gammie and e. ostriker for helpful discussion .
support for this work was provided by grant - in - aid for scientific research ( no .
15740117 ) by the ministry of education , culture , science and technology of japan and nsf grant ast-0307368 . | we investigate numerically the combined effects of supersonic turbulence , strong magnetic fields and ambipolar diffusion on cloud evolution leading to star formation .
we find that , in clouds that are initially magnetically subcritical , supersonic turbulence can speed up star formation , through enhanced ambipolar diffusion in shocks .
the speedup overcomes a major objection to the standard scenario of low - mass star formation involving ambipolar diffusion , since the diffusion time scale at the average density of a molecular cloud is typically longer than the cloud life time . at the same time
, the strong magnetic field can prevent the large - scale supersonic turbulence from converting most of the cloud mass into stars in one ( short ) turbulence crossing time , and thus alleviate the high efficiency problem associated with the turbulence - controlled picture for low - mass star formation .
we propose that relatively rapid but inefficient star formation results from supersonic collisions of somewhat subcritical gas in strongly magnetized , turbulent clouds .
the salient features of this shock - accelerated , ambipolar diffusion - regulated scenario are demonstrated with numerical experiments . |
after an early history marked by vigorous debate ( see , e.g. , @xcite ) , the existence of gravitational radiation has become accepted as a hallmark prediction of einstein s general theory of relativity . despite indirect evidence supporting this prediction , however , gravitational waves have yet to be observed directly . indeed , rough ,
order - of - magnitude calculations indicate a typical passing wave will produce only tiny material strains of order @xmath0 . a new generation of laser - interferometric gravitational wave observatories
have recently been commissioned , and several are already operational , which aim to observe such strains @xcite and thereby detect gravitational waves incident on earth .
these experiments are necessarily extremely delicate and , as a result , both initial detection and the long - term goal of extracting new information about the distant sources of a particular gravitational wave must be aided by detailed theoretical predictions .
a community of theorists have turned to numerical simulations of the full , non - linear einstein equations to identify the characteristic features of the gravitational radiation generated by various sources .
an impressive array of techniques have developed within this young field of numerical relativity which aim to provide the ongoing experimental effort with accurate predictions . however , the program faces a number of challenges , from foundational questions to issues of implementation and interpretation of results . here
, we consider one issue of interpretation .
we ask how , exactly , the well - known theoretical description of gravitational radiation outside quiescent black holes due to teukolsky @xcite might be applied to the results of numerical simulations .
the notion of gravitational radiation in general relativity does not have universal meaning .
rather , a proper definition can be given only in regions of space - time , _ radiation zones _
, whose geometry is characterized by two distinct length scales @xcite : one describing an `` average '' radius of curvature and a second , much shorter scale corresponding to the wavelength of gravitational waves .
because these two scales can be distinguished only when the waves contribute small corrections to the `` average '' curvature , many analyses of gravitational radiation are founded on perturbation theory .
examples include not only the standard analysis in linearized gravity , but the regge
wheeler @xcite and zerilli @xcite approaches , as well as teukolsky s .
even the asymptotic formulation of the gravitational radiation @xcite , which applies to quite general space - times , operates by studying differences between the physical metric and a fixed asymptotic metric near conformal infinity . in numerical relativity , however , all of these analyses are difficult to implement .
the perturbation approaches are complicated because no background metric on space - time is known _ a priori _ , while the asymptotic approach needs access to field values far outside the computational domain .
this paper focusses on teukolsky s perturbative approach since it describes radiation fields near a rotating black hole , the expected end - state of many physical processes producing relatively strong gravitational wave signals .
we first identify the essential elements of the teukolsky formalism which derive from its fixed background geometry .
in particular , we are interested in the preferred tetrad of basis vectors on space - time which underlies the definition of teukolsky s fields .
although this tetrad arises as a perturbation of a canonical , kinnersley tetrad @xcite for the background kerr geometry , we show one can approximate it using only the _ physical _ metric , eliminating any reference to the background . the construction of this approximate kinnersley tetrad occurs naturally in two stages .
the first , which is completed in this paper , fixes the directions of the desired tetrad vectors .
because the final results of our analysis make no mention of a background geometry , they may be applied unambiguously to a broad class of space - times .
we give concrete criteria characterizing this class , and find it includes many examples which differ _ non - perturbatively _ from the kerr geometry . in particular , when a numerical simulation produces a single black hole settling down to equilibrium , this first stage should be viable even at relatively early times after the hole forms .
the problem of the second stage , fixing the scalings of the tetrad vectors along these preferred directions , is described in some detail here but not solved .
we plan to present a solution in a future paper .
however , even the first stage alone provides significant gauge - invariant information partially characterizing the gravitational field encoded in the numerical variables .
an earlier letter @xcite presented one scheme to extract such information , and this paper can be regarded as an elaboration and expansion of that presentation . like the teukolsky formalism itself , several of this paper s results
are most simply expressed in the newman
penrose approach @xcite to space - time geometry .
however , because many existing numerical codes do not currently implement this approach , we strive to present our final results in a form which does not refer explicitly to newman
penrose quantities .
rather , we seek a form which would allow the teukolsky fields to be deduced _
ab initio _ from the physical metric .
there is a price to be paid for this , as certain results appear more complicated in this language .
a complementary , and slightly more comprehensive , analysis based on the newman
penrose approach is given in a companion paper @xcite .
this complementary analysis operates by picking an arbitrary tetrad on a generic space - time and showing how to transform it to yield one of interest in the teukolsky formalism .
we present the two analyses separately to maximize the clarity of each .
the following is an outline of this paper .
first , we review elements of the teukolsky formalism from the traditional point of view , where a preferred background kerr geometry is known _ a priori _ , and formulate in detail the problem to be solved .
second , we introduce the notion of a _
transverse frame_. several such frames exist in a general space - time , all are calculable from the physical geometry , and one is the _ quasi - kinnersley frame _ of interest in the teukolsky formalism . finally , to help deploy these results for numerical relativity , we express these results using only initial - value data on an arbitrary spatial slice .
we conclude with several comments .
finally , an appendix elaborates one of the central calculations of the paper .
the observable signature of a non - trivial gravitational field in general relativity is space - time curvature . however , because no coordinate system is preferred over any other , the coordinate components of the curvature tensor do not generally have any invariant physical meaning . one must be careful to distinguish , for example , between actual gravitational radiation and oscillations in those components caused by a peculiar choice of coordinates .
the teukolsky formalism of first - order perturbation theory @xcite solves this problem using a family of gauge - invariant scalar variables originally proposed by newman and penrose @xcite .
the newman
penrose approach introduces an initially arbitrary _ null tetrad _ @xmath1 on space - time .
the four vector fields here are all null , with the first pair real and the second both complex and conjugate to one another .
their only non - vanishing inner products are @xmath2 it is convenient to restrict attention to _ oriented _ null tetrads satisfying @xmath3 } = \epsilon_{abcd},\ ] ] where @xmath4 denotes the usual space - time volume element .
this restriction can always be made without loss of generality in cases of physical interest .
the independent components of the weyl curvature tensor in such a non - coordinate basis are encoded in five complex _ weyl scalars _ @xmath5 \psi_2 & { } : = c_{abcd } \ , \ell^a \ , m^b \ , \bar m^c \ , n^d \\[.5ex ] \psi_4 & { } : = c_{abcd } \ , \bar m^a \ , n^b \ , \bar m^c \ ,
n^d . \end{array}\quad
\begin{array}{@{}r@{}l@ { } } \psi_1 & { } : = c_{abcd } \ , \ell^a \ , m^b \ , \ell^c \ , n^d \\[.5ex ] \psi_3 & { } : = c_{abcd } \ , \ell^a \ , n^b \ , \bar m^c \ , n^d \end{array}\ ] ] these scalar quantities are naturally independent of any space - time coordinate system , but do clearly depend on the choice of null tetrad .
accordingly , the @xmath6 can acquire direct ( observer - independent ) physical meaning only when a physically preferred null tetrad can be found .
the teukolsky formalism picks a particular null tetrad for the newman
penrose scalars by exploiting special features of its given kerr background .
specifically , any kerr geometry admits exactly two repeated principal null directions emphasize that only the directions of these vectors are determined . ]
@xmath7 and @xmath8 . in any null tetrad @xmath1 whose real null vectors lie in these directions , all the newman
penrose curvature components except @xmath9 vanish , guaranteeing , as we discuss further below , the gauge - invariance of first - order perturbations in @xmath10 and @xmath11 .
it is therefore quite natural to choose the directions of the tetrad fields in this way , but one must proceed to fix their scalings along these directions as well . a technique to do this in an exact kerr , or generally in any petrov type d ,
geometry was first proposed by kinnersley @xcite . to explain this technique , it is useful to recall the generic ambiguity in the null tetrad . at a given point of a general space - time
, one can transform any given oriented null tetrad @xmath1 to any other using elementary transformations of three basic types .
these include two families of elementary _ null rotations _ @xmath12 and @xmath13 , with @xmath14 and @xmath15 arbitrary complex parameters , which preserve either @xmath16 or @xmath17 , respectively , while changing the directions of all three of the other tetrad elements . a third family of _ spin - boost _ transformations @xmath18 , with @xmath19 an arbitrary complex parameter , mutually scales the real null vectors @xmath16 and @xmath17 , leaving their directions unchanged , and rotates the complex elements @xmath20 and @xmath21 by complementary phases .
in addition to these , we highlight a single fourth transformation , the _ exchange _ operation @xmath22 , which interchanges the real and complex tetrad elements in pairs .
the detailed formulae describing these transformations are unimportant here .
they are standard and can be found , for instance , in @xcite .
we do , however , note the effects of these transformations on the weyl scalars @xmath6 : @xmath23 n(b ) : \psi_n \mapsto \check\psi_n & { } = \sum_{m = n}^4\ , { { 4-n } \choose { m - n}}\ , b^{m - n}\ , \psi_m \\[1ex ] s(c ) : \psi_n \mapsto \tilde\psi_n & { } = c^{2 - n } \ , \psi_n \\[1ex ] e : \psi_n \mapsto { \vbox{\mathsurround=0pt \ialign{##\crcr$\hss\rlap{$\scriptscriptstyle\rightharpoonup$}\relax \hbox{$\scriptscriptstyle\leftharpoonup$}\hss$\crcr \noalign{\nointerlineskip}$\displaystyle{\psi}$\crcr}}\relax \vphantom{\psi}}_n & { } = \psi_{4-n } \end{array}\ ] ] where @xmath24 denote the usual binomial coefficients . in the kerr geometry ,
fixing @xmath16 and @xmath17 to lie in the two repeated principal null directions eliminates the null rotations completely , but does not restrict the spin - boost and exchange freedom in any way . to proceed ,
kinnersley implicitly breaks the exchange symmetry by choosing @xmath16 to point outward , toward infinity , in the exterior of the hole and then sets @xmath25 the quantity @xmath26 is one of the standard ricci rotation coefficients used in the newman
penrose approach . setting it to zero
does not eliminate the residual spin - boost ambiguity entirely , however , as one remains free to do spin - boost transformations with parameters @xmath19 satisfying @xmath27 note that , since the right side vanishes , this equation is independent of the scaling of the vector field @xmath16 , and implies only that @xmath19 must be constant along each of its integral curves .
the set of such curves rule the exterior region of the black hole and each intersects a unique point of future null infinity . as a result , adapting the scaling of the tetrad near infinity to stationary observers there uniquely determines the kinnersley tetrad everywhere on kerr space - time . that is , fixing the scalings of @xmath16 and @xmath17 such that their spatial projections relative to a stationary observer have equal magnitude near infinity breaks the remaining spin - boost invariance . in the teukolsky formalism ,
one considers only space - times describing perturbations of a _ known _ kerr metric .
the physical metric , consisting of both background and perturbation , generally does not have repeated principal null directions , and does not admit a preferred kinnserley tetrad .
however , it is nonetheless natural to restrict attention to null tetrads of this metric which differ from the background s kinnersley tetrad only at first order in perturbation theory .
this is teukolsky s approach . in such
a _ perturbed kinnersley tetrad _ ,
all of the curvature components except @xmath9 vanish at leading order , though all may acquire first - order corrections . as a result ,
all components except @xmath9 are invariant under first - order diffeomorphisms , the gauge transformations of the linearized theory .
moreover , the residual , first - order ambiguity in the tetrad is generated by null rotations with parameters @xmath14 and @xmath15 which vanish at leading order , and spin - boosts with parameters @xmath19 equal to unity at leading order .
the exchange operation , being finite , is disallowed .
these residual transformations certainly modify the perturbed tetrad , but do not affect the values of the curvature components @xmath10 , @xmath9 and @xmath11 .
thus , @xmath10 and @xmath11 evaluated in _ any _ such perturbed tetrad are invariant in both senses , and acquire physical meanings .
specifically , using an analysis of the geodesic deviation equation due to szekeres @xcite , they are associated with transverse gravitational radiation propagating along the null directions @xmath17 and @xmath16 , respectively .
the components @xmath28 and @xmath29 in a perturbed kinnersley tetrad are invariant under infinitesimal diffeomorphisms , but may be given any value whatsoever by performing suitable first - order null rotations on the tetrad .
they are therefore pure gauge in the teukolsky formalism . in a variety of situations , including both
the reconstruction of metric perturbations in the first - order theory @xcite and a second - order perturbation analysis @xcite , it is convenient to use the remaining null rotation freedom to set @xmath30 this can always be done , and exhausts the first - order null rotation freedom in the tetrad .
there is no similar convenient criterion to eliminate the first - order spin - boosts , but none is needed in the teukolsky formalism .
let us now focus on the subtleties inherent in applying teukolsky s techniques in numerical relativity .
black holes in general relativity are known to be stable @xcite .
thus , when a numerical evolution produces a space - time containing a single black hole , one should expect it to settle down over time , approaching a quiescent final state .
a description via perturbation theory should accordingly become increasingly viable over time , although in practice it may be far from clear exactly _ how _ a given physical metric should be so described .
one natural response to this difficulty would be to search among _ all _ possible kerr metrics on the given space - time for one in which the perturbation , the difference between the physical metric and that particular kerr metric , is smallest . however , even if it were clear how to conduct such a search systematically , it would likely be an extremely difficult procedure .
we take a different approach here .
the key element of the teukolsky formalism provided by the background kerr metric is its kinnersley tetrad . rather than search for an approximate kerr metric on space - time whose kinnersley tetrad will allow us to exploit the teukolsky formalism
, we search directly for an approximation to that kinnersley tetrad . given
_ only _ the physical metric on a perturbed kerr space - time , a natural way to begin searching for its perturbed kinnersley tetrad is to seek tetrads for the physical metric satisfying eq .
( [ kinn ] ) .
that is , to find a tetrad with @xmath28 and @xmath29 vanishing at leading order , simply demand they vanish to all orders .
this eliminates only gauge degrees of freedom in first - order perturbation theory , and there is no obvious problem of principle seeking such tetrads at an arbitrary point of a general space - time . the components @xmath10 and @xmath11 in such a frame can not generally be set to zero simultaneously by further refining the choice of tetrad . even in perturbation theory , these components are physical and provide a rough , quantitative measure of how `` far '' such a space - time is from kerr @xcite .
motivated by their association with transverse radiation in perturbation theory , we refer to tetrads satisfying eq .
( [ kinn ] ) as _ transverse null tetrads_. we emphasize , however , that in regions of space - time supporting strongly self - gravitating fields , there may be no precise relation between these tetrads and any proper notion of gravitational radiation . even in the comparatively simple case of a perturbed kerr space - time , there may be many different transverse tetrads .
these may differ from one another in two ways .
first , the transversality condition of eq .
( [ kinn ] ) is invariant under the spin - boost and exchange transformations of eq .
( [ trans ] ) whence , whenever one transverse tetrad exists , so do infinitely many others .
in addition to this generic multiplicity , however , there could exist distinct transverse tetrads on a given space - time related by non - trivial null rotations .
this second ambiguity raises a more serious problem in our context since even the _ directions _ of the real null vectors in a transverse tetrad on a perturbed kerr space - time may have nothing to do with the underlying kinnersley tetrad .
the current paper addresses this second difficulty .
it constructs , _ up to spin - boost and exchange _ , a unique tetrad on a broad class of space - times comprising a _ non - perturbative _ neighborhood of the kerr geometry in the space of solutions to einstein s equations .
we refer to the product of this construction as the _ quasi - kinnersley frame_. mathematically , it is an equivalence class @xmath31 of null tetrads which can be transformed into one another using only spin - boost and exchange transformations .
the set of all tetrads at a point is a disjoint union of such equivalence classes , and we refer to these generally as _
null frames_. although this nomenclature may be slightly unfortunate the terms `` null tetrad '' and `` null frame '' are often used interchangeably in the literature we will adhere strictly to it here since the notion of a frame is essential to our argument .
a null frame is completely determined by the common _ directions _ , @xmath7 and @xmath8 , of the real null vectors in its constituent tetrads .
these real null vectors span a time - like 2-plane in the tangent space at each point of space - time , while the complex vectors span the orthogonal space - like 2-plane .
spin - boost and exchange operations may change the bases defined by the tetrad on these two sub - spaces , but not the sub - spaces themselves . since knowledge of either orthogonal sub - space determines the other ,
the real null directions suffice to determine the frame .
mathematically , we can exploit this interrelation by associating a 2-form}$ ] . ] @xmath32 } - m_{[a}\ , \bar m_{b]}.\ ] ] to any given null tetrad @xmath1 .
this complex 2-form is invariant under spin - boosts and only changes sign under exchange .
in addition , it is self - dual and unit : @xmath33 in fact , _ all _ unit self - dual 2-forms arise in this way . for each
, one can find an oriented null tetrad of vectors @xmath1 such that @xmath34 takes the form of eq .
( [ sig ] ) . for a given @xmath34 , we have @xmath35 \sigma_{ab}\ , m^b & = -{{\textstyle\frac{1}{2}}}\ , m_a & \qquad \sigma_{ab}\ , \bar m^b & = { { \textstyle\frac{1}{2}}}\ , \bar m_a . \end{array}\ ] ]
that is , @xmath16 and @xmath17 are the unique _ real _ null eigenvectors of @xmath34 , while @xmath20 and @xmath21 are its unique complex null eigenvectors orthogonal to both @xmath16 and @xmath17 .
every unit , self - dual 2-form has such eigenvectors , each of which is determined only up to scaling , whence the tetrad is not unique .
rather , there is a canonical one - to - one correspondence between null frames and non - null directions in the complex , three - dimensional space of self - dual 2-forms at each point of space - time . a non - null
, self - dual 2-form @xmath34 is associated to a _ transverse _
null frame if and only if @xmath36 where @xmath37 denotes the weyl curvature tensor of space - time .
for most @xmath34 , the right side would include additional terms of the form @xmath38}$ ] and @xmath39}$ ] .
it is straightforward to show that whenever @xmath40 is self - dual on the left side of eq .
( [ eigen ] ) , the right side is self - dual as well .
thus , the search for transverse frames reduces to an eigenvalue equation for the weyl tensor , viewed as a map from the space of self - dual 2-forms to itself .
notably , the eigenvalue for a given eigen - form of the weyl tensor is directly connected to the newman
penrose scalars in the associated null frame : @xmath41 denotes the @xmath9 scalar , which is invariant under both spin - boost and exchange , evaluated in a _ transverse _ frame .
the above result establishes a one - to - one correspondence between transverse frames @xmath31 and non - null , self - dual eigen - forms @xmath34 of the weyl tensor .
the number and nature of the these eigen - forms are very well understood @xcite , and depend solely on the algebraic ( petrov ) class of space - time .
in particular , there are exactly three transverse frames at each point of a algebraically general ( type i ) space - time .
meanwhile , the weyl tensors of type d space - times , such as the kerr solution , have one degenerate eigen - space , leading to infinitely many transverse frames , as well as a distinguished non - degenerate eigen - form associated with the kinnersley frame .
type ii geometries admit only one transverse frame . in each of these three cases , therefore , the possible ambiguity in the transverse frame of interest is _
finite_. one can recognize the kinnersley frame of a type d or type ii geometry immediately , and the only ambiguity occurs in the general case , where there are only three possible choices .
the most specialized algebraic classes , types n and iii ( as well as type 0 , where the weyl tensor vanishes ) , are very rare and correspond to highly specialized fields which are unlikely to arise in numerical relativity unless sought specifically .
we do not consider them further here .
the central open issue is this . at a typical point in the exterior of a black hole formed in a numerical simulation
, the weyl tensor will most likely be of type i. at late times , perturbation theory is expected to become sufficiently accurate to be meaningfully applied and the perturbed kinnersley tetrad , when fixed to first order according to eq .
( [ kinn ] ) , will then belong to a transverse frame of the physical metric . in order to implement the teukolsky formalism in this situation
, one must discover to which of the three transverse frames of the physical geometry this tetrad belongs .
there are a number of ways one might do this in practice @xcite , but all are based on the same basic idea . if it is possible to distinguish the eigen-_value _ of the weyl tensor associated with the quasi - kinnersley frame , then its eigen-_form _ can be singled out using eq .
( [ eigen ] ) .
the frame may then be reconstructed from the form using eq .
( [ recon ] ) .
the three eigenvalues of the weyl tensor at a given space - time point are independent of any basis or coordinate system one might use to evaluate them ; they are algebraic scalar curvature invariants of the weyl tensor . as such
, they should be calculable directly from space - time geometry .
to realize this point explicitly , we calculate the characteristic polynomial @xmath42 of the linear map defined by the weyl tensor from the space of self - dual 2-forms to itself . this calculation is described in some detail in an appendix .
the roots @xmath43 of this polynomial are the three values of @xmath41 in the three transverse frames at a generic point of space - time , which are related to the eigenvalues by eq .
( [ eigen ] ) .
the quantities @xmath44 and @xmath45 appearing here are the well - known scalar curvature invariants defined by @xmath46 j & { } : = { { \textstyle\frac{1}{96 } } } \left ( c_{ab}{}^{cd } \ , c_{cd}{}^{ef } \
, c_{ef}{}^{ab } - i \ , c_{ab}{}^{cd } \ ,
c_{cd}{}^{ef } \,{}^\star\ !
c_{ef}{}^{ab } \right ) , \end{array}\ ] ] where @xmath47 denotes the space - time dual of the weyl tensor .
these invariants are manifestly calculable directly from space - time geometry , with no mention of newman
penrose tetrads , and show implicitly that the three @xmath41 are indeed curvature invariants
. this result may be somewhat surprising at first since the curvature component @xmath9 generally depends entirely on the frame used to evaluate it .
however , we seek here its values only in specific , _ geometrically privileged _ , transverse frames . since these frames are defined invariantly ,
so are the values of @xmath41 in them .
is a triple - cover of the riemann sphere .
the values of @xmath48 on its three sheets give the three values of @xmath43 .
the labels in the figure indicate contiguous regions of @xmath49 pictured on different sheets . ]
the cubic polynomial of eq .
( [ poly ] ) is simple enough to solve analytically , but its solutions will involve multiple - valued functions of the complex variables @xmath44 and @xmath45 . to minimize the subtleties raised by these functions , it is convenient to express the solution in terms of the baker campanelli _ speciality index _
@xcite @xmath50 , yielding @xmath51^{1/3 } + [ w({\mathcal{s}})]^{-1/3}}{\sqrt{s}},\ ] ] where @xmath52 .
the multiple values of @xmath43 then arise from the various branches of @xmath48 .
this function has branch points of order two at @xmath53 and @xmath54 arising from the square roots in @xmath55 and , since @xmath55 does not vanish for any finite value of @xmath56 , only one branch point of order three at @xmath57 .
a careful analysis shows the riemann surface @xmath49 for @xmath48 has the three - sheeted structure depicted in fig .
[ surface ] .
the three eigenvalues of the weyl tensor arise from the values of @xmath48 in these three sheets .
the speciality index was introduced to provide a quantitative measure of whether a given physical metric should admit an accurate description in perturbation theory @xcite .
it equals unity if and only if space - time is either of type d or of type ii .
( it ill - defined for space - times of types n and iii , where @xmath58 , but we do not consider these exotic cases here .
they could be incorporated consistently , however , by working with a multivariate expression for @xmath59 in terms of @xmath44 and @xmath45 . )
one therefore expects @xmath60 at late times in a numerical space - time describing a black hole approaching equilibrium .
only one branch of @xmath48 , associated with the leftmost leaf of @xmath49 in fig .
[ surface ] , is analytic in this limit .
let @xmath61 denote the value of @xmath43 computed in this branch and expand about @xmath62 to find @xmath63.\ ] ] because @xmath48 is analytic at @xmath54 in this leaf of @xmath49 , we find a proper taylor expansion .
on the other hand , @xmath54 is a branch point of order two in the other two leaves , and the expansion there should accordingly be done in powers of @xmath64 rather than @xmath65 . denoting the values of @xmath43 computed in these branches by @xmath66
, we expand to find @xmath67\ ] ] these two expansions are completely independent of how the branch lines for @xmath49 are drawn in the complex @xmath56-plane .
the branch point at @xmath54 makes it impossible to distinguish the roots @xmath66 from one another in any neighborhood of that point , and changing the ambiguous sign of @xmath64 indeed merely interchanges these two roots .
since the eigenvalues @xmath66 are equal when @xmath54 , one expects they must be associated with the degenerate eigen - space of the weyl tensor on a kerr space - time , meaning @xmath61 must be associated with the kinnersley frame .
one can check this explicitly by writing the @xmath44 and @xmath45 in terms of newman
penrose components : @xmath68 j & { } = \psi_0 \psi_2 \psi_4 - \psi_2 ^ 3 + 2 \psi_1 \psi_2 \psi_3 - \psi_0 \psi_3 ^ 2 - \psi_1 ^ 2 \psi_4 .
\end{array}\ ] ] in the kinnersley frame on a type d space - time , only @xmath9 is non - vanishing , and these formulae show that @xmath69 in that frame .
this , of course , is exactly the value of @xmath61 when @xmath54 .
however , the expansion of eq .
( [ expand0 ] ) remains valid within some finite radius of convergence in the complex @xmath56-plane . within this radius
, we can _ rigorously _ identify a _
unique _ eigen - form @xmath70 of the weyl tensor satisfying @xmath71 we then _ define _ the quasi - kinnersley frame as that associated to @xmath70 by eq .
( [ recon ] ) .
this definition is valid whenever @xmath56 is sufficiently close to unity .
this quasi - kinnersley frame may be selected in a number of alternative , but equivalent , ways .
definition 3 of the companion paper @xcite offers one such alternative , essentially using the distinct limits @xmath72 and @xmath73 as @xmath60 to distinguish the desired frame .
that procedure will of course select the same quasi - kinnersley frame described above .
however , we have formulated the definition given here to emphasize a point . in numerical work , one has at one s disposal only the given physical metric and can not actually take the limit @xmath60 at a particular point of the numerical space - time . nonetheless , for a given @xmath74 , eq .
( [ expand0 ] ) gives an unequivocal value for @xmath61 which may be used to select the quasi - kinnersley frame via eq .
( [ qkinn ] ) .
this branch , and therefore the associated frame , does not exist only in the limit @xmath60 , but rather in a finite neighborhood of @xmath54 in @xmath49 given by the radius of convergence of the series of eq .
( [ expand0 ] ) .
thus , although the quasi - kinnersley frame is selected fundamentally by its limiting behavior , it is defined at points of space - time where @xmath56 differs quite substantially from unity . to understand the limitations of the definition , therefore , we need to analyze the series of eq .
( [ expand0 ] ) .
fortunately , the essential features of the series are evident in the structure of the riemann surface @xmath49 .
they are brought out by the following , relatively practical , considerations . in practice ,
the computer can not calculate @xmath61 by summing the ( infinite ) series whose leading terms are given in eq .
( [ expand0 ] ) .
this series can be very slow to converge at points .
rather , it is better to calculate @xmath61 non - perturbatively by picking particular branch cuts and branches for the square- and cube - roots in eq .
( [ soln ] ) such that the series expansion of @xmath43 with those choices yields @xmath61 near @xmath54 .
it turns out the principal branches of all three radicals , where @xmath75{1}$ ] , with the conventional branch lines along the negative real axis yield @xmath61 everywhere in the disk @xmath76 .
the ambiguous sign of @xmath64 causes no problem in this region because changing it , given the branches of the other radicals , merely interchanges the two cube roots in @xmath48 .
thus , @xmath61 is continuous across that part of the real axis joining @xmath53 and @xmath54 .
this is a crude reflection of the absence of any branch line going to @xmath54 in the corresponding leaf of @xmath49 . when @xmath76 , we have given a simple algorithm to calculate @xmath61 , and therefore the quasi - kinnersley frame , non - perturbatively . every perturbed kerr space - time studied by the teukolsky formalism
will certainly satisfy this inequality everywhere outside the black hole . on such space - times
, the quasi - kinnersley frame provides a gauge - fixed version of the perturbed kinnersley frame which is manifestly derived solely from the physical metric .
however , other space - times , which in no way represent small perturbations of the kerr geometry , also admit a quasi - kinnersley frame .
this structure is therefore uniquely and rigorously defined on every space - time in a _
finite _ neighborhood of the kerr solution in the space of solutions to the einstein equations .
it may be possible to extend the definition of the quasi - kinnersley frame to even more general regions of space - time using a continuity argument .
the critical ingredient of the definition which fails when @xmath77 is that it becomes impossible to distinguish @xmath61 from @xmath66 because of the branch point at @xmath53 in @xmath49 . concretely , if one continued to use the principal branches of the radicals in eq .
( [ soln ] ) to define @xmath61 in this broader region , this eigenvalue , and therefore the distribution of associated frames , will be discontinuous at any point of a given curve in space - time where @xmath56 is real and negative .
this discontinuity , however , stems entirely from the branch line in @xmath49 .
if we instead define a function @xmath43 along such a curve which equals @xmath61 up to the point on the curve where @xmath56 is real and negative , and thereafter equals one of the other roots , either @xmath78 or @xmath79 , then @xmath43 _ will _ allow a continuous extension of the quasi - kinnersley frame along the curve .
the roots @xmath66 can be found using the principal branches of the two square roots in eq .
( [ soln ] ) , and the branch of the cube root where @xmath80{1 } = ( -1\pm i\sqrt{3})/2 $ ] . in principle
, it appears this extension of the quasi - kinnersley frame to strong - field regions can be carried out coherently ( i.e. , such that the continuous propagation of the frame from one space - time point to another is independent of the curve used to join them ) provided there is no 3-volume in space - time where the curvature is everywhere either type n or iii . in practice , however , if one is interested in harnessing the conceptual power of the teukolsky formalism in numerical relativity , this may not be worthwhile .
this `` root - substitution '' scheme could incur a high computational expense since it can only be done retroactively , once the late - time evolution is known , and perturbation theory would likely be unreliable anyway when @xmath81 .
nonetheless , there may be situations where the numerical overhead could be justified .
for example , @xcite has advanced a definition of a _ radiation scalar _ based on exactly this sort of approach to defining the quasi - kinnersley frame in strong - field regions of space - time .
that argument yields a scalar invariant @xmath82 , defined throughout space - time , which has a clear interpretation in teukolsky - type radiation zones .
the previous section has defined the quasi - kinnersley frame from a space - time perspective . however , this language may not be the most convenient for numerical relativity .
few numerical codes currently implement a space - time covariant form of the einstein equations , preferring a space+time split where gravity is encoded in a set of fields evolving relative to a fiducial foliation by space - like hypersurfaces .
these fields include at least the intrinsic spatial metric @xmath83 and extrinsic curvature @xmath84 of each leaf of the foliation .
this section describes a simple way to calculate the quasi - kinnersley frame using only these initial - value data , thereby offering a direct path toward numerical implementation .
let @xmath85 denote the unit future - directed normal to a given space - like hypersurface in space - time .
following the convention used in maxwell theory , a general unit self - dual 2-form @xmath34 can be decomposed into _ electric _ and _ magnetic parts _ as @xmath86 } + { { \textstyle\frac{i}{2}}}\ , \epsilon_{abc}\ , \hat\sigma^c.\ ] ] here , @xmath87 denotes the intrinsic spatial volume element and @xmath88 is a _ complex _ spatial vector with real , unit norm .
since @xmath34 is self - dual , its electric part ( technically , @xmath89 ) also determines its magnetic part , and therefore @xmath34 itself , completely .
thus , at each point on a given spatial hypersurface , there is a one - to - one correspondence between _ complex _ spatial unit vectors ( up to sign ) and null frames on space - time at that point .
an equation which determines the @xmath88 corresponding to _ transverse _ frames follows by taking the electric part of both sides of eq .
( [ eigen ] ) : @xmath90 this is once again an eigenvalue problem whose solutions are proportional to the @xmath9 scalar evaluated in a transverse frame . the spatial tensor @xmath91 featured here is symmetric and trace - free , though generally complex .
its real and imaginary parts are the electric and magnetic components of the weyl tensor itself , defined relative to the given spatial slice by @xmath92 b_{ab } & { } : = \hat\tau^m\ , \hat\tau^n\ , { } ^\star\ ! c_{ambn } : = { { \textstyle\frac{1}{2}}}\ , \hat\tau^m\ , \hat\tau^n\ , \epsilon_{amij}\ , c^{ij}{}_{bn } , \end{array}\ ] ] respectively .
both of these tensors are obviously real , and the symmetries of the weyl tensor imply each is separately symmetric and trace - free . the gauss codazzi
mainardi relations of differential geometry yield expressions for @xmath93 and @xmath94 in terms of initial - value data .
the result , which applies regardless of whether the space - time metric satisfies the einstein equations , suggests we consider the quantities @xmath95 \mathcal{b}_{ab } & { } : = -\epsilon_a{}^{mn}\ , d_m\ , k_{nb}. \end{array}\ ] ] here , @xmath96 is the traced extrinsic curvature , while @xmath97 is the ricci curvature of the torsion - free connection @xmath98 compatible with the spatial metric @xmath83 .
the last term in the expression for @xmath99 is the spatial projection of the _ space - time _
einstein tensor .
when the einstein evolution equations hold , therefore , this term can be replaced with the stress tensor of matter in the spatial slice , and thereby expressed in terms of cauchy data for the matter fields .
thus , @xmath99 and @xmath100 are functions only of initial - value data when the equations of motion hold .
moreover , @xmath99 is naturally symmetric and its trace - free part is @xmath93 , while @xmath100 is naturally trace - free and its symmetric part is @xmath94 .
thus , one can compute the tensor @xmath91 in eq .
( [ emred ] ) from initial - value data for both the gravitational and matter fields .
the eigenvalue problem of eq .
( [ emred ] ) is easy enough to solve in practice @xmath91 is , after all , only a @xmath101 matrix once coordinates are chosen but we must once again pick from among its ( generally ) three solutions the one corresponding to the quasi - kinnersley frame . to do so , we first find @xmath102 these formulas give the fundamental algebraic invariants of the space - time weyl tensor as functions of spatial initial - value data .
one can then easily compute the speciality index and , provided @xmath76 , calculate @xmath61 using eq .
( [ soln ] ) . as before
, @xmath61 can be found using the principle branches of the square- and cube - roots in eq .
( [ soln ] ) with the branch cuts for all three radical functions along the negative real axis .
this preferred eigenvalue determines a preferred eigenvector @xmath103 in eq .
( [ emred ] ) associated to the quasi - kinnersley frame .
once the eigenvector @xmath103 has been found , one might like to find the elements of some tetrad in the quasi - kinnersley frame .
this , too , can be done using only spatial , rather than space - time , data .
the real null vectors of an arbitrary tetrad project into a space - like hypersurface to define a pair of real spatial vectors .
while the normalizations of these spatial projections naturally vary if one performs a spin - boost on the tetrad , they define a pair of invariant rays in the tangent space at a point of space - time .
these rays correspond to a pair of real unit vectors @xmath104 and @xmath105 which are generally completely independent of one another .
in particular , they point in exactly opposite directions only when the normal @xmath85 to the spatial slice lies in the space - time tangent 2-plane spanned by @xmath16 and @xmath17 .
the fiducial time foliation used in numerical relativity is overwhelmingly likely be `` boosted '' relative to the 2-plane defined by a given null tetrad , and only parallel unit vectors are actually disallowed .
since @xmath104 and @xmath105 determine the directions of the space - time vectors @xmath16 and @xmath17 , they also suffice to determine the frame associated to @xmath88 , and thus @xmath88 itself .
the explicit relation is @xmath106 note that when @xmath104 and @xmath105 are interchanged , @xmath88 simply changes sign , as it should since @xmath34 changes sign under an exchange operation in the corresponding frame . to invert eq .
( [ sln ] ) and solve for @xmath104 and @xmath105 given @xmath88 , we separate @xmath88 into real and imaginary parts : @xmath107 in general , neither @xmath108 nor @xmath109 is unit , but the normalization condition for @xmath88 demands they be orthogonal with norms satisfying @xmath110 . taking the electric part of eq .
( [ recon ] ) then yields @xmath111 finally , the elements of the most general null tetrad @xmath1 in the frame associated to @xmath88 take the form @xmath112
n^a & { } = \frac{|c|^{-1}}{\sqrt{1 - \hat\lambda \cdot \hat\nu}}\ , ( \hat\tau^a + \hat\nu^a ) \\[2ex ] m^a & { } = \frac{e^{i\theta}}{\sqrt{2}}\ , \frac{\sqrt{1 + \hat\lambda \cdot \hat\nu}}{\sqrt{1 - \hat\lambda \cdot \hat\nu}}\ , ( \hat\tau^a + \hat\mu^a ) , \end{array}\ ] ] where the _ complex _ unit projection of @xmath20 is @xmath113 these vectors are the result of a spin - boost , with parameter @xmath114 , acting on a preferred tetrad determined by setting @xmath115 and @xmath116 .
note that in the degenerate case where @xmath104 and @xmath105 are indeed anti - parallel , the expression for @xmath20 in eq .
( [ lnproj ] ) is ill - defined .
however , the limit as @xmath117 does exist , though it does depend on how the limit is taken .
should this situation arise in practice , it is easy enough to accommodate .
the first two of eqs .
( [ lnproj ] ) are non - singular , and remain unchanged .
one can then choose _ any _ real unit vector @xmath118 in the spatial 2-plane orthogonal to @xmath119 , and take @xmath120 this replaces the third of eqs .
( [ lnproj ] ) to give the remaining two elements of the desired ( oriented ) null tetrad on space - time .
note that there are two arbitrary parameters , the unit vector @xmath118 and the spin parameter @xmath121 , needed to specify @xmath20 in this degenerate case .
they are redundant .
any rotation of @xmath118 in the plane orthogonal to @xmath104 may be compensated by an adjustment to the spin parameter @xmath121 ; as in the non - degenerate case , the spin - boost freedom in the tetrad will remain .
the only cost of degeneracy is the loss of the natural `` gauge - fixing '' of the spin freedom implicitly used in eq .
( [ lnproj ] ) .
that is , we can not specify @xmath122 by insisting that @xmath123 be real and negative since , in the degenerate case , it vanishes for all values of @xmath121 . in any event , this amounts only to a slight mathematical inconvenience .
the residual freedom in the tetrad is the same regardless of whether @xmath119 and , moreover , this circumstance is unlikely to be realized _
exactly _ in numerical relativity .
this section has now shown explicitly that transverse frames can be calculated directly from data on an _ arbitrary _ spatial slice of a given space - time .
in particular , one can find these space - time quantities using the fiducial foliation underlying any numerical evolution scheme based on the familiar arnowit
misner equations or their various descendants .
the central point this paper has argued is that the quasi - kinnersley frame is naturally defined on a broad class of space - times . as long as space - time contains no finite volumes of very high algebraic speciality
, a `` root - substitution '' scheme can propagate the quasi - kinnersley frame continuously from regions supporting fairly weak fields , where there is absolutely no ambiguity in its definition , to arbitrary regions supporting much stronger fields . to avoid potential subtleties inherent to this scheme
, one can restrict attention to situations where @xmath124 throughout space . in this restricted case ,
the quasi - kinnersley frame is particularly easy to find . in the still more restrictive limit @xmath60
, the quasi - kinnersley frame always exists , depends only on the physical metric , and identifies a `` gauge - fixed '' version of the perturbed kinnersley frame satisfying eq .
( [ kinn ] ) .
this provides essential guidance needed to deploy the teukolsky formalism where it is expected to be relevant
. this feature could be useful in extracting geometrically - invariant , _ physically meaningful _
information from the raw numerical data of late - time black hole simulations .
the mathematical structure discussed above could also be useful in constructing observer - independent scalar functions in more general regions where perturbation theory may not apply .
although the physical meaning of these scalars may be somewhat obscure in regions supporting self - interacting fields , they can still be used , in principle , as gauge - invariant observables and thus help clarify invariantly the structure of simulated space - times .
it does not seem possible to establish concrete , analytic limits on the regime where perturbative techniques are justified .
their viability must be tested by some other means .
the internal consistency checks of the lazarus project @xcite , where late - time evolution is done by extracting teukolsky quantities from numerical data on a given time - slice in a non - linear evolution and thereafter using the teukolsky formalism s much more stable evolution equations , provide a natural way to do this .
these consistency checks ensure the physical content of the very late - time data does not depend sensitively on when the extraction took place .
this paper has argued that one can begin attempting extractions whenever one first finds @xmath124 everywhere outside a single black hole . in practice , it appears this can happen surprisingly early in a full numerical evolution @xcite .
in addition to transverse frames , it is possible to consider several other geometrically - preferred frames on a typical space - time .
for example , frames whose real null directions are two of the ( generally four ) principal null directions of the physical metric are quite natural to consider .
the connection between such principle null frames and the transverse frames considered here is discussed much more thoroughly in the companion paper @xcite ( see especially section ii.f ) .
one might wonder whether techniques similar to those outlined here for the quasi - kinnersley frame could be developed to pick some of these other preferred frames on a generic geometry .
the answer , however , is probably in the negative .
the essential feature which has distinguished the quasi - kinnersley frame in this paper is the analyticity of @xmath61 near @xmath62 .
even the other two transverse frames can not be distinguished from one another near algebraic speciality because of the branch cut joining their associated leaves of the riemann surface at this point .
one can show using standard perturbation theory of linear operators that , since @xmath66 fail to vary analytically near speciality , their associated frames also fail to do so @xcite .
this non - analyticity , and our inability to distinguish these frames invariantly , arises precisely because we have perturbed away from a situation where the normally distinct frames degenerate .
all of the other obvious geometrically - preferred frames , such as those based on principal null directions , also degenerate in algebraically special space - times , and thus will share this problem .
the quasi - kinnersley frame is _ unique _ in that it varies analytically when perturbed , and is therefore arguably a singularly useful structure in the sort of analyses contemplated here .
as discussed in the introduction , a second step must be taken to complete the construction begun in this paper and allow an application of the full range of teukolsky s techniques to the results of numerical simulations .
this step will break the spin - boost - invariance of the current results and choose a preferred _ quasi - kinnersley tetrad _ within the quasi - kinnersley frame . unlike the frame - fixing procedure described above , which has been entirely local , done separately at each space - time point , fixing the scalings seems likely to require a _
global _ approach .
setting @xmath125 , as for the kinnersley tetrad on kerr space - time , can still be done generically , and will continue to help pare down the possibilities for a quasi - kinnersley _
tetrad_. however , as in kinnersley s case , this will not determine the tetrad uniquely .
in fact , since the @xmath125 condition ties together the scalings of tetrad elements along the integral curves of @xmath16 , it would seem that fixing the scaling close to the black hole would involve making choices in the evolved fields at the outer boundary at _ much _ later times .
since long - time evolutions of the full , non - linear equations of general relativity are currently unavailable , it seems likely an indirect approach to this problem will be required for the near - term .
it may be possible to found such a approach on a scheme to approximate additional structures of the kerr geometry underlying a given perturbed metric using only the physical metric itself .
work is currently underway to explore whether such an approach could yield a practicable solution to the problem of finding the quasi - kinnersley tetrad .
the authors would like to thank richard price for several stimulating discussions of this work .
cb would also like to thank manuela campanelli , karel kucha , alok laddha , carlos lousto and warner miller for helpful conversations , and the nasa center for gravitational wave astronomy at university of texas at brownsville ( nag5 - 13396 ) for its hospitality during part of the execution of this work .
this work was supported by the national science foundation through grants phy-0244605 to the university of utah and phy-0400588 to florida atlantic university , as well as by the eu network programme through research training network contract hprn - ct-2000 - 00137 and nasa through grant nng04gl37 g to the university of texas at austin .
the characteristic polynomial of eq . ( [ poly ] ) plays a key role in identifying the quasi - kinnersley frame . for definiteness , this appendix will derive it .
the derivation follows closely the discussion given in chapter 4 of @xcite .
the space of self - dual 2-forms at a point of space - time is a three - dimensional complex vector space . to see this
, we recall that in lorentzian signature , any 2-form @xmath126 may be written as a sum of its self - dual and anti - self - dual parts : @xmath127 where @xmath128 $ ] are the _ self - dual _ ( upper sign ) and _ anti - self - dual _ ( lower sign ) _ parts _ of @xmath126 .
one can check the 2-form @xmath129 ( @xmath130 ) is ( anti-)self - dual using the result @xmath131 , which holds for the double dual of any 2-form in lorentzian signature .
the space of all 2-forms in four dimensions is six dimensional , but the decomposition of eq .
( [ decomp ] ) splits it into two subspaces which are complex conjugates of one another . that is , we see by taking the conjugate of the definition of self - duality in eq .
( [ sd ] ) that the complex conjugate of a self - dual 2-form is anti - self - dual .
these two subspaces can not intersect and together form the whole of the space of 2-forms , whence each must be three - dimensional .
the contraction of any self - dual 2-form @xmath34 with the weyl tensor will yield another self - dual 2-form .
this happens because the weyl tensor has the remarkable property that its duals on either pair of indices are equal to one another : @xmath132 thus , the dual of the contraction @xmath133 can be moved to act on the second pair of weyl indices and then to act on @xmath40 , producing a factor of @xmath134 and showing the contraction to be self - dual .
thus , the weyl tensor defines a linear map from a three - dimensional complex vector space to itself . in eq .
( [ dual ] ) , we have shown explicitly how the three - dimensional space of self - dual 2-forms @xmath34 at a point of space - time can be described in terms of complex spatial vectors @xmath135 in an arbitrary spatial hypersurface .
these later objects also form a three - dimensional complex vector space . indeed , the map induced by the weyl tensor on self - dual 2-forms is completely equivalent to the map @xmath136 on complex spatial vectors defined by eq .
( [ emred ] ) .
for conceptual simplicity , then , we may calculate the characteristic polynomial of the spatial map @xmath136 rather of than the space - time weyl tensor restricted to self - dual 2-forms . for the moment , let @xmath136 be _ any _ map on a three - dimensional vector space . using standard techniques in linear algebra
, one can show the its characteristic polynomial can be expressed in terms of the traces of @xmath136 and its powers in the form @xmath137 \\[1ex ] & { } = c^3 - \langle c \rangle\ , c^2 + { \textstyle\frac{1}{2 } } \left [ \langle c \rangle^2 - \langle c^2 \rangle \right ] c \\[1ex ] & \qquad { } - { \textstyle\frac{1}{6 } } \left [ \langle c \rangle^3 - 3 \langle c \rangle \langle c^2
\rangle + 2 \langle c^3 \rangle \right ] , \end{array}\ ] ] where @xmath138 is the usual identity map and @xmath139 denotes the trace of the @xmath140 power of @xmath136 .
returning now to the particular case of the weyl tensor , we see immediately that these coefficients simplify considerably because @xmath136 is trace - free
. moreover , eqs .
( [ ijiv ] ) relate the remaining traces to the curvature invariants @xmath44 and @xmath45 .
this reduces the characteristic equation for the eigenvalues of the weyl tensor to the form @xmath141 since , by eq .
( [ emred ] ) , the eigenvalue @xmath19 is twice @xmath41 ( in a transverse frame ) , we recover the polynomial of eq .
( [ poly ] ) for the values of @xmath43 .
s. rowan and j. hough .
gravitational wave detection by interferometry ( ground and space ) .
_ living rev .
relativity _ * 3 * ( 2000 ) 3 .
online article cited 14 june 2004 : ` http://relativity.livingreviews.org/ articles / lrr-2000 - 3/ ` . | the teukolsky formalism of black hole perturbation theory describes weak gravitational radiation generated by a mildly dynamical hole near equilibrium .
a particular null tetrad of the background kerr geometry , due to kinnersley , plays a singularly important role within this formalism . in order to apply the rich physical intuition of teukolsky s approach to the results of fully non - linear numerical simulations
, one must approximate this kinnersley tetrad using raw numerical data , with no _ a priori _ knowledge of a background .
this paper addresses this issue by identifying the directions of the tetrad fields in a quasi - kinnersley frame .
this frame provides a unique , analytic extension of kinnersley s definition for the kerr geometry to a much broader class of space - times including not only arbitrary perturbations , but also many examples which differ non - perturbatively from kerr .
this paper establishes concrete limits delineating this class and outlines a scheme to calculate the quasi - kinnersley frame in numerical codes based on the initial - value formulation of geometrodynamics . |
recent advances in the study of normal galaxies and active galactic nuclei ( agns ) are ample observational evidence for the existence of central supermassive black holes ( smbhs ) and the relationship between smbhs and bulge properties of host galaxies ( gebhardt et al . 2000 ;
ferrarese & merritt 2000 ; tremaine et al .
2002 ; begelman 2003 ; shen et al .
we can use the stellar and/or gaseous dynamics to derive the smbhs masses in nearby inactive galaxies .
however , it is much more difficult for the case of agns . with the broad emission lines from broad - line regions ( blrs)(e.g .
h@xmath10 , mg ii , civ ; h@xmath11 ) , the reverberation mapping method and the empirical size - luminosity relation can be used to derive the virial smbhs masses in agns ( kaspi et al .
2000 ; vestergaard 2002 ; mclure & jarvis 2002 ; wu et al . 2004 ; greene & ho 2006a ) .
it has been found that nearby galaxies and agns follow the same tight correlation between the central smbhs masses ( @xmath12 ) and stellar bulge velocity dispersion ( @xmath1 ) ( the @xmath13 relation ) ( nelson et al .
2001 ; tremaine et al .
2002 ; greene & ho 2006a , 2006b ) , which also implied that the mass from reverberation mapping method is reliable . according to agns unification model ( e.g. antonucci 1993 ; urry & padovani 1995 ) , agns can be classified into two classes depending on whether the central engine and blrs are viewed directly ( type i agns ) or are obscured by circumnuclear medium ( type ii agns ) . in type
i agns , by using the broad emission lines from blrs ( the reverberation mapping method or the empirical size - luminosity relation ) , we can derive virial smbhs masses .
it is not easy to study their host galaxies because their optical spectra are dominated by the non - stellar emission from the central agns activity .
this is especially true for luminous agns , where the continuum radiation from central source outshines the stellar light from the host galaxy . in type
ii agns , the obscuration of blrs makes both the reverberation mapping method and the empirical size - luminosity relation infeasible to derive smbhs masses .
however , we can use the well - known @xmath13 relation to derive smbhs masses if we can accurately measure the stellar bulge velocity dispersion ( @xmath1 ) .
there are mainly two different techniques to measure @xmath1 , one is the `` fourier - fitting '' method ( sargent et al . 1977 ; tonry & davis 1979 ) , the other is the `` direct - fitting '' method ( rix & white 1992 ; greene & ho 2006b and reference therein ) .
these years it has been successful to derive @xmath1 through fitting the observed stellar absorption features , such as ca h+k @xmath14 3969 , 3934 , mg ib@xmath14 5167 , 5173 , 5184 triplet , and ca
ii@xmath15 8498 , 8542 , 8662 triplet , etc , with the combination of different stellar template spectra broadened by a gaussian kernel ( e.g. kauffmann et al .
2003 ; cid fernandes et al .
2004a ; greene & ho 2006b )
. on the other hand , nelson & whittle ( 1996 ) find that the gaseous velocity dispersion ( @xmath6 ) of [ o iii]@xmath165007 from the narrow - line regions ( nlrs ) is nearly the same as @xmath1 for a sample of 66 seyfert galaxies , and suggest that the gaseous kinematics of nlrs be primarily governed by the bulge gravitational potential .
nelson ( 2001 ) find a relation between @xmath12 and @xmath17}$ ] ( the [ o iii]@xmath165007 velocity dispersion ) for agns , very similar to the relation of @xmath13 , although with more scatter , which strongly suggests that @xmath6 can be used as a proxy for @xmath1 . for lower - redshift type ii agns with @xmath18 , kauffmann et al .
( 2003 ) have investigated the properties of their hosts from the sloan digital sky survey ( sdss ) data release one ( dr1 ) , measured @xmath1 and estimated the smbhs masses from @xmath1 ( brinchmann et al .
2004 ) . by using this sample , greene & ho ( 2005 ) have measured the gaseous velocity dispersion ( @xmath19 ) from multiple transitions ( [ o ii ] @xmath163727 , [ o iii ] @xmath165007 , and [ s ii ] @xmath56716 , 6731 ) and compared @xmath1 and @xmath6 .
they find that @xmath6 from these multiple transitions trace @xmath1 very well , although some emission features they show considerable scatters .
type ii quasars are the luminous analogs of low - luminosity type ii agns ( such as seyfert 2 galaxies ) .
the obscuration of blrs makes quasars appear to be type ii quasars ( obscured quasars ) .
some methods have been used to discover type ii quasars , but only a handful have been found .
recently , zakamsa et al .
( 2003 ) present a sample of 291 type ii agns at redshifts @xmath20 from the sdss spectroscopic data .
about half are type ii quasars if we use the [ o iii ] @xmath165007 line luminosity to represent the strength of the nuclear activity .
what is the @xmath21 relation for type ii quasars ? and what are their smbhs masses and the eddington ratios @xmath22 ( i.e. the bolometric luminosity as a fraction of the eddington luminosity ) ?
here we use the sample of zakamsa et al .
( 2003 ) to study these questions for type ii quasars . in section 2 ,
we introduce the data and the analysis .
our results and discussion are given in sec .
3 . all of the cosmological calculations in this paper assume @xmath23 , @xmath24 , @xmath25 .
with sdss , zakamsa et al .
( 2003 ) presented a sample of 291 type ii agns at redshifts @xmath26 .
we downloaded these spectra from sdss data release four ( dr4 ) and the spectra for 202 type ii agns at redshifts @xmath26 are obtained .
sdss spectra cover 3800 - 9200 @xmath27 , with a resolution ( @xmath28 ) of @xmath29 and sampling of 2.4 pixels per resolution element .
the fibers in the sdss spectroscopic survey have a diameter of 3 " on the sky , for our type ii agns sample at redshifts @xmath26 , the projected fiber aperture diameter typically contains about 90% of the total host galaxy light ( kauffmann & heckman 2005 ) , thus it is feasible to observe significant stellar absorption features , which is the key point to accurately measure the stellar velocity dispersion ( @xmath1 ) in this paper . since the sdss spectra measure the light within a fixed aperture size
, the estimated velocity dispersions of more distant galaxies could be affected by the rotation of stars at larger physical radii than for nearby galaxies . in order to check the rotation contribution to line broadening
, we examine the relation between the stellar velocity dispersion and the redshift , and find no correlation at all , which suggests that the line - broadening contribution from rotation of host galaxies is very small and negligible .
the main reason is that @xmath1 is derived through fitting heavy - element absorption lines ( such as mg ib@xmath16 5173 , and caii k @xmath16 3934 , etc . ) , which are mainly from the old stellar population in the bulge .
we first modelled the stellar contribution in the sdss spectra of type ii agns through the modified version of the stellar population synthesis code , starlight ( version 2.0 , cid fernandes et al .
2001 ; cid fernandes et al .
2004a ; cid fernandes et al .
2004b ; garcia - rissman et al .
2005 ) , which adopted the new stellar library from bruzual & charlot ( 2003 ) .
the code does a search for the linear combination of simple stellar populations ( ssp ) to match a given observed spectrum @xmath30 .
the model spectrum @xmath31 is : @xmath32 \otimes g(v_\star,\sigma_\star)\ ] ] where @xmath33 is the spectrum of the @xmath34 ssp normalized at @xmath35 , @xmath36 is the reddening term , @xmath37 is the population vector , @xmath38 is the synthetic flux at the normalization wavelength , and @xmath39 is the line - of - sight stellar velocity distribution , modelled as a gaussian centered at velocity @xmath40 and broadened by @xmath41 .
the match between model and observed spectra is calculated by @xmath42 ^ 2 $ ] , where the weighted spectrum @xmath43 is defined as the inverse of the noise in @xmath30 . for more detail
, please refer to cid fernandes et al .
( 2005 ) .
prior to the synthesis , the galactic extinction is computed by using the extinction law of cardelli , clayton & mathis ( 1989 ) and the @xmath44 value is taken from schlegel , finkbeiner & davis ( 1998 ) as listed in the nasa / ipac extragalactic database ( ned ) .
the spectra are transformed into the rest frame defined by the redshift given in their sdss fits header .
the spectrum is normalized at 4020@xmath27 and the signal - to - noise ratio is measured in the s / n window between 4730 and 4780 @xmath27 .
masks of @xmath45 @xmath27 around obvious emission lines are constructed for each object individually .
because the redshift coverage of this type ii agns sample , we focus on the strongest stellar absorption features of ca ii k and the g - band , which are less affected by nearby emission lines .
an additional @xmath46 power - law component ( watanabe et al . 2003 ) is used to account for the contribution from the scattered agns continuum emission , a traditional ingredient for modeling seyfert galaxies since koski ( 1978 ) .
finally we check visually our spectral fitting results one by one . for our sample ,
the s / n in the s / n window varies between 0.3 and 21.5 .
the fitting results for high s / n objects are usually better than those for low s / n ones . after inspecting the fitting results , we find that the fitting goodness ( chi - square value ) depends not only on the s / n ( @xmath47 , in the given s / n spectral window ) , but also on the absorption lines equivalent widths ( ew of ca ii k line @xmath48 ) .
at last we select 33 type ii agns , which had shown significant stellar absorption features and are well fitted to derive reliable measurements of stellar velocity dispersion . in order to check whether this sub - sample is representative of the total sample of zakamsa et al .
( 2003 ) with respect to the [ o iii]@xmath165007 luminosity ( @xmath2}$ ] ) , we plot the histograms of the @xmath49}$ ] distribution for the sub - sample and total - sample ( see figure 1 ) . then t - test shows that these two populations are drawn from the same parent population with a possibility of 0.95 .
@xmath2}$ ] is directly adopted from table 1 in zakamsa et al .
( 2003 ) , the adopted raw @xmath49}$ ] values and associated quantities are lower limits . using the @xmath2}$ ] criterion of @xmath50(the logarithm is @xmath51 ) , which corresponds to the intrinsic absolute magnitude @xmath52 ( zakamsa et al .
2003 ) , 20 objects can then be classified as type ii quasars .
2 shows a fitting example of for sdss j150117.96 + 545518.2 with s / n=20.5 .
the final results are presented in table 1 .
all the fittings for 33 type ii agns are appended in the appendix . after subtracting the synthetic stellar components and the agns continuum
, we obtain the clean pure emission - line spectra as shown in the top panel of fig .
2 , where we can analyze the pure emission - line profiles in detail by using the multi - component spectral fitting task specfit ( kriss 1994 ) in the iraf - stsdas package . because of the asymmetric profiles of the [ o iii]@xmath54959 , 5007 lines , two sets of two - gaussian profiles are used in order to remove properly the asymmetric blue / red wings of the [ o iii ] line .
we take the same linewidth for each component , and fix the flux ratio of [ o iii]@xmath164959 to [ o iii]@xmath165007 to be 1:3 . for three objects
, we ca nt fit the [ o iii ] lines for their irregular [ o iii ] lines ( see table 1 ) .
for the [ o ii]@xmath5 3727 , 3729 lines , we use two - gaussian profiles and fix their wavelength separation to the laboratory value ; the ratio of the line intensities is allowed to vary during the fitting .
the decomposition for [ o ii ] lines is more difficult because of relatively low s / n and that the expected line widths are comparable to the pair separation ( 2.4 ) . for objects with available h@xmath11 , we also used one - gaussian profile to fit nii @xmath14 6548,6583 and h@xmath11 6563 lines . and
h@xmath11 and h@xmath11 luminosities would be used to do the extinction correction . for more details
, please refer to bian , yuan & zhao ( 2005 , 2006 ) .
our sample fitting for sdss j150117.96 + 545518.2 is showed in fig .
the derived stellar velocity dispersion was corrected by the instrumental resolutions of both the sdss spectra and the stelib library .
cid fernandes et al .
( 2004a ) have used their stellar population synthesis method to study a sample of 79 nearby galaxies observable from the southern hemisphere , of which 65 are seyfert 2 galaxies .
the s / n in the s / n window varies between 10 and 67 .
they compared their @xmath1 with values from the literature and found the agreement is good .
they estimated that the uncertainty in @xmath1 is typically about 20 km s@xmath53 .
recently , cid fernandes et al .
( 2005 ) applied their synthesis method to a larger sample of 50362 galaxies from the sdss data release 2 ( dr2 ) .
their derived @xmath1 is consistent very well with that of the mpa / jhu group ( kauffmann et al .
the median of the difference between the two estimates is just 9 km s@xmath53 .
we have carefully checked our synthesis fitting result one by one and picked out 33 type ii agns that are well fitted and the stellar velocity dispersion ( @xmath1 ) are reliably derived .
the spectral s / n for these objects are in the range of 5 to 21.5 , most of which larger than 10 , thus the typical uncertainty in @xmath1 should be around 20 km s@xmath53 .
recently , greene & ho ( 2006a , 2006b ) used the direct - fitting method ( barth et al .
2002 ) to study the the systematic biases of @xmath1 from different regions around ca ii triplet , mg ib triplet , and caii h+k stellar absorption features , which are introduced by both template mismatch and contamination from agns .
they argue that the ca ii triplet provides the most reliable measurements of @xmath1 and there is a systematic offset between @xmath54 and @xmath1 derived from other spectral regions . for our higher - redhsift sample and the sdss wavelength coverage 3800 - 9200 , it is impossible to measure @xmath1 from ca ii triplet .
therefore , for higher - redhsift type ii agns , new observations around ca ii triplet are necessary in the future . here
we used the following formula to obtain the corrected velocity dispersion @xmath4 ( greene & ho 2006b ) , @xmath55 for three objects , @xmath56 is near the instrumental resolution and the corrected @xmath57 is unreliable .
these objects are excluded from further analysis , and denoted as @xmath58 in table 1 . the gaseous velocity dispersion ( @xmath19 ) is obtained from full width half maximum ( fwhm ) of emission lines by assuming the gaussian profile : @xmath59}^{g}=fwhm[o ii]/2.35 $ ] and @xmath60}^{g}=fwhm[o iii]/2.35 $ ] . taking into account the sdss spectrum resolution ,
the intrinsic @xmath6 derived from @xmath61)$ ] may be instrumentally broadened .
the intrinsic @xmath19 value can be approximated by @xmath62 , where z is the redshift .
for the spectra from sdss , the mean values of @xmath63 are 74 km s@xmath53(the logarithm is 1.87 dex ) for [ o ii ] , and 60 km s@xmath53(the logarithm is 1.78 dex ) for [ o iii ] ( greene & ho 2005 ) , respectively .
the results of @xmath19 are listed in table 1 ( columns 7 and 8) . after removing the effect of instrumental broadening ,
some objects become unresolved ( see table 1 ) .
measurements of @xmath19 below the resolution limit ( 74 and 60 km s@xmath53for [ o ii ] and [ o iii ] , respectively ) are not reliable .
the error of @xmath19 is derived from the measurement error of the linewidth . for the linewidth ,
the typical error is about 10 per cent .
however , the systematic effects are neglected , e.g. the uncertainties of the continuum subtraction and component decomposition ( bian , yuan & zhao 2005 ) .
the [ o iii ] line has the advantage of being strong and easy to detect in agns .
the [ o iii]@xmath165007 line is usually used as a tracer of agns activity ( e.g. heckman et al .
, 2004 ; greene & ho 2005 ) .
however , the [ o iii]@xmath165007 luminosity is subject to extinction by interstellar dust in the host galaxy and in our galaxy . using sdss data ,
kauffmann et al ( 2003 ) found that for type ii agns at @xmath64 the mean value of the ratio of the corrected luminosity to the observational luminosity ( @xmath65}}{l_{[o iii]}})$ ] , the extinction factor at 5007@xmath27 ) is about 0.85 dex and suggested that the extinction effect should be huge for type ii quasars .
the extinction correction is usually corrected by using the balmer decrement , which is regarded as the best approximation . for 18 out of these 33 objects with available h@xmath11 measurements in the stellar - light subtracted spectra , we calculated the extinction correction of the [ o iii]@xmath165007 luminosity , from the relation ( bassani et al .
1999 ) @xmath66}^{\rm c } = l_{\rm [ o iii ] } \rm [ \frac{(h_\alpha / h_\beta)_{obs}}{(h_\alpha / h_\beta)_{0}}]^{2.94}\ ] ] where an intrinsic balmer decrement @xmath67 is adopted ( gu & huang 2002 ) .
the range of the extinction factors at [ oiii ] is between -0.24 dex and 1.25 dex . the extinction - corrected [
o iii]@xmath165007 luminosity is listed in col . 11 in table 1 . if we used the @xmath2}$ ] criterion of @xmath50 ( zakamsa et al .
2003 ) , 14 out of these 18 objects can then be classified as type ii quasars . using the sdss agn catalogue at mpa / jhu ( kauffmann et al 2003 ) , we obtained the extinction factor at 5007@xmath27 , @xmath65}}{l_{[o iii}})$ ] .
we tried to obtain a relation between @xmath68}}{l_{[o iii}}$ ] and other observational parameters , such as @xmath56 , @xmath69 $ ] , @xmath70 .
however , no correlation is found .
more careful work is required to deal with this problem in the future . using the reverberation mapping method or the empirical size - luminosity relation , it is impossible to estimate the smbhs masses in type ii quasars for the lack of emission lines from blrs .
here we use the formulae to derive the smbhs masses in type ii quasars from the stellar velocity dispersion @xmath1 ( tremaine et al .
2002 ) , which is @xmath71 .
we calculate the eddington ratio , @xmath22 .
we use the extinction - corrected [ o iii ] luminosity ( @xmath72 ) as a surrogate for the agns luminosity ( heckman et al . , 2004 ;
greene & ho 2005 ) , @xmath73}$ ] , to calculate the bolometric luminosity @xmath74 , and @xmath75 .
objects with no extinction corrected [ o iii ] luminosity are discarded in the analysis relative to the eddington ratio .
the results of smbhs masses and eddington ratios in type ii agns are also presented in table 1 ( columns 13 and 14 ) .
we also calculate the smbhs masses and eddington ratios for the lower - redshifts type ii agns at @xmath76 presented by kauffmann et al .
( 2003 ) . for the typical uncertainties of 20 km s@xmath53for @xmath77 km s@xmath53
, the errors of @xmath78 would be about 0.05 dex , corresponding 0.17 dex for @xmath79 , and almost the same for @xmath22 ( bian , yuan , & zhao 2005 ) .
here we do nt consider the cosmology evolution of @xmath13 relation ( e.g. woo et al .
2006 ) , which is a question open to debate . in fig .
4 , we compare the distribution of smbhs masses and eddington ratios in the lower - redshifts sample and the higher - redshifts sample . it is found that the type ii agns at higher - redshifts have higher smbhs masses and higher eddington ratios .
it is well known that the eddington ratio is an important parameter to describe the accretion process in agns .
the [ o iii ] luminosity is usually used as a surrogate for the agns luminosity ( heckman et al .
2004 and the reference therein ) . in fig .
5 , we plot the [ o iii ] luminosity and the corrected [ o iii ] luminosity versus the smbhs masses . using a least - squares regression
, we derive the correlation between @xmath12 and @xmath2}$ ] to be : @xmath80}/l_{\odot})=(6.83\pm0.65)+(0.22\pm0.08)log(m_{bh } /m_{\odot})$ ] .
the correlation coefficient r is 0.45 , with a probability of @xmath81 for rejecting the null hypothesis of no correlation .
however , no correlation is found between @xmath12 and @xmath82}$ ] .
no correlation is found between the eddington ratio and the [ o iii ] luminosity or the corrected [ o iii ] luminosity . from the photoionization model ,
the strength of [ o iii ] is controlled by the nlrs covering factor , density , and ionization parameter ( e.g. baskin & laro 2005 ) .
the relation between the [ o iii ] luminosity and the agns bolometric luminosity is still a question to debate .
the existence of a good correlation between stellar velocity dispersion ( @xmath1 ) and ionized gas velocity dispersion ( @xmath19 ) ( e.g. nelson & whittle 1996 ) suggests that the gaseous kinematics of nlrs in seyfert galaxies be primarily dominated by the bulge gravitational potential , which is further confirmed by nelson & whittle ( 1996 ) .
most recently , greene & ho ( 2005 ) have investigated a large and homogeneous sample of lower - redshift ( 0.02 @xmath84 z @xmath84 0.3 ) type ii agns from the sdss , and found that @xmath19 traces @xmath1 .
though the gas kinematics of nlrs are governed by the gravitational potential of the bulges , they also find that the accretion rate plays an important secondary role .
following greene & ho ( 2005 ) , we study the @xmath85 relation for 33 type ii agns at redshifts 0.3 @xmath84 z @xmath84 0.83 after measuring the gas velocity dispersion ( @xmath19 ) from the narrow emission lines from nlrs , and the stellar velocity dispersion ( @xmath86 ) from the caii h+k , g - band absorption feature , which is shown in fig .
2 . in fig . 6 , we showed the relation between @xmath6 and @xmath4 .
the correlation coefficients are 0.19 , 0.27 , 0.70 , 0.54 for figurs in fig .
6 , from left to right and from top to bottom .
the relation between @xmath6 and @xmath4 becomes much weaker at higher redshifts .
we also found that this poor correlation is not due to the s / n . in order to qualify the comparison between @xmath19 and @xmath86
, we calculate the distribution of @xmath87 .
the value is @xmath8 for the the core component of [ o iii ] line , @xmath9 for the [ o ii ] line .
these suggest that @xmath88 of the the core component of [ o iii ] and [ o ii ] lines can trace @xmath86 within 0.09 and 0.08 dex in the logarithm of @xmath86 , respectively , and that the high - ionization [ o iii ] line traces @xmath86 as well as the low - ionization [ o ii ] line .
if we use the line width of [ o iii ] core component to estimate the smbhs masses from the tremaine s m@xmath89 relation , we would overestimate smbhs masses by about 0.38 dex . for comparison ,
greene & ho ( 2005 ) derived the distribution of @xmath90 of lower - redshift type ii agns with @xmath76 , which are @xmath91 for the [ o iii ] line , @xmath92 for the [ o iii ] core line , and @xmath93 for the [ o ii ] line .
we also calculate the the distribution of @xmath87 for the sample of seyfert galaxies ( nelson & whittle 1996 ) ; the value is @xmath94 , suggesting that @xmath19 of the [ o iii ] line can trace @xmath86 within 0.06 dex in the logarithm of @xmath86 .
our results are thus consistent with theirs . to a first - order approximation , the line widths of the core component of [ o iii ] and [ o ii ] for both low - redshift and high - redshift
type ii agns are primarily controlled by the gravitational potential of the bulges of host galaxies , and can approximately trace the stellar velocity dispersion . as we know , the errors in virial smbhs masses derived from galaxy dynamics or size - luminosity relation is about 0.5 dex ( e.g. magorrian et al . 1998 ; kaspi et al .
2000 ; bian & zhao 2004a , 2004b ) .
we also can use the line - width of the core component of [ o iii ] or [ o ii ] to estimate the black hole mass . in order to find the secondary effect of the line broadening in gas lines from nlrs for our higher - redshift narrow - line agns
, we studied the relation between the deviation of @xmath19 from @xmath86 ( @xmath95 ) and the eddington ratio ( @xmath96 ) ( greene & ho 2005 ) .
using the least - square regression , we find a median strong correlation between these two dimensionless parameters .
for the [ o iii ] line , the relation is : @xmath97 , r=0.71 , @xmath98 . for the [ o ii ]
line , the relation is @xmath99 , r=0.65 , @xmath100 ( see fig .
we also find the comparable correlation between @xmath101 and @xmath102 .
these results confirm that the nuclei accretion process and/or nuclei smbhs would affect the linewidth of gas lines from nlrs , although the primary driver is the gravitational potential of the bulge .
the stellar population synthesis code is used to model the stellar contribution for a sample of 209 type ii agns at redshifts @xmath0 from sdss . according to the @xmath2}$ ] criterion of @xmath50 ,
20 can be classified as type ii quasars .
the main conclusions can be summarized as follows . *
the reliable @xmath1 are measured for 33 type ii agns with significant stellar absorption features .
we use the formula of greene & ho to obtain the corrected stellar velocity dispersions ( @xmath4 and smbhs masses are calculated from the @xmath103 relation . a median strong relation between the [ o iii ] luminosity and the smbh mass is found ( although no correlation between the extinction - corrected [ o iii ] luminosity and the smbh mass ) ; no correlation is found between the eddington ratio and the [ o iii ] luminosity or the extinction - corrected [ o iii ] luminosity .
* the gas velocity dispersion ( @xmath6 ) in nlrs is measured using three sets of two - gaussian profiles to fit [ o iii]@xmath54959 , 5007 and [ o ii]@xmath53727 , 3729 ) in these 33 stellar - light subtracted spectra .
we find that the relation between @xmath6 and @xmath4 becomes much weaker at higher redshifts .
* the distribution of @xmath7 is @xmath8 for the core [ o iii ] line and @xmath9 for the [ o ii ] line , which suggests that @xmath6 can trace @xmath4 within about 0.1 dex in the logarithm of @xmath4 .
the deviation of @xmath6 from @xmath4 is correlated with the eddington ratio .
we thank luis c. ho for his helpful comments .
we thank the anonymous referee for his / her comments and instructive suggestions , which significantly improved our work .
we are grateful to dr .
helmut abt for checking our manuscript .
this work has been supported by the nsfc ( nos .
10403005 and 10473005 ) and the science - technology key foundation from education department of china ( no .
qgu would like to acknowledge the financial supports from china scholarship council ( csc ) and the national natural science foundation of china under grants 10103001 and 10221001 .
funding for the creation and distribution of the sdss archive has been provided by the alfred p. sloan foundation , the participating institutions , nasa , the national science foundation , the us department of energy , the japanese monbukagakusho , and the max planck society .
the sdss web site is http:// www.sdss.org/. this research has made use of the nasa / ipac extragalactic database , which is operated by the jet propulsion laboratory at caltech , under contract with nasa .
antonucci r. , 1993 , ara&a , 31 , 473 barth a. j. , ho l. c. , & sargent w. l. w. , 2002 , aj , 124 , 2607 begelman m. c. , 2003 , science , 300 , 1898 baskin a. , laro a. , 2005 , mnras , 358 , 1043 bassani l. , et al . , 1999 , apjs , 121 , 473 bian w. , zhao y. , 2004a , mnras , 347 , 607 bian w. , zhao y. , 2004b , mnras , 352 , 823 bian w. , yuan q. , zhao y. , 2005 , mnras , 364 , 187 bian w. , yuan q. , zhao y. , 2006 , mnras , 367 , 860 brinchmann j. , et al . , 2004 , apj , 613 , 109 bruzual g. , charlot s. , 2003 , mnras , 344 , 1000 cardelli j. a. , clayton g. c. , mathis j. s. , 1989 , apj , 345 , 245 cid fernandes r. , sodre l. , schmitt h. r. , leao j. r. s. , 2001 , mnras , 325 , 60 cid fernandes r. , gu q. , melnick j. , et al . , 2004a , mnras , 355 , 273 cid fernandes r. , gonzalez delgado r. , storchi - bergmann , t. , pires martins , l. , schmitt h. , 2004b , mnras , 356 , 270 cid fernandes r. , mateus a. , sodre l. , stasinska g. , gomes j. , 2005 , mnras , 358 , 363 ferrarese , l. , merritt d. 2000 , apj , 539 , l9 gebhardt k. , et al .
2000 , apj , 539 , l13 greene j. e. , ho l. c. , 2005 , apj , 627 , 721 greene j. e. , ho l. c. , 2006a , apjl , 641 , l21 greene j. e. , ho l. c. , 2006b , apj , 641 , 117 gu q. , huang j. , 2002 , apj , 579 , 205 gu q. , melnick j. , cid fernandes r. , kunth d. , terlevich , terlevich r. , et al .
2006 , mnras , 366 , 480 heckman t. m. , et al . , 2004 , apj , 613 , 109 kaspi s. , smith p.s . , netzer h. , maoz d. , jannuzi b.t . , giveon u. , 2000 , apj , 533 , 631 kauffmann g. , et al .
2003 , mnras , 346 , 1055 kauffmann , g. , & heckman , t. m. 2005 , philos .
london , a363 , 621 koski a. t. , 1978 , apj , 223 , 56 kriss g. a. 1994 , in asp conf .
61 , astronomical data analysis software and systems iii , ed .
crabtree d. r. , hanisch r. j. , barnes j. ( san francisco : asp ) , 437 maggorian j. , et al .
, 1998 , aj , 115 , 2285 mclure r. j. , jarvis m. j. , 2002 , mnras , 337 , 109 nelson c. h. , 2001 , apj , 544 , l91 nelson c. h. , whittle m. , 1996 , apj , 465 , 96 rix h. w. , & white s. d. m. 1992 , mnras , 254 , 389 sargent w. l. w. , schechter p. l. , boksenberg a. , & shortridge k. 1977 , apj , 212 , 326 schlegel d. j. , finkbeiner d. p. , davis m. , 1998 , apj , 500 , 525 shen z. q. , et al . , nature , 2005 , 438 , 62 tonry j. , & davis m. , 1979 , apj , 84 , 1511 tremaine s. , et al . , 2002 , ap j , 574 , 740 urry c. m. , padovani p. , 1995
, pasp , 107 , 803 vestergaard m. , 2002 , apj , 571 , 733 watanabe m. , et al . , 2003 , apj , 591 , 714 woo j. h. , et al . , 2006 ,
apj , in press , astro - ph/0603648 wu x. b. , et al . , 2004 , a & a , 424 , 793 zakamsa n. l. , et al .
, 2003 , aj , 126 , 2125 | we apply the stellar population synthesis code by cid fernandes et al .
to model the stellar contribution for a sample of 209 type ii agns at @xmath0 from the sloan digital sky survey .
the reliable stellar velocity dispersions ( @xmath1 ) are obtained for 33 type ii agns with significant stellar absorption features . according to the @xmath2}$ ] criterion of @xmath3 , 20 of which
can be classified as type ii quasars .
we use the formula of greene & ho to obtain the corrected stellar velocity dispersions ( @xmath4 ) .
we also calculate the supermassive black holes masses from @xmath4 in these high - redshift type ii agns .
the [ o iii ] luminosity is correlated with the black hole mass ( although no correlation between the extinction - corrected [ o iii ] luminosity and the black hole mass ) , and no correlation is found between the eddington ratio and the [ o iii ] luminosity or the corrected [ o iii ] luminosity .
three sets of two - component profile are used to fit multiple emission transitions ( [ o iii]@xmath54959 , 5007 and [ o ii]@xmath53727 , 3729 ) in these 33 stellar - light subtracted spectra .
we also measure the gas velocity dispersion ( @xmath6 ) from these multiple transitions , and find that the relation between @xmath6 and @xmath4 becomes much weaker at higher redshifts than in smaller redshifts .
the distribution of @xmath7 is @xmath8 for the core [ o iii ] line and @xmath9 for the [ o ii ] line , which suggests that @xmath6 of the core [ o iii ] and [ o ii ] lines can trace @xmath4 within about 0.1 dex in the logarithm of @xmath4 . for the secondary driver , we find that the deviation of @xmath6 from @xmath4 is correlated with the eddington ratio .
galaxies : active galaxies : nuclei quasars : emission lines |
a reliable method of measuring the physical distance to astrophysical objects has always been sought after in observational astronomy ( e.g. , * ? ? ? * ) . in the era of `` precision cosmology '' , the need for accurate physical distance measurements has been amplified ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
accurate and precise distance indicators hold the key to pinning down the value of the hubble constant @xmath6 and many other cosmological parameters ( see discussion in , e.g. , * ? ? ?
a number of methods have been employed to determine extragalactic distances , with varying degree of success ( e.g. , * ? ? ?
the construction and reliability of the `` cosmological distance ladder '' depends crucially on cepheid variables being able to provide precise and accurate distances to nearby ( @xmath7mpc ) galaxies .
the quest for such distances has been an arduous journey for almost a hundred years , with many dramatic twists and turns ( for a review of early years , see * ? ? ? * for a recent review see , e.g. , @xcite ) .
cepheids offer several advantages as distance indicators .
massive stars @xmath8 make an excursion through the instability strip and most , if not all , of them become cepheid variables .
these variable stars are relatively bright ( @xmath9 for a @xmath10day cepheid ) and often have large brightness variations ( amplitude @xmath11 mag ) with a characteristic `` saw - tooth '' shaped light curve .
their intrinsic brightness , combined with their light curve shape and colors , make them easy to distinguish from other classes of variable stars . as a result
, cepheids have been detected and studied in a significant number of star - forming galaxies .
the physical mechanisms underlying cepheid pulsation are well understood , including the observed tight period - luminosity ( pl ) relationship ( e.g. , * ? ? ?
the small scatter in the pl relation allows distance measurements precise to @xmath12 ( e.g. , * ? ? ?
* ) . for these reasons
, cepheids are commonly used to calibrate other distance indicators , forming the base of the cosmological distance ladder . despite their many advantages as a distance indicator
, cepheid distances also have some shortcomings .
most cepheids have an intrinsic brightness of @xmath13 , so with the current instrumentation they can be only used to measure distances to @xmath14mpc ( the largest cepheid distance in @xcite is @xmath15mpc ) .
observations of cepheids in distant galaxies are also hindered by blending @xcite as young stars , cepheids live in close proximity to the crowded star - forming regions of their host galaxies , and are thus likely to have another star of similar brightness on the scale of a typical instrumental point - spread - function ( psf ) .
the effect of blending becomes worse as the square of the distance to the host galaxy @xcite , again limiting the usefulness of cepheids to measuring distances @xmath14mpc even with high resolution instruments such as the hubble space telescopes ( _ hst _ ) .
ideally , we would like to find a distance indicator that shares the good properties of classical cepheids , but is even more luminous , allowing us to observe it further away and be less susceptible to blending .
in this paper we discuss such a possible distance indicator , namely ultra long period ( ulp ) cepheids .
we define ulp cepheids as fundamental mode cepheids with pulsation periods greater than 80 days .
several such cepheids have been already discovered in the pioneering study of @xcite .
however , ulp cepheids have traditionally been ignored for distance measurements as they are pl outliers .
indeed , the observed pl relation flattens for cepheids with periods greater than 100 days ( e.g. , * ? ? ?
* ; * ? ? ?
@xcite suggests that long period leavitt variables could be used for distance measures unfortunately that idea has not permeated through the community .
we argue that the flattening of the pl at long periods actually improves the usefulness of ulp cepheids as distance indicators because it makes them a good standard candle in the traditional sense .
we note several additional advantages of ulp cepheids over lower period cepheids due to their increased luminosity .
ulp cepheids could be used as a stellar distance measure to the hubble flow ( up to @xmath16 mpc) several times the current observational limit . in section [ sec : sample ] we describe our sample compiled from the literature .
the ulp cepheid pl relation is discussed in section [ sec : dm ] .
section [ sec : massloss ] demonstrates how ulp cepheids may provide the additional benefit of testing massive stellar evolutionary models .
we summarize our results in section [ sec : conclusion ] .
llccccccccccccc lmc hv883 & lmc & @xmath17 & @xmath18 & @xmath19 & @xmath20 & @xmath21 & @xmath22 & @xmath23 & @xmath24 & @xmath25 & @xmath26 & @xmath27 & @xmath28 & 1,2,3 + lmc hv2447 & lmc & @xmath29 & @xmath30 & @xmath31 & @xmath32 & @xmath33 & @xmath34 & @xmath23 & @xmath24 & @xmath35 & @xmath36 & @xmath37 & @xmath28 & 1,2,3 + lmc hv2883 & lmc & @xmath38 & @xmath39 & @xmath40 & @xmath41 & @xmath42 & @xmath43 & @xmath23 & @xmath24 & @xmath44 & @xmath45 & @xmath46 & @xmath28 & 1,2,3 + lmc hv5497 & lmc & @xmath47 & @xmath48 & @xmath49 & @xmath50 & @xmath51 & @xmath52 & @xmath23 & @xmath24 & @xmath53 & @xmath54 & @xmath55 & @xmath28 & 1,2,3 + smc hv1956 & smc & @xmath56 & @xmath57 & @xmath58 & @xmath59 & @xmath60 & @xmath61 & @xmath62 & @xmath63 & @xmath64 & @xmath65 & @xmath66 & @xmath67 & 5,2,3;4 + smc hv821 & smc & @xmath68 & @xmath69 & @xmath70 & @xmath50 & @xmath71 & @xmath72 & @xmath62 & @xmath63 & @xmath73 & @xmath74 & @xmath75 & @xmath67 & 1,6,3;4 + smc hv829 & smc & @xmath76 & @xmath77 & @xmath78 & @xmath79 & @xmath26 & @xmath80 & @xmath62 & @xmath63 & @xmath81 & @xmath82 & @xmath83 & @xmath67 & 1,6,3;4 + ngc 6822 - 1 & ngc 6822 & @xmath84 & @xmath85 & @xmath86 & @xmath87 & @xmath88 & @xmath89 & @xmath90 & @xmath91 & @xmath92 & @xmath36 & @xmath93 & @xmath94 & 7,8,9 + ngc 55 - 1 & ngc 55 & @xmath95 & @xmath96 & @xmath97 & @xmath98 & @xmath99 & @xmath100 & @xmath101 & @xmath102 & @xmath103 & @xmath104 & @xmath105 & @xmath106 & 10,11,12 + ngc 55 - 2 & ngc 55 & @xmath107 & @xmath108 & @xmath109 & @xmath110 & @xmath111 & @xmath112 & @xmath101 & @xmath102 & @xmath113 & @xmath114 & @xmath115 & @xmath106 & 10,11,12 + ngc 55 - 3 & ngc 55 & @xmath116 & @xmath117 & @xmath118 & @xmath119 & @xmath120 & @xmath121 & @xmath101 & @xmath102 & @xmath122 & @xmath123 & @xmath124 & @xmath106 & 10,11,12 + ngc 55 - 4 & ngc 55 & @xmath125 & @xmath126 & @xmath127 & @xmath128 & @xmath129 & @xmath130 & @xmath101 & @xmath102 & @xmath131 & @xmath132 & @xmath133 & @xmath106 & 10,11,12 + ngc 55 - 5 & ngc 55 & @xmath134 & @xmath135 & @xmath136 & @xmath137 & @xmath138 & @xmath139 & @xmath101 & @xmath102 & @xmath140 & @xmath141 & @xmath142 & @xmath106 & 10,11,12 + ngc 300 - 1 & ngc 300 & @xmath143 & @xmath144 & @xmath145 & @xmath146 & @xmath147 & @xmath148 & @xmath149 & @xmath150 & @xmath151 & @xmath152 & @xmath153 & @xmath154 & 13,14,15 + ngc 300 - 2 & ngc 300 & @xmath155 & @xmath156 & @xmath157 & @xmath158 & @xmath159 & @xmath160 & @xmath149 & @xmath150 & @xmath161 & @xmath45 & @xmath162 & @xmath154 & 13,14,15 + ngc 300 - 3 & ngc 300 &
@xmath163 & @xmath164 & @xmath165 & @xmath166 & @xmath167 & @xmath168 & @xmath149 & @xmath150 & @xmath169 & @xmath170 & @xmath171 & @xmath154 & 13,14,15 + i zw 18 - 1 & i zw 18 & & & @xmath172 & @xmath173 & @xmath174 & @xmath175 & @xmath176 & @xmath177 & @xmath178 & @xmath179 & @xmath180 & @xmath181 & 16,17 + i zw 18 - 2 & i zw 18 & & & @xmath182 & @xmath183 & @xmath26 & @xmath184 & @xmath176 & @xmath177 & @xmath185 & @xmath186 & @xmath187 & @xmath181 & 16,17 + we have assembled a sample of ulp cepheids from the literature and list their reported positions , periods , and mean @xmath188 and @xmath189 magnitudes ( see table [ tab : ceps ] ) .
we adopt the periods , reddening values , and distance moduli found in these sources except in the case of the magellanic clouds ( see below ) .
our primary criteria for selecting the sample was the existence of @xmath188 and @xmath189 data calibrated on the standard johnson / kron - cousins magnitude system using landolt standards ( with the possible exception of the magellanic clouds ; see below ) .
the ulp distinction is applied to fundamental mode cepheids with periods greater 80 days .
we combed the recent literature for reports of such variable stars . * magellanic clouds * our sample includes four lmc and three smc ulp cepheids .
@xcite calibrated photoelectric observations of these cepheids and transformed them to the johnson / kron - cousins standard system .
the mean flux - weighted photometry for the six cepheids reported in @xcite agrees with the landolt standard star calibrated results of @xcite to within @xmath190 mag , suggesting that the standard photometric system calibration is robust .
mean flux - weighted photometry for hv1956 is obtained from the all sky automated survey ( asas ; * ? ? ?
* ) and @xcite .
the @xmath188 light curves of six of these ulp cepheids were obtained from asas .
hv2883 was not targeted by asas , and its @xmath188 light curve photometry was obtained from @xcite , @xcite , and @xcite .
@xcite provide the @xmath189 light curve data for the entire sample .
we applied the analysis of variance technique @xcite to the seven harvard variable light curves in figure [ fig : lcs ] to obtain the periods listed in table [ tab : ceps ] .
we adopt total reddening values of @xmath191 mag and @xmath192 mag for the lmc and smc , respectively @xcite .
we assume a distance modulus ( dm ) of @xmath193 mag to the lmc for consistency with the sources listed in table [ tab : ceps ] .
the smc dm used is @xmath194 mag @xcite .
the lmc ( smc ) hosts ulp cepheids with periods of 98.6 , 109.2 , 118.7 , and 133.6 ( 84.4 , 127.5 , and 210.4 ) days .
the lmc has gas phase oxygen abundance @xmath195o / h@xmath196 @xcite while the smc is @xmath195o / h@xmath197 @xcite .
the araucaria project @xcite is a photometric survey of local and sculptor group galaxies and their cepheid populations .
the primary goal is to more accurately determine the distances to these galaxies and to characterize the dependence of various stellar distance indicators on metallicity and age .
the araucaria project has observed ulp cepheids in the following galaxies .
* ngc 55 * five ulp cepheids were found in ngc 55 @xcite .
observations were taken with the optical gravitation lensing experiment ( ogle ) detector on the warsaw 1.3 m telescope at las campanas observatory .
they estimate that the calibration procedure used to transform their photometric data from the ogle filters to the standard system produced errors @xmath198 mag .
follow up observations in the ir revealed a total reddening of @xmath199 mag @xcite .
their multi - wavelength pl analysis produced a dm to ngc 55 of @xmath200 mag ( statistical and systematic errors , respectively ) .
ngc 55 hosts ulp cepheids with periods of 85.1 , 97.7 , 112.7 , 152.1 , and 175.9 days .
the oxygen abundance of ngc 55 is @xmath195o / h@xmath201 @xcite . *
ngc 6822 * one ulp cepheid was found in ngc 6822 @xcite .
the filters and telescope used are identical to those of ngc 55 @xcite . similarly , the reported calibration error onto the standard system is @xmath198 mag . as in the multi - wavelength
follow up study of ngc 6822 @xcite , we adopt a total reddening of @xmath202 mag .
the lone ulp cepheid in ngc 6822 has a period of 123.9 days .
@xcite calculate a dm to ngc 6822 of @xmath203 mag ( statistical and systematic errors , respectively ) .
ngc 6822 has a similar oxygen abundance to ngc 55 of @xmath195o / h@xmath204 @xcite .
* ngc 300 * @xcite found three ulp cepheids in ngc 300 .
again , ogle filters were used for the observations .
their calibration onto the standard system has a reported error @xmath198 mag . a multi - wavelength study of ngc 300 @xcite determined a reddening - free dm of @xmath205 mag ( statistical and systematic , respectively ) using a total reddening of @xmath206 mag .
ulp cepheids of 83.0 , 89.1 , and 115.8 days are observed in ngc 300 .
ngc 300 has a strong metallicity gradient ; therefore we adopt mean cepheid radial distance of 4 kpc and apply the averaged gradient of @xcite to obtain a mean oxygen abundance value of @xmath207o / h@xmath208 . *
i zw 18 * @xcite discovered three ulp cepheids from the extremely metal poor galaxy i zw 18 , though they could not confirm one candidate .
a follow - up study @xcite presents flux weighted mean photometry but no data , so the light curves of these objects could not be included in figure [ fig : lcs ] .
the ulp cepheids have periods of @xmath172 and @xmath182 days ( table [ tab : ceps ] ) .
after accounting for extinction @xmath209 mag , @xcite use the red giant branch tip to determine a dm of @xmath210 mag while @xcite find a dm of @xmath211 mag via pulsation models .
we use the former measurement as it is considered more reliable by the authors .
we do not include these two cepheids in the upcoming pl determination as i zw 18 is a full dex more metal poor than the other galaxies in this sample ( @xmath207o / h@xmath212 ; @xcite ) .
there is an increasing amount of support for a metallicity dependent pl ( e.g. * ? ? ?
* ) and including these cepheids in our pl analysis would greatly increase the metallicity dispersion of the host galaxies in our sample .
we do , however , make use of them to examine the ulp pl relation dependence on metallicity . the ulp cepheid sample and its mean , flux - weighted photometry in the standard system can be found in table [ tab : ceps ] .
we assume that the photometric error associated with each ulp cepheid is negligible when compared to the intrinsic scatter of the pl relation .
we transform these measurements to absolute magnitudes via : @xmath213 where @xmath214 is the absolute magnitude in either the @xmath188 or @xmath189 , @xmath215 is the apparent magnitude ; @xmath216 is the reddening free distance modulus ; and @xmath217 is the extinction in the @xmath188 or @xmath189 .
we use the extinction law @xmath218 and @xmath219 @xcite .
we define the wesenheit magnitudes as : w@xmath220 ( e.g. , * ? ? ?
the color - magnitude diagram ( cmd ) highlights several important characteristics of the ulp cepheid data set ( figure [ fig : cmd ] ) .
ulp cepheids are the luminous counterparts to shorter period cepheids in color magnitude space .
our sample clearly populates the luminous region of the instability strip .
future cepheid studies can use ulp cepheids to push cepheid distance measurements well beyond the current @xmath221 mpc limit as our sample s median absolute magnitude is m@xmath3(m@xmath222 ( see section [ sec : conclusion ] ) .
the intrinsic brightness of ulp cepheids makes them ideal candidates for distance indicators to galaxies where classical cepheids can not be observed .
the cepheids in our sample have been ignored as distance indicators as they do not follow to the standard period - luminosity relationship ( e.g. , * ? ? ?
* ) . in this section
, we examine the characteristics of our ulp cepheid sample in the period - luminosity plane and explore their viability as a distance indicator .
we determine pl relations of our sample in the @xmath188 , @xmath189 , and w@xmath3 in section [ sec : pl ] while the metallicity dependence of our results is presented in section [ sec : metal ] . using the data in table [ tab : ceps ]
, we construct pl diagrams in @xmath188 , @xmath189 , and w@xmath3 ( figures [ fig : pl_v ] , [ fig : pl_i ] , [ fig : pl ] , respectively ) . in each case , ulp cepheids are compared to fundamental mode smc cepheids ( hereafter , this control sample will be referred to as ogle cepheids ) .
the ogle cepheid sample contains over @xmath223 fundamental mode cepheids with periods ranging from @xmath224 to @xmath225 days ; however , we only plot the 70 cepheids with periods of greater than 10 days . to quantify our comparison ,
we perform a linear fit on both samples in each pl diagram : the slopes of the ogle cepheid pl relations ( hereafter pl@xmath226 ; dotted lines in the figures ) are adopted from @xcite while the intercepts are chosen to minimize chi - square .
we employ linear least square fitting of the ulp cepheid sample to identify their pl relation ( hereafter pl@xmath227 ) .
we omit errors in distance moduli and extinction in this demonstration as they are small when compared to the overall scatter of the sample .
the parameters of these fits and the rms of each data set in @xmath188 , @xmath189 , and w@xmath3 are listed in table [ tab : fits ] . despite our sample ranging in period from 83 to 210 days ,
ulp cepheids occupy a small region of luminosity space .
we note that the ulp cepheids from i zw 18 are not included in this analysis for reasons outlined in section [ sec : sample ] . the @xmath188 pl diagram is figure [ fig : pl_v ] .
the pl@xmath226 fit has a slope of @xmath228 and the rms of the ogle cepheid sample about this fit is @xmath229 mag .
the ulp cepheid sample has @xmath230 more scatter ( rms=@xmath231 mag ) about the pl@xmath226 fit .
this discrepancy in scatter has led to the standard practice of removing ulp cepheids from pl relation studies ( e.g. , * ? ? ?
* ) and the significant increase in rms suggests that the ulp cepheids do not conform to the classical pl relation . if we determine the pl relation of ulp cepheids alone we find pl@xmath227 has a slope ( @xmath232 ) that is flatter than pl@xmath226(though the slopes are within @xmath233 of each other , see table [ tab : fits ] ) .
the rms of our sample to pl@xmath227 is @xmath234 mag ; only marginally better than the ulp cepheid scatter about the established pl@xmath226 relation . in @xmath188 , the ulp cepheid sample
does not follow a statistically distinct and unique pl .
the longer the wavelength the less reddening is a concern .
the accuracy of distance measurement with cepheids increases going from @xmath188 to @xmath189 ( pl diagram in figure [ fig : pl_i ] ) .
pl@xmath226 has a slope of @xmath235 and the ogle cepheid sample s rms is @xmath236 .
the ulp cepheid scatter about this fit is @xmath237 larger ( @xmath238 mag ) .
the rms of the ulp cepheids is reduced to @xmath239 mag when using the pl@xmath227 fit ( slope@xmath240 ) .
pl@xmath227 is approximately five times as flat as pl@xmath226 and the two slopes are distinct at the @xmath241 level .
while ulp cepheids show the same general trends with regards to period , luminosity , and color as normal cepheids , significant statistical differences between the two populations are apparent in the @xmath189-band pl relation . to further reduce the uncertainty associated with reddening in our pl analysis
, we repeat the procedure using the reddening - free wesenheit index ( w@xmath3 ) introduced in @xcite .
the w@xmath3 pl diagram illuminates the advantages of this reddening - free index for cepheid distance measurements ( figure [ fig : pl ] ) .
the ogle cepheid sample has a very tight relation between period and luminosity ; with small scatter about the pl@xmath226 fit ( slope of @xmath242 ; rms of only @xmath243 mag ) .
the ulp cepheid rms about pl@xmath226 is @xmath244 mag .
the @xmath245-fold increase in scatter implies the ulp cepheid and the ogle cepheid samples do not conform to the same pl relation .
the pl@xmath227 fit slope is flatter than its pl@xmath226counterpart at the @xmath246 level ( @xmath247 vs. @xmath242 ) and the ulp fit slope is consistent with zero slope .
the scatter of the ulp cepheids is only @xmath4 mag about the pl@xmath227 relation .
this scatter is still @xmath248 more than that of the ogle cepheid sample ; however , the ulp sample is relatively small and heterogeneous .
we note that the scatter of ulp cepheids about the pl@xmath227 fit is smaller than that of the ogle cepheidsample about the nominal pl relation in @xmath188 and on par with the same in @xmath189 .
several trends in pl space are apparent as one examines the ulp cepheid sample in @xmath188 , then @xmath189 , and finally in the w@xmath3 index .
as reddening is reduced , the pl@xmath227 parameters are increasingly different from those of pl@xmath226 .
the pl@xmath227 fit is more shallow as one moves from @xmath188 to w@xmath3 . in the w@xmath3
, the pl@xmath227 relation reveals that ulp cepheid luminosity becomes statistically independent of period .
in essence , ulp cepheids behave as bright , standard candles in the reddening - free index .
an uncertainty of the cepheid pl and its derived distance measurement is its sensitivity to the metallicity of the stars ( e.g. , * ? ? ? * ; * ? ? ? * ) .
our sample of ulp cepheids contains six different host galaxies that span a range of @xmath249 dex in @xmath195o / h@xmath250 from 7.22 to 8.39 , providing an opportunity to investigate the dependence of the ulp cepheid pl on metallicity .
we plot the residual of each ulp cepheid to the pl@xmath251 fit listed in table [ tab : fits ] as a function of metallicity ( figure [ fig : res_oh ] ) .
in other studies , linear fits of pl residuals have determined a correction factor , @xmath252 , between @xmath253 and @xmath254 mag dex@xmath255 ( see figure 1 of * ? ? ?
recently , @xcite used the metallicity gradient in ngc 4258 to determine @xmath256 mag dex@xmath255 ( random and systematic errors , respectively ) .
we overlay this relation ( normalized to our data set ) in figure [ fig : res_oh ] for reference .
if the i zw 18 cepheids are confirmed , it suggests a stronger correlation between pl offset and metallicity than is evident in lower period cepheids ( e.g. , * ? ? ?
* ; * ? ? ?
. however , we note that we do not take into account any reddening or dm errors in this analysis . as such
, we do not claim a specific metallicity dependence for ulp cepheids .
we simply demonstrate that the residuals to the pl@xmath227 fit are broadly consistent with the range of values presented in the literature to date .
llllll[ht ] period - luminosity : @xmath188 & smc cepheids & pl@xmath257 & @xmath258 & @xmath228 & @xmath229 + period - luminosity : @xmath188 & ulps & pl@xmath259 & @xmath260 & @xmath232 & @xmath234 + period - luminosity : @xmath189 & smc cepheids & pl@xmath261 & @xmath262 & @xmath235 & @xmath236 + period - luminosity : @xmath189 & ulps & pl@xmath263 & @xmath264 & @xmath265 & @xmath239 + period - luminosity : @xmath266 index & smc cepheids & pl@xmath267 & @xmath268 & @xmath242 & @xmath243 + period - luminosity : @xmath266 index & ulps & pl@xmath251 & @xmath269 & @xmath247 & @xmath4 + at this time we do not apply a metallicity correction to our ulp cepheid pl relations .
precise gas phase oxygen abundance measurements are difficult to obtain ( for a review see * ? ? ?
* ) and we adopt a minimum metallicity error of @xmath270 dex .
the ulp cepheid sample must grow in size and the pl analysis must be more detailed to determine if i zw 18 is an exception and to characterize the functional form of the ulp cepheid pl sensitivity to metallicity .
most cepheid variables cross the instability strip three times ( e.g. , * ? ? ? * ; * ? ? ?
one can determine which crossing a cepheid is undergoing by measuring its period change , @xmath271 .
the first and third crossings are associated with positive @xmath271 while cepheids exhibit a decreasing period on their second crossing .
we investigated the light curves of our sample for signs of period changes by comparing photometry taken over the last @xmath272 years .
one ulp cepheid , hv829 , exhibits a negative period change of about 1.5 days . for this cepheid
, we compiled photometric data taken during 1970 - 1976 from @xcite and @xcite and 2000 - 2004 data from the asas catalog ( see figure [ fig : dpdt ] ) .
this result is confirmed by a measured period of @xmath273 days in @xcite and 85.2 days by @xcite , firmly establishing hv829 as a cepheid on its second crossing .
no other cepheid in our sample exhibited a measurable period change .
ulp cepheids occupy a mass range that is ideal to probe high mass evolutionary models along the instability strip .
we plot the cmd of our ulp cepheid sample and overlay the evolutionary tracks of @xcite ( figure [ fig : mass ] ) .
the location of ulp cepheids in the color magnitude diagram suggest masses between @xmath274 and @xmath275 , depending on the assumed metallicity ( here , we choose smc metallicity ) .
our sample clearly probes a mass range unexplored by current cepheid studies .
note the evolutionary tracks at @xmath276 and @xmath275 only cross the instability strip once , regardless of assumed mass - loss rate .
however , we have shown and confirmed that hv829 is undergoing its _ second _ crossing ; a result at odds with the models .
we note that the evolutionary model represents some `` mean '' behavior based on assumptions of stellar chemical composition , overshooting , and other parameters .
nevertheless , our result is one of the few observational constraints on high mass stellar evolutionary models .
hv829 may indicate that massive stars behave differently than expected .
future stellar evolutionary models in this mass regime should take this into account .
we have presented , for the first time , a collection of ultra long period ( @xmath277days ) cepheids from the literature and demonstrated their viability as a distance indicator . in the past
, ulp cepheids have been ignored as distance indicators , and in fact many stellar variability searches did not extend their cadence and search strategy to allow for their discovery . in @xmath188 , @xmath189 , and especially reddening - free w@xmath3 magnitude ,
ulp cepheids have a relatively flat pl relation compared to the respective relations for classical cepheids @xcite . the dispersion in both the classical and ulp cepheid populations about their respective pl relationships becomes smaller as one moves from @xmath188 to @xmath189 to w@xmath3 , and the discrepancy between the slopes increases .
most notably , the slope of the ulp cepheid w@xmath3 pl relation is significantly shallower than the standard smc pl relation ( slope is @xmath278 vs. @xmath242 ) .
the reddening - free wesenheit index produces the tightest ulp pl relation , with rms residual of only @xmath4 mag ( see figure [ fig : pl ] ) .
other papers ( e.g. * ? ? ?
* ) have found non - linearites in the pl relationship at lower periods . however , the change in slope seen here is much more dramatic and it is unlikely that it shares a physical connection with the pl changes at lower periods .
our ulp pl scatter in w@xmath3 is already less than that of the initial peak brightness vs. absolute magnitude relation for type ia supernova ( @xmath279 mag ; @xcite ) .
a huge amount of effort has been necessary to increase the number of observed type ia sne , refine the calibration technique ( e.g. , * ? ? ?
* ) , and to obtain the low @xmath280-@xmath281 mag scatter in the relation seen today ( e.g. , * ? ? ?
we expect future observational and theoretical studies of ulp cepheids will further decrease the already fairly small scatter found in this first analysis .
we strove to find all the known ulp cepheids in the literature , but our sample is likely not a complete census of ulp cepheids .
we note that the araucaria project found ulp cepheids in three of the five galaxies they observed at the time of this analysis , so the ulp cepheid sample should grow at a reasonable rate as cepheid studies are extended to more galaxies and previous data sets are reanalyzed with longer period searches .
there is also evidence ulp cepheids exist in a broad range of metallicities , as preliminary data analysis of a m81 variability survey with the large binocular telescope has already discovered at least one ulp cepheid ( kochanek , private communication ) .
in addition to following a fairly tight pl relation , the ulp cepheids are also very luminous , with a median absolute magnitude in @xmath189(@xmath188 ) for our sample of @xmath282 . using wfpc3 ,
_ hst _ can obtain @xmath283 photometry at @xmath284 or @xmath285 ( dm @xmath286 distance of @xmath287mpc for the median ulp cepheid ) in about 10 orbits , while only two orbits are needed to reach a dm of @xmath288 .
as one would only need a few orbits per epoch , one could detect the median ulp cepheid ( with a period of @xmath289 days ) at 100 mpc and the brightest ulp cepheids at @xmath16 mpc in a reasonable amount of time .
since the luminosity of an ulp cepheid is a weak function of its period , relatively accurate distances would not require as precise period measurements as is needed for classical cepheids .
we encourage future variability proposals to search for cepheids with periods greater than 100 days .
we note two concerns in using ulp cepheids for distance measurements .
it is obvious from our sample size that ulp cepheids are far less common than classical cepheids .
the relatively small ulp cepheid population will make precision distance measurements less practical .
as the sample size grows and the ulp cepheid pl relations become established , a single ulp cepheid observation may yield distance measurements accurate to @xmath290 .
blending is always a concern in cepheid studies .
blending can compromise classical cepheid measurements as stars of comparable brightness are likely to lie within a single psf @xcite , especially at larger distances .
we expect blending to be less of a problem for our ulp cepheids even at large distances simply because they are so bright and there are very few stars of comparable brightness in a given galaxy . the effect of blending on ulp cepheid observations will need to be investigated further as the sample is increased .
the data set and analysis herein provides meaningful constraints on theoretical work on cepheids in this mass and period range .
we have shown that ulp cepheids show strong evidence for a different , flatter pl relation than their lower period cousins .
several papers have modeled cepheid pulsations and mapped these to theoretical pl relations ( e.g. , * ? ? ?
however , this work has not been reliably extended to the period range of our sample .
pulsation models in this period range would also help determine the intrinsic scatter to the pl@xmath227 relation .
our current sample is not large enough to constrain the sensitivity of the ulp cepheid pl to metallicity .
the pl residuals as a function of metallicity are consistent with the results for shorter period cepheids ( @xmath291 mag dex@xmath255 ; * ? ? ?
note that the median @xmath195o / h@xmath250 value for a galaxy in our sample is about @xmath292dex lower than the corresponding value for the _ hst _ key project cepheid hosts .
period changes of ulp cepheids have powerful implications for stellar evolutionary models in this mass regime .
we examined photometry from two epochs separated by thirty years to check for evidence of period changes in seven magellanic cloud ulp cepheids .
only one , smc hv829 , showed a significant period change from @xmath293 to @xmath294 days .
the negative period derivative indicates that the cepheid is undergoing its second crossing .
assuming the metallicity of the smc , many current evolutionary models do not predict second crossings for cepheids in this mass range ( see figure [ fig : mass ] ) .
future models should incorporate this observational result as it should place limits on several key input parameters .
in addition to their potential as distance indicators , ulp cepheids provide observational constraints on high mass stellar evolution models .
to summarize , ulp cepheids , while often dismissed in the past , are potentially a powerful distance indicator , probe of pl metallicity sensitivity , and also a probe of massive star evolutionary models .
ulp cepheids could provide the first direct stellar distance measurements to galaxies in the 50 - 150 mpc range , extending the cosmological distance ladder well into the hubble flow .
, w. l. , madore , b. f. , gibson , b. k. , ferrarese , l. , kelson , d. d. , sakai , s. , mould , j. r. , kennicutt , jr . , r. c. , ford , h. c. , graham , j. a. , huchra , j. p. , hughes , s. m. g. , illingworth , g. d. , macri , l. m. , & stetson , p. b. 2001 , , 553 , 47 , jr . , r. c. , stetson , p. b. , saha , a. , kelson , d. , rawson , d. m. , sakai , s. , madore , b. f. , mould , j. r. , freedman , w. l. , bresolin , f. , ferrarese , l. , ford , h. , gibson , b. k. , graham , j. a. , han , m. , harding , p. , hoessel , j. g. , huchra , j. p. , hughes , s. m. g. , illingworth , g. d. , macri , l. m. , phelps , r. l. , silbermann , n. a. , turner , a. m. , & wood , p. r. 1998 , , 498 , 181 , a. g. , strolger , l .- g . , tonry , j. , casertano , s. , ferguson , h. c. , mobasher , b. , challis , p. , filippenko , a. v. , jha , s. , li , w. , chornock , r. , kirshner , r. p. , leibundgut , b. , dickinson , m. , livio , m. , giavalisco , m. , steidel , c. c. , bentez , t. , & tsvetanov , z. 2004 , , 607 , 665 , d. n. , verde , l. , peiris , h. v. , komatsu , e. , nolta , m. r. , bennett , c. l. , halpern , m. , hinshaw , g. , jarosik , n. , kogut , a. , limon , m. , meyer , s. s. , page , l. , tucker , g. s. , weiland , j. l. , wollack , e. , & wright , e. l. 2003 , , 148 , 175 , m. , strauss , m. a. , blanton , m. r. , abazajian , k. , dodelson , s. , sandvik , h. , wang , x. , weinberg , d. h. , zehavi , i. , bahcall , n. a. , hoyle , f. , schlegel , d. , scoccimarro , r. , vogeley , m. s. , berlind , a. , budavari , t. , connolly , a. , eisenstein , d. j. , finkbeiner , d. , frieman , j. a. , gunn , j. e. , hui , l. , jain , b. , johnston , d. , kent , s. , lin , h. , nakajima , r. , nichol , r. c. , ostriker , j. p. , pope , a. , scranton , r. , seljak , u. , sheth , r. k. , stebbins , a. , szalay , a. s. , szapudi , i. , xu , y. , annis , j. , brinkmann , j. , burles , s. , castander , f. j. , csabai , i. , loveday , j. , doi , m. , fukugita , m. , gillespie , b. , hennessy , g. , hogg , d. w. , ivezi , . , knapp , g. r. , lamb , d. q. , lee , b. c. , lupton , r. h. , mckay , t. a. , kunszt , p. , munn , j. a. , oconnell , l. , peoples , j. , pier , j. r. , richmond , m. , rockosi , c. , schneider , d. p. , stoughton , c. , tucker , d. l. , vanden berk , d. e. , yanny , b. , & york , d. g. 2004 , , 69 , 103501 | we examine the properties of 18 long period ( @xmath0 days ) and very luminous ( median absolute magnitude of m@xmath1 and m@xmath2 ) cepheids to see if they can serve as an useful distance indicator .
we find that these ultra long period ( ulp ) cepheids have a relatively shallow period - luminosity ( pl ) relation , so in fact they are more `` standard candle''-like than classical cepheids . in the reddening - free wesenheit index ,
the slope of the ulp pl relation is consistent with zero .
the scatter of our sample about the w@xmath3 pl relation is @xmath4 mag , approaching that of classical cepheids and type ia supernovae .
we expect this scatter to decrease as bigger and more uniform samples of ulp cepheids are obtained .
we also measure a non - zero period derivative for one ulp cepheid ( smc hv829 ) and use the result to probe evolutionary models and mass loss of massive stars .
ulp cepheids main advantage over classical cepheids is that they are more luminous , and as such show great potential as stellar distance indicators to galaxies up to @xmath5 mpc and beyond . |
fifteen years from now we will be awash in galaxies .
2df , sloan , and maybe a sloan successor will have given us redshifts for more than a million galaxies in the nearby universe .
deep and vvds will have added @xmath2 galaxies out to redshift @xmath3 .
ngst and 30 m optical / ir telescopes on the ground may have detected thousands of galaxies to @xmath4some of the first sources of light in the universe .
these surveys and others will teach us a tremendous amount about galaxy and structure formation .
our communal efforts so far will seem little more than a prologue to the vast literature on galaxies that will exist in 15 years .
nevertheless i suspect that one of the most fundamental questions in galaxy formation ( and in all astronomy ! ) will remain largely unanswered .
this is the role that supernovae played in shaping the baryonic universe .
the influence of supernovae is thought to account for a wide range of observations throughout our field .
the disruption of star - formation by supernova explosions is the favored explanation for why so few baryons are found in stars today ( e.g. , white & rees 1978 , springel & hernquist 2002 ) .
numerical simulations can not reproduce the large disk galaxies that we observe around us unless they include substantial heat input from supernovae ( e.g. , weil , eke , & efstathiou 1998 ) .
the material between galaxies at high - redshift is hotter than would be expected if gravity and the background radiation field were the only sources of heating ; another source , presumably supernovae , appears to be required ( cen & bryan 2001 ) .
it is difficult to explain why the soft x - ray background is so faint and so dominated by agn without asserting that supernovae blew apart dense clumps of baryons that would otherwise have produced copious free - free emission ( e.g. , pen 1999 ) .
the shape of galaxy clusters x - ray temperature / luminosity relationship differs from naive expectations in a way that suggests that supernovae may have imparted @xmath5 kev of energy to each of the young universe s nucleons ( e.g. , kaiser 1991 ; ponman , cannon , & navarro 1999 ) .
these examples are only a few of many .
we are unable to account for much of what we observe around us without invoking the indistinct notion of strong supernova `` feedback , '' and our understanding of the evolving universe will remain seriously incomplete until we comprehend quantitatively how this feedback works .
the standard picture is that the numerous supernova explosions in a young galaxy create an enormous blast - wave ( or `` wind '' ) that rips through the galaxy and lays waste to its surroundings . but
working through the details of this picture remains challenging even after 30 years of theoretical studies .
there are many complications , but the central problem is that we have little idea of the characteristic energy scale to associate with the blast - waves that supernovae drive .
the energy released by a single supernova , @xmath6 erg , is known , but it is unclear how large a fraction of the energy released by supernova explosions is imparted to nascent winds .
much of it may be harmlessly radiated away by the dense gas it heats .
physical arguments and numerical simulations are at present incapable of estimating a priori the energy of a galaxy s wind to within even an order of magnitude and the energy of the winds is largely what determines how large an impact they have on the evolving baryonic universe . for this reason it is still unclear ( e.g. ) which sorts of galaxies were responsible for seeding the intergalactic medium with metals , or what effect blast - waves have on galaxy formation and evolution , or even whether realistic blast - waves would be physically capable of filling the large role that they are assigned in the standard lore . whether numerical calculations in 2017 will be able to reliably estimate the energy of galaxies winds from first principles is anyone s guess , but in any case we will certainly want empirical support for the numerical results . properties of galaxies ( e.g. , disk sizes ) and of the igm ( e.g. , metal content ) are affected by the strength of supernova winds , and i can imagine convoluted and uncertain chains of reasoning that would provide an estimate the strength of winds from observations of one or the other ; but surely the most straightforward way to estimate the strength of galaxies winds is to observe galaxies and the intergalactic medium simultaneously and see how much winds have disturbed galaxies surroundings .
winds lose energy as they climb out of galaxies potential wells and crash into nearby intergalactic matter , and a robust ( if crude ) measurement of the typical energy of galaxies winds can be obtained by seeing how far they are able to propagate .
simultaneous observations of galaxies and the intergalactic material that surrounds them are easy to obtain , at least in principle : one need only conduct a galaxy redshift survey in a field that contains background qsos whose absorption spectra reveal the locations of hi and metals in the igm .
these sorts of observations , present and future , are the subject of my talk .
figures 1 and 2 show examples of what one can learn about galaxies winds at high redshift from data that are currently available . these data , described more thoroughly in adelberger et al .
( 2002 ) , consist of measured redshifts for 431 lyman - break galaxies in 6 fields at @xmath7 together with high - resolution spectra of a bright background qso in the middle of each field .
the leftmost panel of figure 1 shows the mean lyman-@xmath1 transmissivity of the igm as a function of distance from lyman - break galaxies at redshift @xmath7 .
increases in the hi content of the igm lead to more lyman-@xmath1 absorption and hence to a lower lyman-@xmath1 transmissivity .
( here the igm s lyman-@xmath1 transmissivity at redshift @xmath8 is quantified as the ratio of observed flux in a qso s spectrum at wavelength @xmath9 to the qso s expected flux if there were no absorption from the igm . )
the figure shows that as one approaches a lyman - break galaxy from afar , the density of intergalactic hi at first begins to rise .
this is in accord with the simple view that galaxies ought to be found where the density of matter ( including hi ) is highest . but at small separations ( @xmath10 comoving mpc ) something else happens ; intergalactic hi largely disappears .
one interpretation is that the galaxies winds have largely driven away all material within this radius ; the competing hypothesis that the galaxies light has ionized the hi is ruled out by the large size of region with lowered hi content ( adelberger et al .
2002 ) .
the right panel of figure 1 shows the difference between the neutral hydrogen content of the intergalactic medium in young galaxy clusters and in young voids .
far more neutral hydrogen is found in the young clusters .
the trend appears stronger than would be expected solely from the fact that galaxy clusters ought to contain more matter of all sorts ( adelberger et al .
a possible explanation is that the roiling of the young intracluster medium by galaxies blast - waves creates density inhomogeneities and increases hydrogen s neutral fraction ( adelberger 2002 , 2003 ) .
if mpc - scale winds are responsible for the hi results just discussed , one might expect to find an increase in the density of metals near galaxies .
this appears to be the case , as can be shown in a variety of ways .
consider first the triptych in the left half of figure 2 .
shown are the two dimensional correlation functions of galaxies with galaxies , of galaxies with civ absorption systems , and of galaxies with ly-@xmath1 transmissivity decrements .
the panels show how the densities of hi , civ , and other galaxies vary with spatial separation from a typical lyman - break galaxy at redshift @xmath7 ; one should imagine that the typical galaxy is at the origin and that the shadings indicate the amount of material at various angular and redshift separations .
@xmath11 is defined so that the mean density at any separation @xmath12 is @xmath13 with @xmath14 the global mean .
positive values of @xmath15 correspond to overdensities , negative to underdensities .
one may immediately see that the intergalactic medium near galaxies at @xmath7 contains a large overdensity of civ systems .
now consider the right panel of figure 2 , which shows the same data marginalized in the redshift direction , @xmath16 , where the marginalization helps remove dependence on redshift measurement errors or peculiar velocities .
contemplation of this figure reveals ( a ) that the ratio of civ to hi density increases near galaxies , suggesting perhaps that the intergalactic metallicity is highest close to galaxies , and ( b ) that on large scales the galaxy - civ cross - correlation function is similar to the galaxy - galaxy auto - correlation function , suggesting that civ absorption systems and galaxies may be similar objects .
readers will find a slightly less glib discussion in adelberger et al .
i will not pretend that the data just presented are entirely convincing .
in fact the entire point of my talk is to advertise their inadequacy ! some might worry that the statistical significance of any one of these results is low , but that does not concern me much . the flaw is easily fixed ; existing telescopes will shrink our error bars by a factor of a few in a relatively short time .
i am more concerned with the fundamental limitations of the data .
there are several .
i will mention two . measuring the content of the igm along a single skewer through the galaxy distribution
can tell us something about the characteristic stalling radii of the winds , and hence about their typical energies , but it tells us almost nothing about their geometry .
are they ( e.g. ) isotropic or bipolar ?
this has significant implications but can not be determined if the igm is observed only along a single skewer .
ideally one would have several background qsos whose light pierced each galaxy s wind at a range of angles and impact parameters .
more serious is the fact that working at @xmath17 forces us to rely almost exclusively on the igm s civ content as a proxy for its metal content .
the density of civ is not related in a simple way to the metal content of the igm ( see , e.g. , adelberger et al .
inferences about intergalactic metals based solely on civ will always be somewhat suspect to say the least .
but metals are the most direct signature of supernovae , and measuring the dispersal of metals out of galaxies and into the igm will be central to our attempts at understanding supernova - driven winds . ideally one would be able to measure the density of several metallic species , not merely civ .
this would provide constraints on the ionization parameter and let one say something definitive about metallicity .
the relative abundances of different elements would also ( in principle ) provide information about their nucleosynthetic origin .
but the vast majority of the strongest absorption lines at intergalactic temperatures and densities have wavelengths similar to or shorter than lyman-@xmath1 s .
this means that at redshifts @xmath17 they are buried in the thick lyman-@xmath1 forest and are extraordinarily difficult to detect . only at redshifts @xmath18 , where the decreasing density of the universe slows recombination and thins the forest , can these metallic absorption lines be straightforwardly found .
so this , then , is my picture of the ideal survey for quantifying empirically how supernova winds have affected the universe . find a field with a very large number of qsos at redshifts @xmath0 and obtain high - s : n , high - resolution spectra of each .
( the upper redshift bound comes from our desire that several intergalactic metal lines be easily detectable , the lower bound from the fact that the comoving density of star - formation , and presumably the prevalence of powerful winds , declines rapidly at @xmath19 . )
derive from the qsos absorption lines a detailed picture of the distribution of hi and metals in the intergalactic medium .
conduct a redshift survey of galaxies at @xmath0 in the same field and discover where star - forming galaxies lie amid the jumbled distribution of intergalactic matter . then compare the relative spatial distributions of galaxies and intergalactic material .
to the extent that galaxies send winds into their surroundings , one should find that the intergalactic medium near them is disturbed .
perhaps most of the intergalactic gas will have been driven away ; perhaps large amounts of metals will be found at the winds stalling radii . in any case
, one should be able to answer the following questions ( and others ) : what fraction of the energy released by supernovae appears to have been imparted to galaxies winds ?
what sorts of galaxies drive winds ? are winds from dwarf galaxies really the most important , as many have argued ?
are the winds more isotropic or bipolar
? what is their dominant effect on the the large fraction of the universe s baryons that lies outside of galaxies ? and so on .
now let us think how such a survey might be constructed .
obtaining galaxy redshifts in this redshift range is trivial with 8m - class telescopes ( e.g. , steidel et al .
1996 , adelberger 2002 ) . that part of the problem is solved .
the difficulty lies in obtaining quality spectra of the background qsos . over much of the redshift range @xmath0 , lyman-@xmath1 and ( especially ) the strongest metal lines lie in the observed near- or far - uv and
can only be observed from space .
unfortunately only a small number of qsos are bright enough to be observed with current and planned space - based uv spectrographs . as an illustration
, the left panel of figure 3 compares the rough high - resolution , high - s : n magnitude limits of stis and cos to the apparent magnitude distribution of all known qsos in the ( arbitrary ) redshift range @xmath20 .
only a small fraction of these qsos are bright enough to observe with either spectrograph .
the few hundred sources accessible to cos may seem like more than enough , but this is not true , at least not if the goal ( as it should be ) is to obtain information about the intergalactic distribution of hydrogen and metals along numerous skewers through the foreground galaxy distribution .
this is illustrated by the right panel of figure 3 , which shows , as a function of assumed magnitude limit , the number of unique pointings in the entire sky that contain 5 or more ned qsos at @xmath20 within a circle of radius @xmath21 .
the number of qsos and the circle radius were selected arbitrarily ; the plot is designed only to illustrate how much easier it is to find fields with large numbers of background sources as the qsos limiting magnitude becomes fainter . because ned is increasingly incomplete towards fainter magnitudes , the true gains from a fainter magnitude limit are far larger even than those indicated by this plot .
there are evidently a few known pointings with 5 or more qsos suitable for cos within a @xmath22 circle .
might these pointings be sufficient for the project we are advocating ?
the answer , sadly , is no .
5 qsos per @xmath23 circle works out to roughly one qso per @xmath24 square arcminutes . for an @xmath25 , @xmath26 cosmology at @xmath27
, the distance along the plane of the sky between the qsos in these pointings will typically be more than @xmath28 comoving mpc . for comparison ,
the winds ( apparently ) detected around galaxies at @xmath7 have a radius smaller than @xmath29 mpc .
if our goal is to have a few qsos behind an average galaxy s expanding blast - wave ( in order , e.g. , to say something about the blast - wave s geometry , or learn about the connection between galaxy properties and blast - wave properties on a case - by - case basis ) , we will need a density of background qsos that is at least 100 times higher .
a crude estimate of the required spectrographic sensitivity comes from the following calculation , which could and should be improved .
assume the qso luminosity distribution at the bright end has the shape @xmath30 with @xmath31 ( e.g. , pei 1995 ) .
then the number density of qsos brighter than magnitude limit @xmath32 is proportional to @xmath33 , and an improvement in sensitivity of 2 magnitudes over cos would lead to the desired factor of 100 increase in the density of background qsos .
the good news is that a gain in sensitivity of this size is hardly inconceivable .
the required increase in sensitivity over cos+hst is similar to the increase that will come when stis is replaced by cos , for example . obtaining the data set that i have described , and settling at last the vexing question of how supernovae have changed the evolution of the baryonic universe , is not much beyond our reach .
how long will we have to wait ? this is the bad news .
the necessary qso absorption lines are almost all at wavelengths that can only be observed from space ; the atmosphere is an obstacle that will not change . under current nasa plans these data
will not be available until most of us are retired or dead .
i would like to acknowledge the major contributions of my collaborators chuck steidel , alice shapley , and max pettini , the support and patience of this meeting s organizers , and the hospitality of bukowski s , in boston , where much of this polemic was composed .
adelberger , k.l .
2002 , phd thesis adelberger , k.l . ,
steidel , c.c . ,
shapley , a.e . , & pettini , m. 2002 , , submitted adelberger , k.l .
2003 , , submitted cen , r. & bryan , g. l. 2001 , , 546 , 81 kaiser , n. 1991 , , 383 , 104 pei , y.c .
1995 , , 438 , 623 pen , u .- l . 1999 , , 510 , 1 ponman , t.j . , cannon , d.b . , &
navarro , j.f .
1999 , nature , 397 , 135 springel , v. & hernquist , l. 2002 , , in press steidel , c.c . , giavalisco , m. , pettini , m. , dickinson , m. , & adelberger , k. l. 1996 , , 462 , 17 weil , m. l. , eke , v. r. , & efstathiou , g. 1998 , , 300 , 773 white , s.d.m . & rees , m.j .
1978 , , 183 , 341 | supernovae play a large but poorly understood role in our attempts to explain the evolution of the baryonic universe .
numerous observations throughout astronomy can not be explained if we neglect their influence , yet our quantitative understanding of the ways in which supernovae affect the universe remains remarkably poor .
this is one of the most embarrassing gaps in our knowledge of the cosmos , and planned telescopes and surveys will probably not do much to fill it .
the problem is that these surveys will be optimized to observe galaxies and intergalactic material independently of each other , while ( in the author s view ) by far the best information will come from simultaneous surveys of galaxies and the intergalactic material ( igm ) in their vicinity .
only this will show directly how galaxies affect their surroundings and provide a rough energy scale for supernova - driven winds .
redshifts @xmath0 are ideal for the joint galaxy / igm surveys we advocate , because the comoving density of star formation is near its peak , because the lyman-@xmath1 forest is thin enough for qso spectra to reveal the locations of the dominant metallic species , and because bright background qsos are common . but a new uv - capable spectrograph in space will be required .
journalnameasp conference series @holderastronomical society of the pacific jourvol000 @year2002 @titlehubble s science legacy : future optical - ultraviolet astronomy from space @authork.r .
sembach , j.c .
blades , g.d .
illingworth , r.c .
kennicutt ( eds . )
slug @ [ cols= " < " , ] ps .
@paspcstitlemkbothgobbletwo oddhead oddfoot evenheadoddheadevenfootoddfoot ps .
@myheadingsmkbothgobbletwo oddhead__oddfootevenhead__evenfoot##1##1 leftmark#1#2@upcase#1 rightmark#1#2@upcase#2 # 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
why do scientists publish ? first and foremost , because they want to share their findings and further science .
essential to this process is the ability for other people to ( efficiently ) discover these publications .
the process of information discovery has changed dramatically over the last two decades .
remember the days of spending hours in a library , paging through a&a abstracts , catalogs and tables of contents ? sometimes a publication had to be retrieved via inter library loan , adding more time to the discovery process .
a lengthy discovery process not only means a long journey before finally acquiring enough fuel to set the first step towards your goal , it also means that the discovery process is a significant portion of the cost of employing a scientist .
back then it was virtually impossible to answer the question `` what is the most popular paper on x among people interested in x ? '' or to find a set of review papers on this subject ( within a reasonable amount of time ) .
ease of access is therefore essential for efficient information discovery .
when the digital revolution of the information age culminated in the birth of the internet , followed by the world wide web , the ingredients were there to take information discovery to a new level . at this time , communication started to change from paper to electronic , and this in turn created a fundamental change in society .
information in electronic form resulted in a big shift in ease of access .
but just ease of access is not enough . in order to discover information efficiently
, you need tools to explore this electronic universe .
the combination of ease of access and ( powerful ) tools for information discovery are central to the process of transferring knowledge . in astronomy
, the ads has been central and pivotal in this digital revolution .
it is difficult to objectively quantify the absolute impact of the ads ( or any online service , for that matter ) , but we will highlight a number of facts that will illustrate aspects of the impact of the ads .
firstly , the ease of access , combined with the powerful query capabilities of the ads , has had a direct impact on the scientific process in the form of the amount of time gained that researchers otherwise would have spent going to a library , physically finding an article , xeroxing it and returning to their office . also , access through the world wide web means that communities that historically had little or no access to the scholarly literature , now have access to at least basic meta data , scanned articles , open access literature and full text through e - prints in the arxiv repository ( with which the ads is synchronized every night ) .
thirdly , by diversifying its holdings , the ads provides the astronomy community not only with the essential core journals , but also with publications from the ever - expanding periphery .
fields that seemed to have no overlap with astronomy and astrophysics in the past , suddenly become relevant and core journals for those fields start to have content that should be available to astronomers and astrophysicists .
next , through its digitization efforts , the ads has created access to rare and historical publications .
finally , another measure of impact is the fact that the existence of the ads , and other electronic resources , has had a direct influence on the publication process itself .
we will visit these various modes of impact in more detail in the following sections .
it seems reasonable to assume that an increased efficiency in discovering information will translate into researchers being better informed because , among other things , they will get exposed to a broader range of information sources per search effort . to what level
researchers are informed will most certainly influence the quality of their research .
the increase in efficiency , using the ads , is most dramatic for more complex , but still realistic queries .
for example , finding review papers on a given subject , or the most popular papers among people also interested in a given subject is a matter of just seconds , using the ads `` topic search '' . doing this type of literature research in a traditional , ``
paper '' library would easily take hours .
but also compared to other electronic resources , using the ads will result in increased efficiency and better results .
the increase in efficiency becomes even more pronounced in those cases where the ads provides links to additional sources of information ( like simbad , ned , vizier and on - line data ) . in order to quantify the increase in efficiency
, @xcite developed a metric , based on the concept of equivalent research time gained .
the physical retrieval of an article ( `` overhead '' ) was estimated to be 15 minutes on average , in the `` paper age '' . using the ads
virtually removes this overhead , therefore resulting in a gain of roughly 15 minutes research equivalent time when an article is downloaded .
further estimate that downloading an abstract , citation list or reference list gains one third of the time of an entire article ( 5 minutes ) . based on the worldwide combined ads logs from march 1999
, @xcite found that the impact of the ads on astronomy is 333 fte ( full time equivalent , 2000 hour ) research years per year . in a later study @xcite found that , based on the 2002 ads usage logs , an increase in efficiency of astronomical research by 736 fte researcher equivalent years , or about 7% of all research done in astronomy . if we apply this to the current logs ( extrapolating from the jan 2011 logs to a full year ) and just the four main astronomy journals ( _ apj _ , _ a&a _ , _ mnras _ , _ aj _ ) , we arrive at a number of 985 fte research years per year . whatever the exact interpretation of an fte researcher year is , the impact of the ads on astronomy , and science in general , is clearly substantial .
another example of contributions to increased efficiency are the services _ myads _ and _ myads - arxiv _ , providing the scientific community with a one stop shop for staying up - to - date . in 2003
the ads introduced a notification service , _ myads _ , which uses sophisticated queries ( 2nd order operators ) to give users a powerful tool for staying current with the latest literature in their sub fields of astronomy and physics .
the _ myads - arxiv _ service ( `` daily myads '' ) provides a powerful and unique filter on the enormous amount of bibliographic information added to the ads on a daily basis , as it gets synchronized with arxiv .
in essence _
myads - arxiv _ is a tailor - made , open access , virtual journal ( see @xcite ) .
the _ myads _
services automate an obvious need for most scientists : answering questions like `` who is citing my papers ? '' and `` what are recent , most popular and most cited papers in my field ? '' .
automating these queries and providing an alert service saves time . the popularity of these services is reflected in the steady growth of the number of _ myads _ and _ myads - arxiv _ users ( see figure [ fig : myadsusers ] ) . in 2002 , the harvard - smithsonian center for astrophysics visiting committee reported that the ads `` _ empowered astronomy research in underdeveloped countries and small institutions _ '' ( from report of the cfa visiting committee , 2002 ) . in november of 2005 , the united nations general assembly commended the ads for `` the mirror sites of the nasa - funded astrophysics data system ( ads ) . .
. had been enthusiastically accepted by the scientific community and had become _ important assets for developing countries _ . . . ''
( excerpt from @xcite ) .
the ads has been instrumental in helping to bridge the `` digital divide '' ( see e.g. @xcite ) for astronomical research . in @xcite
we showed that increased internet access , in particular in least developed countries ( un definition ) , has resulted in increased ads usage . in that publication we examined readership in a particular region as a function of gdp per capita ( gpc ) , because science and technology depend heavily on available budgets .
figure [ fig : usagegpc ] ( taken from that paper ) shows the relation between normalized gpc and normalized usage for a specific region . the definition for the region `` least developed countries '' was taken from @xcite .
the numbers have been normalized with respect to the 1997 level , so the diagram shows a relative growth with respect to 1997 .
the general evolution in this diagram is up and to the right , as time progresses .
the data used to construct figure [ fig : usagegpc ] were taken from the ads logs , the `` earth trends database '' ( @xcite ) , the `` world economic outlook '' ( @xcite ) and the `` world statistics pocketbook - least developed countries '' ( @xcite ) .
our logs show that growth in world usage is clearly driven by regions with the biggest potential .
high- and middle - income regions have reached a saturation level in the density of internet users , causing normalized ads usage to increase at a slower rate . clearly , the biggest potential is in low - income regions .
there is a rapid increase in internet user density in these regions , and a similar rapid increase in the number of ads users .
this indicates that the new potential is being used and in this sense there is a bridging happening of the `` digital divide '' , at least from the ads perspective .
another metric for impact is the level of penetration in the scientific community .
in other words : how many people are using the ads regularly ( 10 or more times per month ) ?
figure [ fig : adsusers ] shows that this number of regular users is still increasing .
in @xcite we showed that usage by regular ads users has a median that is fairly constant at about 21 reads per month .
this is an indication for the fact that all frequent ads users on average use the ads on a daily basis .
initially this meant that all professional astronomers use the ads daily , but by now there must be a growing group of professionals outside astronomy ( for example physicists and engineers ) .
this is also indicated by the fact that the current number of frequent ads users is significantly larger than the number of professional astronomers in the world ( the iau currently has just over 10,000 individual members , and there were about 17,000 different authors listed in the main astronomy journals in 2010 ) .
usage data also indicate another type of impact : the ads has become a public service .
ads usage has changed qualitatively over time .
the distribution of reads over users has changed , specifically the ratio of frequent to infrequent users has changed considerably over time .
we feel that the strong increase in infrequent users has an impact on the science education of the general public .
between 35% and 40% of all ads use actually comes from links external to the ads .
the google , google scholar and bing search engines are the largest sources , but the ads is also linked to by thousands of static pages .
for instance , wikipedia has more than 17,000 links to the ads . while some of the page views are scientists using google to find a reference , the vast majority are generated by the general public . in order to stay relevant for its core users
, the ads holdings must accurately reflect the complexity of the fields these people are working in . during the lifetime of the ads
, new fields emerged in astronomy and astrophysics , and existing fields became more complex , reflected in more diffuse boundaries with fields that historically had a tenuous connection with astronomy at best .
the holdings of the ads evolved accordingly .
the number of journals has increased significantly over time and we currently have over 1.8 million records in our astronomy database , distributed over more than 4,500 journals .
our digitization efforts are another aspect of the diversification of the ads holdings .
all the major astronomy journals have been scanned back to volume 1 , and they have recently been re - scanned to capture grey scale and color content .
now the ads is focusing on scanning publications with high scientific impact and that are not otherwise available online .
we have also been collaborating with librarians and observatories to digitize series of historical publications that are difficult to locate and obtain .
the impact of this effort is considerable : were it not for their availability in ads , much of this content would be simply out of the reach of researchers , librarians , and the general public .
consider , for instance , the content that ads has digitized in collaboration with the cfa library , which consists of almost 900,000 pages of historical observatory publications . during 2009 alone
, more than 1 million articles were downloaded from this collection .
in other words , many historical publications have been given a new life thanks to the digitization efforts of the ads .
that this historical material is being read can be illustrated by looking at the obsolescence function of reads .
the ads logs from march and april of 2011 indicate a reads rate for historical publications of about 1.2 reads per paper per year , which agrees with the results found in @xcite .
it seems almost unavoidable that the presence of the ads would also have had an impact on the actual products of the scientific community , specifically publications .
ease of access is very likely to have an impact on the actual writing process , in the sense that publications can probably be classified as being `` published before the introduction of the world wide web '' and `` published after the introduction of the world wide web '' .
one way of finding classifications is to represent publications as nodes in an relational network .
an obvious example is the citation network : articles a and b are directionally connected when `` a cites b '' .
this network turns out to have a very specific trend : high densification over time , i.e. non - linear growth .
this is shown in figure [ fig : densification ] , showing the relation between the number of nodes and the number of edges in the network .
one implication is that , in an average sense , bibliographies have increased in length over time .
this becomes abundantly clear in figure [ fig : bibliographies ] , which shows the average number of references in bibliographies as a function of publication year .
when an overall linear trend is subtracted ( represented by `` x '' in figure [ fig : bibliographies ] ) , a deviation from this trend is observed from the mid to late 1990 s on .
one could argue that the ease of access , offered by web services like the ads , results in more citations ( on average ) . clearly , the ads is not the only source for bibliographic information , but since our data set consisted solely of the major astronomy journals and the ads has 100% penetration in the astronomical community , it is very likely that the ads contributed significantly to the observed trend .
the ads also has had impact on the study of trends in the ( electronic ) publication process and in how literature is being used .
michael j. kurtz has contributed significantly to our understanding of e.g. usage and citation bibliometrics ( see e.g. @xcite and @xcite , @xcite ) and the discussion of the the influence of open access ( see e.g. @xcite and @xcite ) .
how is the ads moving into the future ? according to some , the information age is over and we are moving into the imagination age , where creativity and imagination are becoming the primary creators of economic value , as opposed to thinking and analysis ( see e.g. king 2007 ) .
whatever `` age '' we are in , there will remain a desire with individuals to be able to transfer information freely and to have instant access to information that would have been difficult ( or even impossible ) to retrieve previously .
but there is definitely a new trend .
it is becoming more and more common that we are faced with data collections of such magnitude and complexity , that conventional data and information discovery models brake .
we need innovative ways to explore an enormous , rapidly expanding data universe .
not too long ago gigabytes seemed like a lot .
the panoramic survey telescope and rapid response system ( pan - starrs ) project is expected to produce several terabytes of raw data per night !
the ads holdings will never come close to these amounts of data , but in our case it is the complexity of the data that requires our attention , and also the more complicated demands ( and perhaps expectations ) of people using the ads .
therefore , in order to stay relevant for its end - users , the ads has to innovate and develop new ways to explore the literature universe .
we are doing exactly that in our test environment `` ads labs '' , where we expose our users to new technologies and prototype services .
for example , the `` streamlined search '' allows users to find publications that are review papers for the subject they are interested in .
where in the `` classic '' approach the user had to have knowledge about what he / she is looking for and how to find it , the new streamlined search in `` ads labs '' offers a means to specify beforehand what a user is looking for .
in addition to this , the results will be displayed in a different way .
we use faceted filtering allowing our users to explore the literature by filtering collections of records by a particular property or set of properties .
this is an efficient way to quickly focus on a particular subset of records , from the results of a broad search . in this way
, we provide the user with a custom information environment .
we feel that this approach results in an even more efficient information discovery environment than the classic approach .
in addition to this , the abstract page now also includes recommendations .
these recommendations are based on publication similarity , in combination with article usage information from people who use the ads frequently . the inclusion of recommendations to the usual citations and references links adds an element of serendipity to the usual activity of searching and browsing the literature .
in addition to the new abstract search , ads labs also offers a full text search that includes current full text for all main astronomy journals .
this mode of searching add a whole new dimension to information discovery , hitherto not available .
we discussed a number of metrics for the impact of the ads .
these metrics are indicators for the impact of the ads has on efficiency of information retrieval ( essential for both research and scholarly communication ) , popularization of astronomy and the publication process itself ( both in production and understanding ) . on all levels
, the impact of the ads has been significant .
this impact has also been expressed in the form of recognition by peer organizations and prizes bestowed upon ads staff .
it has been widely recognized ( @xcite ) that in this era of data - intensive science , it is critical for researchers to be able to seamlessly move between the description of scientific results , the data analyzed in them , and the processes used to produce them . as observations , derived
data products , publications , and object metadata are curated by different projects and archived in different locations , establishing the proper linkages between these resources and describing their relationships becomes an essential activity in their curation and preservation .
the ads , in collaboration with the vao , the nasa archives , and the simbad project , is leading the effort of better integrating and linking the research literature with the body of heterogeneous astronomical resources in the vo , allowing users as well as applications to easily cross boundaries between archives .
this endeavor has been named semantic interlinking of astronomical resources ( @xcite ) , and is funded by the virtual astronomical observatory data curation & preservation project . by maintaining its traditional role , but introducing innovations in its querying capabilities , and by taking on this new role in the inter - linking of information sources , the ads intends to keep playing a central , pivotal role within the astronomical community .
henneken , e. , kurtz , m. j. , eichhorn , g. , accomazzi , a. , grant , c. s. , thompson , d. , bohlen , e. , murray , s. s. 2007 .
myads - arxiv a tailor - made , open access , virtual journal . library and information services in astronomy v 377 , 106 .
henneken , e. a. , kurtz , m. j. , eichhorn , g. , accomazzi , a. , grant , c. s. , thompson , d. , bohlen , e. , murray , s. s. , ginsparg , p. , warner , s. 2007 .
e - prints and journal articles in astronomy : a productive co - existence . _ learned publishing _ , 20 , 16 kurtz , m. j. , eichhorn , g. , accomazzi , a. , grant , c. s. , demleitner , m. , murray , s. s. , martimbeau , n. , elwell , b. 2005 .
the bibliometric properties of article readership information .
journal of the american society for information science and technology 56 , 111 .
kurtz , m. j. , eichhorn , g. , accomazzi , a. , grant , c. s. , demleitner , m. , murray , s. s. 2005 .
worldwide use and impact of the nasa astrophysics data system digital library . _ journal of the american society for information science and technology _ , 56 , 36 united nations , 2005 , `` report on the united nations / european space agency / national aeronautics and space administration of the united states of america workshop on the international heliophysical year 2007 '' , copuos report a / ac.105/856 . | the sao / nasa astrophysics data system ( ads ) grew up with and has been riding the waves of the information age , closely monitoring and anticipating the needs of its end - users .
by now , all professional astronomers are using the ads on a daily basis , and a substantial fraction have been using it for their entire professional career .
in addition to being an indispensable tool for professional scientists , the ads also moved into the public domain , as a tool for science education . in this paper
we will highlight and discuss some aspects indicative of the impact the ads has had on research and the access to scholarly publications .
the ads is funded by nasa grant nnx09ab39 g .
keywords : digital libraries ; sao / nasa astrophysics data system ; societal impact |
as the end products of about 97% of the stars of the galaxy , white dwarf stars offer important clues about their prior evolutionary history .
they also provide potentially the age of the galactic disk and of the globular clusters they belong to ( winget et al . 1987 ; harris et al .
2006 ) through the age that can be estimated from cooling sequences ( ruiz & bergeron 2001 ) .
however , determining the age of a white dwarf on its cooling sequence requires knowing its fundamental parameters : the total mass , the effective temperature , the luminosity , the fractional mass of the hydrogen and/or the helium outer layers , the core composition , etc .
asteroseismology of white dwarfs provides a unique tool to explore their internal structure and determine those fundamental parameters .
the method has been successfully applied to the pulsating pre - white dwarf stars of pg 1159 type , the gw vir stars , e.g. the prototype of the group pg 1159 - 035 ( winget et al .
1991 ) , the hottest one rxj 2117 + 3412 ( vauclair et al .
2002 ) and the coolest one pg 0122 + 200 ( fu et al .
2007 , crsico et al . 2007 ) , to the db pulsators , e.g. gd 358 ( winget et al .
1994 , provencal et al . 2009 ) , pg 1351 + 489 ( redaelli et al . 2011 ) , and to the dav stars , either for individual pulsator , e.g. hl tau 76 ( dolez et al . 2006 ) or for global study of the group properties ( castanheira & kepler 2009 ; romero et al .
2012 ) . since @xmath7 80% of the white dwarf stars are of da type , the uncertainties on their fundamental parameters have a strong impact on the derived age estimates .
this justifies the effort in determining precise fundamental parameters of da white dwarfs using asteroseimology of the da pulsators , the dav or zz ceti stars .
there are presently 148 pulsating da white dwarfs known ( castanheira et al .
2010a , 2010b ) .
these stars define a narrow instability strip in the h - r diagram ( or in the log@xmath0-t@xmath8 diagram ) .
this instability strip is a pure " instability strip , which means that all the da white dwarfs falling in this domain of the log@xmath0-t@xmath8 diagram do pulsate ( gianninas et al . 2011 ) .
this is an indication that the internal structure of the zz ceti white dwarfs as derived from asteroseismology is representative of the da white dwarfs as a whole group .
white dwarf stars being the oldest stars in a given stellar population , they can be used to estimate the age of the population they belong to . however , to achieve such a goal one must rely on realistic models of white dwarf stars . the major uncertainties in building such realistic models come from the uncertainty on their total mass and on their hydrogen mass fraction as well as from the approximate treatment of the convection by the mixing length theory parameterized by a mixing length @xmath9 . the asteroseismology is the only method able to determine the value of the hydrogen mass fraction and to give accurate total mass estimates of the zz ceti white dwarfs .
in addition , new theoretical developments on the interaction of the pulsations with convection have opened a way to constrain the efficiency of the convection in white dwarf stars ( wu 2001 , montgomery 2005 , montgomery et al .
2010 ) . the difficulty in determining the fundamental parameters of the zz ceti stars from asteroseismology comes from two main sources : 1 ) the zz ceti stars show generally few modes simultaneously , in contrast with the theoretical calculations which predict much more unstable modes than observed , and 2 ) the pulsation amplitudes become increasingly variable as the zz cetis evolve towards the red edge of the instability strip .
both effects make difficult to find enough appropriate modes to use the method based on the period spacing which needs a large enough number of pulsation modes to be observed and identified . only zz ceti stars close to the blue edge of the instability strip show constant pulsation amplitude .
but in this case , very few modes are unstable since the stars are just entering the instability strip . in those stars ,
the @xmath10-mechanism due to hydrogen partial ionization is responsible for the instability since the fraction of the flux conveyed by convection is negligible ( dolez & vauclair 1981 , winget et al .
as the white dwarfs evolve along their cooling sequence , the fraction of the flux conveyed by convection increases .
the light curves become more complex as a result of the interaction of the convection with the pulsations .
they show signatures of nonlinear effects like non sinusoidal pulse shapes which reflect into linear combinations of real " frequencies in the fourier spectrum .
modeling precisely the light curve resulting from these inteactions can be used to constrain the efficiency of convection ( montgomery 2005 , montgomery et al .
the asteroseismological analysis of various zz ceti stars through the instability strip allows to map the convection efficiency .
the da white dwarf hs 0507 + 0434b is one of those zz ceti star of particular interest which we intend to study in more details .
its parameters as derived from earlier spectroscopy give an effective temperature of 11630@xmath11 k and log@xmath0= 8.17@xmath12 ( fontaine et al .
2003 ; bergeron et al .
2004 ) which placed hs 0507 + 0434b approximately in the middle of the instability strip of the zz ceti pulsators .
fontaine et al . (
2003 ) provide a mass value of 0.71 @xmath13 derived from the models of wood ( 1995 ) for carbon core compositions , helium layers of @xmath14 , and hydrogen layers of @xmath15 , and absolute magnitude in @xmath16 of @xmath17 .
more recent high signal / noise ratio spectroscopy and atmospheric analysis by gianninas et al . (
2011 ) have shifted the zz ceti instability strip to higher effective temperature . with an effective temperature and a surface gravity of 12290 @xmath18 k and log@xmath0= 8.24@xmath12 respectively ,
hs 0507 + 0434b still lies close to the middle of the new zz ceti instability strip .
it is of particular interest because it forms a common proper motion pair with hs 0507 + 0434a , which is a 20000 k da white dwarf ( jordan et al .
gianninas et al .
( 2011 ) give also a higher effective temperature of 21550 @xmath19 k for the a component .
since both members of the pair must have been formed at the same time , this provides one additional constraint on the modeling of their evolution .
hs 0507 + 0434b was discovered to be a zz ceti variable by jordan et al .
it has been observed subsequently from single sites only ( kotak et al .
2002 ; handler et al .
those observations allowed the detection of 10 independent frequencies identified as @xmath20=1 gravity modes and of 38 linear combinations of these frequencies ( handler et al .
the linear combinations of frequencies were interpreted as resulting from the nonlinear interaction of the pulsations with convection .
the 10 independent frequencies are formed of three triplets ( i.e. @xmath20=1 modes split by rotation ) plus one single mode .
this was clearly insufficient to constrain the star internal structure .
to improve this situation , we carried out multisite observation campaigns in 2007 , december 2009 and january 2010 to study hs 0507 + 0434b .
the goal of this paper is to improve the determination of the fundamental parameters of hs 0507 + 0434b using the new data collected during the campaigns of 2007 , 2009 and 2010 .
it is organized as follows : the observations are described in @xmath21 .
@xmath22 analyses the derived amplitude spectrum .
a preliminary asteroseismology of hs 0507 + 0434b is presented in @xmath23 where are discussed successively the frequency identification and their linear combinations , in 4.1 , the period distribution and the period spacing , in 4.2 , the rotational splitting and the rate of rotation , in 4.3 , the inclination of the rotation axis , in 4.4 , and the mode trapping in 4.5 .
@xmath24 discusses the amplitude variations . from the improved list of pulsation modes resulting from this work , we attempt to constrain the fundamental parameters of hs 0507 + 0434b by building a set of models which are described in @xmath25 , from which we derived one `` best fit '' model .
we summarize our results in the conclusions @xmath26 .
hs 0507 + 0434b was observed during a one week multisite campaign in december 2007 , involving the 2.16-m telescope of the national astronomical observatories of china ( naoc ) in xinglong ( xl ) , the 1.8-m telescope of bohyunsan optical astronomy observatory ( boao ) of korea and the 1-m telescope of piszkstet observatory ( po ) of hungary .
later on , a four - week multisite run was carried out from december 2009 to january 2010 with the 2.16-m telescope in xl , the 2.4-m telescope of yunnan astronomical observatory of china in lijiang ( lj ) and the 1.5-m telescope of san pedro mrtir observatory of mexico ( spm ) .
tables 1 and 2 list the journals of the observations . in 2007 ,
an observing cycle of 30 seconds was used with the ccd cameras in boao and po , while a three - channel photoelectric photometer ( pmt ) was used to observe a comparison star , the sky background , and the target simultaneously with the exposure time of 1 second in xl .
these pmt data points were then summed to a similar sampling rate of 30 seconds . in 2009 and 2010 ,
ccd cameras were used for all runs with an observing cycle of approximately 30 seconds .
46.1 hours of data were collected in december of 2007 with a duty cycle of 31% , which leads to the frequency resolution of 1.9 @xmath27hz . in 2009 - 2010
we obtained 158.9 hours of data with a duty cycle of 13.4% and a better frequency resolution of 0.23 @xmath27hz .
@ccccc@ date & telescope & filter & detector & hours + 7 & xl 2.16-m & @xmath28 & pmt & 7.03 + & po 1-m & @xmath29 & ccd & 0.64 + 8 & xl 2.16-m & @xmath28 & pmt & 5.27 + & boao 1.8-m & @xmath29 & ccd & 7.86 + 9 & boao 1.8-m & @xmath29 & ccd & 7.53 + 11 & xl 2.16-m & @xmath28 & pmt & 9.13 + & boao 1.8-m & @xmath29 & ccd & 3.43 + 12 & xl 2.16-m & @xmath28 & pmt & 8.00 + & po 1-m & @xmath29 & ccd & 0.68 + 13 & xl 2.16-m & @xmath28 & pmt & 5.20 + ccc date & telescope & hours + + 13 & xl 2.16-m & 5.18 + 14 & xl 2.16-m & 4.86 + 15 & xl 2.16-m & 7.15 + 16 & xl 2.16-m & 7.09 + 17 & xl 2.16-m & 7.10 + 18 & xl 2.16-m & 5.49 + 25 & lj 2.40-m & 4.61 + 26 & lj 2.40-m & 5.90 + 27 & lj 2.40-m & 6.00 + 28 & lj 2.40-m & 8.18 + 29 & lj 2.40-m & 2.00 + 30 & lj 2.40-m & 4.36 + 31 & lj 2.40-m & 2.52 + + 12 & spm 1.5-m & 5.09 + & xl 2.16-m & 4.14 + 13 & spm 1.5-m & 5.43 + & xl 2.16-m & 3.04 + 14 & spm 1.5-m & 5.10 + & xl 2.16-m & 4.42 + 15 & spm 1.5-m & 5.11 + & xl 2.16-m & 4.58 + 16 & spm 1.5-m & 3.93 + & xl 2.16-m & 4.29 + 17 & spm 1.5-m & 5.13 + & xl 2.16-m & 4.29 + 24 & lj 2.40-m & 5.63 + 26 & lj 2.40-m & 6.10 + 27 & lj 2.40-m & 1.75 + 28 & lj 2.40-m & 2.97 + 29 & lj 2.40-m & 5.38 + 30 & lj 2.40-m & 6.05 + 31 & lj 2.40-m & 5.90 + the ccd data were reduced with the iraf daophot package .
the pmt data were reduced as described in pfeiffer et al .
unfortunately the weather conditions were poor at piszkstet observatory during the campaign in 2007 , so these data were not included in the final light curves of the year .
the data collected in both boao and xl in december 8 and 11 overlapped partially each other . as they were obtained through different filters , johnson @xmath29 for boao and white light for xl ,
the overlapping data were used to calculate the amplitude ratio obtained between the two time - series .
then , the magnitudes of the light curves from xl were calibrated to those in the filter @xmath29 .
hence , data from the two sites were able to be combined together for frequency analysis .
since the noise level of the pmt data was in general higher than that of the ccd data , the overlapping pmt data were not used for the analysis of the light curves .
1(a ) and ( b ) show the calibrated light curves of hs 0507 + 0434b in 2007 and 2009 - 2010 , respectively .
we used the period04 software ( lenz & breger 2005 ) to analyse the light curves and derive the frequencies , amplitudes and phases of the peaks in the fourier transforms .
the amplitude spectra of the light curves in 2007 and 2009 - 2010 are shown in the upper panels of fig .
2(a ) and ( b ) , respectively .
the amplitude spectra of a sinusoidal function sampled at the same rate as the data and having an amplitude equals to unity when there are data and equals to zero in the gaps ( the window functions ) are shown in the insets of upper panels of fig .
2(a ) and ( b ) . from the fourier
transform ( ft ) of the light curves in 2007 , we extracted frequency values of the highest peaks with period04 , fitting the light curves with the frequencies then prewhitenning the corresponding sinusoidal function to look for the next frequencies with significant peaks in the ft .
for the peaks extracted , we followed the criterion of breger et al .
( 1993 ) and kuschnig et al .
( 1997 ) to take the peaks whose signal - to - noise ( s / n ) ratios are higher than 4.0 .
finally , 19 significant frequencies were detected from the light curves collected in 2007 .
the first two columns of table 3 list the resolved frequencies and their amplitudes in 2007 .
in the bottom panel of fig .
2(a ) is shown the ft of the residuals after prewhitenning the 19 frequencies . for the light curves obtained from december of 2009 to january of 2010
, we note that the amplitude spectra of the four weeks of light curves differ significantly from each other , especially for the amplitude values of the peaks .
3 shows the fts of the light curves of the four individual weeks ( namely week1 , week2 , week3 , and week4 , respectively ) .
we discuss those amplitude variations in more details in @xmath24 .
to extract the frequency values from the ft of the combined light curves from december of 2009 to january of 2010 , we made an analysis similar to the one used to analyse the light curves of 2007 .
however , since the amplitude of a given peak is varying from week to week , prewhitenning of each of the peak by one frequency with a unique fitted amplitude and phase does not allow to completely remove that sinusoidal function for all the four weeks of light curves .
the existence of a residual of the prewhitenning by a sinusoidal function creates a secondary peak whose frequency value is very close to that of the resolved frequency with a lower amplitude . in order to make complete prewhitenning , we optimized the amplitude of each frequency for individual weeks of light curves and then prewhitenned the corresponding sinusoidal functions with varying amplitude values .
this process helps us to extract 33 frequencies , that are listed in the third column of table 3 while the fourth column gives the average amplitude values .
the ft of the residuals after prewhitenning by the 33 frequencies is shown in the bottom panel of fig .
we used the same selection criterion of s / n ratios higher than 4.0 as for the analysis of the 2007 data .
as the uncertainties derived by period04 estimate the internal consistency of the solutions , they are believed to be underestimated .
we derived more realistic uncertainties through monte - carlo simulations . for each observed time - series @xmath30 , @xmath31
, we extracted a number of frequencies @xmath32 and the corresponding amplitudes @xmath33 and phases @xmath34 with the software period04 , so the residual time - series @xmath35 were obtained by subtracting the sum of multiple sine functions from the observed time - series as @xmath36 then @xmath37 is regarded as the observation error of @xmath38 .
hence , we constructed 50 simulated time - series @xmath39 for each observed time - series , by taking @xmath39 as a normally distributed random variable with mean @xmath38 and standard variation @xmath40 .
the 50 simulated time - series were fitted with the sum of multiple sine functions by taking @xmath41 as initial values according to least - squares algorithm .
hence 50 sets of new @xmath41 were obtained .
the standard deviations of each parameter of @xmath42 were then calculated .
we define them as the uncertainty estimate of the parameters and include those of frequencies and amplitudes in table 3 .
we compared the uncertainties derived from monte - carlo simulations with those derived from period04 in order to estimate by how much period04 underestimates those uncertainties .
for the 2007 data set , the average uncertainties on the frequencies and on the amplitudes derived from our monte - carlo simulations are 70% larger than those estimated from period04 . for the 2009 - 2010 data sets
, period04 underestimates the frequency uncertainties by 81% and the amplitude uncertainties by 48% , in average .
= 0.8 mm .signals detected from the light curves of hs 0507 + 0434b in 2007 and 2009 - 2010 .
@xmath43=frequency in @xmath27hz .
@xmath44=amplitude in mmag .
the 1@xmath45 uncertainties are estimated from monte - carlo simulations . [ cols= "
> , > , > , > " , ] our results are in good agreement with the analysis of romero et al .
( 2012 ) for the total mass and the luminosity ( m@xmath2/m@xmath3= 0.660 and l / l@xmath3= 2.95@xmath46 @xmath47 ) but disagrees on the hydrogen mass fraction for which romero et al .
( 2012 ) find m@xmath5/m@xmath2= 5.7@xmath48 .
note that in their analysis , romero et al . used the periods of the 4 triplets listed in castanheira & kepler ( 2009 ) , taken from the study of handler et al .
( 2002 ) . in the present paper
we use two more triplets .
note also that our present result for the hydrogen mass fraction is in disagreement with our previous preliminary analysis ( fu et al .
2010 ) in which we also found a best fit for a `` thick '' hydrogen layer ( m@xmath5/m@xmath2 = 10@xmath49 ) . however , in this previous study we introduced the 301 s period in the fit while we are now confident that this period results from a linear combination and is not a genuine mode .
in addition , the @xmath50 analysis in the 4d parameter space exhibits many minima ; our present fits also show a minimum at the previous value of m@xmath5/m@xmath2 = 10@xmath49 but with a much higher @xmath50 value than the one we now find for m@xmath5/m@xmath2= 10@xmath6 .
the value derived in fu et al .
( 2010 ) comes most probably from the identification of a secondary minimum of the @xmath50 .
in estimating the inverse @xmath50 values for the series of model of our grid , we do find maxima which sometimes correspond to `` thick '' hydrogen layer models .
table 7 summarizes the parameters of the five models which give the best fits among the models computed in our grid .
f figure 8 shows the residuals to the average period spacing of the periods computed in each of the five models , from model 1 in the top panel to model 5 in the bottom panel .
all the @xmath1=1 @xmath0-mode periods of the models are shown in the range of periods corresponding to the observed ones , i.e. between 300 s and 750 s. they correspond to radial order @xmath51 from 4 to 14 for the `` thick '' hydrogen layer models 1 and 2 and @xmath51 from 2 to 10 for the `` thin '' hydrogen layer models 3 , 4 and 5 .
the dotted lines indicate the periods of the central component of the six observed triplets .
the observations indicate that the 445 s and the 698 s periods correspond to trapped or `` almost trapped '' modes . in the `` thick '' hydrogen layer model 1
, these periods would best correspond to the modes of order @xmath51= 7 and 13 , respectively , which are not trapped modes . in model 2
, the period fit is the worst and the mode ( @xmath51=7 ) with the closest period to the observed 445 s does not correspond to a trapped mode . in each of the `` thin '' hydrogen layer models 3 , 4 and 5 ,
the computed most trapped mode is found for the order @xmath51=6 , with a period of 500.7 s , 491.7 s and 489.8 s , respectively .
this mode is not detected in our present data sets . as mentioned in @xmath52 , the amplitude of a mode
is not simply related to its linear growth - rate .
so the amplitude of the trapped mode could have been lower than the detection limit during our observing runs . the observed period at 445 s corresponds to the order @xmath51=5 next to the trapped mode .
the period 698 s corresponds to the order @xmath51=9 .
but in model 3 this is not a trapped mode .
the observed trapped or `` almost - trapped '' modes are in better agreement with the modes computed in model 4 and 5 and the global fit estimated from our inverse @xmath50 selects the model 4 as the best - fit model .
this discussion illustrates how sensitive the result is on the number of observed and identified modes . in the data sets presented here , the triplet at 698 s seen in the 2007 data is not present in the 2009 - 2010 data .
in addition , the trapped mode around 500 s that is predicted in our best fit model is absent in both data sets .
this justifies the need for a follow up of this zz ceti star in order to find more modes to better constrain the model and conclude unambiguously about the `` thick '' versus `` thin '' hydrogen content of that star .
however , there are also other arguments in favor of hs 0507 + 0434b having a `` thin '' hydrogen layer .
hs 0507 + 0434a+b form a common proper motion system .
both their similar distance , 49 pc and 48 pc respectively ( gianninas et al .
2011 ) and their similar radial velocity , within their uncertainties ( maxted et al .
2000 , their table 9 ) , confirm that the two stars have been formed together at the same time .
however , in spite of their similar mass ( m@xmath2/m@xmath3=0.67 for the a component according to gianninas et al . and m@xmath2/m@xmath3=0.675 for the b component from our asteroseismic determination ) ,
they show significantly different effective temperature : the component a having 21550 k while the component b , within the same time scale , has cooled to 12290 k ( gianninas et al .
this suggests that hs 0507 + 0434b should have a thinner hydrogen layer than its companion hs 0507 + 0434a .
coming back to the periods at 301.3 s , 247.4 s and 197.7s which we find to fit well the period distribution with the average period spacing of 49.63 s , as discussed in @xmath53 , we checked that neither of the 247.4 s and 197.7s periods could correspond to genuine @xmath1=1 modes since , at least in our claimed `` best - fit model '' , the fundamental @xmath1 = 1 ; @xmath51 = 1 mode has a period of 271.7 s. however , we will discuss below the case of the 197.7s as a possible @xmath1=2 mode .
as far as the 301.3 s is concerned , the corresponding frequency ( 3318 @xmath27hz ) is distant by 150 @xmath27hz from the closest mode in the best - fit model . as this point
, we conclude that our check with the best - fit model confirms these three peaks in the ft as linear combinations .
finally , having selected one model as the best fit model , we checked a posteriori whether some of the further signals or linear combinations listed in table 4 could be identified with either @xmath1=1 and/or @xmath1=2 modes . to do this
we computed the periods of the @xmath1=2 modes of the best fit model for radial orders @xmath51 from 2 to 49 .
we find that some of the further signals could correspond to @xmath1=2 modes : this is the case for the 703.9 s period ( @xmath54 ) , the 972.2 s period ( @xmath55 ) and the 999.7 s period ( @xmath56 ) .
the 1382.7 s period identified as a linear combination ( @xmath57 - @xmath58 ) in table 4 could as well be identified as a @xmath1=2 mode .
the 197.7 s identified as another linear combination ( @xmath57 - @xmath59 ) could also be identified as the @xmath1=2 , @xmath51=2 mode for which we find a period of 198.8 s. table 8 summarizes the identifications that can be proposed for as many modes as possible .
the observed modes correspond to rather low order @xmath51 , between 3 and 10 for the @xmath1=1 modes .
it means that the hs 0507 + 0434b pulsation spectrum is not entirely in the asymptotic regime .
the period distribution does not follow exactly the regular period spacing expected in the asymptotic regime .
this explains that while the `` trapped modes '' are separated by @xmath60k= 5 in table 5 it is @xmath60k=4 in table 8 .
we summarize our results as follows : - we have obtained new photometric time - series of the zz ceti star hs 0507 + 0434b in 2007 , 2009 and 2010 .
the combined power spectra allow us to identify 18 independent pulsation modes consisting of 6 triplets and one single mode , plus a number of their linear combinations .
we identify those triplets as @xmath1=1 @xmath0-modes split by rotation .
- we determine an average period spacing of 49.63 s. the period distribution exhibits the signature of mode trapping .
- from the frequency shifts measured in the triplets , we determine an average rotational splitting of 3.59 @xmath610.57 @xmath27hz from which we infer an average rotation rate of 1.61 @xmath610.26 days .
the amplitude ratio of the triplets components , if interpreted as uniquely due to geometrical effect , could be used to estimate that the angle of the rotation axis on the line of sight is close to @xmath62 .
however , the evidence of amplitude variations casts some doubts on the assumption that only geometrical aspect is responsible for the amplitude ratios .
- the amplitude of the modes vary on week time - scale and we find that the `` pulsation power '' is also time dependent .
- we computed a grid of models and their @xmath1=1 @xmath0-modes in the adiabatic approximation .
the comparison of their theoretical periods with the observed periods through @xmath50 tests allows us to estimate the fundamental parameters of a preliminary `` best fit '' model which has a total mass of m@xmath2/m@xmath3 = 0.675 , a luminosity l / l@xmath3=3.5@xmath4 , a thin hydrogen outer layer of m@xmath5/m@xmath2= 10@xmath6 .
the periods of other models with `` thick '' hydrogen envelope do not fit as well the observed periods according to their @xmath50 .
the `` thin '' hydrogen envelope of the best fit model of hs 0507 + 0434b is in agreement with the fact that its common proper motion companion hs 0507 + 0434a , which must have been formed at the same time and has a similar mass , has a significantly higher effective temperature .
this implies that the b component must have a thinner h envelope than the a component for having cooled down to a lower effective temperature during the same cooling time scale .
this low value of the hydrogen mass fraction suggests that hs 0507 + 0434b has evolved from a last he thermal pulse episode .
further observations of hs 0507 + 0434b are required to refine the modeling , to better constrain the amplitude variation time - scales and identify the physical mechanism driving those variations , and to contribute to the mapping of the convection efficiency through the zz ceti instability strip .
we acknowledge an anonymous referee for his(her ) constructive comments which helped improving a first version of this paper .
jnf acknowledges the support from the national natural science foundation of china ( nsfc ) , through the grants 10878007 and u1231202 .
the research is partially supported by national basic research program of china ( 973 program 2013cb834900 ) and the fundamental research funds for the central universities .
nd , gv and sc acknowledge the support from the programme national de physique stellaire ( cnrs , insu ) .
lfm acknowledges financial support from the unam under grant papiit in104612 and from conacyt by way of grant cc-118611 .
special thanks are given to the technical staff and night assistants of the xinglong station of national astronomical observatories , the lijiang station of yunnan astronomical observatory , the san pedro mrtir observatory , the bohyunsan optical astronomy observatory , and the piszkstet observatory .
this research has made use of the simbad database , operated at cds , strasbourg , france .
45 althaus , l.g . ,
serenelli , a.m. , panei , j.a .
2005 , a&a , 435 , 631 bergeron , p. , wesemael , f. , lamontagne , r. et al .
1995 , apj , 449 , 258 bergeron , p. fontaine , g. , billres , m. et al .
2004 , apj , 600 , 404 bognr , zs , papar , m. , bradley , p.a .
, bischoff - kim , a. 2009 , mnras , 399 , 1954 brassard , p. , fontaine , g. , wesemael , f. , hansen , c.j .
1992 , apjs , 80 , 369 breger , m. , stich , j. garrido , r. et al .
1993 , a&a , 271 , 482 brickhill , a.j . 1975 ,
mnras , 170 , 404 castanheira , b.g . & kepler , s.o .
2009 , mnras , 396 , 1709 castanheira , b.g . ,
kepler , s.o . ,
kleinman , s.j .
2010a , 17th european white dwarfs workshop , aip conference proceedings , 1273 , 500 castanheira , b.g . , kepler , s.o . , kleinman , s.j .
et al . 2010b , mnras , 405 , 2561 crsico , a.h . ,
miller bertolami , m.m . ,
althaus , l.g .
2007 , a&a , 475,619 dolez , n. & vauclair , g. 1981 , a&a , 102 , 375 dolez , n. , vauclair , g. , kleinman , s.j .
2006 , a&a , 446 , 237 fontaine , g. , bergeron , p. , billres , m. et al .
2003 , apj , 591 , 1184 fu , j .-
n . , vauclair , g. , solheim , j .- e .
2007 , a&a , 467 , 237 fu , j .-
n . , dolez , n. , vauclair , g. et al .
2010,17th european white dwarf workshop , aip conference proceedings , 1273 , 516 gianninas , a. , bergeron , p. & ruiz , m.t .
2011 , apj , 743 , 138 goupil , m .- j . , dziembowski , w. & fontaine , g. 1998 , baltic astron . , 7 , 21 handler , g. , romero - colmenero , e. , montgomery , m.h .
2002 , mnras , 335 , 399 harris , h.c . ,
munn , j.a .
, kilic , m. et al .
2006 , a.j . , 131 , 571 hermes , j.j . , mullally , f. , ostensen , r.h .
2011 , apj , 741 , l16 jordan , s. , koester , d. , vauclair , g. et al .
1998 , a&a , 330 , 277 kawaler , s.d . & bradley , p.a .
1994 , apj , 427 , 415 kepler , s.o . ,
nather , r.e . ,
winget , d.e .
2003 , a&a , 401 , 639 kotak , r. , van kerkwijk , m.h . , clemens , j.c .
2002 , a&a , 388 , 219 kuschnig , r. , weiss , w.w . , gruber , r. et al .
1997 , a&a , 328 , 544 lenz , p. , & breger , m. 2005 , comm . in asteroseismology , 146 ,
53 maxted , p.f.l . ,
marsh , t.r . & moran , c.k.j .
2000 , mnras , 319 , 3051 montgomery , m.h .
2005 , apj , 633 , 1142 montgomery , m.h . ,
provencal , j.l .
, kanaan , a. et al .
2010 , apj , 716 , 84 pesnell , w.d .
1985 , apj , 292 , 238 pfeiffer , b. , vauclair , g. , dolez , n. et al .
1996 , a&a , 314 , 182 provencal , j.l . ,
montgomery , m.h . ,
kanaan , a. et al .
2009 , apj , 693 , 564 provencal , j.l . , montgomery , m.h . ,
kanaan , a. et al .
2012 , apj , 751 , 91 redaelli , m. , kepler , s.o . ,
costa , j.e.s .
2011 , mnras , 415 , 1220 romero , a.d . , crsico , a.h . ,
althaus , l.g .
2012 , mnras , 420 , 1462 ruiz , m.t . & bergeron , p. 2001 , apj , 558 , 761 vauclair , g. , moskalik , p. , pfeiffer , b. et al .
2002 , a&a , 381 , 122 vauclair , g. , fu , j .- n . , solheim , j .- e .
2011 , a&a , 528,5 winget , d.e .
, van horn , h.m . , tassoul , m. et al .
1982 , apj , 252 , 65 winget , d.e . ,
hansen , c.j .
, liebert , j. et al .
1987 , apj , 315 , l77 winget , d.e . ,
nather , r.e . , clemens , j.c .
1991 , apj , 378 , 326 winget , d.e . ,
nather , r.e . , clemens , j.c .
1994 , apj , 430 , 839 wood , m.a .
1995 , in white dwarfs , ed .
d. koester & k. werner ( lnp 443 ; berlin ; springer ) , 41 wu , y. 2001 , mnras , 323 , 248 | the pulsating da white dwarfs ( zz ceti stars ) are @xmath0-mode non - radial pulsators .
asteroseismology provides strong constraints on their global parameters and internal structure . since all the da white dwarfs falling in the zz ceti instability strip
do pulsate , the internal structure derived from asteroseismology brings knowledge for the da white dwarfs as a whole group .
hs 0507 + 0434b is one of the zz ceti stars which lies approximately in the middle of the instability strip for which we have undertaken a detailed asteroseismological study .
we carried out multisite observation campaigns in 2007 and from december 2009 to january 2010 . in total , 206 hours of photometric time - series have been collected .
they have been analysed by means of fourier analysis and simultaneous multi - frequency sine - wave fitting . in total , 39 frequency values are resolved including 6 triplets and a number of linear combinations .
we identify the triplets as @xmath1=1 @xmath0-modes split by rotation .
we derived the period spacing , the rotational splitting and the rotation rate . from the comparison of the observed periods with the theoretical periods of a series of models we estimate the fundamental parameters of the star : its total mass m@xmath2/m@xmath3 = 0.675 , its luminosity l / l@xmath3=3.5@xmath4 , and its hydrogen mass fraction m@xmath5/m@xmath2= 10@xmath6 .
[ firstpage ] stars : white dwarfs stars : oscillations stars : individual : hs 0507 + 0434b . |
it is well - known that at asymptotically high baryon densities the ground state of massless three - flavor qcd corresponds to the so - called color
flavor locked ( cfl ) phase @xcite . in this phase quarks of all three flavors as well as three colors undergo pairing near the fermi surface due to the attractive one - gluon exchange potential .
the properties of different collective modes , including nambu goldstone ( ng ) bosons , of the cfl phase were studied already in the framework of weak - coupling qcd @xcite . at intermediate baryon densities , related to compact star physics ,
weak - coupling expansion of qcd is not applicable , so the description of color superconductivity , including the cfl phase , is more adequate in the framework of effective theories for the low energy qcd region . in particular ,
since massless excitations might play an important role in different transport phenomena , such as cooling processes of neutron stars etc , different chiral type effective theories for the pseudoscalar ng bosons of the cfl phase are usually used ( see , e.g. , @xcite ) .
another effective theory approach is based on the nambu jona - lasinio ( njl ) models . since any njl model contains the microscopic quark degrees of freedom , it is especially convenient for the investigation of dynamical processes in dense baryonic matter .
in particular , in the three - flavor njl model the cfl effect was already considered , e.g. , in @xcite , ( see also the review @xcite ) , where some aspects of the phase structure of dense quark matter were discussed , including the influence of the @xmath5-quark bare mass , color- and electric charge neutrality conditions , external magnetic field , etc .
in addition , in @xcite the properties and structure of ng bosons of the cfl phase were considered in the framework of njl models .
one of the most noticeable differences between color superconductivity phenomena with three and two quark species is that the cfl effect is characterized by a hierarchy of energy scales . as it was established in different approaches quoted above , at the lowest scale lie ng bosons , which dominate in all physical processes with energy smaller , than the superconducting gap @xmath6 .
evident contributors at higher energy scales are quark quasiparticles , which in the cfl phase have an energy greater than @xmath6 . however , up to now we know much less about other excitations , whose energy and mass are of the order of @xmath6 in magnitude . among these particles are ordinary scalar and pseudoscalar mesons , massive diquarks etc , i.e. particles which might play an essencial role in dynamical processes of the cfl phase .
in contrast , the properties of mesons and diquarks , surrounded by color superconducting quark matter , were already discussed in the framework of the two - flavor njl model @xcite . in the present paper we are going to study just this type of excitations of the cfl ground state , i.e. mesons and massive diquarks , in the framework of the massless three - flavor njl model . in our previous paper @xcite
we have , in particular , obtained the equations for both scalar and pseudoscalar diquark masses in the cfl phase of the njl model .
there , our consideration was based on the effective action , which is a generating functional for one - particle irreducible green functions .
now , using the same technique , we perform a numerical investigation of diquark masses , as well as of the masses of scalar and pseudoscalar mesons , vs the chemical potential in the cfl phase .
moreover , the octet and singlet structure is established for massive mesons and diquarks ( both scalar and pseudoscalar ) in the cfl phase .
in addition , the masses of diquarks and mesons in the chirally broken quark matter phase are also investigated .
let us consider the following njl model with three massless quark flavors @xmath7q+ g_1\sum_{a=0}^8\big [ ( \bar q\tau_aq)^2+(\bar qi\gamma^5\tau_a q)^2\big ] + \nonumber\\ & + & g_2\!\!\!\sum_{a=2,5,7}\sum_{a'=2,5,7}\big\ { [ \bar q^ci\gamma^5\tau_a\lambda_{a'}q ] [ \bar qi\gamma^5\tau_a\lambda_{a ' } q^c ] + [ \bar q^c\tau_a\lambda_{a'}q ] [ \bar q\tau_a\lambda_{a ' } q^c]\big\}. \label{1}\end{aligned}\ ] ] in ( [ 1 ] ) , @xmath8 is the quark number chemical potential which is the same for all quark flavors , @xmath9 , @xmath10 are charge - conjugated spinors , and @xmath11 is the charge conjugation matrix ( the symbol @xmath12 denotes the transposition operation ) .
the quark field @xmath13 is a flavor and color triplet as well as a four - component dirac spinor , where @xmath14 .
( roman and greek indices refer to flavor and color indices , respectively ; spinor indices are omitted . ) furthermore , we use the notations @xmath15 for gell - mann matrices in the flavor and color space , respectively ( @xmath16 ; @xmath17 is proportional to the unit matrix in the flavor space . clearly , the lagrangian ( [ 1 ] ) as a whole is invariant under transformations from the color group su(3)@xmath18 .
in addition , it is symmetric under the chiral group su(3)@xmath19su(3)@xmath20 ( chiral transformations act on the flavor indices of quark fields only ) as well as under the baryon - number conservation group u(1)@xmath21 and the axial group u(1)@xmath22 . symmetry @xcite .
however , in the present consideration we omit the t hooft term , for simplicity . ] in all numerical calculations below , we use the following values of the model parameters ( see , e.g. , ref .
@xcite ) : @xmath23 mev , @xmath24 and @xmath25 , where @xmath26 is an ultraviolet cutoff parameter in the three - dimensional momentum space .
the linearized version of the lagrangian ( [ 1 ] ) contains collective bosonic fields @xmath27 and looks like @xmath28q -\frac{1}{4g_1}\big [ \sigma_a\sigma_a+ \pi_a\pi_a\big ] - \frac1{4g_2}\big [ \delta^{s*}_{aa'}\delta^s_{aa'}+ \delta^{p*}_{aa'}\delta^p_{aa'}\big ] \nonumber\\ & -&\frac{\delta^{s*}_{aa'}}{2}[\bar q^ci\gamma^5\tau_a\lambda_{a ' } q ] -\frac{\delta^s_{aa'}}{2}[\bar q i\gamma^5\tau_a\lambda_{a ' } q^c]-\frac{\delta^{p*}_{aa'}}{2}[\bar q^c\tau_a\lambda_{a ' } q ] -\frac{\delta^p_{aa'}}{2}[\bar q\tau_a\lambda_{a ' } q^c ] , \label{2}\end{aligned}\ ] ] where here and in the following the summation over repeated indices @xmath29 and @xmath30 is implied . lagrangians ( [ 1 ] ) and ( [ 2 ] ) are equivalent which simply follows from the equations of motion for the bosonic fields @xmath31 in ( 2)-(3 ) @xmath32 and @xmath33 are scalar and pseudoscalar fields , correspondingly .
let us consider the flavor group su(3)@xmath34=su(3)@xmath35 , which is the diagonal subgroup of the chiral group .
then , all complex scalar diquark fields @xmath36 form an @xmath37 multiplet of the su(3)@xmath38su(3)@xmath34 group , i.e. they are a color and flavor antitriplet . the same is true for complex pseudoscalar diquark fields @xmath39 which are also the components of an @xmath40-multiplet of the su(3)@xmath38su(3)@xmath34 group .
evidently , all diquarks @xmath41 have the same nonzero baryon charge .
all the real @xmath42 and @xmath43 fields are color singlets .
moreover , the set of scalar @xmath44 mesons is decomposed into a direct sum of the singlet and octet representations of the diagonal flavor group su(3)@xmath34 .
the same decomposition into multiplets is true for the set of all pseudoscalar @xmath43 mesons .
clearly , in this case the octet is constructed from three pions ( @xmath45 and @xmath46 ) , four kaons ( @xmath47 , @xmath48 and @xmath49 ) and the eta - meson ( @xmath50 ) , whereas the singlet ( @xmath51 ) corresponds to the @xmath52 meson .
the detailed investigations of the thermodynamic potential , performed in our previous paper @xcite for the above accepted model parameter set , shows that at @xmath88 mev the chirally broken quark matter phase with su(3)@xmath89su(3)@xmath90u(1)@xmath21 symmetric ground state is realized in the model ( in this case @xmath67 and @xmath91 mev , where @xmath92 is the dynamical quark mass ) .
however , at @xmath93 the cfl phase of dense baryonic matter arises . in this phase @xmath94 and @xmath6
varies with @xmath0 ( see fig . 2 in @xcite ) .
below we suppose that the gap @xmath6 is a real nonnegative number . in the following we will study the spectrum of meson / diquark excitations both in the cfl and
chirally broken phases of the njl model . since particle masses are calculated by the use of corresponding green functions , it is necessary to put a special attention to the effective action @xmath95 ( [ 12 ] ) which is really a generating functional of the one - particle irreducible ( 1pi ) two - point green functions of mesons and diquarks both in the chirally broken and cfl phases , namely @xmath96 where @xmath97 .
( to obtain a green function ( [ 17 ] ) in the chirally broken phase of quark matter ( @xmath2 ) , one should use in the expression for the quark propagator @xmath83 ( see appendix b ) , entering in ( [ 12 ] ) , @xmath91 mev and @xmath67 , whereas in the cfl phase ( @xmath4 ) all green functions ( [ 17 ] ) correspond to @xmath83 with @xmath94 and values @xmath70 presented in fig . 2 of @xcite . ) in the following , we shall say that in the theory there is a _ mixing _ between two different particles with corresponding fields @xmath98 and @xmath99 , if their 1pi green function @xmath100 is not identically equal to zero .
now , after performing in ( [ 12 ] ) the trace operation over the two - dimensional nambu - gorkov space , we obtain @xmath101 where @xmath102 and @xmath84 are the matrix elements of the quark propagator matrix @xmath83 defined in ( [ 9 ] ) ( see also appendix b ) .
( some necessary explanations concerning the trace - operation over coordinate space in the expressions ( [ 8])-([171 ] ) are given in appendix a ( see ( [ b4 ] ) ) ) .
it follows from these formulae that the effective action ( [ 8 ] ) depends on the mesonic fields @xmath103 only , i.e. it is a generating functional of the two - point 1pi green functions of mesons .
furthermore , the effective action ( [ 170 ] ) is composed from diquark fields only , and the mixing between mesons and diquarks might occur because of the effective action ( [ 171 ] ) .
however , as a detailed analysis of the njl model ( [ 1 ] ) with three massless quarks shows , each green function , containing mixing of mesons and diquarks , is proportional to @xmath104 .
thus , both in the chirally broken phase of quark matter , where @xmath67 , and in the cfl one , where @xmath94 , there is no mixing between mesons and diquarks , and so the term ( [ 171 ] ) can safely be ignored in our present consideration .
there arises a mixing between mesons and diquarks in the cfl phase .
this effect is analogous to the mixing between the @xmath105-meson and the scalar diquark in the color superconducting phase of a two - flavor njl model with nonzero masses of @xmath106- and @xmath107-quarks @xcite . ] finally , note that because of the traces containing an odd number of @xmath108 matrices , there is no mixing between scalar and pseudoscalar particles in the framework of the njl model ( 1 ) , as required by parity conservation.[ii ]
taking into account the remarks from the previous section , we have the following general expressions for the nonzero two - point 1pi green function of mesons which are valid in the chirally broken phase of quark matter as well as in the cfl one ( other two - point mesonic 1pi green functions are zero in the model under consideration ) : @xmath109 where @xmath110 . in momentum space the zeros of the fourier transformations of these functions are connected with meson masses .
in contrast , the two - point green functions for @xmath62 and @xmath134 mesons take another form .
indeed , @xmath135 where @xmath136}{3e_{\delta}^+e_{\delta}^-[p_0 ^ 2- ( e_{\delta}^++e_{\delta}^-)^2]}+\frac{4(e_{2\delta}^++e_{2\delta}^- ) [ e_{2\delta}^+e_{2\delta}^-+e^+e^-]}{3e_{2\delta}^+e_{2\delta}^-[p_0 ^2-(e_{2\delta}^++e_{2\delta}^-)^2]}\right\},\label{4.7}\\ & & { \cal r}=\int\frac{d^3q}{(2\pi)^3}\left\{\frac{32\delta^2(e_{\delta}^++ e_{\delta}^-)}{3e_{\delta}^+e_{\delta}^-[p_0 ^ 2- ( e_{\delta}^++e_{\delta}^-)^2]}+\frac{16\delta^2(e_{2\delta}^++ e_{2\delta}^-)}{3e_{2\delta}^+e_{2\delta}^-[p_0 ^ 2- ( e_{2\delta}^++e_{2\delta}^-)^2]}\right\ } \label{4.8}\end{aligned}\ ] ] ( in ( [ 4.7])-([4.8 ] ) the quantities @xmath137 etc are taken again at @xmath94 ) . evidently , these mesons are singlets with respect to the su(3 ) group , and their masses are presented also in figs 1,2 .
it is clear from these figures that none of the mesons have a zero mass in the cfl phase , i.e. they are not the nambu goldstone bosons ( ng ) of this phase .
moreover , one can see that in the cfl phase there is a singlet - octet mass splitting of pseudoscalar and scalar mesons , which however vanishes in the @xmath138 limit . indeed ,
if the value @xmath138 is used in ( [ 4.1])-([4.8 ] ) , then the 1pi green functions of the octet and singlet mesons are the same , i.e. the mass splitting is absent . + in summary , we have three main conclusions about mesons in the framework of the njl model ( 1 ) with three massless quarks .
firstly , we see that nine pseudoscalar mesons are ng bosons only in the chirally broken phase of quark matter . in the cfl phase
they cease to be ng bosons , since now they acquire finite masses ( see fig .
2 ) . secondly , the cfl breaking of the symmetry generates the octet - singlet mass splitting of mesons .
thirdly , in the cfl phase the mass splitting among scalar and pseudoscalar mesons occurs differently .
indeed , at @xmath4 the mass of the scalar octet mesons is larger than the mass of the scalar singlet meson ( see fig .
1 ) , whereas for pseudoscalar mesons the situation is inverse ( see fig .
2 ) . [ iiib ]
as it follows from the discussion in section [ ii ] , all nonzero two - point 1pi green function of diquarks both in the chirally broken and cfl phases might be determined through the relation @xmath147 where the effective action @xmath148 is given in ( [ 170 ] ) , and @xmath149 or @xmath150 .
the last restriction again means that scalar and pseudoscalar diquarks do not mix in accord with parity conservation .
as shown in our earlier paper @xcite for the case of the two - flavor njl model , any two - point green function of pseudoscalar diquarks differs in the color superconducting phase from a corresponding green function of scalar diquarks by a term which is proportional to @xmath151 , where @xmath129 is the dynamical quark mass in this phase .
the same is true for the cfl phase of the njl model ( 1 ) .
so one can conclude that in the cfl phase of the model ( 1 ) , where @xmath94 , each green function of pseudoscalar diquarks is equal to the corresponding green function of scalar diquarks , e.g. , @xmath152 , etc . hence , to establish the spectrum of the diquark excitations of the cfl phase ,
it is enough to study the set of scalar diquarks ( the mass spectrum of the pseudoscalar diquark excitations will be the same in the cfl phase ) .
let us consider the two - point 1pi green functions of the scalar diquarks .
a more detailed analysis of the effective action ( [ 170 ] ) shows that in the cfl phase , i.e. at @xmath4 , where @xmath94 and @xmath70 , eighteen scalar diquarks ( nine @xmath36 and nine @xmath153 fields ) may be divided into four sectors : @xmath154 , @xmath155 , @xmath156 and @xmath157 .
each of the sectors @xmath158 , where @xmath159 , is composed of @xmath36 , @xmath153 , @xmath160 , and @xmath161 diquark fields , whereas the sector @xmath157 is composed of six fields , @xmath162 , @xmath163 , @xmath164 , @xmath165 , @xmath166 , and @xmath167 .
it turns out that there is a mixing between diquarks entering in the same sector , whereas fields from different sectors are not mixed .
( the analogous situation takes place for the set of pseudoscalar diquarks , which is divided into nonmixing sectors @xmath168 , @xmath169 , @xmath170 and @xmath171 . ) note , the expression @xmath182 which is in the left hand side of the equation ( [ 24 ] ) is a complex - valued function defined on the riemann manifold composed of an infinitely large number of sheets of the variable @xmath130 .
the first ( physical ) sheet is the @xmath130 plane with the cut @xmath183 along the real axis .
( just the integrals ( [ 21])-([23 ] ) supply us with the values of the function @xmath182 on this physical sheet . )
it turns out that apart from the trivial zero , @xmath142 , there are no solutions of the equation ( [ 24 ] ) on this sheet , so there are no stable massive diquark excitations in the @xmath154 sector .
using the procedure of analytical continuation presented in @xcite we could find for each value of the chemical potential @xmath0 a complex point , lying on the second sheet of @xmath130 , where the function @xmath182 turns into zero .
evidently , the real and imaginary parts of this point correspond to the mass and width of a resonance .
hence , as was pointed out above , in the @xmath154 sector there appears a twicely degenerated excitation of the cfl phase , whose mass and width are presented in fig .
3 as functions of @xmath0 . a similar situation occurs in the other four - component sectors @xmath155 and @xmath156 .
namely , for both sectors the 1pi green function matrix has the form ( [ 19 ] ) .
hence , in each of these sectors there are two ng bosons as well as two resonances with the same mass and width , given in fig .
[ iva1 ] in the cfl phase.,scaledwidth=45.0% ] first of all note that there are 36 ( 18 scalar- and 18 pseudoscalar- ) diquark fields ( [ 3 ] ) in our model .
so , there should exist at least 36 elementary diquark excitations both in the chirally broken quark matter phase and in the cfl one .
the masses of 12 ( six of them are ng bosons , the other six are massive resonances ) scalar diquark excitations of the cfl phase were obtained in the previous section .
as it is clear from the discussion made at the beginning of section [ iva ] , the mass spectrum of another 12 particles , which are the cfl ground state excitations in the pseudoscalar diquark sectors @xmath168 , @xmath169 , and @xmath170 , is identical to the mass spectrum of the corresponding scalar diquarks from sectors @xmath158 , where @xmath159 ( see the previous section [ iva1 ] ) .
hence , in addition to the scalar excitations found in the previous section , in the cfl phase there exist six pseudoscalar ng bosons as well as six pseudoscalar resonances , whose mass and width are presented in fig .
3 . concerning the cfl ground state diquark excitations from the sectors @xmath157 and @xmath171
, we should note that the corresponding 1pi green functions form in the rest frame of the momentum representation a nontrivial 6@xmath1846 matrix @xmath185 , which is the same both for the @xmath157 and @xmath171 diquark sectors , with a rather complicated determinant , @xmath186 ^ 2\big\{(2t+r)^2-(2w+p)(2z+q)\big\}. \label{290}\ ] ] an exact expression for the matrix @xmath185 as well as for the 1pi green functions @xmath187 from ( [ 290 ] ) are presented in our previous paper @xcite , where it was also shown that the equation @xmath188 has a three - fold degenerated solution @xmath142 .
so in the diquark sectors @xmath157 and @xmath171 there are six ( three scalar and three pseudoscalar ) ng excitations , and the initial njl model ( 1 ) as a whole has 18 ng bosons in the mass spectrum of the cfl phase .
a more detailed consideration of the quantities @xmath187 shows that @xmath189 , @xmath190 , and @xmath191 , where @xmath192 are given in ( [ 19 ] ) .
so , the square brackets in ( [ 290 ] ) are no more than @xmath193 presented in ( [ 24 ] ) . as a consequence ,
we see , e.g. , that in the @xmath157 sector of the model there exist two massive resonances with the same mass and width , depicted in fig .
moreover , their mass and width are identical to those for the massive resonances from all scalar diquark sectors @xmath158 .
so , all these scalar diquark resonances form in total an octet with respect to the su(3)@xmath71 group .
a similar situation is valid for the pseudoscalar diquarks , where in the mass spectrum there is an octet of the cfl phase excitations with the same mass and width ( see fig .
3 ) . unfortunately
, we did not manage to find nontrivial diquark excitations of the cfl phase from the sectors @xmath157 and @xmath171 , corresponding to a zero of the expression @xmath194 that appears in the braces of ( [ 290 ] ) .
( evidently , both excitations are su(3)@xmath71 singlets . )
the matter is that @xmath195 is an analytical function on a rather complicated riemann manifold of the variable @xmath130 . on its first riemann sheet
there is only a trivial zero , @xmath142 ( which corresponds to the ng bosons , as it was discussed above ) . due to a rather complicated structure of the function @xmath195 , we were not able to perform its continuation onto the second riemann sheet and get any information about the mass and width of the remaining su(3)@xmath71 singlet diquark resonances from the sectors @xmath157 and @xmath171 .
note , the functions @xmath218 and @xmath219 are analytical in the whole complex @xmath220-plane , except for the cut @xmath221 along the real axis .
( in general , these functions are defined on complex riemann surfaces which are to be described by several sheets .
the integral representations for @xmath222 , given in ( [ 48])-([49 ] ) , define its values on the first sheet only . to find values of @xmath222 on the rest of the riemann surfaces , a special procedure of analytical continuation is needed ( see , e.g. , in @xcite ) . ) let us denote by @xmath223 and @xmath224 the solutions of the equations ( [ 50 ] ) and ( [ 51 ] ) , respectively .
of course , they depend on the coupling constant @xmath178 of the diquark channel .
obviously , the stable diquark excitation corresponds to the root @xmath223 which lies on the first riemann sheet and obeys the constraint @xmath225 .
it is fulfilled only if @xmath226 , where @xmath227 and @xmath228 are defined by @xmath229},\nonumber\\ h^{**}&\equiv&\frac { 1}{4\phi_s(0)}=\frac { \pi^2}{4\left [ \lambda\sqrt{m^2+\lambda^2}-m^2\ln ( ( \lambda+\sqrt{m^2+\lambda^2})/m)\right ] } = \frac{3g_1}{2}. \label{52}\end{aligned}\ ] ] ( actually , the last equality in ( [ 52 ] ) , i.e. @xmath230 , is due to the gap equation for @xmath231 . ) in this case @xmath223 is the mass squared of the stable scalar diquark in the vacuum , i.e. at @xmath232 . for a rather weak interaction in the diquark channel ( @xmath233 )
, @xmath223 runs onto the second riemann sheet , and unstable scalar diquark modes ( resonances ) appear . unlike this , a sufficiently strong interaction in the diquark channel ( @xmath234 ) pushes @xmath223 towards the negative semi - axis of the first riemann sheet , i.e. in this case @xmath235 , where @xmath236 is the mass of the diquark in the vacuum .
the latter indicates a tachyon singularity in the scalar diquark propagator , evidencing that the su(3)@xmath237su(3)@xmath238u(1)@xmath21 symmetric ground state of the chirally broken phase is not stable ( in this case there arises a deeper ground state , corresponding to another phase of the model , the cfl phase ) .
a similar observation was made in the framework of a two - flavor njl model , where the chirally broken quark matter phase is unstable if there is a sufficiently strong interaction in the diquark channel @xcite .
indeed , at a very large @xmath178 , as it has been shown in @xcite , the color symmetry is spontaneously broken even at a vanishing chemical potential .
let us ignore for a moment the scalar diquark sector and perform a similar analysis , based on the equation ( [ 51 ] ) , for pseudoscalar diquark excitations .
then , pseudoscalar diquarks are stable excitations of the chirally broken phase only , if the constraint @xmath239 is fulfilled , where @xmath228 is given in ( [ 52 ] ) and @xmath240}. \label{gamma3p}\end{aligned}\ ] ] ( in this case the solution @xmath224 of the equation ( [ 51 ] ) lies inside the interval @xmath241 . )
it turns out that at @xmath242 these excitations are unstable , whereas at @xmath243 a tachyonic instability of the chirally broken phase appears .
now , combining together the above separate considerations of the scalar and pseudoscalar diquark excitations , we may conclude that at a rather weak interaction in the diquark channel ( @xmath244 ) both scalar and pseudoscalar diquark excitations of the chirally broken phase are resonances .
if @xmath226 , then , in addition to mesons , the scalar diquarks are stable particles in this phase ( the pseudoscalar diquarks are unstable as before ) .
note that the initial massless njl model ( 1 ) is parametrized by three independent parameters @xmath26 , @xmath141 and @xmath178 .
so , one may expect that estimates @xmath245 and @xmath228 from ( [ 52 ] ) depend on @xmath26 and @xmath141 .
however , as it was pointed out just after ( [ 52 ] ) , the quantity @xmath246 does not depend really on the cutoff parameter @xmath26 .
in contrast , @xmath245 depends both on @xmath26 and @xmath141 .
in particular , since for the parameter set accepted in sec .
ii we have @xmath247 gev , one can present in this case the quantity @xmath245 in the following form @xmath248 . having a root @xmath249 of the equation ( [ 50 ] )
, one can find in the case @xmath226 two zeros ( with respect to the variable @xmath250 ) of the 1pi green function @xmath251 as well as four zeros of the equation @xmath215 .
they provide us with the following two different mass squared of the excitations in each two scalar @xmath210-diquark system : @xmath252 in particular , for our choice of the model parameters ( see sec .
ii ) we have @xmath253 , where @xmath247 gev . furthermore ,
if @xmath254 then @xmath255 , if @xmath256 then @xmath257 .
since there are nine scalar @xmath208 diquarks as well as nine scalar @xmath258 antidiquarks in our model , we relate @xmath259 in ( [ 53 ] ) to the mass of the diquark nonet with the baryon number @xmath260 and @xmath261 to the mass of the antidiquark nonet with @xmath262 .
the difference between diquark and antidiquark masses in ( [ 53 ] ) is explained by the absence of a charge conjugation symmetry in the presence of a chemical potential @xmath0 . finally , if @xmath234 , then a su(3)@xmath237su(3)@xmath89u(1)@xmath21 symmetric ground state , i.e. the chirally broken phase of quark matter , is not allowed to exist in the model .
the matter is that in this case tachyon singularities of the scalar diquark propagator alone ( at @xmath239 ) , or both of the scalar and pseudoscalar diquark propagators ( at @xmath243 ) appear . as a result , in this case the ground state of the cfl phase is always deeper in comparision with the ground state of the chirally broken phase .
so , only the cfl phase may be realized in the model at sufficiently high values of the coupling constant @xmath178 and arbitrary values of @xmath0 . as a consequence
, one must expect that at @xmath263 the critical value @xmath264 of the chemical potential tends to zero .
the fact that at @xmath256 the diquark mass @xmath236 tends to zero may be considered as a precursor , which appears in the chirally broken phase , of the spontaneous breaking of the su(3)@xmath18 symmetry , taking part at @xmath265 .
in the present paper we have continued the investigation , started in our previous paper @xcite , of the bosonic excitations ( mesons and diquarks ) of the dense quark matter , composed of @xmath106 , @xmath107 , and @xmath5 quarks , at zero temperature .
the consideration is performed in the framework of the massless njl model ( 1 ) , omitting the t hooft six - quark interaction term , for simplicity . in this case , the initial symmetry group of the model , i.e. su(3)@xmath19su(3)@xmath144su(3)@xmath38u(1)@xmath145u(1)@xmath22 does contain the axial u(1)@xmath22 subgroup . as a result , we have shown for the model parameter set accepted in sec .
ii that at sufficiently low values of @xmath0 , @xmath88 mev , the chirally broken quark matter phase with su(3)@xmath237su(3)@xmath238u(1)@xmath21-ground state symmetry is realized and nine massless pseudoscalar mesons ( which are the ng bosons ) , @xmath45 , @xmath46 , @xmath47 , @xmath48 , @xmath49 , @xmath50 and @xmath52 , appear .
( in massless qcd , where u(1)@xmath22 is broken on the quantum level , or in njl models with t hooft interaction the @xmath52-meson is not a ng boson . ) at @xmath4 the original symmetry is spontaneously broken down to su(3)@xmath71 , and the cfl phase does occur . in this case ,
in accordance with the goldstone theorem , eighteen ng bosons must appear in the mass spectrum of the model ( 1 ) .
( in contrast , due to the absence of the unphysical u(1)@xmath22 symmetry , only seventeen ng bosons must appear in massless qcd . ) considering 1pi green functions , we have found nine ng bosons in the sector of scalar diquark excitations .
eight of them have to be considered as non - physical , since in real qcd they supply masses to gluons by the anderson
higgs mechanism .
the remaining scalar ng boson corresponds to a spontaneous breaking of the baryon u(1)@xmath21 symmetry .
the other nine ng bosons are no more pseudoscalar mesons , but now the massless excitations in the pseudoscalar diquark sector of the model .
all that , i.e. the ng boson structure of the model ( 1 ) , is the main result of the paper @xcite which is also confirmed in the present consideration . besides ng diquarks
, we have proved in @xcite the existence of massive diquark excitations in the cfl phase . however , a detailed numerical analysis of the diquark masses vs chemical potential @xmath0 was not done there . in the present paper
we have argued that all massive diquark excitations of the cfl phase are resonances , since the corresponding singularities of their green functions in momentum space lie on the second energy riemann sheet .
moreover , they form scalar- and pseudoscalar su(3)@xmath71 octets and singlets . the mass and width of scalar- and pseudoscalar diquark resonances from octets vs
@xmath0 have been obtained numerically ( see fig .
they are avaluated around 230 mev and 50 mev , correspondingly , i.e. this quantities are at least five times smaller , than the mass and width of the scalar diquark resonance in the color superconducting quark matter composed of @xmath106 and @xmath107 quarks @xcite .
( due to numerical difficulties , we were not able to evaluate the parameters of the above mentioned massive scalar- and pseudoscalar singlet diquark resonances of the cfl phase , however , we guess that their masses are of the same order in magnitude as the masses of the diquark octets . ) to get a more complete view about the diquark properties in the framework of the njl model ( 1 ) , we have considered their masses in the chirally broken phase of quark matter , too .
it follows from our analysis that i ) at sufficiently strong interaction in the diquark channel , i.e. at @xmath266 , the existence of this phase is prohibited in the framework of the njl model ( 1 ) , ii ) depending on the coupling constant @xmath178 , scalar and pseudoscalar diquarks have different properties in this phase .
indeed , at @xmath244 , where @xmath227 is given in ( [ 52 ] ) , both types of diquarks are resonances . however , at @xmath226 the pseudoscalar diquarks remain to be resonances , whereas scalar diquarks are yet stable particles .
as this takes place , there is a splitting between the scalar diquark and scalar antidiquark masses ( see ( [ 53 ] ) ) , which is explained by the violation of the charge conjugation symmetry in the presence of a chemical potential .
( of course , in the chirally broken phase all observable particles are colorless , so one should expect that colored diquarks are confined within baryons ( see e.g. @xcite ) .
thus , one may look at our investigation of the diquark masses in the chirally broken phase as an indication of the existence of rather strong quark - quark correlations inside baryons , which might help in a better understanding of baryon dynamics . ) finally , we have considered in the model ( 1 ) the masses of mesons which are stable particles in both phases . in the chirally broken phase , i.e. at @xmath2 , all nine pseudoscalar mesons are ng bosons , whereas the nine scalar mesons have equal mass @xmath267 mev ( for the parameter set of the model accepted in sec .
ii ) . in the cfl phase , i.e. at @xmath4 , these nonet representations of mesons , reducible with respect to the su(3)@xmath71 group ,
are decomposed into the octet and singlet representations , each with its own mass .
the reason for this octet - singlet mass splitting of mesons is just the color - flavor locked symmetry breaking taking place at @xmath4 . in the cfl phase
the masses of both types of mesons vary in the interval 300@xmath3900 mev , when @xmath0 varies from 330 mev to 500 mev ( see figs 1,2 ) .
however , the mass splitting among the scalar and pseudoscalar mesons occurs in different ways . indeed , as it is easily seen from fig .
1 , the mass of the scalar octet mesons is larger than the mass of the scalar singlet meson , whereas for pseudoscalar mesons the opposite situation takes place ( see fig .
the authors are grateful to the referee for fruitful remarks concerning the multiplet structure of massive diquarks in the cfl phase .
this work has been supported in part by dfg project 436 rus 113/477 and rfbr grant 05 - 02 - 16699 .
the present appendix contains some useful formulae employed in the text . + * i ) determinant : * @xmath268=\det [ da - dbd^{-1}c ] .
\label{b1}\end{aligned}\ ] ] * ii ) inverse matrix : * @xmath269 where @xmath270^{-1}~~,~~n=[db^{-1}a -c]^{-1}~~ , ~~\bar l=[a - bd^{-1}c]^{-1}~~,~~\bar n=[d - ca^{-1}b]^{-1}. \label{b3}\end{aligned}\ ] ] * iii ) variational derivatives : * let @xmath271 are some operators in the coordinate space with matrix elements @xmath272@xmath273 and @xmath274 , respectively .
moreover , let @xmath275 and @xmath276 are some fields .
then , @xmath277 it follows from ( [ b4 ] ) that @xmath278 * iv ) fourier transformations : * for arbitrary function @xmath279 it is possible to define the fourier transformation @xmath280 by the relation @xmath281 taking these relations into account , one obtains from ( [ b5 ] ) that @xmath282 where @xmath283 are fourier transformations of the functions @xmath284 and @xmath285 , correspondingly .
in the nambu gorkov representation the inverse quark propagator matrix @xmath286 is given in ( [ 9 ] ) .
using the techniques , elaborated in @xcite , it is possible to obtain the following expressions for the matrix elements of the quark propagator matrix @xmath287 : @xmath288 where @xmath289 , @xmath290 .
moreover , @xmath291 , @xmath127 , @xmath128 and @xmath292 .
( in these and other similar expressions , @xmath121 is a shorthand notation for @xmath293 , where the limit @xmath294 must be taken at the end of all calculations .
this prescription correctly implements the role of @xmath0 as the chemical potential and preserves the causality of the theory . )
it is clear from ( [ 121])-([124 ] ) that all color- and flavor dependences in the matrix elements @xmath295 and @xmath296 arise only due to the matrix @xmath297 .
it is a @xmath298 matrix in the nine - dimensional space @xmath299 which is the direct production of color and flavor spaces .
note , in the chirally broken quark matter phase , where @xmath138 , @xmath231 , the expressions for the matrix elements ( [ 121])-([124 ] ) have a simpler form .
namely , it is clear that in this phase @xmath300 and @xmath301 i.e. the matrix elements ( [ 1210])-([1240 ] ) are proportional to the unit matrices both in the color and flavor spaces .
m. rho , e. shuryak , a. wirzba , and i. zahed , nucl .
phys . a * 676 * , 273 ( 2000 ) ; v.a .
miransky , i.a .
shovkovy , and l.c.r .
wijewardhana , phys .
d * 63 * , 056005 ( 2001 ) ; v.p . gusynin and i.a .
shovkovy , nucl .
a * 700 * , 577 ( 2002 ) .
r. casalbuoni and r. gatto , phys .
b * 464 * , 111 ( 1999 ) ; c. manuel and m.h.g .
tytgat , phys .
b * 479 * , 190 ( 2000 ) ; s.r .
beane , p.f .
bedaque , and m.j .
savage , phys .
b * 483 * , 131 ( 2000 ) ; d.t .
son and m.a .
stephanov , phys .
d * 61 * , 074012 ( 2000 ) ; g. nardulli , riv .
. * 25n3 * , 1 ( 2002 ) . v. a. miransky and i. a. shovkovy , phys .
lett . * 88 * , 111601 ( 2002 ) ; t. schfer et al , phys .
b * 522 * , 67 ( 2001 ) ; t. brauner , arxiv : hep - ph/0607102 ; j. o. andersen , arxiv : hep - ph/0609020 ; e.j .
ferrer and v. de la incera , arxiv : nucl - th/0703034 ; n. yamamoto , m. tachibana , t. hatsuda , and g. baym , arxiv:0704.2654 ; m.g .
alford , m. braby , and a. schmitt , arxiv:0707.2389 .
r. nebauer , f. gastineau , and j. aichelin , phys .
c * 65 * , 045204 ( 2002 ) ; m. buballa and m. oertel , nucl
. phys . a * 703 * , 770 ( 2002 ) ; f. neumann , m. buballa , and m. oertel , nucl . phys .
a * 714 * , 481 ( 2003 ) ; h. j. warringa , d. boer , and j. o. andersen , phys .
d * 72 * , 014015 ( 2005 ) ; e. j. ferrer , v. de la incera , and c. manuel , phys .
lett . * 95 * , 152002 ( 2005 ) ; nucl .
b * 747 * , 88 ( 2006 ) ; h. j. warringa , arxiv : hep - ph/0606063 ; h. abuki and t. kunihiro , nucl .
a * 768 * , 118 ( 2006 ) .
m. buballa , phys . rept . * 407 * , 205 ( 2005 ) . j. kundu and s. reddy , phys .
c * 70 * , 055803 ( 2004 ) ; k. fukushima and k. iida , phys . rev .
d * 71 * , 074011 ( 2005 ) ; v. werth , m. buballa , and m. oertel , arxiv : hep - ph/0611392 ; v. kleinhaus , m. buballa , d. nickel , and m. oertel , arxiv:0707.0632 .
d. ebert and k.g .
klimenko , phys .
d * 75 * , 045005 ( 2007 ) .
d. blaschke , d. ebert , k.g .
klimenko , m.k .
volkov , and v.l .
yudichev , phys .
d * 70 * , 014006 ( 2004 ) .
d. ebert , k.g .
klimenko , and v.l .
yudichev , phys .
c * 72 * , 015201 ( 2005 ) ; d. ebert and k.g .
klimenko , theor .
phys . * 150 * , 82 ( 2007 ) .
d. ebert , k.g .
klimenko , and v.l .
yudichev , phys .
d * 72 * , 056007 ( 2005 ) ; phys .
d * 75 * , 025024 ( 2007 ) .
l. he , m. jin , and p. zhuang , arxiv : hep - ph/0504148 ; arxiv : hep - ph/0511300 .
d. ebert , k.g .
klimenko , and h. toki , phys .
d * 64 * , 014038 ( 2001 ) ; v.ch .
zhukovsky et al . , jetp lett .
* 74 * , 523 ( 2001 ) ; arxiv : hep - ph/0108185 .
d. ebert and t. jurke , phys .
d * 58 * , 034001 ( 1998 ) ; l.j .
abu - raddad , a. hosaka , d.ebert , and h.toki , phys .
c * 66 * , 025206 ( 2002 ) ; a. haghpayma , arxiv:0708.0763 . | the spectrum of meson and diquark excitations of dense quark matter is considered in the framework of the nambu jona - lasinio model with three types of massless quarks in the presense of a quark number chemical potential @xmath0 .
we investigate the effective action of meson- and diquark fields both at sufficiently large values of @xmath1 mev , where the color flavor locked ( cfl ) phase is realized , and in the chirally broken phase of quark matter ( @xmath2 ) . in the last case all nine pseudoscalar mesons are nambu
goldstone ( ng ) bosons , whereas the mass of the scalar meson nonet is twice the dynamical quark mass . in the chirally broken phase
the pseudoscalar diquarks are not allowed to exist as stable particles , but the scalar diquarks might be stable only at a rather strong interaction in the diquark channel . in the case of the cfl phase ,
all ng bosons of the model are realized as scalar and pseudoscalar diquarks .
moreover , it turns out that massive diquark excitations are unstable for this phase . in particular , for the scalar and pseudoscalar octets of diquark resonances a mass value around 230 mev was found numerically .
in contrast , mesons are stable particles in the cfl phase .
their masses lie in the interval 400@xmath3500 mev for not too large values of @xmath4 . |
the structure and evolution of protoplanetary discs depend upon the rate at which gas can shed its angular momentum and thereby flow inwards .
two widely applicable physical mechanisms are known to lead to the required outward angular momentum transport . if the gas is coupled to a magnetic field , instabilities that inevitably arise in differentially rotating discs ( balbus & hawley 1991 ; chandrasekhar 1961 ; velikhov 1959 ) lead to turbulence and angular momentum transport ( stone et al .
1996 ; brandenburg et al . 1995 ; for a review see e.g. hawley & balbus 1999 ) .
if the disc is massive enough , gravitational instability leads to additional transport ( toomre 1964 ; laughlin & bodenheimer 1994 ; nelson et al .
1998 ; pickett et al .
2000 ) . applying these findings to the construction of protoplanetary disc models
leads to the structure shown schematically in fig . 1 ( after gammie 1996 ) .
in the inner disc , mhd turbulence transports angular momentum .
however , at larger radii of @xmath5 , where the temperature is typically a few hundred k , magnetic field instabilities are suppressed by the low ionization fraction ( matsumoto & tajima 1995 ; gammie 1996 ; gammie & menou 1998 ; livio 1999 ; wardle 1999 ; sano & miyama 1999 ; sano et al .
this leads ( gammie 1996 ) to the formation of a _ layered _ disc structure , in which the gas near the disc midplane is cold , shielded from ionizing high energy radiation , and quiescent ( non - turbulent ) .
turbulence and accretion occurs only in a thin surface layer that is ionized by cosmic rays . moving still further outwards the entire thickness of the disc again
become viscous , either at the radius where the surface density is small enough for cosmic rays to penetrate to the midplane , or where the onset of disc self - gravity provides an alternative non - magnetic source of angular momentum transport .
the predictions of a static layered disc model for the accretion rate and spectral energy distribution of t tauri stars were discussed by gammie ( 1996 ) , and are broadly consistent with observations ( e.g. with the accretion rate for classical t tauri stars measured by gullbring et al .
1998 ) . in this paper
we consider the evolution of the layered disc , which can not be in a steady state ( gammie 1996 , 1999 ; stepinski 1999 ) , and examine the implications for the outflow history of young stars and for the predicted disc mass .
the most significant changes to the disc structure occur at the radii of greatest interest for planet formation ( reyes - ruiz & stepinski 1995 ) , and we discuss the implications for the migration of low mass planets , and for the eccentricity of massive planets interacting with the disc .
describing the evolution of the surface density @xmath6 and midplane temperature @xmath7 of a layered disc requires only minor modifications to the usual time - dependent equations for thin accretion discs .
we denote the surface density of the ` active ' ( viscous ) disc by @xmath8 . if , @xmath9 or , @xmath10 then the disc is viscous throughout its thickness and @xmath11 .
otherwise only the surface layers are viscous and @xmath12 .
the values of these parameters are determined by the requirement that the disc be sufficiently ionized to support mhd turbulence ( gammie 1996 ) .
we adopt @xmath13 , and @xmath14 . for a keplerian disc ,
the angular velocity is @xmath15 , where @xmath16 is the stellar mass .
the surface density evolution is then described by , @xmath17 + { \dot{\sigma } } ( r , t ) , \label{eq3}\ ] ] where @xmath18 is the kinematic viscosity and @xmath19 is the rate of change of the surface density due to infall onto the disc . for the energy equation
, we adopt a simplified form of that used by cannizzo ( 1993 ) , @xmath20 here @xmath21 is the disc specific heat , which for temperatures @xmath22 is given by @xmath23 , where @xmath24 is the gas constant and @xmath25 is the mean molecular weight .
@xmath26 represents local heating due to viscous dissipation , given by , @xmath27 if the entire disc is viscous and @xmath28 otherwise . for @xmath29 , the local cooling rate , we assume that each annulus of the disc radiates as a blackbody at temperature @xmath30 , so that @xmath31 where @xmath32 is the stefan - boltzmann constant
. finally , we include an advective term in the energy equation , which depends on the vertically averaged radial velocity , @xmath33 and the radial temperature gradient .
completing the model requires specification of both the viscosity @xmath18 and the vertical structure , which sets the relation between the central temperature @xmath34 and the surface temperature @xmath30 .
we adopt the simplest , vertically averaged approach , for which , @xmath35 where @xmath36 is the optical depth for a given opacity @xmath37
. when an annulus makes the transition to the layered state , we crudely account for this by replacing @xmath38 in the expression for @xmath39 by @xmath40 .
note that this means that we do not attempt to treat the vertical structure _ during _ the transition consistently .
.opacity regimes in ascending order of temperature , fitted by analytic functions of the form @xmath41 .
we have used fits provided by bell & lin ( 1994 ) , as modified for low temperatures by bell et al .
the maximum temperature @xmath42 for each regime is quoted for a typical disc density of @xmath43 ( where specification of the density is necessary ) . [
cols="^,<,^,<,<",options="header " , ] analytic expressions for low temperature rosseland mean opacities are given by bell et al .
the behaviour of the disc depends primarily on the opacity near the transition temperature @xmath44 , for which the fit is , @xmath45 the full list of opacities used is quoted in table 1 .
these fits have been taken from bell & lin ( 1994 ) , with the modifications for low temperatures quoted in bell et al .
( 1997 ) .
we adopt an alpha prescription ( shakura & sunyaev 1973 ) for the viscosity , @xmath46 where @xmath47 is the midplane sound speed and @xmath48 a dimensionless parameter measuring the efficiency of angular momentum transport . numerical simulations of mhd turbulence in discs suggest that @xmath49 ( stone et al . 1996 ; brandenburg et al . 1995 ) .
an immediate consequence of the layered disc structure is that the accretion rate through the surface layer is a fixed , increasing function of radius ( gammie 1996 ) .
inevitably , therefore , material builds up in the quiescent layer over time , until the surface density becomes high enough that self - gravity of the disc becomes important .
the condition for this is that the toomre @xmath50 parameter ( toomre 1964 ) , given by , @xmath51 is less than some critical value @xmath52 , which is of the order of unity . here
, we adopt @xmath53 .
when @xmath54 , non - axisymmetric instabilities set in and act to transport angular momentum outwards . in doing so , the surface density evolves in such a way that the stability of the disc generally increases , thereby self - regulating the violence of the instability .
in a local approximation , the efficiency of angular momentum transport due to self - gravity can be determined by the requirement that viscous heating balances radiative losses ( gammie 1999 ) .
this approach , in which the cooling time directly determines @xmath48 , is most appropriate for discs whose thermal balance is determined by viscous heating . even in this limit , global effects may be important ( balbus & papaloizou 1999 ) .
a global redistribution of surface density , driven by self - gravity , can also occur even if the disc temperature is set externally ( for example by irradiation ) .
this is the limit that has been studied extensively using isothermal or polytropic simulations ( e.g. laughlin & bodenheimer 1994 ) .
for protoplanetary discs , models suggest that both viscous heating and irradiation can be important at different epochs and radii ( bell et al .
1997 ) . given the above complications , it is evident that attempts to include the effects of self - gravity into one dimensional disc models can only be approximate .
we adopt the form suggested by lin & pringle ( 1987 , 1990 ) , and use a variant of equation ( [ eq11 ] ) along with an effective gravitational ` @xmath48 ' , @xmath55 , with @xmath56 operationally , we solve equations ( [ eq3 ] ) and ( [ eq4 ] ) using an explicit finite difference method , and include the self - gravity as an additional sweep over the grid using @xmath57 and the full surface density @xmath58 in the diffusive part of equation ( [ eq3 ] ) , and as an additional heating term in the energy equation .
for our numerical calculations we adopt @xmath59 , an inner disc radius @xmath60 ( 5 @xmath61 ) , and an outer disc radius @xmath62 ( 40 au ) . the computational grid of 120 radial mesh points is uniform in a scaled radial variable @xmath63 .
a zero torque boundary condition is imposed at @xmath64 , while at @xmath65 we prevent outflow by setting the radial velocity , @xmath66 , to zero . when mass is added to the disc ( representing infall of further material from the molecular cloud ) ,
we take @xmath67 to be a gaussian centered at 10 au with a width of 1 au .
this is exterior to the layered section of the disc that we are interested in studying , so the details of how mass is added are unimportant here .
we add mass assuming that it has the same specific angular momentum and temperature as the local disc material . to demonstrate the evolution of the layered disc model , we first take the infall rate onto the outer disc @xmath68 to be a constant .
we take @xmath69 , as suggested by numerical simulations of mhd disc turbulence , and allow the disc to evolve from an initially low mass ( @xmath70 ) until either a steady state or a limit cycle has developed .
figure 2 shows the accretion rate onto the star for a range of mass infall rates between @xmath71 and @xmath72 . for the highest infall rate ,
the disc is able to transport gas steadily onto the star .
the inner region , in which @xmath73 , is able to be supplied by an outer disc in which @xmath54 and the angular momentum is transported by self - gravity .
@xmath50 is typically around 1.5 in this region , with the transition occurring at a radius of @xmath74 au . for lower infall rates such steady accretion
is not possible .
after an initial phase in which the disc mass increases steadily , a limit cycle is obtained in which outbursts , with a peak accretion rate @xmath75 , alternate with quiescent intervals where @xmath76 .
both the duration of the outbursts , and the recurrence time , vary with the rate at which gas is supplied to the outer disc .
however , to order of magnitude , for @xmath77 we find @xmath78 , and @xmath79 .
the mechanism for these outbursts is shown in figure [ f3 ] , which plots the central temperature during the quiescent interval leading up to an outburst . following the end of one outburst
, there is an extended layered region of the disc between @xmath80 au and @xmath74 au .
this region is stable against gravitational instability , and supports only the small quiescent accretion rate .
gas flowing inwards from the more active self - gravitating region of the disc thus accumulates in the quiescent region , and the transition radius where the disc becomes self - gravitating moves inward .
at the same time , the greater dissipation rate in the self - gravitating disc creates a local peak in the disc temperature , which also moves inwards .
a new outburst is triggered when this peak first exceeds @xmath44 , thereby satisfying the condition for mhd turbulence to be restarted .
once the outburst begins thermal energy is rapidly advected inwards , triggering a transition of the entire inner disc into a hot turbulent state with high accretion rate .
this continues until the reservoir of accumulated mass has been depleted , whereupon a cooling wave sweeps through the disc and returns it to the quiescent , layered , state .
the derived outburst durations are consistent with the expected viscous timescale in protoplanetary discs at radii of a few au . for an outburst triggered at a radius of approximately 2 au
, the viscous timescale once the disc is in outburst , given by @xmath81 where @xmath82 is the disc scale height , is approximately , @xmath83 this is only a rough estimate , but it is of the correct order of magnitude to match the numerical model .
the involvement of disc self - gravity means that relatively large disc masses are required for the limit cycle to operate . in our models
, @xmath84 fluctuates between 0.2 and 0.3 @xmath85 . during an outburst , only a small fraction of the disc mass around 20% is accreted . in more realistic evolutionary models ,
in which the mass infall rate declines with time , the disc will therefore be able to produce a handful of outbursts , with a few hundredths of a solar mass of gas accreted during each , before settling into a permanent quiescent state .
the most striking variability observed in pre - main - sequence stars occurs in fu orionis events ( e.g. kenyon 1995 ; hartmann & kenyon 1996 ) , which are large amplitude outbursts in the system luminosity originating in the accretion disc .
the statistics of these events are subject to considerable uncertainty , but the peak accretion rate during outbursts is of the order of @xmath86 , while the duration is generally thought to be around @xmath87 yr .
the most popular explanation for fu orionis events is in terms of a disc thermal instability ( e.g. bell & lin 1994 ; kley & lin 1999 , and references therein ) , akin to that used to model dwarf novae ( e.g. cannizzo 1993 ) . in the protostellar case
, a thermal instability can only operate at extremely small disc radii , typically at less than 0.1 au . matching the observed timescales
then requires small values of @xmath48 ( @xmath88 to @xmath89 ) , with correspondingly large values for the disc surface density . as noted by gammie ( 1999 )
, it would be attractive to dispense with the need for a thermal instability by invoking an alternative limit cycle of a layered disc .
this might be possible _ if _ the transition to the outburst state was triggered at the extreme inner edge of the layered region .
whether this happens depends on where in the quiescent layer mass preferentially accumulates , and the outcome is therefore sensitive to the detailed , and rather uncertain , physics of the layered disc . in our model , in which matter is being added to the layered region primarily at large radius , the triggering occurs at the outer edge of the layered region , and the timescales are inconsistent with those of fu orionis events .
we note , however , that the accretion rates of @xmath90 that we obtain during our outbursts fall into the thermally unstable regime .
the long outbursts obtained here could then feed shorter fu orionis event of the sort calculated by bell & lin ( 1994 ) .
continuous replenishment of the disc mass from infall is not a realistic model for protostellar accretion . to explore how the accretion rate onto the star may evolve with time
, we adopt a simple model for the infall of gas onto the disc .
the details are model - dependent ( see e.g. larson 1969 ; shu 1977 ; basu 1998 ) , but at early times the infall rate is expected to be of the order of @xmath91 , where @xmath92 is the sound speed in the collapsing cloud . for cloud temperatures
@xmath93 , this implies an infall rate of @xmath94 .
we take an initial infall rate of @xmath95 , and assume that this declines exponentially on the free fall timescale , @xmath96 , where @xmath97 is the cloud density .
we take @xmath98 .
other input parameters are as described earlier . with these parameters , fig .
[ f4 ] shows the accretion rate for the layered disc model .
for comparison , we also show the results from a viscous disc model which has identical parameters , but which has no layered region .
this control model would be appropriate , for example , if angular momentum transport was driven by a purely hydrodynamic mechanism whose efficiency was independent of the disc temperature .
we compute the control model by using the same code and infall history , but with @xmath44 set to zero . initially , when the infall rate is higher than around @xmath99 , both models support an accretion rate onto the star which tracks that with which matter is being added to the disc .
subsequently , accretion through the layered disc occurs in outbursts with properties similar to those described earlier , but with increasing intervals between events . for this model , the last outburst before the disc became permanently quiescent occurred after 2 myr .
the accretion rate onto the star in the control model , on the other hand , declines smoothly with time .
similar results are obtained using other functional forms for the mass infall rate as a function of time .
the timescales of variability predicted by the model are much longer than any direct observational record .
indirect evidence for long timescale variability of young stars , however , is provided by studies of protostellar outflows , since the rate of mass outflow via jets is widely believed to track the accretion rate .
for example ,
observations by reipurth , bally & devine ( 1997 ) suggest that large amplitude variations in the outflow rate occur not only on the @xmath87 yr timescales characteristic of fu orionis events , but also on much longer timescales of @xmath100 yr or greater .
these longer timescales are characteristic of processes occurring at larger disc radii than thermal instabilities , and are consistent with instabilities , of the kind discussed here , originating in the layered disc region .
the _ outburst _ accretion rate in the layered disc model is several orders of magnitude higher than that in the equivalent fully viscous disc model at @xmath101 .
however , the time averaged accretion rate is substantially lower , so that at late times the disc mass in the layered model substantially exceeds that in the viscous model .
we find that @xmath102 after 1 2 myr . by this time
the mass of the fully viscous disc has dropped to only @xmath70 . as shown in fig . 5 , most of this extra mass
is tied up in the quiescent layer at radii of the order of 1 au , where the surface density exceeds the minimum mass solar nebula estimate ( hayashi , nakazawa & nakagawa 1985 ) by about two orders of magnitude .
a disc mass of a few tenths of a solar mass at early times is consistent with the upper end of the disc mass distribution inferred from mm - wavelength observations ( beckwith et al .
1990 ; osterloh & beckwith 1995 ) .
however , the layered disc model predicts that a substantial mass at least a few hundredths of a solar mass remains in the disc throughout most of the typical classical t tauri disc lifetime ( strom 1995 ) .
although recent measurements of the mass of @xmath103 in debris discs indicate that several jupiter masses of gas can persist beyond 10 myr ( thi et al .
2001 ) , our predicted disc masses are in apparent contradiction with mm observations of many systems with disc masses estimated at @xmath104 or smaller . to check whether this is a serious problem for the layered disc model , we plot in fig . 5 the cumulative distribution of disc mass as a function of surface density .
most of the ` additional ' mass in the layered disc , compared to the equivalent fully viscous control model , is locked up in the quiescent layer at small radii and high surface density .
this material is highly optically thick , and thus would not contribute to the observed emission at mm wavelengths . for a range of dust models ,
the opacity of dust at @xmath105 is thought ( menhchikov & henning 1997 ; beckwith et al . 1990 ) to be in the range @xmath106 ( note that this is per gram of _ dust _ ) . for a gas to dust ratio of 100 , this implies that an optical depth of unity at 1 mm is attained for @xmath107 . at this surface density threshold ,
the masses of optically thin gas in the layered and non - layered disc models are very similar .
the masses of protoplanetary discs inferred from mm - wavelength measurements could thus substantially underestimate the true disc masses .
the accretion rates inferred for classical t tauri stars can also provide a test . in the simplest version of the layered disc model , the accretion rate during quiescence does not change over the lifetime of the protostellar phase .
this rate is consistent with the order of magnitude inferred for accreting classical t tauri stars ( gullbring et al .
there have also been suggestions that the observationally inferred accretion rate declines with increasing stellar age ( hartmann et al .
although currently the scatter in measured accretion rates at a given assumed age is very large , and the age estimates themselves are subject to substantial uncertainties ( tout , livio & bonnell 1999 ) , this observation poses difficulties for the simplest layered disc model ( stepinski 1998 ) .
we note , however , that changes in the quiescent accretion rate with time would be expected in more complete models , for example if the opacity from dust decreased with time due to grain growth . including heating of the disc from the changing stellar radiation
would also lead to a decline in the quiescent accretion rate .
the environment which a layered disc presents for planet formation differs in several respects from both previous viscous disc models ( e.g. lin & pringle 1990 ) , and from static models such as the minimum mass solar nebula . at radii between a few tenths of an au , and several au , the gas near the disc midplane is almost always quiescent , unless the settling of cm sized particles itself drives additional instabilities ( goldreich & ward 1973 ; cuzzi , dobrovolskis & champney 1993 ) .
the low viscosity of the disc in this region would reduce the rate at which massive planets migrate inwards , and may allow planet - disc interactions to excite significant eccentricity ( papaloizou , nelson & masset 2001 ) .
other differences include the disc at large radii remaining mildly unstable to gravitational instability for a relatively long time around a myr because the reduced time - averaged accretion rate leads to a larger disc mass at late times when compared to viscous disc models .
outbursts , which persist for a substantial fraction of the disc lifetime , mean that much of the inner disc is expected to be subject to rapid heating and cooling episodes . however , we find no evidence that these cooling waves drive the disc into a state where gravitational collapse ( @xmath108 ) would occur .
the survival of solid bodies in protoplanetary discs is limited by the rate at which they migrate inwards relative to the gas .
migration can be rapid both for cm - sized bodies ( e.g. godon & livio 1999 , and references therein ) , and is particularly problematic for low mass planets , for which migration occurs due to the influence of gravitational torques ( goldreich & tremaine 1979 ) . in standard disc models ,
the migration timescale for earth mass planets at a few au can be as short as @xmath109 , which leaves little time to assemble the cores of giant planets before the putative building blocks are consumed by the star .
the rate of migration depends only very weakly upon the surface density profile ( ward 1997 ) , but _ is _ sensitive to the gradient of the central temperature , and can be halted if there exists a region of the disc where @xmath110 . during the quiescent phase ,
the models we have discussed here possess such a non - monotonic central temperature profile at radii of 1 - 3 au ( fig .
3 ) , leading to the possibility of an enhanced probability of giant planet formation at those radii ( papaloizou & terquem 1999 ) .
current radial velocity surveys are sensitive to massive planets at radii @xmath111 , i.e. within this region .
if migration is as rapid as currently suspected , a consequence of the layered disc model would probably be fewer planets at larger radii than would be expected from an extrapolation from the current data .
in this paper , we have presented models for the evolution of magnetically layered protoplanetary discs . given our present understanding of angular momentum transport in discs ,
this model represents the best guess for the structure of protoplanetary discs at radii of the order of 1 au , where the gas is cool and poorly coupled to the magnetic field .
the evolutionary models presented here suggest a number of important differences with non - layered viscous disc models . 1
. the disc can not be in a steady state for outer disc accretion rates in the range @xmath112 .
a limit cycle is obtained , in which heating when the layered region of the disc becomes self - gravitating periodically restarts mhd turbulence , leading to outbursts of accretion onto the star .
all calculations to date of layered discs obtain strongly episodic accretion , despite variations in the physical processes included in the models ( gammie 1999 ; stepinski 1999 ) .
this is the most robust prediction of the model .
2 . the duration of outbursts depends on where in the quiescent disc they are triggered . in our model ,
the outbursts are long , with duration @xmath113 .
they could drive repeated strong episodes of mass outflow from the inner disc , resulting in ` pulsing ' of the observed jets .
the time - averaged accretion rate is reduced due to the bottleneck created by the layered region of the disc . as a result ,
the disc mass at late times is large , typically @xmath114 at 1 - 2 myr .
most of this mass is locked up in the quiescent layer of the disc at small radii .
the central temperature is strongly modified , and often _ increases _ with radius , near the transition between the layered disc and the outer self - gravitating region .
this could slow the otherwise rapid rate of low mass planetary migration at radii @xmath115 .
5 . a low viscosity in the layered region of the disc
affects the migration rate and eccentricity evolution of massive planets at the radii currently probed by radial velocity surveys .
current observations provide only very limited constraints on the properties of protoplanetary discs at the radii of greatest interest for planet formation . as a result ,
there remains substantial uncertainty in our knowledge of how these discs evolve .
models , such as this one , that attempt to include more realistic disc physics , can lead to very different environments for planet formation and migration than the highly simplified models usually considered .
the most obvious observational predictions of the model are that some very young stars should be accreting at the low ` quiescent ' rate of the order of @xmath116 , and that high accretion rate outbursts should continue , albeit with lesser frequency , during a substantial fraction of the classical t tauri phase .
pja thanks the institute of astronomy for their hospitality during the course of part of this work , and charles gammie , john papaloizou and henk spruit for helpful discussions .
this work was partially supported by the training and mobility of researchers ( tmr ) program of the european commission .
ml acknowledges support from nasa grant nag5 - 6857 .
jep is grateful to stsci for continuing support under their visitors program . | we study protoplanetary disc evolution assuming that angular momentum transport is driven by gravitational instability at large radii , and magnetohydrodynamic ( mhd ) turbulence in the hot inner regions . at radii of the order of 1 au such discs develop a magnetically layered structure , with accretion occurring in an ionized surface layer overlying quiescent gas that is too cool to sustain mhd turbulence .
we show that layered discs are subject to a limit cycle instability , in which accretion onto the protostar occurs in @xmath0 bursts with @xmath1 , separated by quiescent intervals lasting @xmath2 where @xmath3 .
such bursts could lead to repeated episodes of strong mass outflow in young stellar objects .
the transition to this episodic mode of accretion occurs at an early epoch ( @xmath4 ) , and the model therefore predicts that many young pre - main - sequence stars should have low rates of accretion through the inner disc . at ages of a few myr , the discs are up to an order of magnitude more massive than the minimum mass solar nebula , with most of the mass locked up in the quiescent layer of the disc at @xmath5 .
the predicted rate of low mass planetary migration is reduced at the outer edge of the layered disc , which could lead to an enhanced probability of giant planet formation at radii of 1 3 au .
psfig accretion , accretion discs mhd stars : pre - main - sequence stars : formation solar system : formation planets and satellites |
in contrast to most astronomical objects ( such as planets , stars and galaxies ) , the universe as a whole is usually considered to be non - rotational . however , the possibility that the universe rotates should not be ignored , since solutions of gr corresponding to a rotating universe have been found @xcite indicating that a global rotation is physically allowed .
although it is widely assumed that galaxies align randomly and have zero net angular momentum , there have been many investigations on the general alignment of galaxies .
some even claim that a weak alignment of galaxies does exist@xcite .
such an alignment may be used to explain@xcite the recently discovered non - gaussian properties@xcite of the cmba .
furthermore , jaffe et al .
@xcite suggest that the bianchi type vii@xmath3 model with a global rotation may be used to explain some anomalies of the cmba .
in addition , the existence of a global rotation may contradict the inflationary model of the early universe @xcite and mach s principle @xcite .
rotational perturbations may also be used to determine whether or not the universe is open or closed @xcite .
therefore , the study of global rotation is of interest in many different aspects of cosmology , and constraint of the rotation speed of the universe is important .
the most popular approach to constrain the magnitude of the global rotation speed is to make use of data from the cmba because of their precision .
most discussions @xcite have focused on homogeneous cosmological models , i.e. bianchi models @xcite . to be consistent with obeservations , only bianchi type v , vii@xmath4 , vii@xmath3 and ix models , which include the robertson - walker model as a special case , are considered .
the constraints of the global rotation speed obtained depend on the parameters of the models . besides , shear and vorticity are inseparable in these works @xcite , i.e. zero shear automatically implies zero vorticity .
there are many other approaches to constrain the global rotation . based on the idea that a global rotation induces a total net spin of galaxies , the global rotation
can be limited @xcite . moreover ,
empirical relations between angular momenta and mass of galaxies / clusters , such as @xmath5 for spiral galaxies and @xmath6 for clusters can be explained by the global rotation @xcite .
the acceleration caused by the global rotation may be used to explain parts of the accelerating expansion of our universe , and thus the global rotation can be constrained by supernova type ia data @xcite .
recently , some studies of the cmb polarization induced by the global rotation are published @xcite providing potential constraints in the future . to develop a model that preserves the homogeneity and isotropy of the mean cmb
, we study the rotation of the universe as a perturbation in the robertson - walker framework with a cosmological constant in this paper .
unlike the bianchi models , such an approach allows to have non - zero rotation but trivial shear .
since the global rotation does not have any influences on the 1st - order sachs - wolfe effect ( sw effect ) , we need to calculate the metric up to 2nd - order perturbations and the 2nd - order sw effect .
then , we will constrain the angular speed of the rotation using recent data on cmba @xcite .
our model is inhomogeneous with an axial symmetry in general .
the global rotation in our model is not only time - dependent but also radial - dependent .
the line element of a flat rotational universe possesses an axial symmetry and can be written in the form of @xcite @xmath7d\eta^2-[1-h(r,\eta)]dr^2-[1-h(r,\eta)]r^2d\theta^2- [ 1-k(r,\eta)]dz^2\nonumber\\ & ~&+2r^2a(\eta)\omega(r,\eta)d\theta d\eta\},\end{aligned}\ ] ] where @xmath8 and @xmath9 , @xmath10 , @xmath11 is the conformal time defined by @xmath12 with @xmath13 the cosmological time , @xmath14 , @xmath15 and @xmath16 are the cylindrical coordinates in the comoving frame of the universe , @xmath17 is the axis of rotation , @xmath18 is the scale factor of the universe with @xmath19 at the present time , @xmath20 is the angular velocity of the metric observed from an inertial frame whose origin is on the rotational axis , and @xmath21 , @xmath22 and @xmath23 are the perturbations on the ( 0 , 0 ) , ( 1 , 1 ) , ( 2 , 2 ) , ( 3 , 3)-components of the metric due to the rotation . because of the cylindrical symmetry , the perturbation functions due to the rotation are also independent of @xmath15 and @xmath16 . here , we assume that the norm of @xmath24@xmath25@xmath26 , where @xmath26 is the unperturbed metric , is much smaller than that of @xmath24 . explicitly , we assume that the rotation is slow , so that @xmath27 , and we can think of @xmath28 for @xmath14 , @xmath11 within the last scattering surface as the perturbation parameter . by parity consideration , we can see that @xmath20 is composed of only odd powers of @xmath29 , whereas @xmath21 , @xmath22 and @xmath23 , being density and pressure perturbations , only even powers .
since we are interested only up to second - order perturbations , we will consider @xmath20 to be first - order and @xmath21 , @xmath22 and @xmath23 to be second - order .
the metric eq .
( 1.1 ) in ref . @xcite will be recovered if we truncate ours up to the first - order .
since the effect of the rotation on the cmba is independent of the parity , we expect that the sw effect due to rotation occurs in even orders of @xmath29 only .
the einstein field equations ( efes ) for a universe with cosmological constant @xmath0 are @xmath30 where @xmath31 is the stress - energy tensor for a perfect fluid , @xmath32 is the ricci curvature tensor , @xmath33 is the scalar curvature , @xmath34 is the mass - energy density , @xmath35 is the pressure and @xmath36 is the four - velocity of the fluid in the comoving frame . here , we set @xmath37 . if @xmath38 , the universe is homogeneous and @xmath39 expresses the angular velocity of the universe observed anywhere in the comoving frame . otherwise , the universe is inhomogeneous and the observer at the rotating axis passing through the origin is distinct . to solve eq .
( 3 ) up to 2nd - order in @xmath29 , we expand all quantities : @xmath40 where the subscripts indicate the corresponding orders of perturbations .
the zeroth - order efes give rise to the standard friedmann equations : @xmath41 once the equation of state ( eos ) of the fluid is given , we can determine the scale factor @xmath18 , the density @xmath42 and the pressure @xmath43 with the equations above . in this paper
, we consider a universe with @xmath44 .
however , the following formalism can be applied to any fluid with a specified eos . from the temporal - spatial efes and the condition @xmath45
, we have @xmath46 and @xmath47 .
the first - order efes then give : @xmath48=\frac{a^4(\eta)}{8\dot{a}^2(\eta)- 4a(\eta)\ddot{a}(\eta)}[3r\omega ' ( r,\eta)+r^2\omega ' ' ( r,\eta)],\\ 0&=&3\dot{a}(\eta)\omega ' ( r,\eta)+a(\eta)\dot{\omega}'(r,\eta),\end{aligned}\ ] ] where the dots refer to derivatives with respect to the conformal time @xmath11 , and primes mean derivatives with respect to @xmath14 .
as seen from the equations above , a 1st - order rotational perturbation can not generate 1st - order perturbations of the mass - energy density and pressure .
this is expected because @xmath34 and @xmath35 should be unchanged under the inversion of the rotation .
for the same reason , @xmath49 . from eq .
( 11 ) , we see that if @xmath20 is independent of @xmath14 , then @xmath50 .
that is , the fluid in the universe rotates with the metric at the same pace .
nevertheless , an @xmath14-dependent @xmath20 allows us to discuss the centrifugal force for the universe as that discussed in ref .
@xcite for relativistic stars .
( 12 ) implies that @xmath20 must be in the form of @xmath51 , where @xmath52 and @xmath53 are arbitrary functions .
moreover , if the fluid is viscous , the r.h.s . of eq .
( 12 ) will be equal to the 1st - order shear term of @xmath54 and this will free @xmath20 from the form above . without loss of generality , we perform the following transformations : @xmath55 where @xmath56 and @xmath57 are arbitrary functions depending on @xmath21 and @xmath22 .
the first term of @xmath21 comes from the transformation @xmath58 . using these transformations to formulate the second - order efes
, we find that @xmath59 ^ 2}{r^2a^2(\eta ) } + k(r,\eta)+t(r,\eta)\right\},\\ _
2u_1(r,\eta)&=&\frac{-a^2(\eta)}{8\dot{a}^2(\eta)-4a(\eta)\ddot{a}(\eta ) } \{-\dot{a}(\eta)[2k'(r,\eta)+2t'(r,\eta)]\nonumber\\ & ~&+a(\eta)[-2\dot{k}'(r,\eta)+\dot{l}'(r,\eta)]\},\\
_ 2\rho(r,\eta)&=&-\frac{1}{32\pi a^2(\eta)}\{-4\lambda a^2(\eta)[k(r,\eta)+t(r,\eta ) ] + \frac{4\dot{a}(\eta)[3\dot{k}(r,\eta)-2\dot{l}(r,\eta)]}{a(\eta)}\nonumber\\ & ~&-\frac{2[2k'(r,\eta)-l'(r,\eta)]}{r}+12ra^2(\eta)\omega(r,\eta)\omega ' ( r,\eta)+r^2a^2(\eta)\omega ' ^2(r,\eta)-4k''(r,\eta)\nonumber\\ & ~&+2l''(r,\eta)+4r^2a^2(\eta)\omega(r,\eta)\omega ' ' ( r,\eta ) + 4\left[\lambda a^2(\eta)+\frac{3\dot{a}^2(\eta)}{a^2(\eta)}\right]\{k(r,\eta ) + t(r,\eta)\nonumber\\ & ~&-r^2a^2(\eta)\omega^2(r,\eta)+\left[ra(\eta)\omega(r,\eta)- \frac{_1u_2(r,\eta)}{ra(\eta)}\right]^2\}\},\\ _ 2p(r,\eta)&=&\frac{1}{32\pi ra^4(\eta)}\ { 4ra(\eta)[3\dot{a}(\eta)\dot{k}(r,\eta ) -\dot{a}(\eta)\dot{l}(r,\eta)+\dot{a}(\eta)\dot{t}(r,\eta)]+ 2a^2(\eta)[2r\ddot{k}(r,\eta)\nonumber\\ & ~&-r\ddot{l}(r,\eta)+t'(r,\eta ) ] + ra^4(\eta)[-4\lambda k(r,\eta)-4\lambda t(r,\eta)+r^2\omega ' ^2(r,\eta)]\},\\ _ 2p(r,\eta)&=&\frac{1}{32\pi r^2a^6(\eta)}\{-12\dot{a}^2(\eta)[_1u_2(r,\eta)]^2 + 4r^2a^3(\eta)[3\dot{a}(\eta)\dot{k}(r,\eta ) -\dot{a}(\eta)\dot{l}(r,\eta)\nonumber\\ & ~&+\dot{a}(\eta)\dot{t}(r,\eta)]-r^2a^6(\eta)[4\lambda k(r,\eta)+4\lambda t(r,\eta)+3r^2\omega ' ^2(r,\eta)]\nonumber\\ & ~&+a^4(\eta)\{-4\lambda[_1u_2(r,\eta)]^2 + 2r^2[2\ddot{k}(r,\eta)-\ddot{l}(r,\eta)+t''(r,\eta)]\}\},\\
_ 2p(r,\eta)&=&\frac{1}{32\pi ra^4(\eta)}\ { 4ra(\eta)\dot{a}(\eta)[3\dot{k}(r,\eta)-2\dot{l}(r,\eta)+\dot{t}(r,\eta ) ] -ra^4(\eta)[4\lambda k(r,\eta)\nonumber\\ & ~&+4\lambda t(r,\eta)+r^2\omega ' ^2(r,\eta ) ] + 2a^2(\eta)[2r\ddot{k}(r,\eta)-2r\ddot{l}(r,\eta)+l'(r,\eta)+t'(r,\eta)\nonumber\\ & ~&+rl''(r,\eta)+rt''(r,\eta)]\}.\end{aligned}\ ] ] eqs .
( 19)-(21 ) are three different expressions for @xmath60 derived by the three 2nd - order spatial - spatial efes .
( 15 ) is expected by considering the symmetries of @xmath15(odd)- , @xmath16(even)-components of the four - velocity @xmath61 under the inversion of the rotation .
@xmath62 , which is non - zero in general and corresponds to the dynamical changes of @xmath63 and @xmath60 for an @xmath14-dependent rotational speed . in order to calculate these 2nd - order perturbations
, we have to find the solutions of @xmath23 , @xmath57 and @xmath56 .
since the pressure is the same along different directions at one point , eqs .
( 19)-(21 ) are equivalent .
substracting eqs .
( 19)-(21 ) from each other leads to two equations for solving @xmath57 and @xmath56 when @xmath20 is specified while @xmath23 is regarded as an arbitrary function independent of the rotation .
the detailed derivations of these solutions are shown in appendix i.
as the sw effect is invariant under the inversion of the rotation , the first non - zero sw effect due to rotation occurs in 2nd - order perturbations .
the general formalism of the 2nd - oder sw effect has been comprehensively discussed . in the following ,
we will make use of the ideas in @xcite and derive the 2nd - order sw effect of a rotating universe .
the cosmic microwave background ( cmb ) temperature observed at the origin towards a direction @xmath64 can be written as @xmath65 where @xmath66 , the subscripts ( @xmath67 and @xmath68 ) denoting the origin and the last scattering hypersurface ( lsh ) respectively , @xmath36 is the four - velocity of the fluid in the comoving frame , @xmath69 is the wave vector of a light ray in the conformal metric with an affine parameter @xmath70 , @xmath71 is the temperature measured at the point @xmath72 on the lsh , and @xmath73 is the direction of the light ( passing through the point @xmath72 ) observed at the origin .
we show in appendix ii that the 1st - order sw effect due to rotation is zero and the 2nd - order sw effect is @xmath74|_{\eta_{\epsilon}}^{\eta_0}\nonumber\\ & = & \frac{\omega_{\lambda}\sin\phi}{2\lambda(1-\omega_{\lambda } ) } [ 2\dot{a}(\eta_{\epsilon})t'(r_{\epsilon},\eta_{\epsilon})- a(\eta_{\epsilon})\dot{l}'(r_{\epsilon},\eta_{\epsilon})]\nonumber\\ & ~&-\frac{1}{2}\left\{\frac{\omega_{\lambda}^2a^4(\eta_{\epsilon})}{4\lambda^2(1-\omega_{\lambda})^2 } [ 3\omega ' ( r_{\epsilon},\eta_{\epsilon})+r_{\epsilon}\omega ' ' ( r_{\epsilon},\eta_{\epsilon})]^2+t(r_{\epsilon},\eta_{\epsilon})\right\}\nonumber\\ & ~&+\int_{\eta_{\epsilon}}^0\left[-\frac{\dot{t}(-\lambda\sin\phi,\lambda)}{2}+ t'(-\lambda\sin\phi,\lambda)\sin\phi-\frac{\dot{l}(-\lambda\sin\phi,\lambda)}{2}\sin^2\phi\right]d\lambda,\end{aligned}\ ] ] where @xmath75 and @xmath76 denotes the conformal time of the last scattering .
( 23 ) determines the cmba produced by the rotation of the universe once @xmath20 is specified . as an example , we consider the simplest case stationary homogeneous rotation ( i.e. @xmath77 , @xmath53 is an arbitrary function ) .
then , we have @xmath78 and @xmath79 where @xmath80 is a constant .
it is straight - forward to find that @xmath81 to explain this , we recall that @xmath82 if @xmath20 is independent of @xmath14 , which means that the fluid is rotating with the same phase as the metric .
therefore , the effect of the rotating metric cancels the relativistic doppler effect caused by the sources rotating in a stationary metric .
we make use of the previous example , i.e. @xmath83 with @xmath84 in eqs .
( a.10)-(a.18 ) , to constrain the rotation of the universe .
using eqs .
( 23 ) , ( a.10 ) , ( a.16)-(a.18 ) , we expand the cmba as @xmath85 the values of @xmath86 s are listed in table 1 .
+ * table 1 * + [ cols="<,<,<,<,<",options="header " , ] + we notice that the spherical harmonic expansion has non - zero coefficients only when @xmath87 and even @xmath88 for which @xmath89 has cylindrical and parity symmetries .
the result is unlikely to be related to the ` axis of evil ' @xcite , a preferred direction of several low multipoles ( especially quadrupole and octopole ) .
however , in the general case , when we are located off the rotational axis , the cylindrical symmetry is broken and non - zero coefficients for other multipoles are allowed .
for example , the cmba on the two sides of the rotation axis will be affected differently by the rotation in general .
such an asymmetric effect enhances the dipole moment of the cmba .
thus , its potential to explain the ` axis of evil ' can not be eliminated without further study . with @xmath90 , @xmath91km
/ s / mpc and @xmath92 , we constrain @xmath29 to be less than @xmath93 m in si unit . that is , @xmath94 is less than @xmath95 rad yr@xmath2 in usual unit at the last scattering surface . some cmba maps generated with the rotation of the universe are shown in fig . 1 as examples .
nevertheless , our result can be regarded as the first constraint of the rotation of a @xmath0cdm universe . in fig . 2 ,
some normalized 2nd - order perturbed quantities along the light path of the last scattered photons are plotted as a function of @xmath96 with @xmath83 and @xmath97 m. as expected , the perturbed quantities increase with the rotating speed . in fig . 3 , the angular velocity of matter @xmath98 and its difference from that of the metric along the light path of the last scattered photons
are plotted against time .
we can see that the angular velocity of matter can be negative while the rotation speed of the universe is always positive . because of the @xmath14-dependence of @xmath20 , the angular velocity of matter can be different from that of the metric in general as indicated in eq .
these quantities are useful for studying the frame - dragging of the universe in the future .
the distributions of the 2nd - order perturbed densities of matter are shown at two different times in fig .
4 . as shown in fig . 2
, @xmath99 is always positive , which means that matter is moving away from the rotating axis and hence the density is expected to be decreasing with time ( shown in fig . 4 ) . under the mollweide projection with the z - direction pointing to @xmath100 with @xmath97 m , which is the maximum allowed by current cmb data .
the middle map shows the original 5-year wmap map @xcite , and the bottom map is a combined map of the two above.,title="fig:",width=340,height=188 ] under the mollweide projection with the z - direction pointing to @xmath100 with @xmath97 m , which is the maximum allowed by current cmb data .
the middle map shows the original 5-year wmap map @xcite , and the bottom map is a combined map of the two above.,title="fig:",width=340,height=188 ] under the mollweide projection with the z - direction pointing to @xmath100 with @xmath97 m , which is the maximum allowed by current cmb data .
the middle map shows the original 5-year wmap map @xcite , and the bottom map is a combined map of the two above.,title="fig:",width=340,height=188 ] along the light paths of the last - scattered photons , where @xmath101 and @xmath102 .
we set @xmath103.,width=604,height=453 ] and its difference from the rotation speed of the universe are plotted against time .
similar to fig . 2 ,
@xmath101 , @xmath102 and @xmath103 .
we note that the rotation of matter in the universe can be different and even opposite to the rotation of the universe.,width=604,height=453 ] .,width=604,height=453 ]
in this paper , we have developed a cosmological model that has a non - zero rotation but trivial shear in the robertson - walker framework with a cosmological constant .
we have solved the efe s up to 2nd - order perturbations of a flat @xmath0cdm universe with rotation as a 1st - order perturbation .
we also set up the formulation for the 2nd - order sw effect due to the rotational perturbation and find that the effect only influences the spherical harmonics with even @xmath88 s . by making use of recent cmba data
, the angular speed of the rotation is constrained to be less than @xmath1 rad yr@xmath2 at the last scattering surface .
the model of the universe here is different from the bianchi models used in the literatures .
first of all , our model is inhomogeneous with an axial symmetry in general while bianchi models are homogeneous .
moreover , our model is shear - free and thus has the advantage that the sw effect and the constraint obtained are purely due to the global rotation .
compared with previous works , the constraint is much weaker .
for example , barrow et al .
@xcite put a constraint of @xmath104 rad yr@xmath2 on the rotaion of flat bianchi models .
this can be understood mainly because the effects of rotation in our model here show up as 2nd - order sw effects while in previous works they are 1st - order sw effects . for further study
, we notice that our model here produces a 2nd - order outward radial velocity .
it may be used to explain parts of the accelerating expansion of the universe , and therefore , constraints on the global rotation can be obtained with type ia supernova data as proposed in @xcite .
although we only study a flat universe here , it is interesting to study the closed and open cases of our model in view of the significantly different constraints on closed and open bianchi models@xcite .
@xmath105 ^ 2 } \{64r^3\dot{a}^4(\eta)\omega ' ^2(r,\eta ) + 16r^3a^2(\eta)\ddot{a}^2(\eta)\omega ' ^2(r,\eta)\nonumber\\ & ~&+r\lambda a^6(\eta)[3\omega ' ( r,\eta)+r\omega ' ' ( r,\eta)]^2 + ra(\eta)\dot{a}^2(\eta)\{-64r^2\ddot{a}(\eta)\omega ' ^2(r,\eta)\nonumber\\ & ~&+3a(\eta)[3\omega ' ( r,\eta)+r\omega ' ' ( r,\eta)]^2\}\},\\ 0&=&2r\dot{a}^2(\eta)l(r,\eta)+ra(\eta)[-4\ddot{a}(\eta)l(r,\eta)+ 2\dot{a}(\eta)\dot{l}(r,\eta)]\nonumber\\ & ~&+a^4(\eta)[-2r\lambda l(r,\eta ) + r^3\omega ' ^2(r,\eta)]+a^2(\eta)[r\ddot{l}(r,\eta)-l'(r,\eta)\nonumber\\ & ~&-rl''(r,\eta)-rt''(r,\eta)].\end{aligned}\ ] ] these two equations are independent of @xmath23 . that is , @xmath23 is only an arbitrary function unrelated to @xmath20 in general .
physically , @xmath23 comes from the 2nd - order perturbation of mass density ( analogous to the diagonal perturbations of schwarzschild metric ) .
as we are interested in the effects of the rotation only , we set @xmath106 for simplicity . in this paper , we focus on a non - viscous fluid in the universe . using the fact that @xmath107 $ ] for a flat @xmath0cdm universe and eq .
( 8) , we can simplify these two equations further @xmath108 ^
2}{2\lambda(1-\omega_{\lambda})a^3(\eta)},\\ \frac{r}{a(\eta)}\frac{d}{d\eta}[a^2(\eta)\dot{l}(r,\eta)]-a(\eta)[l'(r,\eta ) + rl''(r,\eta)]=ra(\eta)t''(r,\eta)-r^3a^{-3}(\eta)a'^2(r).\end{aligned}\ ] ] here , we remark that @xmath53 , which disappears from these two equations , is arbitrary and does not affect the sachs - wolfe effect .
such an arbitrariness is unrelated to the effects of the rotational universe on the cmba . without loss of generality ,
we expand @xmath52 into taylor series : @xmath109 ( we will explain below why it does not start from @xmath110 ) . substituting the series into eq .
( a.3 ) , @xmath111 where @xmath112 for @xmath113 . by separation of variables , we find that @xmath114,\end{aligned}\ ] ] where @xmath115 as @xmath116 and @xmath117 are arbitrary functions of the homogeneous solutions for eq .
( a.5 ) and thus are independent of @xmath20 , we are free to set them zero .
the series of @xmath52 starts from @xmath118 because the @xmath110 term can be absorbed into @xmath53 while the @xmath119 term , which produces @xmath120 as the particular solution of @xmath56 , is rejected because of the singularity at @xmath121 .
similarly , we expand @xmath122 and substitute it into eq .
( a.4 ) : @xmath123 -(n+1)^2a(\eta)e_{n+1}(\eta)\right\}r^n\nonumber\\ & = & -\sum_{n=5}^{\infty}\sum_{l=2}^{\infty}\frac{l(3n-1)(n - l-1)}{(n-1)a^3(\eta ) } c_lc_{n - l-1}r^{n}\nonumber\\ & ~&-\sum_{n=3}^{\infty}\sum_{l=2}^{\infty } \frac{nl(l+2)(n - l+1)(n - l+3)\omega_{\lambda}}{2\lambda(1-\omega_{\lambda})(n-1 ) a^2(\eta)}c_lc_{n - l+1}r^{n},\end{aligned}\ ] ] where @xmath124 for @xmath125 .
in general , by comparing the @xmath126 terms on both sides , we can obtain the recurrence relations for solving @xmath127 . as an example , we work out the simplest case where @xmath128 ( @xmath129 is a constant ) : @xmath130 - 4a(\eta)e_2(\eta),\\ -\frac{48\omega_{\lambda}\alpha^2}{\lambda(1-\omega_{\lambda})a^2(\eta)}&=&\frac{1}{a(\eta ) } \frac{d}{d\eta}[a^2(\eta)\dot{e}_2(\eta)]-16a(\eta)e_4(\eta),\\ -\frac{14\alpha^2}{a^3(\eta)}&=&\frac{1}{a(\eta ) } \frac{d}{d\eta}[a^2(\eta)\dot{e}_4(\eta)]-36a(\eta)e_6(\eta),\\ & \vdots&\nonumber\\ 0&=&\frac{1}{a(\eta)}\frac{d}{d\eta}[a^2(\eta)\dot{e}_{2n-2}(\eta)]-4n^2a(\eta)e_{2n}(\eta).\end{aligned}\ ] ] we notice that there is a freedom to choose one of @xmath131 arbitrarily , independent of @xmath20 . to prevent infinite series ,
we set @xmath132 so that @xmath133 for @xmath134 and all non - zero @xmath131 s depend on @xmath135 ( due to the rotation ) . we have @xmath136d\eta ' ' d\eta ' , \\
e_0(\eta)&=&\int_{\eta}^0\frac{1}{a^2(\eta ' ) } \int_{\eta ' } ^0 4a^2(\eta ' ' ) e_2(\eta ' ' ) d\eta ' ' d\eta ' , \end{aligned}\ ] ] which can be solved numerically .
( b.1 ) , we need to solve the geodesic equations for the light rays of the cmb , which are @xmath139 as before , we expand the geodesic equations into different orders of @xmath20 . to be consistent , @xmath140 has to be expanded into @xmath141 for the first - order , @xmath144=0,\\ \frac{d}{d\lambda}\left[_0g_{ii}\frac{d_1x^i}{d\lambda}\right]=0,\end{aligned}\ ] ] for @xmath145 . therefore , @xmath146 where @xmath86 s are constants to be determined . for simplicity , we assume that we are located on the rotating axis .
therefore , @xmath102 and @xmath101 due to the cylindrical symmetry .
although the general case that we may be off the rotating axis is more realistic , the constraint here can be regarded as a good approximation provided that our distance to the rotating axis is small compared to that of the last scattering surface .
the general case can be found by assigning a suitable dependence of @xmath15 on @xmath147 and @xmath148 .
the term @xmath149 refers to the angular velocity of the comoving metric . for the second - order , @xmath150 by setting @xmath106 and the requirement of a null geodesic ( @xmath151 )
, we obtain @xmath152 where @xmath153\,d\lambda ' , \\ _ 2k^1(\lambda)&=&\int\left[\frac{t'(r_{\lambda ' } , \eta_{\lambda ' } ) } { 2}+ \frac{l'(r_{\lambda ' } , \eta_{\lambda ' } ) } { 2}\sin^2\phi -\dot{l}(r_{\lambda ' } , \eta_{\lambda ' } ) \sin\phi\right]\,d\lambda ' .\end{aligned}\ ] ] to have a null geodesic , @xmath154 for @xmath155 .
from the results of @xmath156 and @xmath157 , we can easily verify the argument that the 1st - order perturbations of sw effect due to the rotation is zero and calculate the 2nd - order sw effect as @xmath74|_{\eta_{\epsilon}}^{\eta_0}\nonumber\\ & = & \frac{\omega_{\lambda}\sin\phi}{2\lambda(1-\omega_{\lambda } ) } [ 2\dot{a}(\eta_{\epsilon})t'(r_{\epsilon},\eta_{\epsilon})- a(\eta_{\epsilon})\dot{l}'(r_{\epsilon},\eta_{\epsilon})]\nonumber\\ & ~&-\frac{1}{2}\left\{\frac{\omega_{\lambda}^2a^4(\eta_{\epsilon})}{4\lambda^2(1-\omega_{\lambda})^2 } [ 3\omega ' ( r_{\epsilon},\eta_{\epsilon})+r_{\epsilon}\omega ' ' ( r_{\epsilon},\eta_{\epsilon})]^2+t(r_{\epsilon},\eta_{\epsilon})\right\}\nonumber\\ & ~&+\int_{\eta_{\epsilon}}^0\left[-\frac{\dot{t}(-\lambda\sin\phi,\lambda)}{2}+ t'(-\lambda\sin\phi,\lambda)\sin\phi-\frac{\dot{l}(-\lambda\sin\phi,\lambda)}{2}\sin^2\phi\right]d\lambda,\end{aligned}\ ] ] where @xmath75 and @xmath76 denotes the occuring conformal time of the last scattering .
we made use of healpix @xcite to produce fig .
this work is supported by grants from the research grant council of the hong kong special administrative region , china ( project nos .
400707 and 400803 ) .
20 k. gdel , rev .
modern phys .
* 21 * , 3 ( 1949 ) .
s. hawking , mnras * 142 * , 129 ( 1969 ) .
j. d. barrow @xmath158 @xmath159 , mnras * 213 * , 917 ( 1985 ) .
a. de oliveira - costa @xmath158 @xmath159 , phys .
d * 69 * , 063516 ( 2004 ) .
f. k. hansen @xmath158 @xmath159 , astrophys .
j. * 607 * , l67 ( 2004 ) .
p. vielva @xmath158 @xmath159 , astrophys .
j. * 609 * , 22 ( 2004 ) .
h. k. eriksen @xmath158 @xmath159 , astrophys .
j. * 612 * , 633 ( 2004 ) .
d. j. schwarz @xmath158 @xmath159 , phys .
lett . * 93 * , 221301 ( 2004 ) .
k. land @xmath158 @xmath159 , phys .
. lett . * 95 * , 071301 ( 2005 ) . c. j. copi
@xmath158 @xmath159 , phys .
d * 70 * , 043515 ( 2004 ) .
t. r. jaffe @xmath158 @xmath159 , astrophys .
j. * 644 * , 701 ( 2006 ) .
t. r. jaffe @xmath158 @xmath159 , preprint , [ astro - ph/0606046 ] .
r. k. sachs and a. m. wolfe , astrophys .
j. * 147 * , 73s ( 1967 ) . t. pyne and m. birkinshaw , astrophys .
j. * 415 * , 459 ( 1993 ) .
t. pyne and s. m. carroll , phys .
d * 53 * , 6 ( 1996 ) .
s. mollerach and s. matarrese , phys .
d * 56 * , 8 ( 1997 ) . s. matarrese @xmath158 @xmath159 , phys .
d * 58 * , 043504 ( 1998 ) .
j. d. barrow and c. g. tsagas , class .
quantum grav .
* 21 * , 1773 ( 2004 ) .
n. kashikawa and s. okamura , pasj * 44 * , 493 ( 1992 ) .
w. godlowski , mnras * 271 * , 19 ( 1994 ) .
m. j. longo , arxiv preprint , [ astro - ph/0703325 ] .
j. ellis and k. olive , nature * 303 * , 679 ( 1983 ) .
j. d. barrow , mnras * 179 * , 47p ( 1977 ) .
j. d. barrow and a. r. liddle , gen .
* 29 * , 1501 ( 1997 ) .
f. m. gomide , res .
* 3 * , 3 ( 1973 ) .
d. f. raine , mnras * 171 * , 507 ( 1975 ) .
j. d. barrow , canadian j. phys . * 4 * , 152 ( 1986 ) .
e. f. bunn @xmath158 @xmath159 , phys .
lett . * 77 * , 2883 ( 1996 ) .
t. ghosh @xmath158 @xmath159 , phys .
d * 75 * , 083007 ( 2007 ) .
g. f. r. ellis and m. a. h. maccallum , commun .
* 12 * , 108 ( 1969 )
. y. n. obukhov @xmath158 @xmath159 , phys .
d * 66 * , 043518 ( 2002 ) .
* 30 * , 497 ( 1998 ) . w. godlowski . and m. szydlowski , gen .
. grav . * 35 * , 2171 ( 2003 ) . w. godlowski . and m. szydlowski , arxiv preprint , [ astro - ph/0303248v1 ] .
a. pontzen and a. challinor , mnras * 380 * , 1387 ( 2007 ) . http://lambda.gsfc.nasa.gov j. b. hartle and d. h. sharp , astrophys .
j. * 147 * , 317h ( 1967 ) .
s. s. bayin and f. i. cooperstock , phys .
d * 22 * , 10 ( 1980 ) .
j. b. hartle , astrophys .
j. * 150 * , 1005h ( 1967 ) . j. b. hartle and k. s. thorne , astrophys . j. * 153 * , 807h ( 1968 ) .
healpix is available at http://healpix.jpl.nasa.gov | models of a rotating universe have been studied widely since gdel @xcite , who showed an example that is consistent with general relativity ( gr ) . by now , the possibility of a rotating universe has been discussed comprehensively in the framework of some types of bianchi s models , such as type v , vii and ix @xcite , and different approaches have been proposed to constrain the rotation .
recent discoveries of some non - gaussian properties of the cosmic microwave background anisotropies ( cmba ) @xcite , such as the suppression of the quadrupole and the alignment of some multipoles draw attention to some bianchi models with rotation @xcite .
however , cosmological data , such as those of the cmba , strongly prefer a homogeneous and isotropic model .
therefore , it is of interest to discuss the rotation of the universe as a perturbation of the robertson - walker metric , to constrain the rotating speed by cosmological data and to discuss whether it could be the origin of the non - gaussian properties of the cmba mentioned above . here
, we derive the general form of the metric ( up to 2nd - order perturbations ) which is compatible with the rotation perturbation in a flat @xmath0-cdm universe . by comparing the 2nd - order sachs - wolfe effect @xcite due to rotation with the cmba data , we constrain the angular speed of the rotation to be less than @xmath1 rad yr@xmath2 at the last scattering surface .
this provides the first constraint on the shear - free rotation of a @xmath0cdm universe . |
thermotropic liquid crystals exhibit exotic phase behavior upon temperature variation .
the nematic phase is rich with a long - ranged orientational order but lacks translational order . the isotropic - nematic (
i - n ) phase transition , which is believed to be weakly first order in nature with certain characteristics of the continuous transition , has been a subject of immense attention in condensed matter physics and material sciences @xcite .
in contrast , the dynamics of thermotropic liquid crystals have been much less studied , the focus being mostly on the long - time behavior of orientational relaxation near the i - n transition @xcite . a series of optical kerr effect ( oke ) measurements have , however , recently studied collective orientational relaxation in the isotropic phase near the i - n transition over a wide range of time scales @xcite .
the dynamics have been found to be surprisingly rich , the most intriguing feature being the power law decay of the oke signal at short - to - intermediate times @xcite .
the relaxation scenario appears to be strikingly similar to that of supercooled molecular liquids @xcite , even though the latter do not undergo any thermodynamic phase transition .
although the analogous dynamics have been investigated in subsequent studies @xcite , a quantitative estimation of glassy dynamics of rodlike molecules near the i - n transition still eludes us .
the prime objective of this paper is to provide a quantitative measure of glassy dynamics near the i - n transition . to this end , we have undertaken molecular dynamics simulations of a family of model systems consisting of rodlike molecules across the i - n transition in search of glassy behavior . given the involvement of the phase transition to an orientationally ordered mesophase upon lowering the temperature , we choose to probe the single - particle orientational dynamics .
we have defined a fragility index and explored plausible correlation of the features of the underlying energy landscape with the observed fragility in analogy with supercooled liquids .
the systems we have studied consist of ellipsoids of revolution . the gay - berne ( gb ) pair potential @xcite , that is well established to serve as a model potential for systems of thermotropic liquid crystals , has been employed .
the gb pair potential , which uses a single - site representation for each ellipsoid of revolution , is an elegant generalization of the extensively used isotropic lennard - jones potential to incorporate anisotropy in both the attractive and the repulsive parts of the interaction @xcite.in the gb pair potential , @xmath0th ellipsoid of revolution is represented by the position @xmath1 of its center of mass and a unit vector @xmath2 along the long axis of the ellipsiod .
the interaction potential between two ellipsoids of revolution @xmath0 and @xmath3 is given by @xmath4 where @xmath5 here @xmath6 defines the thickness or equivalently , the separation between the two ellipsoids of revolution in a side - by - side configuration , @xmath7 is the distance between the centers of mass of the ellipsoids of revolution @xmath0 and @xmath3 , and @xmath8 is a unit vector along the intermolecular separation vector @xmath9 .
the molecular shape parameter @xmath10 and the energy parameter @xmath11 both depend on the unit vectors @xmath2 and @xmath12 as well as on @xmath13 as given by the following set of equations : @xmath14^{-1/2}\ ] ] with @xmath15 and @xmath16^{\nu } [ \epsilon_{2}({\bf \hat r}_{ij},{\bf e}_{i},{\bf e}_{j})]^{\mu}\ ] ] where the exponents @xmath17 and @xmath18 are adjustable parameter , and @xmath19^{-1/2}\ ] ] and @xmath20\ ] ] with @xmath21 . here
@xmath22 is the aspect ratio of the ellipsoid of revolution with @xmath23 denoting the separation between two ellipsoids of revolution in a end - to - end configuration , and @xmath24 , and @xmath25 , where @xmath26 is the depth of the minimum of the potential for a pair of ellipsoids of revolution aligned in a side - by - side configuration , and @xmath27 is the corresponding depth for the end - to - end alignment .
@xmath28 is the depth of the minimum of the pair potential between two ellipsoids of revolution alligned in cross configuration .
the gb pair potential defines a family of models , each member of which is characterized by the values chosen for the set of four parameters @xmath29 and @xmath18 , and is represented by gb(@xmath30 ) @xcite .
three systems , namely gb(3 , 5 , 2 , 1 ) , gb(3.4 , 5 , 2 , 1 ) , and gb(3.8 , 5 , 2 , 1 ) , that differ in the aspect ratio have been investigated .
molecular dynamics simulations have been performed with each of these systems , consisting of 500 ellipsoids of revolution , in a cubic box with periodic boundary conditions .
each of these systems has been studied along three isochors ( @xmath31 = 0.31 , 0.32 , and 0.33 for @xmath32 = 3.0 ; @xmath31 = 0.25 , 0.26 , and 0.27 for @xmath32 = 3.4 ; @xmath31 = 0.215 , 0.225 , and 0.235 for @xmath32 = 3.8 ) at several temperatures , starting from the high - temperature isotropic phase down to the nematic phase across the i - n phase boundary .
all quantities are given in reduced units defined in terms of the gay - berne potential parameters @xmath28 and @xmath33 : length in units of @xmath33 , temperature in units of @xmath34 , and time in units of @xmath35 , m being the mass of the ellipsoids of revolution .
the mass as well as the moment of inertia of each of the ellipsoids of revolution have been set equal to unity .
the intermolecular potential is truncated at a distance @xmath36and shifted such that @xmath37 , @xmath7 being the separation between two ellipsoids of revolution i and j. the equations of motion have been integrated using the velocity - verlet algorithm with integration time step @xmath38 @xcite .
equilibration has been done by periodic rescaling of linear and angular velocities of particles .
this has been done for a time period of @xmath39 following which the system has been allowed to propagate with a constant energy for a time period of @xmath40 in order to ensure equilibration upon observation of no drift of temperature , pressure , and potential energy .
the data collection has been executed in a microcanonical ensemble . at each state point
, local potential energy minimization has been executed by the conjugate gradient technique for a subset of @xmath41 statistically independent configurations .
the landscape analysis has been done with a system size of @xmath42 ellipsoids of revolution , which is big enough for having no qualitative change in the results due to the system size @xcite .
minimization has been performed with three position coordinates and two euler angles for each particle , the third euler angle being redundant for ellipsoids of revolution .
the single - particle second rank orientational time correlation function ( otcf ) @xmath43 is defined by @xmath44 where @xmath45 is the second rank legendre polynomial , @xmath46 is the unit vector along the long axis of @xmath0th ellipsoid of revolution , and the angular brackets stand for ensemble averaging .
figure [ fig : power ] shows the time evolution of the single - particle second rank otcf for one of the three systems considered here as the temperature is lowered along an isochor from the high - temperature isotropic phase down to the nematic phase across the i - n phase boundary . in the inset
, the average orientational order parameter @xmath47 is shown as a function of temperature along the isochor @xcite .
the variation of @xmath47 with temperature serves to locate the i - n phase boundary . in the present study ,
the i - n transition temperature @xmath48 is taken as the temperature at which @xmath47 of the system is @xmath49 .
for each aspect ratio , three isochors at different densities have been considered .
the qualitative behavior has been found to be the same for all the three systems along all the isochors studied ( data not shown ) .
the emergence of the power law decay in the isotropic phase near the i - n transition is evident in all the cases as a universal characteristic of i - n transition @xcite . as the i - n phase boundary is crossed upon cooling , the advent of two power law decay regimes separated by an intervening plateau at short - to - intermediate times imparts a step - like feature to the temporal behavior of the second rank otcf .
such a feature bears remarkable similarity to what is observed for supercooled liquids as the glass transition is approached from the above @xcite . while for the supercooled liquid the emargence of step - like feature in the otcf is well understood as a consequence of @xmath50 relaxation , the origin of such a feature observed for liquid crystal defied of reliable explanation .
we estimate the orientational correlation time @xmath51 as the time taken for @xmath43 to decay by @xmath52 , i.e. , @xmath53 .
figure [ fig : frgl](a ) shows @xmath51 in the logarithmic scale as a function of the inverse temperature along the three isochors for each of the three systems considered .
we have scaled the temperature by @xmath48 in the spirit of angell s plot , that displays the shear viscosity ( or the structural relaxation time , the inverse diffusivity , etc . ) of glass - forming liquids as a function of the inverse of the scaled temperature , the scaling being done in the latter case by the glass transition temperature @xmath54 @xcite .
for all the three systems , two distinct features are common : ( i ) in the isotropic phase far away from the i - n transition , the orientational correlation time @xmath51 exhibits the arrhenius temperature dependence , i.e. , @xmath55 $ ] , where the activation energy @xmath56 and the pre - factor @xmath57 are both independent of temperature ; ( ii ) in the isotropic phase near the i - n transition , the temperature dependence of @xmath51 shows marked deviation from the arrhenius behavior and can be well described by the vogel - fulcher - tammann ( vft ) equation @xmath58 $ ] , where @xmath57 , @xmath59 , and @xmath60 are constants , independent of temperature .
again these features bear remarkable similarity with those observed for fragile glass - forming liquid .
a non - arrhenius temperature behavior is taken to be the signature of fragile liquids .
for fragile liquids , the temperature dependence of the shear viscosity follows the arrhenius behavior far above @xmath54 and can be fitted to the vft functional form in the deeply supercooled regime near @xmath54 @xcite .
the striking resemblance in the dynamical behavior described above between the isotropic phase of thermotropic liquid crystals near the i - n transition and supercooled liquids near the glass transition has prompted us to attempt a quantitative measure of glassy behavior near the i - n transition .
for supercolled liquids , one quantifies the dynamics by a parameter called fragility index which measures the rapidity at which the liquid s propeties ( such as viscosity ) change as the glassy state is approached . in the same spirit @xcite that offers a quantitative estimation of the fragile behavior of supercooled liquids ,
we here define the fragility index @xmath61 of a thermotropic liquid crystalline system as @xmath62 figure [ fig : frgl](b ) shows the density dependence of the fragility index for the three systems with different aspect ratios . for a given aspect ratio , the fragility index increases with increasing density , the numerical values of the fragility index @xmath61 being comparable to those of supercooled liquids . the change in the fragility index for a given density difference ( @xmath63 ) increases with the decrease in the aspect ratio .
another hallmark of fragile glass - forming liquids is spatially heterogeneous dynamics @xcite reflected in non - gaussian dynamical behavior @xcite .
it is intuitive that the growth of the pseudo - nematic domains , characterized by local nematic order , in the isotropic phase near the i - n transition would result in heterogeneous dynamics in liquid crystals .
we have , therefore , monitored the time evolution of the rotational non - gaussian parameter ( ngp ) @xcite , @xmath64 , which in the present case is defined as @xmath65 where @xmath66 here @xmath67 is the rotation vector like the position vector @xmath1 appears incase of translational ngp of @xmath0th ellipsoid of revolution , the change of which is defined by @xmath68 , @xmath69 being the corresponding angular velocity @xcite , and @xmath70 is the number of ellipsoids of revolution in the system .
ngp will have value equal to zero when system dynamics is spatially homogeneous and will have a non - zero value when the system dynamics is spatially heterogeneous . as a typical behavior , fig .
[ fig : rngp](a ) shows the time dependence of the rotational ngp for one of the systems at several temperatures across the i - n transition along an isochor . on approaching the i - n transition upon cooling ,
a bimodal feature starts appearing with the growth of a second peak , which eventually becomes the dominant one , at longer times .
we further investgate the apearance of this bimodal feature in ngp plot . to this end
we calculate mean square angular deviation ( msad ) of the system at different temperatures starting from high temperature isotropic phase to low temperature nematic phase .
the appearance of the bimodal feature in the rotational ngp is accompanied by a signature of a sub - diffusive regime in the temporal evolution of the mean square angular deviation ( msad ) , the time scale of the short - time peak and that of the onset of the sub - diffusive regime being comparable , as shown in fig . [ fig : rngp](b ) .
we further note that the dominant peak appears on a time scale which is comparable to that of onset of the diffusive motion in orientational degrees of freedom ( odof ) as evident in fig .
[ fig : rngp](b ) and similar feature has been observed recently for glassy systems @xcite also .
we also observe that the time scale at which long - time peak appears is also comparable to the onset of the plateau that is observed in the time evolution of @xmath43 , as shown in fig . [
fig : rngp](c ) .
this striking similarities of the dynamics between liquid crystals near i - n transition and supercooled liquid near glass transition are also supported by the use of mode coupling theory ( mct ) to explain the dynamics of both the systems .
while mct was used first for the supercooled liquid , recently it has been used for liquid crystals also .
mct theory devloped by gottke et al .
@xcite predicts that near i - n transition , the low frequency rotational memory kernel should diverge in a power law fashion .
@xmath71 mean field treatment gives @xmath72 .
invoking the rank ( @xmath73 ) dependence of the memory function , the single particle otcf can be written as @xcite @xmath74^{-1}\ ] ] the above equation can be laplace inverted to obtain a short - to - intermadiate power law decay in @xmath43 which is a universal characteristic of the i - n transition for several model liquid crystals @xcite . recently ,
@xcite have showed that it is also possible to formulate a schematic model that combines short - to - intermediate time relaxation with long time relaxation . in their model , they have expressed the total memory functon ( @xmath75 ) as the sum of mode coupling memory function ( @xmath76 ) and landau - de gennes memory function ( @xmath77 ) .
@xmath78 where @xmath79 @xmath80 is the characteristic frequecny and @xmath81 .
time dependence of @xmath82 can be written expressed in terms of the memory function @xmath83 @xcite and @xmath83 has the following form @xmath84 where @xmath85 is the autocorrelation function of the anisotropy of the polarizability and @xmath86 is the solution of a @xmath87 schematic model for what is referred to as the density correlator . @xmath32 being the coupling constant between them .
now , @xmath77 can be written as @xmath88 here @xmath89 is the relaxation time ( @xmath90 ) and it diverges as @xmath91 as the critical temperature @xmath92 of the i - n transition is approached from the above . following the calculation of ref .
@xcite , one can get two important relaxation equattions . @xmath93 with the initial conditions @xmath94 and @xmath95 and @xmath96 with the initial conditions @xmath97 and @xmath98 .
@xmath99 and @xmath17 are the damping constants .
@xmath100 is identical to one gets from the mct analysis of the supercooled liquids .
difference between this scematic model and one applied for the supercooled liquid is in eq .
@xmath101 . if @xmath102 is set equal to @xmath103 in eq . @xmath101
, the supercooled model is recovered .
@xmath101 is the orientational correlation function coupled to the density correlation function with specific new terms that account for the long - time portion of the relaxation profile that has been previously described by ldg theory .
several studies have attempted to interpret the fragility of glass - forming liquids in terms of the features of the underlying energy landscapes @xcite .
energy landscape analysis gives the potential energy , which devoids of any kind of thermal motions , of inherent structures of the parent liquid and hence provide a better understanding of the structure and dynamics of the paprent liquid .
a recent study on thermotropic liquid crystals has reported the temperature dependent exploration of the energy landscapes of a family of the gay - berne model systems across the mesophases @xcite .
the average inherent structure ( is ) energy @xmath104 has been found to fall as @xmath47 grows across the i - n phase boundary and through the nematic phase in contrast to its insensitivity to the temperature in the high - temperature isotropic phase and the low - temperature smectic - b phase @xcite .
such a fall in the average is energy is consistent with a gaussian form for the number density of inherent structures with energy @xmath105 , that predicts a linear variation of @xmath104 with the inverse temperature : @xmath106 , where @xmath107 and @xmath10 are parameters independent of temperature and @xmath108 is the boltzmann constant @xcite .
note that this has been observed for a glassy system @xcite , where the average is energy also falls over a temperature range @xcite . in fig .
[ fig : lscp](a ) , we demonstrate with the original and the most studied parameterization for the gb pair potential gb@xmath109 that the prediction holds good over the temperature range where @xmath104 is on a decline along all the three isochors studied .
it then follows that a plot of @xmath110 versus @xmath111/@xmath112 would result in a collapse of the @xmath104 data for all densities onto a straight line with negative unit slope .
this is indeed found to be true , as shown in fig .
[ fig : lscp](b ) , implying the validity of the gaussian model in this case as well .
it may be noted that when the distribution of is energy is gaussian , the fragility of glass - forming liquids has been shown to depend on the total number of inherent structures , the width of the gaussian , and the variation of the basin shape with the average is energy @xcite .
the origin of the glassy orientational dynamics , both single particle and collective , of nematogens near the isotropic - nematic transition has been addressed in several publications in recent years @xcite . in these studies ,
the similarity between dynamics of supercooled glassy liquid and liquid crystals has been discussed in detail , but no quantitative measure of the similarity was provided . the fragility index introduced here serves to remove that lacuna .
it is indeed surprising that even the values of the fragility parameter are in the range observed for glassy liquids as well .
this is in agreement with the repeated observation by fayer and coworkers that the values of the power law exponents observed in the two systems are quite similar .
further understanding of the relaxation mechanism has been obtained from a closure look on the heterogeneous dynamics .
as figure 3 demonstrates , the rotational non - gaussian parameter shows a dramatic enhancement of hetrogeneous dynamics as the i - n phase boundary is approached . unlike what is found near the gas - liquid critical point @xcite
, the single - particle dynamics near the i - n phase boundary are observed to be strongly affected by the approaching thermodynamic singularity .
we have discussed mode coupling theory approaches introduced to understand anomalous dynamics observed in this problem . *
acknowledgments * it is a pleasure to thank professor s. sastry for helpful discussions .
this work was supported in parts by grants from dst and csir , india . b.j .
acknowledges csir , india and d.c . acknowledges ugc , india for providing research fellowship | in an attempt to quantitatively characterize the recently observed slow dynamics in the isotropic and nematic phase of liquid crystals , we investigate the single - particle orientational dynamics of rodlike molecules across the isotropic - nematic transition in computer simulations of a family of model systems of thermotropic liquid crystals .
several remarkable features of glassy dynamics are on display including non - exponential relaxation , dynamical heterogeneity , and non - arrhenius temperature dependence of the orientational relaxation time . in order to obtain a _ quantitative _ measure of glassy dynamics in line with the estbalished methods in supercooled liquids ,
we construct a relaxation time versus scaled inverse temperature plot , and demonstrate that one can indeed define a fragility index for thermotropic liquid crystals , that depends on density and aspect ratio .
the values of the fragility parameter are surprisingly in the range one observed for glass forming liquids .
a plausible correlation between the energy landscape features and the observed fragility is discussed . |
cluster ion implantation can be regarded as a forerunner technology as compared to the conventional ion implantation technique used to dope sub - micron devices @xcite . using cluster ions very shallow implantation
can be achieved at very low energy . however , with cluster implantation , nonlinear effects arising in the energy loss processes , as a result of the correlated motion of the constituent atoms , play an important role in deciding the defect structure near the target surface .
in addition to resulting in a nonlinear growth in subsurface damage , cluster ion impact , through sputtering , can also results in kinetic roughening and smoothening of the surface exposed @xcite . in view of all this
, there has been a lot of activities involving low energy cluster ion irradiation related to nonlinear sputtering @xcite , nonlinear damage and defect production @xcite , along with the formation of various kind of surface features @xcite . in connection with the above ,
si , presents itself as a very important material where low energy cluster ions can be used for shallow implantation , of interest to technology . in some earlier work ,
contrary to common expectation , amorphisation upon ion irradiation has been shown to start from the surface rather than the ion projected range @xcite .
results of molecular dynamics ( md ) simulations with 5 kev si , show that the ion impacts produce unrelaxed amorphous patches that have a fast quenched , liquid like structure @xcite . with increase in ion fluence
these regions overlap producing a continuous amorphous layer @xcite .
in fact , with increase in ion fluence , there is a superlinear growth of amorphous volume fraction with a lot of stress build up in the matrix . at high fluence
there is an abrupt transition to a state with a flat amorphous - to - crystalline ( a / c ) interface @xcite .
in such a case , out of plane plastic flow with a reduction in the in - plane stress have been observed @xcite .
all this suggest that ion irradiation induced amorphisation in si is more like a _ phase transition _ , initiated by a spontaneous collapse of the damaged region .
very recent md simulations carried out by marqus _ et al _ show it to be initiated by a high concentration of _ interstitial - vacancy ( iv ) pairs _ or _ bond defects _
, formed in the system @xcite .
similar results have also been shown by nord _
et al _ @xcite who have pointed out that the subsequent transition resulting in a uniform amorphous layer is neither a complete homogeneous nor a complete heterogeneous mechanism .
this makes si an ideal system to study using low energy cluster ions where such a transition to a complete amorphous state is expected at a lower fluence , primarily because of overlapping of collision cascades coming from constituent atoms . in the present paper
we show some results of a systematic study of the subsurface damage produced and the surface features generated in si(100 ) , from si@xmath0 and a similar mass cs@xmath7 ion implantation at 25 kev . channeling rutherford backscattering spectrometry ( crbs ) and atomic force microscopy ( afm )
have been used for sample characterization .
increase in cluster ion fluence has been found to result in a nonlinear growth and saturation in damage leading to amorphisation .
the transition to an amorphised state is found to be associated with a significant drop in the power spectral density of afm data which initially increases with increase in fluence .
[ fig1 ] cleaned si(100 ) wafers ( @xmath8-type , 1 - 2.5 @xmath9 ) were irradiated with 25 kev singly charged negative ions _ viz _
si@xmath0 and cs@xmath7 from a snics - ii ion source ( nec , usa ) using a low energy ion implanter facility . mass analysis of the cluster ions was carried out using a @xmath10 sector magnet ( me / q@xmath11 = 18 mev amu ) .
the base pressure in the target chamber during irradiations was maintained around 2@xmath12 mbar .
all the irradiations were carried out at room temperature with a beam flux of 2 - 3@xmath13 ions @xmath4sec@xmath14 ( ion current of @xmath15 ) at @xmath167@xmath17 off the sample normal . in each case one
part of the sample was kept unimplanted to serve as a reference .
five samples named s1-s5 were systematically irradiated with ions of similar mass ( si@xmath0 or cs@xmath7 ) with gradually increasing ion fluence from 2@xmath18 @xmath4 to 1@xmath19 @xmath4 .
three of these , _ viz _ s1 , s3 and s5 were irradiated using si@xmath0 clusters to fluences of @xmath20 @xmath4 , @xmath21 @xmath4 and @xmath22 @xmath4 respectively .
the remaining two samples , s2 and s4 were irradiated with 25 kev cs@xmath7 ions to fluences of @xmath23 @xmath4 and @xmath24 @xmath4 respectively .
these data are shown in table .
1 . .sample names , ions used and integrated fluence . [ cols="^,^,^ " , ] crbs measurements were carried out on all the samples with 1.35 mev he@xmath25 with a si surface barrier detector placed at 130@xmath17 relative to the incident beam direction .
the measurements were carried out at a steady beam current of 5 @xmath26 , using the 3 mv pelletron accelerator ( 9sdh2 , nec , usa ) facility at iop , bhubaneswar . in case of unirradiated si(100 ) , the reduction in the integrated total yield from random to a channeled spectrum was found to be @xmath165@xmath27 .
following irradiation , the surface topography was examined by afm in the tapping mode , using a multi - mode scanning probe microscope ( nanoscope iv , veeco , usa ) .
measurements were performed in ambient condition using a si cantilever with a nominal tip radius less than @xmath28 .
image processing and analysis of the afm data were carried out using the standard wsxm software package @xcite .
the crbs results as measured for all the five samples _ viz .
_ s1-s5 and a virgin sample ( unirradiated area ) , are presented in fig .
1(a ) . from the figure
, one can observe a gradual increase in surface peak intensity with increase in ion fluence , @xmath1 .
this increase in the area of the surface peak , over and above that in the virgin sample , indicates growth in damage produced with increase in the number of displacements . in line with this ,
the lowest fluence irradiation is found to induce very little damage , too small to be seen through crbs data .
however , increase in irradiation fluence from s1 to s3 ( with cluster ions ) and that for s2 to s4 ( for cs ions ) are seen to result in enhancements of the surface peak , indicating an increase in defect production near the surface . for the sample s5 , with the highest cluster irradiation fluence
, the surface peak is seen to be the most intense .
this sample , with a total single atom fluence of @xmath29 @xmath4 is expected to have an amorphous layer with a thickness of @xmath30 extending from the surface @xcite .
cs@xmath7 irradiation to a fluence , 0.6 times as that of s5 , is seen to produced a damage distribution going deeper into the bulk .
now we look at the difference in the surface peak areas between the irradiated and the unirradiated regions , which at lower irradiation fluence , is proportional to the number of displaced si atoms , in the irradiated lattice .
we denote this as @xmath31 which has been estimated for various samples integrating data ( fig 1 ( a ) ) between channels 340 and 440 .
1(b ) shows ion fluence , @xmath1 , dependence of @xmath31 for various implantation fluence . from the figure
, one can see a clear nonlinear growth and saturation in damage with increase in ion fluence .
similar results , regarding a nonlinear growth in damage production and saturation has been seen earlier in case of 5 kev / atom c@xmath32 cluster irradiations of si(100 ) @xcite .
to get an idea about damage saturation and the amorphisation threshold , as has been done earlier @xcite , we have fitted the three points as obtained for si@xmath33 irradiation to an equation of the form , @xmath34 where @xmath35 is a constant and @xmath2 corresponds to the threshold fluence for damage saturation . the best fit , as indicated by the smooth curve in fig .
1(b ) , yields a value of @xmath2 equal to @xmath36 @xmath4 .
one can see , at a fluence of @xmath37 @xmath4 , the @xmath31 value corresponding to 25 kev cs atoms is almost the same as that expected for a similar mass si@xmath33 cluster at the same energy .
this means , the heavy single atom induced cascades produce almost the same amount of damage or defects as those generated by a similar mass cluster ion . at this high fluence , because of overlapping of damage produced cluster induced nonlinear effects are difficult to be detected . however , in this case there are defects extending into the bulk while those for a cluster ion are better confined in a surface layer .
[ fig3 ] the above @xmath2 value of @xmath3 @xmath4 as obtained for @xmath38 clusters , where damage saturation starts , corresponds to a total atomic fluence of @xmath39 @xmath4 .
this agrees with the finding of agarwal _ et al .
_ @xcite who have shown the amorphisation threshold , for 5 kev si in si , to be @xmath40 @xmath4 .
in fact , with clusters , as expected the present value agrees with the lower limit . in view of this , at an si@xmath33 cluster fluence of @xmath5 @xmath4 ( corresponding to an atomic fluence of @xmath29 @xmath4 ) , we are already well above the threshold for amorphisation . as has been mentioned earlier , afm has been used to study the surface topography in various samples .
some afm pictures ( top view ) , for ( @xmath41 ) scanned areas , taken on the samples are shown in figs . 2 and 3 .
figs 2(a ) and ( b ) correspond to samples s1 and s2 , while figs .
3(a ) and ( b ) correspond to samples s3 and s4 respectively .
usually the features on irradiated surfaces are described through a height - height correlation function which contains three important roughness parameters : ( i ) the vertical correlation length @xmath42 , ( ii ) the lateral correlation length @xmath43 and ( iii ) the roughness exponent @xmath35 @xcite .
the lateral correlation length , @xmath43 , describes the lateral characteristics of the surface , the roughness exponent @xmath35 describing the static scaling properties .
the most commonly reported parameter of surface roughness i.e. @xmath42 , or the root - mean - square ( rms ) roughness , characterizes the surface , only along the vertical direction .
this is defined as standard deviation of the surface height profile , @xmath44 , at each point ( @xmath45 ) of a reference surface plane from the mean height ( @xmath46 ) , as given by , @xmath47^{1/2}\ ] ] where @xmath48 is the number of pixels , @xmath49 being the height at the @xmath50 pixel . in case of the sample s1 , as shown in fig 2(a ) , one can clearly see black dots corresponding to nanometer sized pits and bright spots corresponding to hillocks ( color bars indicating the heights ) .
the observed hillocks are seen to have heights ranging between @xmath51 with average height of @xmath52 .
it has an rms roughness , @xmath42 of @xmath53 .
compared to this , there was hardly any surface feature in the case of a pristine si(100 ) sample ( afm data not shown here ) .
the rms roughness , @xmath42 , of the pristine sample , was found to be @xmath54 , which is about one third of that for s1 , corresponding to a smooth polished surface .
[ fig4 ] in case of the lowest fluence cs implanted sample , s2 , the @xmath42 for surface roughness was found to be @xmath55 , lying between that for s1 ( @xmath56 ) and the pristine sample ( @xmath57 ) .
what is important to see here is that cs implantation to a five times higher fluence is not able to generate a surface roughness as observed in case of s1 .
compared to the above , s3 , implanted with si@xmath33 clusters , to a fluence sixty times that of the sample s1 , was found to have an rms roughness , @xmath42 , of @xmath58 .
however , the above sixty - fold increase in si@xmath33 cluster fluence from s1 to s3 , did not result in a proportionate increase in the number of nanohillocks in s3 .
the surface topography of the sample s4 , implanted with cs to a fluence five times that of s3 is shown in fig .
however , a much smaller number of nanohillocks , compared to the s3 sample , could be seen .
it has a @xmath42 value of @xmath59 which is significantly smaller than that observed for s3 .
the enhanced surface roughness as seen with si@xmath33 implantations in s1 as compared to s2 ( and s3 as compared to s4 ) , even with lower fluence of irradiations , are primarily due to molecular effects coming from overlapping of collision cascades of constituent atoms .
such effects are absent during a single mass cs irradiation which results in lower damage accumulation and lower surface roughness .
compared to all the above , the @xmath42 value of surface roughness observed for the sample s5 , which had the highest cluster ion fluence , turned out to be @xmath60 .
this was quite small compared to the same obtained for s3 , irradiated with an almost one order of magnitude lower fluence . however , the parameter @xmath42 is not enough to give a full characterization of the surface because it is limited by its sensitivity only in the vertical direction .
for example , two images with exactly the same rms roughness values can have different surface morphologies @xcite . in view of this , a power spectral density ( psd ) analysis is often used to look at surface features and their possible origin @xcite .
the psd analysis is accomplished by radially averaging the square magnitude of the coefficients of the two - dimensional fourier transform of the digitized surface profile @xmath44 defined by @xmath61 where @xmath62 .
here @xmath63 is the spatial frequency in reciprocal space , @xmath64 is the scanned area of length @xmath65 and @xmath66 is the height at the position @xmath67 .
the psd , which is the fourier transform of the height - height correlation function , thus turns out to be a function of only one parameter , spatial frequency , @xmath63 .
identification of various processes for surface transport @xcite is carried out using the stochastic rate equations for the surface evolution @xcite . in this analysis @xcite ,
@xmath68 $ ] , where @xmath69 is a term coming from noise correlation , @xmath70 being expansion coefficients . the index @xmath71 has values of 1,2,3 and 4 representing four modes of surface transport _ viz .
_ _ viscous flow _ , _ evaporation - condensation _ , _ volume diffusion _ and _ surface diffusion _ respectively .
this index is further related to the roughness scaling exponent , @xmath35 through the relation @xmath72 .
the psd spectra , @xmath73 , for all the samples s1 to s4 and pristine , are shown in fig .
one can see , there is an increase in spectral strength in the order pristine , s2 , s1 , s4 and s3 where the pristine has the lowest and s3 has the highest values .
this order is exactly the same if one orders the samples according to the @xmath42 values for the surface roughness .
s2 and s4 have five times higher ion fluence as compared to s1 and s3 respectively .
but the the surface features produced are seen to be much reduced .
having seen the above general features we now try to look at other details of @xmath73 .
one can see that the pristine sample ( @xmath74 ) has a cut off value , @xmath75 , which is about 0.001 nm@xmath14 .
this corresponds to a large correlation length , @xmath43 of 1 @xmath76 .
further , it has a @xmath71 close to 1 .
the cs implanted sample , s2 , has also a smooth surface ( @xmath77 ) .
the corresponding @xmath73 indicates surface modulations over a similar length scale as the pristine sample .
compared to this , the si@xmath33 cluster implanted s1 sample , with one fifth of the fluence as in s2 , shows modulations with a higher value of @xmath75 of the order of @xmath78 .
this indicates a correlation length , @xmath43 , of the order of @xmath79 .
as shown earlier , it has a @xmath42 value of @xmath56 . the fourier index @xmath71 is seen to be about 2.5 .
this yields an @xmath35 value of 0.25 , indicating the surface to be self affine with anisotropic scaling along lateral and perpendicular directions .
the @xmath73 for the higher fluence cs irradiated sample , s4 , shows a @xmath75 which is almost similar to that for the lower fluence si@xmath33 implanted sample , s1 , indicating a similar correlation length , @xmath43 , of @xmath79 .
however , it shows a @xmath71 value of around 2.2 resulting in an @xmath35 value @xmath80 . with a @xmath42 value of @xmath59
, it has a higher roughness .
this means , with a 300 times higher fluence , cs ions generate similar surface features as obtained for a low dose si@xmath33 implanted sample , s1 . compared to this ,
the si@xmath33 cluster implanted sample s3 , with a fluence which is one fifth of that in s4 , shows a @xmath71 value close to 4 , indicating an @xmath35 value close to unity .
this indicates the surface modulations to be self similar .
but this sample has the highest surface roughness , @xmath42 of @xmath81 with an average height of @xmath82 .
now we look at what happens when cluster fluence is increased to a value well beyond the amorphisation threshold as in case of s5 .
an afm image of the top view of s5 , taken with a ( @xmath83 ) scan size is shown in fig .
the @xmath73 spectrum for s5 is also shown along with that for s3 in the same figure .
one can clearly see that increasing the cluster ion fluence from 1.2@xmath84 to 1@xmath19 @xmath4 , in going from s3 to s5 , has resulted in no further change in the @xmath71 value which has been found to saturate at 4 .
but the surface modulations changed to have a @xmath42 value of @xmath85 with a correlation length @xmath43 @xmath86 . with an @xmath35 value close to 1 ,
the surface features in s5 are self similar indicating isotropic scaling . however , as compared to the s3 case , @xmath73 shows a significant reduction in the magnitude . at @xmath75 , the ratio between the two is about 33
this can also be seen using the formula @xmath87 $ ] at @xmath88 . since s3 and s5 have almost same @xmath35 as well as @xmath43 values , the ratio of the @xmath73s turn out to be nearly the same as the ratio of the @xmath89 for the two cases .
this is seen to be @xmath90 which is just about right .
it is important to mention that the surface modulations in s5 show a mean height of @xmath91 .
the small value of @xmath73 for s5 , is therefore seen to be coming from a correlation between small but nearly equal heights with a small @xmath42 value .
compared to this @xmath73 in s3 comes from a correlation between higher heights , with an average value of @xmath82 , again with a much higher value of @xmath42 .
earlier , 5 kev si impact on si has been shown to result in creation of amorphous pockets coming form local melting and rapid quenching @xcite .
the stress produced can result in formation of pits and bumps on the surface .
this is in addition to roughening resulting from sputtering .
this local melting and the associated movement of atoms , at lower fluence as in s1 , may be responsible for a @xmath71 value between 2 and 3 .
it also leads to a higher surface roughness as compared to a pristine sample . at lower fluence
there are sparsely distributed amorphous pockets in the matrix .
increase in the implantation fluence results in a fast growth in the number of amorphous patches resulting in a growth in surface roughness , @xmath42 .
this also results in a growth in mean height , @xmath46 , of surface structures produced , resulting in a growth of height - height correlation .
merging of amorphised regions at higher fluence , results in a building up of stress from a large number of _ bond defects_. finally there could be stress relaxation as the damaged lattice becomes unstable .
this could result in a transition to a state with smaller surface features which is achieved by an effective movement of atoms in a lateral direction .
this could be the reason behind getting a @xmath71 value close to 4 as proposed for surface diffusion .
this is probably how a smooth amorphous to crystalline interface can occur in si under high fluence ion irradiation @xcite .
this way crystalline to amorphous transition in si , upon ion irradiation , is more like a phase transition induced by an accumulation of sufficient number of defects which was also suggested by several groups earlier @xcite . in the present case ,
onset of this occurs at a cluster fluence of around 2.5@xmath84 @xmath4 . for a much higher cluster fluence ( as in s5 ) ,
a continuous amorphous layer parallel to the surface is produced , leading to much reduced values for @xmath42 and @xmath73 .
it is therefore not surprising that the @xmath43 , @xmath35 and @xmath71 values as obtained for s5 are very similar to those obtained for a - si films @xcite .
this also confirms that the top surface of the s5 sample is actually amorphised , in agreement with channeling data ( fig .
1(b ) ) where saturation in damage production has been obtained .
it is also important to realize that @xmath31 for s4 is much higher than that in case of s3 ( fig .
1(b ) ) . in fact
it is closer to that in s5 .
however , complete amorphisation does not occur here because the defects produced by cs implantation are distributed over a greater depth resulting in comparatively less defect accumulation near the surface .
to conclude , we have carried out a systematic study of 25 kev si@xmath0 implantation induced damage and surface modifications in si(100 ) where a nonlinear growth in subsurface damage , with fluence , is observed . the damage produced by similar mass cs@xmath7 ions , of the same energy ,
is seen to be distributed over a greater depth leading to much reduced surface features . with si@xmath33 clusters , the threshold fluence for amorphisation of si surface
is found to be @xmath3 clusters @xmath4 , in agreement with earlier published data .
most importantly , at higher cluster fluence a transition to an amorphous state resulting in a much reduced surface roughness is indicated .
the authors would like to thank a.k .
behera for the efficient running of the implanter facility and all the operators of accelerator laboratory , iop for help during crbs runs .
the technical help regarding some preliminary afm measurements and checks by s. varma of iop and s. bhattacharjee of immt , bhubaneswar are also gratefully acknowledged .
10 v. n. popok and e. e. b. campbell , rev . adv .
sci . * 11 * , 19 ( 2006 ) .
i. yamada , j. matsuo , n. toyoda , and a. kirkpatrick , mater .
r : reports , * 34 * , 231 ( 2001 ) .
y. teranishi , k. kondou , y. fujiwara , h. nonaka , m. tomita , k. yamamoto , t. fujimoto , and s. ichimura , , 5528 ( 2006 ) .
h. h. andersen , a. brunelle , s. della - negra , j. depauw , d. jacquet , y. le beyec , j. chaumont , and h. bernas , , 5433 ( 1998 ) ; a. v. samartsev , a. duvenbeck , and a. wucher , , 115417 ( 2005 ) .
l. shao , m. nastasi , x. wang , j. liu , and wei - kan chu , , 503 ( 2006 ) ; j. liu , x. wang , l. shao , x. lu , and w. k. chu , 787 ( 2002 ) .
h. p. lenka , b. joseph , p. k. kuiri , g. sahu , and d. p. mahapatra , , 665 ( 2007 ) .
a. i. titov , a. yu . azarov , l. m. nikulina , and s. o. kucheyev , 064111 ( 2006 ) . s. prasalovich , v. n. popok , p. persson , and e. e. b. campbell , eur .
j. d * 36 * , 79 ( 2005 ) .
j. samela , k. nordlund , v. n. popok , and e. e. b. campbell , , 075309 ( 2008 ) .
h. s. park , h. j. jung , and w. k. choi , 36 ( 2005 ) .
l. p. allen , d. b. fenner , c. santeufemio , w. brooks , and i. yamada , , 3671 ( 2002 ) . v.
n. popok , s. v. prasalovich , and e. e. b. campbell , , 145 ( 2003 ) .
k. nakajima , h. toyofuku , and k. kimura , , 2119 ( 2001 ) .
t. diaz dela rubia and g.h .
gilmer , , 2507 ( 1995 ) .
campisano , s. coffa , v. raineri , f. priolo , and e. rimini , , 514 ( 1993 ) .
o.w . holland , s.j .
pennycook , and g.l .
albert , , 2503 ( 1989 ) .
g. bai and m.a .
nicolet , , 649 ( 1991 ) . c. a. volkert , , 3521 ( 1991 ) .
l. a. marqus , l. pelaz , p. lpez , i. santos , and m. aboy , 153201 ( 2007 ) ; l.a .
marqus , l. pelaz , m. aboy , l. enrquez , and j. barbolla , , 135504 ( 2003 ) ; l.a.marqus , l. pelaz , j. hernndez , j. barbolla , and g.h .
gilmer , 045214 ( 2001 ) .
j. nord , k. nordlund and j. keinonen , , 165329 ( 2002 ) .
wsxm free software .
avilable at _ www.nanotech.es _
i. horcas , r. fernandez , j.m .
gomez - rodriguez , j. colchero , j. gomez - herrero , and a. m. baro , rev .
instrum . * 78 * , 013705 ( 2007 ) .
a. agarwal , t. e. haynes , d. j. eaglesham , h.j .
gossmann , d.c .
jacobson , j. m. poate , and y. e. erokhin , , 3332 ( 1997 ) .
m. dobeli , r , m .
ender , u.s .
fischer , m. suter , h.a .
synal , and d. vetterli , 388 ( 1994 ) .
eklund , e.j .
synder , and r.s .
williums , *285 * , 157 ( 1993 ) .
bray and g.n .
parsons , 035311 ( 2001 ) .
fenner , , 5408 ( 2004 ) .
fang , s. haplepete , w. chen , c.r .
helms , and h. edwards , , 5891 ( 1997 ) .
fang , w. chen , t. yamanaka , and c.r .
heims , , 2837 ( 1996 ) .
r. petri , p. brault , o. vatel , d. henry , e. andre , p. dumas , and f. salvan , , 7498 ( 1994 ) .
tong , r.s .
williams , a. yanase , y. segawa , and m.s .
anderson , , 3374 ( 1992 ) . c. ruppe and a. duparre
; , 8 ( 1996 ) .
m. senthilkumar , n.k .
sahoo , s. thakur , and r.b .
tokas , , 1608 ( 2005 ) . c. herring , , 3521 ( 1950 ) .
tong , e.j .
synder , r.s .
williams , a. yanase , y. segawa , and m.s .
anderson , *277 * , l63 ( 1992 ) .
s.g . mayr and r.s .
averback , 196106 ( 2001 ) .
swanson , j. r. parsons and c.w .
hoelke , radiat .
eff . * 9*,249 ( 1971 ) .
t. motooka , s. harada , and m. ishimaru , , 2980 ( 1997 ) . | the damage growth and surface modifications in si(100 ) , induced by 25 kev si@xmath0 cluster ions , as a function of fluence , @xmath1 , has been studied using atomic force microscopy ( afm ) and channeling rutherford backscattering spectrometry ( crbs ) .
crbs results indicate a nonlinear growth in damage from which it has been possible to get a threshold fluence , @xmath2 , for amorphisation as @xmath3 ions @xmath4 . for @xmath1 below @xmath2 , a growth in damage as well as surface roughness has been observed . at a @xmath1 of @xmath5 ions @xmath4 , damage saturation coupled with a much reduced surface roughness has been found . in this case
a power spectrum analysis of afm data showed a significant drop , in spectral density , as compared to the same obtained for a fluence , @xmath6 .
this drop , together with damage saturation , can be correlated with a transition to a stress relaxed amorphous phase .
irradiation with similar mass cs@xmath7 ions , at the same energy and fluence , has been found to result in a reduced accumulation of defects in the near surface region leading to reduced surface features . |
[ pddefn ] a group @xmath3 of homeomorphisms of a topological space @xmath11 acts _ properly discontinuously _ on @xmath11 if , for every compact subset @xmath12 of @xmath11 , @xmath13 classically , a discrete group @xmath3 of isometries of a riemannian manifold @xmath11 is a crystallographic group if @xmath3 acts properly discontinuously on @xmath11 , and the quotient @xmath14 is compact .
the @xmath3-translates of any fundamental domain for @xmath14 form a tessellation of @xmath11 .
these notions generalize to any homogeneous space , even without an invariant metric .
[ tessdefn ] let * @xmath1 be a lie group and * @xmath0 be a closed subgroup of @xmath1
. a discrete subgroup @xmath3 of @xmath1 is a _ crystallographic group _ for @xmath2 if 1 .
@xmath3 acts properly discontinuously on @xmath2 ; and 2 .
@xmath15 is compact .
we say that @xmath2 has a _ tessellation _ if there exists a crystallographic group @xmath3 for @xmath2 .
crystallographic groups and the corresponding tessellations have been studied for many groups @xmath1 .
( a brief recent introduction to the subject is given in @xcite . ) the classical bieberbach theorems ( * ? ? ?
* chap . 1 )
deal with the case where @xmath1 is the group of isometries of euclidean space @xmath16 . as another example
, the auslander conjecture @xcite asserts that if @xmath1 is the group of all affine transformations of @xmath17 , then every crystallographic group has a solvable subgroup of finite index .
in addition , the case where @xmath1 is solvable has been discussed in @xcite . in this paper , we focus on the case where @xmath1 is a simple lie group , such as @xmath18 , @xmath19 , or @xmath20 . [ standing ] throughout this paper : 1 .
[ standing - g ] @xmath1 is a linear , semisimple lie group with only finitely many connected components ; and 2 .
[ standing - h ] @xmath0 is a closed subgroup of @xmath1 with only finitely many connected components . [
hdisconnected ] because @xmath21 is finite ( hence compact ) , it is easy to see that @xmath2 has a tessellation if and only if @xmath22 has a tessellation .
also , if @xmath2 has a tessellation , then @xmath23 has a tessellation .
furthermore , the converse holds in many situations .
( see [ disconnected ] for a discussion of this issue . )
thus , there is usually no harm in assuming that both @xmath1 and @xmath0 are connected ; we will feel free to do so whenever it is convenient .
on the other hand , because @xmath19 is usually not connected ( it usually has two components ( * ? ? ? * lem . 10.2.4 , p. 451 ) ) , it would be somewhat awkward to make this a blanket assumption .
[ classical ] there are two classical cases in which @xmath2 is well known to have a tessellation . 1 .
if @xmath2 is compact , then we may let @xmath24 ( or any finite subgroup of @xmath1 ) .
2 . [ classical - borel ] if @xmath0 is compact , then we may let @xmath3 be any cocompact lattice in @xmath1 .
( a. borel @xcite proved that every connected , simple lie group has a cocompact lattice . )
thus , the existence of a tessellation is an interesting question only when neither @xmath0 nor @xmath2 is compact .
( in this case , any crystallographic group @xmath3 must be infinite , and can not be a lattice in @xmath1 . ) given @xmath1 ( satisfying [ standing]([standing - g ] ) ) , we would like to find all the subgroups @xmath0 ( satisfying [ standing]([standing - h ] ) ) , such that @xmath2 has a tessellation .
this seems to be a difficult problem in general .
( see the surveys @xcite and @xcite for a discussion of the many partial results that have been obtained , mainly under the additional assumption that @xmath0 is reductive .
) however , it can be solved in certain cases of low real rank . in particular , as we will now briefly explain , the problem is very easy if @xmath25 or @xmath26 .
most of this paper is devoted to solving the problem for certain cases where @xmath27 .
if @xmath25 ( that is , if @xmath1 is compact ) , then @xmath2 must be compact ( and @xmath0 must also be compact ) , so @xmath2 has a tessellation , but this is not interesting . if @xmath28 , then there are some interesting homogeneous spaces , but it turns out that none of them have tessellations . [ calabimarkuscircle ] @xmath29 is transitive on @xmath30 , so @xmath30 is a homogeneous space for @xmath1 .
it does not have a tessellation , for reasons that we now explain .
let @xmath12 be the unit circle , so @xmath12 is a compact subset of @xmath31 .
we claim that @xmath32 , for every @xmath33 ( cf . figure [ circle+ellipse ] ) . to see this , note that , because @xmath34 , the ellipse bounded by @xmath35 has the same area as the disk bounded by @xmath12 , so @xmath35 can not be contained in the interior of the disk bounded by @xmath12 , and can not contain @xmath12 in its interior .
thus , @xmath35 must be partly inside @xmath12 and partly outside , so @xmath35 must cross @xmath12 , as claimed .
let @xmath3 be any discrete subgroup of @xmath1 .
the preceding paragraph implies that @xmath36 , for every @xmath37 .
if @xmath3 acts properly discontinuously on @xmath30 , then , because @xmath12 is compact , this implies that @xmath3 is finite .
so the quotient @xmath38 is not compact .
therefore @xmath3 is not a crystallographic group .
we have shown that no subgroup of @xmath1 is a crystallographic group , so we conclude that @xmath30 does not have a tessellation . ):
@xmath39 , for every @xmath40 , so no infinite subgroup of @xmath41 acts properly discontinuously .
] this example illustrates the _ calabi - markus phenomenon _ : if there is a compact subset @xmath12 of @xmath2 , such that @xmath32 , for every @xmath33 , then no infinite subgroup of @xmath1 acts properly discontinuously on @xmath2 ( see [ calabimarkus ] ) .
thus , @xmath2 does not have a tessellation , unless @xmath2 is compact ( see [ cds->notess ] ) .
we will see in section [ cartansect ] that the following proposition can be proved quite easily from basic properties of the cartan projection .
rrank1-cds [ rrank1-cm ] if @xmath28 , and @xmath0 is not compact , then there is a compact subset @xmath12 of @xmath2 , such that @xmath32 , for every @xmath33 .
[ rank1->notess ] if @xmath28 , and neither @xmath0 nor @xmath2 is compact , then @xmath2 does not have a tessellation .
we now consider groups of real rank two .
the obvious example is @xmath42 , but , in this case , once again , none of the interesting homogeneous spaces have tessellations .
moreover , the same is true when real numbers are replaced by complex numbers or quaternions .
the case where @xmath43 relies on beautiful methods of y. benoist and f. labourie @xcite or g. a. margulis @xcite , which we describe in section [ 1dsect ] .
[ sl3->notess ] if @xmath44 and neither @xmath0 nor @xmath2 is compact , then @xmath2 does not have a tessellation .
it is important to note that some interesting homogeneous spaces do have tessellations .
[ g = lxh ] suppose @xmath45 , and let @xmath3 be a cocompact lattice in @xmath46 .
then @xmath3 acts properly discontinuously on @xmath47 , and @xmath48 is compact .
so @xmath2 has a tessellation .
the following easy lemma generalizes this example to the situation where @xmath1 is a more general product of @xmath46 and @xmath0 , not necessarily a direct product . construct - tess [ g = lh ] let @xmath0 and @xmath46 be closed subgroups of @xmath1 , such that * @xmath49 , * @xmath50 is compact ; and * @xmath46 has a cocompact lattice @xmath3 .
then @xmath2 has a tessellation .
( namely , @xmath3 is a crystallographic group for @xmath2 . ) for @xmath51 or @xmath52 , this lemma leads to some interesting examples found by r. kulkarni ( * ? ? ?
6.1 ) and t. kobayashi ( * ? ? ?
[ kulkarnieg ] there are natural embeddings @xmath53 furthermore , identifying @xmath54 with @xmath55 yields an embedding @xmath56 similarly , identifying @xmath57 with @xmath58 yields an embedding @xmath59 thus , we may think of @xmath60 and @xmath61 as subgroups of @xmath62 ; and we may think of @xmath63 and @xmath64 as subgroups of @xmath65 . with the above understanding , we see that @xmath60 is the stabilizer of a vector of norm @xmath66 .
since @xmath61 is transitive on the set of such vectors , we have @xmath67 similarly , @xmath68 then lemma [ g = lh ] implies that each of the following four homogeneous spaces has a tessellation : * @xmath69 , * @xmath70 , * @xmath71 , and * @xmath72 . when discussing @xmath73 or @xmath52 , we always assume @xmath74 .
this causes no harm , because @xmath75 is locally isomorphic to @xmath76 ( * ? ? ?
* ( x ) , p. 520 ) , and @xmath77 is locally isomorphic to @xmath78 ( * ? ? ?
* ( vi ) , p. 519 ) .
when @xmath79 is even , h. oh and d. witte @xcite provided a complete description of all the ( closed , connected ) subgroups @xmath0 , such that @xmath80 has a tessellation , but their classification is not quite complete when @xmath79 is odd . in this paper , we extend the work of oh and witte to obtain analogous results for homogeneous spaces of @xmath81 . we also give a much shorter proof of the main results of @xcite
the same techniques should yield significant results for homogeneous spaces of the other simple groups of real rank two , although the calculations seem to be difficult . on the other hand ,
the groups of higher real rank require different ideas .
once one knows that a tessellation of @xmath2 exists , it would be interesting to find _ all _ of the crystallographic groups for @xmath2 and , for each crystallographic group , describe the possible tessellations .
these are much more delicate questions , which we do not address at all .
( w. goldman @xcite , f. salein @xcite , t. kobayashi @xcite , and a. zeghib @xcite have interesting results in some special cases . ) in the remainder of this introduction , we state the specific results for homogeneous spaces of @xmath73 and @xmath52 .
[ d(h)-defn ] for any connected lie group @xmath0 , let @xmath82 where @xmath83 is any maximal compact subgroup of @xmath0 .
this is well defined , because all the maximal compact subgroups of @xmath0 are conjugate ( * ? ? ?
15.3.1(iii ) , pp .
180181 ) .
[ d(g ) ] if @xmath0 is semisimple , we have the iwasawa decomposition @xmath84 ( * ? ?
* thm . 6.5.1 , pp .
270271 ) , from which it is obvious that @xmath85 .
this yields the following calculations ( see [ d(su2 ) ] and [ d(sp ) ] ) ) : * @xmath86 .
. * @xmath88 .
* @xmath89 . * @xmath90 .
[ d(h)=dimh ] if @xmath91 ( for some iwasawa decomposition @xmath92 of @xmath1 ) , then @xmath93 ( see [ ansc ] and [ solvable]([solvable - nocpct ] ) ) .
[ simdefn ] for subgroups @xmath94 and @xmath95 of @xmath1 , we write @xmath96 if there is a compact subset @xmath12 of @xmath1 , such that @xmath97 and @xmath98 . note that @xmath99 is not invariant under the equivalence relation @xmath100 . for example
, the cartan decomposition @xmath101 implies that @xmath102 , but we have @xmath103 . the following two theorems state a version of the main results for even @xmath79 .
suf - known [ ow - known ] assume @xmath104 , and let @xmath0 be a closed , connected , subgroup of @xmath1 , such that neither @xmath0 nor @xmath2 is compact .
the homogeneous space @xmath2 has a tessellation if and only if 1 .
@xmath105 ; and 2 . either @xmath106 or @xmath107 .
suf - known [ iw - known ] assume @xmath108 , and let @xmath0 be a closed , connected , subgroup of @xmath1 , such that neither @xmath0 nor @xmath2 is compact .
the homogeneous space @xmath2 has a tessellation if and only if 1 .
@xmath109 ; and 2 .
either @xmath110 or @xmath111 .
the subgroups @xmath0 that arise in theorems [ ow - known ] and [ iw - known ] can also be described more explicitly ( cf .
[ soeventess ] and [ sueventess ] below ) .
t. kobayashi @xcite conjectured that if @xmath0 is reductive and it is impossible to construct a tessellation of @xmath2 by using a generalization of lemma [ g = lh ] ( see [ construct - tess ] ) , then @xmath2 does not have a tessellation .
the following lists three special cases of this general conjecture .
[ notesssu / sp ] the homogeneous spaces 1 . [ notesssu / sp - so2/su1 ] @xmath112 , 2 .
[ notesssu / sp - su2/sp1 ] @xmath113 , and 3 .
[ notesssu / sp - su2/su1 ] @xmath114 do not have tessellations .
if this conjecture is true , then , for odd @xmath79 , there is no interesting example of a homogeneous space of @xmath73 or @xmath52 that has a tessellation .
suf->complete [ su1m->complete ] assume @xmath115 and let @xmath0 be any closed , connected subgroup of @xmath1 , such that neither @xmath0 nor @xmath2 is compact .
if conjecture [ notesssu / sp ] is true , then @xmath2 does not have a tessellation .
the proof of theorem [ su1m->complete ] assumes the following special case proved by r. kulkarni ( * ? ? ?
2.10 ) . in short
, kulkarni noted that the euler characteristic of @xmath15 must both vanish ( because the euler characteristic of @xmath2 vanishes ) and not vanish ( by the gauss - bonnet theorem ) .
( other results in the same spirit , obtaining a contradiction from the study of characteristic classes of @xmath116 , appear in @xcite . )
[ so2n / so1odd - notess ] if @xmath79 is odd , then @xmath117 does not have a tessellation .
let us give a more explicit description of the closed , connected subgroups @xmath0 of @xmath62 or @xmath65 , such that @xmath2 has a tessellation .
this shows that if @xmath79 is even , then the kulkarni - kobayashi examples ( [ kulkarnieg ] ) and certain deformations are essentially the only interesting homogeneous spaces of @xmath73 or @xmath52 that have tessellations . [ kandefn ] fix an iwasawa decomposition @xmath92 .
thus , * @xmath118 is a maximal compact subgroup , * @xmath119 is the identity component of a maximal split torus , and * @xmath120 is a maximal unipotent subgroup .
the following two results are stated only for subgroups of @xmath121 , because the general case reduces to this ( see [ hcanbean ] ) .
the reason is basically that @xmath0 contains a connected , cocompact subgroup that is conjugate to a subgroup of @xmath121 .
( clearly , if @xmath122 is any cocompact subgroup of @xmath0 , then @xmath2 has a tessellation if and only if @xmath123 has a tessellation . )
this is not quite true in general , but the following lemma provides a satisfactory substitute , by showing that it becomes true after enlarging @xmath0 by a compact amount .
hcanbean after replacing @xmath0 by a conjugate subgroup , there is a closed , connected subgroup @xmath124 of @xmath1 , such that @xmath125 and @xmath126 are compact , where @xmath127 denotes the identity component of @xmath128 .
sufeventess [ soeventess ] assume @xmath104 , and let @xmath0 be a closed , connected , nontrivial , proper subgroup of @xmath121 .
the homogeneous space @xmath2 has a tessellation if and only if @xmath0 is conjugate to a subgroup @xmath122 , such that either 1 .
[ soeventess - so ] @xmath129 ; or 2 .
[ soeventess - sp ] @xmath122 belongs to a certain family @xmath130 of deformations of @xmath131 , described explicitly in theorem [ hbthm ] ( with @xmath132 ) .
sufeventess [ sueventess ] assume @xmath108 , and let @xmath0 be a closed , connected , nontrivial , proper subgroup of @xmath121 .
the homogeneous space @xmath2 has a tessellation if and only if @xmath0 is conjugate to a subgroup @xmath122 , such that either 1 .
[ sueventess - su ] @xmath122 belongs to a certain family @xmath133}}}\}$ ] of deformations of @xmath134 , described explicitly in theorem [ suegs ] ; or 2 .
[ sueventess - sp ] @xmath122 belongs to a certain family @xmath130 of deformations of @xmath135 , described explicitly in theorem [ hbthm ] ( with @xmath136 ) .
the proof of theorem [ soeventess ] ( or [ ow - known ] ) in @xcite requires a list @xcite of all the homogeneous spaces of @xmath73 that admit a proper action of a noncompact subgroup of @xmath52 .
( the list was obtained by very tedious case - by - case analysis .
it was extended to homogeneous spaces of @xmath52 in @xcite . ) the following proposition ( [ tessan->dim>1,2 ] ) provides an _ a priori _ lower bound on @xmath137 , and it turns out that the classification of the interesting subgroups of large dimension can be achieved fairly easily ( see [ suflargesect ] ) .
this is the main reason that we are able to give reasonably short complete proofs of theorems [ ow - known ] , [ iw - known ] , [ su1m->complete ] , [ soeventess ] , and [ sueventess ] .
[ tessan->dim>1,2 ] suppose @xmath51 or @xmath52 , and let @xmath0 be a closed , connected , nontrivial subgroup of @xmath121 . if @xmath2 has a tessellation , then @xmath138 this research was partially supported by a grant from the national science foundation ( dms-9801136 ) .
much of the work was carried out during productive visits to the isaac newton institute for mathematical sciences ( cambridge , u.k . )
; we would like to thank the newton institute for the financial support that made the visits possible .
would like to thank the mathematics department of oklahoma state university for their warm and generous hospitality and marc burger both for pointing out a mistake in the original statement and proof of theorem [ sufeventess ] and for many enlightening conversations .
the main problem in this paper is to determine whether or not a homogeneous space @xmath2 has a tessellation .
this requires some method to determine whether or not a given discrete subgroup @xmath3 of @xmath1 acts properly discontinuously on @xmath2 .
y. benoist and t. kobayashi ( independently ) demonstrated that the cartan projection @xmath139 is an effective tool to study this question .
it is the foundation of almost all of our work in later sections . in this section ,
we introduce the cartan projection , and describe some of its basic properties .
first , however , we recall the notion of a proper action ( a generalization of properly discontinuous actions ) and of a cartan - decomposition subgroup . at the end of the section
, we use the cartan projection to briefly discuss the question of when there is a loss of generality in assuming that @xmath1 is connected . a topological group @xmath46 of homeomorphisms of a topological space @xmath11 acts _ properly _ on @xmath11 if , for every compact subset @xmath12 of @xmath11 , @xmath140 it is important to note that a _ discrete _ group of homeomorphisms of @xmath11 acts properly on @xmath11 if and only if it acts properly discontinuously on @xmath11 . for the special case where @xmath141 is a homogeneous space , the following lemma restates the definition of a proper action in more group - theoretic terms
[ proper<>chc ] a closed subgroup @xmath46 of @xmath1 acts properly on @xmath2 if and only if , for every compact subset @xmath12 of @xmath1 , the intersection @xmath142 is compact . if @xmath12 is any compact subset of @xmath1 , then @xmath143 is a compact subset of @xmath2 ; furthermore , any compact subset of @xmath2 is contained in one of the form @xmath144 .
we have @xmath145 this has the following well - known , easy consequence .
[ chcproper ] suppose @xmath0 , @xmath94 , @xmath46 , and @xmath146 are closed subgroups of @xmath1 .
if * @xmath46 acts properly on @xmath2 , and * there is a compact subset @xmath12 of @xmath1 , such that @xmath147 and @xmath148 , then @xmath146 acts properly on @xmath149 .
the following definition describes the subgroups to which the calabi - markus phenomenon applies ( cf .
example [ calabimarkuscircle ] ) .
we say that @xmath0 is a _ cartan - decomposition subgroup _ of @xmath1 if @xmath150 ( see notation [ simdefn ] ) . [ aiscds ] from the cartan decomposition @xmath101 , we know that @xmath119 is a cartan - decomposition subgroup . [ cdsconj ] any conjugate of a cartan - decomposition subgroup is a cartan - decomposition subgroup .
[ calabimarkus ] if @xmath0 is a cartan - decomposition subgroup of @xmath1 , and @xmath3 is a discrete subgroup of @xmath1 that acts properly discontinuously on @xmath2 , then @xmath3 is finite . because @xmath0 is a cartan - decomposition subgroup , there is a compact subset @xmath12 of @xmath1 , such that @xmath151 .
however , from lemma [ proper<>chc ] , we know that @xmath152 is finite
. therefore @xmath153 is finite .
the following well - known , easy fact is a direct consequence of the calabi - markus phenomenon .
it is an important first step toward determining which homogeneous spaces have tessellations .
[ cds->notess ] if @xmath0 is a cartan - decomposition subgroup of @xmath1 , such that @xmath2 is not compact , then @xmath2 does not have a tessellation .
[ a+defn ] * if @xmath1 is connected , let @xmath154 be the ( closed ) positive weyl chamber of @xmath119 in which the roots occurring in the lie algebra of @xmath120 are positive ( cf .
[ kandefn ] ) .
thus , @xmath154 is a fundamental domain for the action of the ( real ) weyl group of @xmath1 on @xmath119 .
* in the general case , let @xmath154 be a closed , convex fundamental domain for the action of the ( real ) weyl group of @xmath1 on @xmath119 , such that @xmath154 is contained in the ( closed ) positive weyl chamber of @xmath119 in which the roots occurring in the lie algebra of @xmath120 are positive .
for each element @xmath155 of @xmath1 , the cartan decomposition @xmath9 implies that there is an element @xmath156 of @xmath154 with @xmath157 .
in fact , the element @xmath156 is unique , so there is a well - defined function @xmath158 given by @xmath159 .
we remark that the function @xmath139 is continuous and proper ( that is , the inverse image of any compact set is compact ) .
the following crucial result of y. benoist provides a uniform estimate on the variation of @xmath139 over disks of bounded radius .
( a related result was proved , independently and simultaneously , by t. kobayashi ( * ? ? ?
the proof is both elementary and elegant .
however , it requires a bit of notation , so we postpone it to [ calcsect - benoist ] ( and , for concreteness , we will assume that @xmath1 is either @xmath73 or @xmath52 in the proof ) .
[ bddchange ] for any compact subset @xmath12 of @xmath1 , there is a compact subset @xmath160 of @xmath119 , such that @xmath161 , for all @xmath33 . for subsets @xmath162 and @xmath163 of @xmath154 , we write @xmath164 if there is a compact subset @xmath12 of @xmath119 , such that @xmath165 and @xmath166 .
[ simvsmu ] for any subgroups @xmath94 and @xmath95 of @xmath1 , we have @xmath96 if and only if @xmath167 .
( @xmath168 ) let @xmath12 be a compact subset of @xmath1 , such that @xmath169 and @xmath170 .
choose a corresponding compact subset @xmath160 of @xmath119 , as in proposition [ bddchange ] .
then @xmath171 and , similarly , @xmath172 .
( @xmath173 ) let @xmath12 be a compact subset of @xmath119 , such that @xmath174 and @xmath175 .
then @xmath176 and , similarly , @xmath177 .
the special case where @xmath178 ( and @xmath94 is closed and almost connected ) can be restated as follows .
[ cdsvsmu ] @xmath0 is a cartan - decomposition subgroup of @xmath1 if and only if @xmath179 . [ rrank1-cds ] assume that @xmath28 .
the subgroup @xmath0 is a cartan - decomposition subgroup of @xmath1 if and only if @xmath0 is noncompact .
( @xmath173 ) we have @xmath180 , and , because @xmath139 is a proper map , we have @xmath181 as @xmath182 in @xmath0 .
because @xmath183 , we know that @xmath154 is homeomorphic to the half - line @xmath184 ( with the point @xmath185 in @xmath154 corresponding to the endpoint @xmath186 of the half - line ) , so , by continuity , it must be the case that @xmath187 .
then corollary [ cdsvsmu ] implies that @xmath0 is a cartan - decomposition subgroup , but we provide the following direct proof that avoids any appeal to proposition [ bddchange ] . from the definition of @xmath139 , we have @xmath188 .
therefore @xmath189 so @xmath0 is a cartan - decomposition subgroup ( by taking @xmath190 in definition [ simdefn ] ) . by using lemma [ proper<>chc ] ,
the proof of corollary [ simvsmu ] also establishes the following .
[ proper<>mu(l ) ] suppose @xmath0 and @xmath46 are closed subgroups of @xmath1 .
the subgroup @xmath46 acts properly on @xmath2 if and only if @xmath191 is compact , for every compact subset @xmath12 of @xmath119 .
as was mentioned in remark [ hdisconnected ] , we may assume , without loss of generality , that @xmath0 is connected . however , it may not be possible to assume that @xmath1 is connected , because , although there are no known examples , it is possible that the following question has an affirmative answer .
[ assumegconn ? ] does there exist a homogeneous space @xmath2 ( satisfying assumption [ standing ] ) , such that @xmath192 has a tessellation , but @xmath2 does not have a tessellation ? if @xmath3 is a crystallographic group for @xmath192 , then it is easy to see that @xmath15 is compact .
however , the following example shows that @xmath3 may not act properly discontinuously on @xmath2 .
let * @xmath193 , * @xmath194 be the automorphism of @xmath195 that interchanges the two factors ( that is , @xmath196 ) , * @xmath197 ( semidirect product ) , and * @xmath3 be a cocompact lattice in @xmath46 ( cf . [ classical]([classical - borel ] ) ) .
then @xmath198 , and @xmath3 is a crystallographic group for @xmath199 ( see example [ g = lxh ] ) . however , @xmath200
, so @xmath3 does not act properly on @xmath2 ( see [ proper<>chc ] with @xmath201 ) .
even so , @xmath2 does have a tessellation , because the diagonal embedding @xmath202 is a crystallographic group for @xmath2 .
thus , this example does not provide an answer to question [ assumegconn ? ] . in this example
, @xmath194 represents an element of the weyl group of @xmath1 that does not belong to the weyl group of @xmath203 .
the following proposition shows that this is a crucial ingredient in the construction .
let @xmath3 be a crystallographic group for @xmath192 .
if the ( real ) weyl group of @xmath1 is same as the ( real ) weyl group of @xmath203 , then @xmath3 is a crystallographic group for @xmath2 . by assumption
, we may choose the same fundamental domain @xmath154 for the weyl groups of @xmath1 and @xmath203 .
let @xmath204 and @xmath205 be the cartan projections ; then @xmath206 is the restriction of @xmath139 to @xmath203 . for simplicity ,
assume , without loss of generality , that @xmath207 ( for example , assume @xmath0 is connected ) .
then , for any compact subset @xmath12 of @xmath119 , we have @xmath208 is finite ( see [ proper<>mu(l ) ] ) .
thus , @xmath3 acts properly discontinuously on @xmath2 ( see [ proper<>mu(l ) ] ) , as desired .
a. borel and j. tits ( * ? ? ?
14.6 , p. 147 ) proved that if @xmath1 is zariski connected , then every element of the weyl group of @xmath1 has a representative in @xmath203 .
also , any element of the weyl group must act as an automorphism of the root system .
thus , we have the following corollary .
let @xmath3 be a crystallographic group for @xmath192 .
if either * @xmath1 is zariski connected , or * every automorphism of the real root system of @xmath203 belongs to the weyl group of the root system , then @xmath3 is a crystallographic group for @xmath2 . 1 .
if @xmath51 , then @xmath1 is zariski connected ( because @xmath209 is connected ( * ? ? ?
* thm . 2.1.9 , p. 60 ) )
, so @xmath2 has a tessellation if and only if @xmath192 has a tessellation .
2 . more generally , if @xmath210 or @xmath52 ( with @xmath211 )
, then every automorphism of the real root system of @xmath203 belongs to the weyl group of the root system ( cf . figure [ rootspict ] ) , so @xmath2 has a tessellation if and only if @xmath192 has a tessellation .
if @xmath212 , where @xmath194 is the cartan involution of @xmath42 , then @xmath194 represents an element of the weyl group of @xmath1 that does not belong to the weyl group of @xmath203 , so the proposition does not apply to @xmath1 .
however , this does not matter : if neither @xmath0 nor @xmath2 is compact , then theorem [ sl3->notess ] implies that @xmath192 has no tessellations , so @xmath2 has no tessellations either .
this section recalls a technical result that often allows us to assume that @xmath0 is a subgroup of @xmath121
. it also recalls some basic topological properties of such subgroups , and also recalls a simple observation relating these subgroups to the root spaces of the lie algebra @xmath213 .
an element @xmath155 of @xmath1 is : * _ hyperbolic _ if @xmath155 is conjugate to an element of @xmath119 ; * _ unipotent _ if @xmath155 is conjugate to an element of @xmath120 ; * _ elliptic _ if @xmath155 is conjugate to an element of @xmath118 . [ jordandecomp ]
each @xmath33 has a unique decomposition in the form @xmath214 , such that * @xmath156 is hyperbolic , @xmath215 is unipotent , and @xmath216 is elliptic ; and * @xmath156 , @xmath215 , and @xmath216 all commute with each other
. [ jordancommute ] if @xmath214 is the real jordan decomposition of some element @xmath155 of @xmath1 , then @xmath156 , @xmath215 , and @xmath216 commute , not only with each other , but also with any element of @xmath1 that commutes with @xmath155 .
this is because the real jordan decomposition of @xmath217 is @xmath218 if @xmath219 , then the uniqueness of the real jordan decomposition of @xmath155 implies @xmath220 , @xmath221 , and @xmath222 .
the following observation is a generalization of the fact that a collection of commuting triangularizable matrices can be simultaneously triangularized .
[ simultaneous ] if @xmath0 is abelian ( or , more generally , solvable ) , and is generated by hyperbolic and/or unipotent elements , then @xmath0 is conjugate to a subgroup of @xmath121 . because of the following result
, we usually assume @xmath91 ( by replacing @xmath0 with a conjugate of @xmath122 ) .
[ hcanbean ] if @xmath0 is connected , then there is a closed , connected subgroup @xmath122 of @xmath1 and a compact , connected subgroup @xmath12 of @xmath1 , such that 1 . [ hcanbean - inan ] @xmath122 is conjugate to a subgroup of @xmath121 ; 2 . [ hcanbean - ch = ch ] @xmath223 is a subgroup of @xmath1 ; and 3 .
[ hcanbean - d(h ) ] @xmath224 ( see notation [ d(h)-defn ] ) . moreover , it is easy to see from ( [ hcanbean - ch = ch ] ) that the homogeneous space @xmath2 has a tessellation if and only if @xmath123 has a tessellation . first , let us note that every connected subgroup of @xmath121 is closed ( see [ solvable]([solvable - h = rn ] ) and [ ansc ] ) , so we do not need to show that @xmath122 is closed .
second , let us note that ( [ hcanbean - d(h ) ] ) is a consequence of ( [ hcanbean - inan ] ) and ( [ hcanbean - ch = ch ] ) . to see this ,
let @xmath225 be a maximal compact subgroup of @xmath226 that contains @xmath12 .
then a standard argument shows that @xmath227 is a maximal compact subgroup of @xmath0 .
( because all maximal compact subgroups of @xmath226 are conjugate , there is some @xmath228 , such that @xmath229 is a maximal compact subgroup of @xmath0 that contains @xmath227 . since @xmath230 , we know that @xmath12 normalizes @xmath225 , so we may assume @xmath231 ; thus , @xmath155 normalizes @xmath0
. then @xmath232 contains @xmath227 .
because @xmath227 is compact , this implies that @xmath155 normalizes @xmath227 .
so @xmath233 is a maximal compact subgroup of @xmath0 . )
therefore @xmath234 similarly , @xmath235 . since @xmath236
, we conclude that @xmath224 , as desired .
[ hcanbean - ss ] assume @xmath0 is semisimple .
we have an iwasawa decomposition @xmath84 ; let @xmath237 and @xmath238 .
assume @xmath239 is a one - parameter subgroup
. let * @xmath240 be the real jordan decomposition of @xmath241 ( see [ jordandecomp ] ) ; * @xmath242 ; and * @xmath243 be the closure of @xmath244 .
( lemma [ simultaneous ] implies that @xmath122 is conjugate to a subgroup of @xmath121 . )
[ hcanbean - abel ] assume @xmath0 is abelian .
we may write @xmath0 as a product of one - parameter subgroups : @xmath245 let @xmath246 be the real jordan decomposition of @xmath247 ( see [ jordandecomp ] ) .
note that @xmath248 , @xmath249 , and @xmath250 commute , not only with each other , but also with every @xmath251 , @xmath252 , and @xmath253 ( see [ jordancommute ] ) .
let @xmath254 and let @xmath255 .
( lemma [ simultaneous ] implies that @xmath122 is conjugate to a subgroup of @xmath121 . ) the general case . from the levi decomposition @xcite , we know that there is a connected , semisimple subgroup @xmath46 of @xmath0 and a connected , solvable , normal subgroup @xmath256 of @xmath0 , such that @xmath257 ( and @xmath258 is finite ) .
let @xmath259 $ ] , so @xmath162 is a connected , normal subgroup of @xmath0 , and @xmath162 is conjugate to a subgroup of @xmath120 ( cf .
* cor . 2.7.1 , p. 51 ) ) .
by modding out @xmath162 , we ( essentially ) reduce to the direct product of cases [ hcanbean - ss ] and [ hcanbean - abel ] . for @xmath0 and @xmath122 as in lemma [ hcanbean ] ,
proposition [ an / h = rd ] ( and [ ansc ] ) implies that if @xmath260 , then @xmath261 is not compact ; also , proposition [ solvable]([solvable - nocpct ] ) ( and [ ansc ] ) implies that if @xmath262 , then @xmath122 is not compact .
therefore : * @xmath263 if and only if @xmath2 is compact ; and * @xmath264 if and only if @xmath0 is compact . thus , if neither @xmath0 nor @xmath2 is compact , then @xmath122 is a nontrivial , proper subgroup of @xmath121 .
everything is this subsection is well known , though somewhat scattered in the literature .
the main results are propositions [ solvable ] and [ an / h = rd ] , which , together with corollary [ ansc ] , show that connected subgroups of @xmath121 and their homogeneous spaces are very well behaved topologically .
corollary [ fiberbundle ] , on the homology of very simple quotient spaces , is also used in later sections .
we begin with the easy case of abelian groups .
this lemma generalizes almost verbatim to solvable groups ( see [ solvable ] ) , but the proof in that generality is not as trivial .
[ abelian ] let @xmath256 be a @xmath26-connected , abelian lie group . 1
. [ abelian - h = rn ] if @xmath0 is a connected subgroup of @xmath256 , then @xmath0 is closed , simply connected , and isomorphic to @xmath265 , for some @xmath266 .
[ abelian - hcapl ] if @xmath0 and @xmath46 are connected subgroups of @xmath256 , then @xmath267 is connected .
[ abelian - nocpct ] if @xmath12 is a compact subgroup of @xmath256 , then @xmath12 is trivial . because @xmath256 is abelian and 1-connected
, the exponential map is a lie group isomorphism from the additive group of the lie algebra @xmath268 onto @xmath256 .
( [ abelian - h = rn ] ) let @xmath269 . because the exponential map is a lie group isomorphism ( hence a diffeomorphism ) , and because @xmath270 is a closed @xmath266-submanifold of @xmath268 , we know that @xmath271 is a closed @xmath266-submanifold of @xmath256 .
of course , @xmath271 is contained in @xmath0 , which is also a @xmath266-submanifold of @xmath256 . because the dimensions are the same , we know that @xmath271 is open in @xmath0 . also , because @xmath271 is closed in @xmath256 , we know that @xmath271 is closed in @xmath0 .
therefore @xmath272 ( because @xmath0 is connected ) . finally , we know that @xmath273 is a diffeomorphism from its domain @xmath274 onto its image @xmath0 .
( [ abelian - hcapl ] ) from ( [ abelianpf - h = rn - exp(h ) ] ) , we have @xmath275 and , similarly , @xmath276 . also , because @xmath277 is bijective ,
we have @xmath278 . therefore @xmath279 is connected .
( [ abelian - nocpct ] ) because @xmath265 is not compact ( for @xmath280 ) , we know , from [ abelian]([abelian - h = rn ] ) , that @xmath281 is trivial ; so @xmath12 is finite . since @xmath282 has no elements of finite order , we conclude that @xmath12 is trivial .
as is usual in the theory of solvable groups , the main results of this section are proved by induction , based on modding out some normal subgroup @xmath46 . to be effective , this method requires an understanding of the quotient space @xmath283 .
the information we need ( even if @xmath46 is not normal ) comes from the following elementary observation , because @xmath256 is a principal @xmath46-bundle over @xmath283 .
[ trivialbundle ] let @xmath284 be a principal @xmath0-bundle over a manifold @xmath11 . 1 .
[ trivialbundle - h = rn ] if @xmath0 is diffeomorphic to @xmath17 , then 1 .
@xmath284 is @xmath0-equivariantly diffeomorphic to @xmath285 , so 2 .
@xmath284 is homotopy equivalent to @xmath11 .
[ trivialbundle - m = rn ] if @xmath11 is diffeomorphic to @xmath17 , then 1 .
@xmath284 is @xmath0-equivariantly diffeomorphic to @xmath285 , so 2 .
@xmath284 is homotopy equivalent to @xmath0 .
any principal bundle with a section is trivial ( * ? ? ?
* cor . 4.8.3 , p. 48 ) .
if either the fiber or the base is contractible , then there is no obstruction to constructing a section ( * ? ? ?
2.7.1(h1 ) , p. 21 ) , so @xmath284 is trivial : @xmath286 .
( the diffeomorphism can be taken to be @xmath0-equivariant , with respect to the natural @xmath0-action on @xmath285 , given by @xmath287 . )
then the conclusions on homotopy equivalence follow from the fact that @xmath17 is contractible ( that is , homotopically trivial ) .
we recall the long exact sequence of the fibration @xmath288 : [ htpyexactfibration ] let @xmath0 be a closed subgroup of a lie group @xmath256 .
there is a ( natural ) long exact sequence of homotopy groups : @xmath289 [ r / hsc ] let @xmath0 be a closed subgroup of a @xmath26-connected lie group @xmath256 .
the homogeneous space @xmath290 is simply connected if and only if @xmath0 is connected .
because @xmath256 is 1-connected , we have @xmath291 , so , from ( [ htpyexactfibration ] ) , we know that the sequence @xmath292 is exact . thus , @xmath293 , so the desired conclusion is immediate . as a step toward proposition [ solvable ] , we prove two special cases that describe the topology of normal subgroups .
[ r = rd ] if @xmath256 is a @xmath26-connected , solvable lie group , then @xmath256 is diffeomorphic to @xmath294 , for some @xmath99
. we may assume the group @xmath256 is nonabelian ( otherwise , the desired conclusion is given by lemma [ abelian]([abelian - h = rn ] ) ) .
then , because @xmath256 is solvable , there is a nontrivial , connected , proper , closed , normal subgroup @xmath46 of @xmath256 . since @xmath283 is simply connected ( see [ r / hsc ] ) , and @xmath295 , we may assume , by induction on @xmath296 , that @xmath283 is diffeomorphic to some @xmath297 . therefore 1 .
[ r = rdpf - r = prod ] @xmath256 is diffeomorphic to @xmath298 and 2 .
rdpf - l = r ] @xmath46 is homotopy equivalent to @xmath256 ( see [ trivialbundle]([trivialbundle - m = rn ] ) ) . because @xmath256 is @xmath26-connected , ( [ r = rdpf - l = r ] ) implies that @xmath46 is @xmath26-connected ; hence
, @xmath46 is a 1-connected , solvable lie group , so we may assume , by induction on @xmath296 , that @xmath46 is diffeomorphic to some @xmath299 .
thus , ( [ r = rdpf - r = prod ] ) implies that @xmath256 is diffeomorphic to @xmath300 , as desired .
[ rnormal ] if @xmath256 is a @xmath26-connected , solvable lie group , then every connected , closed , normal subgroup of @xmath256 is @xmath26-connected .
the following proposition is a nearly complete generalization of lemma [ abelian ] to the class of solvable groups .
there are two exceptions : 1 . of course , subgroups of a solvable group may not be abelian , so the conclusion in [ abelian]([abelian - h = rn ] ) that @xmath0 is isomorphic to some @xmath265 must be weakened to the conclusion that @xmath0 is diffeomorphic to some @xmath265 .
the intersection of connected subgroups is not always connected ( see [ hcapldisconn ] ) , so we add the restriction that @xmath46 is normal to [ abelian]([abelian - hcapl ] ) .
( we remark that no such restriction is necessary if @xmath301 , because the exponential map is a diffeomorphism from @xmath268 onto @xmath256 in this case @xcite . )
[ hcapldisconn ] let @xmath302 @xmath303 then @xmath256 , being diffeomorphic to @xmath304 , is 1-connected ; and @xmath305 and @xmath306 are connected subgroups .
but @xmath307 is not connected .
[ solvable ] let @xmath256 be a @xmath26-connected , solvable lie group . 1 . [ solvable - h = rn ] if @xmath0 is a connected subgroup of @xmath256 , then @xmath0 is closed , simply connected , and diffeomorphic to some @xmath294 .
[ solvable - hcapl ] if @xmath0 and @xmath46 are connected subgroups of @xmath256 , and @xmath46 is normal , then @xmath267 is connected .
[ solvable - nocpct ] if @xmath12 is a compact subgroup of @xmath256 , then @xmath12 is trivial .
( [ solvable - hcapl ] ) we may assume @xmath46 is nontrivial , so @xmath308 .
thus , by induction on @xmath296 , using ( [ solvable - h = rn ] ) , we may assume that @xmath309 is a closed , simply connected subgroup of @xmath283 .
then , since @xmath310 is homeomorphic to @xmath309 , we see that @xmath311 so lemma [ r / hsc ] implies that @xmath267 is connected .
( [ solvable - h = rn ] ) because @xmath256 is solvable , there is a connected , closed , proper , normal subgroup @xmath46 of @xmath256 , such that @xmath283 is abelian .
we know that @xmath46 is 1-connected ( see [ rnormal ] ) , so , by induction on @xmath296 , we may assume that every connected subgroup of @xmath46 is closed and simply connected . from ( [ solvable - hcapl ] )
, we know that @xmath267 is connected , so we conclude that @xmath267 is closed , and @xmath312 from ( [ htpyexactfibration ] ) ( with @xmath0 in the place of @xmath256 , and @xmath46 in the place of @xmath0 ) , together with ( [ pi1(h / hcapl ) ] ) and ( [ pi1(hcapl ) ] ) , we conclude that @xmath313 ; that is , @xmath0 is simply connected .
so ( [ r = rd ] ) implies @xmath0 is diffeomorphic to some @xmath294 . because both @xmath309 and @xmath267 are closed , it is not difficult to see that @xmath0 is closed .
( [ solvable - nocpct ] ) because @xmath256 is solvable , there is a connected , closed , proper , normal subgroup @xmath46 of @xmath256 , such that @xmath283 is abelian .
we know that @xmath283 is 1-connected ( see [ r / hsc ] ) , so @xmath283 has no nontrivial , compact subgroups ( see [ abelian]([abelian - nocpct ] ) ) ; thus , we must have @xmath314 .
therefore , @xmath12 is a compact subgroup of @xmath46 .
then , since @xmath46 is 1-connected ( see [ rnormal ] ) , we may conclude , by induction on @xmath296 , that @xmath12 is trivial .
[ ansc ] @xmath121 is a @xmath26-connected , solvable lie group . because @xmath1 is linear
, it is a subgroup of some @xmath315 . replacing @xmath1 by a conjugate
, we may assume that @xmath121 is contained in the group @xmath316 of upper triangular matrices with positive diagonal entries ( cf .
[ simultaneous ] ) .
the matrix entries provide an obvious diffeomorphism from @xmath316 onto @xmath317 , so @xmath316 is 1-connected .
thus , proposition [ solvable]([solvable - h = rn ] ) implies that @xmath121 is simply connected .
the following observation will be used in sections [ dimhsect ] and [ existencesect ] .
[ fiberbundle ]
let @xmath318 be a connected subgroup of @xmath121 , and suppose we have a proper , @xmath319 action of @xmath318 on a manifold @xmath11 .
then @xmath11 and @xmath320 have the same homology . because the action is proper , we know that the stabilizer of each point of @xmath11 is compact .
however , @xmath318 has no nontrivial compact subgroups ( see [ solvable]([solvable - nocpct ] ) ) .
thus , the action is free .
because the action is free , proper , and @xmath319 , it is easy to see that the manifold @xmath11 is a principal fiber bundle over the quotient @xmath320 ( * ? ? ?
furthermore , the fiber @xmath318 of the bundle is contractible ( see [ solvable]([solvable - h = rn ] ) ) , so lemma [ trivialbundle]([trivialbundle - h = rn ] ) implies that @xmath11 homotopy equivalent to @xmath320 .
therefore , the spaces @xmath11 and @xmath320 have the same homology . for the special case where @xmath320 is a homogeneous space of a solvable group , the following more detailed result describes the topology of @xmath320 , not just its homology .
[ an / h = rd ] if @xmath0 is any connected subgroup of a @xmath26-connected , solvable lie group @xmath256 , then @xmath290 is diffeomorphic to the euclidean space @xmath294 , for some @xmath99 . because @xmath256 is solvable
, it has a nontrivial , connected , closed , abelian , normal subgroup @xmath46 .
since @xmath46 is abelian and @xmath267 is connected ( see [ solvable]([solvable - hcapl ] ) ) , we know that @xmath321 is a 1-connected abelian group ( see [ r / hsc ] ) , so it is isomorphic to some @xmath297 ( see [ abelian]([abelian - h = rn ] ) ) .
we know @xmath0 is closed ( see [ solvable]([solvable - h = rn ] ) ) . also ,
since @xmath46 is nontrivial , we have @xmath308 , so we may assume , by induction on @xmath296 , that @xmath322 is diffeomorphic to some @xmath299 .
now @xmath256 is a principal @xmath323-bundle over @xmath324 . because @xmath325 , this bundle is trivial ( see [ trivialbundle]([trivialbundle - m = rn ] ) ) :
@xmath256 is @xmath323-equivariantly diffeomorphic to @xmath326 .
then @xmath327 as desired .
the following well - known observation puts an important restriction on the subspaces of @xmath329 that are normalized by a torus .
it is an ingredient in our case - by - case analysis of all possible subgroups of @xmath121 in sections [ suflargesect ] and [ proofsect ] .
[ rootdecomp ] let * @xmath330 be the set of weights of @xmath119 on @xmath331 ( in other words , the set of all positive real roots of @xmath1 ) ; * @xmath328 be a subgroup of @xmath119 ; * @xmath332 ; * @xmath333 , where the sum is over all @xmath334 , such that the restriction of @xmath194 to @xmath328 is the same as the restriction of @xmath335 to @xmath328 ; * @xmath336 , where the sum is over all @xmath334 , such that the restriction of @xmath194 to @xmath328 is not the same as the restriction of @xmath335 to @xmath328 . if @xmath337 is any @xmath338-subspace of @xmath329 normalized by @xmath328 , then @xmath339 . since @xmath340 , we know that the elements of @xmath341 are simultaneously diagonalizable ( over @xmath338 ) , so their restrictions to the invariant subspace @xmath337 are also simultaneously diagonalizable ( cf .
* thms . 26 and 27 in 3.12 , pp .
167168 ) ) .
thus , @xmath337 is a direct sum of weight spaces : @xmath342 for each weight @xmath343 of @xmath328 on @xmath337 , we have @xmath344 so @xmath345 and @xmath346 the conclusion follows .
in this section , we prove corollary [ tess->dim>1,2 ] , an _ a priori _ lower bound on @xmath137 . on the way , we recall a result of t. kobayashi that will also be used several times in later sections , and we establish that crystallographic groups have only one end . the following theorem is essentially due to t. kobayashi .
( kobayashi assumed that @xmath0 is reductive , but h. oh and d. witte ( * ? ? ? * thm .
3.4 ) pointed out that , by using lemma [ hcanbean ] , this restriction can be eliminated . )
the proof here is based on kobayashi s original argument and the modifications of oh - witte , but uses less sophisticated topology .
namely , instead of group cohomology and the spectral sequence of a covering space , we use only some basic properties of homology groups of manifolds ( including lemma [ fiberbundle ] ) .
these comments also apply to theorem [ construct - tess ] .
[ noncpctdim ] let @xmath0 and @xmath94 be closed , connected subgroups of @xmath1 , and assume there is a crystallographis group @xmath3 for @xmath2 , such that @xmath3 acts properly discontinuously on @xmath149 . then : 1 .
[ noncpctdim - notess ] we have @xmath347 .
[ noncpctdim - tess ] if @xmath348 , then @xmath349 is compact , so @xmath149 has a tessellation . by lemma [ hcanbean ]
, we may assume @xmath350 .
( so @xmath351 and @xmath352 ( see [ d(h)=dimh ] ) . ) from lemma [ fiberbundle ] , we know that @xmath353 and @xmath354 have the same homology .
therefore @xmath355 with equality if and only if @xmath356 is compact ( * ? ? ?
8.3.4 , p. 260 ) . similarly , we have @xmath357 combining these two statements , we conclude ( [ noncpctdim - notess ] ) that @xmath358 and , furthermore , ( [ noncpctdim - tess ] ) that equality holds if and only if @xmath349 is compact .
[ noncpct - dim - notess ] let @xmath0 and @xmath94 be closed , connected subgroups of @xmath1 , such that @xmath359 . if there is a compact subset @xmath12 of @xmath119 , such that @xmath360 , then @xmath2 does not have a tessellation .
suppose @xmath3 is a crystallographic group for @xmath2 .
( this will lead to a contradiction . ) because @xmath3 acts properly discontinuously on @xmath2 , the assumption on @xmath361 implies that @xmath3 also acts properly discontinuously on @xmath149 ( cf .
[ proper<>mu(l ) ] ) .
so theorem [ noncpctdim]([noncpctdim - notess ] ) yields a contradiction .
it is easy to see that crystallographic groups are finitely generated ; we now show that they have only one end ( see [ tess->1end ] ) .
[ 1enddefn ]
let @xmath318 be a finite generating set for an ( infinite ) group @xmath3 .
we say that @xmath3 has _ only one end _ if , for every partition @xmath362 of @xmath3 into three disjoint sets @xmath363 , @xmath364 , and @xmath12 , such that @xmath363 and @xmath364 are infinite , but @xmath12 is finite , there exists @xmath365 and @xmath366 , such that @xmath367 .
( this does not depend on the choice of the generating set @xmath318 . )
the following observation is a straightforward reformulation of definition [ 1enddefn ] ( obtained by letting @xmath368 and @xmath369 ) .
[ 1endcomplement ] let @xmath318 be a finite generating set for an infinite group @xmath3 . if @xmath363 and @xmath160 are subsets of @xmath3 , such that * @xmath363 is infinite , * @xmath160 is finite , and * @xmath370 , for every @xmath366 , then the complement @xmath371 is finite . definition [ 1enddefn ] is often stated in the language of cayley graphs : the _ cayley graph _ of @xmath3 , with respect to the generating set @xmath318 , is the graph @xmath372 whose vertex set @xmath163 and edge set @xmath373 are given by : @xmath374 the group @xmath3 has only one end if and only if , for every finite subset @xmath12 of @xmath3 , the graph @xmath375 has only one infinite component .
the following lemma is not difficult , but , unfortunately , we do not have a proof that is both short and elementary . [ hn = an ] if @xmath376 , then ,
for some @xmath377 , the conjugate @xmath378 is normalized by @xmath119 .
[ codim>1 ] if @xmath379 , and @xmath2 is not compact , then @xmath2 does not have a tessellation .
it suffices to show that @xmath0 is a cartan - decomposition subgroup of @xmath1 ( see [ cds->notess ] ) .
we may assume , without loss of generality , that @xmath91 ( see [ hcanbean ] ) ; then @xmath380 ( see [ d(h)=dimh ] and [ d(g ) ] ) .
a theorem of b. kostant ( * ? ? ?
5.1 ) implies that @xmath120 is a cartan - decomposition subgroup , so we may assume @xmath381 ; then @xmath382 .
therefore @xmath383 hence @xmath384 , so , from lemma [ hn = an ] , we see that , after replacing @xmath0 by a conjugate subgroup , we may assume that @xmath0 is normalized by @xmath119 .
then , letting @xmath385 and @xmath386 in ( [ rootdecomp ] ) , we see that @xmath387 .
since @xmath384 , we have @xmath388 , so this implies that @xmath389 ; therefore @xmath0 contains @xmath119 . since @xmath119 is a cartan - decomposition subgroup ( see [ aiscds ] ) , this implies @xmath0 is a cartan - decomposition subgroup , as desired .
a topological space @xmath11 is _ connected at @xmath390 _ if every compact subset @xmath391 is contained in a compact subset @xmath392 , such that the complement @xmath393 is connected .
[ tess->1end ] if @xmath3 is a crystallographic group for @xmath2 , then @xmath3 is finitely generated and has only one end .
assume , without loss of generality , that @xmath91 ( see [ hcanbean ] ) .
then @xmath0 is torsion free , so @xmath3 must act freely on @xmath2 ; therefore @xmath15 is a compact manifold ( rather than an orbifold ) . because @xmath3 is essentially the fundamental group of @xmath116 ( specifically , @xmath394 ) , and the fundamental group of any compact manifold
is finitely generated ( * ? ? ?
* thm . 6.16 , p. 95 ) , we know that @xmath3 is finitely generated . from the iwasawa decomposition @xmath92
, we see that @xmath2 is homeomorphic to @xmath395 , and proposition [ an / h = rd ] asserts that @xmath396 is homeomorphic to @xmath294 , for some @xmath99 .
obviously , we must have @xmath397 , and we may assume @xmath2 is not compact ( otherwise , @xmath3 is finite , so the desired conclusion is obvious ) , so corollary [ codim>1 ] implies that @xmath398 .
thus , we conclude that @xmath2 is connected at @xmath390 .
to complete the proof , we use a standard argument ( cf .
* @xmath399 , p. 5 ) ) to show that , because @xmath2 is connected at @xmath390 and @xmath15 is compact , the group @xmath3 has only one end . to begin ,
note that there is a compact subset @xmath391 of @xmath2 , such that @xmath400 .
let @xmath401 ( cf .
* ( ii ) , p. 195 ) ) . because @xmath3 acts properly discontinuously on @xmath2 , we know that @xmath402 is finite ; let @xmath318 be a finite generating set for @xmath3 , such that @xmath403
suppose @xmath404 , with @xmath405 and @xmath406 .
( we wish to show there exist @xmath365 and @xmath407 , such that @xmath408 ; this establishes that @xmath3 has only one end . )
because @xmath2 is connected at @xmath390 , there is a compact subset @xmath392 of @xmath2 , containing @xmath409 , such that @xmath410 is connected .
because @xmath411 , we have @xmath412 because @xmath3 acts properly discontinuously on @xmath2 , we know @xmath413 and @xmath414 are closed ( and neither is contained in @xmath392 ) , so connectivity implies that @xmath415 : there exist @xmath365 and @xmath416 , such that @xmath417 . let @xmath418 ; then @xmath365 , @xmath419 , and @xmath420 so @xmath421 , as desired . [ tess->misswall ] assume @xmath423 .
let * @xmath146 and @xmath424 be the two walls of @xmath154 , and * @xmath3 be a crystallographic group for @xmath2 .
if @xmath0 is not compact , then there exists @xmath425 , such that , for every compact subset @xmath12 of @xmath119 , the intersection @xmath426 is finite . : ( a ) @xmath427 can not be on both sides of @xmath428 , because @xmath3 has only one end .
( b ) therefore , @xmath427 stays away from @xmath429 . ]
suppose there is a compact subset @xmath12 of @xmath119 , such that each of @xmath430 and @xmath431 is infinite .
( this will lead to a contradiction . )
let @xmath318 be a ( symmetric ) finite generating set for @xmath3 ( see [ tess->1end ] ) .
we may assume @xmath12 is so large that @xmath432 for every @xmath37 ( see [ bddchange ] ) .
we may also assume that @xmath12 is convex and symmetric .
because @xmath3 acts properly on @xmath2 , there is a compact subset @xmath391 of @xmath119 , such that @xmath433 ( see [ proper<>mu(l ) ] ) .
furthermore , we may assume that @xmath434 .
let * @xmath435 , * @xmath436 be the union of all the connected components of @xmath11 that contain a point of @xmath146 , and * @xmath437 .
then @xmath363 is infinite ( because @xmath438 is infinite ) .
also , for any @xmath439 and @xmath440 , we have @xmath441 , so @xmath442 . since @xmath3 has only one end ( see [ tess->1end ] ) , this implies @xmath443 is finite ( see [ 1endcomplement ] ) .
because @xmath444 is infinite , we conclude that @xmath445 . because @xmath428 separates @xmath146 from @xmath424 , and every connected component of @xmath436 contains a point of @xmath146
, we conclude that @xmath446 .
this contradicts the fact that @xmath447 .
[ tess->misshk ] assume @xmath423 .
let 1 .
@xmath146 and @xmath424 be the two walls of @xmath154 ; and 2 .
@xmath94 and @xmath95 be closed , connected , nontrivial subgroups of @xmath1 , such that @xmath448 for @xmath449 .
if @xmath0 is not compact , then any crystallographic group for @xmath2 acts properly discontinuously on either @xmath149 or @xmath450 .
suppose @xmath3 acts properly discontinuously on * neither * @xmath149 * nor * @xmath450 .
( this will lead to a contradiction . ) from proposition [ proper<>mu(l ) ] , we know there is a compact subset @xmath12 of @xmath119 , such that each of @xmath451 and @xmath452 is infinite . then , since @xmath453 , we may assume ( by enlarging @xmath12 ) that each of @xmath438 and @xmath431 is infinite .
this contradicts the conclusion of proposition [ tess->misswall ] .
[ tess->dim>1,2 ] assume @xmath423 .
let 1 .
@xmath146 and @xmath424 be the two walls of @xmath154 ; and 2 .
@xmath94 and @xmath95 be closed , connected , nontrivial subgroups of @xmath1 ; such that @xmath448 for @xmath449 .
if @xmath2 has a tessellation , and @xmath0 is not compact , then @xmath454 the desired conclusion is obtained by combining corollary [ tess->misshk ] with theorem [ noncpctdim]([noncpctdim - notess ] ) . for @xmath455 , there does not exist a connected subgroup @xmath456 , such that @xmath453 ( cf .
[ sl3-b+ ] ) .
thus , corollary [ tess->dim>1,2 ] does not provide a lower bound on @xmath422 in this case .
although the following conjecture does not seem to have been stated previously in the literature , it is perhaps implicit in @xcite .
[ d(h)=1->notess ? ] if @xmath457 , then @xmath2 does not have a tessellation . in this section , we establish that the conjecture is valid in two cases : if either @xmath458 ( see [ 1drank2 ] ) or @xmath1 is almost simple ( see [ 1dsimple ] ) .
each of these illustrates a general theorem : for groups of real rank two , the conjecture follows from a theorem of y. benoist and f. labourie that is based on differential geometry ; h. oh and d. witte observed that , for simple groups , the conjecture follows from a theorem of g. a. margulis that is based on unitary representation theory . the following example is the only case of conjecture [ d(h)=1->notess ? ] that is needed in later sections .
( it is used in the proof of theorem [ sl3->notess ] . ) because @xmath459 and @xmath460 is almost simple , this example is covered both by the theorem of benoist - labourie and by the theorem of margulis , but it would be interesting to have an easy proof .
[ wallinsl3 ] assume @xmath461 , for @xmath132 , @xmath462 , or @xmath463 , and let @xmath464 then @xmath149 does not have a tessellation .
let us begin with an easy observation .
[ 1drank1 ] if @xmath457 and @xmath28 , then @xmath2 does not have a tessellation .
we may assume @xmath91 ( see [ hcanbean ] ) . from ( [ rrank1-cds ] )
, we know that @xmath0 is a cartan - decomposition subgroup , so lemma [ cds->notess ] implies that @xmath2 must be compact ; thus , the trivial group @xmath185 is a crystallographic group for @xmath2 .
however , since @xmath465 ( see [ d(g ) ] ) , and @xmath185 acts properly discontinuously on @xmath466 , this contradicts theorem [ noncpctdim]([noncpctdim - notess ] ) .
the proof of lemma [ 1drank1 ] shows that the dimension of every connected , cocompact subgroup of @xmath1 is at least @xmath467 .
this is a result of m. goto and h.c .
wang ( * ? ? ?
* ( 1.2 ) , p. 263 ) .
[ center->notess ] if @xmath0 is reductive and contains an element of @xmath119 in its center , then @xmath2 does not have a tessellation .
to illustrate the idea behind theorem [ center->notess ] , we give a direct proof of the following special case ( under an additional technical assumption ( see [ 1da->notesspf - h = z ] ) ) , which is sufficient for our needs .
( note that condition ( [ 1da->notesspf - h = z ] ) is satisfied for the subgroup @xmath94 of proposition [ wallinsl3 ] . )
benoist and labourie prove the general case by using a slightly different 1-form in place of the form @xmath335 that we define in step [ 1da->notesspf - omega ] .
[ 1da->notess ] if @xmath0 is a one - dimensional subgroup of @xmath119 , then @xmath2 does not have a tessellation .
suppose @xmath3 is a crystallographic group for @xmath2 .
( this will lead to a contradiction . ) by passing to finite - index subgroups , we may assume that @xmath1 and @xmath0 are connected .
we will construct a volume form @xmath468 on @xmath15 that is exact : @xmath469 .
the integral of @xmath468 over @xmath15 is the volume of @xmath15 , which is obviously not @xmath186 , but stokes theorem implies that the integral of any exact form over a closed manifold is @xmath186 .
this is a contradiction .
[ 1da->notesspf - omega ] construction of a certain @xmath470-form @xmath471 on @xmath116 .
let @xmath472 be the orthogonal complement to @xmath270 , under the killing form .
then @xmath473 is an @xmath474-invariant subspace of @xmath213 , such that @xmath475 and @xmath476 .
let @xmath477 be the projection with kernel @xmath473 , and let @xmath335 be the corresponding left - invariant @xmath270-valued 1-form on @xmath1 .
the space @xmath1 is a homogeneous principal @xmath0-bundle over @xmath2 .
it is well known ( * ? ? ?
ii.11.1 , p. 103 ) that @xmath335 is the connection form of a @xmath1-invariant connection on this bundle , and that the curvature form @xmath478 of this connection is given by @xmath479 \bigr)$ \qquad for $ x , y \in { \mathfrak{\lowercase{m}}}$.}\ ] ] also , because @xmath0 is abelian , the structure equation ( * ? ? ?
ii.5.2 , p. 77 ) implies @xmath480 by identifying the @xmath26-dimensional lie algebra @xmath270 with @xmath338 , we may think of @xmath335 and @xmath478 as ordinary ( that is , @xmath338-valued ) differential forms . because @xmath335 and ( hence ) @xmath478 are left - invariant , they determine well - defined forms @xmath481 and @xmath482 on @xmath483 .
( note that @xmath481 is a connection form on the principal @xmath0-bundle @xmath353 over @xmath15 , and the curvature form of this connection is @xmath482 . ) because @xmath0 is abelian , we have @xmath484 for all @xmath485 ( cf .
ii.5.1(c ) , p. 76 ) ) , so the horizontal form @xmath482 determines a well - defined form @xmath471 on the base space @xmath15 .
construction , for a certain @xmath486 , of a certain @xmath486-form @xmath487 on @xmath15 . identifying @xmath270 with @xmath338
provides an ordering on the weights of @xmath270 . let * @xmath488 be the @xmath186 weight space of @xmath489 on @xmath473 ; * @xmath490 be the sum of the positive weight spaces of @xmath491 on @xmath473 ; * @xmath492 be the sum of the negative weight spaces of @xmath491 on @xmath473 ; * @xmath493 ; and * @xmath139 be a nontrivial left - invariant @xmath486-form on @xmath1 , such that @xmath494 because @xmath139 is left - invariant , and @xmath495 , the form @xmath139 determines a well - defined differential form @xmath487 on @xmath15 .
( we remark that , because @xmath496 , condition ( [ mu(h+m ) ] ) implies that the form @xmath139 is unique up to a scalar multiple . )
[ 1da->notesspf - volume ] for a certain @xmath497 , the wedge product @xmath498 is a volume form on @xmath15 .
let @xmath499 .
it suffices to show that the restriction of @xmath478 to @xmath500 is a symplectic form .
it is obviously skew , so we need only show that it is nondegenerate .
thus , letting @xmath501 \bigr ) = 0 \,\ } & \mbox{{{\upshape(}\ref{omega(m , m)}{\upshape ) } } } \\ & = \{\ , x \in { \mathfrak{\lowercase{m}}}_+ \oplus { \mathfrak{\lowercase{m}}}_- \mid [ x , { \mathfrak{\lowercase{m}}}_+ \oplus { \mathfrak{\lowercase{m}}}_- ] \subset { \mathfrak{\lowercase{m } } } \,\ } & \mbox{(definition of~$\omega$ ) } \\ & = \{\ , x \in { \mathfrak{\lowercase{m}}}_+ \oplus { \mathfrak{\lowercase{m}}}_- \mid \langle { \mathfrak{\lowercase{h } } } \mid [ x , { \mathfrak{\lowercase{m}}}_+ \oplus { \mathfrak{\lowercase{m}}}_- ] \rangle_{\text{killing } } = 0 \,\ } & \mbox{(${\mathfrak{\lowercase{m } } } = { \mathfrak{\lowercase{h}}}^\perp$ ) } , \end{aligned}\ ] ] we wish to show @xmath502 .
there is no harm in passing to the complexification @xmath503 of @xmath213 .
let @xmath504 be a cartan subalgebra of @xmath503 that contains @xmath270 . because @xmath504 preserves the killing form , centralizes @xmath270 , and normalizes @xmath500 , we know that @xmath505 is @xmath506-invariant ; thus , @xmath505 is a sum of root spaces .
suppose there exists a nonzero element @xmath507 of @xmath505 , such that @xmath507 belongs to some root space @xmath508 .
( this will lead to a contradiction . )
there exists @xmath509 , such that @xmath510 \rangle_{\text{killing } } = \alpha(t)\ ] ] for all @xmath511 ( * ? ? ?
8.3(c ) , p. 37 ) . because @xmath512 , we have @xmath513 . then @xmath514 , so @xmath515 . we now know that @xmath516 and @xmath517 , so @xmath518 \rangle_{\text{killing } } = 0 $ ]
. we therefore conclude , from the definition of @xmath519 , that @xmath520 .
this is a contradiction .
the form @xmath471 is exact : we may write @xmath521 . let @xmath522 be the connection form of a flat connection on @xmath483 over @xmath15 .
( since the principal bundle is trivial ( see [ trivialbundle ] ) , it is obvious that there is a flat connection . ) for any vector field @xmath507 on @xmath15 , let @xmath523 be the lift of @xmath507 to a vector field on @xmath353 that is horizontal with respect to the flat connection @xmath522 .
since @xmath0 is abelian , there is a well - defined 1-form @xmath524 on @xmath525 given by @xmath526 then @xmath527 \bigr ) \right ) & \mbox{(definition of~$d$ ) } \\ & = \frac{1}{2 } \left ( \widetilde x \bigl
( \overline{\omega}(\widetilde y ) \bigr ) - \widetilde y \bigl ( \overline{\omega}(\widetilde x ) \bigr ) - \overline{\omega } \bigl ( \widetilde{[x , y ] } \bigr ) \right ) & \mbox{(definition of~$\breve{\phi}$ ) } \\ & = \frac{1}{2 } \left ( \widetilde x \bigl ( \overline{\omega}(\widetilde y ) \bigr ) - \widetilde y \bigl ( \overline{\omega}(\widetilde x ) \bigr ) - \overline{\omega } \bigl ( [ \widetilde x,\widetilde y ] \bigr ) \right ) & \mbox{($\breve{\omega}_0 $ is flat ) } \\ & = d \overline{\omega}(\widetilde x , \widetilde y ) & \mbox{(definition of~$d$ ) } \\ & = \overline{\omega}(\widetilde x , \widetilde y ) & \mbox{{{\upshape(}\ref{structeq}{\upshape ) } } } \\ & = \breve{\omega}(x , y ) & \mbox{(definition of~$\breve{\omega}$ ) } .
\end{aligned}\ ] ] for simplicity , assume that @xmath528 [ 1da->notesspf - closed ] we have @xmath529 .
let * @xmath530 be a basis of @xmath490 , and * @xmath531 be the dual basis of @xmath492 , with respect to the symplectic form @xmath478 on @xmath500 .
thus , @xmath532 .
let @xmath533 be a basis of @xmath534 , write @xmath535 = \sum_\ell a_{j , k}^\ell z_\ell \pmod { { \mathfrak{\lowercase{h } } } + { \mathfrak{\lowercase{m}}}_+ + { \mathfrak{\lowercase{m}}}_- } , \ ] ] and define @xmath536 = \sum_{j,\ell } a_{j , j}^\ell z_\ell .\ ] ] [ 1da->notesspf - closed - windep ] @xmath537 is independent of the choice of the basis @xmath538 of @xmath490 ( with the understanding that @xmath539 must be the dual basis of @xmath492 ) .
let @xmath540 for some @xmath541 with @xmath542 .
then @xmath543 so @xmath544 + [ x'_2,y'_2 ] + \sum_{j \ge 3 } [ x'_j , y'_j ] \\ & = \bigl ( [ x_1,y_1 ] + ( b / a)[x_2,y_1 ] \bigr ) + \bigl ( [ x_2,y_2 ] - ( b / a ) [ x_2,y_1 ] \bigr ) + \sum_{j \ge 3 } [ x_j , y_j ] \\ & = \sum_j [ x_j , y_j ] \\ & = w . \end{aligned}\ ] ] since @xmath530 can be transformed into any other basis of @xmath490 by a sequence of elementary operations as in ( [ elemop ] ) , we conclude that @xmath537 is independent of the choice of basis , as desired .
we have @xmath545 .
substep [ 1da->notesspf - closed - windep ] implies that @xmath537 is centralized by @xmath546 , so @xmath537 is in the center of @xmath547 .
let @xmath194 be a cartan involution of @xmath1 with @xmath548 for @xmath549 .
substep [ 1da->notesspf - closed - windep ] implies @xmath550 .
( from substep [ 1da->notesspf - closed - windep ] , we see that @xmath537 depends only on @xmath270 and the chosen identification of @xmath270 with @xmath338 ; @xmath194 reverses the choice of identification . )
thus , @xmath537 is a hyperbolic element of the center of @xmath547 . by assumption
[ 1da->notesspf - h = z ] , this implies @xmath545 , as desired . completion of step [ 1da->notesspf - closed ] .
let @xmath551 be the basis of @xmath552 dual to @xmath553 we may assume @xmath554 . then , because @xmath555 \subset { \mathfrak{\lowercase{m}}}_-$ ] and @xmath556 \subset { \mathfrak{\lowercase{m}}}_+$ ] , we have @xmath557 so @xmath558 therefore @xmath559 from the choice of @xmath560 , we have @xmath561 , so @xmath562 and @xmath563 hence @xmath564 since @xmath545 , we have @xmath565 , so the desired conclusion follows .
[ 1da->notesspf - volexact ] @xmath566 is exact .
we have @xmath567 a contradiction . from step [ 1da->notesspf - volume ] , we know
that @xmath568 on the other hand , step [ 1da->notesspf - volexact ] implies that this integral is zero .
this is a contradiction .
[ oppinv ] let @xmath569 be the opposition involution in @xmath154 ; that is , for @xmath570 , @xmath571 is the unique element of @xmath154 that is conjugate ( under an element of the weyl group ) to @xmath572 .
thus , for all @xmath573 , we have @xmath574 see ( [ sl3-oppinv ] ) for an explicit description of the opposition involution in @xmath455 .
for some groups , such as @xmath51 , we have @xmath575 for all @xmath573 ( see [ mu(h-1 ) ] ) ; in such a case , the opposition involution is simply the identity map on @xmath154 .
[ 1drank2 ] if @xmath457 and @xmath458 , then @xmath2 does not have a tessellation .
suppose @xmath3 is a crystallographic group for @xmath2 .
( this will lead to a contradiction . ) from ( [ 1drank1 ] ) , we know @xmath423 .
let @xmath146 and @xmath424 be the two walls of @xmath154 and , for @xmath425 , let @xmath576 . because @xmath429 is a ray ( that is , a one - parameter semigroup ) , it is clear that @xmath456 is a subgroup of @xmath119 . from proposition [ tess->misswall ] , we know that there is some @xmath577 , such that @xmath578 for every compact subset @xmath12 of @xmath119 . since @xmath579
, we have @xmath580 , so this implies that @xmath581 for every compact subset @xmath12 of @xmath119 .
also , because @xmath582 , we have @xmath583 , so @xmath584 therefore @xmath585 for every compact subset @xmath12 of @xmath119 .
hence , corollary [ proper<>mu(l ) ] implies that @xmath3 acts properly discontinuously on @xmath586 .
then , because @xmath587 ( see [ d(h)=dimh ] ) , theorem [ noncpctdim]([noncpctdim - tess ] ) implies that @xmath586 has a tessellation .
this contradicts corollary [ 1da->notess ] .
[ tempered - def ] the subgroup @xmath0 is _ tempered _ in @xmath1 if there exists a ( positive ) function @xmath588 ( with respect to a left - invariant haar measure on @xmath0 ) , such that , for every unitary representation @xmath589 of @xmath1 , either * @xmath590 for all @xmath591 and all @xmath118-fixed vectors @xmath592 and @xmath343 ; or * some nonzero vector is fixed by every element of @xmath593 . for many examples of tempered subgroups of simple lie groups ,
see @xcite .
[ margulistempered ] if @xmath0 is noncompact and tempered , then @xmath2 does not have a tessellation .
suppose @xmath3 is a crystallographic group for @xmath2 .
( this will lead to a contradiction . ) to simplify the notation somewhat ( and because this is the only case we need ) , let us assume @xmath239 is a one - parameter subgroup of @xmath1 . because @xmath1 and @xmath3 are unimodular ( recall that @xmath1 is semisimple ( see [ standing ] ) and @xmath3 is discrete )
, there is a @xmath1-invariant measure ( in fact , a @xmath1-invariant volume form ) on the homogeneous space @xmath483 ( * ? ? ?
* lem 1.4 , p. 18 ) .
thus , the natural representation @xmath589 of @xmath1 on @xmath594 , defined by @xmath595 is unitary . because @xmath0 is noncompact , and acts properly on @xmath353
, we know that any compact subset of @xmath353 has infinitely many pairwise - disjoint translates ( all of the same measure ) , so we see that @xmath596 therefore , @xmath589 has no nonzero @xmath1-invariant vectors , so , because @xmath0 is tempered , we know that there is some @xmath597 , such that @xmath598 for all @xmath599 and all @xmath118-invariant @xmath600 . because @xmath15 is compact , there is a compact subset @xmath12 of @xmath1 , such that @xmath601 ; let @xmath602 be the image of @xmath12 in @xmath353 . from the choice of @xmath12 , we know , for each @xmath603 , that there is some @xmath604 , such that @xmath605 because @xmath606 is a compact subset of @xmath353 , there is a positive , continuous function @xmath592 on @xmath353 with compact support , such that @xmath607$,}\ ] ] and , by averaging over @xmath118 , we may assume that @xmath592 is @xmath118-invariant .
fix some large @xmath608 . because @xmath609 is compact , and @xmath353 has infinite volume ( see [ margulistemperedpf - notcpct ] )
, there is some @xmath118-invariant continuous function @xmath610 on @xmath353 , such that @xmath611 , @xmath612 and @xmath613 we have @xmath614 however , because @xmath597 , we know that @xmath615 . this is a contradiction .
we state the following well - known result of representation theory without proof . as is explained in (
* , p. 140 ) , it can be obtained by combining work of r. howe ( * ? ? ?
* cor . 7.2 and 7 ) and m. cowling ( * ? ? ?
( the assumption that @xmath616 can be relaxed : it suffices to assume that @xmath1 is not locally isomorphic to @xmath617 or @xmath618 . ) fix any matrix norm @xmath619 on @xmath1 ; for example , we may let @xmath620 .
[ expdecay ] if @xmath1 is almost simple , and @xmath616 , then there are constants @xmath621 and @xmath622 such that , for every unitary representation @xmath589 of @xmath1 , either 1 .
@xmath623 for all @xmath33 and all @xmath624-fixed vectors @xmath592 and @xmath343 ; or 2 .
some nonzero vector is fixed by every element of @xmath593 .
although we can not prove theorem [ expdecay ] here , we present an elementary proof of the following related result , which , unfortunately , is qualitative , rather than quantitative . on the other hand
, this simple result applies to all vectors , not only the @xmath118-fixed vectors , and it applies to all semisimple groups , including @xmath617 and @xmath618 .
it was first proved by r. howe and c. moore ( * ? ? ?
5.1 ) and ( independently ) r. zimmer ( * ? ? ?
[ matcoeffs->0 ] if * @xmath1 is connected and almost simple ; * @xmath589 is a unitary representation of @xmath1 on a hilbert space @xmath625 , such that no nonzero vector is fixed by @xmath593 ; and * @xmath626 is a sequence of elements of @xmath1 , such that @xmath627 , then @xmath628 , for every @xmath629 .
[ matcoeffs->0-a ] assume @xmath630 . by passing to a subsequence , we may assume @xmath631 converges weakly , to some operator @xmath373 ; that is , @xmath632 let @xmath633 for @xmath634 , we have @xmath635 so @xmath636
. therefore @xmath637 .
we have @xmath638 so the same argument , with @xmath639 in the place of @xmath373 and @xmath640 in the place of @xmath641 , shows that @xmath642 . because @xmath589 is unitary , we know that @xmath631 is normal ( that is , commutes with its adjoint ) for every @xmath643 ; thus , the limit @xmath373 is also normal : we have @xmath644
. therefore @xmath645 so @xmath646
. thus , @xmath647 by passing to a subsequence of @xmath626 , we may assume @xmath648 ( see [ horosubseq ] ) .
then @xmath649 , so @xmath650 .
hence , for all @xmath651 , we have @xmath652 as desired . the general case . from the cartan decomposition @xmath101 , we may write @xmath653 , with @xmath654 and @xmath655 . because @xmath118 is compact , we may assume , by passing to a subsequence , that @xmath656 and @xmath657 converge : say , @xmath658 and @xmath659
. then @xmath660 by case [ matcoeffs->0-a ] .
the following example illustrates lemma [ horosubseq ] .
let @xmath455 , define @xmath94 as in proposition [ wallinsl3 ] , and suppose @xmath626 is some sequence of elements of @xmath94 , such that @xmath661 .
we may write @xmath662 where @xmath663 . by passing to a subsequence
, we may assume that either @xmath664 or @xmath665 .
if @xmath664 , then , in the notation of ( [ horodefn ] ) , we have @xmath666 if @xmath665 , then @xmath162 and @xmath667 are interchanged .
thus , in either case , @xmath337 is the sum of two root spaces of @xmath213 , and @xmath668 is the sum of the two opposite root spaces .
it is not difficult to see that @xmath669 $ ] is the sum of @xmath670 and the remaining two root spaces .
therefore , we have @xmath671 , so @xmath672 .
[ horosubseq ] if @xmath1 and @xmath626 are as in theorem [ matcoeffs->0 ] , and @xmath673 , then , after replacing @xmath626 by a subsequence , we have @xmath674 , where @xmath162 and @xmath667 are defined in ( [ horodefn ] ) . by passing to a subsequence , we may assume @xmath626 is contained in a single weyl chamber , which we may take to be @xmath154 . then , by passing to a subsequence yet again , we may assume , for every positive real root @xmath675 , that either @xmath676 or @xmath677 is bounded .
let * @xmath330 be the set of positive real roots ; * @xmath678 be the set of positive simple real roots ; * @xmath679 ; and * @xmath680 . there is a compact subset @xmath12 of @xmath119 , such that @xmath681 , so , because @xmath661 , we know that @xmath328 is not trivial . for each real root @xmath675 ,
let @xmath682 be the corresponding root subspace of @xmath213 .
then @xmath683 now , for @xmath684 , we have @xmath685 if and only if @xmath675 is in the linear span of @xmath686 .
thus , we see that @xmath337 is precisely the unipotent radical of the standard parabolic subalgebra @xmath687 corresponding to the set @xmath686 of simple roots ( * ? ? ?
* , pp . 8586 ) .
similarly , @xmath668 is the unipotent radical of the opposite parabolic algebra @xmath688 . because @xmath1 is simple ,
the unipotent radicals of opposite parabolics generate @xmath213 ( * ? ? ?
* prop . 4.11 , p. 89 ) , so @xmath689 , as desired .
[ 1d->tempered ] assume @xmath1 is simple , and @xmath616 . if @xmath0 is a one - parameter subgroup of @xmath121 , then either 1 .
@xmath0 is tempered ; or 2 .
@xmath690 .
write @xmath239 . from the real jordan decomposition ( [ jordandecomp ] ) , we may assume , after replacing @xmath0 by a conjugate subgroup , that @xmath691 , where @xmath692 is a hyperbolic one - parameter subgroup , and @xmath693 is a unipotent one - parameter subgroup , such that @xmath694 and @xmath695 commute with each other .
we may assume @xmath696 , so @xmath694 is nontrivial .
since the growth of the hyperbolic one - parameter subgroup @xmath694 is exponential , while that of the unipotent one - parameter subgroup @xmath695 is polynomial , there is some @xmath697 , such that @xmath698 for large @xmath599 . since the function @xmath699 is in @xmath700 , it follows from theorem [ expdecay ] that @xmath0 is tempered , as desired .
[ 1dinn->notess ] if @xmath457 and @xmath690 , then @xmath2 does not have a tessellation .
we have @xmath701 ( see [ d(h)=dimh ] ) , so @xmath0 is a connected , one - dimensional , unipotent subgroup .
hence , the jacobson - morosov lemma ( * ? ? ?
9.7.4 , p. 432 ) implies that there exists a connected , closed subgroup @xmath94 of @xmath1 , such that @xmath94 contains @xmath0 , and @xmath94 is locally isomorphic to @xmath41 .
then @xmath0 is a cartan - decomposition subgroup of @xmath94 ( see [ rrank1-cds ] ) , so there is a compact subset @xmath12 of @xmath119 , such that @xmath702 ( see [ bddchange ] ) . also , we have @xmath703
. therefore , theorem [ noncpct - dim - notess ] applies .
[ 1dsimple ] if @xmath457 and @xmath1 is simple , then @xmath2 does not have a tessellation .
we may assume @xmath91 ( see [ hcanbean ] ) , so @xmath701 ( see [ d(h)=dimh ] ) . * if @xmath704 , then lemma [ 1drank1 ] applies .
* if @xmath690 , then lemma [ 1dinn->notess ] applies . *
if @xmath616 and @xmath696 , then corollary [ 1d->tempered ] implies that @xmath0 is tempered , so theorem [ margulistempered ] applies .
y. benoist ( * ? ? ?
1 ) and g.a . margulis ( unpublished ) proved ( independently ) that @xmath707 does not have a tessellation .
using benoist s method , h. oh and d. witte ( * ? ? ?
* prop . 1.10 ) generalized this result by replacing @xmath41 with any closed , connected subgroup @xmath0 , such that neither @xmath0 nor @xmath708 is compact .
the same argument applies even if @xmath338 replaced with either @xmath462 or @xmath463 .
however , the proof of benoist ( which applies in a more general context ) relies on a somewhat lengthy argument to establish one particular lemma . here
, we adapt benoist s method to obtain a short proof of theorem [ sl3->notess ] that avoids any appeal to the lemma .
[ sl3-oppinv ] assume @xmath461 , for @xmath132 , @xmath462 , or @xmath463 .
* let @xmath569 be the opposition involution in @xmath154 ( see [ oppinv ] ) ; * let @xmath709 . more concretely
, we have @xmath710 [ 1dinsl3 ] if @xmath461 and @xmath457 , then @xmath2 does not have a tessellation .
since @xmath423 , the desired conclusion follows from corollary [ 1drank2 ] ; since @xmath1 is simple , it also follows from corollary [ 1dsimple ] .
however , we give a proof that requires only the special case described in proposition [ wallinsl3 ] , rather than the full strength of ( [ 1drank2 ] ) or ( [ 1dsimple ] ) .
suppose @xmath3 is a crystallographic group for @xmath2 .
( this will lead to a contradiction . )
let @xmath146 and @xmath424 be the two walls of @xmath154 . from proposition [ tess->misswall ] , we know that there exists @xmath425 , such that @xmath711 is finite , for every compact subset @xmath12 of @xmath119 . because @xmath712 , we have @xmath713 . on the other hand , @xmath569 interchanges @xmath146 and @xmath424 .
thus , the preceding paragraph implies that @xmath714 is finite , for every compact subset @xmath12 of @xmath119 . for @xmath94 as in ( [ sl3wallsubgrp ] )
, we have @xmath715 , so the conclusion of the preceding paragraph implies that @xmath3 acts properly discontinuously on @xmath149 ( see [ proper<>mu(l ) ] ) . now theorem [ noncpctdim]([noncpctdim - tess ] ) implies @xmath349 is compact ; thus , @xmath149 has a tessellation .
this contradicts proposition [ wallinsl3 ] . for completeness
, we include the proof of the following simple proposition .
[ sl3-b+ ] assume @xmath461 .
if @xmath0 is a closed , connected subgroup of @xmath121 with @xmath716 , then @xmath717 . since @xmath91 and @xmath718 , it is easy to construct a continuous , proper map @xmath719 \times { \mathord{\mathbb{r}}}^+ \to h$ ] such that @xmath720 , for all @xmath721 ( cf.figure [ constructphifig ] ) .
for example , choose two linearly independent elements @xmath215 and @xmath722 of @xmath270 , and define @xmath723 . ]
if we identify @xmath119 with its lie algebra @xmath724 , then @xmath154 is a convex cone in @xmath724 and the opposition involution @xmath569 is the reflection in @xmath154 across the ray @xmath725 .
thus , for any @xmath570 , the points @xmath156 and @xmath726 are on opposite sides of @xmath725 , so any continuous curve in @xmath154 from @xmath156 to @xmath726 must intersect @xmath725 .
in particular , for each @xmath721 , the curve @xmath727 from @xmath728 to @xmath729 must intersect @xmath725 .
thus , we see , from an elementary continuity argument , that @xmath730 \times { \mathord{\mathbb{r}}}^+ \bigr ) \bigr]$ ] contains @xmath725 .
therefore , @xmath725 is contained in @xmath428 .
: @xmath428 is on both sides of @xmath725 , so it must contain @xmath725 . ] : @xmath427 stays away from @xmath725 , because @xmath717 .
also , half of @xmath427 is on each side of @xmath725 , because @xmath580 .
this contradicts the fact that @xmath3 has only one end . ]
suppose @xmath3 is a crystallographic group for @xmath2 .
( this will lead to a contradiction . )
we may assume @xmath91 ( see [ hcanbean ] ) .
let @xmath318 be a ( symmetric ) finite generating set for @xmath3 , and choose a compact , convex , symmetric subset @xmath12 of @xmath119 so large that @xmath731 for every @xmath37 ( see [ bddchange ] ) . from lemma [ 1dinsl3 ]
, we know that @xmath718 , so proposition [ sl3-b+ ] implies that @xmath717 .
then , because @xmath3 acts properly on @xmath2 , we conclude that @xmath732 is finite ( see [ proper<>mu(l ) ] ) .
since @xmath139 is a proper map , this implies that @xmath733 is finite .
let @xmath363 and @xmath364 be the two components of @xmath734 . because @xmath712 , we know that @xmath713 . then , because @xmath569 interchanges @xmath363 and @xmath364 , we conclude that @xmath735 .
therefore , @xmath736 and @xmath737 have the same cardinality , so they must both be infinite .
so @xmath738 because @xmath3 has only one end ( see [ tess->1end ] ) , this implies there exist @xmath739 such that @xmath740 for some @xmath440 .
then @xmath741 , @xmath742 , and @xmath743 using the fact that @xmath12 is symmetric and the fact that @xmath12 contains the identity element @xmath185 , we conclude that @xmath744 therefore @xmath745 intersects both @xmath363 and @xmath364 .
since @xmath725 separates @xmath363 from @xmath364 , and @xmath12 is connected , this implies that @xmath745 intersects @xmath725 ; hence @xmath746 .
this contradicts the fact that @xmath747 ( see [ sl3->notesspf - gammaf ] ) .
from this point on , we focus almost entirely on @xmath73 and @xmath52 .
( the only exception is that some of the examples constructed in section [ existencesect ] are for other groups . ) in this section , we define the group @xmath750 , which allows us to provide a fairly unified treatment of @xmath73 and @xmath52 in later sections .
* we use @xmath751 to denote either @xmath338 or @xmath462 .
* let @xmath752 , so @xmath753 .
* we use @xmath754 to denote the purely imaginary elements of @xmath751 , so @xmath755 * for @xmath756 , there exist unique @xmath757 and @xmath758 , such that @xmath759 .
( * warning : * in our notation , the imaginary part of @xmath760 is @xmath761 , _ not _ @xmath762 . ) * for @xmath756 , we use @xmath763 to denote the conjugate @xmath764 of @xmath592 .
( if @xmath132 , then @xmath765 . )
* for a row vector @xmath766 , or , more generally , for any matrix @xmath767 with entries in @xmath751 , we use @xmath768 to denote the conjugate - transpose of @xmath767 .
[ sufdefn ] for @xmath769 we define @xmath770 and @xmath771 then : * @xmath772 is a realization of @xmath73 , * @xmath773 is a realization of @xmath52 , and * @xmath774 is the lie algebra of @xmath750 .
we choose * @xmath119 to consist of the diagonal matrices in @xmath750 that have nonnegative real entries , * @xmath120 to consist of the upper - triangular matrices in @xmath750 with only @xmath26 s on the diagonal , and * @xmath775 .
a straightforward matrix calculation shows that the lie algebra of @xmath121 is @xmath776 [ tworows ] from , we see that the first two rows of any element of @xmath777 are sufficient to determine the entire matrix .
in fact , it is also not necessary to specify the last entry of the second row of the matrix .
[ d(su2 ) ] from ( [ d(g ) ] ) and ( [ suf - an ] ) , we see that @xmath778 .
because @xmath120 is simply connected and nilpotent , the exponential map is a diffeomorphism from @xmath331 to @xmath120 ( indeed , its inverse , the logarithm map , is a polynomial ( * ? ? ?
* thm . 8.1.1 , p. 107 )
, so each element of @xmath120 has a unique representation in the form @xmath779 with @xmath780 .
thus , each element @xmath781 of @xmath120 determines corresponding values of @xmath592 , @xmath767 , @xmath782 , @xmath783 , @xmath784 and @xmath785 ( with @xmath786 ) .
we write @xmath787 for these values .
[ simpleroots ] we let @xmath675 and @xmath788 be the simple real roots of @xmath750 , defined by @xmath789 for a ( diagonal ) element @xmath156 of @xmath119 .
thus , the positive real roots ( see figure [ rootspict ] ) are @xmath790 concretely : * the root space @xmath682 is the @xmath592-subspace in @xmath331 , * the root space @xmath791 is the @xmath782-subspace in @xmath331 , * the root space @xmath792 is the @xmath767-subspace in @xmath331 , * the root space @xmath793 is the @xmath783-subspace in @xmath331 , * the root space @xmath794 is the @xmath785-subspace in @xmath331 ( this is 0 if @xmath132 ) , and * the root space @xmath795 is the @xmath784-subspace in @xmath331 ( this is 0 if @xmath132 ) . and ( b ) @xmath796 . ]
let @xmath797 and , for a given lie algebra @xmath798 , @xmath799 note that if @xmath800 for every @xmath801 , then @xmath802 \subset { { { \mathfrak{\lowercase{d}}}}_{{\mathfrak{\lowercase{h}}}}}$ ] and @xmath803 is contained in the center of @xmath270 ( cf .
[ [ u , v ] ] ) . by definition ( see [ a+defn ] ) , we have @xmath804 therefore , from the definition of @xmath675 and @xmath788 ( see [ simpleroots ] ) , we see that @xmath805 for @xmath806 , the division algebra of real quaternions , the group @xmath807 is a realization of @xmath808 .
most of the work in this paper carries over , but the upper bound on @xmath137 given in theorem [ maxnolinear ] is not sharp in this case ( and it does not seem to be easy to improve this result to obtain a sharp bound ) .
thus , we have not obtained any interesting conclusions about the nonexistence of tessellations of homogeneous spaces of @xmath808 .
we now describe how the four important families of homogeneous spaces of example [ kulkarnieg ] are realized in terms of @xmath750 .
[ su1ndefn ] let * @xmath811 ; * @xmath812 ; * @xmath813 ; and * @xmath814
. then , for an appropriate choice of the embeddings in example [ kulkarnieg ] , we have @xmath815 and ( if @xmath816 ) we have @xmath817 [ d(sp ) ] from ( [ d(g ) ] ) , ( [ su1nan ] ) and ( [ sp1man ] ) , we see that * @xmath818 and * @xmath819 .
the arguments in later sections often require the calculation of @xmath820 , for some @xmath821 , or of @xmath822 $ ] , for some @xmath823 .
we now provide these calculations for the reader s convenience .
for @xmath824 we have @xmath825 when @xmath826 , this simplifies to : @xmath827 similarly , when @xmath828 , we have @xmath829 for @xmath830 we have @xmath831 } [ u , \tilde u ] = \begin{pmatrix } 0 & 0 & \phi \tilde y - \tilde \phi y & - x \tilde y^{\dagger } + \tilde x y^{\dagger } + \phi \tilde { { \mathord{\mathsf{y}}}}- \tilde \phi { { \mathord{\mathsf{y } } } } & -2 { \operatorname{im}}(x \tilde x^{\dagger } + \phi { \overline{\tilde \eta } } - \tilde \phi { \overline{\eta } } ) \\ & 0 & 0 & -2 { \operatorname{im}}(y \tilde y^{\dagger } ) & \tilde y x^{\dagger } - y \tilde x^{\dagger } + \tilde { { \mathord{\mathsf{y}}}}{\overline{\phi } } - { { \mathord{\mathsf{y}}}}{\overline{\tilde \phi } } \\ & & & \cdots \\ \end{pmatrix } .\ ] ] [ conjugation ] for @xmath832 , we have @xmath833 + \frac{1}{2 } \bigl [ [ u , v ] , v \bigr ] + \frac{1}{3 ! } \bigl [ \bigl [ [ u , v ] , v \bigr ] , v \bigr ] + \cdots .\ ] ] combining this with allows us to calculate the effect of conjugating by an element of @xmath120 .
for example , suppose @xmath821 , with @xmath800 and @xmath834 , and suppose @xmath835 .
we see , from ( [ [ u , v ] ] ) , that @xmath836 } = 0 $ ] and that @xmath837 } = y_{[u , v ] } = 0 $ ] , so @xmath838 , v \bigr ] = 0 $ ] ( see [ [ u , v ] ] )
. therefore @xmath839 .\ ] ]
y. benoist ( * ? ? ?
* lem . 2.4 ) showed that calculating values of the cartan projection @xmath139 is no more difficult than calculating the norm of a matrix ( see [ mucalc ] ) . in this section
, we describe this elegant method and some of its consequences , in the special case @xmath840 . throughout this section ,
we assume @xmath840 .
we employ the usual big oh and little oh notation : for functions @xmath841 on a subset @xmath507 of @xmath1 , we say @xmath842 if there is a constant @xmath12 , such that , for all @xmath843 with @xmath844 large , we have @xmath845 .
( the values of each @xmath846 are assumed to belong to some finite - dimensional normed vector space , typically either @xmath462 or a space of complex matrices .
which particular norm is used does not matter , because all norms are equivalent up to a bounded factor . )
we say @xmath847 if @xmath848 as @xmath182 .
( we use @xmath182 to mean @xmath849 . ) also , we write @xmath850 if @xmath851 and @xmath852 .
we use the following norm on @xmath750 , because it is easy to calculate .
the reader is free to make a different choice , at the expense of changing @xmath853 to @xmath854 in a few of the calculations . for @xmath855 , we define @xmath856 to be the maximum absolute value among the matrix entries of @xmath781 .
that is , @xmath857 [ rhodefn ] define @xmath858 by @xmath859 , so @xmath860 is the second exterior power of the standard representation of @xmath750 .
thus , we may define @xmath861 to be the maximum absolute value among the determinants of all the @xmath862 submatrices of the matrix @xmath781 .
that is , @xmath863 from ( [ exp ] ) , ( [ exp(phi=0 ) ] ) , and ( [ exp(y=0 ) ] ) , it is clear that the @xmath864 minor in the top right corner is often larger than the other @xmath862 minors , so we give it a special name .
[ deltadefn ] for @xmath865 , define @xmath866 [ calcrho ] for @xmath867 , we have @xmath868 and @xmath869 . from ( [ suf - an ] ) , we see that @xmath870 thus , from ( [ a+ ] ) , we see that @xmath871 for @xmath872 ( and , since @xmath156 is diagonal , we have @xmath873 for @xmath874 ) . therefore , the desired conclusions follow from the definitions of @xmath875 and @xmath876 .
[ mu = rho ] we have @xmath877 for @xmath855 .
choose @xmath878 , such that @xmath879 .
because @xmath880 for @xmath881 , and @xmath882 ( since @xmath118 is compact ) , we have @xmath883 and @xmath884 so ( [ h = mu(h ) ] ) holds .
similarly , we have @xmath885 so ( [ rho(h)=rho(mu ) ] ) holds . for @xmath867 , we know , from ( [ calcrho ] ) , that @xmath886 and @xmath887 . thus , letting @xmath888 , and using ( [ h = mu(h ) ] ) and ( [ rho(h)=rho(mu ) ] )
, we see that @xmath889 and @xmath890 as desired .
proposition [ mu = rho ] generalizes to any reductive group @xmath1 ( * ? ? ?
however , one may need to use a different representation in the place of @xmath860 .
in fact , if @xmath891 , then @xmath497 representations of @xmath1 are needed ; for @xmath840 , we have @xmath423 , and the two representations we use are @xmath860 and the identity representation @xmath892 .
[ gn = hn ] let @xmath893 and @xmath894 be two sequences of elements of @xmath750 .
we have @xmath895 if and only if @xmath896 ( @xmath168 ) let @xmath897 . from ( [ mucalc ] )
, we see that @xmath898 for @xmath899 , so , using ( [ ajj ] ) , we have @xmath900 therefore @xmath901 , as desired .
( @xmath173 ) because @xmath12 is compact , we have @xmath902 ( cf . proof of ( [ h = mu(h ) ] ) and ( [ rho(h)=rho(mu ) ] ) )
. then the desired conclusions follow from ( [ h = mu(h ) ] ) and ( [ rho(h)=rho(mu ) ] ) . because @xmath12 is compact
, we have @xmath903 and @xmath904 for any @xmath905 ( cf
. proof of ( [ h = mu(h ) ] ) and ( [ rho(h)=rho(mu ) ] ) ) .
thus , the desired conclusion follows from corollary [ gn = hn ] .
because of proposition [ mu = rho ] , we will often need to calculate @xmath844 and @xmath906 . the following observation and
its corollary sometimes simplifies the work , by allowing us to replace @xmath781 with @xmath907 .
[ mu(h-1 ) ] we have @xmath575 for @xmath855 .
define @xmath908 as in ( [ sufdefn ] ) , and choose @xmath878 , such that @xmath879 . for any @xmath867 , we see , using ( [ suf - an ] ) or ( [ ajj ] ) , that @xmath909 , so @xmath910 note that @xmath911 . also , we have @xmath912 and @xmath913 , so it is obvious that @xmath914 and @xmath915 .
therefore @xmath916 thus , from the definition of @xmath139 , we conclude that @xmath575 , as desired .
the following corollary is obtained by combining lemma [ mu(h-1 ) ] with corollary [ gn = hn ] .
[ rho(h-1 ) ] we have @xmath917 and @xmath918 for @xmath855 . for @xmath425 ,
set @xmath919 from ( [ a+ ] ) , we see that @xmath146 and @xmath424 are the two walls of @xmath154 . from ( [ calcrho ] )
, we have @xmath920 we reproduce the proof of the following result , because it is both short and instructive .
( although we have no need for it here , let us point out that the converse of this proposition also holds , and that there is no need to assume @xmath91 . )
because of this proposition ( and corollary [ cds->notess ] ) , section [ suflargesect ] will study the existence of curves @xmath241 , such that @xmath921 , for @xmath922 .
[ cds<>h_m ] let @xmath0 be a closed , connected subgroup of @xmath121 in @xmath750 .
if , for each @xmath425 , there is a continuous curve @xmath241 in @xmath0 , such that @xmath923 as @xmath924 , then @xmath0 is a cartan - decomposition subgroup . :
if @xmath428 contains a curve near each wall of @xmath154 , then it also contains the interior . ] by hypothesis , there is a continuous , proper map @xmath925 , such that @xmath926 . because @xmath91 , we know that @xmath0 is homeomorphic to some euclidean space @xmath927 ( see [ solvable]([solvable - h = rn ] ) )
suppose , for the moment , that @xmath43 .
( this will lead to a contradiction . )
we know that @xmath928 for @xmath929 . because @xmath917 and @xmath918 ( see [ rho(h-1 ) ] )
, we must also have @xmath928 for @xmath930 .
there is no harm in assuming @xmath931 ; then @xmath932 ( because @xmath43 ) , so we conclude that @xmath928 for all @xmath591 .
this contradicts the fact that @xmath933 for @xmath934 .
we may now assume @xmath718 .
then , because @xmath0 is homeomorphic to @xmath927 , it is easy to extend @xmath935 to a continuous and proper map @xmath936 \times { \mathord{\mathbb{r}}}^+ \to h$ ] . from and ( [ gn = hn ] ) , we know that the curve @xmath937 stays within a bounded distance from the wall @xmath429 ; say @xmath938 < c$ ] for all @xmath939 .
we may assume @xmath12 is large enough that @xmath940 for all @xmath941 $ ] .
then an elementary homotopy argument shows that @xmath942 \times { \mathord{\mathbb{r}}}^+ \bigr ) \bigr]$ ] contains @xmath943 so @xmath944 \times { \mathord{\mathbb{r}}}^+ \bigr ) \bigr ] \approx a^+$ ] . because @xmath945 \times { \mathord{\mathbb{r}}}^+ \bigr ) \bigr]$ ] , we conclude from theorem [ cdsvsmu ] that @xmath0 is a cartan - decomposition subgroup
when @xmath28 , the weyl chamber @xmath154 has only one point at infinity .
thus , if @xmath0 is any noncompact subgroup , then the closure of @xmath428 must contain this point at infinity .
this is why it is easy to prove that any noncompact subgroup of @xmath1 is a cartan - decomposition subgroup ( see [ rrank1-cds ] ) .
the idea of proposition [ cds<>h_m ] is that if @xmath423 , then the points at @xmath390 of the weyl chamber @xmath154 form a closed interval .
if the closure of @xmath428 contains the two endpoints of this interval , then , by continuity , it must also contain all the points in between . unfortunately , we have no good substitute for this proposition when @xmath946 . the points at @xmath390 of
@xmath154 form a closed disk ( topologically speaking ) .
it is easy to define a map @xmath947 from one disk to another , such that the image of @xmath947 contains the entire boundary sphere , but does not contain the interior of the disk .
thus , it does not suffice to show only that the closure of @xmath428 contains the boundary of the disk at @xmath390 ; rather , one needs additional homotopical information to guarantee that no interior points are missed .
[ mu(suorsp ) ] let @xmath948 , and fix some @xmath949 .
then @xmath950 and @xmath951 are the two walls of @xmath154 .
we have 1 .
[ mu(suorsp)-su ] @xmath928 for @xmath952 ; and 2 .
[ mu(suorsp)-sp ] @xmath933 for @xmath953 .
let @xmath954 or @xmath810 .
then @xmath955 is a maximal compact subgroup of @xmath0 . from the cartan decomposition @xmath956 and the definition of @xmath139
, we conclude that @xmath957 . in the notation of ,
we see ( from definition [ su1ndefn ] ) that @xmath958 , where @xmath959 then , since @xmath960 ( see [ mu(h-1 ) ] ) and @xmath961 for @xmath962 , we conclude that @xmath963 is a wall of @xmath154 .
furthermore , we have @xmath964 for @xmath965 ( see [ rho(l ) ] ) , so @xmath966 for @xmath591 ( see [ mu = rho ] ) . [ su1inh ] if there is a continuous curve @xmath967 in @xmath0 , such that @xmath968 , then there is a compact subset @xmath12 of @xmath1 , such that @xmath969 .
for any ( large ) @xmath970 , we see from continuity ( more precisely , from the intermediate value theorem ) that there exists @xmath721 , such that @xmath971 then , by assumption and from [ mu(suorsp)]([mu(suorsp)-su ] ) , we have @xmath972 so there is a compact subset @xmath160 of @xmath119 , such that @xmath973 ( see [ gn = hn ] ) . therefore @xmath974 as desired .
the following corollary can be proved by a similar argument .
( recall that the equivalence relation @xmath100 is defined in ( [ simdefn ] ) . )
[ hsimsuorsp ] assume @xmath0 is not compact . 1
. [ hsimsuorsp - su ] we have @xmath975 if and only if @xmath928 for @xmath591 .
[ hsimsuorsp - sp ] we have @xmath976 if and only if @xmath977 for @xmath591 .
because of proposition [ cds<>h_m ] , we will often want to show that a curve @xmath241 satisfies @xmath978 , for some @xmath577 .
the following lemma does half of the work .
[ owalls ] let @xmath507 be a subset of @xmath750 . 1
. [ owalls - linear ] if @xmath979 for @xmath843 , then @xmath928 for @xmath843 .
[ owalls - square ] if @xmath980 for @xmath843 , then @xmath981 for @xmath843 . from ( [ calcrho ] ) and ( [ a+ ] )
, we have @xmath982 and @xmath983 for @xmath867 .
thus , letting @xmath888 , and using ( [ h = mu(h ) ] ) and ( [ rho(h)=rho(mu ) ] ) , we have : @xmath984 and @xmath985 so @xmath986 and @xmath987 .
the desired conclusions follow . for convenience , we record the following simple observation .
( for the proof , cf . the proof of ( [ h = mu(h ) ] ) and ( [ rho(h)=rho(mu ) ] ) .
) [ conjcurve ] let * @xmath425 , * @xmath33 , and * @xmath967 be a continuous curve in @xmath0 . if @xmath988 , then @xmath989 .
the following well - known , elementary observation is used frequently in the later sections . [ o(linear ) ] let @xmath990 be a subspace of a finite - dimensional real vector space @xmath163 , and let @xmath991 and @xmath992 be linear transformations . 1
. [ o(linear)-o ] if @xmath993 ( or , more generally , if @xmath994 ) , then there is a linear transformation @xmath995 , such that @xmath996 for all @xmath997 .
therefore @xmath852 on @xmath990 .
[ o(linear)-= ] if @xmath998 , then @xmath999 on @xmath990 .
( [ o(linear)-o ] ) by passing to a subspace , we may assume @xmath1000 . then , by modding out @xmath1001 , we may assume @xmath1002 is an isomorphism onto its image .
define @xmath1003 by @xmath1004 , and let @xmath995 be any extension of @xmath1005 . for @xmath997 , we have @xmath1006 so @xmath852 .
( [ o(linear)-= ] ) from ( [ o(linear)-o ] ) , we have @xmath852 and @xmath1007 , so @xmath999 .
let @xmath270 be a real lie subalgebra of @xmath329 , and assume there does not exist a nonzero element @xmath215 of @xmath270 , such that @xmath1008 and @xmath834 .
then there exist @xmath338-linear transformations @xmath1009 , such that @xmath1010 for all @xmath801 .
( similarly , @xmath1011 , @xmath1012 , and @xmath1013 are also functions of @xmath1014 . ) furthermore , we have @xmath1015 . the following well - known result is a generalization of the fact that all norms on a finite - dimensional vector space are equivalent up to a bounded factor .
[ quadrforms ] if @xmath163 is any finite - dimensional real vector space , and @xmath1016 are two continuous , homogeneous functions of the same degree , such that @xmath1017 , then @xmath999 . by continuity
, the function @xmath1018 attains a non - zero minimum and a finite maximum on the unit sphere . because @xmath1018 is homogeneous of degree zero , these values bound @xmath1018 on all of @xmath1019 .
in this section , we show how to construct several families of homogeneous spaces that have tessellations .
all of these examples are based on a method of t. kobayashi ( see [ construct - tess ] ) that generalizes example [ kulkarnieg ] . as explained in the comments before theorem
[ noncpctdim ] , the following theorem is essentially due to t. kobayashi . [ construct - tess ] if * @xmath0 and @xmath46 are closed subgroups of @xmath1 , with only finitely many connected components ; * @xmath46 acts properly on @xmath2 ; * @xmath1020 ; and * there is a cocompact lattice @xmath3 in @xmath46 , then @xmath2 has a tessellation .
( namely , @xmath3 is a crystallographic group for @xmath2 . )
because @xmath3 is a closed subgroup of @xmath46 , we know that it acts properly on @xmath2 ( see [ chcproper ] ) . thus , it suffices to show that @xmath116 is compact . from lemma [ hcanbean ]
, we see that there is no harm in assuming @xmath1021 , and that there is a closed , connected subgroup @xmath1022 of @xmath1 , such that * @xmath1022 is conjugate to a subgroup of @xmath121 , * @xmath1023 , and * @xmath1024 , for some compact subset @xmath12 of @xmath1 .
( unfortunately , we can not assume @xmath1025 : we may not be able to replace @xmath46 with @xmath1022 , because there may not be a cocompact lattice in @xmath1022 .
for example , there is not lattice in @xmath121 , because any group with a lattice must be unimodular ( * ? ? ? * rem .
1.9 , p. 21 ) . )
it suffices to show that @xmath1026 is compact .
( because @xmath1027 is compact , and @xmath1028 is compact , this implies that @xmath15 is compact , as desired . )
we know that @xmath1022 acts properly on @xmath2 ( see [ chcproper ] ) , so @xmath1029 acts properly on @xmath1 , with quotient @xmath1026 .
therefore , lemma [ fiberbundle ] implies that @xmath1026 has the same homology as @xmath1 ; in particular , @xmath1030 from the iwasawa decomposition @xmath92 , and because @xmath121 is homeomorphic to @xmath1031 ( see [ ansc ] and [ r = rd ] ) , we know that @xmath1 is homeomorphic to @xmath1032 .
since @xmath1031 is contractible , this implies that @xmath1 is homotopy equivalent to @xmath118 , so @xmath1 and @xmath118 have the same homology ; in particular , @xmath1033 since @xmath1034 this implies that the top - dimensional homology of the manifold @xmath1035 is nontrivial .
therefore @xmath1026 is compact ( * ? ? ?
8.3.4 ) , as desired .
our results for @xmath1036 are based on the following special case of the theorem .
the converse of this corollary is proved in section [ proofsect ] ( see [ suf - known ] ) .
recall the equivalence relation @xmath100 , introduced in notation [ simdefn ] .
[ sueventessexists ] let @xmath0 be a closed , connected subgroup of @xmath1036 .
if * @xmath1037 ; and * either @xmath1038 or @xmath976 , then @xmath2 has a tessellation .
let @xmath1039 and @xmath1040 . by assumption
, we have @xmath1041 , for some @xmath1042 ; let @xmath1043 . because @xmath1044 and @xmath1045 are the two walls of @xmath154 ( see [ mu(suorsp ) ] ) , we know that @xmath1046 acts properly on @xmath1047 ( see [ proper<>mu(l ) ] ) ; since @xmath1048 , this implies that @xmath46 acts properly on @xmath2 ( see [ chcproper ] ) . also , we have @xmath1049 ( see [ d(sp ) ] and [ d(g ) ] ) , and there is a cocompact lattice in @xmath46 ( cf .
[ classical]([classical - borel ] ) ) . thus , the desired conclusion follows from theorem [ construct - tess ] .
the homogeneous spaces described here were found by h. oh and d. witte ( * ? ? ?
* thms . 4.1 and 4.6 ) , ( * ? ? ?
* thm . 1.5 ) .
[ hb - defn ] for any @xmath338-linear @xmath1050 , we define @xmath1051 we write @xmath1052 , rather than @xmath1053 , because @xmath767 is a row vector .
it is easy to see , using , for instance , the formula for the bracket in , that if @xmath1054 then @xmath1055 is a real lie subalgebra of @xmath1056 ; we let @xmath1057 denote the corresponding connected lie subgroup of @xmath121 . from ( [ d(h)=dimh ] ) , we have @xmath1058 [ hb = sp1 m ] assume @xmath1059 . by comparing ( [ sp1man ] ) with ( [ hb - defn ] )
, we see that there is a @xmath338-linear map @xmath1060 , such that @xmath1061 ( and @xmath1062 satisfies ( [ vbwb ] ) ) .
thus , in general , @xmath1057 is a deformation of @xmath1063 .
[ hbthm ] let @xmath1064 be @xmath338-linear .
if * condition ( [ vbwb ] ) holds , and * [ hbthm - xbnotinfx ] @xmath1065 , for every nonzero @xmath1066 , then 1 .
[ hbthm - mu ] @xmath933 for @xmath1067 ; and 2 .
[ hbthm - tess ] @xmath1068 has a tessellation .
( [ hbthm - mu ] ) given @xmath1067 , write @xmath1069 , with @xmath1070 and @xmath1071 .
we may assume that @xmath1072 ( by replacing @xmath781 with @xmath907 if necessary ( see [ rho(h-1 ) ] ) ) .
it suffices to show @xmath1073 ( for then @xmath1074 , so lemma [ owalls]([owalls - square ] ) applies ) .
[ hb - hinn ] assume @xmath156 is trivial . from and ( [ hb - defn ] )
, we see that @xmath1075 so @xmath1076 from ( [ exp(phi=0 ) ] ) and ( [ deltadefn ] ) , we have @xmath1077 from ( [ xbnotinfx ] ) , we see that @xmath1078 for every nonzero @xmath1066 , so lemma [ quadrforms ] implies @xmath1079 also , because @xmath1080 ( and @xmath1081 ) , we have @xmath1082 thus ,
@xmath1083 so @xmath1084 , as desired .
the general case . from case [ hb - hinn ] , we know @xmath1085 .
then , because @xmath1086 , we have @xmath1087 then , since @xmath1088 and @xmath1089 , we conclude that @xmath1090 , as desired .
( [ hbthm - tess ] ) from ( [ hbthm - mu ] ) and [ hsimsuorsp]([hsimsuorsp - sp ] ) , we see that @xmath1091 .
then , because @xmath1092 ( see [ d(hb ) ] ) , theorem [ sueventessexists ] implies that @xmath1093 has a tessellation . [ bsymplectic ] let @xmath1050 be @xmath338-linear .
condition ( [ vbwb ] ) holds if and only if either 1 .
@xmath132 ; or 2 .
@xmath136 and @xmath1094 , where @xmath1095 and we use the natural identification of @xmath1096 with @xmath1097 .
assume @xmath132 .
because @xmath1098 for every @xmath1099 , it is obvious that ( [ vbwb ] ) holds .
assume @xmath136 .
if ( [ vbwb ] ) holds , then
@xmath1100 so @xmath1101 is symplectic .
the argument is reversible .
[ xbnotinfx ] * for @xmath132 , the assumption that @xmath1065 simply requires that @xmath316 have no real eigenvalues . * for @xmath136 , we do not know a good description of the linear transformations @xmath316 that satisfy @xmath1065 , although it is easy to see that this is an open set ( and not dense ) .
a family of examples was constructed by h. oh and d. witte ( see [ hbeg ] below ) . *
if @xmath79 is odd , then there does not exist @xmath1102 satisfying the assumption that @xmath1065 . for @xmath132 ,
this is simply the elementary fact that a linear transformation on an odd - dimensional real vector space must have a real eigenvalue . for @xmath1103 , see step [ dim(u)<2npf - u / z ] of the proof of proposition [ dim(u)<2n ] . *
if @xmath79 is even , then , by varying @xmath316 , one can obtain uncountably many pairwise non - conjugate subgroups @xmath1057 , such that @xmath1104 has a tessellation . for @xmath132 , this is proved in ( * ? ? ?
* thm . 1.5 ) ) . for @xmath136 ,
a similar argument can be applied to the examples constructed in ( [ hbeg ] ) below .
[ hbeg ] assume @xmath79 is even , let @xmath1105 , such that @xmath1101 has no real eigenvalue , and define an @xmath338-linear map @xmath1106 by @xmath1107 .
let us verify that @xmath316 satisfies the conditions of theorem [ hbthm ] ( for @xmath136 ) .
let @xmath1108 .
from the definition of @xmath316 , and because @xmath1105 , we have @xmath1109 suppose @xmath1110 , for some @xmath1111 . because @xmath1112 , we must have @xmath1113 . then @xmath1114 because @xmath1101 has no real eigenvalues , we know that @xmath26 is not an eigenvalue of @xmath1101
, so we conclude that @xmath1115 .
similarly , because @xmath1116 is not an eigenvalue of @xmath1101 , we see that @xmath1117 .
therefore @xmath1118 these examples are new for @xmath136 , but provide nothing interesting for @xmath132 ( see [ hc]([hc - real ] ) ) .
[ suegsdefn ] for @xmath1119 $ ] , we define @xmath1120}}}= { \left\{\ , \begin{pmatrix } t & \phi & x & { \operatorname{re}}\phi + c { \operatorname{im}}\phi & { { \mathord{\mathsf{x}}}}\\ & 0 & 0 & 0 & * \\ & & \dots \\ \end{pmatrix } \mathrel{\left| \vphantom { \left\ { \begin{pmatrix } t & \phi & x & { \operatorname{re}}\phi + c { \operatorname{im}}\phi & { { \mathord{\mathsf{x}}}}\\ & 0 & 0 & 0 & * \\ & & \dots \\ \end{pmatrix } \mid \begin{matrix } t \in { \mathord{\mathbb{r } } } , \\ \phi \in { \mathbb{f } } , \\ x \in { \mathbb{f}}^{n-2 } , \\ { { \mathord{\mathsf{x}}}}\in{{\mathbb{f}}_{\text{imag}}}\end{matrix } \right\ } } \right . } \begin{matrix } t \in { \mathord{\mathbb{r } } } , \\ \phi \in { \mathbb{f } } , \\ x \in { \mathbb{f}}^{n-2 } , \\ { { \mathord{\mathsf{x}}}}\in{{\mathbb{f}}_{\text{imag}}}\end{matrix } \,\right\ } } .\ ] ] it is easy to see , using , for instance , the formula for the bracket in , that @xmath1121}}}$ ] is a real lie subalgebra of @xmath1056 ( even without the assumption that @xmath1122 ) ; we let @xmath1123}}}$ ] be the corresponding connected lie subgroup of @xmath121 . from ( [ d(h)=dimh ] ) , we have @xmath1124 } } } ) & = \dim { { { { \mathfrak{\lowercase{h}}}}_{[c]}}}\\ & = \dim { \mathord{\mathbb{r}}}+ \dim { \mathbb{f}}+ \dim { \mathbb{f}}^{n-2 } + \dim { { \mathbb{f}}_{\text{imag}}}\\ & = 1 + { q}+ { q}(n-2 ) + ( { q}-1 ) \\ & = { q}n . \end{split}\ ] ] [ hc ] let @xmath1125 be embedded into @xmath774 as in [ su1nan ] . 1 . [ hc - real ] if @xmath1126 , then @xmath216 is irrelevant in the definition of @xmath1121}}}$ ] ( because @xmath1127 ) ; therefore @xmath1121}}}={\operatorname{{\mathfrak{\lowercase{su}}}}}(1,n;{\mathord{\mathbb{r}}})\cap ( { \mathfrak{\lowercase{a } } } + { \mathfrak{\lowercase{n}}})$ ] .
2 . if @xmath1128 , then @xmath1129 } } = { \operatorname{{\mathfrak{\lowercase{su}}}}}(1,n;{\mathord{\mathbb{c}}})\cap ( { \mathfrak{\lowercase{a } } } + { \mathfrak{\lowercase{n}}})$ ] . thus , in general , @xmath1121}}}$ ] is either @xmath1130 or a deformation of it .
[ suegs ] assume @xmath136 , and @xmath1059 is even . if @xmath1119 $ ] , then 1 .
[ suegs - linear ] @xmath928 for @xmath1131}}}$ ] ; and 2 .
[ suegs - tess ] @xmath1132}}}$ ] has a tessellation .
( [ suegs - linear ] ) given @xmath1131}}}$ ] , it suffices to show that @xmath979 ( see [ owalls]([owalls - linear ] ) ) .
write @xmath1069 , with @xmath1070 and @xmath1133 .
we may assume that @xmath1072 ( by replacing @xmath781 with @xmath907 if necessary ( see [ rho(h-1 ) ] ) ) .
let @xmath1134 be the real quadratic form @xmath1135 and let @xmath163 be the @xmath338-subspace of @xmath1136 defined by @xmath1137 [ suegspf - q = x2+phi2 ] for @xmath1138 , we have @xmath1139 . for @xmath1140 , we have @xmath1141 ( because @xmath1142 and @xmath1143 is purely imaginary ) . thus , the restriction of @xmath1144 to @xmath163 is positive definite , so the desired conclusion follows from lemma [ quadrforms ] .
[ suegspf - u = x2+phi2+xx ] we have @xmath1145 . from ( [ exp(y=0 ) ] ) ( with @xmath1146 ) , we have @xmath1147 and @xmath1148 then , from step [ suegspf - q = x2+phi2 ] , we see that @xmath1149 as desired .
completion of the proof . from step [ suegspf - u = x2+phi2+xx
] , we have @xmath1150 also , from ( [ exp(y=0 ) ] ) , we have @xmath1151 thus , it is easy to see that @xmath1152 so the desired conclusion follows from lemma [ owalls]([owalls - linear ] ) .
( [ suegs - tess ] ) from ( [ suegs - linear ] ) and [ hsimsuorsp]([hsimsuorsp - su ] ) , we see that @xmath1123}}}\sim { \operatorname{su}}(1,n)$ ] .
then , because @xmath1153 } } } ) = 2n$ ] ( see [ d(hc ) ] ) , theorem [ sueventessexists ] implies that @xmath1132}}}$ ] has a tessellation .
proposition [ hcuncountable ] shows that if @xmath136 , then @xmath1123}}}$ ] is not conjugate to @xmath1154}}$ ] unless @xmath1155 ( for @xmath1156 $ ] ) .
thus , theorem [ suegs]([suegs - tess ] ) implies that , by varying @xmath216 , one obtains uncountably many nonconjugate subgroups @xmath1123}}}$ ] , such that @xmath1157}}}$ ] has a tessellation .
[ g1xg2-tess ] let @xmath1158 be the direct product of two connected , linear , almost simple lie groups @xmath1159 and @xmath1160 of real rank one , with finite center , and let @xmath0 be a nontrivial , closed , connected , proper subgroup of @xmath121 .
the homogeneous space @xmath2 has a tessellation if and only if , perhaps after interchanging @xmath1159 and @xmath1160 , there is a continuous homomorphism @xmath1161 , such that @xmath1162 ( @xmath168 ) we may assume @xmath1163 ( by interchanging @xmath1159 and @xmath1160 if necessary ) .
[ hcapboth ] assume @xmath1164 and @xmath1165 . for @xmath1166 , we know that @xmath1167 is not compact ( see [ solvable]([solvable - nocpct ] ) )
, so corollary [ rrank1-cds ] implies that there is a compact subset @xmath1168 of @xmath1169 , such that @xmath1170 .
then , letting @xmath1171 , we have @xmath151 , so proposition [ cds->notess ] implies that @xmath2 does not have a tessellation .
this is a contradiction .
assume @xmath1164 and @xmath1172 . from corollaries [ rrank1-cds ] and [ cdsvsmu ] , we know that there is a compact subset @xmath12 of @xmath1173 , such that @xmath1174
. therefore , corollary [ noncpct - dim - notess ] ( with @xmath1159 in the place of @xmath94 ) implies @xmath1175 .
then , because @xmath1172 ( and @xmath91 ) , we conclude that @xmath0 is the graph of a homomorphism from @xmath1176 to @xmath1177 , as desired .
assume @xmath1178 . from corollary [ tess->dim>1,2 ] , we know that @xmath1179 . then , since @xmath1178 , we conclude that @xmath0 is the graph of a homomorphism from @xmath1177 to @xmath1176 . interchanging @xmath1159 and @xmath1160 yields the desired conclusion .
( @xmath173 ) we verify the hypotheses of theorem [ construct - tess ] , with @xmath1160 in the role of @xmath46 .
let @xmath1180 be the image of @xmath0 under the natural homomorphism @xmath1181 . because @xmath91 , we know that @xmath1180 is closed ( see [ solvable]([solvable - h = rn ] ) ) .
it is well known ( and follows easily from ( [ proper<>chc ] ) ) that any closed subgroup acts properly on the ambient group , so this implies that @xmath1180 acts properly on @xmath1182 . from the definition of @xmath0
, we have @xmath1172 , so we conclude that @xmath1183 acts properly on @xmath1182 ; equivalently , @xmath1160 acts properly on @xmath2 ( cf . [ proper<>chc ] ) . because @xmath1184 , we have @xmath1185 . also , we have @xmath93 ( see [ d(h)=dimh ] ) and , from the definition of @xmath0 , we have @xmath1186 .
therefore @xmath1187 there is a cocompact lattice in @xmath1160 ( cf .
[ classical]([classical - borel ] ) ) .
so theorem [ construct - tess ] implies that @xmath2 has a tessellation .
t. kobayashi observed that , besides the examples with @xmath1188 or @xmath6 ( see [ kulkarnieg ] ) , theorem [ construct - tess ] can also be used to construct tessellations of some homogeneous spaces @xmath2 in which @xmath1 and @xmath0 are simple lie groups with @xmath1189 .
he found one pair of infinite families , and several isolated examples .
[ kobayashibigeg ] each of the following homogeneous spaces has a tessellation : 1 .
[ kobayashibigeg - so4/sp1 ] @xmath1190 ; 2 .
@xmath1191 ; 3 .
@xmath1192 ; 4 .
@xmath1193 ; 5 .
@xmath1194 ; 6 .
@xmath1195 ; 7 . @xmath1196 ; 8 .
. it would be very interesting to find other examples of simple lie groups @xmath1 with reductive subgroups @xmath0 and @xmath46 that satisfy the hypotheses of theorem [ construct - tess ] .
let @xmath1198 and @xmath1199 . from [ kobayashibigeg]([kobayashibigeg - so4/sp1 ] ) , we know that @xmath2 has a tessellation
. h. oh and d. witte ( * ? ? ?
4.6(2 ) ) pointed out that the deformations @xmath1200 ( where @xmath1057 is as in theorem [ hbthm ] , with @xmath136 ) also have tessellations , but it is not known whether there are other deformations of @xmath123 that also have tessellations .
it does not seem to be known whether the other examples in theorem [ kobayashibigeg ] lead to nontrivial deformations , after intersecting @xmath0 with @xmath121 .
this section presents a short proof of the results we need from @xcite and @xcite .
those papers provide an approximate calculation of @xmath428 , for every closed , connected subgroup @xmath0 of @xmath73 or @xmath52 , respectively , but here we consider only subgroups of large dimension .
also , we do not need a complete description of the entire set @xmath428 ; we are only interested in whether or not there is a curve @xmath241 , such that @xmath988 , for some @xmath577 .
the main results of this section are theorem [ maxnolinear ] ( for @xmath1201 ) and theorem [ bestnosquare ] ( for @xmath1202 ) .
they give a sharp upper bound on @xmath422 , for subgroups @xmath0 that fail to contain such a curve , and , if @xmath79 is even , also provide a fairly explicit description of all the subgroups of @xmath121 that attain the bound . because of the limited scope of this section , the proof here is shorter than the previous work , and we are able to give a fairly unified treatment of the two groups @xmath73 and @xmath52 .
the arguments are elementary , but they involve case - by - case analysis and a lot of details , so they are not pleasant to read .
[ standingsu2f ] throughout this section : 1 .
we use the notation of [ coordssect ] .
( in particular , @xmath132 or @xmath462 , and @xmath752 . )
3 . @xmath1203 .
4 . @xmath0 is a closed , connected subgroup of @xmath121 that is compatible with @xmath119 ( see [ compatibledefn ] ) , so @xmath1204 ( see [ d(h)=dimh ] ) .
( note that @xmath162 is connected ( see [ solvable]([solvable - hcapl ] ) ) . )
we use the notation of [ calcsect ] .
( in particular , @xmath906 is defined in ( [ rhodefn ] ) and @xmath1207 is defined in ( [ deltadefn ] ) . )
except in subsection [ compatiblesubsect ] , @xmath0 is compatible with @xmath119 ( see [ compatibledefn ] ) .
recall that the real jordan decomposition of an element of @xmath1 is defined in ( [ jordandecomp ] ) ; any element @xmath155 of @xmath121 has a real jordan decomposition @xmath1208 ( with @xmath216 trivial ) .
if @xmath156 is an element of @xmath119 , rather than only conjugate to an element of @xmath119 , we could say that @xmath155 is compatible with @xmath119 .
" we now define a similar , useful notion for subgroups of @xmath121 .
lemma [ conjtocompatible ] shows there is usually no loss of generality in assuming that @xmath0 is compatible with @xmath119 , and lemma [ not - semi ] shows that the compatible subgroups can be described fairly explicitly .
[ compatibledefn ] let us say that @xmath0 is _ compatible _ with @xmath119 if @xmath1209 , where @xmath1210 , @xmath1205 , and @xmath1211 denotes the centralizer of @xmath328 in @xmath120 . in preparation for
the proofs of the main results , let us state a lemma that records a few of the nice properties of jordan components .
[ aucgood ] let @xmath214 be the real jordan decomposition of an element @xmath155 of @xmath1
. then 1 .
the real jordan decomposition of @xmath1212 is @xmath1213 ; and 2 .
@xmath156 , @xmath215 , and @xmath216 all belong to the zariski closure of @xmath1214 . therefore : 1 .
[ aucgood - norm ] @xmath156 , @xmath215 , and @xmath216 each normalize any connected subgroup of @xmath1 that is normalized by @xmath155 ; and 2 .
[ aucgood - id ] if @xmath1215 , for some @xmath1216 and some @xmath1217-invariant subspace @xmath537 of @xmath213 , then @xmath1218 [ conjtocompatible ] @xmath0 is conjugate , via an element of @xmath120 , to a subgroup that is compatible with @xmath119 . [ conjtocompatiblepf - algebraic ] assume , for the real jordan decomposition @xmath1069 of each element @xmath781 of @xmath0 , that @xmath156 and @xmath215 belong to @xmath0 . let @xmath328 be a maximal split torus of @xmath0 .
( recall that a split torus is a subgroup consisting entirely of hyperbolic elements . )
then @xmath328 is contained in some maximal split torus of @xmath1 , that is , in some subgroup of @xmath1 conjugate to @xmath119 ; replacing @xmath0 by a conjugate , we may assume @xmath1219 .
in other words , we now know that @xmath1220 is a maximal split torus of @xmath0 .
given @xmath591 , we have the real jordan decomposition @xmath1069 . by assumption , @xmath1221 ; thus , @xmath156 belongs to some maximal split torus @xmath1222 of @xmath0 .
a fundamental result of the theory of solvable algebraic groups implies that all maximal split tori of @xmath0 are conjugate via an element of @xmath1223 ( * ? ? ?
* thm . 4.21 ) , so there is some @xmath1224 , such that @xmath1225
. then @xmath1226 , being a subgroup of @xmath119 , is a split torus .
thus , the maximality of @xmath328 implies that @xmath1227 ; let @xmath1228 .
then @xmath1229 since @xmath591 is arbitrary , we conclude that @xmath1230 so @xmath0 is compatible with @xmath119 . the general case .
let @xmath1231 be the subgroup of @xmath121 generated by the jordan components of the elements of @xmath0 .
( of course , since every element of @xmath0 has a jordan decomposition , we have @xmath1232 . )
then case [ conjtocompatiblepf - algebraic ] applies to @xmath1180 , so , replacing @xmath0 by a conjugate , we may assume @xmath1233 , where @xmath1234 and @xmath1235 ( see [ conjtocompatiblepf - h = tu ] ) . because @xmath1180 normalizes @xmath0 ( see [ aucgood]([aucgood - norm ] ) ) , we know that @xmath1236 acts as the identity on @xmath1237 , for all @xmath1238 .
hence , lemma [ aucgood]([aucgood - id ] ) implies that @xmath1239 acts as the identity on @xmath1237 , for all @xmath1240 ; therefore @xmath1241 \subset h$ ] . also , we have @xmath1242 \subset [ an , an ] \subset n$ ] . thus , letting @xmath1205 , we have @xmath1243 \subset h
\cap n = u .\ ] ] because @xmath1244 and @xmath1245 is @xmath1246-invariant , the adjoint action of @xmath1247 on @xmath1245 is completely reducible , so ( [ conjtocompatiblepf - comm ] ) implies that there is a subspace @xmath1248 of @xmath1245 , such that @xmath1249 = 0 $ ] and @xmath1250 .
therefore , @xmath1251 , so @xmath1252 let @xmath1253 be the projection with kernel @xmath120 , and let @xmath1254 .
then @xmath1255 so @xmath1256 .
for any @xmath591 , we know , from ( [ conjtocompatiblepf - hbarcompat ] ) , that there exist @xmath1257 , @xmath1258 and @xmath1259 , such that @xmath1260 . because @xmath1261 , we must have @xmath1262 and , because @xmath1263 , we have @xmath1264 . therefore ,
we conclude that @xmath1266 , so @xmath0 is compatible with @xmath119 .
the preceding proposition shows that @xmath0 is conjugate to a subgroup @xmath122 that is compatible with @xmath119 . the subgroup @xmath122 is usually not unique , however .
the following lemma provides one way to change @xmath122 , often to an even better subgroup .
[ conjuomega ] assume that @xmath0 is compatible with @xmath119 , and let @xmath1210 . if @xmath1267 , then @xmath1268 is compatible with @xmath119 .
let @xmath1269 , @xmath1270 , and @xmath1271 . because @xmath215 centralizes @xmath328
, we have @xmath1272 also , because @xmath1133 , and @xmath120 is normal , we have @xmath1273 , so @xmath1274 since @xmath1133 , we have @xmath1275 , so @xmath1276 and @xmath1277 thus , @xmath1278 as desired .
[ not - semi ] if @xmath0 is compatible with @xmath119 , then either 1 .
[ not - semi - tu ] @xmath1279 ; or 2 .
[ not - semi - not ] there is a positive root @xmath335 , a nontrivial group homomorphism @xmath1280 , and a closed , connected subgroup @xmath162 of @xmath120 , such that 1 . [ not - semi - codim1 ] @xmath1281 ; 2 .
[ not - semi - normal ] @xmath162 is normalized by both @xmath1282 and @xmath1283 ; and 3 .
[ not - semi - disjoint ] @xmath1284 . because @xmath0 is compatible with @xmath119 , we have @xmath1266 , where @xmath1285 and @xmath1205 .
we may assume that @xmath1286 , for otherwise ( [ not - semi - tu ] ) holds . therefore @xmath1287 .
because @xmath1288 is a sum of root spaces , this implies that there is a positive root @xmath335 , such that @xmath1289 . because @xmath423 , we have @xmath1290 , so we must have @xmath1291 ( otherwise we would have @xmath1292 , so @xmath1293 ; hence ( [ not - semi - tu ] ) holds )
. therefore , @xmath1294 . because @xmath1295 , we have @xmath1296 $ ] , so
there is a nontrivial one - parameter subgroup @xmath1297 in @xmath1298 that is not contained in @xmath162 . because @xmath328 centralizes @xmath1211 , we may write @xmath1299 where @xmath1300 is a one - parameter subgroup of @xmath328 and @xmath1301 is a one - parameter subgroup of @xmath1211 . furthermore , this decomposition is unique , because @xmath1302 .
( in fact , @xmath1299 is the real jordan decomposition of @xmath1303 . )
define @xmath1304 by @xmath1305 for all @xmath599 .
( [ not - semi - codim1 ] ) for all @xmath599 , we have @xmath1306 , which establishes one inclusion of ( [ not - semi - codim1 ] )
. the other will follow if we show that @xmath1307 , so suppose @xmath1308 . then lemma [ dimt]([dimt - a ] ) implies that @xmath1309 , so it follows from lemma [ rootdecomp ] ( with @xmath386 and @xmath385 ) that @xmath1310 , contradicting our assumption that @xmath1311 .
( [ not - semi - normal ] ) because @xmath1312 , we know that each of @xmath694 and @xmath695 normalizes @xmath0 ( see [ aucgood]([aucgood - norm ] ) ) .
being in @xmath121 , they also normalize @xmath120 .
therefore , they normalize @xmath1313 .
( [ not - semi - disjoint ] ) suppose @xmath1314 . because the intersection @xmath1315 is connected ( see [ not - semi - normal ] and [ solvable]([solvable - hcapl ] ) ) , and @xmath1316 , we must have @xmath1317 .
therefore @xmath1318 , so @xmath1319 .
this contradicts our assumption that @xmath1286 .
[ tnormsu ] if @xmath0 is compatible with @xmath119 , then @xmath1320 normalizes @xmath1223 .
[ dimt ] if @xmath1321 , then 1 .
[ dimt - a ] @xmath0 contains a conjugate of @xmath119 ; and 2 . [ dimt - cds ] @xmath0 is a cartan - decomposition subgroup .
( [ dimt - a ] ) let @xmath1253 be the projection with kernel @xmath120 , and let @xmath1180 be the zariski closure of @xmath0 . from the structure theory of solvable algebraic groups ( * ? ? ?
10.6(4 ) , pp .
137138 ) , we know that @xmath1322 is the semidirect product of a torus @xmath328 and and unipotent subgroup @xmath1323 .
replacing @xmath0 by a conjugate under @xmath120 , we may assume that @xmath340 .
since @xmath1324 we must have @xmath1325 , so @xmath1326 normalizes @xmath0 ( see [ zar - norm ] ) .
then , since @xmath1325 , we conclude that @xmath1327 ( see [ rootdecomp ] ) .
( [ dimt - cds ] ) from ( [ dimt - a ] ) , we see that , by replacing @xmath0 with a conjugate subgroup , we may assume @xmath1309 . because @xmath119 is a cartan - decomposition subgroup ( see [ aiscds ] ) , this implies @xmath0 is a cartan - decomposition subgroup .
the following basic result was used twice in the above arguments .
[ zar - norm ] if @xmath0 is a closed , connected subgroup of @xmath1 , then the zariski closure of @xmath0 normalizes @xmath0 .
our goal is to prove theorem [ maxnolinear ] ; we begin with some preliminary results .
first , an observation that simplifies the calculations in some cases , by allowing us to assume that @xmath1008 .
[ x=0 ] let @xmath1328 .
if @xmath1329 and @xmath1330 , then there is some @xmath1331 , such that 1 .
@xmath1332 , 2 .
@xmath1333 , and 3 .
@xmath1334 .
because @xmath1329 and @xmath1330 , there is some @xmath1335 , such that @xmath1336 .
let * @xmath722 be the element of @xmath682 with @xmath1337 , * @xmath1338 , and * @xmath1339 . from ( [ conjugation ] ) , we see that * @xmath1340 , * @xmath1341 , and * @xmath1342 , as desired . [ hinn - linear ] if there does not exist a continuous curve @xmath967 in @xmath162 , such that @xmath1343 , then 1 .
[ hinn - linear - phi=0&eta2= ] for every nonzero element @xmath1344 of @xmath803 , we have @xmath1345 ; and 2 .
[ hinn - linear - phi=0&=0 ] for every element @xmath215 of @xmath1346 , such that @xmath1347 , we have @xmath1348 ( [ hinn - linear - phi=0&eta2= ] ) suppose there is a nonzero element @xmath1344 of @xmath803 with @xmath1349 . let @xmath1350 ( see [ exp(phi=0 ) ] ) .
we have @xmath1351 then , because @xmath1352 , it is easy to see that @xmath1353 .
also , we have @xmath1354 , so @xmath1355 , as desired .
( [ hinn - linear - phi=0&=0 ] ) suppose there is an element @xmath215 of @xmath1346 , such that @xmath1356 , and @xmath1357 let @xmath1358 .
[ hinnlinearpf - x=0 ] assume @xmath1008 . because @xmath1356 , we must have @xmath1330 . then , from , we know that @xmath1359 .
so , from , we see that * @xmath1360 , * @xmath1361 whenever @xmath1362 , and * @xmath1363 whenever @xmath1364 and @xmath1365 .
this implies that @xmath1366 .
[ hinnlinearpf - y=0 ] assume @xmath834 .
this is similar to case [ hinnlinearpf - x=0 ] .
( in fact , this can be obtained as a corollary of case [ hinnlinearpf - x=0 ] by replacing @xmath0 with its conjugate under the weyl reflection corresponding to the root @xmath675 . )
assume @xmath1330 . because @xmath1356 , lemma [ x=0 ] implies there is some @xmath1331 , such that , letting @xmath1339 , we have @xmath1367 , @xmath1368 , and @xmath1369 . we show below that is satisfied with @xmath1370 in the place of @xmath215 , so , from case [ hinnlinearpf - x=0 ] ,
we conclude that @xmath1371 .
thus , the desired conclusion follows from lemma [ conjcurve ] ( with @xmath1201 ) . to complete the proof ,
we now show that is satisfied with @xmath1370 in the place of @xmath215 .
( this can be verified by direct calculation , but we give a more conceptual proof . ) because @xmath1372 , multiplication by @xmath1373 on the left performs a row operation on the first two rows of @xmath781 ; likewise , multiplication by @xmath155 on the right performs a column operation on the last two columns of @xmath781 .
these operations do not change the determinant @xmath1207 : thus @xmath1374 from and the definition of @xmath678 , we see that @xmath1375 because @xmath1347 , we have @xmath1376 , so this simplifies to @xmath1377 thus , is equivalent to the condition that @xmath1378 .
then , since @xmath1379 we conclude that is also valid for @xmath1370 .
[ dim(u+z)<3 ] if there does not exist a continuous curve @xmath967 in @xmath162 , such that @xmath1343 , then @xmath1380 furthermore , if equality holds , and @xmath136 , then @xmath1381 and @xmath1382 .
assume @xmath136 . because @xmath1383 is a quadratic form of signature @xmath1384 on @xmath1385 ,
we know , from [ hinn - linear]([hinn - linear - phi=0&eta2= ] ) , that @xmath1386 .
thus , we may assume @xmath1387 , so there is some @xmath1388 , such that @xmath1389 .
[ dim(u+z)<3pf - yy=0 ] assume there exists @xmath1390 , such that @xmath1391 and @xmath1392 . from ( [ [ u , v ] ] )
, we see that @xmath1393 \in { { { \mathfrak{\lowercase{d}}}}_{{\mathfrak{\lowercase{h}}}}}$ ] , with @xmath1394 } = -{\operatorname{im}}(\phi_u { \overline{\eta_z } } ) \neq 0 $ ] and @xmath1395 } = \eta_{[u , z ] } = 0 $ ] .
this contradicts [ hinn - linear]([hinn - linear - phi=0&eta2= ] ) .
assume there exists @xmath1390 , such that @xmath1396 . from ( [ [ u , v ] ] )
, we see that @xmath1393 $ ] is an element of @xmath803 , such that @xmath1395 } = 0 $ ] , and @xmath1397 } = \phi_u { { \mathord{\mathsf{y}}}}_z$ ] is a purely imaginary multiple of @xmath1011 .
so subcase [ dim(u+z)<3pf - yy=0 ] applies ( with @xmath1393 $ ] in the place of @xmath1344 ) .
assume @xmath1391 and @xmath1398 , for all @xmath1390 . from [ hinn - linear]([hinn - linear - phi=0&eta2= ] )
, we see that @xmath1399 , so the assumption of this subcase implies @xmath1400 .
thus , @xmath1401 so the desired inequality holds .
if equality holds , then @xmath1402 and @xmath1403 .
thus , we may choose @xmath1390 , such that @xmath1404 , and @xmath1405 , such that @xmath1406 . from the assumption of this subcase , we know that @xmath1398 ; thus , @xmath1407 .
therefore , subcase [ dim(u+z)<3pf - yy=0 ] applies , with @xmath1408 in the place of @xmath215 .
assume @xmath132 . because @xmath1409 and @xmath1410 , the desired inequality holds unless @xmath1411 and @xmath1387 .
thus , we may assume there is some nonzero @xmath1390 and some @xmath1328 , such that @xmath1389 .
assume @xmath834 .
we may assume @xmath1412 ( by replacing @xmath215 with @xmath1413 , if necessary ) . let @xmath1414 . from , we see that @xmath1415 , but @xmath1416 therefore @xmath1417 , so lemma [ owalls]([owalls - linear ] ) implies that @xmath1343 .
this is a contradiction .
assume @xmath1330 .
let @xmath722 be the element of @xmath791 with @xmath1418 , and let @xmath1419 .
then @xmath1368 ( see [ conjugation ] and [ [ u , v ] ] ) .
thus , by replacing @xmath0 with the conjugate @xmath1420 ( see [ conjcurve ] ) , we may assume @xmath1008 .
for any large real number @xmath939 , let @xmath1421 be the element of @xmath1422 that satisfies @xmath1423 .
then , from , we see that @xmath1424 clearly , we have @xmath1425 .
a calculation shows that @xmath1426 , and certain other @xmath864 minors also have cancellation . with this in mind
, it is not difficult to verify that @xmath1427 ( see ( * ? ? ?
* case 3 of pf . of 5.12(@xmath1428 ) ) for details ) .
this is a contradiction .
[ dim(u)<2n ] if there does not exist a continuous curve @xmath967 in @xmath1429 , such that @xmath1343 , then @xmath1430 furthermore , 1 . [ dim(u)<2n - eqeven ] if equality holds , and @xmath79 is even , then @xmath1431 , for every @xmath1432 ; 2 .
[ dim(u)<2n - eq=3 ] if equality holds , and @xmath1433 , then @xmath1434 . by passing to a subgroup ,
we may assume @xmath1381 .
let @xmath163 be the projection of @xmath337 to @xmath1435 ; then @xmath1436 .
[ dim(u)<2n-<xy > pf ] assume @xmath1437 for every @xmath1438 .
assume @xmath79 is even . from theorem [ hinn - linear]([hinn - linear - phi=0&eta2= ] ) , we know that @xmath163 does not intersect @xmath791 ( or @xmath792 , either , for that matter ) , so @xmath1439 therefore @xmath1440 as desired .
( if equality holds , then we have conclusion ( [ dim(u)<2n - eqeven ] ) . ) [ dim(u)<2npf-<xy>-n=2 ] assume @xmath79 is odd .
[ dim(u)<2npf - u / z ] we have @xmath1441 .
suppose not : then @xmath1442 ( this will lead to a contradiction . ) let @xmath1443 , so @xmath507 is a @xmath338-subspace of @xmath1444 . for each @xmath1445 , there is some @xmath1446 , such that @xmath1447 ;
define @xmath1448 . by the assumption of this case ,
we know @xmath1449 so @xmath1450 is uniquely determined by @xmath767 ; thus , @xmath1451 is a well - defined @xmath338-linear map . also , again from the assumption of this case , we know that @xmath1452 because @xmath1453 , we have @xmath1454 if @xmath132 ( that is , if @xmath1455 ) , this implies @xmath1456 , so @xmath947 is defined on all of @xmath1457 . because @xmath79 is odd , this implies that @xmath947 has a real eigenvalue , which contradicts .
we may now assume @xmath136 .
let * @xmath1458 , where @xmath1459 , * @xmath1460 be the projective space of the real vector space @xmath507 , and * @xmath1461 \in { \mathord{\mathbb{p}x}}$ ] , for @xmath1462 , so @xmath1463 is a vector bundle over @xmath1460 .
define @xmath1464 by @xmath1465 .
any @xmath338-linear transformation @xmath1466 is a continuous function , such that @xmath1467 for all @xmath1445 ; that is , a section of @xmath1463 .
thus , @xmath1468 , @xmath947 , and @xmath155 each define a section of @xmath1463 . furthermore , these three sections are pointwise linearly independent over @xmath338 , because implies that @xmath767 , @xmath1469 , and @xmath1470 are linearly independent over @xmath338 , for every nonzero @xmath1445 . on the other hand , the theory of characteristic classes ( * ? ? ?
. 4 , p. 39 ) implies that @xmath1463 does not have three pointwise @xmath338-linearly independent sections ( see ( * ? ? ?
8.2 ) for details ) .
this is a contradiction .
completion of the proof of subcase [ dim(u)<2npf-<xy>-n=2 ] . from step [ dim(u)<2npf - u / z ]
, we see that the desired inequality holds .
we may now assume @xmath1433 and @xmath1471 . since @xmath1472 , we must have @xmath1455 , so @xmath132 .
therefore @xmath1473 , so @xmath1474 as desired .
assume there is some @xmath1475 , such that @xmath1476 .
[ dimupf - x=0 ] assume @xmath1477 .
since @xmath1478 , we must have @xmath1479 .
then @xmath1480 for every @xmath1390 ( otherwise [ hinn - linear]([hinn - linear - phi=0&=0 ] ) yields a contradiction ) ; this implies @xmath1481 and @xmath1482 because @xmath1483 , we know that @xmath1484 ; so @xmath1485 since @xmath1486 for every @xmath1390 , but @xmath1487 ( see [ hinn - linear]([hinn - linear - phi=0&eta2= ] ) ) , we must have @xmath1488 for every @xmath1390 .
therefore @xmath1489 let @xmath1490 be the natural projection .
note that @xmath1491 ( if @xmath1492 , with @xmath1493 , then there is some @xmath599 , such that @xmath1494 .
we also have @xmath1495 , so , from [ hinn - linear]([hinn - linear - phi=0&=0 ] ) , we see that @xmath1496 .
thus @xmath1497 .
so @xmath1498 is 1-dimensional . ) because @xmath1486 for every @xmath1390 , and @xmath337 is a lie algebra , we see , from ( [ [ u , v ] ] ) , that @xmath1499 must be a totally isotropic subspace for the symplectic form @xmath1500 , so @xmath1501 therefore @xmath1502 this completes the proof if @xmath1503 : * if @xmath79 is even , then , because @xmath1504 , we have @xmath1505 . *
if @xmath1506 is odd , then @xmath1507 , so @xmath1508 .
now let @xmath1433 , and suppose @xmath1509 .
( this will lead to a contradiction . ) because equality is attained in the proof above , we must have @xmath1510 , so there exists @xmath1511 with @xmath1512 . for @xmath599 ,
let @xmath1513 .
then @xmath1514 thus , this expression changes sign , so it must vanish for some @xmath939 . on the other hand , since @xmath1433 , we have @xmath1515 for every @xmath1516 . thus [ hinn - linear]([hinn - linear - phi=0&=0 ] ) yields a contradiction .
[ dimupf - y=0 ] assume @xmath1517 .
this is similar to subcase [ dimupf - x=0 ] .
( in fact , this can be obtained as a corollary of subcase [ dimupf - x=0 ] by replacing @xmath0 with its conjugate under the weyl reflection corresponding to the root @xmath675 . )
assume @xmath1479 .
because @xmath1518 , lemma [ x=0 ] implies there is some @xmath1331 , such that , letting @xmath1519 , we have @xmath1367 , @xmath1368 , and @xmath1520 .
there is no harm in replacing @xmath0 with @xmath1521 ( see [ conjcurve ] ) .
then subcase [ dimupf - x=0 ] applies ( with @xmath1370 in the place of @xmath722 ) .
[ maxnolinear ] recall that assumptions [ standingsu2f ] are in effect .
if there does not exist a continuous curve @xmath967 in @xmath0 , such that @xmath1343 , then @xmath1522 furthermore , if equality holds , and @xmath79 is even , then 1 .
[ maxnolinear - tu ] @xmath1523 ; 2 .
[ maxnolinear - phi0 ] @xmath800 for every @xmath1328 ; 3 .
[ maxnolinear-<xy > ] @xmath1524 , for every @xmath1438 ; 4 .
[ maxnolinear - z ] @xmath1525 for every nonzero @xmath1390 ; 5 .
[ maxnolinear - dimu / z ] @xmath1526 ; and 6 .
[ maxnolinear - dimz ] @xmath1527 .
let @xmath1528 from lemmas [ dimt]([dimt - a ] ) and [ dim(u+z)<3 ] , and proposition [ dim(u)<2n ] , we have @xmath1529 this implies the desired inequality , unless @xmath1433 , @xmath136 , and we have equality in both lemma [ dim(u+z)<3 ] and proposition [ dim(u)<2n ] .
this is impossible , because equality in lemma [ dim(u+z)<3 ] requires @xmath1530 , but proposition [ dim(u)<2n]([dim(u)<2n - eq=3 ] ) implies @xmath1531 . in the remainder of the proof , we assume that equality holds in , and that @xmath79 is even
. proposition [ hinn - linear]([hinn - linear - phi=0&eta2= ] ) implies ( [ maxnolinear - z ] ) .
[ maxnolinearpf - f = c ] assume @xmath136 . because equality holds , lemma [ dim(u+z)<3 ] implies ( [ maxnolinear - phi0 ] ) and ( [ maxnolinear - dimz ] )
. then proposition [ dim(u)<2n]([dim(u)<2n - eqeven ] ) implies ( [ maxnolinear-<xy > ] ) ( because @xmath1381 ) . since @xmath1381 ( see [ maxnolinear - phi0 ] ) and equality holds in , we have @xmath1532 and @xmath1533 so ( [ maxnolinear - dimu / z ] ) and ( [ maxnolinear - dimz ] ) hold .
let @xmath1210 .
corollary [ tnormsu ] implies that @xmath328 normalizes @xmath337 , so , from ( [ maxnolinear-<xy > ] ) and lemma [ rootdecomp ] , we see that @xmath1534 . on the other hand , @xmath1535 , so , from equality in , we conclude that @xmath1536 .
therefore @xmath1537 .
suppose @xmath1538 is any continuous group homomorphism , such that @xmath1539 normalizes @xmath162 . from ( [ maxnolinear-<xy > ] ) and ( [ [ u , v ] ] )
, we see that @xmath1540 , so @xmath343 must be trivial .
this implies that [ not - semi]([not - semi - not ] ) can not apply here , so [ not - semi]([not - semi - tu ] ) yields ( [ maxnolinear - tu ] ) .
assume @xmath132 .
proposition [ dim(u)<2n]([dim(u)<2n - eqeven ] ) implies that @xmath1541 for every @xmath1432 .
suppose ( [ maxnolinear - phi0 ] ) is false .
then there is some @xmath1328 , such that @xmath1389 . also , because @xmath1542 , we may fix some @xmath1543
then , letting @xmath1544 $ ] , we see , from ( [ [ u , v ] ] ) , that @xmath1545 and @xmath1512 , so @xmath1546 . this contradicts the conclusion of the preceding paragraph .
conclusion ( [ maxnolinear-<xy > ] ) follows from ( [ maxnolinear - phi0 ] ) and [ dim(u)<2n]([dim(u)<2n - eqeven ] ) .
conclusion ( [ maxnolinear - tu ] ) can be established by arguing as in the last two paragraphs of case [ maxnolinearpf - f = c ] .
equations ( [ maxnolinearpf - u / z ] ) and ( [ maxnolinearpf - dimz ] ) establish ( [ maxnolinear - dimu / z ] ) and ( [ maxnolinear - dimz ] ) . our goal is to prove theorem [ bestnosquare ] ; we start with two preliminary results .
[ hinn - square ] if there does not exist a continuous curve @xmath967 in @xmath162 , such that @xmath1547 , then 1 .
[ hinn - square - indep ] for every element @xmath215 of @xmath1346 , we have @xmath1548 ; 2 . [ hinn - square - eta2neq ] for every element @xmath1344 of @xmath803 , we have @xmath1549 ; and 3 .
[ hinn - square - y=0+yy=0 ] for every element @xmath215 of @xmath337 , such that @xmath1389 , @xmath834 , and @xmath1550 , we have @xmath1551 .
( [ hinn - square - indep ] ) suppose there is an element @xmath215 of @xmath1346 , such that @xmath1552 .
let @xmath1553 .
then , from , we see that @xmath1554 .
furthermore , @xmath1555 because @xmath1552 , we have @xmath1556 , so @xmath1557 ; therefore @xmath1558 , so @xmath1559 so lemma [ owalls]([owalls - square ] ) implies that @xmath1547 , as desired .
( [ hinn - square - eta2neq ] ) suppose there is an element @xmath1344 of @xmath803 , such that @xmath1560 ; in other words , we have @xmath1561 . let @xmath1562 ( see [ exp(phi=0 ) ] ) .
then @xmath1563 and @xmath1564 so @xmath1565 so lemma [ owalls]([owalls - square ] ) implies that @xmath1547 , as desired .
( [ hinn - square - y=0+yy=0 ] ) suppose there is an element @xmath215 of @xmath337 , such that @xmath1566 , @xmath834 , @xmath1550 , and @xmath1567 .
let @xmath1414 . from ( [ exp(y=0 ) ] ) , we see that @xmath1568 ( note that , because @xmath1567 , we have @xmath1569 ) . then @xmath1570 and @xmath1571 so @xmath1572 . thus , lemma [ owalls]([owalls - square ] ) implies that @xmath1547 , as desired .
the following lemma obtains a dimension bound from condition [ hinn - square]([hinn - square - indep ] ) .
[ dimv < d(n-2 ) ] if @xmath163 is a @xmath338-subspace of @xmath1573 , such that @xmath1574 for every @xmath1575 , then either 1 .
[ dimv < d(n-2)-n>3 ] @xmath1576 ; or 2 .
[ dimv < d(n-2)-n=3 ] @xmath1433 and @xmath1577 .
because @xmath1578 , we may assume that there exist nonzero @xmath1579 , such that @xmath1580 and @xmath1581 ( otherwise , the projection to one of the factors of @xmath1573 is injective when restricted to @xmath163 , so ( [ dimv < d(n-2)-n>3 ] ) holds ) .
then @xmath1582 , so , by assumption , we have @xmath1583 . because @xmath1584 and @xmath1585 are nonzero , this implies @xmath1586 . [ dimv < d(n-2)pf - yinfx ] for all @xmath1575 , we have @xmath1587
. we may assume @xmath1588 ( otherwise the desired conclusion is obvious ) .
then , since @xmath1589 , we conclude that @xmath1590 .
similarly , because @xmath1591 we must have @xmath1592 .
therefore @xmath1593 since @xmath1594 , this implies @xmath1595 , so @xmath1587 , as desired .
[ dimv < d(n-2)pf - v = fxf ] we have @xmath1596 . given @xmath1575 , step [ dimv < d(n-2)pf - yinfx ] asserts that @xmath1587 . by symmetry ( interchanging the two factors of @xmath1597 )
, we must also have @xmath1598 .
so @xmath1599 , as desired .
completion of the proof . from step [ dimv
< d(n-2)pf - v = fxf ] , we have @xmath1600 if @xmath1504 , then ( [ dimv < d(n-2)-n>3 ] ) holds ; otherwise , ( [ dimv < d(n-2)-n=3 ] ) holds .
[ bestnosquare ] recall that assumptions [ standingsu2f ] are in effect .
if there does not exist a continuous curve @xmath967 in @xmath0 , such that @xmath1547 , then @xmath1601 .
furthermore , if equality holds , then @xmath0 is of the form @xmath1602 , where 1 .
[ bestnosquare - t ] @xmath1603 , 2 .
[ bestnosquare - u ] @xmath1604 , and 3 .
[ bestnosquare - x ] @xmath1605 for every @xmath1606 .
note that @xmath1607 ( see [ dimt]([dimt - a ] ) ) and @xmath1608 we have @xmath1609 .
suppose not .
let @xmath163 be the projection of @xmath1346 to @xmath1610 .
we have @xmath1611 and , for every @xmath1612 with @xmath1613 , we have @xmath1614 ( see [ hinn - square]([hinn - square - indep ] ) ) , so lemma [ dimv < d(n-2 ) ] implies that @xmath1433 . therefore @xmath1615 .
then , because @xmath1616 , we know that @xmath1617 and @xmath1618 ; thus , there exist @xmath1619 , such that * @xmath1008 , @xmath1330 ; and * @xmath1620 , @xmath1517 . therefore @xmath822 $ ] is a nonzero element of @xmath793 ( see [ [ u , v ] ] ) , so @xmath1621 \bigr ) \neq 0 $ ] .
this contradicts lemma [ hinn - square]([hinn - square - eta2neq ] ) .
we have @xmath1622 .
suppose not : then , because @xmath1623 , there is some @xmath1624 , and , because @xmath1625 , there is some nonzero @xmath1626 , such that @xmath1627 .
we must have @xmath1628 ( otherwise [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction ) ; thus @xmath1629 .
we must have @xmath1630 ( otherwise [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction ) .
thus , we see that @xmath1631 is nonconstant as a function of @xmath599 , so [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction . the desired inequality .
we have @xmath1632 as desired . in the remainder of the proof , we assume that @xmath1633
. we must have equality throughout the preceding paragraphs .
[ vinalpha+beta ] we have @xmath1634 .
suppose not : then there is some @xmath1635 , such that @xmath1636 .
let @xmath1637 and @xmath1544 $ ] .
then , from , we see that @xmath1545 and @xmath1512 , and that @xmath1638 \in { { \mathfrak{\lowercase{n}}}}_{\alpha+2\beta } + { { \mathfrak{\lowercase{n}}}}_{2\alpha+2\beta}$ ] . from [ hinn - square]([hinn - square - indep ] ) , we have @xmath1639 and @xmath1640 , so @xmath1641 therefore @xmath1642 , so @xmath1643 } \neq 0 $ ] ( see [ [ u , v ] ] ) , so [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction .
we have @xmath1644 . from step [ vinalpha+beta ] , together with the fact that @xmath1645 we conclude that @xmath1646 . therefore , @xmath1647 so @xmath1648 \\ & = [ { \mathfrak{\lowercase{u}}}_{\phi=0 } + { { \mathfrak{\lowercase{d } } } } , { \mathfrak{\lowercase{u}}}_{\phi=0 } + { { \mathfrak{\lowercase{d } } } } ] \\ & = [ { { \mathfrak{\lowercase{n}}}}_{\alpha+\beta } + { { \mathfrak{\lowercase{d } } } } , { { \mathfrak{\lowercase{n}}}}_{\alpha+\beta } + { { \mathfrak{\lowercase{d } } } } ] \\ & = [ { { \mathfrak{\lowercase{n}}}}_{\alpha+\beta},{{\mathfrak{\lowercase{n}}}}_{\alpha+\beta } ] \\ & = { { \mathfrak{\lowercase{n}}}}_{2\alpha+2\beta } . \end{aligned}\ ] ] because @xmath1649 , we must have @xmath1644 .
we have @xmath1644 .
let @xmath1650 be the projection of @xmath0 to @xmath119 .
then there exists @xmath1651 , such that @xmath1652 , and , in the notation of lemma [ rootdecomp ] , we have @xmath1653 because @xmath328 normalizes @xmath337 ( see [ tnormsu ] ) , we know , from lemma [ rootdecomp ] , that @xmath1654 . since @xmath1655 , we know that @xmath1656 projects nontrivially ( in fact , surjectively ) to @xmath682 . on the other hand , we know that @xmath1657 ( otherwise [ hinn - square]([hinn - square - y=0+yy=0 ] ) yields a contradiction ) .
therefore @xmath1658 , so there must be a positive root @xmath1659 , such that @xmath1660 . then @xmath1661 ; since @xmath1662 , we must have @xmath1663 .
because @xmath1664 , we have @xmath1665 then , since @xmath1666 , we must have @xmath1667 by inspection , we see that this implies @xmath1668 , so we conclude that @xmath1651 , as desired .
[ bestnosquarepf - sigma ] we have @xmath1669 , @xmath1603 , and @xmath1670 . since @xmath1671 , it suffices to show @xmath1672 .
thus , let us suppose @xmath1673 .
( this will lead to a contradiction . )
we have @xmath1674 ( and recall that @xmath1675 ) , so there is some @xmath1328 , such that @xmath1389 and @xmath1330 . because @xmath1646 , we have @xmath1676 so there is some @xmath1635 , such that @xmath1677 and @xmath1627 .
then @xmath822 \in { { \mathfrak{\lowercase{n}}}}_{\alpha+2\beta } + { { \mathfrak{\lowercase{n}}}}_{2\alpha + 2\beta}$ ] , with @xmath1678 } \neq 0 $ ] ( see [ [ u , v ] ] ) , so [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction .
[ bestnosquarepf - tu ] we have @xmath1679 .
suppose not : because @xmath1603 , we conclude that there is some nonzero @xmath1680 , such that @xmath1370 normalizes @xmath337 ( see [ not - semi ] ) . if @xmath1681 , then , because @xmath1646 , there is some @xmath1682 , such that @xmath1683 and @xmath1627 .
then @xmath1684 \in { { \mathfrak{\lowercase{n}}}}_{\alpha+2\beta } + { { \mathfrak{\lowercase{n}}}}_{2\alpha + 2\beta}$ ] , with @xmath1685 } \neq 0 $ ] ( see [ [ u , v ] ] ) , so [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction .
if @xmath1545 , then , since @xmath1686 , we must have @xmath1687 .
there is some @xmath1688 with @xmath1689 .
then @xmath1684 \in { { \mathfrak{\lowercase{n}}}}_{\alpha+2\beta } + { { \mathfrak{\lowercase{n}}}}_{2\alpha + 2\beta}$ ] , with @xmath1685 } \neq 0 $ ] ( see [ [ u , v ] ] ) , so [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction .
completion of the proof .
( [ bestnosquare - t ] ) from step [ bestnosquarepf - tu ] , we know that @xmath1690 , and , from step [ bestnosquarepf - sigma ] , that @xmath1603 .
( [ bestnosquare - u ] ) since @xmath1691 , it suffices to show @xmath1692 : given @xmath1693 , we wish to show @xmath1629 . because @xmath1646 , we know that @xmath1694 .
thus , all that remains is to show that @xmath1627 .
if not , then choosing @xmath1328 with @xmath1389 , we see that @xmath1678 } \neq 0 $ ] ( see [ [ u , v ] ] ) .
so [ hinn - square]([hinn - square - eta2neq ] ) yields a contradiction .
( [ bestnosquare - x ] ) from lemma [ hinn - square]([hinn - square - y=0+yy=0 ] ) , we know that @xmath1551 for every @xmath1606 .
this section proves two main results .
both assume that @xmath1 is either @xmath73 or @xmath52 . 1 .
theorem [ suf->complete ] shows that if @xmath79 is odd , and one or two specific homogeneous spaces of @xmath1 do not have tessellations , then no interesting homogeneous space of @xmath1 has a tessellation .
2 . theorem [ sufeventess ] shows that if @xmath79 is even , then certain deformations of the examples found by r. kulkarni and t. kobayashi ( see [ kulkarnieg ] ) are essentially the only interesting homogeneous spaces of @xmath1 that have tessellations .
assume conjecture [ notesssu / sp ] is true , and suppose @xmath3 is a crystallographic group for @xmath2 .
( this will lead to a contradiction . )
let @xmath1697 ( see [ su1ndefn ] ) . from ( [ d(sp ) ] ) , we have @xmath1698 and @xmath1699 , where @xmath752 .
we may assume that @xmath91 ( see [ hcanbean ] ) , and that @xmath0 is compatible with @xmath119 ( see [ conjtocompatible ] ) .
because @xmath0 is not a cartan - decomposition subgroup ( see [ cds->notess ] ) , the contrapositive of proposition [ cds<>h_m ] implies , for some @xmath577 , that there does * not * exist a continuous curve @xmath1700 in @xmath0 , such that @xmath988 .
therefore , either theorem [ maxnolinear ] ( if @xmath1201 ) or theorem [ bestnosquare ] ( if @xmath1202 ) implies that @xmath1701 .
theorem [ noncpctdim]([noncpctdim - tess ] ) ( combined with the fact that @xmath1702 ) implies that @xmath149 has a tessellation .
this contradicts either theorem [ so2n / so1odd - notess ] ( if @xmath132 ) or conjecture [ notesssu / sp][notesssu / sp - su2/su1 ] ( if @xmath136 ) . from lemma [ mu(suorsp ) ] , we know that @xmath361 and @xmath1703 are the two walls of @xmath154 , so corollary [ tess->misshk ] ( combined with the assumption of this case ) implies that @xmath3 acts properly discontinuously on @xmath450 .
therefore , since conjecture [ notesssu / sp][notesssu / sp - so2/su1][notesssu / sp - su2/sp1 ] asserts that @xmath450 does not have a tessellation , the contrapositive of theorem [ noncpctdim]([noncpctdim - tess ] ) ( with @xmath95 in the role of @xmath94 ) implies that @xmath1704 .
hence , the contrapositive of theorem [ maxnolinear ] implies there is a continuous curve @xmath1700 in @xmath0 , such that @xmath1343 .
thus , there is a compact subset @xmath12 of @xmath1 , such that @xmath147 ( see [ su1inh ] ) .
since @xmath3 acts properly discontinuously on @xmath2 , this implies that @xmath3 acts properly discontinuously on @xmath149 ( see [ chcproper ] ) .
this contradicts the assumption of this case . 1 . [ sufeventess - sp ]
there is an @xmath338-linear map @xmath1050 , such that 1 .
[ sufeventess - sp - xb ] @xmath1706 for every @xmath1707 ( see [ bsymplectic ] ) , and 2 .
[ sufeventess - sp-<x , y > ] @xmath1065 , for every nonzero @xmath766 ( see [ xbnotinfx ] ) , and 3 .
[ sufeventess - sp - hb ] @xmath0 is conjugate to @xmath1057 ( see [ hb - defn ] and [ hb = sp1 m ] ) ; or 2 .
[ sufeventess - sur ] @xmath132 and @xmath0 is conjugate to @xmath1708 ( see [ su1ndefn ] ) ; or 3 .
[ sufeventess - suc ] @xmath136 and there exists @xmath1119 $ ] , such that @xmath0 is conjugate to @xmath1123}}}$ ] ( see [ suegsdefn ] ) .
also , we may assume @xmath0 is compatible with @xmath119 ( see [ conjtocompatible ] ) . because @xmath0 is not a cartan - decomposition subgroup ( see [ cds->notess ] ) , proposition [ cds<>h_m ] implies that one of the following two cases applies . 1 .
[ sufeventesspf - tu ] @xmath1537 ; 2 .
[ sufeventesspf - phi0 ] @xmath800 for every @xmath1328 ; 3 .
[ sufeventesspf-<xy > ] @xmath1524 , for every @xmath1438 ; 4 .
[ sufeventesspf - z ] @xmath1525 for every nonzero @xmath1390 ; 5 .
[ sufeventesspf - dimu / z ] @xmath1526 ; and 6 .
[ sufeventesspf - dimz ] @xmath1527 .
for any @xmath1390 with @xmath1716 , we know , from ( [ sufeventesspf - z ] ) , that @xmath1099 ; therefore , lemma [ o(linear)]([o(linear)-o ] ) implies there exist @xmath338-linear maps @xmath1717 and @xmath1718 , such that @xmath1719 for all @xmath1390 .
more concretely , we may say that there exist @xmath1111 and @xmath1720 , such that @xmath1721 for all @xmath1390 .
let @xmath722 be the element of @xmath682 with @xmath1722 , and let @xmath1723 be the conjugate of @xmath0 by @xmath1724 . then @xmath124 satisfies the conditions imposed on @xmath0 ( note that @xmath124 , like @xmath0 , is compatible with @xmath119 ( see [ conjuomega ] ) ) , so there exist @xmath1725 and @xmath1726 , such that @xmath1727 for all @xmath1728 .
given @xmath1728 with @xmath1729 , let @xmath1730 .
because @xmath1731 , we have @xmath1732 , -v \bigr ] = 0 $ ] , so , from remark [ conjugation ] and , we see that @xmath1733 @xmath1734 and @xmath1735 therefore @xmath1736 since @xmath1737 is arbitrary , this implies @xmath1738 .
thus , by replacing @xmath0 with @xmath124 , we may assume that @xmath1739 .
this means that @xmath1740 for all @xmath1390 . from ( [ sufeventesspf - dimz ] )
( and because @xmath136 , so @xmath1741 ) , we know that @xmath1742 , so there is some nonzero @xmath1743 , such that @xmath1744 .
( so @xmath1745 . ) then @xmath1746 , so we see , from ( [ sufeventesspf - z ] ) , that @xmath1747 for every nonzero @xmath1390 . now , since @xmath1748 there is some nonzero @xmath1390 , such that @xmath1749 .
we have @xmath1750 because @xmath1751 is pure imaginary , we know that @xmath1752 , so this implies that @xmath1753 .
thus , replacing @xmath0 by a conjugate under a diagonal matrix , we may assume @xmath1754 , as desired .
thus , in the notation of lemma [ rootdecomp ] , we have @xmath1760 and @xmath1761 , so @xmath1762 , as desired .
( note that this is a direct sum of vector spaces , not of lie algebras : we have @xmath1763 \subset { { { \mathfrak{\lowercase{d}}}}_{{\mathfrak{\lowercase{h}}}}}$ ] . ) for any @xmath1764 with @xmath1008 , we have @xmath1765 so @xmath1766 ( see [ sufeventesspf-<xy > ] ) ; therefore , lemma [ o(linear)]([o(linear)-o ] ) implies there is a @xmath338-linear map @xmath1767 , such that @xmath1768 for all @xmath1764 . then , because @xmath1769 ( see [ sufeventesspf - dimu / z ] ) , we must have @xmath1770 combining this with ( [ sufeventesspf - tu ] ) and the conclusions of steps [ sufeventesssppf - xx = yy ] and [ sufeventesspf - u+z ] , we see that @xmath1771 .
therefore @xmath1772 , so conclusion ( [ sufeventess - sp - hb ] ) holds .
letting @xmath1773 $ ] , for any @xmath1774 , we see , from , that @xmath1775 and @xmath1776 from step [ sufeventesssppf - xx = yy ] , we know that @xmath1777 , so this implies that conclusion ( [ sufeventess - sp - xb ] ) holds . let @xmath1782 ( so @xmath1783 ) .
let @xmath1144 be the sesquilinear form ( or bilinear form , if @xmath132 ) on @xmath1784 defined by @xmath1785 let @xmath1786 from ( [ pdqf ] ) , we see that the restriction of @xmath1787 to @xmath1788 is a ( positive - definite ) inner product .
let @xmath1789 be the @xmath1790-orthogonal complement to @xmath1788 . as a form over @xmath751
, @xmath1144 has signature @xmath1791 .
thus , as a form over @xmath338 , @xmath1787 has signature @xmath1792 . since @xmath1793
we conclude that @xmath1789 is a @xmath1794-dimensional @xmath338-subspace on which @xmath1787 is negative - definite .
choose some nonzero @xmath1795 .
multiplying by a real scalar to normalize , we may assume @xmath1796 . because @xmath1797 is transitive on the vectors of norm @xmath1116
, there is some @xmath1798 , such that @xmath1799 .
thus , letting @xmath1800 and @xmath1801 , we have @xmath1802 , so , by replacing @xmath270 with the conjugate @xmath1803 , we may assume @xmath1804 .
choose some nonzero @xmath1807 , such that @xmath722 is @xmath1790-orthogonal to @xmath215 .
multiplying by a real scalar to normalize , we may assume @xmath1808 . by replacing @xmath722 with @xmath1809 if necessary
, we may assume @xmath1810 .
because @xmath722 is @xmath1790-orthogonal to @xmath1804 , we have @xmath1811 ( see [ uperp ] ) .
let @xmath1812 and @xmath1813 .
then @xmath1814 so @xmath1815 also , @xmath1816 so @xmath1817 .
thus , @xmath486 and @xmath939 are of opposite signs so , because @xmath1818 therefore , we may choose @xmath1819 $ ] , such that @xmath1820 let @xmath1821 then @xmath1822 and @xmath1823 hence , there is some @xmath1824 , such that @xmath1825 and @xmath1826 .
thus , replacing @xmath270 with the conjugate @xmath1827 ( cf .
[ ghat ] ) , we may assume @xmath1828 . therefore @xmath1829 by combining this with and comparing with ( [ suegsdefn ] ) ( with @xmath136 ) , we conclude that @xmath1830 } } $ ] . by comparing dimensions
, we see that equality must hold ; this establishes conclusion ( [ sufeventess - suc ] ) ( because @xmath1831 ) .
( [ suf - known - sim ] ) in each case , there is some @xmath425 , such that @xmath966 for @xmath591 ( see [ hbthm]([hbthm - mu ] ) , [ mu(suorsp)]([mu(suorsp)-su ] ) , and [ suegs]([suegs - linear ] ) )
. then corollary [ hsimsuorsp ] implies either that @xmath1038 ( if @xmath1835 ) or that @xmath976 ( if @xmath1202 ) .
suppose @xmath1837}}}g = { h_{[c']}}$ ] , for some @xmath1838 .
because all maximal split tori in @xmath1154}}$ ] are conjugate , we may assume that @xmath155 normalizes @xmath1839 .
since all roots of @xmath1839 on both @xmath1121}}}$ ] and @xmath1840}}$ ] are positive , @xmath155 can not invert @xmath1841 , so we conclude that @xmath155 centralizes @xmath1841 ; that is , @xmath1842 . in the notation of case [ sufeventesspf - nosquare ] of the proof of theorem [ sufeventess ] ,
define @xmath1843 ( cf .
[ ghat ] ) . then @xmath1844
, so we may assume @xmath1845 ( because @xmath1841 , being a subgroup of @xmath1123}}}$ ] , obviously normalizes @xmath1123}}}$ ] ) .
write @xmath1846 . then , because @xmath1837}}}g = { h_{[c']}}$ ] , we must have @xmath1847 } } } } ) = v_{{{{\mathfrak{\lowercase{h}}}}_{[c']}}}$ ] ; hence @xmath1847}}}}^\perp ) = v_{{{{\mathfrak{\lowercase{h}}}}_{[c']}}}^\perp$ ] .
for any basis @xmath1848 of @xmath1849}}}}^\perp$ ] with @xmath1850 and @xmath1851 , we have @xmath1852 similarly for any @xmath1790-orthonormal basis @xmath1853 of @xmath1854}}}^\perp$ ] . because @xmath1824 , this implies @xmath1855 . because @xmath1156 $ ] , we conclude that @xmath1836 , as desired .
m. cowling , sur les coefficients des reprsentations unitaires des groupes de lie simple , in : p. eymard , j. faraut , g. schiffmann , and r. takahashi , eds . :
_ analyse harmonique sur les groupes de lie ii _ ( sminaire nancy - strasbourg 197678 ) , lecture notes in math .
# 739 , springer , new york , 1979 , pp .
132178 .
r. ellis and m. nerurkar , enveloping semigroup in ergodic theory and a proof of moore s ergodicity theorem , in : j. c. alexander . , ed .
, _ dynamical systems ( college park , md , 198687 ) _ , lecture notes in math .
# 1342 , springer , new york , 1988 , pp .
172179 .
m. gromov , asymptotic invariants of infinite groups , in : g. a. niblo and m. a. roller , eds .
, _ geometric group theory , vol .
2 _ ( sussex , 1991 ) , london math . soc .
lecture notes # 182 , cambridge univ . press , cambridge , 1993 , pp
. 1295 .
r. howe , a notion of rank for unitary representations of the classical groups , in : a. fig talamanca , ed . : _ harmonic analysis and group representations _ , ( cime 1980 ) , liguori , naples , 1982 , pp .
223331 .
t. kobayashi , discontinuous groups and clifford - klein forms of pseudo - riemannian homogeneous manifolds , in : b. rsted and h. schlichtkrull , eds . , _ algebraic and analytic methods in representation theory , _ academic press , new york , 1997 , pp .
99165 .
f. labourie , quelques rsultats rcents sur les espaces localement homognes compacts , in : p. de bartolomeis , f. tricerri and e. vesentini , eds . , _ manifolds and geometry _
, symposia mathematica , v. xxxvi , cambridge u. press , 1996 . | let @xmath0 be a closed , connected subgroup of a connected , simple lie group @xmath1 with finite center .
the homogeneous space @xmath2 has a _ tessellation _ if there is a discrete subgroup @xmath3 of @xmath1 , such that @xmath3 acts properly discontinuously on @xmath2 , and the double - coset space @xmath4 is compact .
note that if either @xmath0 or @xmath2 is compact , then @xmath2 has a tessellation ; these are the obvious examples .
it is not difficult to see that if @xmath1 has real rank one , then only the obvious homogeneous spaces have tessellations .
thus , the first interesting case is when @xmath1 has real rank two .
in particular , r. kulkarni and t. kobayashi constructed examples that are not obvious when @xmath5 or @xmath6 . h. oh and d. witte constructed additional examples in both of these cases , and obtained a complete classification when @xmath7 .
we simplify the work of oh - witte , and extend it to obtain a complete classification when @xmath8 .
this includes the construction of another family of examples .
the main results are obtained from methods of y. benoist and t. kobayashi : we fix a cartan decomposition @xmath9 , and study the intersection @xmath10 .
our exposition generally assumes only the standard theory of connected lie groups , although basic properties of real algebraic groups are sometimes also employed ; the specialized techniques that we use are developed from a fairly elementary level .
# 1 latex@info#1 font@info#1 |
let @xmath8 be a smooth , nonorientable , compact surface of genus @xmath9 with @xmath10 boundary components and @xmath11 punctures . if @xmath10 and/or @xmath11 is zero then we omit it from the notation . if we do not want to emphasise the numbers @xmath12 , we simply write @xmath7 for a surface @xmath8 . recall that @xmath13 is a connected sum of @xmath9 projective planes and @xmath8 is obtained from @xmath14 by removing @xmath10 open disks and specifying a set @xmath15 of @xmath11 distinguished points in the interior of @xmath7 .
let @xmath16 be the group of all diffeomorphisms @xmath17 such that @xmath18 is the identity on each boundary component and @xmath19 .
by @xmath20 we denote the quotient group of @xmath16 by the subgroup consisting of the maps isotopic to the identity , where we assume that the isotopies fix @xmath15 and are the identity on each boundary component .
@xmath20 is called the _ mapping class group _ of @xmath7 .
the mapping class group of an orientable surface is defined analogously , but we consider only orientation preserving maps .
usually we will use the same letter for a map and its isotopy class .
in contrast to mapping class groups of orientable surfaces , the nonorientable case has not been studied much .
the first significant result is due to lickorish @xcite , who proved that the mapping class group of a closed nonorientable surface is generated by dehn twists and a so called crosscap slide ( or a y homeomorphism ) .
later this generating set was simplified by chillingworth @xcite , and extended to the case of punctured surfaces by korkmaz @xcite .
korkmaz also computed the first homology group of the mapping class groups of punctured nonorientable surfaces @xcite .
it is also known that the group @xmath21 is generated by involutions @xcite . at first glance
it seems that it should be possible to derive some properties of @xmath20 from the properties of the mapping class group of its orientable double cover .
surprisingly , although it is known that @xmath20 is isomorphic to the centraliser of some involution in the mapping class group of the double cover of @xmath7 ( see @xcite ) , this idea has not led to any significant results .
one of the most fundamental properties of the mapping class group is that it acts on the set @xmath22 of isotopy classes of circles . in the case of an orientable surface
this observation leads to the most powerful tools in the study of mapping class groups .
for example the set @xmath22 has simple structures of a simplicial complex , which lead to definitions of complexes of curves .
this idea was the basic tool in finding a presentation of the mapping class group and also in obtaining some descriptions of its ( co)homology groups ( cf @xcite and references there ) .
another example is the extension of the action of the mapping class group on @xmath22 to the action on equivalence classes of measured foliations .
this idea leads to the thurston theory of surface diffeomorphisms ( cf @xcite ) . in either of these examples
, it is of fundamental importance to understand the action of generators of @xmath20 on a single circle . throughout this paper , we concentrate on a very basic result in this direction , namely on the well known formula for the intersection number @xmath23 which holds for any two circles @xmath1 and @xmath2 on an orientable surface and any integer @xmath3 ( cf proposition 3.3 of @xcite ) .
our first result provides a formula for the action of a twist on a nonorientable surface , similar to ( cf theorem [ tw : index ] ) . to be more precise
, we show that for generic two sided circles @xmath1 and @xmath2 on @xmath7 such that @xmath24 , and any integer @xmath25 , we have @xmath26 where @xmath27 are nonnegative integers depending only on the mutual position of @xmath1 and @xmath2 . as an application of this result
, we prove in section [ sec : twist ] some algebraic properties of twists on nonorientable surfaces .
finally , in section [ sec : centr ] we show that up to a finite number of exceptions , the centraliser of the subgroup generated by the twists is equal to the center of @xmath28 and is generated by @xmath10 boundary twists ( cf theorem [ tw : center ] ) .
we end the paper with an appendix , which contains the description of two rather exceptional mapping class groups , namely those of a klein bottle with one puncture and of a klein bottle with one boundary component .
all the results presented are well known in the orientable case ( cf @xcite ) , but for nonorientable surfaces they are new .
moreover , we believe that the methods we develop will contribute to a further study of mapping class groups of nonorientable surfaces .
since the strategy we follow is similar to that in @xcite , in some cases we omit detailed proofs referring the reader to the above article .
by a _ circle _ on @xmath7 we mean an oriented simple closed curve on @xmath29 , which is disjoint from the boundary of @xmath7 .
usually we identify a circle with its image . if @xmath30 and @xmath31 are isotopic , we write @xmath32 . if two circles @xmath1 and @xmath2 intersect , we always assume that they intersect transversely . according to whether a regular neighbourhood of a circle is an annulus or a mbius strip
, we call the circle _ two sided _ or _
one sided _ respectively .
we say that a circle is _ essential _ if it does not bound a disk disjoint form @xmath15 , and _ generic _ if it bounds neither a disk with fewer than two punctures nor a mbius strip disjoint from @xmath15 .
notice that the nonorientable surface @xmath8 admits a generic two sided circle if and only if @xmath33 with @xmath34 and @xmath35 .
following @xcite we will say that circles @xmath1 and @xmath2 _ cobound a bigon _ if there exists a disk whose boundary is the union of an arc of @xmath1 and an arc of @xmath2 .
moreover , we assume that except the end points , these arcs are disjoint from @xmath36 .
for every two circles @xmath1 and @xmath2 we define their _ geometric intersection number _ as follows : @xmath37 in particular , if @xmath1 is a two sided circle and @xmath38 then @xmath39 .
the following proposition ( cf proposition 3.2 of @xcite ) provides a very useful tool for checking if two circles are in a minimal position ( with respect to @xmath40 ) .
[ prop : bigon ] let @xmath1 and @xmath2 be essential circles on @xmath7
. then @xmath41 if and only if @xmath1 and @xmath2 do not cobound a bigon .
let @xmath1 be a two sided circle . by definition ,
a regular neighbourhood of @xmath1 is an annulus @xmath42 , so if we fix one of two possible orientations of @xmath42 , we can define the _ dehn twist _ @xmath0 about @xmath1 in the usual way .
we emphasise that since we are dealing with nonorientable surfaces , there is no canonical way to choose the orientation of @xmath42 .
therefore by a twist about @xmath1 we always mean one of two possible twists about @xmath1 ( the second one is then its inverse ) . by a _
boundary twist _ we mean a twist about a circle isotopic to a boundary component .
if @xmath1 is not generic then the dehn twist @xmath0 is trivial .
we will show that the converse is also true ( cf corollary [ prop : tw : inf ] ) .
other important examples of diffeomorphisms of a nonorientable surface are the _ crosscap slide _ and the _ puncture slide_. they are defined as a slide of a crosscap and of a puncture , respectively , along a one sided circle ( for precise definitions and properties see @xcite ) .
for the rest of this section let us fix two sided generic circles @xmath1 and @xmath2 such that @xmath41 . by a _ segment _ of @xmath2 ( with respect to @xmath1 ) we mean any unoriented arc @xmath43 of @xmath2 satisfying @xmath44 .
similarly we define an _
oriented segment_. if @xmath43 is an oriented segment , by @xmath45 we mean the segment equal to @xmath43 as an unoriented segment but with reversed orientation , and by @xmath46 the unoriented segment determined by @xmath43 .
we call a segment @xmath43 of @xmath2 _ one sided _ [ _ two sided _ ] if the union of @xmath43 and an arc of @xmath1 connecting @xmath47 is a one sided [ two sided ] circle .
an oriented segment is one sided [ two sided ] if the underlying unoriented segment is one sided [ two sided ] .
oriented segments @xmath48 and @xmath49 ( not necessarily distinct ) of @xmath2 are called _ adjacent _ if both are one sided and there exists an open disk @xmath50 on @xmath29 with the following properties : 1 .
@xmath51 consists of the segments @xmath48 , @xmath49 of @xmath2 and the arcs @xmath52 , @xmath53 of @xmath1 ; 2 .
@xmath50 is disjoint from @xmath54 ( see figure [ fig : adj ] ) . ]
[ uw : rigid : disk ] let @xmath55 be oriented segments such that @xmath43 is adjacent to @xmath56 through a disk @xmath50 and @xmath57 is adjacent to @xmath58 through @xmath59 . then since @xmath50 and @xmath59 are disjoint from @xmath54 , either @xmath60 or @xmath61 .
in particular if @xmath62 and @xmath63 then @xmath61 .
oriented segments @xmath64 are called _ joinable _ if there exist oriented segments @xmath65 such that @xmath66 , @xmath67 and @xmath68 is adjacent to @xmath69 for @xmath70 .
unoriented segments are called adjacent [ joinable ] if they are adjacent [ joinable ] as oriented segments for some choice of orientations .
[ uw : dwa : el : adj ] observe that if @xmath43 is a segment of @xmath2 then there are at most two segments of @xmath2 adjacent to @xmath43 ( one on each side of @xmath43 ) .
we now define a graph @xmath71 , which will help us to measure how much @xmath72 differs from @xmath73 ( cf formula ) .
the vertices of @xmath71 correspond to one sided unoriented segments of @xmath2 .
if we have two segments which are adjacent through the disk @xmath50 , we join the vertices corresponding to these segments by an edge ( labelled @xmath50 ) .
so in particular , we do not exclude the possibility that there are multiple edges or loops .
observe that segments @xmath64 are joinable if and only if the corresponding vertices of @xmath71 can be connected by a path . having the above definitions
, we could formulate the relationship between the action of a twist and the intersection number .
[ tw : index ] let @xmath1 and @xmath2 be two sided generic circles and let @xmath27 be the numbers of vertices in the connected components of @xmath71 . then for every integer @xmath25
@xmath74 the rest of this section is devoted to the proof of the above theorem .
the idea of the proof is very simple : construct the circle @xmath75 , perform all obvious reductions of @xmath76 and count them , finally prove that there are no further reductions .
however the details of the proof are quite involved , and we first need some preparations . for two oriented joinable segments @xmath43 and @xmath56 define the _ distance _ between @xmath43 and @xmath56 to be the minimal @xmath77 such that there exist oriented segments @xmath65 with @xmath66 , @xmath67 and @xmath68 adjacent to @xmath69 for @xmath70 . the following three lemmas , which contain the crucial properties of joinable segments , will be proved simultaneously .
[ lem:1 ] if @xmath43 is an oriented segment of @xmath2 then @xmath43 and @xmath45 are not joinable .
[ lem:00 ] let @xmath43 and @xmath56 be oriented , joinable segments of @xmath2 at distance @xmath77 , and let @xmath65 be oriented segments such that @xmath66 , @xmath67 and @xmath68 is adjacent to @xmath69 for @xmath70 .
then @xmath78 if @xmath79 .
[ rem : adj : seg ] let @xmath80 be oriented segments of @xmath2 such that @xmath81 is adjacent to @xmath82 through a disk @xmath83 for @xmath70
. moreover , assume that @xmath84 and the distance between these two segments is equal to @xmath77 .
then @xmath85 for @xmath79 , and the interior @xmath50 of @xmath86 is an open disk with the following properties : 1 .
@xmath51 consists of the segments @xmath87 , @xmath88 of @xmath2 and the arcs @xmath89 , @xmath90 of @xmath1 ; 2 .
@xmath91 ; 3 .
each of the sequences @xmath92 and @xmath93 is strictly monotone with respect to some orientation of @xmath1 ( cf figure [ fig : nn01 ] ) . ]
first observe that we have the implications : 1 .
lemma [ lem:1 ] @xmath94 lemma [ lem:00 ] , 2 .
lemma [ lem:00 ] @xmath94 lemma [ rem : adj : seg ] .
in fact , in order to prove ( a ) , let oriented segments @xmath65 of @xmath2 be as in lemma [ lem:00 ] .
since @xmath95 ( by the definition of joinability ) and the sequence @xmath65 is minimal with respect to @xmath77 , @xmath96 for @xmath79 . moreover , by lemma [ lem:1 ] , we have @xmath97 for @xmath79 . to prove ( b ) , observe that by lemma [ lem:00 ] , @xmath98 for @xmath99 . hence by remark [ uw : rigid : disk
] , @xmath85 for @xmath79 and one can think of @xmath50 as the interior of a standard rectangle ( obtained by gluing all @xmath100 s along common boundary components ) with two opposite sides glued to @xmath1 .
now it is clear that @xmath50 satisfies conditions ( 1)(3 ) above .
observe that the proofs of the above implications preserve distance , in the sense that if lemma [ lem:1 ] is true for segments of distance @xmath101 ( i.e @xmath43 and @xmath45 are not joinable with distance @xmath101 ) , then lemma [ lem:00 ] is also true for segments of distance @xmath101 .
similarly for implication ( b ) . the rest of the proof will be by induction ( simultaneous for all three lemmas ) on the distance between joinable segments .
suppose first that @xmath102 .
we will prove lemma [ lem:1 ] ; lemmas [ lem:00 ] and [ rem : adj : seg ] will follow by implications ( a ) and ( b ) above . if @xmath43 is adjacent to @xmath45 then there exists an open disk @xmath50 with boundary consisting of @xmath43 , @xmath45 and two arcs of @xmath1 connecting @xmath47 .
the best way to think about such a situation is that we have a rectangle ( corresponding to @xmath50 ) with two opposite sides glued by an orientation reversing map ( these sides correspond to @xmath43 and @xmath45 ) .
what we get is a mbius strip with @xmath1 as the boundary circle , which is a contradiction , since @xmath1 is generic .
let @xmath103 , and assume that lemmas [ lem:1][rem : adj : seg ] are true for joinable segments of distance less than @xmath77 . by implications ( a ) and ( b ) it is enough to show that @xmath43 and @xmath45 are not be joinable with distance @xmath77 .
suppose that oriented segments @xmath65 of @xmath2 are such that @xmath68 is adjacent to @xmath69 for @xmath104 , @xmath105 and the distance between @xmath106 and @xmath107 is equal to @xmath77 . if @xmath108 then @xmath106 and @xmath107 would have distance @xmath109 contrary to @xmath103 . hence @xmath110 and we can apply lemma [ rem : adj : seg ] to the segments @xmath106 and @xmath111 .
let @xmath112 be an open disk provided by that lemma and let @xmath113 be a disk given by adjacency of @xmath111 and @xmath105 . by lemma [ lem:00 ] , @xmath114 for @xmath79 , @xmath115 , hence if we assume that @xmath116 , then remark [ uw : rigid : disk ] and the construction of @xmath112 implies that @xmath113 is a disk given by adjacency of @xmath117 and @xmath111 ( this is because this is the only disk composing @xmath112 which has @xmath111 as a boundary component ) .
but this is impossible since @xmath118 ( otherwise the distance between @xmath106 and @xmath107 would be less than @xmath77 ) .
therefore @xmath119 and we claim that @xmath120 is a mbius strip with boundary equal to @xmath1 , which leads to a contradiction , since @xmath1 is generic .
in fact , @xmath50 is obtained from a rectangle ( corresponding to @xmath121 ) by identifying its two opposite sides ( corresponding to @xmath106 and @xmath107 ) by an orientation reversing map and then gluing the remaining side to @xmath1 .
since @xmath1 is two sided , we have the notion of _ being on the same side of @xmath1 _ for germs of transversal arcs starting at the points of @xmath1 . in particular , if @xmath122 is an end point of a segment @xmath43 and @xmath123 of @xmath56 then by _ @xmath122 and @xmath123 being on the same side of @xmath1 _ , we mean that the germs of @xmath43 and @xmath56 starting at @xmath122 and @xmath123 respectively are on the same side of @xmath1 .
[ uw:1 ] initial [ terminal ] points of oriented joinable segments of @xmath2 are on the same side of @xmath1 .
in fact , otherwise there would exist a path , arbitrary close to @xmath1 , connecting points on different sides of @xmath1 which is disjoint from @xmath1 ( cf lemma [ rem : adj : seg ] ) .
[ lem:2 ] let @xmath43 and @xmath124 be oriented segments such that @xmath56 begins at the terminal point of @xmath43 .
then @xmath43 and @xmath56 are not joinable .
suppose @xmath43 and @xmath56 are joinable . then @xmath43 and @xmath56 are one sided and by lemma [ uw:1 ] , the initial points of @xmath43 and @xmath56 are on the same side of @xmath1 .
hence the initial and the terminal points of @xmath43 are on different sides of @xmath1 . since @xmath43 and @xmath56 are joinable , by lemma [ rem : adj : seg ] , there exists a disk @xmath50 with boundary consisting of @xmath43 , @xmath56 and arcs of @xmath1 connecting the initial point of @xmath43 with the terminal point of @xmath43 and the terminal point of @xmath43 with the terminal point of @xmath56 . in order to imagine possible configurations of @xmath1 , @xmath43 and @xmath56 , it is convenient to think of a rectangle with two opposite sides @xmath43 and @xmath56 such that the remaining sides are glued to different sides of @xmath1 in such a way , that @xmath43 and @xmath56 are one sided and the terminal point of @xmath43 coincides with the initial point of @xmath56 .
there are two possibilities to do it ( see figure [ fig : rn03 ] ) : either the initial point of @xmath43 is between the end points of @xmath56 , or the terminal point of @xmath56 is between the end points of @xmath43 ( the third possibility , that the initial point of @xmath43 and the terminal point of @xmath56 coincide , is impossible since @xmath2 is generic ) . and
@xmath1 lemma [ lem:2 ] ] geometrically , it is quite clear that the situation shown in figure [ fig : rn03 ] is not possible .
in fact , this figure implies that @xmath2 , , winds infinitely many times along the core of a mbius strip ( it can not turn back , because @xmath1 and @xmath2 do not cobound a bigon , and since everything is smooth there is no risk of pathologies ) . in order to have a more formal argument , recall that lemma [ rem : adj : seg ] implies that @xmath125 consists of @xmath77 segments @xmath80 of @xmath2 such that @xmath126 and @xmath127 . in particular , each of the arcs @xmath89 and @xmath128 of @xmath129 contains @xmath77 points of @xmath2 . but
this is impossible since either @xmath130 and @xmath131 ( figure [ fig : rn03](i ) ) , or @xmath132 and @xmath133 ( figure [ fig : rn03](ii ) ) . by a _
double segment _ of @xmath2 we mean an unordered pair of two different oriented segments of @xmath2 which have the same initial point .
clearly each point of @xmath36 determines exactly one double segment , so in particular , there are @xmath40 double segments .
two double segments are called _ joinable _ if there exists an oriented segment @xmath43 in the first double segment and @xmath56 in the other such that @xmath43 and @xmath56 are joinable .
[ biarc ] suppose @xmath134 .
then for each double segment @xmath122 there exits a double segment @xmath135 which is not joinable to @xmath122 .
assume that every double segment is joinable to @xmath122 .
let @xmath136 be oriented segments forming @xmath122 . since @xmath134 , @xmath137
let us adopt the notation of consecutive segments of @xmath2 as in figure [ fig : ans02 ] .
lemma [ biarc ] ] we have the following relationships : * @xmath11 and @xmath138 are joinable : this is because by lemma [ lem:1 ] , @xmath107 is not joinable to @xmath106 and by lemma [ lem:2 ] , it is not joinable to @xmath138
. therefore @xmath11 must be joinable either to @xmath106 or @xmath138 ( since we assumed that every double segment is joinable to @xmath122 ) .
lemma @xmath139 implies that it is joinable to @xmath138 .
* the initial and terminal points of @xmath106 are on the same side of @xmath1 : this follows by lemma [ uw:1 ] , from joinability of @xmath11 and @xmath138 .
* @xmath56 and @xmath106 are joinable : this is because @xmath140 is joinable neither to @xmath138 ( lemma [ lem:1 ] ) nor @xmath106 ( lemma [ lem:2 ] ) , and @xmath56 is not joinable to @xmath138 ( lemma [ lem:2 ] ) . * the initial and terminal points of @xmath138 are on the same side of @xmath1 : this follows by lemma [ uw:1 ] , from joinability of @xmath56 and @xmath106 . *
@xmath10 and @xmath138 are joinable : this is because @xmath141 is joinable neither to @xmath138 ( lemma [ uw:1 ] ) nor @xmath106 ( because @xmath56 is joinable to @xmath106 ) , and @xmath10 is not joinable to @xmath106 ( lemma [ uw:1 ] ) .
figures [ fig : ans04](i)(iii ) show reconstruction of @xmath1 and @xmath2 due to the above properties ( here , as in the proof of lemma [ lem:2 ] , one should think of joinability as a rectangle with two edges glued to @xmath1 ) .
lemma [ biarc ] ] let @xmath142 be an annulus with sides : @xmath143 , the arc of @xmath1 connecting the initial point of @xmath138 with the terminal point of @xmath11 and the arc of @xmath1 connecting the terminal points of @xmath10 and @xmath138 , i.e. @xmath142 is the shaded region in figure [ fig : ans04](iv ) .
clearly this figure implies that @xmath2 winds infinitely many times along the core of @xmath142 . more formally , as in lemma [ lem:2 ] ,
lemma [ rem : adj : seg ] implies that each of the two sides of @xmath142 contained in @xmath1 contains the same number of points of @xmath36 , which is impossible .
recall that a _ cycle _ in a graph with the set of vertices @xmath144 is any sequence of different edges @xmath145 , where @xmath146 .
[ prop : acyclic ] every vertex in @xmath71 has degree at most @xmath109 .
moreover @xmath71 is a forest , i.e. it does not contain cycles ( in particular there are neither loops nor multiple edges ) .
the first statement follows from remark [ uw : dwa : el : adj ] .
suppose that there is a cycle in @xmath71 .
by lemma [ lem:1 ] , this means that there exists a sequence @xmath65 of oriented segments of @xmath2 such that @xmath147 and @xmath68 is adjacent to @xmath69 through a disk @xmath83 for @xmath70 .
moreover , since every vertex has degree at most @xmath109 , our cycle is simple ( i.e. all its vertices are different ) , hence @xmath114 for @xmath79 , @xmath148 .
suppose first that @xmath149 , i.e. there exists a loop in @xmath71 and @xmath112 is a disk given by adjacency of @xmath106 to itself .
now think of @xmath150 as obtained by the following construction : identify two opposite sides ( corresponding to @xmath106 ) of a rectangle ( corresponding to @xmath50 ) this gives us an annulus , and then we have to glue the remaining sides to @xmath1 .
there are two possibilities to do it and we obtain either a torus or a klein bottle .
the first case is not possible since @xmath106 is one sided and in the second case @xmath151 ( because there is only one isotopy class of generic two sided circles on a klein bottle cf corollary [ cor : klein : circ ] ) . if @xmath152 , since @xmath114 for @xmath79 , @xmath148 and @xmath153 , we have @xmath154 for @xmath155
( cf remark [ uw : rigid : disk ] ) . therefore if @xmath50 is an open disk obtained by applying lemma [ rem : adj : seg ] to the segments @xmath106 and @xmath111 , then @xmath156 .
hence we can complete the reasoning as in the case @xmath149 , but with @xmath157 .
the following proposition shows that @xmath71 could be defined not only for circles @xmath158 but for their isotopy classes .
since we will not use this result we skip its proof .
let @xmath159 be two sided circles on @xmath7 such that @xmath160 , @xmath161 and @xmath162 .
then @xmath71 is isomorphic to @xmath163 .
the theorem is trivial if @xmath39 , so assume that @xmath164 .
let @xmath42 and @xmath165 be oriented regular neighbourhoods of @xmath1 and @xmath2 respectively such that @xmath166 is a regular neighbourhood of @xmath54 .
define also @xmath167 to be a collar neighbourhood of @xmath2 and let @xmath168 be the boundary component of @xmath169 different from @xmath2 .
in particular , @xmath2 and @xmath168 are disjoint , isotopic and @xmath170 .
the set @xmath171 consists of @xmath172 disjoint 4gons .
we can label their vertices by @xmath173 for @xmath174 in such a way that the following conditions are satisfied : 1 .
@xmath175 and @xmath176 are arcs of @xmath2 and @xmath168 , respectively ; 2 .
the orientation of the 4gon @xmath177 , induced by a cyclic ordering of vertices agrees with the orientation of @xmath42 ( see figure [ fig : tetra ] ) . and
@xmath169 ] let also @xmath179 and let us adopt the convention that unless otherwise stated the arc @xmath175 ( or @xmath180 ) means this of the two arcs of @xmath2 ( or @xmath168 ) with end points @xmath181 ( or @xmath182 ) which is contained in @xmath42 .
outside @xmath42 the twist @xmath0 acts as the identity , so the circle @xmath183 has the following properties : 1 .
outside @xmath42 , @xmath184 is equal to @xmath168 ; 2 .
each arc of @xmath185 , circles @xmath186 times around @xmath42 . due to the above properties ,
each of the @xmath187 arcs @xmath180 of @xmath184 crosses @xmath2 in @xmath188 points ( see figure [ fig : cons ] ) . in particular @xmath189
observe that the notation is chosen in such a way that every time @xmath184 enters the neighbourhood @xmath42 through a point @xmath190 , it crosses @xmath175 ( cf figure [ fig : cons ] ) . and
@xmath2 ] now we are going to define a class of circles which contains @xmath184 and is closed under certain deformations ( defined later ) .
suppose @xmath192 is a circle contained in @xmath166 and such that outside @xmath42 , @xmath192 consists of @xmath187 disjoint arcs each of which is disjoint from @xmath2 and has end points on different components of @xmath193 , where @xmath122 is the component of @xmath194 containing this arc . moreover , if we identify @xmath42 with @xmath195 $ ] so that each of the arcs @xmath196 has constant first coordinate , then we assume that each arc of @xmath197 is monotone with respect to the first coordinate .
we then call @xmath192 _ admissible_. observe that in particular , @xmath168 and @xmath184 are admissible
. we can extend the notion of [ oriented ] segments to any admissible circle @xmath192 , defining them to be components of @xmath198 .
moreover , since @xmath42 is orientable , we could speak about one sided [ two sided ] segments . in addition
every oriented segment of @xmath192 uniquely determines an oriented segment of @xmath2 , so we have a well defined map from the set of oriented segments of @xmath192 into the set of oriented segments of @xmath2 .
denote this map by @xmath199 .
clearly @xmath199 induces a map between the sets of unoriented segments of @xmath192 and of @xmath2 . by abuse of notation
we also use the symbol @xmath199 for this map .
we will use the notion of an [ oriented ] segment of @xmath2 starting at @xmath200 ( or @xmath201 ) meaning the [ oriented ] segment of @xmath2 with initial point @xmath202 which passes through @xmath200 ( or @xmath201 ) .
the constructed circle @xmath184 , in contrast to the oriented case , usually does not satisfy @xmath203 .
however we will define two types of reduction which will enable us to deform , in a very controlled way , @xmath184 into a circle @xmath204 satisfying @xmath205 .
let @xmath43 be an oriented one sided segment of @xmath2 with initial point @xmath202 and terminal point @xmath206 .
let @xmath56 be an oriented segment of an admissible circle @xmath192 such that @xmath207 .
suppose further that if we orient the arc @xmath208 of @xmath192 complementary to @xmath56 in such a way that it has the same initial and terminal points as @xmath56 then the first intersection point of @xmath209 lying on @xmath208 is on @xmath175 and the last one is on @xmath210 .
moreover , assume that between @xmath43 and @xmath56 there are no other segments of @xmath192 ( see figure [ fig : red1 ] ) .
now we see that we can push the segment @xmath56 of @xmath192 towards @xmath43 to obtain a circle @xmath211 isotopic to @xmath192 such that @xmath212 .
observe also that @xmath211 is admissible and @xmath213 ( modulo the identification of @xmath56 and its deformation @xmath58 ) .
we call every such deformation of @xmath192 a _ reduction of type i_. suppose now that we have two adjacent oriented segments @xmath214 of @xmath2 with initial points @xmath215 and terminal points @xmath216 respectively .
let @xmath56 be an oriented segment of an admissible circle @xmath192 such that @xmath217 .
suppose further that if @xmath208 is constructed as above then the first intersection point of @xmath209 lying on @xmath208 is on @xmath175 and the last one is on @xmath218 .
moreover , assume that between @xmath43 and @xmath56 there are no other segments of @xmath192 ( see figure [ fig : red2 ] ) . as before we can push @xmath56 towards @xmath43 obtaining a circle @xmath211 isotopic to @xmath192 such that @xmath212 .
observe also that @xmath211 is admissible and if we denote by @xmath58 the segment resulting from the deformation of @xmath56 , we have @xmath219 whereas @xmath217 . outside the segments @xmath56 for @xmath199 and @xmath58 for @xmath220 these two maps are identical .
we call every such deformation of @xmath192 a _ reduction of type ii_. let @xmath221 be an oriented segment of @xmath2
. then by the construction of @xmath184 , there exists a unique oriented segment @xmath56 of @xmath184 with @xmath222 .
suppose further that @xmath43 and @xmath56 determine a reduction of type i ( see figure [ fig : red1 ] ) .
we claim that if @xmath58 is obtained from @xmath56 by performing this reduction , then @xmath43 and @xmath58 do not allow a reduction of type i. in fact , if we orient the arc @xmath223 complementary to @xmath58 in such a way that it has the same initial and terminal points as @xmath58 , then the first point of @xmath224 on @xmath223 can not be on @xmath175 ( because before @xmath223 goes back to @xmath175 it must intersect each @xmath225 , for @xmath226 ) .
therefore if @xmath65 are all segments of @xmath2 which determine a reduction of type i ( with respect to @xmath184 ) , and @xmath227 is the circle obtained form @xmath184 by performing these @xmath77 reductions , then @xmath227 admits no further reductions of type i. in order to determine the number @xmath77 , observe that if @xmath228 and @xmath229 then a segment @xmath43 of @xmath2 is one sided if and only if both its end points are in @xmath230 or @xmath231 ( see figure [ fig : ans08 ] ) .
moreover , if @xmath57 is a segment of @xmath184 with @xmath232 then @xmath43 and @xmath57 determine a reduction of type i if and only if both end points of @xmath43 are in @xmath230 ( cf figure [ fig : cons ] ) . ]
observe also that the above characterisation of one sided and two sided segments of @xmath2 in terms of their end points , shows that the number of one sided segments with end points in @xmath230 is equal to the number of one sided segments with end points in @xmath231 ( they alternate along @xmath2 ) .
since the total number of one sided segments of @xmath2 is @xmath233 , where @xmath27 are the numbers of vertices in the connected components of @xmath71 , we see that @xmath234 .
therefore @xmath235 notice also that @xmath227 is admissible and @xmath236 ( up to the obvious identification of domains ) . by proposition [ prop : acyclic ] ,
every connected component @xmath237 of @xmath71 is a path , so every such component determines a sequence @xmath238 ( @xmath239 being the number of vertices in @xmath237 ) of segments of @xmath2 such that @xmath68 is adjacent to @xmath69 for @xmath240 .
therefore we see that @xmath237 determines @xmath241 reductions of @xmath227 of type ii ( see figure [ fig : concomp ] ) . and @xmath227 corresponding to @xmath237 ] let @xmath204 be the circle obtained by performing all these reductions , in particular @xmath243 we claim that @xmath204 admits no further reductions .
first observe that every reduction of type ii is determined by two adjacent segments @xmath43 and @xmath57 of @xmath2 . by the construction of @xmath204
at least one of the preimages @xmath244 or @xmath245 is empty .
hence @xmath204 admits no reduction of type ii . in order to show that @xmath204 admits no reduction of type i
, suppose that @xmath221 is an oriented one sided segment of @xmath2 and @xmath56 is an oriented segment of @xmath204 such that @xmath246 .
denote also by @xmath208 the arc complementary to @xmath56 oriented in such a way that @xmath202 is its initial point . by the construction of @xmath204
it is clear that the first point of of @xmath247 lying on @xmath208 can not be on @xmath175 ( because before @xmath208 goes back to @xmath175 it must intersect each @xmath225 for @xmath226 ) .
to finish the proof it is enough to show that @xmath248 , i.e. that @xmath2 and @xmath204 do not cobound a bigon ( cf proposition [ prop : bigon ] ) .
denote by @xmath249 the points of intersection of @xmath204 with the boundary of @xmath42 corresponding to the points @xmath250 of @xmath184 ( i.e. the segment @xmath251 of @xmath204 is the deformation of @xmath180 ) .
before we proceed further we need the following corollary of lemma [ biarc ] .
[ lem : seg ] if @xmath134 , then for every @xmath252 , each of the arcs @xmath175 and @xmath253 intersects the set @xmath254 ) . for a fixed @xmath255 , by lemma [ biarc ]
, there exists a double segment @xmath122 which is not joinable to the double segment determined by @xmath175 .
assume that @xmath122 is determined by an arc @xmath210 for some @xmath256 ( see figure [ fig : lem : seg ] ) .
] now one should think that @xmath204 is obtained from @xmath184 by unwinding along adjacent segments .
since no oriented segment of @xmath122 is joinable to an oriented segment of the double segment @xmath175 , @xmath257 can not unwind along @xmath175 and vice versa ( crossed disks in figure [ fig : lem : seg ] represent obstacles to the unwinding ) .
hence the arc @xmath258 intersect @xmath175 and @xmath253 intersect @xmath210 .
suppose that @xmath2 and @xmath204 cobound a bigon @xmath50 with vertices @xmath259 .
assume that @xmath260 is on the arcs @xmath175 , @xmath258 and @xmath261 on @xmath218 , @xmath262 .
first consider the case @xmath263 .
since there are at least two points of intersection @xmath264 , we have @xmath265 .
observe that since there are no one sided segments of @xmath2 , we have @xmath266 .
now there are two possibilities : either the arc @xmath267 is contained in @xmath42 or it passes through @xmath268 and @xmath269 .
similarly , there are two possibilities for the position of the second arc of @xmath51 ( see figure [ fig : case_m=1 ] ; observe that cases ( ii ) and ( iii ) are possible only if @xmath270 ) . in each of these cases ,
the path indicated in figure [ fig : case_m=1 ] ( running along @xmath2 ) connects points on different sides of @xmath51 and is disjoint from @xmath271 , a contradiction . and
@xmath2 if @xmath272 therefore we further assume that @xmath134 .
now the proof splits into two cases .
case 1 : @xmath273 .
there are two arcs of @xmath2 joining @xmath260 and @xmath261 : the one contained in @xmath42 and another one , running through @xmath200 and @xmath201 .
observe that only the first one can be a boundary arc of the bigon @xmath50 .
this follows from the observation that by the assumption @xmath134 and by lemma [ lem : seg ] , @xmath260 and @xmath261 can not be consecutive on the second of these arcs . now depending on the position of the second boundary arc of @xmath50 , we deduce that either the boundary of @xmath50 is a nonseparating circle , or @xmath1 and @xmath2 cobound a bigon see figure [ fig : case_i = k ] . ]
case 2 : @xmath274 .
since @xmath260 and @xmath261 are consecutive on @xmath2 , there exists an arc of @xmath2 with end points @xmath260 and @xmath261 whose interior is disjoint from @xmath204 . by lemma [ lem : seg ] , this arc outside @xmath42 is equal to the segment @xmath43 of @xmath2 with end points @xmath202 and @xmath275 .
if @xmath276 then @xmath1 and @xmath2 would cobound a bigon ( see figure [ fig : case_i_neq_k1 ] ) , so @xmath277 .
, @xmath276 ] since @xmath260 and @xmath261 are consecutive on @xmath204 , there exists an arc of @xmath204 with end points @xmath260 and @xmath261 whose interior is disjoint from @xmath2 . as before , by lemma [ lem : seg ] , this arc outside @xmath42 is equal to the segment @xmath56 of @xmath204 with one end point @xmath279 or @xmath280 and the other one @xmath281 or @xmath282 see figure [ fig : case_i_neq_k2 ]
. , @xmath277 ] first observe that @xmath56 is one sided .
in fact , otherwise the arc @xmath284 of @xmath285 corresponding to @xmath56 at one end would intersect @xmath286 and at the other would not ( see figure [ fig : ans08 ] ) .
this would imply that @xmath246 and @xmath287 at the same time a contradiction . from this
it follows that @xmath43 is also one sided ( otherwise @xmath271 would be one sided ) .
therefore the existence of @xmath50 implies that if @xmath288 then @xmath286 and @xmath43 are adjacent and we can perform a reduction of type ii
see figure [ fig : case_i_neq_k3](i ) . in case
@xmath246 it is possible to perform a reduction of type i see figure [ fig : case_i_neq_k3](ii ) . hence in both cases we obtain a contradiction with the construction of @xmath204 . , @xmath277 ] [ rem : generic ] observe that if @xmath151 , i.e. if a regular neighbourhood of @xmath54 is orientable , then @xmath289 and the proof of theorem [ tw : index ] works without the assumption that @xmath1 and @xmath2 are generic ( lemma [ lem : seg ] is not needed ) . in particular , if @xmath290 , theorem [ tw : index ] implies that @xmath291 , hence @xmath1 is generic .
[ prop : inter ] let @xmath25 be an integer
. then 1 .
@xmath292 if @xmath293 ; 2 .
@xmath294 ; 3 .
@xmath295 . in particular ,
if @xmath296 , then @xmath297 .
the assertion is trivial for @xmath39 , so let @xmath164 . by the proof of theorem [ tw : index ] , @xmath298 . if @xmath293 then @xmath299 , which proves ( 1 ) .
the inequality ( 2 ) follows form ( 1 ) if @xmath293 , and if @xmath300 then by lemma [ lem : seg ] , @xmath301 . in order to prove ( 3 ) ,
first observe that if @xmath27 are as in the statement of theorem [ tw : index ] , then by ( 2 ) , @xmath302 ( otherwise @xmath303 ) .
therefore @xmath71 is not a path with @xmath304 vertices , i.e. @xmath305 .
now it is an easy exercise that if @xmath1 and @xmath2 are positive integers , such that @xmath306 , then @xmath307 .
hence @xmath308 by theorem [ tw : index ] , the above inequality yields ( 3 ) .
[ sec : twist ] [ lem : curv ] assume that @xmath309 if @xmath310 , and let @xmath311 be generic two sided circles on @xmath312 such that : 1 .
@xmath313 , if @xmath79 ; 2 .
@xmath314 is isotopic neither to @xmath315 nor to @xmath316 if @xmath79 ; 3 .
none of the @xmath314 is isotopic to a boundary component of @xmath7 ; 4 .
if we cut @xmath7 along those @xmath314 which separate @xmath7 , then every component homeomorphic to a klein bottle with one boundary component is disjoint from @xmath30
. then there exists a generic two sided circle @xmath2 such that @xmath317 if @xmath318 , and @xmath319 .
let @xmath320 be the connected component of @xmath321 containing @xmath30 .
clearly it is enough to construct a generic two sided circle @xmath2 on @xmath320 such that @xmath319 . now
if we cut @xmath320 open along @xmath30 we obtain a surface @xmath322 with two more boundary components ; denote them by @xmath323 and @xmath324 .
moreover , if we fix the orientation of @xmath30 , then @xmath323 and @xmath324 inherit orientations from @xmath30 .
consider two cases : case 1 : _
@xmath322 is connected . _
if @xmath322 is nonorientable then we can represent @xmath322 as a connected sum of an oriented surface and a number of projective planes . now depending on mutual orientations of @xmath323 and @xmath324 , one of the two curves indicated in figure [ const_b_nonort ] projects to a two sided circle @xmath2 on @xmath320 ( the shaded disk in figure [ const_b_nonort ] represents a crosscap on @xmath322 ) .
if @xmath322 is nonorientable ] if @xmath322 and @xmath320 are orientable then the construction of @xmath2 is shown in figure [ const_b_ort](i ) . if @xmath322 is orientable and @xmath320 is nonorientable then either @xmath322 has genus at least @xmath325 or by assumption , it has at least two punctures / boundary components different from @xmath323 and @xmath324 .
the construction of @xmath2 in each of these cases is shown in figures [ const_b_ort](ii ) and [ const_b_ort](iii ) respectively . if @xmath322 is orientable ] case 2 : _ @xmath322 is disconnected .
_ let @xmath326 and @xmath327 be connected components of @xmath322 such that @xmath328 is a boundary component of @xmath329 , @xmath330 .
observe that for @xmath330 , we have : 1 .
if @xmath329 has genus @xmath331 then it has at least two punctures / boundary components different from @xmath328 ( since @xmath30 is generic and isotopic neither to a boundary component of @xmath7 nor to any of @xmath332 , @xmath333 ) ; 2 . if @xmath329 is nonorientable of genus @xmath325 , then it has at least one puncture or a boundary component different from @xmath328 ( since @xmath30 is generic ) ; 3 .
if @xmath329 is orientable of genus at least @xmath325 or nonorientable of genus at least @xmath109 , then @xmath329 is a connected sum of a torus / klein bottle with boundary component @xmath328 and some other surface . therefore , in any case we can construct an arc @xmath334 on each of @xmath329 , @xmath330 , such that projections of @xmath335 and @xmath336 onto @xmath320 give a two sided circle @xmath2 such that @xmath337 ( see figure [ const_b_dis ] ) .
if @xmath322 is disconnected ] observe that in each case , @xmath338 , hence by remark [ rem : generic ] , @xmath2 is generic .
[ rem : separ ] it is easy to prove that if @xmath1 is a generic two sided circle on a klein bottle with one boundary component , which is not isotopic to the boundary , then @xmath1 is nonseparating ( cf lemma [ app_a_lem1 ] and its proof ) .
therefore , if @xmath311 are generic two sided circles on a closed surface , satisfying all assumptions of the above lemma but ( 4 ) , then @xmath30 is nonseparating and @xmath315 is separating for some @xmath339 . for each nonorientable surface @xmath7 ,
let @xmath340 be the surface obtained by gluing a torus minus a disk to each boundary component of @xmath7 .
then @xmath340 has no boundary and the following , very useful , property ( cf proposition 3.5 of @xcite ) : [ prop : subsur ] suppose @xmath1 and @xmath2 are circles on @xmath7
. then @xmath1 is isotopic to @xmath2 in @xmath7 if and only if they are isotopic in @xmath340 .
it is an easy observation that the only nontrivial dehn twist on a klein bottle has order @xmath109 .
the next proposition shows that except for this example , dehn twists about disjoint circles generate a free abelian group ( we will use this result in the proof of theorem [ tw : center ] ) .
[ prop : z : free ] suppose @xmath341 if @xmath310 , and let @xmath311 be generic two sided circles on @xmath312 such that @xmath313 , if @xmath79 , and @xmath314 is isotopic neither to @xmath315 nor to @xmath316 if @xmath79 . consider the function @xmath342 defined by @xmath343 then @xmath18 is an injective homomorphism .
clearly @xmath18 is a homomorphism , so let us prove that it is injective
. for @xmath7 being a klein bottle with puncture the assertion follows from proposition [ app_a_prop1 ] and theorem [ app : tw : kl : punct ] , so assume that @xmath7 is not a klein bottle with a puncture .
suppose @xmath344 in @xmath6 .
clearly @xmath344 also in @xmath345 , where @xmath340 is the surface described above . without loss of generality
we can assume that the first @xmath77 of the circles @xmath311 are separating on @xmath340 and the remaining ones does not .
we will prove by induction on @xmath255 that @xmath346 .
suppose that @xmath347 for @xmath348 . by proposition [ prop : subsur ] and by remark [ rem : separ ] , the circles @xmath349 and the surface @xmath340 satisfy the assumptions of lemma [ lem : curv ] .
therefore , there exists a circle @xmath2 on @xmath340 , such that @xmath350 for @xmath351 and @xmath352 . now if @xmath353 , proposition [ prop : inter ] yields @xmath354 hence @xmath346 , which completes the proof .
[ prop : tw : inf ] suppose @xmath341 if @xmath310 , and let @xmath1 be a generic two sided circle on @xmath312 .
then the dehn twist @xmath0 has infinite order in @xmath20 .
if @xmath1 and @xmath2 are circles on an orientable surface and @xmath355 nonzero integers , then it is known ( cf theorem 3.143.15 of @xcite ) that : 1 .
@xmath356 if and only if @xmath38 and @xmath357 ; 2 .
@xmath358 if and only if @xmath39 .
moreover if @xmath359 then 1 .
@xmath360 if and only if @xmath293 and @xmath361 .
clearly the , , if clauses of ( 1 ) and ( 2 ) also hold on nonorientable surfaces . in case ( 3 ) observe that if @xmath1 and @xmath2 are two sided circles on a nonorientable surface and @xmath362 , then the regular neighbourhood of @xmath54 is a torus with a boundary component , so it makes sense to assume that the orientations of regular neighbourhoods @xmath363 of @xmath1 and @xmath2 agree . under this assumption also the , , if clause of ( 3 ) holds ( it is just a braid relation ) .
the next three propositions show that under some obvious assumptions , also the , , only if clauses of the above statements hold on nonorientable surfaces . [ prop : twist : not ] let @xmath1 and @xmath2 be generic two sided circles on @xmath312 .
if @xmath364 and @xmath77 are nonzero integers such that @xmath356 , then @xmath1 is isotopic to @xmath365 . moreover if @xmath341 for @xmath310 and the orientations of regular neighbourhoods of @xmath1 and @xmath2 are such that @xmath366 , then @xmath357 . if @xmath367 then by proposition [ prop : inter ] , @xmath368 and @xmath369 .
therefore @xmath39 .
suppose @xmath1 is not isotopic to @xmath365 . by proposition [ prop : subsur ] , @xmath1 is not isotopic to @xmath365 in @xmath340 . since on a klein bottle or a klein bottle with one puncture
there is only one generic two sided circle ( up to isotopy and reversing orientation cf proposition [ app_a_prop1 ] and corollary [ cor : klein : circ ] ) , @xmath340 is neither of these surfaces .
now either @xmath370 or @xmath371 satisfy the assumptions of lemma [ lem : curv ] ( cf remark [ rem : separ ] ) . in the first case we have a circle @xmath184 on @xmath340 such that by proposition [ prop : inter ] , @xmath372 and @xmath373 . hence @xmath374 .
the second case can be handled in exactly the same way .
the last statement follows form corollary [ prop : tw : inf ] . [ prop : com ] let @xmath1 and @xmath2 be generic two sided circles on @xmath7 .
if @xmath364 and @xmath77 are nonzero integers such that @xmath358 , then @xmath39 .
the assertion is trivial for a klein bottle ( cf corollary [ cor : klein : circ ] ) , so assume that @xmath7 is not a klein bottle . if @xmath375 then @xmath376 . by corollary [ prop : tw :
inf ] , @xmath377 , hence @xmath184 is generic . therefore , by proposition [ prop : twist : not ] , @xmath184 is isotopic to @xmath378 .
if we assume that @xmath290 then by proposition [ prop : inter ] , @xmath379 a contradiction .
[ prop : barid ] let @xmath1 and @xmath2 be generic two sided circles on @xmath312 such that @xmath380 . if @xmath364 and @xmath77 are nonzero integers such that @xmath360 , then @xmath293 .
moreover , if @xmath41 and the orientations of regular neighbourhoods of @xmath1 and @xmath2 agree , then @xmath361 . since there is only one isotopy class of circles on a klein bottle ( cf corollary [ cor : klein : circ ] ) , @xmath341 if @xmath310 .
moreover , we can assume that @xmath41 .
if @xmath39 then @xmath381 , and by proposition [ prop : twist : not ] , @xmath382 . therefore @xmath290 .
if @xmath383 then @xmath384 hence by corollary [ prop : tw : inf ] , @xmath184 is generic , and by proposition [ prop : twist : not ] , @xmath385 .
this gives @xmath386 therefore by inequality ( 3 ) of proposition [ prop : inter ] , @xmath387 this easily implies that @xmath388 .
suppose first that @xmath389 .
if @xmath151 then by theorem [ tw : index ] , @xmath390 contrary to .
therefore @xmath71 has two vertices .
this implies that the regular neighbourhood of @xmath54 is a klein bottle with two boundary components , i.e. the configuration of @xmath158 and their regular neighbourhood is as in the left hand part of figure [ fig : braid ] . and
@xmath383 lemma [ prop : barid ] ] the right hand part of the same figure shows the circle @xmath383 ( strictly speaking , since we have ambiguity in the choice of orientations of neighbourhoods of @xmath1 and @xmath2 , it is one of the possible circles @xmath383 ; however , other choices yield similar pictures ) .
in particular @xmath391 and by theorem [ tw : index ] , @xmath392 contradicting .
therefore @xmath293 , and by statement ( 1 ) of proposition [ prop : inter ] , @xmath393 hence by , @xmath394 .
now if the orientations of neighbourhoods of @xmath1 and @xmath2 agree , then @xmath395 .
therefore by and proposition [ prop : twist : not ] , @xmath361 .
[ sec : pants ] to decompose nonorientable surfaces , beside standard pantalons of type i iii ( see figure [ fig : pantalons ] and section 4 of @xcite ) , we need two more surfaces , namely a mbius strip @xmath396 with one puncture and a mbius strip @xmath397 with an open disk removed , which we call _ ( nonorientable ) skirts of type i and ii _ , respectively .
the mapping class group of a skirt of type ii is generated by the boundary twists , and the mapping class group of a skirt of type i is generated by a puncture slide @xmath398 such that @xmath399 is a twist about the boundary component .
a decomposition of a surface into pantalons and skirts , is called a _
p - s decomposition_. a p - s decomposition is called _ separating _ if each of the circles defining it , is a boundary of two different pantalons / skirts .
the reason for considering separating p - s decompositions is that if we know that some diffeomorphism @xmath400 preserves such a decomposition , then from the structure of the mapping class groups of pantalons / skirts we can conclude that @xmath401 is of very simple form .
this remark will be of great importance in the proof of theorem [ tw : center ] . for precise definitions of pantalons of type
i iii and a pantalon decomposition , we refer the reader to @xcite . the euler characteristic of a pantalon or skirt is @xmath402 .
therefore , none of the surfaces : @xmath403 with @xmath404 , @xmath405 nor @xmath406 admits a p - s decomposition .
apart from these exceptions , every nonorientable surface admits a p - s decomposition .
let us now specify some such decompositions : * _ projective plane @xmath403 with @xmath407 and @xmath408 .
_ if @xmath7 is not a skirt , we cut off a mbius strip with a puncture / boundary component ; there remains a disk with at least two punctures / boundary components and we can decompose it into pantalons .
the resulting decomposition is separating . * _ klein bottle @xmath409 with @xmath410 .
_ we cut @xmath7 into pantalons of type ii and iii . if @xmath407 , this decomposition is separating . * _ nonorientable surface @xmath8 with @xmath411 odd .
_ we decompose @xmath7 into one skirt of type ii and some number of pantalons of type ii and iii ( see figure [ p - s : odd ] ; the shaded disk represents a crosscap ) . if @xmath412 or @xmath410 , this decomposition is separating .
+ * _ nonorientable surface @xmath8 with @xmath413 even .
_ we decompose @xmath7 into two pantalons of type iii and some number of pantalons of type ii and iii ( see figure [ p - s : even ] ) .
this decomposition is separating .
+ in the following , by a p - s decomposition we will always mean one of the decompositions listed above .
[ sec : centr]let @xmath414 be the _ twist subgroup _ of @xmath20 , i.e. the subgroup of @xmath20 generated by all dehn twists . in the case of a closed nonorientable surface , @xmath414 is a subgroup of index @xmath109 ( cf @xcite ) . if @xmath415 then the index of @xmath416 is @xmath417 ( cf corollary 6.2 of @xcite ) .
we now compute the centraliser @xmath418 .
this will allow us to compute the centre of @xmath20 .
observe that , as in the orientable case , boundary twists are central in @xmath20 .
we are going to prove that up to a finite number of exceptions , there are no other elements of @xmath20 which centralise @xmath414 . before we state the main theorem , we need to consider some exceptional cases .
the mapping class group of a projective plane and of a mbius strip is trivial ( cf theorem 3.4 of @xcite ) .
the projective plane with one puncture , a skirt of type i , a skirt of type ii and the klein bottle have abelian mapping class groups ( respectively @xmath419 ) so @xmath420 is equal to @xmath6 .
if @xmath7 is a projective plane with two punctures , then @xmath421 is trivial , so @xmath420 is equal to @xmath6 i.e. to the dihedral group @xmath422 ( of order @xmath423 ) ( cf corollary 4.6 of @xcite ) . if @xmath7 is a klein bottle with one puncture or a klein bottle with one boundary component then the description of @xmath424 follows from corollaries [ app_wn1 ] and [ app_wn2 ] .
we will now examine the case of a closed nonorientable surface @xmath7 of genus @xmath425 .
@xmath7 has a double cover @xmath426 which is an orientable surface of genus @xmath109 .
suppose that @xmath426 is embedded in @xmath427 in such a way that it is invariant under reflections across the @xmath428 , @xmath429 and @xmath430 planes ( see figure [ genus3:1 ] ) . and its double cover ]
let @xmath431 be the isotopy class of a diffeomorphism @xmath432 induced by the central symmetry of @xmath427 : @xmath433 . by @xcite , @xmath6 is isomorphic to the quotient group @xmath434 , where @xmath435 is the centraliser of @xmath436 in @xmath437 .
moreover , this isomorphism is induced by the projection @xmath438 , where @xmath439 is the orbit space , which from now on will be our model for @xmath7 .
let @xmath440 be the hyperelliptic involution , i.e. the isotopy class of a diffeomorphism induced by the half turn about the @xmath441-axis ( see figure [ genus3:1 ] ) .
since @xmath442 is central , it induces a central element @xmath443 of @xmath444 .
observe that if @xmath1 is a circle on @xmath7 as in figure [ genus3:1 ] , then @xmath445 and @xmath443 preserves the local orientation of a neighbourhood of @xmath1 .
now let @xmath17 represent an element of the centraliser @xmath446 of the twist subgroup . since @xmath447 , proposition [ prop : twist : not ] implies that @xmath448 is isotopic to @xmath449 .
so we can assume that @xmath450 .
moreover , @xmath18 must preserve the local orientation of a neighbourhood of @xmath1 .
therefore we can choose @xmath451 such that @xmath452 is isotopic to the identity in a neighbourhood of @xmath1 .
now we can cut @xmath7 open along @xmath1 , and conclude from the mapping class group of the skirt of type ii that @xmath453 for some integer @xmath77 .
now by lemma [ lem : curv ] , there exists a two sided generic circle @xmath2 such that @xmath290 ( see figure [ genus3:1 ] ) .
since @xmath454 commutes with the twist @xmath455 , proposition [ prop : com ] implies that @xmath456 .
therefore we have proved the following : let @xmath7 be a closed nonorientable surface of genus 3 .
the centre of @xmath6 is equal to the centraliser @xmath420 of the twist subgroup and is generated by the involution @xmath443 .
now we are ready to prove the general result concerning the centraliser @xmath420 .
[ tw : center ] suppose that @xmath457 and let @xmath458 be the boundary curves of @xmath312 .
then the centraliser @xmath420 of the twist subgroup is equal to the centre of @xmath6 .
moreover , @xmath420 is generated by @xmath459 and is isomorphic to @xmath460 .
since the proof follows the lines of the proof of theorem 5.6 of @xcite , we only sketch it .
the isomorphism @xmath461 follows from proposition [ prop : z : free ] , so it is enough to prove that @xmath462 .
let @xmath311 be the circles defining a separating p - s decomposition of @xmath7 ( cf section [ sec : pants ] ) . if @xmath463 then @xmath464 , hence by proposition [ prop : twist : not ] , @xmath465 for @xmath466 .
now we can assume that in fact @xmath467 ( cf proposition 3.10 of @xcite ) , hence @xmath18 permutes pantalons / skirts .
first suppose that @xmath18 interchanges some two components @xmath326 and @xmath327 of the p - s decomposition .
if @xmath326 and @xmath327 are both pantalons of type ii glued along a circle @xmath315 , then the remaining boundary curves @xmath468 and @xmath469 must be glued together .
in fact , since @xmath467 for every @xmath255 , and @xmath18 interchanges @xmath470 and @xmath471 , we have @xmath472 .
therefore @xmath7 is a klein bottle with two punctures .
observe that @xmath18 must preserve orientations of regular neighbourhoods of @xmath315 and @xmath473 and this is possible only if @xmath18 does not interchange @xmath326 and @xmath327 . if @xmath326 and @xmath327 are both pantalons of type iii , then as before we argue that @xmath7 is a closed nonorientable surface of genus @xmath474 and @xmath18 do not interchange @xmath326 and @xmath327 . observe that by our choice of p - s decompositions ( cf section [ sec : pants ] ) , and since @xmath7 is nonorientable , @xmath326 and @xmath327 can be neither a pantalon of type i nor a skirt .
therefore we proved that @xmath18 maps every pantalon / skirt onto itself .
moreover , since @xmath18 centralises their boundary twists , the restriction of @xmath18 to each pantalon preserves its orientation .
if @xmath33 then the p - s decomposition of @xmath7 contains neither a pantalon of type i nor a skirt of type i ( cf section [ sec : pants ] ) . by the structure of the mapping class groups of pantalons of type ii / iii and skirt of type ii
, @xmath475 now for each fixed @xmath476 , by lemma [ lem : curv ] , there exists a generic two sided circle @xmath2 such that @xmath477 and @xmath350 , for @xmath256 . therefore @xmath455 commutes with @xmath478 for @xmath256 .
it also commutes with all @xmath479 and with @xmath18 , hence it commutes with @xmath480 . by proposition
[ prop : com ] , this yields @xmath481 , which completes the proof in this case .
it remains to consider the case of @xmath7 being a projective plane with @xmath482 punctures @xmath483 . for each @xmath484
, there exists a two sided circle @xmath485 on @xmath7 , such that @xmath486 has two components , one of which is a mbius strip with a puncture @xmath487 , and the other is a disk with @xmath488 punctures . since @xmath18 centralises the twist about @xmath184 , it satisfies @xmath489 . because the components of @xmath486 are not homeomorphic , @xmath18 can not interchange them , so in particular , @xmath490 . therefore @xmath18 fixes @xmath15 pointwise .
now the p - s decomposition of @xmath7 consists of one skirt of type
i and a number of pantalons of type i and ii ; assume that @xmath314 is the circle which cuts off the skirt .
since @xmath18 preserves the orientation of every pantalon , by the structure of the mapping class groups of the pantalons and of the skirt , @xmath491 where @xmath398 is a boundary slide . then @xmath492 now a similar argument as before yields @xmath493 .
suppose @xmath494 .
then the centre of @xmath495 is trivial .
for the rest of this subsection let @xmath496 denote a klein bottle with one puncture @xmath43 .
[ app_a_lem1 ] let @xmath184 be a generic two sided circle on @xmath7 .
then @xmath486 is connected and orientable .
suppose that @xmath486 has two components @xmath326 and @xmath327 .
then both @xmath326 and @xmath327 have exactly one boundary component and one of them has a puncture .
the euler characteristics of @xmath326 and @xmath327 satisfy @xmath497 without loss of generality we can assume that @xmath498 and therefore @xmath499 . if @xmath500 then @xmath326 is a disk , which is impossible since @xmath184 is generic . if @xmath501 then @xmath326 is either a disk with a puncture or a mbius strip .
both cases are impossible . since @xmath502 and @xmath486 has two boundary components and one puncture , if we glue a disk to each of the boundary components and remove the puncture , we obtain a surface of euler characteristics @xmath109 , i.e. the sphere .
therefore @xmath486 is orientable .
[ app_a_lem2 ] if @xmath1 and @xmath2 are generic two sided circles on @xmath7 , then there exists @xmath503 such that @xmath504 . by the previous lemma , @xmath505 and @xmath506 are diffeomorphic as punctured surfaces .
we can choose a diffeomorphism @xmath507 which extends to @xmath508 .
then @xmath509 .
[ app_a_prop1 ] there are exactly two isotopy classes of generic two sided circles on @xmath7 . by lemma [ app_a_lem2 ] ,
it is enough to prove that if we fix some generic two sided circle on @xmath7 then @xmath510 and for any @xmath511 , @xmath448 is isotopic either to @xmath1 or @xmath512 . to prove this ,
let us describe generators of @xmath20 .
following @xcite , represent @xmath7 as the one point compactification of a plane with two crosscaps and a puncture ( see figure [ fig : klein : punct ] ) .
let @xmath513 be closed curves indicated in figure [ fig : klein : punct ] .
in particular , @xmath514 and @xmath192 are one sided , while @xmath515 and @xmath1 are two sided .
define @xmath516 to be the puncture slides along @xmath514 and @xmath192 , and the crosscap slide along @xmath515 respectively . then by theorem 4.9 of @xcite , @xmath20 is generated by @xmath516 and @xmath0 .
it is straightforward to check that @xmath510 , @xmath517 and @xmath518 .
[ cor : klein : circ ] there is exactly one isotopy class of generic two sided circles on a klein bottle @xmath406 .
consider another model of @xmath7 , namely the one shown in figure [ fig : klein : punct2 ] .
define @xmath1 and @xmath514 as shown in the figure , and let @xmath398 be the puncture slide along @xmath514 .
if we cut @xmath7 along @xmath1 , we obtain a cylinder with a puncture .
reflection of this cylinder across the circle parallel to boundary components and passing through the puncture induces a diffeomorphism @xmath519 such that @xmath520 .
[ app : tw : kl : punct ] let @xmath7 be a klein bottle with a puncture and @xmath521 as above . then @xmath6 is the product @xmath522 and is isomorphic to @xmath523 . by proposition [ app_a_prop1 ] ,
if @xmath524 is any diffeomorphism , then @xmath448 is isotopic either to @xmath1 or to @xmath512 , so the subgroup @xmath525 consisting of maps which do not interchange the sides of @xmath1 is of index @xmath109 in @xmath6 .
moreover , @xmath526 and @xmath527 . all maps @xmath528 such that @xmath448 is isotopic to @xmath1 form the subgroup @xmath142 of index @xmath109 in @xmath529 , and @xmath530 .
if @xmath531 is any diffeomorphism then we can assume that @xmath532 and @xmath77 preserves sides of @xmath1 .
if we cut @xmath7 open along @xmath1 , we conclude from the mapping class group of the cylinder that @xmath533 for some @xmath534 .
therefore @xmath529 is generated by @xmath398 and @xmath0 .
since @xmath399 is a twist about the boundary of a mbius strip , @xmath398 is of order @xmath109 .
moreover , @xmath398 reverses the orientation of a regular neighbourhood of @xmath1 , so @xmath535 .
therefore @xmath536 since @xmath537 , @xmath538 , to complete the proof it is enough to show that @xmath0 is of infinite order .
this can be shown by computing the induced homomorphism on homology .
[ app_wn1 ] let @xmath7 be a klein bottle with one puncture and @xmath539 as above .
then the centre of @xmath6 is equal to the group of order @xmath109 generated by @xmath540 .
the centraliser @xmath420 of the twist subgroup is generated by @xmath0 and @xmath540 , and is isomorphic to @xmath541 .
now let @xmath542 denote the klein bottle with one boundary component @xmath2 . observe that if @xmath320 is a klein bottle with a puncture , then the inclusion @xmath543 induces a homomorphism @xmath544 which extends every @xmath545 by the identity on @xmath546 ( see figure [ fig : klein : hole ] ; note that this time the shaded disk does not represent a crosscap but a disk ) .
we claim that the kernel of @xmath547 is generated by the boundary twist @xmath455 on @xmath7 .
in fact , if @xmath548 then @xmath549 in @xmath320 . by proposition 3.5 of @xcite , we also have @xmath549 in @xmath7 . moreover ,
@xmath18 preserves the orientation of a neighbourhood of @xmath1 , so it does not interchange sides of @xmath1 . therefore @xmath18 is induced by a mapping of @xmath550 , hence by the structure of the mapping class group of a pantalon of type iii , @xmath551 .
now @xmath552 and by corollary [ prop : tw : inf ] , @xmath553 . since every diffeomorphism of @xmath7 fixes the boundary , the image of @xmath547 consists of elements of @xmath554 which preserve the local orientation around the puncture on @xmath320 .
all such elements form a subgroup @xmath555 of index @xmath109 , which is generated by @xmath0 and @xmath556 .
observe that we can use the same definitions as for the maps @xmath557 , and @xmath398 to define diffeomorphisms @xmath558 such that @xmath559 .
the problem is that @xmath560 and @xmath561 do not fix the boundary of @xmath7 .
however , if we define @xmath562 where @xmath563 is a half twist about the boundary circle @xmath2 , then @xmath564 fixes the boundary and @xmath565 . now from the exact sequence @xmath566 and easily verifiable relations @xmath567 we obtain the following theorem : let @xmath7 be a klein bottle with one boundary component and @xmath568 as above .
then the mapping class group of @xmath7 is the semidirect product @xmath569 and is isomorphic to @xmath570 .
[ app_wn2 ] let @xmath7 be a klein bottle with one boundary component and @xmath571 as above .
then the centre of @xmath6 is the cyclic group generated by @xmath455 .
the centraliser @xmath420 of the twist subgroup is generated by @xmath0 and @xmath455 , and is isomorphic to @xmath572 .
the author wishes to thank the referee for his / her helpful suggestions . | let @xmath0 be the dehn twist about a circle @xmath1 on an orientable surface .
it is well known that for each circle @xmath2 and an integer @xmath3 , @xmath4 , where @xmath5 is the geometric intersection number .
we prove a similar formula for circles on nonorientable surfaces . as a corollary
we prove some algebraic properties of twists on nonorientable surfaces .
we also prove that if @xmath6 is the mapping class group of a nonorientable surface @xmath7 , then up to a finite number of exceptions , the centraliser of the subgroup of @xmath6 generated by the twists is equal to the centre of @xmath6 and is generated by twists about circles isotopic to boundary components of @xmath7 . |
ngc 4261 is an early - type galaxy in the virgo west cloud , at the distance of 32 mpc ( @xcite ) .
it belongs to a group of galaxies ( @xcite ) that does not show prominent sign of gravitational interactions .
the h - band luminosity of ngc 4261 is @xmath0 corresponding to a dynamical mass of @xmath1 ( using a mass to light ratio of about 1.2 , from @xcite ) , ranking this galaxy among the most massive galaxies in virgo .
the optical spectrum is consistent with a dominant population of old stars ( @xmath2 gyr , @xcite ) and the light distribution is smooth with no evidence of sub - structure ( @xcite ; @xcite ) .
ngc 4261 is a fri radio galaxy ( 3c 270 ) showing two radio jets emanating from a nucleus hosting a supermassive black hole with a mass of about 5 @xmath3 10@xmath4 m@xmath5 ( @xcite ) surrounded by a dusty torus and a nuclear disk ( @xcite ; @xcite ) .
the x - ray nuclear power from the underlying agn has been well studied by _
xmm _ and _ chandra _ : the x - ray luminosity is l@xmath6 erg s@xmath7 , and the x - ray spectrum is a power law absorbed by a column density n@xmath8 @xmath9 .
the source appears to be embedded in diffuse hot gas ( @xmath10 kev ) and shows low amplitude , rapid variability ( @xcite ; @xcite ; @xcite ) .
thanks to the high spatial resolution of _ chandra _ , @xcite ( zhfm03 hereafter ) have discovered the presence of about sixty bright off - nuclear x - ray sources in ngc 4261 , which stand out for their large - scale anisotropic pattern .
the anisotropy extends over an angle of @xmath11 , corresponding to a linear dimension of @xmath12 kpc ( @xmath13 effective core radii large , since @xmath14 kpc for ngc 4261 ; @xcite ) .
zhfm03 interpret this anisotropy as evidence that the x - ray sources are associated with young stellar population , possibly formed in a recent episode of star formation triggered by some dynamical agent along tidal tails . in spite of any optical evidence of either a young population ( the population in the nuclear regions is as old as 15 gyrs ; see @xcite and @xcite ) or of recent merging events , zhfm03 suggest a high mass binary origin for the majority of the sources , based on the significantly higher efficiency of the formation of high mass x - ray binaries relative to the low mass ones .
this interpretation makes ngc 4261 a rather unique example , in clear contrast with _
observations of other early - type galaxies , where the x - ray sources are generally distributed nearly isotropically , trace the dominant ( old ) stellar population , and are believed to belong to the low mass x - ray binary population ( lmxb , e.g. @xcite ; @xcite ; @xcite ) .
in addition , a significant fraction ( @xmath15 20% -70% ) of these lmxbs is associated to globular clusters ( gcs ; @xcite ; @xcite ; @xcite ; @xcite ; @xcite ) , with a preference to gcs with redder color index , i.e. , clusters that are more metal rich ( @xcite ; @xcite ; @xcite ) . in this perspective , the case of ngc 4261 stands out as a remarkable exception worth of further study .
we have therefore chosen to study its case again , in order to deepen our understanding of the the nature of the x - ray sources in external galaxies .
for this reason we have used archive observations of _ chandra _ , of the isaac newton telescope ( int ) , and of the hubble space telescope ( _ hst _ ) ( see table 1 for a summary of the dataset used ) to search for optical counterparts of the x - ray sources , and to study their relation with the gc population of ngc 4261 .
the outline of the paper is as follows .
in @xmath16 we describe the data reduction , in the x - ray and optical bands . in @xmath17
we proceed on studying the properties of the optical point - like sources , and we cross - correlate the optical data with the x - ray data .
we then derive the luminosity function of the x - ray source population . in @xmath18
we investigate on the azimuthal anisotropy of the x - ray sources , and on that of the underlying gc candidate population . in the context of our new findings we discuss in @xmath19
the nature of the sources and their potential in tracing the history of assembly of ngc 4261 in the virgo cluster .
ngc 4261 was observed by _ chandra _
@xcite with the acis - s3 ccd on 2000 may 6 for a total exposure of 37,210 s ( see table [ tab : obs ] for details ) .
the data were collected in 1/2 sub - array mode to avoid the pile - up of the central agn source .
the active field of view is shown in figure [ fig1 ] over - plotted on an optical image of the galaxy .
the data reduction was performed with the package ` ciao ` v. 3.0.2 and ` caldb ` v. 2.2 .
data were cleaned for high background periods due to particle induced flares and the resulting exposure time is of @xmath20 ks .
individual sources were detected using a wavelet - detection algorithm ( _ wavdetect _ in ` ciao ` ) , with the same parameters used by zhfm03 , in three energy bands : full band ( @xmath21 kev ) , soft band ( @xmath22 kev ) and hard band ( @xmath23 kev ) .
the catalog that we obtained includes 54 sources detected above the probability threshold of 10@xmath24 that implies @xmath25 1 false detection due to statistical fluctuations .
the positions of the x - ray sources are shown in figure [ fig2 ] . in table
[ tab : sources ] we give the source number ( column 1 ) , the right ascension and declination j2000 ( column 2 and 3 ) .
count rates ( column 4 ) are converted into unabsorbed luminosities ( @xmath26 kev ) assuming a power - law model with photon index 1.7 and a galactic line of sight column density of n@xmath27= 5.8 @xmath3 10@xmath28 @xmath9 ( @xcite ) .
zhfm03 do not publish a catalog of their x - ray sources , so a comparison of the two source lists is not possible .
however , a visual inspection of their figure 1 indicate consistency .
data relative to the central region of this galaxy has also been analyzed by @xcite and @xcite , who have mostly focused on the agn emission in the x - ray , radio and optical bands .
in particular @xcite detect 5 off - nuclear sources : a , b , c , e coincide with our sources 38 , 33 , 31 , and 14 respectively , and d , which we do not detect ( our source 19 is the closest , at @xmath29 ) . although these authors comment of the fact that they do not find optical counterparts for any of their sources , they do not comment on the possible connection of these sources with the jet , in spite of the evidence that two of them , ( a and e ) are aligned and at opposite ends of it .
we have used archive observations of int and _ hst _
wfpc2 for a systematic search of the optical counterparts of the x - ray sources .
we extracted a deep @xmath30 band image from the int - wide field camera ( wfc ) archive , consisting of three exposures of 500 seconds each , for a total integration time of 1500 seconds .
the image covers the whole galaxy ( the field of view of the int wfc is of about @xmath31 deg@xmath32 ) and has a mean seeing of @xmath33 ( @xcite ) .
the reduction of the images was carried out using standard tasks in the ` iraf ` package .
bias subtraction and flat - field normalization were applied to each exposure using the median of several bias frames and flat - field exposures obtained during the observations .
the three exposures were combined using a median filter , that automatically removes cosmic rays . using reference catalogues of astrometric stars
, we obtained a wcs solution for the whole field having a mean rms of @xmath151 arcsec .
wfpc2 observations cover the central portion of the galaxy , as shown by the white boxes in figure [ fig1 ] .
the south - east ( se ) field was observed using three different filters : f675w , f547 m , f791w while only the f702w filter is available for the north - west ( nw ) field .
the exposure times of the four images are given in table [ tab : obs ] .
for our analysis we started from pipelined processed data .
hst _ wfpc2 and int ,
we first fitted ellipses to the isophote of the galaxy using the task ` ellipse ` from the iraf package , and subtracted the model from the data .
the resulting image was then used to detect sources with sextractor ( @xcite ) .
the last isophote has a major and minor axis of @xmath34 and @xmath35 arcmin respectively .
we improve the relative hst astrometry from @xmath36 to @xmath37 arcsec comparing the overlap region between the nw and se field . on the contrary we did not try to make any improvement in the relative astrometry between chandra and int / hst images .
we note that , for both hst and int images , the model subtraction leaves a quadrupole structure in the central region .
although this region is smaller in the hst data , we exclude a wider area of @xmath150.5 square arcmin , due to the presence of residuals in the int data that could induce spurious results in our comparisons .
this structure was already noticed by @xcite for this object , and was attributed to the boxyness of the galaxy .
using sextractor we extracted point - like objects from the model subtracted images of both int and _ hst _ datasets .
we set a sextractor detection threshold of 2.5 @xmath38 over the local background .
we select as point - like sources objects with the following properties : ( i ) a magnitude error smaller than 0.1 , ( ii ) ellipticity smaller than 0.5 , ( iii ) fwhm between 1 and 4 pixels ( 0.1 - 0.4 arcsec ) for _ hst _ , as suggested by @xcite , and fwhm between 3 and 10 pixels ( 1 - 3 arcsec ) for int . note that for the _
hst_-se field we based our selection upon the f675w filter owing to its longer exposure . as a final step ,
we have checked the sextractor results by visual inspection , and we have removed objects on or close to the ccds edges for the hst images .
for the se field of the _ hst _ observation , we have been able to estimate the color of the sources presented in @xmath39 .
the instrumental magnitudes , provided by sextractor , were converted in the johnson - cousin v , r and i systems using the synthetic transformations given in @xcite . in order to estimate the photometric error due to model subtraction we measured the photometry on the unsubtracted image using local background and apertures from the subtracted image .
the difference in the v - i color of each globular cluster obtained from the two different estimates is shown in figure [ fig3 ] .
the distribution appears fairly gaussian with a mean difference of @xmath40 mag ( see figure [ fig3 ] ) . the total photometric uncertainty is thus given by the combination of the error on the measure of gcs magnitudes ( given by sextractor ) , the error on the zero - point ( taken from @xcite ) and the uncertainty due to the model subtraction ( 0.07 mag , fig.[fig3 ] ) .
this results in a mean error on the v - i color of @xmath15 0.11 mag .
we used f547 m and f791w filters to determine the color index ( v - i ) of the extracted objects .
we select 325 objects detected in each of the three filters with v - i color between 0 and 2 ( @xcite ) , that we consider gc .
their color index and magnitude distributions shown in figure [ fig4 ] are consistent with those of the globular cluster populations observed in other massive early - type galaxies ( @xcite ; @xcite ; @xcite ; @xcite ; @xcite ) , reinforcing our definition .
the color index distribution spreads from v - i @xmath41 to @xmath42 , indicating that both populations of gc ( red and blue ) are present . thus the unimodal distribution observed in the v - i color histogram is probably due to the blending of different gcs populations .
since the color information is restricted to the _ hst_-se field , we can not classify as gcs the sources extracted from the other optical images at the same confidence .
however , we have first considered sources in the region common to all three datasets ( @xmath43 arcmin@xmath32 ) .
the cross - correlation between the three lists indicates that @xmath44% ( 37/38 ) and @xmath45% ( 49/59 ) of the sources in the _
hst_-nw and int field respectively coincide with a gc in the _ hst_-se field .
given this high fraction , we are confident that even the point - like sources from the int dataset are good `` gc candidates '' , in spite of the lower resolution of the instrument . in
what follows , we will use the `` gc candidate '' list ( from the int observation ) , that provides an homogeneous set of optical sources over the whole galaxy . as a final check ,
we have also compared the sky projected distribution of the int and _ hst _ gc candidates . since _ hst_-se and _ hst_-nw fields have different exposure times , we restrict the _ hst _ sample to a magnitude @xmath46 = @xmath47 ( i.e. the magnitude limit of the shallowest field ) .
the comparison is done through an adaptive kernel density analysis .
this technique is a two - step procedure which first applies a pilot smoothing to estimate the local density of gcs , starting from the list of their positions only , and then uses a smoothing window variable in size that decreases with increasing local density . in this way
the statistical noise in low - density regions can be reduced without smearing out the high density peaks @xcite . as shown in figure [ fig5 ] , the iso - density contours from the adaptive kernel analysis are in very good agreement , confirming the reliability of the int gcs dataset .
what is apparent in the figure is the peculiar distribution of the gc population , that shows two main concentrations ne and sw of the nuclear region .
the presence of the same structures in both datasets ensures that it is not a feature derived from the poor subtraction in the central region ( the black area in figure [ fig5 ] ) since the equivalent area in the hst data is negligible . in order to investigate the nature of the x - ray sources we have looked for optical counterparts .
the central source ( n. 29 ) is identified with the agn ( @xcite ; @xcite ; @xcite ) , and sources no .
27 , 43 and 53 with background / foreground galaxies .
the positions of the remaining x - ray sources have been cross - correlated with the gc sample , from the hst fields , with a matching radius of @xmath31 arcsec , and from the int dataset , with a matching radius of @xmath48 arcsec .
these choices take into account only the intrinsic astrometric uncertainties of the three datasets ( see sections 2.1 and 2.2 ) and not the relative uncertainties eventually present between the optical and x - ray observations . source 54 falls outside the model subtracted area , while sources no .
19 , 31 and 33 fall in the central region excluded from our previous analysis .
we found an optical counterpart for 23 of the 46 discrete x - ray sources that we assume all belonging to ngc 4261 .
the identification is indicated in table [ tab : sources ] . to calculate the chance coincidence probability we performed 1000 monte - carlo simulations by shifting the positions of the int and hst optical samples by a random offset and performing again the cross - correlation .
we adopted a shift larger than the search radius but small enough so the sources do not fall outside the distribution of the x - ray sources ( @xmath151.5 arcmin ) .
the probability of finding more than one spurious association is @xmath152.5% and @xmath150.1% in the int and hst sample , respectively . in figure [ fig6 ]
we give a few examples that show the positions of the x - ray sources over - plotted onto the _ hst _ se , _ hst _ nw and int fields .
the figure also demonstrates the good correspondence between the _ hst _ and int counterparts , indicating again that the identifications obtained in the int field not covered by _
hst _ are not severely hampered by the lower resolution of the int image .
note that , where the color information is available , we found that all the x - ray sources inhabit a red ( color index v - i @xmath49 ) gc , see figure [ fig4 ] .
the optical identification with gc candidates for @xmath50% of the x - ray sources suggests a lmxb origin for them ( @xcite ) .
we verified the above statement by comparing the cumulative x - ray luminosity function ( xlf ) of our sources with those derived from observations of other galaxies ( figure [ fig7 ] ) .
in particular , we use the functional forms derived by @xcite and @xcite considered by these authors as a signature of a universal behavior for the hmxb and the lmxb populations , respectively ; namely a single power - law for hmxb , with slope of @xmath51 , and a broken power - law for lmxb , with slopes of @xmath52 and @xmath53 for the high luminosity end . in testing the consistency with these universal functions
, we have derived the xlf using all sources in table [ tab : sources ] , except for the agn ( source no .
29 ) and the 3 interlopers ( sources no .
27 , 43 and 53 ) .
we performed a log - log maximum likelihood fitting procedure for both hmxb and lmxb functional forms assuming fixed slopes .
we considered as free parameter the normalization factors and , in the case of a broken power law , the luminosity of the break .
the resulting xlfs are shown in figure [ fig7 ] .
it is unquestionable that a single power - law can not reproduce the lf of the sources in ngc 4261 .
it is also apparent that the lmxb broken - power law is a good representation of the data given also the value of the break luminosity , that we have left as a free parameter in the fitting procedure , and that is consistent with the gilfanov value of @xmath54 .
this result is consistent with the hypothesis that the x - ray sources in this galaxy are associated to a population of lmxbs .
note that we do not take into account incompleteness at low luminosities @xcite ; however , we point out that at least at the high luminosity end ( where incompleteness should not significantly affect the xlf slope ) the luminosity function is much steeper than the xlf of the hmxbs .
zhfm03 have shown with a kolmogorov - smirnov test that the spatial distribution of the x - ray sources in ngc 4261 is anisotropic . using our list of 50 sources ( sources n. 29 , 27 , 43 and 53 excluded )
we confirm the spatial anisotropy over a scale of about @xmath55 with very high confidence ( p@xmath56% from the kolmogorov - smirnov test ) . in the preceding sections
we have shown that the properties of the x - ray sources in ngc 4261 are consistent with a population of lmxbs .
this however does not explain yet their spatial anisotropy , which is in stark contrast with the smooth isophotal distribution of the optical light .
given the close association with the gc population , which we find also has a peculiar distribution , we have considered the possibility that the anisotropy is nothing but a manifestation of an anisotropy seeded in the gc population . in order to test this hypothesis we have first done a statistical analysis on the spatial distribution of the gc candidates . in order to investigate both the details of the spatial distributions of the x - ray and optical sources separately and their relation we have compared the normalized surface density of sources again with an adaptive kernel density analysis on the two sets of sources . in figure [ fig8 ]
we show the iso - intensity contours extracted from the kernel analysis on the gc candidate sample ( thin line ) and on the x - ray sources ( thick line ) . with this figure as a guide ,
we have defined 6 regions ( see figure [ fig9 ] , right panel ) in which we have simply computed the surface number density of sources for both gcs and x - ray sources , which plot in figure [ fig9 ] ( left panel ) as a function of the region number .
the errors associated with each entry reflects the total number of sources in the region , and gives a direct measure of the significance of the observed region - to - region variations , we also test the correlation between the two datasets using the iraf task ` crosscor ` and we find that the highest correlation between the two datasets is consistent with a null shift .
our combined analysis of the x - ray and optical material available for ngc 4261 shows that the large majority of the x - ray sources in ngc 4261 are accreting lmxbs .
about 50% of these sources have an optical counterpart , that we identify with a gc . where color information is available , the sources are associated with the red ( metal - rich ) population , as already observed in several other galaxies .
further evidence of a lmxb origin comes from the interpretation of the luminosity function , consistent with the `` universal '' distribution proposed by @xcite .
the photometric analysis of the gc population of ngc 4261 carried on a portion of the galaxy ( _ hst_-se field ) shows that the properties of the gcs are similar to those observed in other early - type galaxies ( @xcite ; @xcite ; @xcite ) .
furthermore this similarity extends to the connection between x - ray sources and gc population ( @xcite ) .
in particular we note that the x - ray sources are preferentially associated with the brightest and reddest gcs . as in other ellipticals , we find that the gcs hosting a x - ray source are @xmath157% and that the x - ray sources associated with gcs are @xmath1550% , consistent with the result of @xcite .
it is therefore not surprising that ngc 4261 ranks among those galaxies having a sizable fraction of x - ray sources in gcs .
the peculiarity of ngc 4261 resides mainly in the large scale anisotropy of the x - ray sources .
our adaptive kernel density analysis has highlighted the presence of a peculiar spatial distribution in the gc candidate population , and the existence of a correlation between the spatial anisotropy of the x - ray sources and the main over - densities seeded in the gc system .
to our knowledge , this is the first time that x - ray sources indicate the presence of substructures in the distribution of gc .
a non uniform spatial distribution of the gc system of ngc 4261 may reflect a peculiar history of formation of this galaxy . in the paradigm of hierarchical clustering cosmogony
, elliptical galaxies may be the end result of binary or multiple major mergers among gas - rich progenitors .
major mergers likely redistribute the original gc systems of the progenitor galaxies and trigger the formation of new gcs in the center as well as along the tidal tails of the new object arising from the merger .
this would result in the coexistence of two populations , metal rich and metal poor clusters , that have different photometric properties , spatial distribution and dynamical characteristics , as shown in a recent work by @xcite .
hierarchical build - up also implies a second way by which galaxy can grow , namely accretion of dwarf satellites that would lose their gcs to the primary system .
gcs coming from shredded satellites could also contribute to spatial anisotropies and would be mostly metal - poor if such satellites are similar to nearby dwarfs .
evidence that either of these two mechanisms can account for a large fraction of the gc systems in ellipticals is still lacking ( @xcite ) .
a close inspection of the photometric and kinematical properties of the gcs can in principle help disentangle many aspects of such complex galaxy genesis .
ngc 4261 could be an excellent target for further investigations : combined chandra and hst observations can thus potentially shed light on how early - type galaxies were assembled . in parallel
, we are undertaking a theoretical study running high - resolution simulations of galaxy mergers similar to those carried on in @xcite , including a gc formation algorithm .
we thank the anonymous referee for useful comments that helped us to improve the paper .
we thank marco scodeggio for providing us the implementation of the algorithm used for the adaptive kernel density analysis , and michele bellazzini for sharp comments and suggestions .
this work was supported by the italian ministry of university and research ( miur ) under the national program prin 2003 . | _ chandra _ images of the elliptical galaxy ngc 4261 have revealed an anisotropy in the spatial distribution of the off - nuclear x - ray sources , interpreted by zezas et al .
as evidence of an association with a young stellar population .
our independent analysis of archive x - ray ( _ chandra _ ) and optical ( int and _ hst _ ) observations confirms the anisotropy of the x - ray sources but conducts to a different interpretation for their origin .
we find that nearly half of the x - ray sources are associated to a globular cluster ( optical counterpart ) suggesting that they are accreting low - mass x - ray binaries .
where color index information is available , the x - ray sources are found to reside in red ( metal - rich ) systems .
the luminosity function of the x - ray sources is also consistent with the one drawn from a population of low mass x - ray binaries .
we further investigate the properties of the sample of point - like sources obtained from archival optical images that we suggest are good globular cluster candidates and for which we find that the projected spatial distribution is non - homogeneous .
in addition , we show that the distributions of the optical and x - ray populations are very similar , which leads us to conclude that the spatial anisotropy of the x - ray sources in ngc 4261 is mostly a reflection of the anisotropy of the globular cluster population . |
light bridges ( lbs ) are bright structures in the otherwise dark umbra that often exhibit a granular - like morphology @xcite .
they represent a discontinuity or interruption in the regular umbral field @xcite .
lbs are known to harbor weak and inclined fields @xcite , but their origin and magnetic nature is still a matter of debate @xcite .
one of the reasons that make lbs interesting is their chromospheric activity , seen as surges in h@xmath1 @xcite , localized brightenings in h filtergrams @xcite , and brightness enhancements in the upper chromosphere and transition region @xcite .
the stressed magnetic configuration associated with lbs is perceived to be responsible for this activity , but the precise relationship , if any , is yet to be established .
unfortunately , there is a lack of simultaneous vector magnetic field measurements in the photosphere to diagnose these events . here
we analyze _ hinode _ spectropolarimetric observations and h filtergrams of a sunspot lb in an attempt to relate its chromospheric activity to the photospheric magnetic field .
we discover patches of supersonic downflows in the photospheric layers of the lb and show that some of them are associated with strong h brightness enhancements .
interestingly , the supersonic flows produce spectral signatures never seen before in lbs .
on 2007 may 1 , the leading spot of noaa active region 10953 was observed with _ hinode _
@xcite at a heliocentric angle of 8@xmath2 ( @xmath3 ) .
between 10:46 and 12:25 ut , the _ hinode _
spectropolarimeter @xcite recorded the four stokes profiles of the iron lines at 630 nm with a spectral sampling of 21.55 m , a pixel size of 016 , and an exposure time of 4.8 s per slit position ( normal map mode ) .
the observations were corrected for dark current , flat field , thermal flexures , and instrumental polarization using routines included in the solarsoft package .
simultaneously , the broadband filter imager of _ hinode _ took h filtergrams with a cadence of 1 minute to monitor the chromosphere of the lb .
the effective pixel size of the ca images is 011 .
[ sunspot_image ]
the left panel of figure [ sunspot_image ] shows a continuum map of the spot and the lb . we have inverted the observed stokes profiles using the sir code ( stokes inversion based on response functions ; * ? ? ?
sir computes perturbations in the physical quantities at specific locations across the optical depth grid called _ nodes _ , and then carries out an interpolation to yield values at all grid points . to determine the global structure of the lb and the surroundings
, we performed a one - component inversion setting the magnetic and dynamic parameters to be constant with depth .
the temperature stratification was perturbed with two nodes .
a total of 9 parameters were retrieved from the observed profiles , including height - independent micro- and macro - turbulent velocities and a stray - light factor .
the three components of the vector magnetic field ( strength , inclination , and azimuth ) deduced from the inversion are shown in the second and third columns of figure [ sunspot_image ] .
all the angles are expressed in the local reference frame after a manual disambiguation of the line - of - sight ( los ) azimuths .
as can be seen , the lb is characterized by weaker and more inclined fields than the umbra .
this confirms earlier results by , e.g. , @xcite and @xcite . in the upper half of the lb ,
the magnetic field is parallel to the axis of the bridge .
both photometrically and magnetically , the lb looks like an extension of the penumbra protruding into the umbra .
@xcite detected a horizontal flow along the lb that starts in the adjacent penumbra , demonstrating that the two structures are also connected dynamically . at the lower end of the lb , where the lb fields pointing south encounter sunspot fields oriented toward the north , one observes an isolated region with relatively weak magnetic fields .
in addition , there is a discontinuity in the field azimuth running parallel to the west edge of the lb .
the los velocity map displayed in the third column of figure [ sunspot_image ] reveals the existence of strong , localized downflows in the lb with velocities of up to 4 km s@xmath0 .
interestingly , the downflows occur close to the weak - field region and the azimuth discontinuity described above , i.e. , at positions where the magnetic field changes orientation very rapidly ( fourth column of figure [ sunspot_image ] ) .
some of the downflowing patches coincide with chromospheric h brightness enhancements , as can be seen in figure [ calcium ] .
the filtergram displayed there was taken during the polarimetric scan of the lb and shows a strong h line - core brightening at the position and time of the largest photospheric velocities ) .
these structures show brightenings , but not as intense as those associated with the strongest downflows . ] .
noaa ar 10953 produced many other long - lasting chromospheric plasma ejections on april 29 and 30 @xcite .
the stokes @xmath4 profiles associated with the downflows have two peaks in the red lobe , i.e. , they exhibit a total of three peaks .
hereafter they will be labelled as type 1 . in the lb
one also finds anomalous linear polarization profiles with normal stokes @xmath4 signals which are designated as type 2 .
type 3 profiles are essentially a combination of the other two classes .
examples of these profiles are given in figure [ combo_plots ] , together with their spatial distribution .
milne - eddington - like atmospheres such as the ones used to determine the global structure of the lb can not reproduce the complex shapes of type 1 profiles .
for this reason , the velocities given in section [ sir ] are only approximate .
here we obtain more reliable values with the help of two - component inversions .
are shown in blue .
the arrows indicate the transverse component of the field in the local reference frame for every alternate pixel , after solving the 180@xmath2 azimuth ambiguity .
white and black contours represent fields weaker than 750 g and bright continuum structures , respectively .
the white dashed lines mark thin chromospheric threads whose ends are located in or near the downflowing patches . , scaledwidth=44.0% ] we start with a simple two - component model in which the magnetic field and the los velocity do not vary with height . the results of this inversion are presented in figure 4 .
as can be seen , one of the components has supersonic velocities of @xmath5 km s@xmath0 and the other is essentially at rest .
the model explains the observed profiles in a very satisfactory way , but the assumption of two magnetic components separated horizontally in the resolution element may not be realistic because of the relatively large sizes of the downwflowing patches ( 0.5 - 0.7 arcsec ) .
an alternative scenario is that they are stacked in the vertical direction , one on top of the other . to investigate this possibility we consider a different model atmosphere with two magnetic components .
one of them has height - independent parameters and the other features a discontinuity in the stratifications at a certain optical depth .
the amplitude and the location of the discontinuity are free parameters .
these inversions have been carried out using sirjump , an extension of the sir code ( bellot rubio , in preparation ) .
the results show that the fast component always has velocities exceeding 10 km s@xmath0 . as an example
, figure [ inv_results2 ] displays the stratifications of one such model with the downflows occurring in the lower half of the photosphere .
while the two - component models used in this section differ in complexity , all of them indicate supersonic velocities and magnetic fields of the same polarity .
in fact , the existence of supersonic flows is obvious from the shapes of the observed stokes @xmath4 profiles , which are similar to those emerging from the outer part of the penumbra where the evershed flow exceeds the sound speed .
the inclined red wings of stokes @xmath6 also demonstrate the occurrence of strong velocities .
all these spectral signatures are well reproduced by the two - component inversions . to the best of our knowledge ,
this is the first time that supersonic downflows are found in lbs .
subsonic speeds of @xmath71 km s@xmath0 have been reported earlier by @xcite , @xcite , and @xcite . in regions where the magnetic field is relatively horizontal ,
the supersonic flows are associated with type 3 profiles .
they have similar stokes @xmath6 and @xmath4 spectra as type 1 profiles , but also complex linear polarization signals which can not be seen when the field is more vertical .
we observe anomalous stokes @xmath8 profiles with up to five peaks and signatures of strong redshifts ( figure [ combo_plots ] , left ) .
the multiple peaks indicate that different magnetic components coexist in the resolution element at the position of the downflows .
type 3 profiles are common near the azimuth discontinuity on the west side of the lb , where the lb fields meet the umbral fields ( figure [ combo_plots ] , right ) .
thus , it is likely that these are the magnetic components giving rise to the complex stokes @xmath8 signals .
they may lie horizontally next to each other in the resolution element , or could be stacked one on top of the other . in both cases the large variations of field azimuth that occur on very small spatial scales should facilitate the reconnection of magnetic field lines from the lb and the adjacent umbra , even if they have the same polarity @xcite .
the interaction is expected to produce bidirectional jets , and we speculate that the supersonic downflows observed near the azimuth discontinuity and the weak - field region are the signatures of reconnection jets directed toward the photosphere .
this would agree with @xcite , who believe that the lb is essentially a horizontal , twisted flux tube that reconnects with the umbral field at a low altitude .
however , our measurements can not confirm the reconnection scenario unambiguously because the upward jet components are not always visible in the h images .
indeed , another possibility is that the supersonic downflows are produced by a mechanism similar to that driving the evershed flow at the outer end of penumbral filaments .
we have determined the global magnetic and dynamic properties of a sunspot light bridge from the inversion of stokes profiles observed by _
a one - component model atmosphere with height - independent parameters ( except for the temperature ) has been used to that end .
our analysis reveals patches of strong downflows exceeding 4 km s@xmath0 in the lb , associated with complex stokes @xmath4 spectra having double red lobes .
two - component inversions of the profiles indicate supersonic velocities of 10 km s@xmath0 or more .
a milne - eddington inversion by @xcite of the same lb on the previous day shows only a small downflow patch with velocities of 0.7 km s@xmath0 .
the supersonic downflows occur in regions where the magnetic field of the lb meets sunspot fields with rather different orientations . in those locations
one observes anomalous linear polarization signals , but only when the field is relatively horizontal .
moreover , the supersonic downflows are sometimes associated with transient chromospheric brightenings . the interaction of the lb field and the sunspot fields may create current sheets , leading to magnetic reconnection in the upper photosphere / lower chromosphere .
the strong photospheric flows could represent the downward directed jets generated in this process .
however , such an explanation remains speculative because the associated upward jets are not always seen as chromospheric brightenings .
the discovery of supersonic downflows in lbs has been possible due to the high spatial resolution and stability of _ hinode _ , but their interpretation is still unclear .
additional observations should be performed to investigate the chromosphere and photosphere of lbs simultaneously , in an attempt to detect small brightness enhancements that would confirm the reconnection scenario or favor alternative explanations .
our sincere thanks to the _ hinode _ team for providing the data used in this letter , and to the referee for valuable suggestions .
hinode is a japanese mission developed and launched by isas / jaxa , with naoj as domestic partner and nasa and stfc ( uk ) as international partners .
it is operated by these agencies in co - operation with esa and nsc ( norway ) .
we acknwoledge support by the spanish micinn through projects esp2006 - 13030-c06 - 02 and pci2006-a7 - 0624 , and by junta de andaluca through project p07-tep-2687 . | we report the discovery of supersonic downflows in a sunspot light bridge using measurements taken with the spectropolarimeter on board the _ hinode _ satellite .
the downflows occur in small patches close to regions where the vector magnetic field changes orientation rapidly , and are associated with anomalous circular polarization profiles .
an inversion of the observed stokes spectra reveals velocities of up to 10 km s@xmath0 , making them the strongest photospheric flows ever measured in light bridges .
some ( but not all ) of the downflowing patches are cospatial and cotemporal with brightness enhancements in chromospheric h filtergrams .
we suggest that these flows are due to magnetic reconnection in the upper photosphere / lower chromosphere , although other mechanisms can not be ruled out . |
since the seminal work by vaiana et al .
( 1981 ) , it has become clear that all late - type stars share the same basic coronal characteristics : hot thermal plasma with temperatures around 110 mk covering stellar surfaces , magnetic confinement , the presence of flares , etc .
a systemic investigation indicates that most active stars have x - ray luminosity up to 2 orders of magnitude higher than that of solar corona ( scelsi et al .
2005 ; peres et al . 2004 ) .
present popular assumption for interpretation of such difference can be attributed to their different composition in terms of various kinds of coronal structures ( ranging from the relatively faint and cool structures of the background corona to the very bright and hot flaring regions ) and to the number of x - ray emitting coronal structures present .
the solar corona as observed with the modern x - ray and xuv telescopes on board yohkoh , soho , or trace , is found to be extremely structured , and even in the high angular resolution trace images , it appears to be spatially unresolved fine structure .
yet spatially resolved x - ray observations of stellar coronae are currently not feasible .
some information on the spatial distribution of stellar coronae was inferred from x - ray light curves of suitably chosen stars such as eclipsing binaries ( white et al .
1990 ; schmitt @xmath1 k@xmath10rster 1993 ; g@xmath10del et al .
1995 , 2003 ; siarkowski et al .
1996 ) , but such analyses can only be carried out for very special systems with advantageous geometries , and the actual information derivable from such data is rather limited . another method to infer the structure in spatially unresolved data is spectroscopic measurements of the electron density , that is , x - ray spectra allow us to get the structure information for a various stellar coronae .
nevertheless , the temperature distribution ( or emission measure distribution emd ) and the coronal abundance could be estimated from low - resolution spectra by an application of global fitting approaches . while the previous measurements did not allow measuring density @xmath11 and emission measure _ em _ independently because the information from spectral lines was not available from spectra with the low resolution , such that no emitting volumes _ v _ could be estimated from _
em = @xmath12_. therefore the information about loop size was nt accessible . direct spectroscopic information on plasma densities at coronal temperatures on stars other than the sun firstly became possible with the advent of high - resolution " spectra ( @xmath13 ) obtained by the _ extreme ultraviolet explorer _ ( euve ) which is capable of separating individual spectral lines .
even with this resolution , the available diagnostics have often tended to be not definitive , owing to the poor signal - to - noise ratio ( snr ) of the observed spectra or blended lines .
after the launch of new generation satellites _
chandra _ and xmm-_newton _ , the high - resolution spectra coupled with the large effective area has made it is possible to measure individual lines in the x - ray range for a large sample of stars in the same fashion as x - ray emission lines from the solar corona obtained and analyzed for many years ( doyle 1980 ; mckenzie @xmath1 landecker 1982 ; gabriel et al .
the emissivity of selected lines depends on the density .
some lines may be present only in low - density plasmas , such as the forbidden line in he - like triplet , while other lines may appear only in high - density plasmas ( such as lines formed following excitations from excited levels ) .
ness et al . ( 2002 , 2004 ) systemically investigated the coronal density using the characteristic of he - like triplet for stars with various activity covering inactive and active levels .
for the hot - temperature region , the density is estimated from emission lines of carbon - like iron , and a typical density ranging 10@xmath14-10@xmath15@xmath5 was obtained .
for the low - temperature region , the density can be derived from low-@xmath16 elements with low ionization energies such as c v , which is lower by at least an order of magnitude than that of the hot - temperature region .
a typical electron density ranging 10@xmath1710@xmath18@xmath5 has been derived for solar and solar - like coronae by authors ( audard et al .
2001 , brinkmann et al .
2000 ) . in inactive stars , emission lines of si vii
si xiii have also been clearly detected , and some lines of si x show a high snr in the wavelength range covered by letgs .
in the collision ionization equilibrium ( mazzatto et al . 1998 ) condition , the temperature of peak fractional abundance of si x is very close to that of c v. this means that they form in the same region and share the same electron density .
therefore , the density derived from si x should be comparable to that from he - like c v. in our recent work ( liang et al .
2005 ) , we noticed that the line intensity ratio @xmath3 of si x at the lines at 50.524 and 50.691 originating from @xmath19 transitions is sensitive to the density , whereas it is insensitive to the temperature .
so an application of this ratio @xmath3 in several stellar coronal spectra is performed . in this paper , we derive the electron densities for stars : procyon , @xmath0 cen a@xmath1b , capella and @xmath2 eri using this ratio @xmath3 for the first time , and compare the derived densities with those from he - like c v. the paper is structured as follows : we present our sample and a detailed description of line flux measurements in sect .
a brief description of theory model is introduced in sect .
3 . diagnostic of the electron density and discussions are presented in sect .
the conclusions are given in sect .
the new generation of x - ray telescopes and x - ray spectrometers on board _ chandra _ and xmm-_newton _ has opened the world of spectroscopy to the x - ray astronomy with high - resolution and high effective collecting areas .
spectroscopic measurements can be performed with instruments , such as the reflection grating spectrometer ( rgs ) on board xmm-_newton _ , and the high and medium energy grating ( heg and meg ) spectrometers , and the low energy transmission grating spectrometer ( letgs ) on board _
chandra_. though rgs , heg , and meg provide large effective areas in the high energy region ( 5 to 40 ) with high - resolution , the letgs covers a much larger wavelength range 5175 with the high spectral resolution .
one specific advantage of the letgs resulting from its covered large range of wavelength is that the o vii triplet at around 21 , the c v triplet at around 41 and @xmath19 transition lines at 50 of si x can be measured in one spectrum . in this wavelength range ,
a number of emission lines from highly ionized si ions have been found in the coronal spectra ( raassen et al .
2002 , 2003 ) . here
, we pay a special attention on the emission lines at 50.524 and 50.691 of si x from @xmath19 transitions .
our sample consists of 5 stars including three normal dwarf stars , i.e. , procyon and @xmath0 cen a and b , an active late - type dwarf star , @xmath2 eri , and an active binary system capella .
the properties of our sample , along with obsid and exposure time are summarized in table 1 .
all observations adopt grating of letgs combined with hrc instrument on board _ chandra _ observatory . in case of capella ,
an additional observation ( with obsid=55 ) with acis - s instrument is used . because acis - s instrument has significant energy resolution to separate overlapping spectral orders , letgs+acis - s observations are better choice for determination of the electron density .
however only one observation is available from _ chandra data archival center _ for our sample .
another goal of analysis for the acis - s spectrum of capella is to validate whether there is significant contamination from high - order ( refer to @xmath20 ) spectra around the selected lines ( 50.524 and 50.691 ) .
if line fluxes derived from the two different observations are comparable , no contamination from high - order spectra can be concluded .
this also backs up our analyses for other stars .
reduction of the letgs datasets uses ciao3.2 software with the science threads for letgs / hrc - s observations . for the acis - s spectrum of capella , destreak tool was used to clean the events induced by detector artifacts on ccd 8 ( streaks " ) . for
the extraction of the spectra of @xmath0 cen a&b , similar procedure as employed by raassen et al .
( 2003 ) is adopted here under the ciao environment .
figure 1 shows the spectra with background subtracted for procyon , @xmath0 cen a@xmath1b , capella and @xmath2 eri in the wavelength range of 5055 .
the line fluxes were derived by modelling the spectrum locally with narrow gaussian profiles , in which the instrumental effective area extracted from ciao3.2 calibration files with fullgarf tool has been folded , together with a constant representing the background and ( pseudo)continuum emissions which were determined in the line - free region ( 5152 ) . here , only the positive order spectra has been adopted because there is a plate gap around 52 for the negative spectra .
the average instrumental fwhm of lines is about 0.06 for letgs observations .
the fluxes have been obtained after correction for the effective area . in the fitting , 1@xmath21 uncertainty is adopted to determine statistical errors for the line fluxes , and no wavelength correction has been used to the dispersion relation of the hrc - s / letg .
the error bars determined in this method are slightly underestimated due to the adopted functional forms of the line profiles , which introduce a certain of systematic errors .
table 2 shows the theoretical wavelengths and measured fluxes of the two prominent lines of si x , together with results of fe xvi lines detected in capella and @xmath2 eri stars .
since the centered wavelength for a given line is also a thawed parameter in the fitting , the best - fit result is different for the different stars , yet it is consistent with theoretical identification within 30m .
so only the theoretical wavelength is given for each emission line in this table .
for normal stars such as procyon and @xmath0 cen a@xmath1b , a detailed description has been presented by raassen et al .
( 2002 , 2003 ) and a conclusion that the coronal plasmas are dominated by plasmas with temperatures ranging 13 mk was obtained .
@xmath0 cen a and b were firstly spectrally resolved from observation with letgs instrument , a detailed description about it is presented by raassen et al .
( 2003 ) . in the long wavelength region
@xmath22 , @xmath19 transitions of si x at 50.524 and 50.691 are the prominent lines , which ensures the correct measurements of line fluxes .
in order to measure the fluxes of the two lines correctly , contamination from higher order spectral lines of high ionized fe ions should be taken into account .
fortunately , emission from high charged fe ions is either weak or absent in first order ( and particularly in third order which is damped by a factor @xmath23 ) for inactive stars . in addition
, the temperature of peak emissivity of fe xvi lines ranging this wavelength region is about 4 mk , which is higher than that of x - ray emitting plasma .
such that no emission lines of fe xvi are detected for the three stars .
in order to obtain the line fluxes , two gaussian profiles with the same fwhm value , combining with a constant were used in fitting the spectra in 5051 .
the continuum level is around 1.2@xmath24photon s@xmath25@xmath26@xmath25 for the three normal stars , which is determined by fitting with a constant model in line - free wavelength region 5152 .
the best - fit values of line fluxes are listed in table 2 , along with statistic errors ( with percentage ) derived with 1@xmath21 uncertainty .
the observed wavelengths are within 10 m , when compared with experimental ones .
the reduced @xmath27 are 0.76 , 0.63 and 0.67 , respectively , which indicates present fitting is good for the three normal stars . for active star capella , numerous studies exist in the literatures for the active star capella ( mewe et al .
2001 , canizares et al .
2000 , audard et al . 2001 ) , which reported its properties using observations with different instruments .
most observations adopted this star as test platform to calibrate instruments on board satellites ( brinkmann et al . 2000 ) .
the observation adopting the acis - s instrument eliminates the contamination from high - order spectral lines for the selected lines ( si x ) . in order to validate whether there is contamination from the high - order spectral lines for the selected si x lines , the observation with acis - s instrument
is also analyzed . here ,
the -1 order spectrum is adopted , because no photon has been detected in + 1 order spectrum above 30 .
nevertheless , the contamination from lines of fe xvi around 50 is inevitable , since they have become the strongest lines as shown in fig .
1 . for much more active star ,
the emissions of si x completely disappear . in the fitting , four components ( with same fwhm ) combined with continuum emission ( 5.7@xmath24photon cm@xmath25s@xmath25@xmath25 determined in the line - free region )
were used for the acis - s and hrc spectra in 5051 .
the reduced @xmath27 for acis - s and hrc observations are 0.75 and 0.71 , respectively .
2 shows the best - fit ( dotted line ) to the acis - s spectrum . in this plot ,
the hrc spectrum of capella is also overlapped ( dashed histogram ) for the visual comparison though their effective areas are slightly different .
the best - fit results of the two different observations indicate that there is no contamination from the high - order spectra .
the derived line fluxes are comparable each other , such as for line at 50.367 , the derived fluxes are [email protected] and [email protected]@xmath24phot.@xmath26s@xmath25 for acis and hrc observations , respectively , for line at 50.691 , they are [email protected] and [email protected]@xmath24phot.@xmath26s@xmath25 .
such consistency gives us confidence to use the line fluxes derived from hrc spectra for other stars , though no acis - s observation data is available for our sample . in case of capella , acis - s spectrum was adopted for the analysis in the following . in the spectra observed with acis - s and hrc , the line of si x at 50.691 has been clearly detected . is the observed flux at 50.576 completely from contribution of si x at 50.524 ?
in the available database such as chianti , mekal and aped , we note that fe xvi line at 50.555 may also be the significant source of contribution to the observed flux at 50.576 , because present instrument can not resolve the two emission lines obviously .
the most prominent lines at 50.367 , 54.136 and 54.716 in this range were identified to be emissions of fe xvi by raassen et al .
( 2002 ) , which further confirms that fe xvi contribute to the observed flux at 50.576 .
so an another problem is that how much fraction of the flux at 50.576 comes from si x emission at 50.524 .
the emission measure distribution ( emd ) of capella has been derived from different observations with different instruments with high - resolutions such as heg ( canizares et al .
2000 ) , letgs ( mewe et al . 2001 ) on board _ chandra _ and rgs ( audard et al . 2001 ) on board xmm-_newton_. their results indicate that the emd has a sharp peak around 7 mk , together with a smaller one grouping around 1.8 mk which is necessary to explain the detected o vii he - like triplet and c vi ly@xmath0 line .
argiroffi et al .
( 2003 ) investigated its variability during different phase , and found that the emission is compatible with a constant source . in this work
, we adopt the emd derived from rgs observation by audard et al .
( 2001 ) , with combination of emissivity of fe xvi included in ( astrophysical plasma emission code ) apec model to derive the contributions of fe xvi lines .
we found that fe xvi line at 50.555 contributes @xmath3062% to the total observed fluxes ( [email protected]@xmath31phot.@xmath26s@xmath25 ) .
the remaining flux at this wavelength should come from si x line at 50.524 .
additionally , we notice that there is a shoulder around 50.576 .
so we fit the spectrum again by adding an additional component .
we initially set two artificial components around 50.57 .
the best - fit ( smooth solid line in fig .
2 ) of the spectrum shows a better result than the fitting with four components ( dotted line ) by the visual inspection .
the reduced @xmath27 is also more close to unity which changes from 0.75 to 0.79 .
though the obtained flux ( [email protected]@xmath31phot.@xmath26s@xmath25 ) of fe xvi line at 50.576 is slightly higher than the theoretical prediction ( 1.65 ) , the two values from different procedure are comparable .
the line flux of si x line at 50.524 listed in table 2 , is from the fitting procedure .
for the line flux at 50.367 we further note that about @xmath3073% contribution comes from fe xvi line at 50.350 .
a search of apec shows that the remaining flux may be from contribution of fe xvii lines ( 50.342 and 50.361 ) and si x ( 50.333 , 50.336 and 50.359 ) . as the strongest lines of si x at 50.524 appear to be weak in capella ,
we suggest that the remaining flux of the observed flux at 50.367 is from emission lines of fe xvii . using the emd model derived by audard et al .
( 2001 ) and the emissivity of fe xvii predicted by the apec model , we predict the total fluxes of the four lines are about 0.68@xmath24photon @xmath26 s@xmath25 , which is consistent with the remaining flux ( 0.97@xmath24photon @xmath26 s@xmath25 ) within statistical error .
the detailed analysis for capella reveals that the contamination due to high - order spectral lines is negligible .
however , the emission from fe xvi lines must be taken into account .
four components and a constant value are used to fit the spectrum of @xmath2 eri in wavelength range 5051 .
the continuum emission determined in the line - free region is 1.2@xmath24photon cm@xmath25s@xmath25@xmath25 .
the best - fit ( reduced @xmath27=0.62 ) results of line fluxes are listed in table 2 , along with 1@xmath21 statistical error . for the flux extraction of si x line at 50.524 , the similar procedure as for capella is applied to estimate the contribution of fe xvi lines . using the emissivity of fe xvi extracted from apec and normalization according to the isolated line at 54.134
, we estimate that the contributions of fe xvi to the observed fluxes around 50.350 and 50.555 are about 53% and 39% , respectively . for line flux around 50.550 ,
the remaining flux is from contribution of si x line ( 50.524 ) .
in our previous paper ( liang et al . 2005 ) , we describe properties and application of si x spectrum in detail , and found that six line ratios are sensitive to electron density . one of these ratios @xmath3 has a good application on density diagnostic because the two lines are the strongest lines in the spectrum of collision dominated low - density plasma . for consistency
, we describe the property of the ratio @xmath3 briefly , and the model considered in our calculation .
the ratio @xmath3 is defined as @xmath32 radiative rates , and electron impact excitation as well as de - excitation among 320 levels of si x have been included in the calculation of line intensities at the steady state condition for different electron densities .
the energy level is replaced by experimental value if it is available .
excitation rates by protons of foster , keenan @xmath1 reid ( 1997 ) are also taken into account , which are calculated by using the close - coupled impact parameter method . in the observed spectra ,
the two lines ( 50.691 and 50.703 ) are blended .
so the sum of the intensities of these two lines is used in definition and calculation .
figure 3 shows the ratio as a function of the electron density @xmath11 at three different logarithmic electron temperature ( log@xmath33(k ) ) : 5.9 , 6.1 and 6.3 .
the ratio is sensitive to the electron density in the range from 10@xmath34 @xmath5 to 10@xmath18 @xmath5 which matches conditions that known to appear in various stellar object .
the effect of electron temperature on the ratio appears to be important at a low - density limit @xmath35@xmath5 , while the effect is negligible above @xmath36@xmath5 as shown in fig .
this is an important feature for density diagnostics .
the line intensity at 50.703 holds the sensitivity to the electron temperature and density . at the high - density , it is negligible when compared with that of line at 50.691 , but it appears to be comparable at low - density limit .
so the slight sensitivity to the electron temperature is present in the low - density region as shown in fig .
based on the measurements of line fluxes for individual lines of si x , the observed line ratios are obtained for our sample , which are listed in table 3 .
all these ratios are located in the density - sensitive range ( @xmath3762.7 ) , so the electron density of x - ray emitting layer of stellar coronae can be derived . for capella
, the emission lines of si x appear to be weak , and one feature ( 50.524 ) is strongly contaminated by emission line of fe xvi as shown in sect.2 , so the observed ratio shows a relatively large uncertainty . by comparing
observed ratios with the theoretical prediction , electron densities of stellar coronae are derived for our sample which are listed in table 3 .
figure 4 shows such comparison , in which the solid line is prediction at the logarithmic electron temperature 6.1 of maximum fraction of si x in the ionization equilibrium of mazzatto et al .
( 1998 ) , symbols with errors are observed ratios of our sample .
all the observed ratios are around unity within uncertainty , which result in that the diagnosed electron densities distribute in the range of 10@xmath1710@xmath38 @xmath5 .
one component ( @xmath0 cen b ) of the binary @xmath0 centauri shows a same ratio with that of procyon , while the other component shows a value close to that of @xmath2 eri .
therefore the electron densities for the two stars share the same value as shown in fig .
4 . the work of raassen et al . ( 2003 ) also indicates the similar properties between @xmath0 cen b and procyon .
the ratio between _ ly_@xmath0 of h - like o viii and resonance lines ( or triplet ) of he - like o vii for @xmath0 cen b is very close to that of procyon ( see top pannel of fig . 9 in work of raassen et al .
( 2003 ) ) . a systematic analyses of extreme ultraviolet spectra for 28 stellar coronae performed by sanz - forcada et al .
( 2003 ) , reveals that the stellar coronae consistent with two basic classes of magnetic loops : solar - like loops with maximum temperature around log@xmath33(k)@xmath306.3 and lower electron densities ( @xmath3910@xmath40@xmath5 ) , and hotter loops peaking around log@xmath33(k)@xmath306.9 with higher electron densities ( @xmath41@xmath5 ) . for the hotter temperature region ,
the electron densities are usually derived from highly charged iron ions with higher peak temperature in the ionization equilibrium ( mazzotta et al . 1998 ) .
sanz - forcada et al .
( 2003 ) derived the electron densities are more than @xmath42@xmath5 from euv spectra observed with euve satellite .
an upper limit ( @xmath43@xmath5 ) for stellar coronae was concluded by ness et al .
( 2004 ) from carbon - like iron lines with fluxes extracted from low energy transmission grating spectrometer ( letgs ) spectra . from resolved triplets of he - like magnesium and silicon , upper limits @xmath44 and
@xmath45@xmath5 for capella were derived by canizares et al .
( 2000 ) from the heg spectra . through de - blending the contamination for ne ix triplet , ness et al .
( 2003 ) place an upper limit of 2.0@xmath46@xmath5 for capella .
ness et al . ( 2002 , 2004 ) further made a detailed investigation for the coronal density using he - like triplet , and found that the derived densities from he - like c , n and o do not deviate from low - density limits for inactive stars .
for the cool x - ray emitting region , o vii triplet is extensively used to derive the electron density for stellar coronae because of its high snr around 21 . whereas the c v triplet forms at a much cooler emitting layer which should carry out information in this emitting layer . for inactive and some active stars
, the c v triplet can be detected and the corresponding electron density has been published in literatures(ness et al .
2001 ) . in ionization equilibrium of mazzotta
et al . ( 1998 ) condition , the temperature ( 1.26 mk ) of maximum fractional abundance of si x is close to that derived from h- and he - like c. therefore the electron densities from the two approaches should be compatible with each other .
here we make a comparison between our results and the values constrained by ness et al .
( 2001 ; 2002 ) for our sample , as shown in fig . 5 .
the figure reveals that our results are consistent with those of ness et al .
( 2001 , 2002 ) within 1@xmath21 statistical uncertainty , which suggests that electron densities derived from spectra of ions with similar formation temperature are compatible with each other .
for three inactive stars , the density derived from the two methods shows an excellent agreement .
the uncertainties of our results are much smaller than those constrained by he - like c v. the large uncertainty in the electron densities derived from he - like c v is due to a faint inter - combination line presented , which results in that the observed ratio @xmath47 between the inter - combination ( @xmath48 ) and forbidden ( @xmath49 ) lines located near the low - density limit . the flux measurement for @xmath48 line is strongly depended on the determination of continuum level , so a relatively larger uncertainty appears in constrained electron densities . for selected lines in this study , no faint emission lines are involved .
the line fluxes of the two features are comparable for these inactive stars .
the observed ratios are close to unity , which corresponds to the density - sensitive region . in case of @xmath0
cen a ( g2v ) , only upper limit of density can be obtained from c v triplet because a faint inter - combination line is present .
however present work further constrains the mean electron density for the cooler layer of the star . in case of capella
, the derived electron density also agrees with that derived from c v , whereas our result shows a larger error bar .
one reason is due to that the emission line of si x at 50.524 is severely contaminated by line of fe xvi at 50.555 .
another possible reason may be the determination of continuum level , because the emission lines of si x have become weak for active stars .
for @xmath2 eri , the electron density is not available from c v because the forbidden line is severely contaminated .
the electron density of 1.29@xmath50 @xmath5 constrained by si x could be used to represent the density in c v forming region .
for inactive stars , such as procyon and @xmath0 cen a and b , emission lines of si x at 50.524 and 50.691 are the prominent lines with high snr in the wavelength range 3570 .
our previous paper indicates that the ratio @xmath3 between the two lines is sensitive to electron density , while it is insensitive to electron temperature .
this is a good feature for diagnostic of the electron density for hot astrophysical and laboratory plasmas . in the low - density limit region ,
the obvious temperature effect of the ratio is due to contamination by @xmath19 transition of si x at wavelength 50.703 . compared with the intensity of line at 50.691 , the line intensity at 50.703
steeply decreases with increasing the electron density . in case of capella ,
the line at 50.691 is clearly detected , while the line at 50.524 is severely contaminated by line of fe xvi .
based on the conclusion of argiroffi et al .
( 2003 ) that the emission of capella is compatible with a constant source , we use the emd derived by audard et al .
( 2001 ) and line emissivity included in apec1.3.1 model to estimate the contribution of fe xvi line at 50.555 .
about 62% flux is estimated to be from the contribution of fe xvi , the remaining flux is from si x line at 50.524 .
the contribution of fe xvi line at 50.367 is simultaneously estimated , which contributes about 73% . when the emd is folded with emissivities of fe xvii lines , the estimated flux of fe xvii lines around 50.367
is 0.68@xmath24photon @xmath26 s@xmath25 which is consistent with the remaining flux within statistical error . applying the procedure to @xmath2 eri
, we conclude that the contributions of fe xvi lines at 50.36 and 50.54 are about 53% and 39% , respectively .
so the flux of si x line at 50.524 is about 0.64@xmath24 photon @xmath26 s@xmath25 .
the observed ratios of our sample are close to the unity which corresponds the density - sensitive range of @xmath3 .
comparison of our results with those derived from he - like c v shows a good agreement .
moreover , the derived electron densities and error bars are similar for the inactive stars , but the error bars are large for the active stars . in conclusion ,
the emission lines of si x have potential density - diagnostic application in stellar coronae and other x - ray sources .
argiroffi , c. , maggio , a. , & peres , g. , 2003 , a&a , 404 , 1033 audard , m. , behar , e. , g@xmath51del , m. , et al .
2001 , a&a , 365 , l329 brinkmann , a. c. , gunsing , c. j. t. , kaastra , j. s. , et al .
2000 , apj , 530 , l111 canizares , c. r. , huenemoerder , d. p. , davis , d. s. , et al .
2000 , apj , 539 , l41 doyle , j. g. 1980 , a&a , 87 , 183 foster , v. j. , keenan , f. p. , reid , & r. h. g. 1997 , at .
data nucl .
data tables , 67 , 99 gabriel , a. h. , bely - dubau , f. , faucher , p. , et al .
1988 , j. phys . , 49 , 235 g@xmath51del , m. , schmitt , j. h. m. m. , benz , a. o. , et al .
1995 , a&a , 301 , 201 g@xmath51del , m. , audard , m. , & mewe , r. 2003 , a&a , 403 , 155 liang , g. y. , zhao , g. , shi , j. r. 2005 , mnras , 2005 , ( in referee ) mckenzie , d. l. , & landecker , p. b. 1982 , apj , 259 , 372 mewe , r. , raassen , a. j. j. , drake , j. j. , et al .
2001 , a&a , 368 , 888 ness , j. -u . , brickhouse , n. s. , drake , j. j. , et al .
2003 , apj , 598 , 1277 ness , j. -u . , g@xmath51del , m. , schmitt , j. h. m. m. , et al . 2004 , 427 , 667 ness , j. -u . , mewe , r. , schmitt , j. h. m. m. , et al . 2001 , a&a , 367 , 282 ness , j. -u . ,
schmitt , j. h. m. m. , burwitz , v. , et al .
2002 , a&a , 394 , 911 peres , g. , orlando , s. , reale , f. , et al .
2004 , apj , 612 , 472 raassen , a. j. j. , mewe , r. , audard , m. , et al .
2002 , a&a , 389 , 228 raassen , a. j. j. , ness , j. -u . , mewe , r. , et al .
2003 , a&a , 400 , 671 sanz - forcada , j. , brickhouse , n. s. , dupree , a. k. 2003 , apjs , 145 , 147 siarkowski , m. , pres , p. , drake , s. a. , et al .
1996 , apj , 473 , 470 scelsi , l. , maggio , a. , peres , g. , et al .
2005 , a&a , 432 , 671 schmitt , j. h. , m. m. , & k@xmath51rster , m. 1993 , science , 262 , 215 smith , r. k. , brickhouse , n. s. , liedahl d. a. , et al .
2001 , apj , 556 , l91 vaiana , j. s. , et al .
1981 , apj , 245 , 163 white , n. , e. , shafer , r. , a. , parmar , a. n. , et al .
1990 , apj , 350 , 776 @xmath52 } & 10^{28}{\rm erg / s } \\
\hline { \rm procyon } & 61421 & 63 & 69.6 & { \rm f5.01v - v } & 3.5 & 6540 & 34.46 & 2.06 & 2.43\\ { \rm \alpha~cen~a } & 128620 & 29 & 79.5 & { \rm g2.0v } & 1.34 & 5780 & 33.77 & 1.23 & 0.65 \\ { \rm \alpha~cen~b } & 128621 & 29 & 79.5 & { \rm k0.0v } & 1.34 & 5780 & 33.28 & 0.8 & 0.52 \\ { \rm capella } & 34029 & 1248 & 84.7 & { \rm g1.0iii / k0.0i } & 12.94 & 5850 & 35.71 & 9.2/13 & 255 \\ { \rm \epsilon~eri } & 22049 & 1869 & 105.3 & { \rm k2.0v } & 3.22 & 4780 & 33.11 & 0.81 & 20.9
\\ \hline \end{array}\ ] ] | electron density diagnostics based on a line intensity ratio of si x are applied to the x - ray spectra of procyon , @xmath0 cen a@xmath1b , capella and @xmath2 eri measured with the low energy transmission grating spectrometer ( letgs ) combined with high - resolution camera ( hrc ) on board the _ chandra x - ray observatory_. the ratio @xmath3 of the intensities of the si x lines at 50.524 and 50.691 is adopted .
a certain of the temperature effect in @xmath3 appears near the low - density limit region , which is due to the contamination of si x line at 50.703 . using the emission measure distribution ( emd ) model derived by audard et al .
( 2001 ) for capella and emissivities calculated with apec model by smith et al .
( 2001 ) , we successfully estimate contributions of fe xvi lines at 50.367 and 50.576 ( 73% and 62% , respectively ) . a comparison between observed ratios and theoretical predictions constrains the electron densities ( in logarithmic ) for procyon to be 8.61@xmath4 @xmath5 , while for @xmath0
cen a@xmath1b , capella and @xmath2 eri to be 8.81@xmath6 @xmath5 , 8.60@xmath7 @xmath5 , 9.30@xmath8 @xmath5 and 9.11@xmath9 @xmath5 , respectively .
the comparison of our results with those constrained by the triplet of he - like carbon shows a good agreement . for normal stars ,
our results display a narrow uncertainty , while for active stars , a relatively larger uncertainty , due to the contamination from fe xvi lines , is found .
another possible reason may be that the determination of the continuum level , since the emission lines of si x become weak for the active stars . for @xmath2 eri , an electron density in the c v forming region
was estimated firstly through si x emissions . |
the idea that the observed fundamental forces in 4-dimensions can be understood in terms of the dynamics of a simpler higher dimensional theory is now nearly 90 years old @xcite .
starting from a 5-dimensional theory on a manifold @xmath0 , where @xmath1 is a curved 4-dimensional space - time and the fifth dimension is a perfect circle with radius @xmath2 , and taking the 5-dimensional line element to be @xmath3 : @xmath4 where @xmath5 is a 4-dimensional vector potential , the 5-dimensional einstein action reduces to @xmath6 where @xmath7 is a @xmath8 field strength in 4-dimensions and @xmath9 . if we now introduce extra matter , e.g. a scalar field @xmath10 , and perform a harmonic expansion on @xmath11 , @xmath12 then the 5-dimensional kinetic term for @xmath10 gives rise to an infinite tower of massive fields in @xmath1 , @xmath13 , with masses @xmath14 .
a non - abelian generalisation of the kaluza - klein idea uses a @xmath15-dimensional manifold @xmath16 , with @xmath17 compact lie groups .
the co - set space @xmath18 has isometry group @xmath19 and holonomy group @xmath20 . performing the integral @xmath21 over the internal space , with @xmath22 the @xmath19-invariant measure on @xmath18 , leads to yang - mills gauge theory in 4-dimensions with gauge group @xmath19
; e.g. @xmath23 , with @xmath24 isometry and @xmath8 holonomy , gives 4-dimensional einstein - yang - mills theory with gauge group @xmath24 , see e.g. @xcite .
alternatively , one can start from @xmath15-dimensional yang - mills theory on @xmath25 with gauge group @xmath26 .
forgcs and manton @xcite showed that interesting symmetry breaking effects can occur if @xmath27 and one chooses a specific embedding @xmath28 . integrating over @xmath29
then gives a yang - mills - higgs system on @xmath30 , with a gauge group @xmath31 which is the centraliser of @xmath20 in @xmath26 , i.e. @xmath32 with @xmath33=0 $ ] ( see also @xcite ) .
upon dimensional reduction the internal components of the @xmath15-dimensional gauge field @xmath34 play the r^ ole of higgs fields in @xmath35-dimensions and a higgs potential is generated from the @xmath15-dimensional yang - mills action : @xmath36 ( here @xmath37 are co - ordinates on @xmath1 , @xmath38 co - ordinates on @xmath18 ) .
the full @xmath15-dimensional yang - mills action , with field strength @xmath39 , reduces as @xmath40 where @xmath41 denotes trace over the @xmath15-dimensional gauge group @xmath26 and @xmath42 is over the @xmath35-dimensional gauge group @xmath31 .
furthermore the higgs potential can break @xmath31 dynamically .
in particular if @xmath43 , then @xmath44 breaks @xmath31 spontaneously to @xmath45 , the centraliser of @xmath19 in @xmath26 , @xmath46=0 $ ] .
consider again the simplest case @xmath23 , where @xmath47 and @xmath48 .
for example if @xmath49 then indeed @xmath43 and in the first step @xmath50 : @xmath51 breaking @xmath52 to @xmath53 . upon reduction
the @xmath35-dimensional higgs doublet , @xmath54 , @xmath55 , dynamically breaks @xmath56 , which is the centraliser of @xmath57 in @xmath49 . going beyond @xmath24 symmetry on the co - set space , a harmonic expansion of , for example , a scalar field @xmath10 on @xmath23
, @xmath58 generates a tower of higher modes , @xmath59 , which have masses @xmath60 in @xmath35-dimensions .
much of the steam was taken out of the co - set space dimensional reduction programme with witten s proof that spinors on @xmath25 _ can not _ give a chiral theory on @xmath1 @xcite .
reviews of co - set space dimensional reduction are given in @xcite and @xcite .
equivariant dimensional reduction is a systematic procedure for including internal fluxes on @xmath18 ( instantons and/or monopoles of @xmath20-fields ) which are symmetric ( equivariant ) under @xmath19 @xcite .
it relies on the fact that , with suitable restrictions on @xmath19 and @xmath20 , there is a one - to - one correspondence between @xmath19-equivariant complex vector bundles over @xmath61 @xmath62 and @xmath20-equivariant bundles over @xmath1 , @xmath63 where @xmath19 acts on the space @xmath61 via the trivial action on @xmath1 and by the standard left translation action on @xmath18 ( we shall restrict ourselves to the case where @xmath19 and @xmath20 are compact and the embedding @xmath64 is maximal ) . if @xmath65 and @xmath66 are @xmath67 vector bundles there is a commutative diagram of bundle maps @xmath68_r & \ e \
\ar[d ] \ar[r]^{\rm induce } & \ b \ar[d ] \ & \ar[l]^s \ { \bf c}^k \\ & \m_4 \
& \ar[l]^{\rm restrict } \
\m_d & } \ ] ] where the induction map is defined by @xmath69 in general the reduction gives rise to quiver gauge theories on @xmath30 .
including spinor fields , coupling to background equivariant fluxes , can give rise to chiral theories on @xmath30 .
one expects zero modes of the dirac operator on @xmath18 to manifest themselves as massless chiral fermions in @xmath30 but , as we shall see , yukawa couplings are induced and the dimensional reduction can give masses to some zero modes @xcite .
consider once again the simplest non - trivial example with @xmath47 and @xmath48 , giving a 2-dimensional sphere @xmath23 ( or projective line @xmath70 ) , and with @xmath71 . choosing an embedding @xmath72 gives a decomposition @xmath73 , where @xmath74 , associated with the @xmath75-dimensional irreducible representation of @xmath24 .
let @xmath76 , @xmath77 and @xmath78 .
then , as a @xmath79 matrix , @xmath80 decomposes as @xmath81 where @xmath24 acts on @xmath80 as a @xmath82 block matrix .
each subspace @xmath83 transforms under @xmath84 and carries a @xmath8 charge @xmath85 , @xmath86 . introducing a complex co - ordinate @xmath87 on @xmath88 ( of radius @xmath2 ) ,
@xmath89 we write the potential and field strength for a monopole of charge @xmath90 in these co - ordinates as @xmath91 the @xmath92 gauge potential , a lie algebra valued 1-form @xmath34 on @xmath93 , now splits into @xmath94 blocks @xmath95 where @xmath96 , @xmath97 , @xmath98 is a @xmath99 gauge connection on @xmath30 , and @xmath100 will acquire the interpretation as a set of higgs fields . as a @xmath82 block matrix @xmath101 where each @xmath102 is a @xmath103 matrix transforming under @xmath104 . as a @xmath82 matrix
the higgs field is @xmath105 dimensional reduction generates a 4-dimensional higgs potential , @xmath106\bigg)^2,\ ] ] where @xmath107 is the 6-dimensional gauge coupling .
the minimisation of the higgs potential gives a vacuum structure that depends on the monopole charges @xmath85 . as a concrete example , consider the case with @xmath109 and @xmath110 ( fundamental of @xmath24 ) , so that @xmath111 and @xmath112 , @xmath113 . in this case
there is one unit charge monopole and one anti - monopole sector in the internal space which give a symmetry breaking pattern @xmath114 so @xmath115 is broken dynamically to @xmath8 ( for details , see @xcite ) .
there is only one higgs multiplet , @xmath116 , which is a 2-component vector , and the minimum of @xmath117 is at @xmath118 in a suitable gauge .
perturbing around this vacuum gives @xmath119 , with @xmath120 real , and the higgs mass works out to be @xmath121 .
the three gauge boson masses are @xmath122 while the weinberg angle evaluates to @xmath123 .
clearly this is not a phenomenologically viable model for electroweak interactions , as the gauge boson masses and the weinberg angle are wrong , but it is nevertheless instructive . as a second example take @xmath125 .
let @xmath126 ( adjoint of @xmath24 ) and choose @xmath127 , so that @xmath128 .
there are now three sectors in the internal space , one charge two monopole , its anti - monopole , and a trivial sector .
the symmetry breaking scheme in this case is @xmath129 there are two higgs multiplets , @xmath130 and @xmath131 , both of which are @xmath79 matrices .
the higgs potential is @xmath132 and we expand @xmath102 around the vacuum as @xmath133 with @xmath134 .
diagonalising the higgs mass matrix produces two distinct eigenvalues @xmath135 , @xmath136
. there are @xmath137 gauge bosons with mass @xmath138 , @xmath137 with @xmath139 , while two @xmath140-bosons acquire masses @xmath141 and @xmath142
. this construction generates quiver gauge theories on @xmath30 . writing the lie algebra of @xmath24 in the form @xmath143=\pm 2 j_\pm$ ] ,
the higgs fields give rise to a chain of bundle maps @xmath144 : @xmath145 -10pt the isometry group @xmath24 is rather special in that there is only one raising and one lowering operator , so the quiver diagram is always a chain .
higher rank isometry and holonomy groups generate more complicated quiver diagrams in general . as a more general example consider @xmath146 ( for details see @xcite and @xcite ) .
label the irreducible representations of @xmath52 by @xmath147 , corresponding to the young tableau @xmath148 denote irreducible representations of @xmath149 by @xmath150 , with @xmath151 ( isospin ) and @xmath152 ( hypercharge ) . then under the embedding @xmath153 ,
the irreducible representations decompose as @xmath154 , where @xmath155 represents the set of all @xmath149 irreducible representations in @xmath147 .
for example , @xmath156 has two elements : @xmath157 .
the root diagram for @xmath52 is for any given irreducible representation @xmath147 , @xmath158 and @xmath159 map between elements of @xmath155 with different isospin and can be decomposed into components that increase the isospin and components that decrease it : @xmath160 with @xmath161 choosing a basis of orthonormal 1-forms for @xmath162 which is compatible with the complex structure , @xmath163 , @xmath164 , @xmath165 , @xmath166 , define the lie - algebra valued 1-forms @xmath167 , together with their complex conjugates , via the relations @xmath168 there is then a higgs field , @xmath169 , associated with each @xmath167 .
for example , the adjoint representation @xmath170 of @xmath52 decomposes as @xmath171 where the different @xmath149 representations are also indicated by their usual particle physics notation .
choosing the gauge group to be @xmath172 with @xmath173 there are four higgs fields mapping between the @xmath149 representations and the quiver diagram assumes the form for illustrative purposes , we further specialise to the case @xmath174
. then dimensional reduction gives @xmath175 , @xmath176 and @xmath177 are @xmath178 complex matrices acted on by some @xmath179 subgroup .
the symmetry is further reduced by dynamical symmetry breaking @xmath180 and the higgs potential minimised by @xmath181 where @xmath182 , @xmath183 , @xmath184 , @xmath185 are four unitary matrices satisfying one extra condition @xmath186 for a general @xmath52 irreducible representation , @xmath147 , the quiver diagram is the total number of higgs matrices ( blue links ) is @xmath187 , while the number of gauge groups ( green dots ) is @xmath188 . if @xmath189 are all equal , then all higgs fields are @xmath178 matrices and @xmath44 is minimised by those higgs fields all proportional to unitary matrices , with constraints of the form ( [ higgscondition ] ) on the unitary matrices around any plaquette . interpreting the higgs fields as a @xmath190 lattice gauge field on the quiver lattice , the constraints are satisfied by demanding the trivial gauge configuration on the quiver lattice .
to study how dimensionally reduced fermions and yukawa couplings emerge in these models , we first consider the simplest non - trivial example of @xmath88 . represent the dirac operator for a fermion with unit charge in the presence of a magnetic monopole on @xmath88 of charge @xmath90 by @xmath192 .
mathematically , this is the dirac operator twisted with the @xmath90-th tensor power of the tautological line bundle @xmath193 @xcite . for a given @xmath90
, the eigenspinors will be denoted by @xmath194 and have eigenvalues @xmath195 so that @xmath196 for @xmath90 even the quantum number @xmath197 is half - integral while for odd @xmath90 it is integral : in both cases @xmath198 and the degeneracy is @xmath199 , labelled by @xmath200
. the eigenspinors can be decomposed into their positive and negative chirality components @xmath201 where the sign corresponds to the sign of the eigenvalue .
in addition , for the special value @xmath202 when @xmath203 , there are @xmath204 zero modes : for @xmath205 there are @xmath90 negative chirality modes , which we denote by @xmath206 while for @xmath207 there are @xmath204 positive chirality modes , @xmath208 for a given monopole charge , the index of the dirac operator is @xmath209 the dirac operator on @xmath210 splits up into the direct sum of 4-dimensional and 2-dimensional dirac operators @xmath211 at first sight zero modes of the dirac operator on @xmath88 might be expected to manifest themselves as massless fermions for the dirac operator on @xmath30 , but we shall see below that this is not always the case .
after dimensional reduction a fermion on @xmath210 , e.g. in the fundamental of @xmath92 , will decompose as @xmath212 where the @xmath213 signs refer to the @xmath88 chirality , not 4-dimensional or 6-dimensional chirality .
indeed @xmath214 itself could be either dirac or weyl in 6-dimensions . in the equivariant dimensional reduction framework
only zero modes on @xmath88 are compatible with @xmath24 symmetry : @xmath215 correspond to higher harmonics which do not have this symmetry and correspond to 4-dimensional fermions with masses of order @xmath216 .
focusing on zero modes , the 6-dimensional fermions @xmath217 decompose as @xmath218 where @xmath219 and @xmath220 are either dirac spinors in 4-dimensions , if @xmath214 is dirac in 6-dimensions , or weyl spinors of opposite chirality , if @xmath214 is weyl in 6-dimensions .
not all of the 4-dimensional fermions @xmath219 and @xmath220 are massless however @xcite . in 6 dimensions the dirac operator involves the 6-dimensional gauge field , which includes the higgs field after dimensional reduction , and these induce 4-dimensional yukawa couplings , allowing for the possibility of generating mass terms for 4-dimensional fermions through dynamical symmetry breaking . if , and only if , @xmath221 is odd there is a @xmath35-dimensional yukawa coupling linking @xmath222 to @xmath223 through @xmath224 for the example in [ threetwoone ] , @xmath225 , we had @xmath112 , @xmath113 , and @xmath110
. in this case @xmath222 transforms as @xmath226 under @xmath149 , @xmath223 as @xmath227 , and @xmath228 as @xmath226 .
these 4-dimensional fermions pick up a mass @xmath229 via the higgs vacuum expectation value , which is of the same order as the masses of the higher harmonic fermions arising from non - zero eigenvalues of the dirac operator on @xmath88 and therefore should be removed from consideration if we are assuming higher harmonics are too heavy to be relevant to the physics at low energies .
the issue of fermions on @xmath162 is complicated because there is a topological obstruction to the existence of a spin structure : due to the fact that the second stiefel - whitney class is non - vanishing @xcite there is a global obstruction to defining spinors on @xmath232 for even @xmath233 .
nevertheless , fermions can be defined by coupling them to monopoles and/or instantons ( spin@xmath234 structures ) . the full spectrum of the twisted dirac operator is complicated but for equivariant dimensional reduction we only need the zero modes . for fermions coupling to an equivariant monopole of magnetic charge @xmath221 and an equivariant instanton of topological charge @xmath233 , the index of the dirac operator on @xmath162 is @xcite @xmath235 the fact that this is not an integer if @xmath233 and @xmath221 have the same parity , i.e. they are either both even or both odd ( e.g. @xmath236 ) , is related to the lack of spin structure on @xmath162 . under the embedding @xmath237 , @xmath238 , @xmath233 and @xmath221
always have the same parity , so any equivariant monopole / instanton background arising from the embedding will not admit global spinors .
we therefore allow for a further twist with a monopole of charge @xmath239 ( @xmath240 odd ) and the index for this twisted gauge field configuration is @xmath241 we shall denote the positive and negative chirality zero modes of this operator , with a given fixed @xmath242 , by @xmath243 and @xmath244 respectively ( for notational clarity the degeneracy is not indicated ) .
for @xmath245 we have @xmath246 , and choosing for example @xmath247 results in @xmath248 for example , the case @xmath128 with @xmath249 gives a single @xmath178 higgs matrix and the symmetry reduction scheme @xmath250 with @xmath251 , @xmath252 and @xmath253 are the only zero modes giving the equivariant decomposition @xmath254 where @xmath255 and @xmath256 are either 4-dimensional dirac spinors on @xmath30 , if @xmath214 is dirac in 8-dimensions , or chiral spinors of opposite chirality in 4-dimensions , if @xmath214 is chiral in 8-dimensions .
the induced 4-dimensional yukawa couplings generate a mass term for these spinors given by @xmath257 a different choice of @xmath242 leads to a different conclusion . taking @xmath258 results in @xmath259
there is no analogue of @xmath255 in this case and yukawa couplings can not generate a mass term in 4-dimensions .
starting from the adjoint representation @xmath260 consider the symmetry breaking scheme @xmath261 choosing , for example , @xmath262 gives @xmath263 in this case yukawa couplings generate a mass coupling the 4-dimensional spinors @xmath264 and @xmath265 , but the 8 flavours @xmath266 remain massless .
we have shown that equivariant dimensional reduction with a simple gauge group @xmath26 gives the following : * gauge symmetry reduction @xmath267 with only one gauge coupling in 4-dimensions , even if @xmath31 is semi - simple . * further dynamical symmetry breaking @xmath268 where the vacuum and symmetry breaking patterns , including higgs and gauge boson masses and weinberg angles ,
can be deduced uniquely from group theory and induced representation theory . * in certain cases the vacuum configuration is related to gauge dynamics on the quiver lattice : the higgs vacuum corresponds to zero flux on the quiver lattice .
* when fermions are included , chiral theories with families emerge naturally from non - trivial fluxes on @xmath18 .
* chiral fermions on @xmath93 do not allow direct mass terms , but yukawa couplings can give 4-dimensional masses to some of the resulting fermions on @xmath30 .
yukawa couplings can even give masses to some , but not all , zero modes .
the gauge and fermion structure of equivariant dimensionally reduced field theories is clearly very rich .
standard model type yukawa couplings , with different chiralities belonging to different irreducible representations of the gauge group , arise quite naturally in the models presented here , but an exhaustive analysis of all possibilities would be an ambitious programme and remains to be tackled .
we thank a. chatzistavrakidis and h. steinacker for helpful discussions .
the work of bpd is supported in part by the eu research training network in noncommutative geometry ( eu - ncg ) .
the work of rjs is supported in part by grant st / g000514/1 `` string theory scotland '' from the uk science and technology facilities council .
e. witten , in : _ proceedings of the 1983 shelter island conference on quantum field theory and the fundamental problems of physics _ ,
r. jackiw , n.n .
khuri , s. weinberg and e. witten ( mit press , 1985 ) , p. 227 .
d. kapetanakis and g. zoupanos , phys .
* 219 * ( 1992 ) 1 .
kubyshin , j.m .
mourao , g. rudolph and i.p .
volobujev , _ dimensional reduction of gauge theories , spontaneous compactification and model building _
( springer , 1989 ) . l. lvarez - cnsul and o. garca - prada , j. reine angew . math .
* 556 * ( 2003 ) 1 [ arxiv : math.dg/0112160 ] ; commun . math . phys .
* 238 * ( 2003 ) 1 [ arxiv : math.dg/0112161 ]
. o. lechtenfeld , a.d .
popov and r.j .
szabo , progr .
* 171 * ( 2007 ) 258 [ arxiv:0706.0979 [ hep - th ] ] .
b.p . dolan and r.j .
szabo , jhep * 03 * ( 2009 ) 059 [ arxiv:0901.2491 [ hep - th ] ] .
b.p . dolan and r.j .
szabo , jhep * 08 * ( 2009 ) 038 [ arxiv:0905.4899 [ hep - th ] ] .
r. bott and l.w .
tu , _ differential forms in algebraic topology _
( springer , 1982 ) . | we review recent applications of equivariant dimensional reduction techniques to the construction of yang - mills - higgs - dirac theories with dynamical mass generation and exactly massless chiral fermions .
( based on invited talk given by the first author at the 2nd school on `` quantum gravity and quantum geometry '' session of the 9th hellenic school on elementary particle physics and gravity , corfu , greece , september 1320 2009 . to be published in _ general relativity and gravitation_. ) |
ls i + 61 303 is a high mass x - ray binary ( hmxb ) that consists of an optical star with spectral type b0 ve and an unknown compact companion in a highly eccentric , 26.5 day orbit @xcite . while the system has a relatively low x - ray luminosity for a hmxb , ls i + 61 303 is the 15th brightest @xmath0-ray source included in the _ fermi _ lat 1-year point source catalogue ( @xcite ) . the be disk interacts with the compact companion , producing emission that has been observed to vary with orbital phase at every wavelength across the electromagnetic spectrum , from radio to tev ( eg . @xcite , @xcite ) .
@xcite found periodic radio outbursts that peak near @xmath2 , and they defined the arbitrary reference for zero phase at hjd 2,443,366.775 that remains the conventional definition for ls i + 61 303 .
periastron occurs at @xmath3 @xcite .
during 2008 october and november , we performed an intense multiwavelength observing campaign on ls i + 61 303 supported by a _
cycle 1 program .
we obtained optical h@xmath1 spectra of ls i + 61 303 at the kpno coud feed telescope over 35 consecutive nights to study the evolution of the emission during a complete orbit @xcite , @xcite .
the h@xmath1 line profile exhibits a dramatic emission burst near @xmath4 , observed as a redshifted shoulder in the line profile ( see fig .
[ gray ] ) as the compact source moves almost directly away from the observer .
smaller temporal changes in the red spectra suggest additional h@xmath1 emission variability , so we subtracted the mean emission line profile to investigate the residuals carefully ( see fig .
[ diff ] ) .
during about half of the orbit , @xmath5 , the difference spectra reveal a partial s - shaped pattern similar to a spiral density wave that is commonly observed in be star disks @xcite .
@xcite also observed a strong blue peak near @xmath6 , which supports the development of a spiral density wave near periastron .
after this phase , the peculiar red shoulder develops .
we measured the equivalent width of h@xmath1 , @xmath7 , for each spectrum by directly integrating over the line profile .
( we use the convention that @xmath7 is negative for an emission line . )
the errors in @xmath7 are typically about 10% due to noise and placement of the continuum .
figure [ eqwidth ] shows that during our coud feed run , @xmath7 decreased slightly just before periastron .
since @xmath7 is correlated to the radius of a be star s circumstellar disk @xcite , we interpret the decline in emission as a slight decrease in disk radius as gas is stripped away by the compact companion .
@xmath7 then rises dramatically with the onset of the red shoulder emission component near @xmath4 .
figure [ eqwidth ] also compares our recent @xmath7 with those measured by @xcite .
their data were accumulated over six different observing runs over 19982000 , and the long term differences in emission strength are substantial .
also during 2008 october and november , g. pooley obtained nearly simultaneous radio flux coverage with the arcminute microkelvin imager ( ami ) array .
the 15 ghz ami light curve ( fig .
[ radio ] ) reveals emission that peaks at the same time as the h@xmath1 `` red shoulder '' outburst .
contemporaneous _ rxte _
light curves from @xciteand _ fermi _ light curves ( @xcite ) also reveal orbitally modulated emission that peaks just before the h@xmath1 red shoulder , although their wide phase bins may mask a true correlation .
the h@xmath1 emission clearly traces the high energy emission region in this system .
the unusual broadness of the h@xmath1 red shoulder emission is consistent with a balmer - dominated shock ( bds ; @xcite ) .
bds are traditionally observed around supernova remnants but are also sometimes produced within pulsar wind nebulae and other evolved stellar systems .
they form when high velocity ( 2009000 km s@xmath8 ) shocks collide with the interstellar medium , manifesting themselves as optically emitting filaments .
energetic particles and/or photons may be generated in the post - shock region of the collisionless , non - radiative shock .
direct collisional excitation of the pre - shock atoms produces a narrow emission line component that reflects thermal conditions within the pre - shock gas .
if the energetic particles exceed the shock velocity , the pre - shock hydrogen atoms also exchange electrons with post - shock protons , manifesting themselves as broad neutral hydrogen lines ( widths @xmath9 km s@xmath8 ) .
the h@xmath1 line structure in ls i + 61 303 is complicated by the superposition of emission from the circumstellar disk ; however , the broad red shoulder is consistent with such a bds .
the temporary nature of the red shoulder , as well as the correlated gev
radio emission , suggests that the bds only forms when a high density tidal mass stream interacts with a pulsar wind in ls i + 61 303 .
we thank di harmer and the staff at kpno for their hard work to schedule and support the coud feed observations .
guy pooley , christina aragona , tabetha boyajian , amber marsh , and rachael roettenbacher helped collect the data presented here and should be cheered for their heroic efforts .
this work is supported by nasa dpr numbers nnx08av70 g , nng08e1671 , nnx09at67 g , and an institutional grant from lehigh university . | the @xmath0-ray binary ls i + 61 303 is one of the brightest fermi sources , with orbitally modulated emission across the electromagnetic spectrum .
here we present h@xmath1 spectra of ls i + 61 303 that exhibit a dramatic emission burst shortly before apastron , observed as a redshifted shoulder in the line profile . a correlated burst in radio , x - ray , and gev emission
is observed at the same orbital phase .
we interpret the source of the emission as a compact pulsar wind nebula that forms when a tidal mass stream from the be circumstellar disk interacts with the relativistic pulsar wind .
the h@xmath1 emission offers an important probe of the high energy emission morphology in this system . |
the phenomenon of screening has long been known in presence of long - range coulomb interactions , and it is generally expected to render the observed behavior very similar to that of systems with short - range interactions . indeed , recent computational work has investigated the critical behavior close to charge ordering in lattice coulomb systems @xcite , suggesting the ising universality class .
in addition , analytical arguments have been presented @xcite supporting this view , for a broad class of lattice models with the generalized coulomb interaction of the form @xmath5 in @xmath3 dimensions .
we should mention , however , that long - range interactions are generally expected to produce mean - field like critical behavior for @xmath6 , while for @xmath7 one expect short - range critical behavior .
the nontrivial effect of long - range interactions upon the critical behavior is possible only for @xmath8 , because in this regime the interaction is `` integrable '' ( no neutralizing background is needed ) , and screening becomes inoperative .
should one expect any interesting or novel physics in coulomb - like models ( @xmath1 ) , in comparison to the short - range situation ?
our work confirms that these models indeed feature conventional critical behavior in the narrow critical region @xmath9 .
we show , however , that a striking new behavior is uncovered in a broader temperature interval , reflecting strong geometric frustration inherent to such long - range interactions .
first , we find a dramatic decrease of melting temperature of lattice coulomb gas as a result of the level of frustration in the system .
this can be easily understood by noting that our lattice coulomb gas maps into an antiferromagnetic ising model with long - range interactions . here ,
all the spin tend to anti - align with all other spins , but this can not be achieved for very long interactions ( @xmath1 ) , resulting in very low melting temperature .
indeed , for half - filled coulomb systems ( @xmath10 ) , the melting temperature @xmath11 is one order of magnitude smaller @xcite than the generalized coulomb energy @xmath12 , where @xmath13 is the lattice constant .
continuum models @xcite show an even more dramatic behavior , with the melting temperature being as much as two orders of magnitude smaller then the coulomb energy .
this striking behavior , although well documented in several model studies , is not widely appreciated or understood in simple terms .
the second robust feature of these models , which has only recently been discovered @xcite , is the emergence of the `` pseudogap phase '' .
this pseudogap phase is a specific feature of long - range type interaction with @xmath1 , and is observed in a broad temperature range @xmath14 ( see fig .
1 ) , where @xmath15 is the pseudogap temperature where the gap in the single particle density of state ( dos ) starts to open .
we show that the physical coulomb interaction ( @xmath10 ) lays deep in the regime of very long range interaction @xmath16 , the regime where our analytical `` extended dynamical mean - field theory '' ( edmft ) approach becomes is asymptotically exact @xcite .
this observation explains the surprising accuracy of this analytical scheme both when applied to clean models @xcite , and in previous applications to coulomb glasses @xcite .
lattice model with interactions @xmath17 .
the charge ordering temperature @xmath18 , as obtained from edmft theory ( full line ) and monte - carlo simulations ( open symbols ) . the pseudogap temperature @xmath19(dashed line )
remains finite as @xmath20 ; a broad pseudogap phase emerges at @xmath21 .
we also show @xmath22 for the same model with short - range interactions ( dotted line ) .
the inset shows corresponding single particle density of state ( dos ) for different temperatures .
the edmft results ( dashed line ) show excellent agreement with monte - carlo simulation ( solid line ) above the melting temperature @xmath2 .
@xcite , width=283 ]
our model is described by the hamiltonian of particles living on a half - filled hypercube lattice ( lattice spacing @xmath13 ) that interact via coulomb - like interaction @xmath23 , @xmath24 here @xmath25 is the distance between lattice sites @xmath26 and @xmath27 expressed in the units of the lattice spacing . in general , our unit of energy the nearest - neighbor repulsion @xmath28 .
we focus on the half - filled system @xmath29 that maps the problem into `` antiferromagnetic - like '' ising model .
in addition to numerically - exact ( classical ) monte - carlo simulations , we also utilize an analytical approach , called so - called extended dynamical mean field theory ( edmft ) @xcite which is expected @xcite to be accurate for very long interaction @xmath30 , where the effective coordination number becomes very large .
the inter - site charge correlations are included in this approach , and are treated on the same footing as the local ones @xcite . here
, the lattice problem is mapped into an effective impurity problem embedded in a self - consistently determined bath containing both fermionic and bosonic excitations @xcite . in our case , this bosonic bath describes the plasmon excitations induced by the inter - site charge correlations . of plasmon modes on a 3d hypercubic lattice .
the low - energy branch of the spectrum , which describes the `` sheer '' charge fluctuations close to the ordering wavevector , features an @xmath31-dependent energy scale , one that sets the melting temperature @xmath32 .
the scaling becomes exact for very long interaction range @xmath33 , but is approximately valid for all @xmath34 , explaining the simple @xmath31-dependence of all quantities.,width=278 ] in this paper , we focus on the classical limit , where the origin of the pseudogap phase is most pronounced and the accuracy of the edmft can be tested directly with simple classical monte carlo simulations .
here , the density - density correlation function becomes @xmath35 and the self - consistent condition @xmath36 where we define the classical plasmon - mode spectral density ( fig .
2 ) @xmath37 by solving numerically the self - consistent condition in equation ( 3 ) using the spectral density @xmath38 , the @xmath39 dependence of dimensionless parameter @xmath40 is found to be power - law dependent ( fig .
the analytic solution can be derived at the limit of @xmath41 , spectral density becomes @xmath42 , where @xmath43 ( @xmath44 in 2d , and @xmath45 in 3d ) is the minimum energy in momentum space ( center of brillouin zone ) . by solving equation ( 2 ) at this limit , we find the analytic solution @xmath46 the result qualitatively holds all the way to @xmath47 as shown in fig .
the charge ordering temperature @xmath48 is defined by the @xmath49 at the corresponding ordering wave vector @xmath50 ; in @xmath51 the system forms a bcc structure , and a checkerboard patter in @xmath52 .
the depression of melting temperature as a function of interaction parameter @xmath53 can be understood in simple way by noting that the spectral density @xmath38 has a simple scaling form @xmath54 as shown in fig .
the sharp peak of @xmath55 below characteristic energy scale @xmath56 , physically correspond to `` sheer '' plasmon modes with wave vector @xmath57 .
@xcite this energy scale plays a role of an effective debye temperature .
this tells us why the ordering temperature decreases @xmath58 in agreement with the estimation of lindermann criterion applied to the sheer mode . at the lowest energy near @xmath59 the dispersion relation assumes the standard form @xmath60 , same as for short - range interactions
this can also be understood from the form of the potential surface in momentum space , as shown in fig .
4 . the potential surface is flatter for smaller @xmath31 and the values are close to the minimum energy level at @xmath61 .
this picture clearly explains the origin of the sharp peak in the spectral density @xmath38 . for different interaction range @xmath62 .
in the limit of @xmath63 , we find parameter @xmath64 , both in d=2 and d=3.,width=278 ] in the classical limit , the single particle density of state ( dos ) is nothing more than the probability distribution function of electrostatic potentials computed on different sites in the lattice @xmath65 within the edmft approach it assumes the form @xmath66}\right.\\ \left.+\exp{\left[-\frac{\beta^{2}}{2d}\left(\omega-\frac{d}{2\beta}\right)^{2}\right]}\right\ } .\label{classdos}\end{gathered}\ ] ] the coulomb pseudogap ( `` plasma dip '' ) @xmath67 , which is the distance between the gaussian peaks that start to open at @xmath68 . as shown in fig.1 ,
both @xmath69 and @xmath19 remain finite for @xmath53 , since @xmath70 in this limit ( fig .
this results , for long range interactions , in the formation of a wide pseudogap phase , since the melting temperature decreases as @xmath71 while pseudogap temperature remains finite ( @xmath72 ) as @xmath63 .
we define @xmath15 based on the change of curvature in the dos at @xmath73 , i.e. the vanishing of the second derivative of the dos @xmath74 , and finally use @xmath75 from the power - law dependent of @xmath40 ( fig .
( @xmath76 [ reverse potential surface ] ) according to equation ( 8,9 ) for different interaction range .
the lowest energy state near @xmath44 feature parabolic like surface for interactions @xmath31 equal to 1 ( red / flattest),2(green / middle),3(blue / curviest ) and only change rapidly near the edges of the first brillouin zone .
the edges are related to the tail on the plasmon spectral density of fig . 2 and the parabolic surface related to the peak on the plasmon spectral density .
at @xmath77 , the long - range dispersion expansion does not have rapid change near the edges and is more like short - range dispersion .
( b ) short - range dispersion expansion does not have singularity near the edges for any interaction range .
, title="fig:",width=278 ] ( @xmath76 [ reverse potential surface ] ) according to equation ( 8,9 ) for different interaction range .
the lowest energy state near @xmath44 feature parabolic like surface for interactions @xmath31 equal to 1 ( red / flattest),2(green / middle),3(blue / curviest ) and only change rapidly near the edges of the first brillouin zone .
the edges are related to the tail on the plasmon spectral density of fig . 2 and the parabolic surface related to the peak on the plasmon spectral density .
at @xmath77 , the long - range dispersion expansion does not have rapid change near the edges and is more like short - range dispersion .
( b ) short - range dispersion expansion does not have singularity near the edges for any interaction range .
, title="fig:",width=278 ]
the fact that models with long - range interactions with @xmath1 belong to the same universality class as the short - range ising model can be clearly seen from carefully examining the form of the interaction in momentum space .
to determine its precise form in quantitative detail one must use ewald summation @xcite and split the appropriate lattice into a short - range part @xmath78 and long - range part @xmath79 where @xmath80 is the incomplete gamma function , and @xmath81 is reciprocal lattice unit vector .
the resulting dispersion is shown , for different @xmath31 in fig .
4 . at the center of the brillouin zone ,
the long - range part for different interaction @xmath31 is equally parabolic for all k vectors and only start to drop sharply when approaching the vicinity of the corners of momentum space @xmath82 .
the singularity points near the edges of the brillouin zone where @xmath83 diverges to @xmath84 , leading to positively diverging energy for corresponding ( longitudinal ) fluctuation modes .
this mode has very small statistical weight .
it is related to the tail on the spectral density ( fig .
2 ) , and it does not contribute in the critical behavior of the system . the critical behavior , in contrast ,
is determine by the dispersion of `` sheer '' modes in the vicinity of the ordering wavevector , a quantity that remains of the same qualitative form as for short - range interactions @xcite .
the role of long - range interaction in suppressing critical temperature @xmath2 is an important feature of systems with long - range interactions , behavior which directly reflects the generic presence of strong geometric frustration .
while these effects do not modify the universality class of the narrow critical region , novel behavior is found well above the melting temperature .
here we uncovered the emergence of a broad pseudogap phase , a feature unique to long - range interactions ; this regime is very accurately described by our analytical edmft theory , and we find excellent agreement with ( numerically exact ) monte carlo simulations .
the authors thank seng cheong , misha fogler , daniel khomskii , andy millis , joerg schmalian , dan tsui , and kun yang for useful discussions , and also to the organizers especially natasha kirova who incorporated all of our interest .
this work was supported by the national high magnetic field laboratory ( yp , ht , sp , em , and vd ) and the nsf through grants dmr-0542026 and dmr-1005751 ( yp , ht , and vd ) . | we study a class of models with long - range repulsive interactions of the generalized coulomb form @xmath0 .
we show that decreasing the interaction exponent in the regime @xmath1 dramatically depresses the charge ordering temperature @xmath2 in any dimension @xmath3 , reflecting the strong geometric frustration produced by long - range interactions .
a nearly frozen coulomb liquid then survives in a broad pseudogap phase found at @xmath4 , which is characterized by an unusual temperature dependence of all quantities .
in contrast , the leading critical behavior very close to the charge - ordering temperature remains identical as in models with short - range interactions .
pseudogap , coulomb , universality class , edmft |
during last ten years , there have been numerous published research articles devoted to the precise determination of the strong form factors and coupling constants of meson vertices via qcd sum rules ( qcdsr ) @xcite .
qcdsr formalism have also been successfully used to study some of the `` exotic '' mesons made of quark- gluon hybrid ( @xmath0 ) , tetraquark states ( @xmath1 ) , molecular states of two ordinary mesons , glueballs and many others @xcite .
coupling constants can provide a real possibility for studying the nature of the bottomed and charmed pseudoscalar and axial vector mesons .
more accurate determination of these coupling constants play an important role in understanding of the final states interactions in the hadronic decays of the heavy mesons .
our knowledge of the form factors in hadronic vertices is of crucial importance to estimate hadronic amplitudes when hadronic degrees of freedom are used . when all of the particles in a hadronic vertex are on mass - shell , the effective fields of the hadrons describe point - like physics .
however , when at least one of the particles in the vertex is off - shell , the finite size effects of the hadrons become important .
the following coupling constants have been determined by different research groups : @xmath2 @xcite , @xmath3@xcite , @xmath4@xcite , @xmath5@xcite , @xmath6 @xcite , @xmath7@xcite , @xmath8 @xcite , @xmath9 , @xmath10 @xcite , @xmath11 @xcite and @xmath12 , @xmath13 , @xmath14 and @xmath15@xcite , in the framework of three point qcd sum rules .
it is very important to know the precise functional form of the form factors in these vertices and even to know how this form changes when one or the other ( or both ) mesons are off - shell @xcite . in this review ,
we focus on the method of three point qcd sum rules to calculate , the strong form factors and coupling constants associated with the @xmath16 , @xmath17 , @xmath18 , @xmath19 , @xmath20 and @xmath21 vertices , for both the bottom ( charm ) and pion states being off - shell .
the three point correlation function is investigated in two phenomenological and theoretical sides . in physical or phenomenological part ,
the representation is in terms of hadronic degrees of freedom which is responsible for the introduction of the form factors , decay constants and masses . in qcd or theoretical part , which consists of two ,
perturbative and non - perturbative contributions ( in the present work the calculations contributing the quark - quark and quark - gluon condensate diagrams are considered as non - perturbative effects ) , we evaluate the correlation function in quark - gluon language and in terms of qcd degrees of freedom such as , quark condensate , gluon condensate , etc , by the help of the wilson operator product expansion(ope ) . equating two sides and applying the double borel transformations , with respect to the momentum of the initial and final states , to suppress the contribution of the higher states and continuum ,
the strong form factors are estimated .
the outline of the paper is as follows . in section ii , by introducing the sufficient correlation functions , we obtain qcd sum rules for the strong coupling constant of the considered @xmath16 , @xmath17 and @xmath18 vertices . with the necessary changes in quarks
, we can easily apply the same calculations to the @xmath19 , @xmath20 and @xmath21 vertices . in obtaining the sum rules for physical quantities ,
both light quark - quark and light quark - gluon condensate diagrams are considered as non - perturbative contributions . in section iii ,
the obtained sum rules for the considered strong coupling constants are numerically analysed .
we will obtain the numerical values for each coupling constant when both the bottom ( charm ) and pion states are off - shell . then taking the average of the two off - shell cases
, we will obtain final numerical values for each coupling constant . in this section
, we also compare our results with the existing predictions of the other works .
in order to evaluate the strong coupling constants , it is necessary to know the effective lagrangians of the interaction which , for the vertices @xmath22 , @xmath23 and @xmath24 , are@xcite : @xmath25 from these lagrangians , we can extract elements associated with the @xmath22 , @xmath23 and @xmath24 momentum dependent vertices , that can be written in terms of the form factors : @xmath26 where @xmath27 and @xmath28 are the four momentum of the initial and final mesons and @xmath29 , @xmath30 and @xmath31 are the polarization vector of the @xmath32 and @xmath33 mesons .
we study the strong coupling constants @xmath22 , @xmath23 and @xmath24 vertices when both @xmath34 and @xmath35 $ ] can be off - shell .
the interpolating currents @xmath36 , @xmath37 , @xmath38 and @xmath39 are interpolating currents of @xmath34 , @xmath40 , @xmath32 , @xmath33 mesons , respectively with @xmath41 being the up or down and @xmath42 being the heavy quark fields .
we write the three - point correlation function associated with the @xmath22 , @xmath23 and @xmath24 vertices . for the off - shell @xmath35 $ ]
meson , fig.[f1 ] ( left ) , these correlation functions are given by : @xmath43 @xmath44 @xmath45 and for the off - shell @xmath34 meson , fig.[f1 ] ( right ) , these quantities are : @xmath46 @xmath47 @xmath48 ( 130,23 ) ( -17,-15 ) correlation function in ( eqs .
( [ eq22 ] - [ eq27 ] ) ) in the ope and in the phenomenological side can be written in terms of several tensor structures .
we can write a sum rule to find the coefficients of each structure , leading to as many sum rules as structures . in principle
all the structures should yield the same final results but , the truncation of the ope changes different structures in different ways . therefore some structures lead to sum rules which are more stable . in the simplest cases , such as in the @xmath16 vertex , we have five structures @xmath49 , @xmath50 , @xmath51 , @xmath52 and @xmath53 .
we have selected the @xmath49 structure . in this structure
the quark condensate ( the condensate of lower dimension ) contributes in the case of bottom meson off - shell .
we also did the calculations for the structure @xmath52 and the final results of both structures in predicting of @xmath49 are the same for @xmath54 and in the @xmath17 vertex , we have two structure @xmath55 and @xmath56 .
the two structures give the same result for @xmath57 .
we have chosen the @xmath55 structure . in the @xmath18 vertex
we have only one structure @xmath58 is written as : @xmath59 where @xmath60 denotes other structures and higher states .
the phenomenological side of the vertex function is obtained by considering the contribution of three complete sets of intermediate states with the same quantum number that should be inserted in eqs.([eq22 ] - [ eq27 ] ) .
we use the standard definitions for the decay constants @xmath61 ( @xmath62 , @xmath63 , @xmath64 and @xmath65 ) and are given by : @xmath66 the phenomenological part for the @xmath49 structure associated to @xmath67 vertex , when @xmath68 is off - shell meson is as follow : @xmath69 the phenomenological part for the @xmath70 structure related to the @xmath71 vertex , when @xmath72 is off - shell meson is : @xmath73 the phenomenological part for the @xmath58 structure related to the @xmath74 vertex , when @xmath75 is off - shell meson is : @xmath76 in the eqs.([eq210 ] - [ eq212 ] ) , h.r .
represents the contributions of the higher states and continuum . with the help of the operator product expansion ( ope ) in euclidean region ,
where @xmath77 , we calculate the qcd side of the correlation function ( eqs .
( [ eq22 ] - [ eq27 ] ) ) containing perturbative and non - perturbative parts . in practice ,
only the first few condensates contribute significantly , the most important ones being the 3-dimension , @xmath78 , and the 5-dimension , @xmath79 , condensates . for each invariant structure ,
i , we can write @xmath80 where @xmath81 is spectral density , @xmath82 are the wilson coefficients and @xmath83 is the gluon field strength tensor .
we take for the strange quark condensate @xmath84 @xcite and for the mixed quark - gluon condensate @xmath85 with @xmath86 @xcite .
furthermore , we make the usual assumption that the contributions of higher resonances are well approximated by the perturbative expression @xmath87 with appropriate continuum thresholds @xmath88 and @xmath89 .
the cutkosky s rule allows us to obtain the spectral densities of the correlation function for the lorentz structures appearing in the correlation function .
the leading contribution comes from the perturbative term , shown in fig.[f1 ] . as a result
, the spectral densities are obtained to the double discontinuity in eq.([eq214 ] ) for vertices that are given in appendix a. we proceed to calculate the non - perturbative contributions in the qcd side that contain the quark - quark and quark - gluon condensate . the quark - quark and quark - gluon condensate is considered for when the light quark is a spectator @xcite , therefore only three important diagrams of dimension 3 and 5 remain from the non - perturbative part contributions when the bottom meson are off shell . these diagrams named quark - quark and quark - gluon condensate are depicted in fig.[f2 ] . for the pion off - shell , there is no quark - quark and quark - gluon condensate contribution .
after some straightforward calculations and applying the double borel transformations with respect to the @xmath90 and @xmath91 as : @xmath92 where @xmath93 and @xmath94 are the borel parameters , the contribution of the quark - quark and quark - gluon condensate for the bottom meson off - shell case , are given by : @xmath95 the explicit expressions for @xmath96}^{bottom}$ ] associated with the @xmath97 , @xmath98 and @xmath99 vertices are given in appendix b. ( 50,10 ) ( -70,-40 ) the gluon - gluon condensate is considered when the heavy quark is a spectator @xcite , and the bottom mesons are off - shell , and there is no gluon - gluon condensate contribution .
our numerical analysis shows that the contribution of the non - perturbative part containing the quark - quark and quark - gluon diagrams is about @xmath100 and the gluon - gluon contribution is about @xmath101 of the total and the main contribution comes from the perturbative part of the strong form factors and we can ignore gluon - gluon contribution in our calculation .
the qcd sum rules for the strong form factors are obtained after performing the borel transformation with respect to the variables @xmath102 and @xmath103 on the physical ( phenomenological ) and qcd parts and equating these two representations of the correlations , we obtain the corresponding equations for the strong form factors as follows .
@xmath104 for the @xmath105 form factors : @xmath106 @xmath107 @xmath104 for the @xmath108 form factors : @xmath109 @xmath110 @xmath104 for the @xmath111 form factors : @xmath112 @xmath113 where @xmath114 , @xmath115 and @xmath116 are the continuum thresholds and @xmath117 and @xmath118 are the lower limits of the integrals over @xmath119 as : @xmath120
in this section , the expressions of qcd sum rules obtained for the considered strong coupling constants are investigated .
we choose the values of the meson and quark masses as : @xmath121 , @xmath122 , @xmath123 , @xmath124 , @xmath125 , @xmath126 , @xmath127 , @xmath128 , @xmath129 . also the leptonic decay constants used in this calculation are taken as : @xmath130@xcite , @xmath131 , @xmath132@xcite , @xmath133 , @xmath134@xcite , @xmath135 , @xmath136@xcite . for
a comprehensive analysis of the strong coupling constants , we use the following values of the quark masses @xmath137 and @xmath138 in two sets : seti , @xmath139@xcite , @xmath140@xcite and set ii , @xmath141@xcite , @xmath142@xcite . the expressions for the strong form factors in eqs.([eq219]-[eq224 ] ) should not depend on the borel variables @xmath93 and @xmath94 .
therefore , one has to work in a region where the approximations made are supposedly acceptable and where the result depends only moderately on the borel variables . in this work
we use the following relations between the borel masses @xmath143 and @xmath144 @xcite : @xmath145 for bottom meson off - shell and @xmath146 for pion meson off - shell .
the values of the continuum thresholds @xmath147 and @xmath148 , where m is the @xmath34 mass , for @xmath149 $ ] off - shell and the @xmath149 $ ] meson mass , for @xmath34 off - shell and @xmath150 varies between : @xmath151 @xcite .
using @xmath152 , @xmath153 and fixing @xmath154 , we found a good stability of the sum rule in the interval @xmath155 for the two cases of bottom and pion being off - shell .
the dependence of the strong form factors @xmath54 , @xmath57 and @xmath156 on borel mass parameters for off - shell bottom and pion mesons are shown in fig.[f31 ] .
( 110,26 ) ( -20,-20 ) = 7.5 cm = 7.5 cm we have chosen the borel mass to be @xmath157 .
having determined @xmath158 , we calculated the @xmath159 dependence of the form factors .
we present the results in fig.[f32 ] for the @xmath54 , @xmath57 and @xmath156 vertices . in this
figures , the small circles and boxes correspond to the form factors in the interval where the sum rule is valid .
as it is seen , the form factors and their fit functions coincide together , well . ,
@xmath57 and @xmath156 on @xmath159 for the bottom off - shell and the pion off - shell mesons .
the small circles and boxes correspond to the form factors via the 3psr calculations.,title="fig:",width=192,height=188 ] , @xmath57 and @xmath156 on @xmath159 for the bottom off - shell and the pion off - shell mesons .
the small circles and boxes correspond to the form factors via the 3psr calculations.,title="fig:",width=192,height=188 ] , @xmath57 and @xmath156 on @xmath159 for the bottom off - shell and the pion off - shell mesons .
the small circles and boxes correspond to the form factors via the 3psr calculations.,title="fig:",width=192,height=188 ] we discuss a difficulty inherent to the calculation of coupling constants with qcdsr .
the solution of eqs.([eq219]-[eq224 ] ) is numerical and restricted to a singularity - free region in the @xmath160 axis , usually located in the space - like region .
therefore , in order to reach the pole position , @xmath161 , we must fit the solution , by finding a function @xmath162 which is then extrapolated to the pole , yielding the coupling constant .
the uncertainties associated with the extrapolation procedure , for each vertex is minimized by performing the calculation twice , first putting one meson and then another meson off - shell , to obtain two form factors @xmath163 and @xmath164 and equating these two functions at the respective poles .
the superscripts in parenthesis indicate which meson is off - shell . in order to reduce the freedom in the extrapolation and constrain the form factor ,
we calculate and fit simultaneously the values of @xmath162 with the pion off - shell .
we tried to fit our results to a monopole form , since this is often used for form factors @xcite . for the off - shell pion meson ,
our numerical calculations show that the sufficient parametrization of the form factors with respect to @xmath159 is : @xmath165 and for off - shell bottom meson the strong form factors can be fitted by the exponential fit function as given : @xmath166 .parameters appearing in the fit functions for the @xmath54 , @xmath57 and @xmath156 vertices for @xmath167 and @xmath139 ( set i ) and @xmath141 ( set ii ) . [ cols="^,^,^,^,^",options="header " , ]
in this article , we analyzed the vertices @xmath16 , @xmath17 , @xmath18 , @xmath19 , @xmath20 and @xmath21 within the framework of the three point qcd sum rules approach in an unified way .
the strong coupling constants could give useful information about strong interactions of the strange bottomed and strange charmed mesons and also are important ingredients for estimating the absorption cross section of the @xmath168 by the @xmath34 mesons .
in this appendix , the perturbative contributions for the sum rules defined in eqs.([eq219]-[eq224 ] ) are : @xmath169 , \nonumber \\ \rho^{b_0(\pi)}_{b_1b_0\pi}&= & 4 n_c i_0\left[b_2\left(m_2m_3-km_1m_2+km_1m_3-m_3 ^ 2+\delta-\frac{u}{2}\right)+km_3 ^ 2-m_3m_1-k\frac{\delta}{2}\right ] , \nonumber \\ \rho^{b_1(\pi)}_{b_1b_1\pi}&= & 4i n_c i_0\left[b_1\left(m_3-km_1\right)+b_2\left(m_2+m_3\right)+m_3\right],\end{aligned}\ ] ] the explicit expressions of the coefficients in the spectral densities entering the sum rules are given as : @xmath170,\nonumber \\ b_1 & = & \frac{1}{\lambda(s , s',q^2 ) } \left [ 2 s ' \delta -\delta ' u\right],\nonumber \\ b_2 & = & \frac{1}{\lambda(s , s',q^2 ) } \left [ 2 s \delta ' -\delta u\right],\end{aligned}\ ] ] where @xmath171 , @xmath172 , @xmath173 , @xmath174 for bottom meson off - shell and @xmath175 , @xmath172 , @xmath176 , @xmath177 for pion meson off - shell , @xmath178 represents the color factor . in this appendix ,
the explicit expressions of the coefficients of the quark - quark and quark - gluon condensate of the strong form factors for the vertices @xmath16 , @xmath17 and @xmath18 with applying the double borel transformations are given .
@xmath179 @xmath180 @xmath181 ii coskun aydin , a.hakan yilmaz , mod.phys.lett .
a19,2129 - 2134(2004 ) , v.v.braguta , a.i.onishchenko , phys.lett .
b,591 , 267 - 276 ( 2004 ) , takumi doi , yoshihiko kondo , makoto oka , phys.rept.398,253-279 ( 2004 ) , r.d .
matheus , f.s .
navarra , m. nielsen , r. rodrigues da silva , arxiv : hep - ph/0310280 .
p. colangelo and a. khodjamirian , in at the frontier of particle physics / handbook of qcd , edited by m. shifman ( world scientific , singapore , 2001 ) , vol .
3 , pp . 14951576 ; a.v . radyushkin , in proceedings of the 13th annual hugs at cebaf , hampton , virginia , 1998 , edited by j. l. goity ( world scientific , singapore , 2000 ) , pp . 91150 .
a. bazavov , c. bernard , c. m. bouchard , c. detar , m. di pierro , a. x. el - khadra , r. t. evans , e. d. freeland , e. gmiz , steven gottlieb , u. m. heller , j. e. hetrick , r. jain , a. s. kronfeld , j. laiho , l. levkova , p. b. mackenzie , e. t. neil , m. b. oktay , j. n. simone , r. sugar , d. toussaint , and r. s. van de water , phys .
d 85 , 114506 ( 2012 ) . | in this article , the three point qcd sum rules is used to compute the strong coupling constants of vertices containing the strange bottomed ( charmed ) mesons with pion .
the coupling constants are calculated , when both the bottom ( charm ) and pion states are off - shell .
a comparison of the obtained results of coupling constants with the existing predictions is also made .
key words : strong coupling constant , meson , qcd sum rules , bottom , charm . |
argumentation frameworks ( afs ) are a convenient way to represent conflicting information and to deduce which subset of the information can be inferred .
for instance , they can be used to model dialogs between several agents @xcite or to analyze on - line discussion between social network users @xcite .
argumentation can also be useful in a mono - agent setting , for instance to infer non - trivial conclusions from an inconsistent knowledge base @xcite .
the domain called _ dynamics of argumentation _ has become a hot topic in recent years , with numerous publications about it .
the first ones consider really classical debate scenarios as the source of the dynamic process @xcite .
these approaches are perfectly well - suited for classical exchange of arguments between agents .
then , some approaches have proposed to consider new scenarios , closer to what happens with belief change in logical settings @xcite : these approaches propose to question the existing relation between arguments , and to modify this relation if it is required @xcite .
+ these works directly deal with the structure of the abstract afs .
an interesting question is `` what does af revision mean when we consider logic - based afs ? '' .
indeed , it is not obvious that attacks between arguments can be changed , since they stem from the logical inference relation ; for instance , if arguments @xmath0 and @xmath1 attack each other because their claims are the negation of each other ( _ rebuttal _ attack ) , then it is not accurate to consider that the attack between @xmath0 and @xmath1 could be removed .
but this is only the case when we consider completely specified deductive arguments @xcite .
as argued in the literature @xcite , the arguments which are used in real situations are often enthymemes , which are partially specified arguments : some parts of the support or some parts of the claim are not described , because it is supposed that they belong to some `` common knowledge ''
. there may be different reasons for an agent not to share some part of her knowledge , such as some cost on the communication process .
then , the agent who receives an enthymeme must decide how to complete the content of the enthymeme to be able to use it .
but if the missing formulae to complete the enthymeme are not part of the agent s beliefs ( or at least , are not considered by the agent as the most accurate way to complete the enthymeme ) , then she will use a badly completed enthymeme in her argumentation framework .
we see with this situation that , even with an underlying logical belief base , the nature of arguments and attacks is not absolute ; it depends on the agent s beliefs and on her way to complete enthymemes .
+ so we propose to consider the use of enthymemes in the argumentation process to explain the questionability of some attacks .
we illustrate the possibility that a logic - based argumentation framework contains mistaken attacks .
then we show that the existing work on the dynamics of afs can be used on such enthymeme - based afs , as soon as a distinction between classical deductive arguments and enthymemes is done in the abstract af , and that this distinction is used in the revision process .
+ the paper is organized as follows .
the first section presents the background notions required to the understanding of the paper . in particular
, we describe briefly belief revision , abstract argumentation and revision of afs , and logic - based argumentation .
then in the second section , we focus on enthymemes in logic - based afs ; we explain how using enthymemes can be a source of mistaken attacks in the resulting af . the following section illustrates the revision process on logic - based afs which contain enthymemes . after the description of a basic approach in which each attack concerning an enthymeme is questionable , we propose a refinement of this approach based on the notion of fixed part of an enthymeme .
finally , the last section concludes the paper and sketches some interesting future work .
belief revision is well - known when an agent s beliefs are represented in a logical setting .
the intuitive idea is `` how can an agent incorporate a new piece of information into her beliefs ? '' , which is not a trivial question when the agent s previous beliefs and the new piece of information are conflicting .
one of the most influencial works on this topic is the agm framework @xcite , which gives rationality postulates for belief change operators , when the beliefs are represented as deductively closed sets of formulae .
here we are interested in the adaptation of agm revision to finite propositional logic by katsuno and mendelzon .
they explain that revising a formula @xmath2 by a formula @xmath3 is equivalent to selecting some models of @xmath3 which are minimal w.r.t . some plausibility relation .
this relation has to satisfy some properties .
a faithful assignment is a mapping from a formula @xmath2 to a total pre - order between interpretations @xmath4 such that : 1 . if @xmath5 and @xmath6 , then @xmath7 ; 2 . if @xmath5 and @xmath8 , then @xmath9 ; 3 . if @xmath10 , then @xmath11 .
then , a km revision operator @xmath12 is a mapping from two formulae @xmath13 to a new formula such that @xmath14 for instance , the dalal revision operator can be defined through the pre - order built on the hamming distance .
the _ hamming distance _ between two propositional interpretations @xmath15 is the number of assignments which differ between @xmath16 and @xmath17 , formally : @xmath18 .
+ the total pre - order @xmath19 is defined by @xmath20 the dalal revision operator @xmath21 is a mapping from two formulae @xmath13 to a new formula such that @xmath22 let us illustrate the behavior of the dalal revision operator .
consider @xmath23 and @xmath24 \wedge \neg d$ ] .
the models of @xmath2 are @xmath25 .
we revise @xmath2 by @xmath26 .
the models of @xmath3 are @xmath27 .
table [ tab : distancemodels ] gives the hamming distance between models of @xmath2 and models of @xmath3 .
.hamming distance between models of @xmath2 and @xmath3[tab : distancemodels ] [ cols="^,^,^",options="header " , ] since the minimal hamming distance between @xmath28 and a model of @xmath2 is @xmath29 ( @xmath30 for instance ) , while the distance between @xmath31 and any model of @xmath2 is at least @xmath32 , then @xmath33 , and so @xmath34 .
an abstract af is a directed graph which represents the arguments and the attacks between them .
the usual problem to solve with such an abstract af is `` how to determine which arguments are accepted ? '' .
this question is tackled in the seminal paper by dung .
an argumentation framework ( af ) is a pair @xmath35 where @xmath36 is a set of abstract entities called _ arguments _ , and @xmath37 is the _ attack relation _ which represents the conflicts between arguments .
+ given a _
@xmath38 , the @xmath38-extensions of @xmath39 , denoted @xmath40 , are subsets of @xmath36 which can be accepted .
an argument is then _ skeptically accepted _ by @xmath39 w.r.t .
@xmath38 iff it belongs to each @xmath38-extension of @xmath39 .
+ in this paper , we illustrate our approach on the _ stable semantics _ : @xmath41 is a _ stable extension _ of @xmath39 ( denoted by @xmath42 ) iff * @xmath43 s.t . @xmath44 ; * @xmath45 , @xmath46 s.t . @xmath44 . given the set of arguments @xmath47 , the af @xmath48 with @xmath49
is given in figure [ fig : ex - graphe ] .
its stable extensions are @xmath50 .
as explained in the introduction , the question of change in afs has been tackled by several approaches .
here we use the _ translation - based revision _ from @xcite .
the idea of this method is to translate the af and the semantics into a propositional formula , to use a km revision operator to perform the expected change , and then to decode the models of the revised formula to obtain a set of revised afs .
the propositional encoding is a generalization of a result from besnard and doutre .
they have defined a formula @xmath51 , built on propositional variables corresponding to the arguments , such that the set of models of @xmath51 exactly correspond to the set of stable extensions of an af . @xcite
generalize this encoding with the addition of two other kinds of variables @xmath52 .
@xmath53 means that there is an attack from the argument @xmath54 to the argument @xmath55 , and @xmath56 means that the argument @xmath54 is skeptically accepted .
given an af @xmath57 , the stable encoding of @xmath39 is @xmath58 where @xmath59 \end{array}\ ] ] in general , @xmath60 can be defined for any semantics @xmath38 as soon as the formula @xmath51 exists ; for semantics with a complexity higher than np , we can consider for instance qbf encodings to define @xmath51 .
then , the revision operator is defined as follow : given @xmath12 a km revision operator and @xmath2 a propositional formula built from the set of variables @xmath61 , the _ translation - based revision operator _
@xmath62 is defined as @xmath63 with @xmath64 a mapping from a formula @xmath65 to a set of afs @xmath66 such that each af @xmath67 corresponds to one of the models @xmath68 of @xmath65 : @xmath69 appears in @xmath70 iff @xmath53 is true in @xmath68 .
this general definition allows to change any attack and argument status as long as it is compatible with @xmath38 .
if additional constraints should be satisfied , the use of a constrained version is possible : given @xmath12 a km revision operator and @xmath71 two propositional formulae built from the set of variables @xmath61 , the _ constrained translation - based revision operator _ @xmath72 is defined as @xmath73 to conclude , let us mention a particular revision operator proposed by @xcite : we call @xmath74 ( resp .
@xmath75 ) the translation - based ( resp . constrained translation - based ) revision operator which gives priority to the minimal change of the attack relation .
this operator is similar to the dalal - based revision revision operator @xmath76 , but it uses a weighted version of the hamming distance such that changing the value of a single @xmath53 variable is more expensive than changing the value of each @xmath56 variable .
we consider the af @xmath77 given in figure [ fig : ex - graphe ] .
we suppose the existence of an integrity constraint @xmath78 , which means that the attacks from @xmath79 and @xmath80 to @xmath81 must not be removed .
the result of the revision @xmath82 is the af @xmath83 described in figure [ fig : ex - constrained - revision ] .
now the extensions are @xmath84 , so @xmath81 is skeptically accepted .
we focus on this kind of revision operators because @xcite already proposes a way to incorporate a constraint on the attack relation , which is required by our approach .
other revision or update operators could be used instead , but we should adapt their definition to take into account the constraint .
the question of the exact nature of arguments and attacks is tackled by several approaches which can be gathered under the name _ structural argumentation_. here we focus on one of the most prominent ones : deductive argumentation @xcite .
[ def : deductive - arg ] a deductive argument built from a belief base @xmath85 is a pair @xmath86 , where @xmath87 is called the _ support _ and @xmath3 the _ claim _ , such that : 1 .
@xmath88 , 2 .
@xmath89 , 3 .
@xmath90 , 4 .
@xmath87 is minimal with respect to @xmath91 among the sets of formulae which satisfy items 1 .
2 . and 3 .
there is an intuitive explanation to this definition .
first the agent is supposed to use her beliefs to justify her claim , which explains the first condition .
the second and third conditions guarantee that the claim is actually supported by the beliefs of the agent , but not by conflicting beliefs ( for instance , the sentence `` it is raining and it is not raining , so i am the queen of england . ''
is not an argument at all ) .
finally , the last condition ensures that there is no useless piece of information in the support : `` it is raining , when it is raining i should use an umbrella , and i love chocolate .
so i will use my umbrella . ''
is not accurate either .
+ the conflicts between deductive arguments may have different natures .
the most general sort of conflict is defined as follow : a _ defeater _ for an argument @xmath86 is an argument @xmath92 such that @xmath93 , for some @xmath94 .
it is possible to use deductive arguments to build an _ argument tree _ with the arguments and counterarguments which attack and defend a given claim .
+ we can also build a full argumentation framework from the set of deductive arguments generated from a belief base .
given @xmath36 the set of deductive arguments generated from the belief base @xmath85 , the _ exhaustive graph _ associated with @xmath85 is the af@xmath35 with @xmath95 is a defeater for @xmath96 .
here we focus on the defeater relation , which is the most general one , but exhaustive graphs can be generated with another attack relation which guarantees additional properties for the defeaters ( undercut , rebuttal , and so on )
. moreover , these graphs may be infinite in general ; besnard and hunter propose an approach to circumvent this problem .
see @xcite for more details .
before formalizing our approach , we want to explain intuitively , with natural language arguments , why agents can disagree on the attack relation , and more generally why attacks could be questionable .
let us consider the following arguments : ( c ) : : the us army is preparing a secret plan to retreat from afghanistan ( source : wikileaks ) .
( b ) : : our informed sources say that the wikileaks documents are fake ( source : ny times ) .
( a ) : : the media can not be trusted on military issues ( source : n. chomsky ) .
now we consider three agents @xmath97 ; each of them may have some personal beliefs which are not shared with the other agents .
+ * @xmath98 thinks that chomsky is the most credible source , and considers that wikileaks is a media more reliable than ny times .
so her af is the one given in fig .
[ fig : agent - a1 ] .
* @xmath99 thinks that chomsky is a more credible source than ny times , and ny times is a more credible source than wikileaks .
she also believes that wikileaks can not be seen as a media .
so her af is the one given in fig .
[ fig : agent - a2 ] . *
finally , @xmath100 thinks that ny times is the most credible source , and that chomsky is not reliable on this topic .
so her af is the one given in fig .
[ fig : agent - a3 ] .
these personal afs may depend on many different parameters ( additional information which is not available to each agent , preferences , context , previous experience of each agent , and so on ) .
+ of course , under the assumption that the agents share all their knowledge and beliefs , the personal beliefs of the agents can be represented as additional arguments and we obtain a single af representing the whole information about a topic .
but we think that this assumption is too strong for at least three reasons .
first , there may be technical issues with this information sharing ; for instance , there may be some cost on communication between agents , or the global amount of information in the network may be too important to be stored in a centralized way .
then , for strategical reasons , agents may _ choose _ not to share their knowledge and beliefs .
also , if argument are mined from natural language ( for instance , for an analysis of social networks debates ) , there are likely some implicit pieces of information used in the argumentation process . this explains why some attacks may be questionable .
for instance , if the agents @xmath98 and @xmath100 consider that agent @xmath99 is trustworthy , then they could have to change the attack relation in their own afs if they receive from agent @xmath99 the information `` @xmath101 should be accepted '' . on the opposite , if agents vote to determine the arguments statuses , there will be a majority of agents ( @xmath98 and @xmath100 ) voting against @xmath101 ( meaning that @xmath101 is rejected in their afs ) , so agent @xmath99 should modify the attack relation to incorporate this piece of information in her af . [ [ enthymemes - with - partial - support ] ] enthymemes with partial support + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + now let us formalize this notion of `` arguments with partial knowledge '' , and their role in the existence of mistaken attacks .
hunter defines what he calls _
approximate arguments _ , which are pairs @xmath86 which do _ not _ satisfy the four conditions of deductive arguments . he classifies them depending on which properties they satisfy , and then focuses on enthymemes
enthymeme _ is a pair @xmath86 such that @xmath102 , but there is a set @xmath103 such that @xmath104 is a deductive argument .
intuitively , @xmath105 represents some `` common knowledge '' that the agent supposes to be known by her opponents .
then it is not useful for the agent to state the full deductive argument to be able to exchange information and to reach her goal ( persuading her opponent , helping to take a decision , negotiating , and so on ) . to illustrate this concept , we borrow a simple example of real life use of enthymemes from @xcite
let us consider john and his wife yoko , who is going outside without an umbrella .
if john tells her `` you should take your umbrella , because the weather report predicts rain '' , there is no formal reason to consider that @xmath106 ( with @xmath107 and @xmath108 ) is an argument .
it is in fact an enthymeme , because john supposes that @xmath109 is part of the knowledge he shares with yoko .
one of the questions tackled in @xcite is `` how does the agent knows that @xmath105 is _ actually _ part of the common knowledge ? '' .
hunter supposes that each agent has a way to evaluate the plausibility that a given formula will be part of the knowledge shared between her and another agent . for each agent @xmath110
whose beliefs are expressed in the propositional language @xmath111 , * @xmath112 denotes her own _ personal base _ , * and for each other agent @xmath113 , @xmath114 is a mapping from the language @xmath111 to @xmath115 $ ] , such that @xmath116 represents the certainty that @xmath3 is common to both agents @xmath110 and @xmath113 . [ [ on - enthymemes - and - mistaken - attacks ] ] on enthymemes and mistaken attacks + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + this mapping @xmath114 is used by the agent to build her arguments and decide whether they should be fully specified deductive arguments , or whether enthymemes can be used .
the idea is simply to keep only the formulae @xmath2 in the support such that the associated value @xmath117 is less than a given threshold @xmath118 ; the other ones can be omited because they are supposed to be known by agent @xmath113 .
this process may lead to some problems in the exchange of arguments .
there are at least two sources of mistakes . 1 .
the mapping @xmath114 describes the _ perception _ that agent @xmath110 has of her common knowledge with @xmath113 .
if this perception is wrong , then there could be some exchange of enthymemes that the agent @xmath113 can not decode accurately .
2 . even with a good evaluation of the common knowledge by @xmath114 , the choice of a bad threshold could also lead to enthymemes that the other agent can not decode . in both these situations , agent @xmath113 receives some `` argument '' @xmath119
which is not fully specified , and then she has to complete the support with some @xmath120 from her own belief base , which could of course lead to the addition of some attacks from an existing argument @xmath1 to this new argument @xmath0 , for instance if the claim of @xmath1 is the formula @xmath121 , for some @xmath122 . even if for low - level treatments , it can be represented as @xmath123 ( for instance to determine if possibly new incoming arguments attack it ) , at a higher level it is still the argument @xmath0 which is used .
indeed , this @xmath124 is not an argument that @xmath113 has built by herself , since some of the premises are not part of her belief base .. ] then , agent @xmath113 can receive a new piece of information about the argument @xmath0 which is incompatible with the attack from @xmath1 to @xmath0 ( the simplest example being `` @xmath0 and @xmath1 should be accepted together '' ) .
so she has to build a new internal state @xmath125 from a subset @xmath126 from her belief base ; for the same reason as previously , at a higher level it is still the argument @xmath0 originally built from agent @xmath110 s beliefs .
+ [ [ enthymemes - with - partial - claim ] ] enthymemes with partial claim + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we have seen that enthymemes are a way to communicate arguments with partial support .
black and hunter also give some examples of enthymemes with a partial claim .
borrowing their example , let us consider the sentence @xmath127 `` john has bought the times '' .
the enthymeme @xmath128 can be interpreted in at least two ways , which lead to different claims : 1 .
@xmath129 with @xmath130 `` john has bought a copy of the newspaper the times '' ; 2 .
@xmath131 with @xmath132 `` john has bought the company which publishes the newspaper the times '' .
similarly to what we have described for enthymemes with a partial support , if an agent receives an argument @xmath0 which is in fact an enthymeme with a partial claim , some mistakes in the attack relation can appear .
for instance , she may consider that @xmath0 attacks an argument @xmath1 because some part of @xmath1 s support is conflicting with the completed claim ( either @xmath133 or @xmath134 in our example ) .
if she later receives a piece of information which is not compatible with this attack , then she may have to consider a removal of this attack ( for instance , because the chosen claim is not accurate ) .
+ so we can formally define the class of enthymemes ( with partial claims and partial supports ) as follows : given @xmath135 a deductive argument , an approximate argument @xmath136 is an enthymeme for @xmath137 iff @xmath138 and @xmath139
. stated otherwise , the pair @xmath86 is an enthymeme for @xmath140 , with @xmath87 the partial support and @xmath3 the partial claim . in the rest of this paper , we call such a pair @xmath86 a _ non - completed enthymeme _ and @xmath140 a _ completed enthymeme_. a completed enthymeme may not satisfy the conditions stated in definition [ def : deductive - arg ] , since the set of formulae @xmath87 comes from another agent s belief base .
moreover , contrary to a fully specified argument stemming from the agent s beliefs , a completed enthymeme can be questioned .
for several reasons , the use of an abstract af by the agent is interesting , even when she uses an underlying belief base .
for instance , the developement of efficient approaches to solve abstract argumentation problems permits to obtain the conclusion of the agent s af with respect to several semantics and inference policies ( see for instance the competition of argumentation solvers @xcite ) . but to avoid the loss of information about the nature of arguments and attacks , we propose to refine the definition of the af .
given @xmath141 and @xmath142 which denote respectively the agent s deductive arguments and enthymemes , the agent s _ enthymeme - based af _
is @xmath143 with * @xmath144 ; * @xmath145 ; * @xmath146 the set of certain attacks ( between deductive arguments ) ; * @xmath147 the set of questionable attacks ( concerning at least one enthymeme ) .
computing the extensions of such an af is identical to the process for classical dung s afs ; differentiating both kinds of attacks is useful only for the dynamics scenarios such as revision . in this
setting , each attack can be added or removed as soon as it concerns at least one enthymeme .
we will refine this later .
[ ex : enthymeme - based - af ] let @xmath148 be the enthymeme - based af presented in fig .
[ fig : enthymeme - based - af ] .
arguments with rounded corners are the enthymemes while the other ones are deductive arguments .
similarly , the dashed arrows represent the questionable attacks , while the other ones are the certain attacks . in this example
, we suppose that the agent has received the enthymemes in the following way : * @xmath149 , which has been completed by @xmath150 ; * @xmath151 , which has been completed by @xmath152 in the support and @xmath153 in the claim . with this af ,
the accepted arguments are @xmath154 .
the existence of mistaken attacks in an enthymeme - based af can be tackled through some approaches of the _ dynamics of abstract argumentation _ @xcite . in the case when some arguments and the relations between them are certain ( in particular , when they are fully specified arguments instead of enthymemes ) , integrity constraints can simply be added to these revision / update / enforcement operators to ensure that forbidden attacks will not be added , and mandatory attacks will not be removed .
since it is already defined by @xcite , we will exemplify the dynamics of argumentation with their _ constrained revision _ approach , presented previously .
we can encode an integrity constraint to fix the attacks and non - attacks concerning the deductive arguments into the setting from @xcite .
given @xmath155 an enthymeme - based af , the integrity constraint on deductive arguments is @xmath156 now , if the agent receives some piece of information about the arguments statuses or the attack relation , then she can use the af revision operator @xmath157 as defined previously in the case when this new piece of information disagrees with the current af .
this revision operator guarantees that the relations between deductive arguments will not be modified during the revision process , which is desirable since they are directly stemming from the logical inference relation .
ex : enthymeme - based - af we continue the previous example .
the agent receives the piece of information `` @xmath158 should be accepted '' , which corresponds to the formula @xmath159 .
the integrity constraint is @xmath160 , which ensures that the attacks between the deductive arguments @xmath161 and @xmath162 will not be modified .
the possible results are given in fig .
[ fig : possible - results - revision ] .
+ the exact _ change operator _ which should be used depends on the the properties expected for the process , for instance it is well - known that performing an update @xcite is accurate when the change is explained by an evolution of the world , while performing a revision @xcite is accurate when the evolution only concerns the agent s beliefs about the world ; thus these operations do not satisfy the same properties .
similarly , among the different approaches in the state of the art about the dynamics of afs , each of them do not have the same expressivity .
for instance , the revision approach described in @xcite permits to revise by a formula concerning the extensions , while the translation - based approach illustrated here permits to revise by a formula concerning skeptical acceptance of arguments and attacks at the same time .
so , the choice of a change operator completely depends on the application and the agent s needs and preferences . in the following ,
we continue to consider revision to be consistent with the previous example , but update and extension enforcement @xcite could be considered as well . after obtaining the result of the revision process
, the agent should now decode this result to determine which of the revised afs is the most plausible real af corresponding to her beliefs , and which enthymemes should be internally modified ( and _ how _ they should be internally modified ) to ensure that the abstract af and the logic - based af coincide .
let @xmath66 be the set of afs obtained from the revision process .
for each @xmath67 , @xmath70 is called an _ acceptable af _
iff for each attack which differs between the original af @xmath39 and @xmath70 , the agent s belief base contains some formulae which allow to complete the enthymemes s.t .
this new completion is consistent with the attacks in @xmath70 .
ex : enthymeme - based - af continuing the previous example , let us suppose that the agent s belief base contains the formulae @xmath163 . then the enthymeme @xmath158 can be completed into @xmath164 , which leads to the acceptable af @xmath165 given in fig .
[ fig : acceptable1 ] .
similarly , if the agent s belief base contains the formulae @xmath166 , then the agent can consider the acceptable af @xmath167 given in fig .
[ fig : acceptable2 ] , since @xmath168 can be completed into @xmath169 .
when the set of acceptable afs is not a singleton , we can consider two different solutions : * the agent can keep the whole set as the result , to express the uncertainy of the result of the revision , as suggested by @xcite which consider that revising or updating an af can lead to a set of afs ; * the agent can use external information ( preferences between afs , preferences between formulae in the enthymemes , and so on ) to select a single acceptable af as the result .
none of them is in general more desirable than the other one , the choice depends on the situation ( specific application , user s preferences , computational issues,@xmath170 ) . in the previous parts , we suppose that each attack concerning an enthymeme is questionable .
but we can be more precise in the definition of the enthymeme - based af .
indeed , even when we consider an enthymeme @xmath171 , some of the attacks concerning it may be certain .
we know that some parts of @xmath171 , that we have called previously the partial support and the partial claim , are fixed .
if the reason of an attack between @xmath171 and a deductive argument is a logical conflict involving one of these fixed parts of @xmath171 , then this attack can be considered as certain .
similarly , when we consider another enthymeme @xmath172 , if there is an attack between @xmath171 and @xmath172 which is stemming from the fixed part of @xmath171 and the fixed part of @xmath172 , then this attack can not be removed either .
+ let us first formalize this notion of fixed part .
if @xmath173 is a deductive argument or a non - completed enthymeme , then the _ fixed part _ of @xmath0 is @xmath174 .
+ if @xmath175 is a completed enthymeme , then @xmath174 .
so if we consider a fully specified deductive argument or a non - completed enthymeme , the fixed part is the set of all the formulae involved in it .
but when we consider an enthymeme _ completed with the agent s beliefs _ , then the fixed part is the set of formulae which appear in the enthymeme that the agent has originally received , but do not appear in the completed version of it .
ex : enthymeme - based - af let us consider again the arguments @xmath161 , @xmath162 , @xmath158 and @xmath168 .
the fixed parts of the deductive arguments are trivially the union of their support and their claim .
+ the result is more interesting for the enthymemes : * @xmath176 ; * @xmath177 ; now let us define the involved part of an argument in an attack .
let @xmath173 and @xmath178 be two arguments ( deductive arguments , completed enthymemes or non - completed enthymemes ) .
if there is an attack between @xmath0 and @xmath1 , then the _ involved part _ of @xmath0 in the conflict between @xmath0 and @xmath1 , denoted by @xmath179 , is the set @xmath180 such that @xmath181 and @xmath105 is minimal w.r.t .
@xmath91 . otherwise , @xmath182 .
so , if we have a rebuttal conflict between @xmath0 and @xmath1 ( meaning that the claims are the contradiction of each other ) then @xmath183 and @xmath184 .
if the conflict is an undercut from @xmath0 to @xmath1 ( meaning that the claim @xmath3 of @xmath0 is conflicting with some part @xmath185 of the support of @xmath1 ) , then @xmath183 and @xmath186 .
ex : enthymeme - based - af now we can see which parts of the arguments @xmath161 , @xmath162 , @xmath158 and @xmath168 are involved in conflicts .
the certain attack @xmath187 comes from the contradiction between @xmath188 and @xmath189 , so @xmath190 and @xmath191 .
concerning the questionable attacks , we have : * @xmath192 and @xmath193 ; * @xmath194 and @xmath195 .
now we can refine the definition of an enthymeme - based af . given @xmath141 and @xmath142 which denote respectively the agent s deductive arguments and enthymemes , the agent s _ refined enthymeme - based af _
is @xmath143 with * @xmath144 ; * @xmath196 ; * @xmath197 : the set of certain attacks ; * @xmath198 : the set of questionable attacks .
we use @xmath199 as a notation for @xmath200 , which is the set of attacks between deductive arguments . of course ,
if an argument is a fully specified deductive argument , then the part of it which is involved in conflicts is a fixed part .
so to refine the af , we need to check if it is the case with the enthymemes .
ex : enthymeme - based - af studying the relations between involved parts and fixed parts for the enthymemes @xmath158 and @xmath168 , we obtain the following : * @xmath201 ; * @xmath202 .
so none of the attacks @xmath203 and @xmath204 is certain .
but we can exhibit more interesting cases , for which the use of a refined enthymeme - based af leads to another result than the basic enthymeme - based af .
let @xmath205 be a deductive argument , and @xmath206 an enthymeme , which can be completed for instance by the additional support @xmath207 .
+ it is easy to see here that @xmath208 , so the conflict between @xmath209 and @xmath210 is not questionable , and the af corresponding to these arguments is @xmath211 given in fig .
[ fig : refined - af ] .
\(a ) @xmath212 ; ( b ) [ below of = a ] @xmath213 ; \(a ) edge[bend right ] ( b ) ( b ) edge[bend right ] ( a ) ; when we consider these refined enthymeme - based afs in the revision process , the integrity constraint must be adapted to take into account each certain attack , and not only the ones between deductive arguments : given @xmath155 a refined enthymeme - based af , the integrity constraint on certain attacks is @xmath214 this new constraint ensures that a certain attack will not be removed during the revision process , and that attacks between deductive arguments will not be added if they do not belong to the original af . to conclude ,
let us formalize the intuitive `` chomsky example '' , showing the different completions of enthymemes which lead to the different agents afs .
we use the following propositional variables : @xmath215 means that the us army will retreat ; @xmath216 means that the * w*i*k*ileaks information about * r*etreat is true ; @xmath217 means that the * w*i*k*ileaks documents are * f*ake ; @xmath218 means that * m*edia can * n*ot be * t*rusted on military issues .
as they are stated , the arguments @xmath0 , @xmath1 and @xmath101 which are shared by the agents are these ones : @xmath219 all of them are enthymemes .
for all the agents , the completion of @xmath101 is @xmath220 . but they disagree on the completion of the other enthymemes .
agent @xmath98 considers that @xmath221 and @xmath222 .
agent @xmath99 completes the enthymemes as follows : @xmath223 and @xmath224 .
finally , agent @xmath100 uses these completions of enthymemes : @xmath225 and @xmath224 .
these completions of enthymemes lead to the afs described in figure [ fig : chomsky - example ] , with all arguments which are enthymemes , and all attacks which are questionable .
so here , in case of a revision , the revision operator is used with the integrity constraint @xmath226 , which is equivalent to a revision without a constraint .
+ we mentioned in the introduction two scenarios which require to use dynamics of argumentation techniques .
first , we suppose that agent @xmath99 is considered to be trustworthy by other agents .
then , when she says that @xmath101 should be accepted ( which is represented by the formula @xmath227 ) , the other agents have to revise their af with this new piece of information
. the result of the revision for @xmath98 , with a corresponding completion of enthymemes which are modified because of the revision , is given in figure [ fig : revision - chomsky - a1 ] .
similarly , for @xmath100 , figure [ fig : revision - chomsky - a3 ] describes the possible revised afs , with the modified enthymemes corresponding to it . ' '' '' finally , the other scenario was a vote on the acceptance status of @xmath101 .
since the majority of agents rejects @xmath101 , @xmath99 has to revise her af by @xmath228 to find an agreement with the majority .
possible results are described in figure [ fig : revision - chomsky - a2 ] . ' '' ''
in this paper , we argue that , in realistic situations , agents do not share _ all _ their knowledge and beliefs .
there are different possible reasons , among them , technical and strategical reasons seem to be the most intuitive explanations . also , implicit information is frequently used in natural language argumentation ( on social networks for instance ) .
when this situation occurs , there is some uncertainty in the resulting argumentation frameworks built by the agents .
it is likely that agents opinion about arguments meaning and relations between arguments will differ ; there may be some misunderstanding in the communication process which leads to mistakes in the generation of arguments and attacks . here ,
this is formalized with the use of enthymemes , instead of deductive arguments , to represent the uncertain nature of some arguments .
for this reason , the reception of a new piece of information ( supposed to be reliable ) can force the agents to question the current attack relation to obtain a result which is compatible with the new piece of information .
we have described formally how the use of enthymemes in the argumentation process can lead to the existence of these mistaken attacks , and how to define an argumentation framework which makes the distinction between the certain attacks and the questionable attacks .
then , we have seen that the existing works on the dynamics of abstract afs can be used to perform a change on an enthymeme - based af when it is required to incorporate a new piece of information . here , we exemplify it with the translation - based revision from @xcite , but it can be adapted to any revision , update or enforcement approach as soon as it is possible to consider an integrity constraint on the attack relation . +
this paper only presents some preliminary work on this question .
many future works can be envisioned .
first , we want to model the uncertainty by other means than enthymemes . for instance , using weights could lead to the definition of original change operators which define the notion of minimal change w.r.t . these weights ; it would then be more expensive to change a single attack which has a high weight than to change several attacks with low weights .
determining what can be the origin of these weights is particularly interesting . combined with the use of enthymemes
, we think that giving a low weight to an attack which is easy to modify ( because there are many possible completions of enthymemes in the belief base ) is an interesting way to tackle the problem . for the eafs defined in this paper , as well as the weighted approach
mentioned above , several questions are opened .
we have made the simplifying hypothesis that the agents will have some possible completion of arguments at their disposal , which is not always the case in real world situations .
similarly , the revision operators may lead to an empty - set of results ( because of the integrity constraint ) , or on the opposite , to a non - singleton set of results .
all these cases must be investigated to ensure the possibility of some practical applications .
the complexity of revising such framework , compared to the original revision approach in the abstract setting , will also be studied to be able to identify which approaches can be used for real applications .
we also plan to use some of the existing pieces of software for the dynamics of afs ( in particular the one described in @xcite for extension enforcement ) , to study the scalability of the approaches on practical examples .
but it requires first an important work to build argumentation graphs from logical knowledge - bases , since the existing works focus only on argumentation trees @xcite .
this work as been supported by the austrian science fund ( fwf ) under grants p25521 and i1102 . | _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ in this paper , we present a preliminary work on an approach to fill the gap between logic - based argumentation and the numerous approaches to tackle the dynamics of abstract argumentation frameworks .
our idea is that , even when arguments and attacks are defined by means of a logical belief base , there may be some uncertainty about how accurate is the content of an argument , and so the presence ( or absence ) of attacks concerning it .
we use enthymemes to illustrate this notion of uncertainty of arguments and attacks .
indeed , as argued in the literature , real arguments are often enthymemes instead of completely specified deductive arguments .
this means that some parts of the pair ( support , claim ) may be missing because they are supposed to belong to some `` common knowledge '' , and then should be deduced by the agent which receives the enthymeme .
but the perception that agents have of the common knowledge may be wrong , and then a first agent may state an enthymeme that her opponent is not able to decode in an accurate way .
it is likely that the decoding of the enthymeme by the agent leads to mistaken attacks between this new argument and the existing ones . in this case
, the agent can receive some information about attacks or arguments acceptance statuses which disagree with her argumentation framework .
we exemplify a way to incorporate this new piece of information by means of existing works on the dynamics of abstract argumentation frameworks . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ = 1 |
entangled photons are essential building blocks for optical quantum information processing , such as quantum computing ( qc ) @xcite and quantum key distribution ( qkd ) @xcite .
conventionally , entangled photons have been generated using a myriad of techniques , most notably by using the process of spontaneous parametric down - conversion ( spdc ) utilizing second order nonlinearities in crystals @xcite .
properties such as brightness , scalability , compact form - factor and room temperature operation play key roles in enabling us to fully profit from entangled photon sources in applications such as qc and qkd . as such , the physics and technology of generating and manipulating entangled photons in monolithic settings have recently been topics of immense interest .
harnessing such effects in a monolithic form - factor also enables further incorporation of other photonic components that may be necessary for the aforementioned applications @xcite .
this provided the drive that motivated the early work on implementing entangled sources in waveguides of crystals with appreciable second order nonlinearities such as lithium niobate @xcite . realizing entangled photon sources in monolithic settings enables much more than the inclusion of numerous necessary components simultaneously
: it can enable the direct generation of novel and useful photonic quantum states with specified properties , without moving parts , while benefiting from the accurate alignment of nano - lithography , precision of epitaxial growth and thin film deposition techniques .
for example , monolithic platforms offer opportunities to provide photons that are entangled in one or several degrees of freedom simultaneously without the need for any extra component on the chip @xcite . in addition
, monolithic sources can offer significant control over the spectral - temporal properties of the entangled photons with relative ease and high precision @xcite .
this in turn provides a powerful tool for tailoring the temporal correlation or the spectral bandwidth of the photon states .
such states can be of extremely short correlation times , which can enhance the accuracy of protocols for quantum positioning and timing @xcite and the sensitivity offered by quantum illumination @xcite .
the same integrated sources can generate states with extremely large temporal correlation times .
this in turn leads to narrow spectral bandwidth , which can provide a more efficient atom - photon interface and improved sources for long - haul qkd @xcite .
the vast majority of the aforementioned applications use polarization entangled photon sources .
entanglement in the polarization degree of freedom has been the most widely utilized to implement entangled sources for experiments and applications that probe or exploit quantum effects .
photon pairs in polarization entangled sources need to be indistinguishable in every degree of freedom , except for polarization , which is challenging to achieve for states produced directly in waveguides @xcite . for photon pairs generated in a type - ii process ,
in which the down - converted photons are cross - polarized , the birefringence in the group velocities of the modes , where the photons propagate , will cause a temporal walk - off between the pair , allowing polarization to be inferred from the photon arrival time . on the other hand , for photon pairs generated in a type-0 or type - i process , where the photons in a pair are co - polarized ,
there is a lack of two orthogonal polarizations necessary for polarization entanglement . as a result
, most waveguide sources of photon pairs require an off - chip compensation setup @xcite or an interferometer @xcite to generate polarization entanglement , which increases the source complexity and decreases the system stability significantly .
recently , several techniques have been demonstrated to generate polarization entangled photons from a monolithic chip @xcite .
the approaches which use spontaneous four - wave mixing ( sfwm ) in si - based chips utilize integrated photonic components such as on - chip polarization rotators @xcite or 2d grating couplers @xcite , and benefit from mature fabrication technologies .
however , the indirect bandgap of si presents significant challenges for further integration with the pump lasers . to this end ,
iii - v semiconductor material systems offer an optimal solution in terms of functionality to tailor the dispersion and birefringence as well as monolithic integration with the pump lasers @xcite .
techniques using the counterpropagating phase - matching ( pm ) scheme @xcite and modal pm in bragg reflection waveguides ( brws ) @xcite based on algaas have been demonstrated . in the former case
, however , the requirement of two pump beams with strictly controlled incidence angles and beam shapes imposes significant challenge for further integration , while in the latter case , the spectral distinguishability and walk - off due to modal birefringence compromises the quality of entanglement . in this work ,
we demonstrate how the waveguiding physics associated with brws can be used to simultaneously produce two polarization entangled photon sources using alternative approaches in a single self - contained , room - temperature semiconductor chip .
the waveguide structure utilized is schematically shown in fig .
[ fig : structure_spdc_sem](a ) .
the chip , based on a single monolithic semiconductor brw , is straightforward to design and implement and has no moving parts .
the technique allows direct polarization entanglement generation using an extremely simple setup without any off - chip walk - off compensation , interferometer , or even bandpass filtering .
the first source is achieved through the concurrent utilization of two second order processes , namely type-0 and type - i spdc processes , pumped by a single waveguide mode @xcite as opposed to two modes of different polarizations @xcite or modes propagating in opposite directions @xcite .
this approach permits the integration of the pump with the source in a monolithic form .
within the same waveguide , there exists a second source based on type - ii process due to the lack of material birefringence @xcite .
the virtual energy diagrams of the two sources are also schematically shown in fig .
[ fig : structure_spdc_sem](a ) . as such
, in this approach , by varying the pump polarization and wavelength , one can select between both polarization entangled sources or use them concomitantly .
the direct generation of both bell states @xmath0 and @xmath1 on a single chip can be envisaged .
in addition , by lithographically tuning the waveguide ridge width , one can tune the degree of entanglement of the first source .
for concurrent type-0 and type - i processes with a shared tm polarized pump , paired photons can be either generated in tm polarizations via the type-0 process , or in te polarizations via type - i process . in the ideal case
, photon pairs can be produced from the two processes with the same efficiency and identical spectrum , which renders them in a maximally entangled state @xmath2 .
this is the approach which we shall pursue to obtain a chip - based entangled source using the type-0 and type - i interactions .
the algaas structure used to demonstrate these sources was grown on a [ 001 ] gaas substrate and the waveguide direction was oriented along [ 110 ] . due to the zinc - blende crystal symmetry , the nonlinear tensor @xmath3 is non - zero only when @xmath4 , with @xmath5 being the crystal coordinates . as a result , three spdc processes , namely type-0 , type - i , and type - ii
could coexist provided pm is satisfied . among them , type-0 process depends on the electric field components of the interacting modes along the propagation direction , which are usually negligible in weakly guided waveguides . in brws , however , due to the index variations between different layers , the efficiency of type-0 process can be significant and can even be markedly tuned by engineering the epitaxial structure @xcite . in order to achieve concurrent pm of type-0 and type - i processes ,
the effective indices of the pump @xmath6 should be equal to those of the down - converted photons @xmath7 and @xmath8 simultaneously , i.e. , @xmath9 , with @xmath10 indicating the degenerate pm frequency of the down - converted photons , and te , tm indicating the polarizations .
this requirement can be satisfied lithographically by tuning the ridge width .
the two photon state generated via the two concurrent spdc processes is given by @xcite @xmath11 , \label{eq : state}\end{aligned}\ ] ] where @xmath12 , @xmath13 are the generation rates ( pairs per pump photon ) of the two processes after taking into account the losses . @xmath14 and @xmath15 are the biphoton wavefunctions , with the subscripts indicating the photon polarizations , and satisfy the normalization condition @xmath16 .
when spatially separated , the paired photons are polarization entangled .
the two spectra are found to be almost identical , as shown in fig .
[ fig : spectra](a ) . as a result ,
( [ eq : state ] ) is maximally entangled when the generation rates are the same , i.e. , @xmath17 . in this case , there is no way , even in principle , to tell in which process a pair of photons are generated unless polarizations are measured .
therefore , polarization entangled photons can be generated inherently on the chip without the need of any extra component .
polarization entanglement can also be produced by the type - ii process on the same chip .
following the same formalism , the two - photon state of the type - ii process can be explicitly written as @xmath18 .
\label{eq : state_type - ii}\end{aligned}\ ] ] for maximal entanglement , it requires @xmath19 .
this is not satisfied for the waveguide under test . however , due to the lack of material birefringence , and thus very small temporal walk - off , there exists a significant amount of overlap between @xmath20 and @xmath21 .
the corresponding spectra are shown in fig .
[ fig : spectra](b ) . as a result
, entanglement exists even without any compensation and bandpass filtering .
and @xmath22 polarized photons generated via the type - i and type-0 process , respectively ; and ( b ) those of photons generated via the type - ii process .
the waveguide length is assumed to be 1.05 mm , the same as the waveguide tested in the experiment . ]
photons in a pair need to be spatially separated . in this work , we used a 50:50 beamsplitter to separate photons non - deterministically followed by post - selection .
however , paired photons can also be separated deterministically by a dichroic mirror or integrated dichroic splitter as was done in @xcite .
for an ideal dichroic mirror which has a splitting wavelength at the degenerate point , the degree of entanglement is identical to that using a 50:50 beamsplitter .
as a proof of principle demonstration , a wafer designed for type - i pm around 1550 nm @xcite is used to demonstrate this entangled photon source .
waveguides of this structure are lithographically tuned in order to satisfy concurrent pm of type-0 and type - i processes , with an etch depth of 6.5 @xmath23 m and multiple ridge widths centered at 1.5 @xmath23 m with a step size of 20 nm .
an sem image of a waveguide is shown in fig .
[ fig : structure_spdc_sem](b ) .
numerical simulations predict that concurrent pm of the two types can be achieved at a wavelength around 1.63 @xmath23 m with a ridge width of @xmath241.5 @xmath23 m .
note that redesigning the epitaxial structure can shift the center wavelength to 1550 nm @xcite .
the sample under test had a length of 1.05 mm . in order to select the waveguide that has the best alignment of their pm wavelengths , second harmonic generation ( shg ) for both type-0 and type - i processes
were tested .
the experiment was carried out on a standard end - fire coupling setup by pumping the waveguides with an optical parametric oscillator ( opo ) pumped by a femtosecond pulsed ti : sapphire laser .
the normalized shg tuning curves of the waveguide under test are shown in fig .
[ fig : pm wavelength](a ) . according to fig .
[ fig : pm wavelength](a ) , pm wavelengths of both types are near 1640 nm , the longest achievable wavelength of the opo used . due to the large bandwidth of the pump pulse , the exact pm wavelengths could not be accurately identified .
spdc was then carried out by pumping the waveguides with a cw ti : sapphire laser where the dependence of single photon count rate on the pump wavelength was measured .
this allowed us to locate the waveguide which has the best overlap in pm wavelengths among all tested waveguides .
the results of the waveguide selected , shown in fig . [ fig : pm wavelength](b ) , indicates that the pm wavelengths of both types are @xmath25 nm . the uncertainty in the pm wavelength measurement was due to the pump power fluctuation in the waveguide because of fabry-prot resonances , which could not be resolved with instrumentation available . to generate
polarization entangled photons via concurrent type-0 and type - i processes , the tm polarized pump beam from a cw ti : sapphire laser was coupled into the waveguide using a 100@xmath26 ( n.a.=0.90 ) objective lens , with a power of 1.13 mw before the lens .
photon pairs generated were collected by a 40@xmath26 objective lens at the output facet and passed through long pass filters to eliminate the pump .
after their separation using a 50:50 beamsplitter , the signal and idler photons were collected into multimode fibers and detected by two ingaas single photon detectors . the signal arm detector ( id220 , i d quantique ) operates in a free - running mode and provided 20% efficiency at 1550 nm .
the idler detector ( id210 , i d quantique ) was operating in a gated mode and provided 25% at 1550 nm .
it was externally triggered by the detection events of the first detector .
an optical delay was added before the second detector to compensate for the electronic delay between the two detectors .
both detectors had a deadtime of 20 @xmath23s .
the coincidence counts were recorded with the help of a time - to - digital converter ( tdc ) circuitry . at the degenerate wavelength of @xmath241635 nm ,
the detection efficiencies are around 4% and 5% , respectively .
a pair of quarter - wave plates ( qwps ) and polarizers were used to measure the polarizations of the photon pairs .
the schematic of the experimental setup is illustrated in fig .
[ fig : setup ] . considering the transmission coefficients of the output objective lens ( 90% ) , long pass filters ( 70% ) , beamsplitter ( 43% ) , qwps and polarizers ( 75% ) and fiber collection efficiencies in each path ( 53% and 34% ) , the overall collection efficiency of photon pairs with respect to the position right after the waveguide output facet was found to be @xmath241.5% .
typical coincidence histograms are given in fig .
[ fig : histograph](a ) for two @xmath27 polarized photons and fig .
[ fig : histograph](b ) for two @xmath22 polarized photons for a pump wavelength of 816.76 nm and an integration time of 3 minutes .
the coincidence peaks indicate that photon pairs were generated via both type - i and type-0 processes .
the high level of accidental counts is due to detector dark counts and broken photon pairs due to waveguide losses as well as limited collection ( 1.5% ) and detection ( 0.2% ) efficiencies . by blocking the idler arm
, we found the dark counts of the second detector consists of 83% of the total accidental counts .
thus we can expect a much higher coincidence - to - accidental ( car ) ratio by redesigning a sample which generates photon pairs in region where the detectors are more efficient ( e.g. at 1550 nm ) . and ( b ) @xmath28 basis for an integration time of 3 minutes .
the red bars around the peaks represent the counts in the coincidence window of @xmath240.5 ns . ] the net coincidence count rates for both type-0 and type - i processes are around 0.7 hz , after subtracting the accidental counts . taking into account the signal arm detector s dead time , single count rate ( 16 khz ) , as well as the overall collection and detection efficiencies
, we estimate the photon pair generation rates after the waveguide are @xmath29 pairs / s .
the input objective lens has a transmission of 70% and the coupling efficiency into the pump bragg mode is estimated to be 6% , resulting in an internal pump power of 47.3 @xmath23w right after the input facet .
therefore , the photon pair production rates are both around @xmath30 pairs / s / mw with respect to the internal pump power and external photon pairs , or equivalently , @xmath31 pairs / pump photon .
the fact that both processes have roughly the same generation rates , as opposed to type - i being more efficient than type-0 according to theoretical calculation , could be because that the tm mode has a smaller loss than that of the te mode in this type of deeply etched waveguides .
we confirmed this by measuring the losses using fabry-prot method and found the losses are 4.3 @xmath32 and 2.5 @xmath32 for te and tm modes , respectively .
it could also be because the pump wavelength is slightly detuned from the degenerate pm wavelength of the more efficient process , as the two processes may not have exactly identical pm wavelengths .
quantum state tomography measurements were then subsequently performed by projecting the paired photons into 16 polarization combinations and the density matrix was reconstructed using the maximal likelihood method @xcite .
the net density matrix @xmath33 of the output two - photon state , given by fig .
[ fig : rho ] , is found to have a concurrence @xcite of @xmath34 .
the maximum fidelity with a maximally entangled state @xmath35 , defined by @xmath36 , is 0.89 with a corresponding phase angle @xmath37 .
the non - zero phase @xmath38 is because of the slightly dissimilar degenerate pm wavelengths of the two processes .
theoretical calculation according to eq .
( [ eq : state ] ) shows that this phase value is due to the type - i pm wavelength being @xmath240.02 nm shorter than that of the type-0 process .
the imperfection of the entanglement could be mainly because the pump wavelength is not optimal , causing extra spectral distinguishability between the two processes .
this can be improved by using a tunable diode laser which has a fine spectral tunability .
in addition , the mechanical drift of the characterization setup could result in an increase of mixture and a decrease of entanglement for measurements longer than a few minutes .
the fact that a slightly different pm wavelengths causes one of the processes to take place below its maximal generation rate can be used to balance the generation rates of the two processes and therefore increase the degree of entanglement .
although engineering efforts can be made to achieve almost identical efficiencies , in reality , there may still be differences especially with the existence of polarization dependent losses .
in such a case , the waveguide ridge width can be lithographically tuned such that the stronger process has a shorter pm wavelength , while the pump wavelength should be tuned to the degenerate pm wavelength of the weaker process . to illustrate this point
, we consider an example where the type - i process is twice efficient as the type-0 process , i.e. , @xmath39 . in the case where the two pm wavelengths are identical
, the theoretical fidelity to a maximally entangled state is 0.96 , and the concurrence is 0.93 .
the degree of entanglement can be increased to near maximum by tuning the ridge width such that the pm wavelength of type - i process is 0.55 nm below that of type-0 process .
the corresponding tuning curves of the two processes weighted by their efficiencies are shown in fig .
[ fig : tuning_curves_and_spectra](a ) .
the pump wavelength , marked by the white line , is fixed at the degenerate pm wavelength of type-0 process . in this case , the spectra , given by fig .
[ fig : tuning_curves_and_spectra](b ) , show almost the same amplitude , indicating the two processes having the same spectral brightness near the degenerate wavelength . by utilizing a weak spectral filtering with a bandwidth of 80 nm , the calculated fidelity and concurrence
are increased to 0.997 and 0.995 , respectively , with @xmath40 , indicating the generation of maximally entangled photons . without bandpass filtering ,
the maximum fidelity and concurrence can still reach 0.99 and 0.97 , respectively , with a phase @xmath41 , when the type - i pm wavelength is 0.36 nm below that of type-0 process .
in addition , this tuning approach could be used to generate non - maximally entangled states @xmath42 with a tunable value of @xmath43 , which offers significant advantages over maximally entangled states in some applications such as closing the detection loophole in quantum nonlocality tests @xcite .
lastly , we show the generation of cross - polarized entangled photons on the same chip via the type - ii process . for
a te polarized pump at 812.92 nm , which is only less than 4 nm below those of type-0 and type - i , the output state , again without any compensation and bandpass filtering , shows a concurrence of @xmath44 and a fidelity of 0.74 to the maximally entangled state @xmath45 .
the degree of entanglement is comparable with that obtained in @xcite . the reconstructed density matrix is given by fig .
[ fig : rho_type - ii ] .
the utility of this device becomes most apparent when one considers its capability to have all three types of pm simultaneously achieved at the same wavelength via the tuning of the epitaxial structure combined with ridge width control @xcite .
the unique ability to achieve this monolithically allows the generation of both co - polarized and cross - polarized polarization entangled photons on the same chip at the same pump wavelength , by simply selecting the pump polarization .
nevertheless , a few nanometres difference in the pm wavelength can be covered by a femtosecond pump laser .
in conclusion , we have demonstrated how the waveguiding physics associated with brws can be used to simultaneously produce two polarization entangled photon sources using alternative approaches in a single self - contained , room - temperature semiconductor chip .
direct generation of polarization entangled photons from a monolithic compound semiconductor chip via concurrent type-0 and type - i spdc processes has been characterized .
simultaneous pm of the two processes was achieved using simple lithographic control on the ridge width of brws . without the need of off - chip compensation , interferometer , and bandpass filter ,
the degree of entanglement is among the highest in previous demonstrations from monolithic iii - v and si chips .
in addition , the same device can also directly generate polarization entanglement via the type - ii process , with a pump wavelength of only 4 nm shorter . further improving the device performance relies largely on improved fabrications . by reducing the waveguide losses ,
the generation rates can potentially be increased by more than two orders of magnitudes , as predicted in @xcite for lossless waveguides .
in addition , the degree of entanglement can be increased by fine - tune the ridge width via more precise fabrications , as we have shown that the entanglement can be nearly maximal in the ideal case @xcite .
the degree of entanglement can also be easily increased by bandpass filtering @xcite .
note that previous work on ferro - electric waveguides also studied the diversity of multiple pm processes to generate quantum states of particular properties , such as using two quasi - phase matching ( qpm ) gratings to generate polarization entangled photons @xcite and using different spatial modes to achieve mode entanglement @xcite .
while they are fundamentally important , realizing such effects in monolithic fashions , especially on iii - v semiconductor platforms as in this work , ushers a new era of practical optical quantum information processing .
the authors would like to acknowledge r. marchildon , g. egorov , f. xu , e. zhu , and z. tang for helpful discussions .
this work was supported by natural sciences and engineering research council of canada ( nserc ) .
j. w. silverstone , d. bonneau , k. ohira , n. suzuki , h. yoshida , n. iizuka , m. ezaki , c. m. natarajan , m. g. tanner , r. h. hadfield , v. zwiller , g. d. marshall , j. g. rarity , j. l. obrien , and m. g. thompson , nature photon .
* 8 * , 104 ( 2014 ) s. sauge , m. swillo , s. alber - seifried , g. b. xavier , j. waldeback , m. tengner , d. ljunggren and a. karlsson , opt .
express * 15 * , 6926 ( 2007 ) .
f. kaiser , a. issautier , l. a. ngah , o. dnil , h. herrmann , w. sohler , a. martin , and s. tanzilli , new j. phys .
* 14 * , 085015 ( 2012 ) .
b. g. christensen , k. t. mccusker , j. b. altepeter , b. calkins , t. gerrits , a. e. lita , a. miller , l. k. shalm , y. zhang , s. w. nam , n. brunner , c. c. w. lim , n. gisin , and p. g. kwiat , phys .
lett . * 111 * , 130406 ( 2013 ) . | generating nonclassical states of photons such as polarization entangled states on a monolithic chip is a crucial step towards practical applications of optical quantum information processing such as quantum computing and quantum key distribution .
here we demonstrate two polarization entangled photon sources in a single monolithic semiconductor waveguide .
the first source is achieved through the concurrent utilization of two spontaneous parametric down - conversion ( spdc ) processes , namely type-0 and type - i spdc processes .
the chip can also generate polarization entangled photons via the type - ii spdc process , enabling the generation of both co - polarized and cross - polarized entangled photons in the same device . in both cases , polarization entanglement is generated directly on the chip without the use of any off - chip compensation , interferometry or bandpass filtering .
this enables direct , chip - based generation of both bell states @xmath0 and @xmath1 simultaneously utilizing the same pump source .
in addition , based on compound semiconductors , this chip can be directly integrated with it own pump laser .
this technique ushers an era of self - contained , integrated , electrically pumped , room - temperature polarization entangled photon sources . |
the 3d disk models initially extend between 4 and 20 au .
the central star is modeled as a softened point mass with a mass of @xmath5 .
there are no fixed boundaries ; the disk is free to expand and contract and both the central star and the disk can respond to their mutual gravitational field .
the disk surface density profile is of the type @xmath6 .
disk models have masses between @xmath7 and @xmath8 and initial temperature profiles as those used in ma02 .
the minimum temperatures are reached at the outer edge of the disk and range from 36 to 60 k. more details on the setup of the initial conditions are explained in ma03 .
models are evolved with a locally isothermal or with an adiabatic equation of state ; the new runs adopt an adiabatic equation of state with the addition of radiative cooling and heating . in all cases
we include a standard monaghan artificial viscosity to model shocks , with typical values of the coefficients @xmath9 and @xmath10 .
the artificial viscosity term appears in both the momentum and the energy equation , hence irreversible shock heating is included , except in the locally isothermal equation of state , where by construction the thermal energy of any given particle is assumed to be constant .
radiative cooling is implemented using a prescription similar to that used by rice et al .
( 2003a ) ( see also pickett et al .
2003 ) ; the cooling timescale is simply assumed to be proportional to the local orbital frequency of particles , @xmath11 .
in addition we introduce a density dependent criterion , so that when a region grows beyond a specified threshold , radiative cooling is completely switched off . in the runs presented here the density threshold
is fixed at @xmath12 g/@xmath13 this is a conservative choice based on the recent calculations by boss ( 2001 ) with radiative transfer , which show that at such densities the temperature evolves in a nearly adiabatic fashion . in runs that are evolved using a locally isothermal equation of state we simply switch to an adiabatic equation of state throughout the disk once such density threshold is reached somewhere ( see also ma02 ) .
in the runs with radiative cooling we heat the inner part of the disk by means of another radially dependent term ( this goes exponentially to zero at @xmath14 au ) so that a gravitationally stable disk ( @xmath15 ) develops a temperature profile similar to that used in the initial conditions of the locally isothermal runs ( the latter was indeed motivated by the results of the radiative transfer models of boss ( 1996 ) that include irradiation from the central star and compressional heating due to material infalling onto the disk from the molecular cloud , see ma03 ) .
the simulations are run with gasoline , a parallel tree - based gravity code with sph hydrodynamics which uses multistepping to evolve efficiently a wide density range ( wadsley , stadel & quinn 2003 ) .
in what follows we describe the main results of our large suite of numerical simulations , describing first the locally isothermal and purely adiabatic runs , and then those including radiative cooling and heating .
a detailed description of the former can be found in ma03 .
disks evolved with a locally isothermal equation of state and with @xmath16 fragment after 6 - 7 orbital times ( we used the orbital time at 10 au , 28 years , as a reference ) , the others ( @xmath17 ) develop only from very strong to moderate spiral patterns which reach a peak amplitude after 6 - 7 orbital times and then fade and saturate towards a nearly stationary pattern ( see ma02 ) . with @xmath16 , clump formation proceeds even when the equation of state is switched from locally isothermal to adiabatic once the critical density threshold is reached ( see previous section ) , although the clumps that survive and become gravitationally bound are fewer due to strong shock heating along the spiral arms ( see ma03 ) .
clumps form on eccentric orbits ( these follow the path of the spiral arms ) at scales from below to just about one jupiter mass , for disks with masses in the range @xmath18 . for the same value of @xmath19 , lighter and colder disks produce clumps with appreciably smaller mass ;
the minimum scale of fragmentation is indeed set by the local jeans mass , and it can be shown that this scales as @xmath20 for a fixed value of @xmath19 ( see ma03 ) . the higher the mass resolution
( higher number of particles ) the higher is the number of gravitationally bound condensations that arise . on the other end , @xmath21 marks the threshold between fragmenting and non - fragmenting disks in a way that is independent on resolution ; disks with @xmath22 or higher
were evolved with increasing number of particles , always formed strong spiral patterns but these never broke up into clumps .
the degree of fragmentation depends very weakly on the magnitude of the coefficients of artificial viscosity ; there is a trend of stronger fragmentation with lower viscosity but once again this does not affect the threshold @xmath19 ( see ma02 ) .
we investigate how the outcome of our simulations depends on the way we set up the initial conditions by running a test in which an initially very light disk ( @xmath23 ) is grown in mass at a constant rate over tens of orbital times until it reaches the same ( ten times bigger ) mass of one of our standard disk models undergoing clump formation ( the temperature is kept fixed and is equivalent to that used for the latter model ) .
fragmentation occurs even in such growing disk once the outer regions approach @xmath24 ( figure 1 ) ; this shows that weaker non - axisymmetric torques occurring at higher values of @xmath25 while the disk grows do not lead to self - regulation of the disk at values of @xmath25 higher than 1.4 through mass redistribution .
the results of this experiment weaken considerably one of the arguments against gravitational instability , namely that in real disks spiral instabilities would always saturate before fragmentation becomes possible ( laughlin & rozyczka , 1996 ) .
a few simulations with 200,000 particles were carried out for as many as 30 orbital times to probe the evolution of the system of protoplanets up to about 1000 years .
several mergers occur over a few orbital times after the fragmentation has started , leaving 2 or 3 protoplanets with masses between @xmath26 and @xmath27 on orbits with eccentricities in the range @xmath28 .
the surviving protoplanets continue to accrete mass at a rate which is strongly dependent on the equation of state adopted in this later part of the evolution . for adiabatic conditions the accretion is negligible and
the values we just quoted should well represent the final masses of the planets , whereas for isothermal conditions the accretion rate can be as high as @xmath29/yr , and so protoplanets can reach brown dwarf - like masses in a few thousand years ( the latter is the estimated lifetime of the disk before it is accreted onto the central star due to the strong gravitational torques , see ma03 ) .
we also performed a number of runs in which disks are evolved with an adiabatic equation of state since the very beginning .
we explore different initial @xmath19 and different values of @xmath30 , in the range @xmath31 .
we find fragmentation only in runs starting from a very massive disk ( @xmath32 with unrealistically low values of @xmath25 , @xmath33 , and only for @xmath34 or lower .
in particular only if @xmath35 do clumps survive for several orbital times and become gravitationally bound . at these low values of @xmath25 mass redistribution and shock heating
are indeed so efficient that the disk quickly departs from the initial conditions ; such configurations are therefore unrealistic .
. in these runs ( @xmath3 particles ) fragmentation
is obtained for sufficiently short cooling timescales ( figure 2 ) . for any given initial disk model
we first run a simulation with a long cooling timescale which is expected to yield a stable configuration ( see rice et al .
2003a ) and then we re - simulate the same disk with increasingly smaller cooling timescales until we enter a regime where fragmentation takes place .
we find that the latter occurs for critical cooling timescales in the range @xmath36 depending on disk mass ( this varies between @xmath37 and @xmath38 ) and on the value of @xmath30 used in the equation of state . at larger masses
the higher disk self - gravity can amplify non - axisymmetric perturbations more effectively and more rapidly , so lower cooling rates are needed to counteract heating from strong compressions and shocks along the spiral arms . for values of @xmath39
our results are in very nice agreement with those of rice et al .
( 2003a , b ) , who used lower resolution sph simulations and a different setup of the initial conditions ; we both find that @xmath40 or smalller is necessary to trigger fragmentation in a disk with a mass @xmath41 ( see figure 2 ) . instead , the critical cooling timescale for the same disk rises by more than @xmath42 if @xmath43 like in boss ( 2002 ) , supporting his claim that convective transport of energy at a rate comparable to the orbital time is enough to sustain the instability .
like in the runs described in the previous section , overdense regions grow rapidly in mass and reach fragmentation provided that @xmath19 drops below @xmath44 . when comparing the same disk models , while most features like the type of spiral pattern and the number of clumps formed are comparable in these radiative runs and in those started with a locally isothermal equation of state , the spiral arms appears thinner and more filamemtary in these new runs , owing to sharper density and pressure gradients .
the different density profile across the spiral arms is probably due to the fact that shock heating begins as soon as the first non - axisymmetric structure appears in the new runs , whereas it is completely inhibited below the critical density threshold in the locally isothermal runs .
we note that disks that undergo fragmentation in the simulations of rice et al .
( 2003a , b ) do that in a much stronger fashion compared to ours , with several tens of gravitationally bound clumps instead of the few ( between 1 and 10 , see figure 2 ) that survive the first violent phase of the gravitational instability in our runs .
this difference certainly arises because we shut off cooling in the overdense regions once they grow beyond the density threshold , while rice et al .
do not . in our runs
the sites of formation of the protoplanets coincide with those of the strongest pressure and temperature gradients along the spiral arms ( see figure 2 ) , thus it is not surprising that only the highest overdensity peaks survive ( see also pickett et al .
2003 ) ; later , while the instability starts to fade away , these few gravitationally bound clumps will be able to survive for long timescales quite irrespective of the thermodynamical scheme adopted because the disk enters a more quiet phase of its evolution and strong compressions do not occur anymore ( see ma03 ) .
however , even during the subsequent evolution we expect that allowing or not allowing cooling within the densest regions will make a difference ; clumps would contract nearly isothermally if strong cooling is always active and will reach higher mass concentrations and thereby smaller effective sizes compared to the case in which cooling has been inhibited ( see also figure 2 of ma03 that compares two runs with a locally isothermal equation of state , one with and one without the switch to adiabatic later in the evolution ) . a different size implies a different cross section for mergers , and therefore a different mass spectrum and number of the clumps after many orbital times .
mass accretion rates of protoplanets are quite close to those found in previous runs in which the disk was evolved adiabatically after fragmentation ( see ma03 ) .
one simulation with @xmath3 particles was carried out for 600 years ( equivalent to more than 20 orbital times at 10 au ) ; in that we measured accretion rates @xmath45yr during the late stage , although there is considerable scatter when we look at the individual `` histories '' of the protoplanets protoplanets that venture inside 10 au are heated considerably and they can even lose some mass at the pericenter of their orbit , whereas those that spend most of the time at @xmath46 au have the highest accretion rate since they easily sweep the cold gas along their trail .
gravitational instability continues to remain a very attractive mechanism to explain the origin of gas giants , especially those found in extrasolar planetary systems . with our new runs
we showed that the mass range of bound condensations arising along the spiral arms is significantly broad once the relevant disk parameters are changed ; the outcome is not necessarily `` superjupiters '' , protoplanets even as small as saturn can be formed .
the general picture is confirmed with runs that directly model radiative heating and cooling ; the only necessary requirement for the instability to proceed is that the cooling time must be comparable or only slightly larger than the orbital time at some point of disk evolution .
once protoplanets are formed , mergers during the first few orbital times and accretion of disk gas over longer timescales are the two ways by which they can grow in mass .
how much they can grow will depend a lot on the details of heating and cooling .
our results suggest that , even if the cooling time remains comparable to the orbital time in the outer disk , a mass of @xmath47 might be an upper limit if gravitational torques cause the dissipation of most of the disk on timescales of about @xmath48 years ( see also ma03 ) .
bate , m.r . , & burkert , a. , 1997 , mnras , 288 , 1060 bate , m.r . ,
lubow , s.h . ,
ogilvie , g.i . , & miller , k.a .
, 2003 , mnras , in press boss , a.p . , 1996 ,
apj , 469 , 906 boss , a.p . , 2001 , apj , 563 , 367 boss , a.p . , 2002 ,
apj , 576 , 462 cameron , a.g.w . , 1978 , moon planets , 18 , 5 guillot , t. , 1999 , science , 286 , 72 haisch , k. r. jr . ,
lada , e. a. , & lada , c.j .
, 2001 , aj , 121 , 2065 laughlin , g. & rozyczka , m. , 1996 , apj 456 , 279 lissauer , j.j . , 1993 ,
ara&a , 31 , 129 nelson , a.f . , 2003 , in asp conf .
294 , scientific frontiers in research on extrasolar planets , ed .
d.deming ( san francusco : asp ) , in press mayer , l. , quinn , t. , wadsley , j. , & stadel , j.j . , 2002 ,
science , 298 , 1756 mayer , l. , quinn , t. , wadsley , j. , & stadel , j.j . , 2003 ,
apj , submitted , astro - ph/0310771 marcy , g.w . , & butler , r.p . , 1998 , araa , 36 , 57 pickett , b.k . , cassen , p. , durisen , r.h . , & link , r. 2000 , apj , 529 , 1034 pickett , b.k . ,
meija , a. , durisen , r.h . ,
cassen , p.m. , berry , d.k . , & link , r.p . , 2003 ,
apj , 590 , 1060 nelson , r.p , papaloizou j.c.b . , masset , f. & kley , w. , 2000 , mnras , 318 , 18 rice , w.k.m . , armitage , p.j . ,
bate , m.r . , & bonnell , i.a .
, 2003a , mnras , 339 , 1025 rice , w.k.m . , armitage , p.j . ,
bate , m.r . , & bonnell , i.a .
, 2003b , mnras , in press , astro - ph/0310679 throop , h.b . , bally , j. , esposito , j.l.w . , mccaughrean , m.j . , 2001 , science , 292 , 1686 wadsley , j. , stadel , j. , & quinn , t.r . , new astronomy , 9 , 137 , in press | we present the results of high resolution sph simulations of the evolution of gravitationally unstable protoplanetary disks .
we report on calculations in which the disk is evolved using a locally isothermal or adiabatic equation of state ( with shock heating ) , and also on new simulations in which cooling and heating by radiation are explicitly modeled .
we find that disks with a minimum toomre parameter @xmath0 fragment into several gravitationally bound protoplanets with masses from below to a few jupiter masses .
this is confirmed also in runs where the disk is given a quiet start , growing gradually in mass over several orbital times .
a cooling time comparable to the orbital time is needed to achieve fragmentation , for disk masses in the range @xmath1 .
after about 30 orbital times , merging between the bound condensations always leads to 2 - 3 protoplanets on quite eccentric orbits .
# 1_#1 _ # 1_#1 _ = @scaling.95 # 1@scaling#1 # 1=@scaling the formation of gas giants as a result of fragmentation in a protoplanetary disk , an old idea ( kuiper 1951 , cameron 1978 ) , has been recently revived by a number of studies that are finally laying the ground for a quantitative understanding of such process ( boss 2001 , 2002 ; mayer et al .
2002 , 2003;pickett et al . 2000 , 2003 ) .
the renewed consideration of this mechanism stems from several problems faced by the conventional model of giant planet formation , in which first a rocky core is assembled over @xmath2 years and then a gaseous envelope is accreted in a few million years or more , the exact timescale being dependent on the details of the models like the disk surface density and the opacity of the mixture of gas and dust ( lissauer 1993 ) .
these timescales are an order of magnitude too long to form planets before the disk is dissipated by photoevaporation in highly irradiated environments like the orion nebula ( throop et al .
2001 ) , and are a bit too tight even when compared to the typical lifetime of disks in more quiet environments like taurus ( haisch , lada & lada 2001 ) .
one of the strong points of the core accretion model , namely the prediction that gas giants have solid cores , is considerably weakened now that new models of the interior of jupiter are consistent with the total absence of such a core ( guillot 1999 ) .
the discovery of extrasolar planets ( marcy & butler 1998 ) has worsened the situation further because now we need to explain the existence of very massive planets , up to ten times larger than jupiter , that in the standard model would either take too long to be formed , might not form at all ( bate et al .
2003 ) or could migrate towards the central star before being able to accrete enough mass ( nelson et al .
2000 ) .
the distribution of their orbital eccentricities must also be explained ; whilst in the past few years several papers have proposed a variety of explanations , sometimes tuned to the properties of one particular system , none of these is valid in general .
mayer et al . ( 2002 , hereafter ma02 ) showed for the first time that if a massive disk remains cold long enough , until the growing overdensities reach some density threshold , fragmentation takes place even when heating from compression and shocks is included the resulting clumps survive for tens of orbital times , leading to systems of only a few protoplanets with masses and orbital eccentricities in the range of observed extrasolar planets .
the strong point of the 3d smoothed particle hydrodynamics ( sph ) simulations of ma02 was the high resolution ( up to @xmath3 gas particles ) that allowed to resolve the local jeans length across a wide range of densities , including the regime in which fragmentation takes place ( bate & burkert 1997 ; nelson 2003 ) .
however , these simulations where quite simplified in the thermodynamics ( see pickett et al . 2000 ; 2003 ) , since only two extreme conditions , a locally isothermal or an adiabatic equation of state ( with shock heating ) , were employed .
boss ( 2001 , 2002 ) , by using radiative transfer in the diffusion approximation , has shown that efficient cooling in the disk midplane can occur thanks to convective transport of heat to the disk atmosphere .
he finds that the timescale for convective cooling is comparable to the orbital time in the outer , colder regions of a disk of mass @xmath4 . here
we review the main results obtained with the large survey of simulations extensively described in mayer et al .
( 2003 , hereafter ma03 ) , and we present the first results of new simulations in which radiative cooling and heating are directly implemented . following recent work by pickett et al .
( 2003 ) and rice et al .
( 2003a , b ) , we will investigate how fast cooling has to be for fragmentation into gravitationally bound planets to occur . |
in spite of significant progress in recent years , the nature of gamma - ray bursts ( grb ) , discovered by @xcite more than 30 years ago , is still enigmatic .
although the light curves and emission spectra of grbs are very diverse , they seem to split into two groups of possibly different origin : long bursts ( @xmath1 s ) with a softer spectrum and short bursts ( @xmath2 s ) with a harder spectrum @xcite .
the long grbs are often believed to be associated with star - formation regions @xcite .
only these regions can host massive stars that have astronomically very short lifetime and die soon after the birth .
in fact , recent observations have provided strong arguments in favour of the connection of grbs with the deaths of massive stars .
light curves of many grb optical afterglows show features inherent in the supernovae events ; moreover , several long grbs have been firmly associated with particular supernovae , the most popular examples being grb 980425 and sn 1998bw @xcite .
even more convincing evidence exists in the case of the low red shift grb 030329 ( @xmath3 ; @xcite ) and its associated supernova , sn 2003dh @xcite .
the spectra of these supernovae show exceptionally broad emission lines indicating abnormally high velocity of the ejecta , typical of the recently proposed `` hypernovae '' class objects .
the most popular model of the central engine of these sources is based on the `` failed supernova '' stellar collapse scenario implying that the iron core of the progenitor star forms a black hole @xcite . if the progenitor does not rotate , its collapse is likely to happen silently until all the star has been swallowed up by the black hole .
if , however , the specific angular momentum of the equatorial part of the stellar envelope exceeds that of the last stable orbit of the black hole , then the collapse becomes highly anisotropic . while in the polar region it may proceed more or less uninhibited , the equatorial layers form a dense and massive accretion disk .
then the collapse of the layers is delayed , and the gravitational energy released in the disk can be very large and drive grb outflows , predominantly in the polar directions where mass density of the accreting matter can be much lower @xcite . however , the actual process responsible for the grb outflows is not established and remains a subject of ongoing investigations . the main mechanisms proposed to explain grb outflows are neutrino pair annihilation heating @xcite , magnetized disk wind @xcite , and magnetic braking of the central black hole rotation @xcite
. high - precision self - consistent models of disk dynamics and neutrino propagation are required in order to obtain reliable results in the neutrino - driven supernova explosion theory . by now only relatively crude calculations have been carried out , and they show that the neutrino heating need not have a dominant role .
@xcite studied the heating rate due to annihilation of neutrinos emitted by neutrinospheres of various prescribed geometries and temperatures .
the energy deposition rates obtained in this paper lie in the range @xmath4 erg / s , and the typical annihilation efficiency seems to be rather low , about @xmath5 .
neutrino heating from a geometrically thin standard accretion disk @xcite is calculated in a recent article by @xcite .
it is shown that the process efficiency strongly depends on the rotation parameters of the black hole and rapidly decreases with the distance as @xmath6 .
other aspects of the collapsar model were considered in @xcite .
@xcite carried out general relativistic mhd simulations of accretion disks with the masses @xmath7 .
they found that the disk opacity for neutrinos was high , resulting in low ( @xmath8 ) efficiency of neutrino emission itself , as most neutrinos generated in the disk could not escape from it and accreted in the black hole . @xcite
considered both neutrino heating and cooling in their newtonian simulations of collapsars .
they concluded that the neutrino energy deposition was insufficient to drive grb explosions and that the magnetic mechanism was more promising . in the last few years
the role of magnetic field in driving the black hole accretion and relativistic outflows has been a subject of active investigations via numerical simulations , that produced numerous interesting and important results @xcite . in these studies
the initial distribution described a keplerian disk or equilibrium torus threaded with a relatively weak poloidal field , whose lines followed the iso - density or iso - pressure contours of the disk .
the disk accretion was found to be driven by magnetic stresses via development of magneto - rotational instability ( mri ) @xcite .
in addition to the disk , the numerical solution considered two other generic structures - the magnetized disk corona and the highly magnetized polar funnel that hosted the relativistic outflow from the black hole .
these studies applied simple adiabatic equations of state and did not take into account effects of radiative cooling that may be small for some types of active galactic nuclei , but not for disks of collapsing stars .
recently @xcite have carried out two - dimensional general relativistic mhd simulations in kerr metrics with a realistic equation of state and taking into account the neutrino cooling , but the physical time span of their computations was rather short , only @xmath9 s. in this article , we model the situation ( similar to @xcite ) when a massive stellar core collapses and the outer envelop forms a massive accreting torus . in section 2 , we consider physical processes , in section 3 initial conditions , in section 4 results , section 5 is the discussion .
for the simulations we use the equation of state @xmath10 where @xmath11 is the boltzmann constant , @xmath12 is the radiation energy density constant , @xmath13 is concentration of baryons , @xmath14 is the mass of proton , @xmath15 is the pressure , @xmath16 is the density and @xmath17 is the temperature ; @xmath18 describes the ground level for the cold degenerate matter , the expression for @xmath18 is an approximation of the table function from @xcite .
the specific energy ( per mass unit ) was defined thermodynamically as @xmath19 the value of @xmath20 was taken from @xcite .
this equation of state allows us to deal with high - density matter .
a more accurate eos is significantly more expensive in calculations , while the disregarding of the @xmath21-pair component in the plasma , for instance , can give a relative error in definition of the temperature less than @xmath22 at high temperatures @xcite . on the grounds discussed in the introduction we ignore effects of neutrino heating in our simulations and take into account only the energy loss due to emission of neutrinos .
approximation formulas for the cooling rates due to @xmath21 pair annihilation @xmath23 , neutrino photo - production @xmath24 , and plasma mechanism @xmath25 are taken from @xcite , the cooling rate due to urca processes @xmath26 from @xcite , and the cooling rate due to synchrotron neutrino emission @xmath27 from @xcite : @xmath28 the neutrino cooling is introduced by the source term in the energy - momentum equation @xmath29 where @xmath30 is the plasma four - velocity , @xmath31 is the determinant of the metric tensor , and @xmath32 is the cooling rate as measured in the fluid frame .
one of the aims of our research was the investigation of sensitivity of the torus accretion model to the neutrino cooling processes .
we compare two cases : with no cooling at all and the optically thin case with neutrino cooling .
the first instance is the most popular @xcite and less realistic .
the second instance well suits for a light torus with the mass of a few per cent of the solar .
it could be realized when two neutron stars or a neutron star and a black hole merge .
the intermediate optical depth case is considered in @xcite and can be suitable for a collapsar model when the disk mass reaches several solar masses .
we make our calculation in the ideal relativistic mhd approximation using an upwind conservative scheme that is based on a linear riemann solver and the constrained transport method to evolve the magnetic field .
the details of this numerical method and its testing results are expounded in @xcite .
the gravitational attraction of the black hole is introduced via kerr metric in the kerr - schild coordinates , @xmath33 .
we set the black hole mass as @xmath34 which corresponds to @xmath35 km ( we define @xmath36 ) .
the two - dimensional axisymmetric computational domain is @xmath37 , where @xmath38 , @xmath39 is the dimensionless angular momentum of the black hole ( @xmath40 is the black hole angular momentum ) , and @xmath41 km .
we adopt free - outflow as the outer and the inner boundary conditions .
the inner boundary is located just inside the event horizon and adopts the free - flow boundary conditions .
it is worthy of notice that the inner boundary is inside of the outer event horizon - this choice is possible since the horizon coordinate singularity is absent in the kerr - schild coordinates .
the total mass within the domain is small ( less than 25% ) as compared with the mass of the black hole , which allows us to ignore its self - gravity .
the grid is uniform over @xmath42 , where it has 320 cells , and is almost uniform over @xmath43 , where it has 459 cells , the linear size of each cell being the same in both directions . as the initial distribution we consider an equilibrium torus @xcite which is a `` torus '' of plasma with the black hole in the centre .
the value of the specific angular momentum @xmath44 determines the total effective potential which can be written as @xmath45 where @xmath46 and @xmath47 our numerical scheme can not operate with vacuum . to avoid this difficulty
, we have to introduce minimum possible values of the density and pressure that limit how small they can be .
we also check the magnetic field and do not allow it to drop below the value defined by the equations @xmath48 and @xmath49 . for the magnetization typical of the task considered these inequalities are usually satisfied with a margin of three orders .
the magnetic field strength is limited by the pair production by the neutrino - antineutrino annihilation and the compton up scattering pair production @xcite . through the horizon for the models : _ low torus momentum _ ( dashed line ) , _ neutrino cooling _
( solid line ) , and _ quadrupole 2 _ ( dot - dashed - line),width=321 ] the parameter we are most interested in is the energy release effectiveness @xmath50 where @xmath51 is the total energy flux at radius @xmath52 and @xmath53 is the accretion rate on the horizon of the black hole . in order to test the effectiveness dependence on the accretion conditions , we introduce several models with various initial magnetic fields and other parameters
. the initial magnetic field structure in the disk is mainly determined by the dynamo effect .
dynamo in accretion disks was investigated by many authors @xcite in the context of @xmath54-@xmath55 model .
depending on the dynamo parameters , stationary dipole or quadrupole magnetic fields were obtained .
some models showed time oscillations .
the calculations gave only a rough estimation of the magnetic field strength .
to reproduce this simulations , we have chosen three different topologies of the initial magnetic field ( similar to @xcite ) . due to a strong dependence of the toroidal field component on the poloidal one
, we initially introduce a purely poloidal magnetic field .
this field can be described by a vector potential with a single nonzero component @xmath56 . in our calculations
we use @xmath57 where @xmath58 and @xmath59 are the parameters of the vector potential .
we adopt a dipole initial magnetic field for _ low torus momentum _ , _ high torus momentum _ , _ neutrino cooling _ , _ schwarzschild _ , and _ low magnetized _ models .
quadrupole 1 _ possesses a quadrupole - like field generated by two dipoles : one is above and the second is below the equatorial plane .
_ quadrupole 2 _ model also has a quadrupole field , but the two dipoles are situated closer and farther from the black hole , which reproduces time oscillating magnetic field in the accretion disk . the topology of magnetic field is presented in fig .
[ top_field ] . in all the cases the initial field is normalized so that the maximum of the magnetic to gas pressure ratio @xmath60 does not exceed @xmath61 , except for the _ low magnetized _ model , where @xmath62 .
we have chosen the small value of @xmath63 in order to avoid strong influence of the magnetic pressure on the initial hydrodynamical equilibrium of the configuration .
the dimensionless angular momentum of the black hole is @xmath64 for all the models except for the _ schwarzschild _ one , where @xmath65 .
the collapsing torus specific angular momentum and mass are equal to @xmath66 for _ low torus momentum _ , _ neutrino cooling _ , _ quadrupole 1 _ , _ quadrupole 2 _ models and @xmath67 for _ high torus momentum _ , _ schwarzschild _ , _ low magnetized _ models .
in all the models , except for _ neutrino cooling _ , we neglect the neutrino cooling .
the summary of the model parameters see in tab . 1 .
[ cols="<,^,^,^,^,^,^,^,^,^",options="header " , ] here @xmath68 is the dimensionless angular momentum of the accreting matter ; @xmath69 accretion rate @xmath70 ; @xmath71 wind mass loss rate @xmath72 at radius @xmath52 ; @xmath73 , @xmath74 , @xmath75 are electromagnetic energy , matter energy , and the total energy fluxes at radius @xmath52 , @xmath76 ; @xmath77 accretion effectiveness ( [ eta ] ) ( for the _ low magnetized _ model we present separately the effectiveness of the electromagnetic energy extraction ) . notice that in the case of the _ low magnetized _ model the accretion starts only after @xmath78 s. , while the calculation ends at @xmath79 s.
topology and strength of the magnetic field play a crucial role in the process of accretion on a black hole . in our simulations
the energy release effectiveness varies over two orders of magnitude when we change the magnetic field topology from the dipole ( _ neutrino cooling _ , _ high torus momentum _
models ) to the quadrupole one(_quadrupole 1 _ , _ quadrupole 2 _ models ) . due to the 2d approach
we can not consider the dynamo effect .
however , we can see temporal increases of the magnetic field intensity owing to the mri - driven turbulence growth and magnetic line twist .
the blandford - znajek mechanism plays a leading role in accreting black hole energy release . for a force - free monopole magnetosphere
the blandford - znajek power is given by @xmath80 we assume @xmath81 , were @xmath82 is the angular velocity of the black hole and @xmath83 is the magnetic flux ( [ baushev ] ) threading the black hole .
our estimations of the energy release effectiveness differ by a factor of @xmath84 from @xcite and by a factor of up to @xmath85 from @xcite .
we explain it by distinctions between the initial magnetic fields : in @xcite , for instance , the magnetic field flux was @xmath86 times higher .
neutrino cooling processes significantly suppress the wind from the accretion disk , but do not diminish appreciably the energy release ; moreover , sometimes they can stabilize the outflow . the neutrino driven oscillations ( fig . [ oscil ] )
look like an artificial effect of turning on the initial conditions : more realistic conditions like @xcite do not lead to significant radial oscillations .
but even if the effect is real , the oscillations @xcite can scarcely be observed by gravitational wave detectors .
highly magnetized jets appear in all the models considered , where the black hole possesses angular momentum .
the obtained parameters of the plasma acceleration in the jets ( fig .
[ lor_c ] ) are in good agreement with @xcite . the boundary between the jet and the wind is unstable ( the perturbation growth is most pronounced in fig . [ lor_c ] , _ neutrino cooling _ case , in the south hemisphere in the radius interval @xmath87 .
@xcite reported about a similar effect ) .
the jet speed is much higher than the sound speed in the wind , and the boundary perturbations generate strong shock waves propagating to the wind region ( fig.[fl_sl ] ) .
the shocks effectively heat the wind and can produce a hot corona around the accretion disks like in the ss433 system @xcite .
if we adopt the effectiveness coefficient @xmath88 of the models _ neutrino cooling _ and _ high torus momentum _ , it allows us to estimate the total energy release of the torus accretion as @xmath89 erg .
this value of @xmath77 however , need not to be universal : as we saw , other magnetic field topologies can have a significantly different efficiency .
the scenario considered in this article is a possible alternative to the magnetar - driven hypernova scheme @xcite .
both the scenarios predict the appearance of collimated jets energetic enough to explain grb as well as hypernovae phenomena .
accurate calculations are needed to find characteristic features and distinguish these two schemes . by
now we can say that the chemical composition of matter in the jets should be different . in the collapsar scenario
we expect a highly poynting - dominated jet ( @xmath90 , where @xmath91 and @xmath92 are the field and the matter energy flux densities , respectively ) , mainly made up of @xmath21-pairs , while in the magnetar - driven scenario the jet is baryon - dominated and intermediately magnetized ( @xmath93 )
. the effect of the magnetic field sign change found in the jet of the _ quadrupole 2 _ model , may be the key to understanding of the `` surprising evolution of the parsec - scale faraday rotation gradients in the jet of the bl lac object b1802 + 784 '' @xcite .
complicated topology of the magnetic field in the accretion disk can lead to variations of the magnetic field direction in the jet . in this case
, the jet constitutes a chain of regions with various magnetic field orientations .
the accretion rate and energy release are governed by four main parameters : magnetic field structure , angular momentum of the black hole , specific angular momentum of the accreting matter , accretion disk mass or accreting part of the star envelop . to have a long - time accretion , angular momentum of the accreting matter should be high enough to form an accretion disk around the black hole . as it has been demonstrated by the _
schwarzschild _ model ,
if the black hole has deficient angular momentum , the powerful magnetized jet does not appear at all , which totally changes the picture of the accretion .
moreover , a comparison of the _ low _ and _ high torus momentum _
models shows us that a respectively small variation of the angular momentum of the accreting matter leads to a change of the accretion efficiency .
neutrino losses do not prevent energy release and can even stabilize the magnetized jet outflow .
the least understood and a very important factor defining the accretion efficiency is the disk magnetic field structure . as we could see from the calculations , a mere replacing of the dipole field by the quadrupole one decreases the energy release by two orders of magnitude .
the magnetic flux @xmath83 governs the energy extraction from ergosphere ( [ ebz ] ) , while the black hole magnetosphere is formed and supported by the disk . in our calculations we preset the initial magnetic field in the torus
it is clear , however , that self - consistent numerical calculations of the magnetic dynamo in the disk are necessary in order to elucidate the problem .
nevertheless , our calculations show that a very intensive electromagnetic and matter energy flux can be evolved during massive torus accretion on a rotating black hole with mass @xmath94 .
the total energy budget is quite sufficient to explain hypernova explosions like grb 980425 or grb 030329 .
the magnetized matter accretion on a rotating black hole is a very promising model of the central engine of these phenomena .
in addition , we have found that the instability of the boundary between the jet and the wind generates shock waves propagating to the wind region and heating it .
this effect can give rise to a hot corona around the binary system similar to ss433 .
we would like to express our deep gratitude to prof .
s. s. komissarov for helping in the computational code creation , and for very fruitful discussions .
we thank prof .
g. s. bisnovatyi - kogan for helpful discussions .
the calculations were fulfilled at the st .
andrews uk mhd cluster and at the white rose grid facilities ( everest cluster ) .
this work was supported by the pparc ( grant `` theoretical astrophysics in leeds '' ) and by the rfbr ( russian foundation for basic research , grant 08 - 02 - 00856 ) . | we present numerical simulations of the axisymmetric accretion of a massive magnetized plasma torus on a rotating black hole .
we use a realistic equation of state , which takes into account neutrino cooling and energy loss due to nucleus dissociations .
we simulated various magnetic field configurations and torus models , both optically thick and thin for neutrinos .
it is shown that the neutrino cooling does not significantly change either the structure of the accretion flow or the total energy release of the system .
the calculations evidence heating of the wind surrounding the collapsar by the shock waves generated at the jet - wind border .
this mechanism can give rise to a hot corona around the binary system like ss433 .
angular momentum of the accreting matter defines the time scale of the accretion . due to the absence of the magnetic dynamo in our calculations ,
the initial strength and topology of the magnetic field determines magnetization of the black hole , jet formation properties and the total energy yield .
we estimated the total energy transformed to jets as @xmath0 ergs which was sufficient to explain hypernova explosions like grb 980425 or grb 030329 .
gamma - rays : bursts , methods : numerical , ( magnetohydrodynamics ) mhd , black hole physics . |
the seminal experiments on electricity and magnetism in the @xmath1@xmath2 centuries were codified into a set of physical laws , gauss s law , ampre s law , faraday s law , and thompson s law , that form the foundation of early classical electromagnetism .
in the course of time , a crisis emerged when it was determined that ampre s law is not consistent with the continuity equation for charge and current . in 1865 , maxwell @xcite proposed a modification to ampre s law that resolved the inconsistency and established a new paradigm for electrodynamics that led to the prediction of electromagnetic wave phenomena . over two decades later , hertz @xcite demonstrated the existence of electromagnetic waves , providing experimental validation for the maxwell theory . with the replacement of ampre s law by the maxwell ampre law , the collective laws of classical electrodynamics are known as maxwell s equations , appearing in a microscopic formulation that applies to microscopic fields in the vacuum and in a macroscopic formulation as equations of motion for averaged fields in continuous media .
the maxwell equations for the components of the fields were recast as vector equations by heaviside and by minkowski . the maxwell
minkowski equations are the dominant paradigms of microscopic classical electrodynamics and of macroscopic classical continuum electrodynamics . for over fifteen decades ,
maxwellian continuum electrodynamics has been extraordinarily successful in predicting and describing real - world phenomena that occur in dielectric materials .
there are , however , anomalies .
the anomalies include the abraham minkowski momentum controversy @xcite and rosen s refractive index - dependent special theory of relativity @xcite .
it has long been known that the macroscopic maxwell minkowski equations are inconsistent with conservation of total momentum : that is the origin story of the century - old abraham minkowski controversy @xcite .
the difference between the total momentum , which is decided by temporal invariance of the total momentum in a thermodynamically closed system , and the electromagnetic momentum that is systematically derived from the macroscopic maxwell - minkowski equations has generally been attributed to the existence of a material momentum that is associated with the movement of matter in either a macroscopic or a microscopic picture of the dielectric @xcite .
while the physical origin and characteristics of the phenomenological material momentum have been debated for decades , experimental demonstrations of a material momentum has been variously inconsistent , contradictory , and refuted @xcite likewise , theoretical treatments of the abraham minkowski controversy @xcite have not provided a conclusive consistent physical solution . in 2007 ,
pfeifer , nieminen , heckenberg , and rubinsztein - dunlop @xcite reviewed the state of the field and concluded that the issue had been settled for some time : `` when the appropriate accompanying energy momentum tensor for the material medium is also considered , experimental predictions of the various proposed tensors will always be the same , and the preferred form is therefore effectively a matter of personal choice . '' in 2010 , barnett @xcite offered a more restrictive resolution of the abraham minkowski conundrum by contending that the total momentum for the medium and the field is composed of either the minkowski canonical momentum or the abraham kinetic momentum , each supplemented by the appropriate canonical or kinetic material momentum . according to the scientific method , a scientific hypothesis should result in a unique testable prediction of physical quantities .
there are many examples in the extant experimental record in which the interpretation of momentum experiments is unrestricted with experiments that prove the minkowski electromagnetic momentum later being shown to prove the abraham electromagnetic momentum and vice versa @xcite .
likewise , experiments that disprove the minkowski momentum are later re - analyzed to confirm the minkowski momentum and similarly for the abraham momentum @xcite .
although the barnett @xcite and pfeifer , nieminen , henckenberg , and rubinsztein - dunlop @xcite theories are based on fundamental principles , barnett s restriction to two simultaneous physically motivated momentums is not compatible with the mathematical tautology that underlies pfeifer , _
et al_. the non - uniqueness of the electromagnetic momentum and the non - uniqueness of the material momentum that are derived for the propagation of light through a dielectric are contrary to popper s criterion of falsifiability . in a thermodynamically closed system ,
the total momentum , like the total energy , is a known quantity that is uniquely determined by temporal invariance .
then the construction of the unique total energy momentum tensor gets us right with the scientific method and definitively resolves the abraham
minkowski controversy .
except that there are problems here , too . applying the four - divergence operator to the total energy
momentum tensor , one obtains conservation laws for the total energy and the total linear momentum of an antireflection - coated simple linear medium that is positioned in free space @xcite .
it is easily argued that the conservation law for the total energy , so obtained , can not be incommensurate with the poynting theorem that is systematically derived from the maxwell field equations .
it is also easily proven that _ i _ )
the energy conservation law , derived from the four - divergence of the total energy momentum tensor , is self - inconsistent because its two non - zero terms depend on different powers of the refractive index and _ ii _ ) the conservation law for the total energy that is derived from the four - divergence of the total energy
momentum tensor is undeniably incommensurate with the poynting theorem @xcite .
the current author @xcite made the _
ansatz _ of a ravndal @xcite refractive index - dependent material four - divergence operator and demonstrated consistency between the field equations and total energy
momentum conservation , at the cost of apparent problems with special relativity .
then we have fixed the abraham minkowski momentum problem only to have it re - appear in a different form elsewhere .
we must go deeper into the relations between electrodynamics , conservation , and special relativity in a simple linear dielectric medium in order to obtain any meaningful resolution of the abraham
minkowski controversy .
classical continuum electrodynamics can be treated as a formal system in which the macroscopic maxwell minkowski equations and the constitutive relations are the axioms .
theorems are derived from the axioms using algebra and calculus .
poynting s theorem of energy continuity is one example of a theorem of axiomatic continuum electrodynamics as is the continuity equation for the minkowski momentum .
the scalar energy continuity equation and the scalar component equations of the vector minkowski momentum continuity equation can be written using linear algebra as a single matrix differential equation that is a valid theorem within the formal system of maxwellian continuum electrodynamics . in this article
, we derive valid theorems of the formal theory of maxwellian continuum electrodynamics from explicitly stated axioms using simple substitutions of explicitly defined quantities ; there are no implicit axioms and no manifest deficiencies in the derivations . noting that conservation laws , special relativity , and the maxwell field equations are distinct laws of physics , we show that valid theorems of maxwellian continuum electrodynamics are proven false by the energy
momentum conservation laws for an unimpeded , inviscid , incompressible continuous flow of non - interacting particles , photons .
we also show that valid theorems of the formal theory of maxwellian continuum electrodynamics are proven false by von laue s application of the einstein theory of special relativity to a dielectric . when a valid theorem of an axiomatic formal theory is proven false by an accepted standard
, then it is proven that one or more axioms of the formal theory are false .
alternatively the standard with which the theorem is being compared is false , or both the theory and the standard can be simultaneously false . therefore
, the extant theoretical treatments of energy momentum conservation , special relativity , and continuum electrodynamics must be regarded as being mutually inconsistent in a region of space in which the effective speed of light is @xmath0 .
this is a _ significant result of this article _ and it was clearly presaged by the abraham
minkowski controversy @xcite and by the contradiction between the von laue @xcite and rosen @xcite treatments of special relativity in a dielectric .
having proven the existing theories to be inconsistent , it is customary to propose alternatives . in the continuum limit ,
the field and dielectric are continuous at all length scales .
we define an isotropic , homogenous , absorptionless , linear dielectric - filled , flat , non - minkowski , continuous material spacetime @xmath3 , where the time - like coordinate @xmath4 is identified with @xmath5 . in this spacetime ,
the basis functions of propagating electromagnetic fields @xmath6 are what we would expect for monochromatic , or quasimonochromatic , fields propagating at speed @xmath0 in a simple linear dielectric , instead of @xmath7 that are the basis functions of maxwellian electrodynamics .
consequently , there is a fundamental difference between an organically continuum - based electrodynamics and microscopic fields interacting with individual particles in the vacuum .
as in statistical and thermal physics , the macroscopic model and the microscopic model are fundamentally disjoint : the assumptions , approximations , and averages that are implicit in the macroscopic model are not reversible .
we apply lagrangian field theory in the isotropic , homogeneous , linear dielectric - filled , flat , non - minkowski , continuous material spacetime @xmath3 to systematically and rigorously derive equations of motion for macroscopic fields in a simple linear dielectric .
these equations of motion are the axioms of a new formal theory of continuum electrodynamics .
there is sufficient commonality between the new equations of motion and the macroscopic maxwell
minkowski equations that the extensive theoretical and experimental work that is correctly described by the macroscopic maxwell
minkowski equations has an equivalent formulation in the new theory .
in particular , the wave equation has the same form in both versions of continuum electrodynamics .
more interesting is the work that we can do with the new formalism of continuum electrodynamics in a thermodynamically closed system that is improperly posed in the standard maxwell theory of continuum electrodynamics .
these cases will typically involve the invariance or tensor properties of the set of coupled equations of motion for the macroscopic fields .
this interpretation is borne out in our common experience : the macroscopic maxwell
minkowski equations produce exceedingly accurate experimentally verified predictions of simple phenomena like reflection , refraction , wave velocity , _ etc .
_ , but fail to render a unique , uncontroversial , verifiable prediction of energy momentum conservation .
conservation laws are distinct from the energy
momentum evolution theorems that are systematically derived from the maxwell field equations . in continuum dynamics , the continuity equations , often called conservation laws , are derived for the case an unimpeded , inviscid , incompressible continuous flow of non - interacting particles ( dust , fluid , photons , _ etc .
_ ) in an otherwise empty volume @xcite . applying the divergence theorem to a taylor series expansion of a density field of a conserved property in an empty minkowski spacetime results in a continuity equation for the conserved property .
we show that applying the same derivation procedure to a non - empty , linear dielectric - filled , isotropic , homogeneous , flat , non - minkowski , continuous material spacetime produces a continuity equation for a conserved property of a simple linear dielectric that requires differentiation with respect to the independent variable @xmath4 , which is identified with @xmath5 , instead of @xmath8 .
the difference can be attributed to the fact that the photon fluid is not incompressible : there is a change in energy density , and a corresponding change in photon density , as the field crosses the interface between the vacuum and a simple linear dielectric .
we then construct the diagonally symmetric , traceless total energy momentum tensor as an element of a valid theorem of the new formal theory of continuum electrodynamics .
we demonstrate that the spatial integrals of the total energy density and the total momentum density are temporally invariant and that valid theorems of the new formal theory correspond to continuity equations for the conserved properties in a non - empty , dielectric - filled spacetime .
the tensor properties of the new formulation of continuum electrodynamics constitute a definitive resolution of the abraham minkowski dilemma . the unique total energy
momentum tensor that is derived by the new formal theory is entirely electromagnetic in nature .
consequently , there is no need to assume any `` splitting '' of the total energy , the total momentum , or the total angular momentum into field and matter subsystems in order to satisfy the energy
momentum conservation laws for a simple linear dielectric : _ i _ ) in the case of quasimonochromatic optical radiation incident on a stationary homogeneous simple linear dielectric draped with a gradient - index antireflection coating , the surface forces are negligible .
then , the total energy , the total linear momentum , and the total angular momentum are purely electromagnetic and the dielectric remains internally and externally stationary .
_ ii _ ) in the absence of an antireflection coating , the dielectric block acquires momentum due to the optically induced surface pressure that is associated with fresnel reflection and this will be treated by boundary conditions , not by a hypothetical internal material momentum , in a separate publication .
special relativity is the remaining issue on our list of three theories that are mutually inconsistent in a dielectric , the other two being macroscopic electrodynamics and conservation laws .
einstein s theory of special relativity was derived using coordinate transformations between inertial coordinate systems in an infinite vacuum spacetime .
we can , in principle , start with einstein s theory of relativity for microscopic particles in an infinite vacuum spacetime , as a fundamental basis , and describe the interaction of each particle of the material with every other particle and describe how the interaction of every particle with the microscopic field affects the propagation of a light signal .
if we were even capable of such a monumental task , we would then justify some assumptions about limiting behavior and make some approximations and averages that would allow an effective macroscopic description of the full microscopic treatment .
typically , the process will invoke the continuum limit , evaluating discrete sums by integrals and extending the region of integration to spatial infinity @xcite . here , we derive a theory of relativity based on the propagation of a light signal in an infinite , isotropic , homogeneous , continuous dielectric spacetime in which the speed of light is @xmath0 and make a very reasonable assumption that the theory holds for the large finite material spacetime that is a model for a simple linear dielectric .
the result confirms rosen s @xcite phenomenological index - dependent lorentz factor .
this result is consistent with the symmetry of the new equations of motion that are derived from first principles for macroscopic fields in a dielectric - filled spacetime .
we also show how the macroscopic rosen theory of dielectric special relativity can be related to the von laue theory of dielectric special relativity that is based on einstein s theory of special relativity in a vacuum .
the mutually consistent alternative formulations of three connected physical principles , relativity , conservation , and electrodynamics , in a dielectric medium that is represented by a continuous , linear dielectric - filled , flat , non - minkowski material spacetime allows a self - consistent , fully macroscopic , fully electromagnetic resolution of the abraham
minkowski controversy .
the axioms of the formal theory of continuum electrodynamics are the macroscopic maxwell minkowski equations @xmath9 @xmath10 @xmath11 @xmath12 [ eqt2.01 ] and the constitutive relations @xmath13 @xmath14 [ eqt2.02 ] for an absorptionless linear medium with no free charges or free currents .
the exact mechanisms that create the macroscopic polarization and macroscopic magnetization are fundamentally outside the scope of continuum electrodynamics : by definition , the microscopic constituents are treated as having been averaged in a continuum limit and are represented by continuum abstractions for the permittivity @xmath15 and the permeability @xmath16 . for clarity and concision ,
we treat a limiting case in which the pulse is sufficiently monochromatic and the center frequency of the exciting field is sufficiently far from material resonances that dispersion can be treated parametrically .
we define a simple linear dielectric as an idealized model of a linear dielectric that can be treated as having a real , time - independent refractive index in the parameter region of interest .
we limit our attention to simple linear dielectric media and take @xmath17 @xmath18 [ eqt2.03 ] where @xmath19 is real and time - independent , as the final two axioms of our theory of dielectric continuum electrodynamics .
the rules for the construction of valid theorems from the axioms are algebra and calculus . for a simple linear dielectric , we can write the macroscopic maxwell minkowski equations as @xmath20 @xmath21 @xmath22 @xmath23 [ eqt2.04 ] by the direct substitution of eqs .
( [ eqt2.02 ] ) and ( [ eqt2.03 ] ) into the macroscopic maxwell minkowski equations , eqs .
( [ eqt2.01 ] ) .
equations ( [ eqt2.04 ] ) are valid theorems of the formal theory of maxwellian continuum electrodynamics , individually and collectively .
we can re - parameterize the dielectric maxwell
minkowski equations , eqs .
( [ eqt2.04 ] ) , using a simple change of variable @xmath24 substituting eq .
( [ eqt2.05 ] ) into eqs .
( [ eqt2.04 ] ) , we obtain @xmath25 @xmath26 @xmath27 @xmath28 [ eqt2.06 ] despite their unusual appearance , eqs .
( [ eqt2.06 ] ) are valid theorems of the formal theory of maxwellian continuum electrodynamics , individually and collectively .
specifically , eqs .
( [ eqt2.06 ] ) are mathematically equal to the dielectric maxwell minkowski equations , eqs .
( [ eqt2.04 ] ) , with a simple change of notation that is effected by eq .
( [ eqt2.05 ] ) . consequently , eqs .
( [ eqt2.06 ] ) are valid for the same physical configurations and the same conditions as eqs .
( [ eqt2.04 ] ) and do not suffer from any limiting behavior or any boundary conditions that do not also affect eqs .
( [ eqt2.04 ] ) in the same manner .
the equations of motion for macroscopic electromagnetic fields in a dielectric , eqs .
( [ eqt2.06 ] ) , were the penultimate results of ref .
however , those results were based on a phenomenological construction of the evolution equations for energy and momentum based on temporal invariance of the total energy and the total momentum in a thermodynamically closed system and the _ ansatz _ of a ravndal @xcite index - dependent four - divergence operator . the unusual appearance of eqs . ( [ eqt2.06 ] ) and the implications of eqs .
( [ eqt2.06 ] ) for energy momentum conservation and for special relativity led to questions about the derivation in ref .
@xcite , specifically questions about whether there were implicit axioms or other errors involved in the derivation .
the point of casting the derivation in terms of the formal theory of continuum electrodynamics , here , is so that there can be no question about the derivation : it is a two - line derivation consisting of the simple substitution of the re - parameterization from eq .
( [ eqt2.05 ] ) into the theorem , eq .
( [ eqt2.04 ] ) , that is systematically derived by simple and direct substitution of explicitly stated axioms , eqs .
( [ eqt2.02])([eqt2.03 ] ) into the maxwell
minkowski equations , eqs .
( [ eqt2.01 ] ) . in the following section , sec .
iii , we will examine the validity of eqs .
( [ eqt2.06 ] ) that were derived phenomenologically in ref .
@xcite and were derived formally , just above , in the current article .
we will find that _ i _ ) the energy
momentum matrix differential equation that is systematically and rigorously derived from eqs .
( [ eqt2.06 ] ) is the total energy
momentum tensor continuity equation , _ ii _ ) the total energy
momentum tensor continuity equation is incommensurate with , and directly contradicts , the conservation laws of total energy and total momentum in an unimpeded continuous flow , and _ iii _ ) the symmetry of eqs .
( [ eqt2.06 ] ) imposes the rosen index - dependent lorentz factor that is incommensurate with and is directly contradicted by the von laue theory of special relativity in a dielectric .
obviously , the field equations can not be inconsistent with conservation principles and special relativity ; this is simply a further manifestation of the abraham
minkowski controversy .
after presenting the new facets of the controversy , the remainder of the article is devoted to a self - consistent description of macroscopic electrodynamics , energy momentum conservation , and special relativity in a region of space in which the speed of light is @xmath0 .
the abraham minkowski controversy involves a long - standing problem of a non - unique momentum and a non - unique energy momentum tensor for electromagnetic fields in a dielectric .
it has been the practice for the theoretically indeterminate momentum properties of the electromagnetic system to motivate the assumption of a material momentum in the microscopic structure of the dielectric in order to preserve compliance with the temporal invariance of total momentum @xcite , even though the assumption of a microscopic material subsystem is adverse to the continuum limit that is the foundational principle of continuum electrodynamics .
we continue with the formal theory of maxwellian continuum electrodynamics and systematically and rigorously derive a matrix differential equation from the field equations , eqs .
( [ eqt2.06 ] ) .
we subtract the scalar product of eq .
( [ eqt2.06c ] ) with @xmath29 from the scalar product of eq .
( [ eqt2.06a ] ) with @xmath30 and apply common vector identities to produce a valid theorem @xmath31 of maxwellian continuum electrodynamics for our simple dielectric medium .
we then form another valid theorem @xmath32 @xmath33 from the sum of cross products of eqs .
( [ eqt2.06a ] ) and ( [ eqt2.06c ] ) with macroscopic fields . in
what follows , the index convention for greek letters is that they belong to @xmath34 and lower case roman indices from the middle of the alphabet are in @xmath35 .
coordinates @xmath36 are @xmath37 , as usual . substituting eq .
( [ eqt2.06b ] ) multiplied by @xmath29 and eq .
( [ eqt2.06d ] ) multiplied by @xmath30 into eq .
( [ eqn3.02 ] ) , we obtain a vector differential equation , @xmath38 in component form . here , @xmath39 is a diagonally symmetric matrix with components @xmath40 equation ( [ eqn3.03 ] ) is another valid theorem of maxwellian continuum electrodynamics . as a matter of linear algebra , we can row - wise write the scalar differential equation , eq .
( [ eqn3.01 ] ) , and the three scalar differential equations that comprise the vector differential equation , eq .
( [ eqn3.03 ] ) , as a single matrix differential equation @xmath41 where the material four - divergence operator @xcite @xmath42 is defined by context and is distinct from the usual four - divergence operator @xmath43 also , @xmath44 is a four - force density and @xmath45 \label{eqn3.09}\ ] ] is a traceless diagonally symmetric matrix . the matrix differential equation , eq .
( [ eqn3.05 ] ) , is a valid theorem of the formal theory of maxwellian continuum electrodynamics in which eqs .
( [ eqn3.06 ] ) , ( [ eqn3.08 ] ) , and ( [ eqn3.09 ] ) are useful combinations of operators and variables that are used for notational convenience .
now , let us perform a similar derivation from the equations of motion , eqs .
( [ eqt2.04 ] ) , that are mathematically equal to eqs .
( [ eqt2.06 ] ) . taking the scalar product of eqs .
( [ eqt2.04a ] ) and ( [ eqt2.04c ] ) with macroscopic fields and subtracting the results , we obtain @xmath46 we denote the macroscopic electromagnetic energy density @xmath47 and the poynting energy flux vector @xmath48 as usual , to obtain poynting s theorem @xmath49 poynting s theorem , eq .
( [ eqn3.13 ] ) is a valid theorem of the formal theory of maxwellian continuum electrodynamics , as is eq .
( [ eqn3.10 ] ) .
we can form another valid theorem @xmath50 from the sum of cross products of eqs .
( [ eqt2.04a ] ) and ( [ eqt2.04c ] ) with macroscopic fields .
substituting eq .
( [ eqt2.04b ] ) multiplied by @xmath29 and eq .
( [ eqt2.04d ] ) multiplied by @xmath51 into eq .
( [ eqn3.14 ] ) and writing @xmath52 yields @xcite @xmath53 yet another valid theorem of maxwellian continuum electrodynamics . as a matter of linear algebra , we can write , row - wise , eq .
( [ eqn3.10 ] ) and the three scalar differential equations that comprise the vector differential equation , eq .
( [ eqn3.16 ] ) , as a single matrix differential equation @xmath54 where @xmath55 is the usual four - divergence operator defined by eq .
( [ eqn3.07 ] ) , @xmath56 is the minkowski four - force density , and @xmath57 \label{eqn3.19}\ ] ] is , by construction , a four - by - four matrix . the minkowski matrix differential equation , eq .
( [ eqn3.17 ] ) , is a valid theorem of the formal theory of maxwellian continuum electrodynamics . for historical reasons , the four - by - four matrix , eq .
( [ eqn3.19 ] ) , is known as the minkowski energy
momentum tensor .
a number of other formulations for the electrodynamics of fields , energy , and momentum in a dielectric appear in the scientific literature @xcite . in order to provide a basis for comparison with aspects of the prior work
, we use the formal theory of maxwellian continuum electrodynamics to derive the abraham energy momentum theory @xcite that was contemporaneous with the minkowski theory @xcite .
we subtract @xmath58 from both sides of eq .
( [ eqn3.14 ] ) .
we combine that result , row - wise , with eq .
( [ eqn3.10 ] ) to obtain a new matrix differential equation @xmath59 that is also valid theorem of the formal theory of maxwellian continuum electrodynamics , where @xmath60 \label{eqn3.22}\ ] ] is a diagonally symmetric four - by - four matrix and @xmath61 is the abraham four - force density .
although the time average of the force given by eq .
( [ eqn3.20 ] ) is essentially zero @xcite , the force should not be treated as negligible because that would mean that the left - hand side of eq .
( [ eqn3.14 ] ) is also negligible thereby limiting the theory to non - propagating static fields .
the matrix , eq .
( [ eqn3.22 ] ) , is commonly known as the abraham energy
momentum tensor .
conservation laws are distinct from the energy
momentum theorems that were systematically derived from the maxwell field equations in the preceding subsection . in continuum dynamics , a continuity equation reflects the conservation of a scalar property in the continuum limit of an unimpeded inviscid incompressible flow of non - interacting particles ( dust , fluid , photons , etc . ) in terms of the equality of the net rate of flux out of the volume and the time rate of change of the property density field @xcite . for
a conserved scalar property , the continuity equation of the property density @xmath62 @xmath63 is derived by applying the divergence theorem to a taylor series expansion of the property density field @xmath62 and the property flux density field @xmath64 to unimpeded flow in an otherwise empty volume @xcite . for unimpeded flow in a thermodynamically closed system , we have a conserved scalar property , the total energy , and a conserved vector property , the total momentum , belonging to the same closed system . as a matter of linear algebra
, we can construct a total energy
momentum tensor @xmath65 with the following characteristics that cause the conservation laws to be obeyed in an unimpeded inviscid incompressible flow of non - interacting particles in the continuum limit : \1 ) the continuity equations of the total energy and total momentum are generated by the four - divergence of the total energy
momentum tensor @xcite , @xmath66 continuity equations are sometimes known as conservation laws , but there are additional conditions on conserved properties in a continuous flow .
\2 ) the integrals over all - space @xmath67 of the total energy density @xmath68 and the total linear momentum density @xmath69 are invariant in time @xcite .
if @xmath70 is not invariant in time then @xmath71 is not the total energy
momentum tensor of a thermodynamically closed system .
temporal invariance of total energy and total linear momentum are sometimes known as conservation laws , but there are additional conditions on conserved properties in a continuous flow .
\3 ) a non - symmetric energy
momentum tensor corresponds to an unbalanced shear force and is not consistent with conservation of total angular momentum .
the total energy
momentum tensor is symmetric @xmath72 insuring conservation of total angular momentum @xcite .
if @xmath71 is not symmetric then @xmath71 is not the total energy momentum tensor of a thermodynamically closed system .
there are arguments in the abraham
minkowski controversy literature that a symmetric energy
momentum tensor is not necessary for conservation of total angular momentum , although such magic always comes with a price .
for example , obukhov @xcite describes an ` abrahamization ' procedure to obtain a symmetric energy momentum tensor by splitting a non - symmetric energy
momentum tensor into two parts coupled by a ` force ' , such as eq .
( [ eqn3.20 ] ) .
these forces are not always negligible as discussed at the end of the preceding subsection , section iii , subsection a. \4 ) the trace of the total energy
momentum tensor is proportional to the mass density @xmath73 : @xmath74 for light , the trace of the energy
momentum tensor is zero corresponding to the continuum limit of a flow of massless non - interacting photons @xcite .
\5 ) substituting the symmetry condition , eq .
( [ eqn3.27 ] ) , into eq .
( [ eqn3.25 ] ) , we obtain a redundant condition @xmath75 that is often seen in place of eq .
( [ eqn3.25 ] ) @xcite .
however , it should be remembered that eq .
( [ eqn3.29 ] ) is conditional on eq .
( [ eqn3.25 ] ) and on symmetry of the energy
momentum tensor , eq .
( [ eqn3.27 ] ) . as a matter of linear algebra ,
we can construct a matrix differential equation , eq .
( [ eqn3.25 ] ) , from four continuity equations in the form of eq .
( [ eqn3.24 ] ) .
each continuity equation corresponds to a row of the matrix differential equation .
we could , in principle , construct a matrix differential equation from four continuity equations of different physical systems or subsystems .
this is prevented here by the conditions on the symmetry and trace , eqs .
( [ eqn3.27 ] ) and ( [ eqn3.28 ] ) , that couple the rows of @xmath65 .
the continuity equations that correspond to the rows of @xmath65 are likewise coupled by these conditions .
if all these requirements are met then @xmath65 is the total energy momentum tensor of an isolated , thermodynamically closed system .
the macroscopic maxwell
minkowski equations and conservation of energy momentum are distinct fundamental laws of physics . in section
iii , subsection a , we derived equations of motion related to the flow of energy and momentum in a simple linear dielectric as valid theorems of the macroscopic maxwell minkowski equations . here , we compare the results that were systematically derived from the macroscopic maxwell
minkowski equations with the conditions of section iii , subsection b for conservation of energy and momentum in an unimpeded inviscid incompressible flow of non - interacting photons in the continuum limit . in the plane - wave limit , eq .
( [ eqn3.17 ] ) becomes @xmath76 in agreement with the conservation condition 1 ) , eq .
( [ eqn3.25 ] ) . in that limit
, it can be asserted that eqs .
( [ eqn3.10 ] ) and ( [ eqn3.14 ] ) are electromagnetic energy and momentum conservation laws .
it is also frequently claimed in the scientific literature that the minkowski linear momentum is nearly conserved @xcite , in general , and the transient nature of the minkowski four - force , eq .
( [ eqn3.18 ] ) , supports that assertion . at this point
, we would like to make it very clear that these assertions are false : the minkowski momentum , when the field is in the dielectric , is a factor of @xmath77 greater that the incident momentum strongly violating the conservation condition 2 ) , eq .
( [ eqn3.26 ] ) , that the total linear momentum is temporally invariant .
1 is a graphical representation of a quasimonochromatic field in the vacuum traveling to the right at some time @xmath78 before entering a dielectric medium .
the representation is in terms of the envelope of the vector potential , before entering the dielectric .
the field , on the left , is traveling to the right .
the dielectric , draped with a gradient - index anti - reflection coating , is represented by the outline of the refractive index ( n-1 ) on the right . ]
figure 2 is a representation of the same field at some time @xmath80 after it has entered a linear isotropic homogenous dielectric through a gradient - index antireflection coating . when the field is entirely within the dielectric . ]
finite - difference time - domain integration of the wave equation was used to propagate the vector potential in the plane - wave limit @xcite . according to the figures ,
the amplitude of the vector potential in the dielectric is the amplitude of the incident vector potential divided by @xmath81 and the nominal width of the pulse in the dielectric is the width of the incident pulse reduced by a factor of @xmath77 . numerically integrating , we find that @xmath82 the transients at the leading and trailing edges of the pulse are included in the numerical integration . for our quasimonochromatic field
, the width can be taken as full - width half - maximum and we can visually integrate the fields in figs . 1 and 2 to obtain the same result as the numerical integration , eq .
( [ eqn3.31 ] ) . for quasimonochromatic fields in the continuous wave limit
, we define the amplitude of the vector potential @xmath83 by @xmath84 .
the amplitudes of the fields , @xmath85 , @xmath86 , @xmath87 , are defined in like manner .
then , @xmath88 , @xmath89 , and @xmath90 . substituting the relations between the fields and the vector potential amplitudes @xcite into the minkowski electromagnetic energy formula @xmath91 and the minkowski electromagnetic momentum formula @xmath92 and comparing the results with eq .
( [ eqn3.31 ] ) , we find that the minkowski electromagnetic energy is conserved but the minkowski electromagnetic momentum is greater than the incident momentum by a factor of @xmath77
. it could not be otherwise because eqs .
( [ eqn3.32 ] ) and ( [ eqn3.33 ] ) have the same quadratic dependence of the fields but differ by a factor of @xmath77 in magnitude .
consequently , @xmath93 is very different from the total momentum and @xmath94 is not and approximation of the total energy momentum tensor .
let us now perform of a similar analysis using the theorem , eq .
( [ eqn3.05 ] ) . due to the temporal invariance demonstrated by eq .
( [ eqn3.31 ] ) , both the electromagnetic energy @xmath95 and the electromagnetic linear momentum @xmath96 are temporally invariant satisfying the condition 2 ) , eq .
( [ eqn3.26 ] ) , for conservation of total energy
momentum in a closed system .
the matrix @xmath97 , eq .
( [ eqn3.09 ] ) , is symmetric proving conservation of total angular momentum , condition 3 ) , eq .
( [ eqn3.27 ] ) in a closed system .
furthermore , the trace of @xmath97 is zero , thereby satisfying condition 4 ) , eq .
( [ eqn3.28 ] ) , for the energy momentum tensor of a system that is entirely composed of massless photons . in the ordinary course of physics
, transients are generally treated as negligible . in this limit ,
( [ eqn3.05 ] ) , the time - like coordinate of @xmath98 is dependent on the refractive index making eq .
( [ eqn3.36 ] ) incommensurate with the conservation condition 1 ) , eq .
( [ eqn3.25 ] ) . if the transient forces in eqs .
( [ eqn3.05 ] ) and ( [ eqn3.17 ] ) are negligible then fundamental laws of physics are inconsistent as shown just above .
then we are forced to conclude that the transient forces are non - negligible and we now examine the consequences for that case . for a given electromagnetic energy momentum four - tensor @xmath99 that is systematically derived from the macroscopic maxwell minkowski equations , there is an associated four - force density @xmath100 such that @xmath101 because our system is thermodynamically closed , the four - force @xmath100 represents coupling to a sub - system @xcite . because the four - force is associated with the gradient of the refractive index ,
the dynamics of the sub - system are based on a material four - tensor @xmath102 such that @xmath103 the total energy momentum tensor @xmath104 must obey the conservation law 1 ) , eq .
( [ eqn3.25 ] ) , @xmath105 a wide variety of physical models have been employed in an effort to fully resolve the problem of momentum conservation in a dielectric @xcite .
selected examples are discussed in the next subsection of this article .
typically , one assumes a microscopic model of the material dynamics in a dielectric and applies an averaging technique to derive the macroscopic momentum of the material .
the correctness of the results are affirmed by the fundamental nature of the physical laws that are used as the basis of the analysis .
adding the electromagnetic and material tensors , one obtains the total energy momentum tensor for the thermodynamically closed system .
the total linear momentum , eq .
( [ eqn3.35 ] ) and the total energy , eq .
( [ eqn3.34 ] ) , are known quantities by temporal invariance . using the conservation laws , eqs .
( [ eqn3.25])([eqn3.28 ] ) to relate the total energy density and the total momentum density to elements of the total energy
momentum tensor , we write @xmath106 \ , .
\label{eqn3.41}\ ] ] the total energy momentum tensor , eq .
( [ eqn3.41 ] ) is diagonally symmetric and traceless .
consequently , the conditions for conservation 3 ) and 4 ) , eqs .
( [ eqn3.27 ] ) and ( [ eqn3.28 ] ) are satisfied in addition to condition 1 ) , eq .
( [ eqn3.25 ] ) , that is satisfied by definition , eq . ( [ eqn3.40 ] ) . due to the temporal invariance demonstrated by eq .
( [ eqn3.31 ] ) , both the total energy @xmath107 and the total linear momentum @xmath108 are temporally invariant satisfying the remaining condition 2 ) , eq .
( [ eqn3.26 ] ) .
applying the remaining energy
momentum conservation condition , eq . ( [ eqn3.25 ] ) we find that the energy continuity equation @xmath109 that is obtained for @xmath110 is false because the two non - zero terms depend on different powers of @xmath77 in addition to being incommensurate with the poynting theorem .
this result is based on the total energy
momentum tensor and is therefore independent of the particular representation , abraham , minkowski , einstein
laub , _ etc . _ , that is used .
consequently , the macroscopic maxwell
minkowski equations and the conservation laws are fundamental laws of physics that are proven to be mutually contradictory in the case of a thermodynamically closed system consisting of quasimonochromatic light incident on , and propagating through , a simple linear dielectric that is draped with a gradient - index antireflection coating
. some readers may argue that pure induction without experimental support is not a method of theoretical physics .
however , the matrix energy
momentum evolution equations are derived by formal theory directly from the laws of continuum electrodynamics .
then these theorems were demonstrated to be contrary to global conservation laws .
this proof by mathematical contradiction is far stronger than an experimental demonstration .
one might recall that the 1887 michelson
morley experiment @xcite was initially interpreted to prove the existence of ether drift and was later deemed to support the absence of ether in the einstein relativity theory . likewise , the experiments that were originally viewed as support for the abraham energy momentum theory or the minkowski theory , or both , will be shown in sec .
vii to validate the new theory that is derived in sec .
the relationship of special relativity to the new treatment of continuum electrodynamics will be derived in sec .
vi . the century - long history of the abraham
minkowski controversy @xcite is a search for some provable description of momentum and momentum conservation for electromagnetic fields in dielectric media .
a wide variety of physical principles have been applied to establish the priority of one type of momentum over another , or to establish that the abraham and minkowski formulations are equally valid .
the modern resolution of the abraham minkowski momentum controversy is to adopt a scientific conformity in which the minkowski momentum and the abraham momentum are both correct forms of electromagnetic momentum with the understanding that neither is the total momentum @xcite . the total momentum , eq .
( [ eqn3.39 ] ) , is conserved in a thermodynamically closed system consisting of an antireflection coated block of a linear dielectric material situated in free - space that is illuminated by a quasimonochromatic field @xcite .
then , either the minkowski momentum or the abraham momentum can be used as the momentum of the electromagnetic field as long as that momentum is accompanied by the appropriate material momentum @xcite .
the material momentum is specific to a particular material and we will consider several well - known models that have appeared in the scientific literature in order to circumscribe the area of difficulty . in a quasi - microscopic approach , the material momentum is often modeled as the aggregated kinematic momentum of individual particles of matter . the total energy
momentum tensor is the sum of the electromagnetic energy
momentum tensor and the material energy momentum tensor .
in one particular example , pfeifer , nieminen , heckenberg , and rubinsztein - dunlop @xcite , posit that the total energy momentum tensor is the sum of the macroscopic abraham energy
momentum tensor , eq .
( [ eqn3.22 ] ) , and the dust energy
momentum tensor @xmath111 \ , . \label{eqn3.45}\ ] ] here , @xmath112 is a mass density and @xmath113 is a velocity field .
the dust tensor , eq .
( [ eqn3.45 ] ) , is usually applied to a thermodynamically closed system consisting of non - interacting neutral massive particles in an incoherent unimpeded flow such that @xmath114 in the current context , however , the total tensor energy
momentum continuity equation is @xcite @xmath115 clearly , it is intended that the dust tensor is coupled to the abraham electromagnetic tensor through the abraham force density such that @xcite @xmath116 performing the substitution of the dust tensor , eq .
( [ eqn3.45 ] ) , and the abraham energy
momentum tensor , eq .
( [ eqn3.22 ] ) , into eq .
( [ eqn3.47 ] ) results in @xcite @xmath117 + \nabla\cdot(\rho_0 { \bf v } ) + \nabla\cdot ( { \bf e}\times{\bf b})=0 \label{eqn3.49}\ ] ] for the @xmath110 component of the total tensor energy momentum continuity equation , eq .
( [ eqn3.47 ] ) .
pfeifer , nieminen , heckenberg , and rubinsztein - dunlop @xcite then use conservation of momentum arguments to phenomenologically relate the material momentum density to the electromagnetic momentum density with the _ ansatz _ @xmath118 substituting eq .
( [ eqn3.50 ] ) into eq .
( [ eqn3.49 ] ) produces @xmath117 + \nabla\cdot ( n{\bf e}\times{\bf b})=0 \ , .
\label{eqn3.51}\ ] ] the total energy and total momentum are both quadratic in the fields and must have the same dependence on the refractive index @xmath77 .
we note that if the particle density @xmath112 is constant then eq .
( [ eqn3.51 ] ) is false because the two nonzero terms would depend on different powers of the refractive index .
although pfeifer , nieminen , heckenberg , and rubinsztein - dunlop @xcite do not propose a time - dependent model for the particle density @xmath112 , the two non - zero terms of the energy continuity equation will be incommensurate unless eq .
( [ eqn3.51 ] ) becomes @xmath119 + \nabla\cdot ( n{\bf e}\times{\bf b})=0 \label{eqn3.52}\ ] ] requiring @xmath120 the corresponding tensor continuity equation for the total energy and the total momentum @xmath121 is false because the presence of the material four - divergence operator violates the conservation condition , eq .
( [ eqn3.25 ] ) for @xmath122 .
this result shows that temporal invariance of the total energy and the total momentum does not guarantee that the evolution equations for the total energy and the total momentum satisfy the conservation laws . in an influential 1973 article ,
gordon @xcite uses a microscopic model of the dielectric in terms of electric dipoles . assuming a dilute vapor in which the dipoles do not interact with each other or their host
, gordon writes the microscopic lorentz dipole force on a particle with linear polarizability @xmath123 as @xcite @xmath124 in the vacuum , where @xmath125 is the microscopic electric field and @xmath126 is the microscopic magnetic field .
the material momentum density is obtained by spatially averaging the force on a single dipole and integrating with respect to time .
then the material momentum density is @xmath127 where @xmath128 is the dipole density .
the fields acting on the dipoles inside a dielectric are not the same as the fields in free space . for the purpose of presenting the prior work ,
we accept , without proof , that the material momentum density is @xcite @xmath129 gordon assumes that the total momentum density is the sum of the abraham momentum density and the material momentum density .
making a transformation to retarded time @xcite , gordon @xcite derives @xmath130 for the total momentum density and then assumes a pseudo - momentum in order to force agreement with the minkowski form of momentum . in the gordon model , and similar derivative models ,
the dipoles are free particles that are accelerated by the lorentz dipole force at the leading edge of the field and travel at constant velocity until decelerated by the lorentz dipole force at the trailing edge of the force . in a real dielectric
, the motion of dipoles will be considerably impeded .
consequently , it is assumed that a traveling deformation of the material , rather than the unrestrained motion of dipoles , will contribute the requisite material momentum @xcite . the linear momentum @xmath131 that is obtained by spatially integrating the gordon momentum density and the minkowski momentum that is obtained by adding a hypothetical pseudomomentum concludes the derivation .
comparing eq .
( [ eqn3.59 ] ) to the temporal invariance displayed in eq .
( [ eqn3.31 ] ) , the gordon momentum @xmath132 is temporally invariant in the case of propagation of a quasimonochromatic field through a gradient - index antireflection - coated simple linear dielectric @xcite .
consequently , we now have a plausible model for the material momentum .
however , there are several problems with the derivation presented in ref .
@xcite in addition to the assumptions that are described above .
_ i _ ) in eq .
( [ eqn3.55 ] ) , the force density @xmath133 has been improperly retained because several other terms of the same order have been dropped in the dipole approximation @xcite .
moreover , this small term is divided into two large terms of nearly equal magnitude and opposite sign .
_ ii _ ) temporal invariance of the total linear momentum is only one of the four conditions of the energy
momentum conservation laws , eqs .
( [ eqn3.25])([eqn3.28 ] ) .
_ iii _ ) there is a factor of 2 error in the susceptibility used by gordon . in the corrected version of the gordon derivation , milonni and boyd @xcite prove that the sum of the electromagnetic and material momentums is the minkowski momentum , eq .
( [ eqn3.33 ] ) , not the gordon momentum , eq .
( [ eqn3.59 ] ) .
barnett and loudon @xcite and barnett @xcite present a model in which the abraham momentum and the minkowski momentum are both appropriate momentums for the field in a dielectric .
each of the classical electromagnetic momentums is accompanied by a material momentum , different in each case , and identified with either a canonical or kinetic phenomenology .
the material momentum densities , @xmath134 and @xmath135 , are defined implicitly by conservation of total momentum , such that @xmath136 although providing a descriptive model for construction of the total linear momentum , the total linear momentum , @xmath137 , is unique by temporal invariance and it is already a known quantity in terms of macroscopic fields , eq .
( [ eqn3.43 ] ) . as in the previous example , temporal invariance of the total linear momentum is only one of the four conditions of the energy momentum conservation laws , eqs .
( [ eqn3.25])([eqn3.28 ] ) .
if we use either of the canonical or kinetic models of eq .
( [ eqn3.60 ] ) for the total linear momentum , then the total energy momentum tensor will be the same as eq .
( [ eqn3.41 ] ) . applying the conservation condition , eq .
( [ eqn3.25 ] ) , the four - divergence of the total energy momentum tensor will produce an energy continuity equation that is the same as eq .
( [ eqn3.44 ] ) that was shown to be false because the two nonzero terms depend on different powers of the refractive index and is incommensurate with the poynting theorem .
ramos , rubilar , and obukhov @xcite utilize a fully relativistic 4-dimensional tensor formalism to discuss the energy
momentum of a system that consists of an antireflection - coated rigid slab of dielectric with a final constant velocity @xmath113 . their total energy
momentum tensor is @xmath138 @xmath139 @xmath140 where @xmath141 is the four - velocity @xmath142 .
ramos , rubilar , and obukhov @xcite claim that the total energy momentum tensor satisfies the energy momentum balance equation @xmath143
where @xmath144 is a four - current density , that the energy momentum tensor of the complete system is conserved , and that the system is thermodynamically closed if @xmath145 .
then the total four - momentum of the whole system is conserved and @xmath146 is a time - independent quantity @xcite . in order to test the validity of the total energy
momentum tensor , eq .
( [ eqn3.61 ] ) , we consider a quasimonochromatic field in the plane - wave limit to be normally incident on a simple linear dielectric through an antireflection coating .
the final velocity of the block depends on its mass , an independent quantity . in order to proceed unencumbered ,
we take the low - velocity limit .
evaluating the @xmath147 element of eq .
( [ eqn3.62 ] ) , we obtain @xmath148 by substitution from the total energy
momentum tensor , eq .
( [ eqn3.60 ] ) , using quantities from the field tensor @xmath149 . because @xmath62 is constant in time @xcite , eq .
( [ eqn3.64 ] ) is equivalent to the poynting theorem and we can see that the two non - zero terms are quadratic in the fields and depend on the same power of the refractive index @xmath77 .
we substitute quantities from the total energy
momentum tensor in the low - velocity limit and the field tensor @xmath149 into eq .
( [ eqn3.63 ] ) and obtain the abraham momentum . because the abraham momentum is a factor of @xmath77 smaller than the total momentum , eq .
( [ eqn3.63 ] ) is not temporally invariant , disproving the claim by ramos , rubilar , and obukhov @xcite that eq .
( [ eqn3.61 ] ) is the total energy
momentum tensor .
moreover , the total energy momentum tensor for a given thermodynamically closed system is unique and the elements of eq .
( [ eqn3.61 ] ) , in the appropriate limit , are not in agreement with the total energy momentum tensor of ref .
@xcite or with the total energy
momentum tensor , eq .
( [ eqn3.41 ] ) , that is derived by temporal invariance .
a fully microscopic model of the interaction of light with ponderable matter is unique , valid , and beyond our capabilities .
the examples above are representative of the many diverse quasi - microscopic treatments of the abraham
minkowski controversy as there is no unique quasi - microscopic model for the material properties of a simple linear dielectric @xcite .
there are many ways to couple and average the quasi - microscopic material properties with the electromagnetic properties that are systematically derived from the macroscopic maxwell minkowski equations . in the end ,
the total linear momentum is the sum of the electromagnetic momentum and the material momentum .
the correctness of the procedure is rooted in the fundamental basis of the model and the derivation .
however , the conservation laws that are given in subsection b of section iii by the relations ( [ eqn3.25])([eqn3.28 ] ) are not necessarily satisfied .
we provided specific examples , using the models of pfeifer , nieminen , heckenberg , and rubinsztein - dunlop @xcite , and of barnett @xcite where the continuity equation for the total energy is proven to be false . in another example
, we showed that the total momentum that is derived from the fully relativistic 4-dimensional tensor formalism of ramos , rubilar , and obukhov @xcite is not temporally invariant by a not insignificant factor of @xmath77 .
more importantly , these are general results .
any construction of the total energy
momentum tensor must be based on energy and momentum densities corresponding to the temporally invariant total energy , eq .
( [ eqn3.42 ] ) , the temporally invariant total momentum , eq .
( [ eqn3.43 ] ) , and diagonal symmetry .
then the four - divergence of the total energy
momentum tensor will always result in an energy continuity equation whose terms depend on different powers of the refractive index as shown in subsection c. the einstein theory of special relativity defines transformations between different inertial reference frames moving at constant velocities in vacuum @xcite .
the lorentz factor @xmath150 plays an important role in determining whether laws of physics satisfy the relativity principle .
einstein showed that the maxwell equations for electromagnetic fields in the vacuum exhibit lorentz invariance .
subsequently , von laue @xcite showed that the einstein velocity sum rule explained eighteenth - century experiments of light dragging by a moving dielectric thereby significantly contributing to the acceptance and development of einstein s theory of special relativity .
specifically , von laue s work cemented the identification of the vacuum lorentz factor , eq .
( [ eqn3.65 ] ) , with special relativity in a dielectric @xcite . due to the 1907 work by von laue @xcite ,
it is well - known that the vacuum lorentz factor , eq .
( [ eqn3.65 ] ) , is associated with relativistic transformations in a dielectric in addition to its role in the vacuum .
consequently , the vacuum lorentz factor , eq .
( [ eqn3.65 ] ) , has been associated with the symmetry of the macroscopic maxwell
minkowski equations , eqs .
( [ eqt2.04 ] ) .
there have been , from time to time , phenomenological arguments @xcite that the lorentz factor for a dielectric should depend on the refractive index .
this is confirmed in the current work because the symmetry of equations eqs .
( [ eqt2.06 ] ) , which are valid theorems of maxwellian continuum electrodynamics , selects the material lorentz factor @xmath151 this result is contradicted by the von laue @xcite description of the fizeau @xcite water tube experiment in terms of the relativistic velocity addition theorem @xcite . as a result , the material lorentz factor , eq .
( [ eqn3.66 ] ) , is ruled out experimentally as a violation of special relativity .
consequently , the equations of motion for the macroscopic fields given by eqs .
( [ eqt2.06 ] ) are proven false by special relativity . because the set of equations , eqs .
( [ eqt2.06 ] ) , is a valid theorem of axiomatic continuum electrodynamics , the axioms of the theory , the maxwell
minkowski equations , eqs .
( [ eqt2.01 ] ) , are also disproved for a simple linear dielectric .
the index - dependent lorentz factor can not be dismissed by an argument that it only applies to a specific unfeasible physical system of a linear isotropic homogeneous dielectric material that fills all space .
both sets of equations , eqs .
( [ eqt2.04 ] ) and ( [ eqt2.06 ] ) , are formally and mathematically equal under the same physical conditions and yet have incommensurate lorentz factors .
it is not a matter of which version of the lorentz factor is correct or whether one might better fit relativity under certain conditions .
the cause of this direct contradiction is that operations of the formal theory of maxwellian continuum electrodynamics alter the symmetry of the macroscopic maxwell minkowski equations .
this issue of relativity in a dielectric is complementary to the abraham - minkowski controversy in which operations of the formal theory of maxwellian continuum electrodynamics alter the energy
momentum tensor for electromagnetic fields propagating in a simple linear dielectric .
at the fundamental , microscopic level , dielectrics consist of tiny bits of host and polarizable matter , embedded in the vacuum , with interactions of various types . according to lorentz
, the seat of the electromagnetic field is the empty space .
if a light pulse is emitted from a point @xmath152 at time @xmath153 , then wavefronts are defined by @xmath154 in a flat four - dimensional minkowski spacetime @xmath155 .
equation ( [ eqf4.01 ] ) underlies classical electrodynamics and its relationship to special relativity .
although light always travels at speed @xmath156 , eq .
( [ eqf4.01 ] ) is only valid at short range , before the light is scattered by the various features of the dielectric .
while the microscopic picture is always valid , there are practical difficulties in treating all of the interactions as light traverses a dielectric . in continuum
electrodynamics , the dielectric is treated as continuous at all length scales and the macroscopic refractive index @xmath77 is defined such that light travels with an effective speed of @xmath0 .
in such a medium , wavefronts from a point source at @xmath152 and emitted at time @xmath153 , are defined by @xmath157 here , we postulate eq .
( [ eqf4.02 ] ) , instead of eq .
( [ eqf4.01 ] ) , as the basis of a theory of continuum electrodynamics and derive the consequences for classical dynamics , field theory , and electrodynamics .
we consider an arbitrarily large , but finite , region of space to be filled with a simple linear dielectric that is characterized by a linear refractive index @xmath77 . for clarity and concision
, we will work in a regime in which dispersion can be treated parametrically and is otherwise negligible such that @xmath77 is a real constant for a transparent dielectric .
we define an inertial reference frame @xmath158 with orthogonal axes , @xmath159 , @xmath160 , and @xmath161 and require that the origin of the reference frame is significantly inside the volume that is defined by the surface of the dielectric medium .
we denote a timelike coordinate in the medium as @xmath4 .
if a light pulse is emitted from the origin at time @xmath162 , then @xmath163 describes wavefronts in @xmath164 .
the four - vector @xmath3 represents the position of a point in a four - dimensional flat non - minkowski material spacetime in which @xmath165 .
the material spacetime reduces to ordinary minkowski spacetime if @xmath166 . for a system of particles ,
the transformation of the position vector @xmath167 of the @xmath168 particle to @xmath169 independent generalized coordinates is @xmath170 applying the chain rule , we obtain the virtual displacement @xmath171 and the velocity @xmath172 of the @xmath168 particle in the new coordinate system .
substitution of @xmath173 into the identity @xmath174 yields @xmath175 for a system of particles in equilibrium , the virtual work of the applied forces @xmath176 vanishes and the virtual work on each particle vanishes leading to the principle of virtual work @xmath177 and dalembert s principle @xmath178 using eqs .
( [ eqf4.05 ] ) and ( [ eqf4.09 ] ) and the kinetic energy of the @xmath168 particle @xmath179 we can write dalembert s principle , eq . ( [ eqf4.11 ] ) , as @xmath180 \delta q_j = 0 \label{eqf4.13}\ ] ] in terms of the generalized forces @xmath181 if the generalized forces come from a generalized scalar potential function @xmath182 @xcite , then we can write lagrange equations of motion @xmath183 where @xmath184 is the lagrangian .
the canonical momentum is therefore @xmath185 in a linear medium .
comparable derivations for the vacuum case , @xmath186 , appear in goldstein @xcite and marion @xcite , for example .
this version of canonical momentum differs from the existing vacuum formula because the dielectric time @xmath187 appears instead of the vacuum time @xmath188 .
the field theory @xcite is based on a generalization of the discrete case in which the dynamics are derived from a lagrangian density @xmath189 .
the generalization of the lagrange equation , eq .
( [ eqf4.15 ] ) , for fields in a linear medium is @xcite @xmath190 we take the lagrangian density of the electromagnetic field in the medium to be @xmath191 constructing the components of eq .
( [ eqf4.17 ] ) , we have @xmath192 @xmath193 @xmath194_j \label{eqf4.21}\ ] ] for the lagrangian density given in eq .
( [ eqf4.18 ] ) . substituting the individual terms , eqs .
( [ eqf4.19])([eqf4.21 ] ) , into eq .
( [ eqf4.17 ] ) , the lagrange equations of motion for the electromagnetic field in a dielectric are the three orthogonal components of the vector wave equation @xmath195 for fields , the canonical momentum density @xmath196 supplants the discrete canonical momentum defined in eq .
( [ eqf4.16 ] ) .
we can write the second - order equation , eq .
( [ eqf4.22 ] ) , as a set of first - order differential equations . to that end
, we introduce macroscopic field variables @xmath197 @xmath198 obviously , @xmath30 is the canonical momentum field density whose components were defined in eq .
( [ eqf4.23 ] ) after making the substitutions indicated by eq .
( [ eqf4.19 ] ) . substituting the definition of the canonical momentum field @xmath30 , eq .
( [ eqf4.24 ] ) , and the definition of the magnetic field @xmath29 , eq .
( [ eqf4.25 ] ) , into eq .
( [ eqf4.22 ] ) , we obtain a maxwell ampre - like law @xmath199 the divergence of eq .
( [ eqf4.25 ] ) and the curl of eq .
( [ eqf4.24 ] ) produce thompson s law @xmath200 and a faraday - like law @xmath201 respectively .
we posit the charge continuity law @xmath202 that corresponds to conservation of free charges with a free charge density @xmath203 in the continuum limit .
the divergence of the variant maxwell ampre law , eq .
( [ eqf4.26 ] ) , @xmath204 is combined with the charge continuity law , eq .
( [ eqf4.29 ] ) , to obtain @xmath205 integrating eq .
( [ eqf4.31 ] ) with respect to the temporal coordinate yields a modified version of gauss s law @xmath206 where @xmath207 is a constant of integration corresponding to a bound charge density .
this completes the set of first - order equations of motion for the macroscopic fields , eqs .
( [ eqf4.26])([eqf4.28 ] ) and ( [ eqf4.32 ] ) .
we have included free charges and a free - charge current as a historical imperative due to their common appearance in the macroscopic maxwell equations .
however , the notion of charges moving unimpeded in a continuous material is dubious because the displacement of the otherwise continuous dielectric by the charges is not treated in the theory .
consolidating the equations of motion and dropping the charges and currents , we have the equations of motion for macroscopic electromagnetic fields in a simple linear dielectric , @xmath208 @xmath209 @xmath210 @xmath211 [ eqf4.33 ] derived from first principles for quasimonochromatic electromagnetic fields in a dielectric - filled flat non - minkowski continuous material spacetime .
based on the derivation of these equations , it is required that the source terms in eqs .
( [ eqf4.33c ] ) and ( [ eqf4.33d ] ) that involve the gradient of the refractive index are , at most , perturbative , see sec .
v. the opposite limiting case , a piecewise homogeneous medium , will be treated in a separate publication .
next we want to relate the results of our derivation , eqs .
( [ eqf4.33 ] ) , to the ordinary macroscopic maxwell equations . according to maxwellian continuum electrodynamics , eqs .
( [ eqf4.33 ] ) are equal to the macroscopic maxwell equations , eqs .
( [ eqt2.04 ] ) , under a simple change of variable . in the current formalism , however , the usual macroscopic maxwell fields @xmath51 and @xmath212 are not definable , except as a trivial case .
if one does define the electric field @xmath213 in accordance with maxwellian continuum electrodynamics , then eqs . ( [ eqf4.16])([eqf4.24 ] ) require @xmath214 and the result is limited to the vacuum where @xmath166 . because the refractive index @xmath77 defines the time - like coordinate of a non - minkowski spacetime , the change of variable @xmath215 that was used in sec .
ii is an improper spacetime transformation .
there are special cases in which the new [ @xmath30,@xmath29 ] continuum electrodynamics can be phenomenologically related to the usual macroscopic maxwell equations .
these typically involve situations in which only one of the equations of macroscopic electrodynamics is needed , such as the wave equation @xmath217 that is derived by substituting the definitions of the macroscopic fields , eqs .
( [ eqf4.24 ] ) and ( [ eqf4.25 ] ) , into the maxwell ampre law , eq .
( [ eqf4.33a ] ) .
therefore , we can be assured that the extensive theoretical and experimental work that is correctly described by the macroscopic maxwell theory of continuum electrodynamics has an equivalent , or nearly equivalent , expression in the new theory .
nevertheless , we must be very careful about integrating established concepts of electromagnetism into the new version of continuum electrodynamics , as demonstrated in the previous paragraph .
more interesting is the work that we can do with the new formalism of continuum electrodynamics that was unreachable or incorrect , or improperly posed , in the standard maxwell theory of continuum electrodynamics .
these cases will typically involve the invariance or tensor properties of the set of coupled equations of motion .
this interpretation is borne out in our common experience : the macroscopic maxwell
minkowski equations produce exceedingly accurate experimentally verified predictions of simple phenomena like reflection , refraction , fresnel relations , wave propagation , _ etc .
_ , but fail to render a unique , uncontroversial , experimentally verifiable prediction of energy momentum conservation . in the next sections , we will demonstrate the utility of the new formalism of continuum electrodynamics by addressing energy momentum conservation and special relativity in a dielectric and then a reinterpretation of the balazs thought experiment and the jones
richards experiment .
the derivation of the continuity equation of a property flux density is described following eq .
( [ eqn3.24 ] ) as applying the divergence theorem to a taylor series expansion of the property density field @xmath62 and the property flux density field @xmath64 to unimpeded flow in an otherwise empty volume @xcite . here
, we are modeling a dielectric as a dielectric - filled spacetime .
as shown in eq .
( [ eqf4.04 ] ) , the generalized temporal coordinate in this spacetime is @xmath187 .
then a continuity equation has the form @xmath219 in a region of space in which light travels at @xmath0 .
we can compare the conservation laws in a dielectric - filled spacetime with the conservation laws 1)5 ) , eqs .
( [ eqn3.25])([eqn3.29 ] ) , in an empty volume .
\1 ) continuity equations in a dielectric have the form of eq .
( [ eqf5.01 ] ) , instead of eq .
( [ eqn3.24 ] ) . writing a scalar continuity equations for energy and a scalar continuity equation for each of the three components of the momentum , row - wise , we obtain @xmath220 instead of eq .
( [ eqn3.25 ] ) as a condition for conservation of energy and momentum properties in a continuous unimpeded flow in a dielectric - filled spacetime , as opposed to a vacuum spacetime .
the surface of a dielectric is a boundary between two regions of different photon density , consequently the usual assumption of incompressible flow that underlies eq .
( [ eqn3.25 ] ) is broken and there is a different continuity equation , eq .
( [ eqf5.02 ] ) , for each value of @xmath77 .
\2 ) the total energy and the total linear momentum are invariant in time in a dielectric , as well as in vacuum , such that @xmath221 \3 ) the total energy
momentum tensor is symmetric @xmath222 otherwise an unbalanced shear force would cause a violation of conservation of total angular momentum
. see the discussion following eq .
( [ eqn3.27 ] ) .
\4 ) the trace of the energy
momentum tensor is proportional to the mass density @xmath73 : @xmath223 \5 ) the extra continuity equation @xmath224 is obtained by substituting the symmetry condition 3 ) , eq .
( [ eqf5.04 ] ) , into the continuity condition 1 ) , eq .
( [ eqf5.02 ] ) .
several of the conservation conditions are unchanged for dynamics that take place in a dielectric - filled spacetime .
the continuity equations , eqs .
( [ eqf5.02 ] ) and ( [ eqf5.06 ] ) , now depend on an index - dependent material four - divergence operator , eq .
( [ eqn3.06 ] ) , instead of the vacuum four - divergence operator , due to the index - dependent velocity of light in the medium .
we can demonstrate consistency of the new formulation of continuum electrodynamics with the conservation laws inside a dielectric - filled spacetime .
the equations of motion for the macroscopic fields , eqs .
( [ eqf4.33a ] ) and ( [ eqf4.33c ] ) , can be combined in the usual manner , using algebra and calculus , to write an energy continuity equation @xmath225 + \nabla\cdot \left ( { \bf b}\times{\bf \pi}\right ) = \frac{\nabla n}{n } \cdot({\bf b}\times{\bf \pi } ) \label{eqf5.07}\ ] ] in terms of an electromagnetic energy density @xmath226 an electromagnetic momentum density @xmath227 and a perturbative transient power density @xmath228 likewise , we can combine eqs .
( [ eqf4.33 ] ) to derive the momentum continuity equation @xmath229 where the stress - tensor @xmath230 is @xmath231 and @xmath232 is a force density .
then the continuity equations , eqs .
( [ eqf5.07 ] ) and ( [ eqf5.11 ] ) , can be written as @xmath233 where @xmath234 \label{eqf5.15}\ ] ] is the electromagnetic energy
momentum tensor and @xmath235 is the four - force density .
now , we require for the variation of the refractive index to be sufficiently small that the force density and power density can be neglected , such that @xmath236 because the right - hand side of eq .
( [ eqf5.17 ] ) is nil , there is no mechanism in the theory to couple to any sub - system by a source or sink of energy or momentum .
therefore , the electromagnetic system is thermodynamically closed . integrating the total electromagnetic momentum density of eq .
( [ eqf5.09 ] ) over all - space @xmath67 , we have the temporally invariant total momentum @xmath237 likewise , the temporally invariant total energy @xmath238 is obtained by integrating the energy density eq .
( [ eqf5.08 ] ) over all - space @xmath239 .
all of the quantities that constitute the total energy density , total momentum density , and total energy momentum tensor are electromagnetic quantities with the caveat that the gradient of the refractive index is small .
although rigorous results are restricted to a limiting case , the the real - world necessity of a non - zero gradient adds only a small perturbative effect .
the opposite limit of a piecewise homogeneous medium without an antireflection coating will be considered in a separate article .
there are only two macroscopic fields in the theory , @xmath30 and @xmath29 , and we disavow the use of macroscopic fields @xmath51 and @xmath212 and the permittivity @xmath15 .
obviously , we would like to informally compare and contrast the new theory with maxwellian continuum electrodynamics .
there is sufficient commonality with the classic theories that simple phenomena like reflection , refraction , and wave velocity have equivalent formulations in the new theory .
informal relations aside , the macroscopic maxwell equations , eqs .
( [ eqt2.04 ] ) or ( [ eqt2.02 ] ) , can not be derived as valid theorems from our new axioms , eqs .
( [ eqf4.33 ] ) , and we can not insert expressions that are derived from maxwellian continuum electrodynamics into the new theory .
in a 1952 article , rosen @xcite argues that there should be a number of theories of relativity , each associated with an isotropic homogeneous medium in which a limiting speed is associated with the phenomena that take place in the medium . in rosen s special relativities @xcite , and earlier work by michels and patterson @xcite ,
the vacuum speed of light that appears in the lorentz factor is phenomenologically replaced by the speed of light @xmath0 in the dielectric .
rosen uses a model of an arbitrarily large dielectric medium in which the observers have no contact with the vacuum and proposes the index - dependent material lorentz factor @xmath240 , @xmath241 where @xmath77 is the macroscopic refractive index and @xmath242 is the relative speed of the two coordinate systems in the dielectric .
more recent work arrives at the material lorentz factor @xmath240 by a consideration of the transformation symmetry of the macroscopic maxwell equations @xcite .
the symmetry of eqs .
( [ eqf4.33 ] ) shows that their invariance will depend on the material lorentz factor , eq .
( [ eqf6.01 ] ) , not the einstein , vacuum , version of the lorentz factor eq .
( [ eqn3.65 ] ) . in this section ,
we explore coordinate transformations between inertial systems in an arbitrarily large simple dielectric medium and derive the previously phenomenological material lorentz factor , eq .
( [ eqf6.01 ] ) .
we find that the material lorentz factor , eq .
( [ eqf6.01 ] ) , is appropriate for events that occur in the dielectric if the observer is likewise positioned inside the dielectric . using boundary conditions ,
we show how an outside observer , in the vacuum of a laboratory frame of reference , would view events that happen in the dielectric .
the case of the vacuum - based observer viewing events in a dielectric is re - analyzed using the vacuum lorentz factor , eq .
( [ eqn3.65 ] ) .
the points to be made here are _ i _ ) that the material lorentz invariance of dielectric special relativity confirms our derivation of continuum electrodynamics based on a flat non - minkowski material spacetime and _ ii _ ) that boundary conditions are necessary to refer physical processes that occur inside a dielectric to a laboratory frame of reference in which physical measurements can be made in the vacuum .
we consider two inertial reference frames , @xmath158 and @xmath243 , in a standard configuration @xcite in which @xmath159 and @xmath244 are collinear , @xmath160 is parallel to @xmath245 , @xmath161 stays parallel to @xmath246 , and @xmath247 translates at a constant speed in the direction of the positive @xmath159-axis .
the origins of the two systems coincide at some initial time . at each point in each coordinate system , time is measured by an idealized clock and all the clocks in each coordinate system have been synchronized by one of the usual methods .
as we are studying coordinate transformations in a simple linear dielectric , both coordinate axes are embedded in an arbitrarily large dielectric - filled region of space . at time @xmath248 , a directional light pulse is emitted from the common origin , labeled @xmath249 , along the @xmath160- and @xmath245-axes . in the rest frame of the dielectric , @xmath164 , the pulse is reflected by a mirror in the dielectric at point @xmath250 and returns to the origin at time @xmath251 as shown in fig .
then the distance from the origin to the mirror is @xmath252 , where @xmath253 is the speed of light in the rest frame @xmath164 of the dielectric . in the dielectric . ]
the trajectory of the light pulse in the @xmath247 frame of reference is shown in fig .
the translation of the @xmath247 frame is transverse to the @xmath160-axis so the distance from the mirror at @xmath254 to the @xmath244-axis is @xmath255 , the same as the distance from the mirror at @xmath250 to the @xmath159-axis .
viewed in the @xmath247 frame , the light pulse is emitted from the point @xmath249 at time @xmath256 , is reflected from the mirror at point @xmath254 , and is detected at the point @xmath257 at time @xmath258 . during that time
, the point of emission / detection has moved a distance @xmath259 . by symmetry ,
the light is reflected from the mirror at a time @xmath260 making the distance the light travels from the origin to the mirror @xmath261 , where @xmath262 is the speed of light in the direction @xmath263 in the @xmath247 frame of reference . by the pythagorean theorem , we have @xmath264 we write the previous equation as @xmath265 and define the lorentz factor @xmath240 by @xmath266 such that @xmath267 at this point , there are more unknowns than equations and we can proceed no further without some additional condition .
when einstein faced the equivalent problem for free space , he postulated that light travels at a uniform speed @xmath156 in the vacuum , regardless of the motion of the source . here , the isotropy of an arbitrarily large region of space in which the speed of light is @xmath253 leads us to postulate that light travels at a uniform speed @xmath253 in a homogeneous dielectric .
the fizeau experiment @xcite and its interpretation in terms of fresnel drag tells us that the refractive index in a dielectric depends on the velocity of the dielectric .
if that were to be the case , then we could determine the absolute motion of the block in relation to a preferred reference frame in violation of relativity .
in the next subsection , we show that fresnel drag is a consequence of making measurements in a `` laboratory '' frame of reference in the vacuum outside the dielectric .
basically , observers in different lab frames of reference would measure different speeds for the same ray of light in a dielectric . here , for an observer in an arbitrarily large dielectric
, we can substitute @xmath268 into eq .
( [ eqf6.05 ] ) to obtain @xmath269 we could argue that the lorentz factor is always the vacuum lorentz factor , eq .
( [ eqn3.65 ] ) , because the dielectric can always be modeled as particles and interactions in the vacuum where special relativity is valid . however , our model is not the microscopic model of particles and interactions in the vacuum and we must deal with the macroscopic model that is before us . then , in the limit of continuum electrodynamics , the macroscopic lorentz factor for coordinate transformation in an arbitrarily large simple linear dielectric is given by eq .
( [ eqf6.06 ] ) .
now , the speed of light @xmath253 will be different in different dielectrics .
we are considering only materials in which the speed of light is inversely proportional to some constant @xmath77 and we obtain @xmath270 as our material lorentz factor .
the material lorentz factor , eq .
( [ eqf6.07 ] ) , resolves the apparent contradiction between the equations of motion of the macroscopic fields , eqs .
( [ eqf4.33 ] ) , and special relativity . in the dielectric . ] at this point , we know how to perform two types of transformations : transformations between coordinate systems in the vacuum and transformations between coordinate systems in a dielectric .
now we want to investigate how an outside observer , in the vacuum , would treat events that happen in the dielectric . to this end
, we use the same set of axes @xmath164 and @xmath247 to simultaneously perform both types of transformations by requiring that the @xmath159 , @xmath244 , @xmath161 , and @xmath246 axes lie on the surface of a semi - infinite dielectric , figs . 5 and 6 .
then the upper half - space , @xmath271 and @xmath272 , is modeled with a real macroscopic index of refraction @xmath77 in the rest - frame @xmath164 where the speed of light is @xmath0 . the lower half - space , @xmath273 and @xmath274 , is vacuum in which the speed of light is @xmath156 . at time @xmath275 , a bi - directional light pulse is emitted from the common origin @xmath249 along the @xmath276- and @xmath277-axes . the pulse is reflected by a mirror in the vacuum at @xmath278 and returns to the origin at time @xmath279 .
the pulse is also reflected by a mirror in the dielectric at @xmath280 and returns to the origin at time @xmath281 .
the locations of the mirrors are adjusted so that both reflections return to the origin at the same time , such that @xmath282 by construction . at the dielectric / vacuum boundary . ]
the trajectory of the light pulse in the @xmath247 frame of reference is shown in fig .
the translation of the @xmath247 frame is transverse to the @xmath160-axis so the distance from the mirror at @xmath283 to the @xmath244-axis is @xmath284 , the same as the distance from the mirror at @xmath285 to the @xmath159-axis .
viewed from the @xmath247 frame , the light pulse is emitted from the point @xmath249 at time @xmath286 , is reflected from the mirror at point @xmath283 , and is detected at the point @xmath287 at time @xmath288 . during that time , the point of emission / detection has moved a distance @xmath289 . at the dielectric / vacuum boundary . ]
events that occur at both the same time and the same place in one inertial reference frame occur simultaneously in all inertial reference frames .
the pulse that travels through the vacuum is reflected back to the origin and arrives at the same time that the pulse makes a round trip through the dielectric , eq .
( [ eqf6.08 ] ) .
the principle of simultaneity gives us the condition @xmath290 one can then square both sides of eq .
( [ eqf6.09 ] ) .
substituting @xmath291 from the vacuum theory and eqs .
( [ eqf6.03 ] ) and ( [ eqf6.08 ] ) into the square of eq .
( [ eqf6.09 ] ) yields @xmath292 grouping terms containing @xmath293 , the previous equation becomes @xmath294 then , from eq .
( [ eqf6.09 ] ) , we obtain @xmath295 the speed of light in the dielectric is @xmath296 as observed from a point in the vacuum , outside the dielectric .
now , let us return to eq .
( [ eqf6.04 ] ) and make a different assumption by adopting the vacuum lorentz factor , regardless of the refractive index . equating eqs .
( [ eqn3.65 ] ) and ( [ eqf6.05 ] ) , we obtain @xmath297 comparing eqs .
( [ eqf6.14 ] ) and ( [ eqf6.15 ] ) , we see that the usual vacuum lorentz factor , eq .
( [ eqn3.65 ] ) , can be used for events that occur in a dielectric if the observer is in the vacuum .
however , the material lorentz factor , eq .
( [ eqf6.06 ] ) , is appropriate if observers are within the dielectric .
the isotropy and homogeneity of the material spacetime that was assumed in subsection a is obviously violated by the presence of the boundary with the vacuum .
likewise , the isotropy and homogeneity of free space that is assumed in the einstein theory of special relativity is violated by the presence of the dielectric . by treating two abutting semi - infinite slabs of dielectric and vacuum spacetime
, one can set up the equations of special relativity in each half - space .
light is generated on the boundary but does not cross the boundary , consequently , the two half - spaces are isotropic and homogeneous in the separate regions that are connected by boundary conditions .
we can add additional boundary conditions to derive the theory of dielectric special relativity for a finite block , instead of a semi - infinite slab , of dielectric in free space .
consequently , the theory is valid for all spacetime .
we can never place a matter - based observer , no matter how small , in a continuous dielectric because the model dielectric is continuous at all length scales and will always be displaced .
consequently , the necessity to make non - optical measurements in a vacuum leads to the establishment of a laboratory frame of reference .
for an observer in the vacuum of a lab frame of reference , the symmetry is that of vacuum lorentz transformations corresponding to the vacuum lorentz factor .
an observer in the vacuum , such as fizeau @xcite , will measure the speed of light in a dielectric to be dependent on the velocity of the dielectric relative to the laboratory frame of reference , an effect that fresnel attributed to ether drag . for an arbitrarily large dielectric ,
the observer , in the form of an immaterial inertial coordinate system , is in the medium and measures the speed of light in a dielectric to be @xmath0 .
it is not surprising that coordinate transformations are different in the different physical configurations .
in 1953 , balazs @xcite proposed a thought experiment to resolve the abraham minkowski controversy .
the thought experiment was based on the law of conservation of momentum and a theorem that the center of mass , including the rest mass that is associated with the energy , moves at a uniform velocity .
the total energy @xmath298 becomes the einstein formula @xmath299 for massive particles in the limit @xmath300 .
for massless particles , like photons , eq .
( [ eqt7.01 ] ) becomes @xmath301 where @xmath302 is a unit vector in the direction of motion .
equation ( [ eqt7.02 ] ) defines the instantaneous momentum of `` a '' photon between scattering events in a microscopic model of a dielectric .
the description of the momentum of a field in terms of the momentums of constituent photons is difficult because the effective momentum of a photon in the direction of propagation is different from its instantaneous momentum due to scattering .
there are other complications , including an indefinite photon number , that cause us to choose to choose a macroscopic classical description for light propagation in a dielectric . as an electromagnetic field propagates from vacuum into a simple linear dielectric , the effective velocities of photons in the field are reduced due to scattering .
there is a corresponding increase in photon density in the dielectric .
likewise , the classical energy density @xmath303 and the classical momentum density @xmath304 are enhanced by a factor of @xmath77 in the dielectric , compared to the vacuum . for finite pulses in a dielectric , the enhanced energy density is offset by a narrowing of the pulse so that the electromagnetic energy @xmath305 is invariant for quasimonochromatic fields in the plane wave limit .
the electromagnetic energy is the total energy by virtue of being invariant in time .
likewise , the electromagnetic momentum @xmath306 is invariant and is the total momentum .
the center - of - energy velocity of the field slows to @xmath307 . invoking the einstein mass energy equivalence ,
it is argued in the scientific literature @xcite that some microscopic constituents of the dielectric must be accelerated and then decelerated by the field ; otherwise the theorem that the center of mass energy moves at a constant velocity is violated . for a distribution of particles of mass @xmath308 and velocity @xmath309 ,
the total momentum @xmath310 is the sum of the momentums of all the particles @xmath311 in the distribution .
if the mass of each particle @xmath308 is constant , the statement that the velocity of the center of mass @xmath312 is constant is a statement of conservation of total momentum .
we know from section iii that the momentum , eq .
( [ eqt7.04 ] ) , is temporally invariant even though the center - of - energy travels at speed @xmath0 in the dielectric .
because of the enhanced momentum density of the field in a dielectric , the differential electromagnetic momentum @xmath313 that is contained in an element of volume @xmath314 is a factor of @xmath77 greater than in the vacuum . for a finite pulse , the narrower pulse width and enhanced momentum density offset allowing the electromagnetic momentum to be invariant in time as the field enters , and exits , the dielectric through the gradient - index antireflection coating .
consequently , there is no need to hypothesize any motion of the material constituents of the dielectric to preserve conservation of momentum .
one of the enduring questions of the abraham
minkowski controversy is why the minkowski momentum is so often measured experimentally while the abraham form of momentum seems to be so favored in theoretical work .
we now have the tools to answer that question .
the minkowski momentum is not measured directly , but inferred from a measured index dependence of the optical force on a mirror placed in a dielectric fluid @xcite . because the field is completely reflected at the mirror , the force on the mirror is @xmath315 the measured force on the mirror is directly proportional to the refractive index @xmath77 of the fluid @xcite in agreement with eq .
( [ eqt7.08 ] ) . on the other hand ,
if we were to assume @xmath316 , which is the relation between momentum and force in an otherwise empty spacetime , then we can write eq .
( [ eqt7.08 ] ) as @xmath317 then one might infer from eq .
( [ eqt7.09 ] ) that the momentum of the field in the dielectric fluid is the minkowski momentum .
instead , we see that the electromagnetic momentum that is obtained from an experiment that measures the optical force on a mirror depends on the theory that is used to interpret the results .
however , based on the changes to continuum electrodynamics that are necessitated by conservation of energy and momentum by the propagation of light in a continuous medium , we find that eq .
( [ eqt7.08 ] ) is the correct relation between the force on the mirror and the momentum of the field in a dielectric .
in this article , we treated maxwellian continuum electrodynamics as an axiomatic formal theory and showed that valid theorems of the formal theory are contradicted by conservation laws and showed that valid theorems of the formal theory are contradicted by special relativity .
axiomatic formal theory is a cornerstone of abstract mathematics and the contradiction of valid theorems of maxwellian continuum electrodynamics by other fundamental laws of physics proves , unambiguously , that continuum electrodynamics , special relativity , and energy
momentum conservation laws , as currently applied to dielectrics , are mutually inconsistent .
we then established a rigorous basis for a reformulation of theoretical continuum electrodynamics by deriving equations of motion for the macroscopic fields from lagrangian field theory adapted for a dielectric - filled spacetime .
we reformulated the conservation laws , which were originally derived for the case of an unimpeded inviscid incompressible flow of non - interacting particles ( dust , fluid , etc . ) in the continuum limit in an otherwise empty volume , for the continuous flow of photons in a dielectric - filled volume .
we used coordinate transformations between inertial reference frames in a dielectric - filled volume to derive a theory of dielectric special relativity .
the reformulated versions of continuum electrodynamics , special relativity , and energy
momentum conservation laws are mutually consistent in a dielectric - filled volume .
we used the new theory to provide a lucid re - interpretation of the balazs thought experiment and the jones
richards experiment for the electromagnetic momentum in a dielectric medium .
the abraham
minkowski controversy is trivially resolved because the tensor total energy momentum continuity theorem , the total energy momentum tensor , the total momentum , and the total energy are fully electromagnetic and unique for a thermodynamically closed system consisting of a simple linear dielectric block draped with a gradient - index antireflection coating illuminated by quasimonochromatic light .
j. c. maxwell , a dynamical theory of the electromagnetic field , phil .
, 155 , 459512 ( 1865 ) .
r. hertz , ueber die einwirkung einer geradinigen electrischen schwingung auf eine benachbarte strombahn , ann .
( berlin ) , 270 , 155 - 170 ( 1888 ) .
h. minkowski , die grundgleichungen fr die elektromagnetischen vorgnge in bewegten krpern , natches .
gttingen , 53 - 111 ( 1908 ) .
m. abraham , zur elektrodynamik bewegter krper , rend .
circ . mat .
palermo , 28 , 1 - 28 ( 1909 ) .
d. j. griffiths , resource letter em-1 : electromagnetic momentum , am .
, 80 , 718 ( 2012 ) .
k. t. mcdonald , bibliography on the abraham
minkowski debate , http://physics.princeton.edu/@xmath318mc- donald/ examples / ambib.pdf ( 2015 ) .
i. brevik , experiments in phenomenological electrodynamics and the electromagnetic energy
momentum tensor , phys .
, 52 , 133 - 201 ( 1979 ) .
p. w. milonni and r. w. boyd , momentum of light in a dielectric medium , adv .
photonics , 2 , 519553 ( 2010 ) .
b. a. kemp , macroscopic theory of optical momentum , progress in optics , 20 , 437488 ( 2015 ) .
r. n. c. pfeifer , t. a. nieminen , n. r. heckenberg , and h. rubinsztein - dunlop , colloquium : momentum of an electromagnetic wave in dielectric media , rev .
, 79 , 11971216 ( 2007 ) . s. m. barnett and r. loudon , the enigma of optical momentum in a medium , phil . trans .
a , 368 , 927939 ( 2010 ) . b. a. kemp , resolution of the abraham
minkowski debate : implications for the electromagnetic wave theory of light in matter , j. appl .
phys . , 109 , 111101 ( 2011 ) .
y. n. obukhov , electromagnetic energy and momentum in moving media , ann . phys .
( berlin ) , 17 , 830851 ( 2008 ) .
t. ramos , g. f. rubilar and y. n. obukhov , first principles approach to the abraham
minkowski controversy for the momentum of light in general linear non - dispersive media , j. opt .
, 17 , 025611 ( 2015 ) .
m. mansuripur , radiation pressure and the linear momentum of the electromagnetic field , opt .
, 12 , 53755401 ( 2004 ) . s. m. barnett , resolution of the abraham minkowski dilemma , phys .
lett . , 104 , 070401 ( 2010 ) .
r. peierls , proc .
a , the momentum of light in a refracting medium , 347 , 475491 ( 1976 ) .
m. e. crenshaw , electromagnetic momentum and the energy
momentum tensor in a linear medium with magnetic and dielectric properties , j. math .
phys . , 55 , 042901 ( 2014 ) .
n. l. balazs , the energy
momentum tensor of the electromagnetic field inside matter , phys .
, 91 , 408411 ( 1953 ) .
j. p. gordon , radiation forces and momenta in dielectric media , phys .
rev . a , 8 , 1421 ( 1973 ) .
m. e. crenshaw and t. b. bahder , energy
momentum tensor for the electromagnetic field in a dielectric , opt .
, 284 , 24602465 ( 2011 ) . c. mller , _ the theory of relativity _ , oxford university press , london ( 1972 ) .
i. brevik , what do experiments in optics tell us about photon momentum in media ? , http://arxiv.org/pdf/1310.3684v1.pdf ( 2013 )
. w. she , j. yu , and r. feng , observation of a push force on the end face of a nanometer silica filament exerted by outgoing light , phys .
lett . 101 , 243601 ( 2008 ) .
l. zhang , w. she , and u. leonhardt , experimental evidence for abraham pressure of light , new j. phys .
17 , 053035 ( 2015 ) .
m. mansuripur , comment on `` observation of a push force on the end face of a nanometer silica filament exerted by outgoing light , '' phys .
, 103 , 019301 ( 2009 ) . i. brevik , comment on `` observation of a push force on the end face of a nanometer silica filament exerted by outgoing light , '' phys .
, 103 , 219301 ( 2009 ) . w. she , j. yu , and r. feng , she _
_ reply : , phys .
, 103 , 219302 ( 2009 ) .
m. bethune - waddell and k. j. chau , `` simulations of radiation pressure experiments narrow down the energy and momentum of light in matter , '' rep .
, 78 , 122401 ( 2015 ) .
n. rosen , special theories of relativity , am .
j. phys , 20 , 161164 ( 1952 ) .
w. c. michels and a. l. patterson , special relativity in refracting media , phys .
, 60 , 589592 ( 1941 ) . f. ravndal , effective electromagnetic theory for dielectric media , http://arxiv.org/pdf/0804.4013v3.pdf ( 2008 ) . v. a. redko and , g. j. spix , on the different forms of the maxwell s electromagnetic equations in a uniform media , http://arxiv.org/pdf/hep-th/0604080v1.pdf ( 2006 ) . g. weinstein , albert einstein and the fizeau 1851 water tube experiment , http://arxiv.org/pdf/1204.3390.pdf ( 2012 ) .
m. laue , die mitfhrung de lichtes durch bewegte krper nach dem relativittsprinzip , ann .
( berlin ) 23 , 989990 ( 1907 ) .
r. w. fox and a. t. mcdonald , _ introduction to fluid dynamics _ , wiley , new york ( 1978 ) .
m. e. crenshaw , quantum optics of dielectrics in momentum - space , opt .
commun . , 235 , 153161 ( 2004 ) .
j. d. jackson , _ classical electrodynamics , _
john wiley , new york ( 1975 ) .
p. kinsler , a. favaro , and m. w. mccall , four poynting theorems , eur .
j. phys . , 30 , 983 ( 2009 ) .
p. penfield , jr . and
h. a. haus , _ electrodynamics of moving media _ , mit press , cambridge ( 1967 ) .
i. brevik , minkowski momentum resulting from a vacuum - medium mapping procedure , and a brief review of minkowski momentum experiments , http://arxiv.org/pdf/1610.03882.pdf ( 2012 ) .
n. g. c. astrath , l. c. malacarne , m. l. baesso , g. v. b. lukasevicz , and s. e. bialkowski , unravelling the effects of radiation forces in water , nature communications , 5 , 4363 ( 2014 ) . v. p. torchigin and a. v. torchigin , interrelation between various types of optically induced forces , opt .
commun . , 301 - 302 , 147151 ( 2013 ) . m. mansuripur , the force law of classical electrodynamics : lorentz versus einstein and laue , optical trapping and optical micromanipulation x , ed . k. dholakia and g. c. spalding , proc . of spie vol .
8810 , 88100k-1:18 ( 2013 ) . l. d. landau and e. m. lifshitz , _ the classical theory of fields _
, 4th ed . ,
elsevier , amsterdam ( 2006 ) .
d. j. griffiths , _ introduction to electrodynamics , _ prentice - hall , englewood cliffs , new jersey ( 1981 ) .
a. a. michelson and e. w. morley , on the relative motion of the earth and the luminiferous ether , am .
, 34 , 333 - 345 ( 1887 ) .
a. icsevgi and w. e. lamb , jr . ,
propagation of light pulses in a laser amplifier , phys .
, 185 , 517545 ( 1969 ) .
m. e. crenshaw , decomposition of the total momentum in a linear dielectric into field and matter components , ann .
( n y ) , 338 , 97106 ( 2013 ) .
w. rindler , _ introduction to special relativity _
, oxford university press , new york ( 1982 ) .
p. m. schwarz and j. j. , schwarz , _ special relativity from einstein to strings _ , cambridge university press , cambridge ( 2004 ) .
r. shankland , conversations with albert einstein , am .
j. phys . , 31 , 4757 ( 1963 ) . m. h. fizeau , on the hypotheses relating to the luminous aether , and an experiment which appears to demonstrate that the motion of bodies alters the velocity with which light propagates itself in their interior , phil .
, 2 , 568573 ( 1851 )
. h. goldstein , _ classical mechanics _
, addison - wesley , reading ( 1980 ) . j. b. marion , _ classical dynamics of particles and systems _ , academic , new york ( 1970 ) . c. cohen - tannoudji , j. dupont - roc and t. grynberg , _ photons and atoms _ , wiley , new york ( 1989 ) . m. hillery and l. mlodinow , quantization of electrodynamics in nonlinear dielectric media , phys .
a , 55 , 678689 ( 1984 ) .
r. v. jones and j. c. s. richards , the pressure of radiation on a refracting medium , proc .
london , ser .
a , 221 , 480498 ( 1954 ) . | we treat continuum electrodynamics as an axiomatic formal theory based on the macroscopic maxwell
minkowski equations applied to a thermodynamically closed system consisting of an antireflection - coated block of a simple linear dielectric material situated in free - space that is illuminated by a quasimonochromatic field .
we prove that valid theorems of the formal theory of maxwellian continuum electrodynamics are inconsistent with energy
momentum conservation laws for the unimpeded , inviscid , incompressible flow of non - interacting particles , photons , in the continuum limit .
we also show that valid theorems of maxwellian continuum electrodynamics are contradicted by the index - independent lorentz factor of von laue s application of the einstein theory of special relativity to a dielectric .
therefore , the extant theoretical treatments of macroscopic electrodynamics , special relativity , and energy momentum conservation must be regarded as being mutually inconsistent in a region of space in which the effective speed of light is @xmath0 .
having proven the existing theories to be inconsistent , it is customary to propose alternatives .
we derive , from first principles , a self - consistent alternative theoretical treatment of electrodynamics , special relativity , and energy momentum conservation in an isotropic , homogeneous , linear dielectric - filled , flat , non - minkowski , continuous material spacetime .
there is sufficient commonality with the classic theories that the extensive theoretical and experimental work that is correctly described by maxwellian continuum electrodynamics , conservation , and von laue
einstein dielectric special relativity has an equivalent formulation in the new theory .
the more complex issues of the abraham minkowski momentum controversy and rosen s refractive index - dependent dielectric special relativity theory have robust resolutions in the new theory .
0.5 cm keywords : abraham minkowski controversy , energy momentum tensor , macroscopic maxwell equations , material lorentz factor , non - minkowski spacetime , rosen s special relativities 0.5 cm conflict of interest
: the author declares that there is no conflict of interest . |
giant pulses ( gps ) have been reported in four pulsars ( b0531 + 21 , b1937 + 21 , b1821@xmath124 and b0540@xmath169 ) todate ( staelin & reifenstein 1968 ; lundgren et al . 1995 ; cognard et al .
1996 ; romani & johnston 2001 ; johnston & romani 2003 ) .
three of these pulsars are millisecond pulsars ( msps ) and also show strongly pulsed hard x - ray profiles ( takahashi et al .
the radio gps occur in a narrow phase window close to the high energy non - thermal pulse indicating a common magnetospheric origin .
all these pulsars have a high value of @xmath0 .
we have used gmrt to search for gps in candidate msps with a non - thermal high energy emission and a range of @xmath0 and report detection of such pulses in two more pulsars , psr j0218 + 4232 and b1957 + 20 .
we obtained about 3600 s of data on each pulsar using 20 to 22 gmrt antennae in an incoherent mode at 610 mhz with 16 mhz bandwidth .
the expected rms noise in the above configuration of gmrt for a sampling time of 258 @xmath2 is about 1 jy . the data in two subbands ( 8 mhz each ) were dedispersed to a common sky frequency ( 610 mhz ) .
the periods with a peak greater than 3.5 times rms in both bands at the same sample were identified as large amplitude pulses ( laps ) .
this procedure disperses any narrow interference spike discriminating against interference .
a number of marginal laps , i.e. pulses with a peak between 3.0 to 3.5 times rms , were also identified .
we searched 2.2 million periods for psr j0218 + 4232 ( @xmath3 3.2 @xmath4 g ) and found three significant laps .
figure 1a shows the integrated profile for this pulsar and the detected laps are marked with filled circles .
the largest of these had an intensity 51 times the mean intensity ( intensity of lap , @xmath5 jy-@xmath6s ) .
our data also consisted of 9 marginal laps , 7 of which occur between phase 0.89 - 1.2 , which is the phase interval corresponding to one of the high energy peaks .
we searched about 1 million periods for psr b1957 + 20 ( @xmath3 3.8 @xmath4 g ) and found one significant lap , shown in figure 1b , with an intensity 129 times the mean intensity ( @xmath7 jy-@xmath6s ) .
in addition , we also detected 5 marginal laps .
psrs b1957 + 20 and j0218 + 4232 have the fourth and sixth highest values of @xmath0 respectively of known radio pulsars .
hence , these new detections support a connection between the magnitude of @xmath0 and the existence of gps and laps .
the data for pulsars with @xmath0 marginally below @xmath8 g are being analyzed currently . , i. et al .
1996 , apj , 457 , 81 johnston , s. , & romani , r. w. 2003 , private communication lundgren , s. c. et al .
1995 , apj , 453 , 433 romani , r. w. , & johnston , s. 2001 , apj , 557 , l93 staelin , d. h. , & reifenstein , e. c. 1968 , science , 162 , 1481 , m. et al .
2001 , apj , 554 , 316 | giant pulses ( gps ) , occasional individual pulses with an intensity 100 times the average intensity , have been detected in four pulsars todate .
their origin is not well understood , but studies suggest a connection between the strength of magnetic field at the light cylinder @xmath0 and the existence of gps . here , we report on detection of significant large amplitude pulses ( laps ) in two more pulsars with high values of @xmath0 , psrs j0218 + 4232 and b1957 + 20 , observed using giant meterwave radio telescope ( gmrt ) .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
exotic isotopes along the neutron and proton drip lines are important for our understanding of the formation of elements and they constitute tests of our understanding of nuclear structure .
the proton- and neutron - rich regimes in the chart of nuclei are therefore the focus of existing and forthcoming experimental facilities around the world @xcite .
the emergence of new degrees of freedom is one important feature of these systems ; exemplified , e.g. , by the discovery of several nuclear halo states along the drip lines @xcite .
halo states in nuclei are characterized by a tightly bound core with weakly attached valence nucleon(s ) .
universal structures of such states can be considered a consequence of quantum tunneling , where tightly - bound clusters of nucleons behave coherently at low energies and the dynamics is dominated by relative motion at distances beyond the region of the short - range interaction . in the absence of the coulomb interaction , it is known that halo nuclei bound due to a large positive s - wave scattering length will show universal features @xcite . in the case of proton halo nuclei , however , the coulomb interaction introduces an additional momentum scale @xmath1 , which is proportional to the charge of the core and the reduced mass of the halo system .
the low - energy properties of proton halos strongly depend on @xmath1 .
halo effective field theory ( eft ) is the ideal tool to analyze the features of halo states with a minimal set of assumptions .
it describes these systems using their effective degrees of freedom , i.e. core and valence nucleons , and interactions that are dictated by low - energy constants @xcite . for s - wave proton halo systems
there will be a single unknown coupling constant at leading order , and this parameter can be determined from the experimental scattering length , or the one - proton separation energy .
obviously , halo eft is not intended to compete with _ ab initio _ calculations that , if applicable , would aim to predict low - energy observables from computations starting with a microscopic description of the many - body system .
instead , halo eft is complementary to such approaches as it provides a low - energy description of these systems in terms of effective degrees of freedom .
this reduces the complexity of the problem significantly . by construction
, it can also aid to elucidate the relationship between different low - energy observables .
furthermore , halo eft is built on fields for clusters , which makes it related to phenomenological few - body cluster models @xcite .
the latter have often been used successfully for confrontation with data for specific processes involving halo nuclei .
a relevant example in the current context is the study of proton radiative capture into low - lying states states of @xcite .
a general discussion of electromagnetic reactions of proton halos in a cluster approach was given in @xcite .
the emphasis of an eft , however , is the systematic expansion of the most general interactions and , as a consequence , the ability to estimate errors and to improve predictions order by order .
the structure and reactions of one- and two - neutron halos have been studied in halo eft over the last years ( see , e.g. , refs .
however , concerning charged systems only unbound states such as @xmath2 @xcite and @xmath3 @xcite have been treated in halo eft . in this letter
, we apply halo eft for the first time to one - proton halo nuclei . we restrict ourselves to leading order calculations of systems that are bound due to a large s - wave scattering length between the core and the proton .
the manuscript is organized as follows : in sec .
[ sec : theory ] , we introduce the halo eft and discuss how coulomb interactions are treated within this framework . in the following section , we present our results and calculate , in particular , the charge form factor and charge radius at leading order .
furthermore , we derive expressions for the radiative capture cross section .
we apply our general formulae to the excited @xmath0 state of and compare our numerical results with existing data for this system .
we conclude with an outlook and a discussion on the importance of higher - order corrections .
in halo eft , the core and the valence nucleons are taken as the degrees of freedom .
for a one - proton halo system , the lagrangian is given by @xmath4 here @xmath5 denotes the proton field with mass @xmath6 and @xmath7 the core field with mass @xmath8 , @xmath9 denotes the leading order ( lo ) coupling constant , and the dots denote derivative operators that facilitate the calculation of higher order corrections .
the covariant derivative is defined as @xmath10 , where @xmath11 is the charge operator .
the resulting one - particle propagator is given by @xmath12^{-1}~.\ ] ] for convenience , we will also define the proton - core two - particle propagator @xmath13^{-1}~,\ ] ] where @xmath14 denotes the reduced mass of the proton - core system .
we include the coulomb interaction through the full coulomb green s function @xmath15 .
the dashed line denotes a core propagator , the solid line a proton propagator and the wave line denotes the exchange of a coulomb photon.,width=302 ] where @xmath16 is the coulomb four - point function defined recursively in fig .
[ fig : fourpointgamma ] . to distinguish coordinate space from momentum space states
we will denote the former with round brackets , i.e. @xmath17 .
in coordinate space , the coulomb green s function can be expressed via its spectral representation @xmath18 where we define the coulomb wave function through its partial wave expansion @xmath19 here we have defined @xmath20 and @xmath21 with the coulomb momentum @xmath22 and also the pure coulomb phase shift @xmath23 for the coulomb functions @xmath24 and @xmath25 , we use the conventions of ref .
the regular coulomb function @xmath24 can be expressed in terms of the whittaker m - function according to @xmath26 with the @xmath27 defined as @xmath28}}{2(2l+1)!}~.\ ] ] we shall also need the irregular coulomb wave function , @xmath25 , which is given by @xmath29 where @xmath30 is the whittaker w - function and the coefficient @xmath31 is defined as @xmath32 to obtain the fully - dressed two - particle propagator , that includes strong and coulomb interactions , we calculate the irreducible self - energy shown in fig .
[ fig : irredbubble ] .
@xmath33 the expression above is known and is given by @xmath34 this integral was solved in ref .
@xcite , using dimensional regularization in the power divergence subtraction ( pds ) scheme , as @xmath35 with a divergent part @xmath36 @xmath37-\frac{m_\mathrm{r}\mu}{2\pi}~,\ ] ] where @xmath38 is the space dimension , @xmath39 is the euler constant , and @xmath40 is the pds regulator .
the function @xmath41 is defined as @xmath42 with @xmath43 being the polygamma function .
note that the divergent part in eq .
( [ eq : jdivergent ] ) is energy independent .
this will become important later when the derivative of @xmath44 , with respect to the energy , will be required .
the coupling constant @xmath9 can be determined by matching to a two - body observable , such as the coulomb corrected proton - core scattering length @xcite : @xmath45 since we stay at leading order , however , an explicit expression for @xmath9 will not be required for the calculation of electromagnetic observables in the next section .
.,height=113 ] in our calculation of the charge form factor , we follow the derivation of the deuteron form factor presented in ref . @xcite . the form factor is obtained by calculating the matrix element @xmath46 for momentum transfer @xmath47 in the breit frame , where no energy is transferred by the photon .
it was shown in ref .
@xcite that this matrix element can be expressed as in our eq .
( [ eq : ff1 ] ) compared to eq .
( a10 ) of @xcite is that we have an extra @xmath48 in our definition of the irreducible self - energy . ]
@xmath49 where @xmath50 denotes the irreducible three - point function shown in fig .
[ figlogamma ] , and @xmath51 is the derivative of the self - energy with respect to the total energy evaluated at the energy @xmath52 , where @xmath53 is the proton separation energy or core - proton binding energy . with the proton - core mass ratio @xmath54 , the three - point function @xmath50 is given by @xmath55~ , \label{eqlogamma}\end{aligned}\ ] ] and the derivative of the self - energy can be written as @xmath56 evaluating @xmath50 at zero momentum transfer , by using eq .
( [ eq : cgfspectral ] ) and orthonormality of the wavefunctions , and comparing with eq .
( [ eq : dsigmae ] ) shows that the charge form factor is properly normalized to one in this limit .
we find that eq .
( [ eqlogamma ] ) can be simplified by writing the coulomb green s function for negative energy using the whittaker w - function .
this is achieved by demanding proper asymptotics and using that only the s - wave part can contribute to propagation to zero separation , that is @xmath57}{\rho'\rho}\right ) \nonumber\\ & = & i\frac{m_{\mathrm{r}}\gamma(1+k_{\mathrm{c}}/\gamma_0)}{2\pi}\frac{w_{-k_{\mathrm{c}}/\gamma_0,1/2}(2\gamma_0r)}{r}~ , \label{eq : simplcoulgreen}\end{aligned}\ ] ] where we have introduced the binding momentum @xmath58 .
the resulting integral is then @xmath59\nonumber\\ & & \times w_{-k_{\mathrm{c}}/\gamma_0,1/2}(2\gamma_0 r)^2~ , \label{eqbintegralnprime}\end{aligned}\ ] ] where @xmath60 are the spherical bessel functions .
once the parameters of the proton halo system are fixed , the equation @xmath61 is used to calculate the charge form factor and the corresponding charge radius numerically .
we have calculated these quantities for the excited @xmath0 state of , which has a proton separation energy of @xmath62@xcite .
note that the proton separation energy is the only non - trivial experimental input at lo .
the charge form factor is related to the charge radius via the expansion @xmath63 and we find for the charge radius squared @xmath64 since the proton and core are treated as structureless fields in halo eft , this quantity corresponds to the charge radius difference according to @xmath65 where @xmath66 is the charge radius squared corresponding to the particle @xmath67 , , @xmath68 .
analog to the deuteron case , the charge radii of proton and enter at higher orders in the calculation via counter terms .
the error of the eft can be estimated by comparing the momentum scale @xmath69 of the halo with the break - down scale @xmath70 of the eft .
the latter is given by the closest interfering state .
for the halo system the break - down scale is given by the bound state , @xmath71 below the @xmath0 state @xcite .
thus , the expected lo error is @xmath72 for the halo state in @xmath73 , which is comparable to the lo error of a pionless eft calculation for the two - nucleon system .
our approach can easily be applied to low - energy radiative capture .
the differential cross section for this reaction is @xmath74 where @xmath75 is the vector amplitude for the sum of the diagrams shown in fig .
[ fig : capture ] , where a proton is captured by a core while a real photon is emitted .
the relative momentum of the proton - core system is @xmath76 and the four - momentum of the photon is @xmath77 , with associated polarization vectors @xmath78 and @xmath79 .
the factor @xmath80 is the wavefunction renormalization , or lsz reduction factor .
the vector amplitude @xmath75 can be expressed as the integral @xmath81~,\end{aligned}\ ] ] where the @xmath82 has emerged from the feynman rule of the vector photon coupling and acts on the coulomb wavefunction due to a partial integration .
evaluating the angular integrals and multiplying with the polarization vector , the integral is simplified to @xmath83\nonumber\\ & & + \big[(f\to 1-f),~(z_{\mathrm{c}}\to 1)\big]\bigg|^2~ , \label{eq : captureintsimpl}\end{aligned}\ ] ] where the angles @xmath84 and @xmath85 will be integrated over to give the total cross section . for a given physical system
, we can solve the integral in eq .
( [ eq : captureintsimpl ] ) numerically using eq .
( [ eq : simplcoulgreen ] ) for the coulomb green s function . ) @xmath86 is presented by the solid black line .
the theoretical result is compared with the data by chow _
_ @xcite and morlock _ et al .
_ @xcite shown by blue triangles and green dots , respectively .
the calculation by bennaceur _
@xcite is shown by the dashed curve.,width=377 ] radiative capture into low - lying states of has been measured by rolfs _
@xcite , chow _ et al .
_ @xcite and by morlock _
_ @xcite . in fig .
[ fig : sfactor ] we show the astrophysical s - factor , defined as @xmath87 the figure shows the halo eft results of our lo calculation compared to experimental data for capture into the @xmath0 excited state and a phenomenological calculation using the shell model embedded in the continuum . at threshold , we find that @xmath88 .
our lo results are slightly low , but consistent with the experimental data , within the expected 50% error .
we anticipate that the next - to - leading order correction will increase the radiative capture cross section through the appearance of a finite effective range at this order .
it can also be noted that the results agree qualitatively with the predictions obtained in the shell model embedded in the continuum @xcite .
in this work , we have shown that coulomb effects can be included in halo eft , and that thereby static and dynamical observables of proton halo nuclei become accessible .
we have calculated the charge radius and the radiative proton capture cross section of s - wave proton halo nuclei at leading order in halo eft .
our results can be applied to any one - proton halo system whose interaction is dominated by s - waves .
in particular , the excited @xmath0 state in is known to have a large s - wave component .
we have calculated the charge radius for this system . while this observable is not yet experimentally accessible , this result provides a prediction for _ ab initio _ calculations using modern nucleon - nucleon interactions .
in addition , we have compared our results for radiative capture into the excited @xmath0 state of with experimental data and found good agreement within the expected error .
furthermore , we found that halo eft gives the same qualitative behavior for this observable as previous calculations that have employed phenomenological models . for a quantitative description of the experimental data ,
higher order corrections are required . in a future publication
, we will address how these corrections are included within halo eft in the presence of coulomb interactions .
the size of these contributions will strongly be affected by the relative size of the effective range and the coulomb momentum @xmath89 , which provides an additional scale in systems with coulomb interactions .
our calculation is also a first step towards a calculation of properties of within halo eft .
this system requires the inclusion of two low - energy constants at leading order since it interacts dominantly in the p - wave @xcite .
finally , our approach might prove useful for heavier systems whose static observables can be calculated using _ ab initio _ approaches , but for which continuum properties are not accessible within the same framework due to the computational complexity . in this scenario ,
_ ab initio _ predictions of , e.g. , the one - proton separation energy could be used to fix the halo eft parameters , which in turn could be used to predict continuum observables such as the radiative capture cross section . in the case of neutron halos ,
such an approach was recently carried out to predict novel features in the calcium isotope chain using halo eft @xcite .
we thank h. esbensen and s. knig for helpful discussions , p. mohr and k. bennaceur for supplying relevant data . this work was supported by the swedish research council ( dnr .
2010 - 4078 ) , the european research council under the european community s seventh framework programme ( fp7/2007 - 2013 ) / erc grant agreement no .
240603 , the office of nuclear physics , u.s .
department of energy under contract nos .
de - ac02 - 06ch11357 , by the dfg and the nsfc through the sino - german crc 110 , by the bmbf under contract 05p12pdfte , and by the helmholtz association under contract ha216/emmi . | we use halo effective field theory to analyze the universal features of proton halo nuclei bound due to a large s - wave scattering length . with a lagrangian built from effective core and valence - proton fields ,
we derive a leading - order expression for the charge form factor . within the same framework we also calculate the radiative proton capture cross section .
our general results at leading order are applied to study the excited @xmath0 state of fluorine-17 , and we give results for the charge radius and the astrophysical s - factor .
halo nuclei , charge radius , radiative capture , effective field theory |
the specific curve pairs are the most popular subjects in curve and surface theory and involute - evolute pair is one of them .
we can see in most textbooks various applications not only in curve theory but also in surface theory and mechanic . in this study ,
the spherical indicatrices of involute of a space curve are given . in order to make a involute of a space curve and its evolute curve slant helix , the feature that spherical indicatrices curve s need to have
are examined .
let @xmath0 be a curve with @xmath1 , where @xmath2 @xmath3 .
the arc - lenght @xmath4 of a curve @xmath5 is determined such that @xmath6 let us denote @xmath7 and we call @xmath8 a tangent vector of @xmath9 at @xmath10 .
we define the curvature of @xmath11 by @xmath12 . if @xmath13 then the unit principal normal vector @xmath14 of the curve at @xmath10 is given by @xmath15 .
the unit vector @xmath16 is called the unit binormal vector of @xmath11 at @xmath17 .
then we have the frenet - serret formulae @xmath18where @xmath19 is the torsion of @xmath11 at @xmath10 @xcite . the curve @xmath11 is called evolute of @xmath20 if the tangent vectors are orthogonal at the corresponding points for each @xmath21 in this case , @xmath20 is called involute of the curve @xmath11 and there exists a relationship between the position vectors as@xmath22where @xmath23 is the distance between the curves @xmath11 and @xmath24 at the corresponding points for each @xmath25 the pair of ( @xmath24 , @xmath11 ) is called a involute - evolute pair .
@xmath26 is not a constant for involute - evolute pairs@xcite . on the other hand , izumiya and takeuchi
have introduced the concept of slant helix by saying that the normal lines make a constant angle with a fixed straight line .
they characterize a slant helix if and only if the geodesic curvature of the principal image of the principal normal indicatrix@xmath27is a constant function , where @xmath28@xcite . in this study , we denote @xmath29 , @xmath30 , @xmath31 @xmath32 , @xmath33 and @xmath34 , @xmath35 , @xmath36 @xmath37 , @xmath38 are the frenet equipments of @xmath11 and @xmath39 respectively .
tangent , principal normal and binormal vectors are described for the spherical curves which are called tangent , principal normal and binormal indicatrices both the curves @xmath11 and @xmath39 respectively . throughout this study ,
both involute and evolute curves are regular .
in this section , we introduced the spherical indicatrices of involute curve of a curve in euclidean 3-space and gave considerable results by using the properties of the curves , similar to the previous section .
let @xmath11 be a curve with its involute curve @xmath20 then @xmath40where @xmath41and @xmath42 is definitely positive .
let @xmath43 be the sign of @xmath42 such that if @xmath44 , @xmath45 and if @xmath46 , @xmath47 we differentiate the equation ( [ 2 ] ) with respect to @xmath4 , we get@xmath48since @xmath29 and @xmath49 are orthogonal , there is no any component of @xmath50 on @xmath29 .
thus @xmath43 has to be @xmath51 .
[ t1]let @xmath20 be involute of a space curve , then we have frenet formula:@xmath52where @xmath53with the parametrization @xmath54@xmath55and the curvature and torsion of @xmath56 are @xmath57the geodesic curvature of the the principal image of the principal normal indicatrix of involute curve is@xmath58 from ( [ 5 ] ) , it is obvious that involute of @xmath11 is a planar curve if and only if @xmath11 is a generalized helix . for further usage
we denote @xmath59 as @xmath60 . by using ( [ 1 ] ) and
( [ 5 ] ) we obtained the relation@xmath61and so we have @xmath62thus we have the following theorem .
[ t2]if the frenet frame of the tangent indicatrix @xmath68 of involute of @xmath10 is @xmath69 , we have frenet formula:@xmath70where@xmath71with the parametrization@xmath72and the curvature and torsion of @xmath73 are@xmath74the geodesic curvature of the principal image of the principal normal indicatrix of @xmath73 is@xmath75 let @xmath20 be involute of a space curve @xmath11 then spherical image of the tangent indicatrix of @xmath20 is a spherical helix if and only if involute of @xmath11 is a slant helix . in this case ,
spherical image of the tangent indicatrix of @xmath20 is a slant helix on unit sphere too . if the frenet frame of the principal normal indicatrix @xmath82 of involute of the curve @xmath10 is @xmath83 , we have frenet formula:@xmath84where@xmath85with the parametrization@xmath86and the curvature and torsion of @xmath87 are@xmath88 + \left [ \left ( \tfrac{-% \widetilde{f}^{^{\prime } } \left ( 1+\widetilde{f}^{2}\right ) ^{\frac{3}{2}}}{% \rho } \right ) \left ( \tfrac{\widetilde{\kappa } ^{2}\left ( 1+\widetilde{f}% ^{2}\right ) ^{\frac{5}{2}}}{\rho } \right ) ^{^{\prime } } \right ] + \left [ \left ( \tfrac{\widetilde{\kappa } \widetilde{f}^{^{\prime } } \left ( 1+% \widetilde{f}^{2}\right ) ^{\frac{3}{2}}}{\rho } \right ) ^{^{\prime } } \left ( \tfrac{\widetilde{\kappa } \left ( 1+\widetilde{f}^{2}\right ) ^{\frac{5}{2}}}{% \rho } \right ) \right ] \right\ } \notag\end{aligned}\ ] ] where @xmath89@xmath90the geodesic curvature of the principal image of the principal normal indicatrix of @xmath87 is@xmath91where if the frenet frame of the binormal indicatrix @xmath96 of involute of the curve @xmath10 is @xmath97 , we have frenet formula:@xmath98where @xmath99with the parametrization @xmath100and the curvature and torsion of @xmath101 are@xmath102the geodesic curvature of the principal image of the principal normal indicatrix of @xmath101 is@xmath103 let @xmath11 be a space curve and @xmath20 be its involute with nonzero torsion then spherical image of binormal indicatrix of @xmath24 is a circle on unit sphere if and only if @xmath104 is a generalized helix .
let @xmath112 and @xmath113be two regular curves in @xmath114then @xmath115 and @xmath116 are similar curves with variable transformation if and only if the principal normal vectors are the same for all curves @xmath117 under the particular variable transformation @xmath118@xmath119of the arc - lengths .
let @xmath120 and @xmath121be two regular curves in @xmath114then @xmath73 and @xmath122 are similar curves with variable transformation if and only if the principal normal vectors are the same for all curves @xmath123 under the particular variable transformation @xmath118@xmath124of the arc - lengths . in @xcite , the general equation of spherical helix family
is obtained by monterde which is , @xmath125where @xmath126 @xmath127 in 4b , kula et al .
obtained the general equation of a slant helix family similar to following@xmath128where @xmath129}{% 2w\left ( w+1\right ) } + \frac{\left ( w+1\right ) \sin [ \left ( w-1\right ) t]}{% 2w\left ( w-1\right ) } \\ \gamma _ { \mu } ^{2}(s ) & = & \frac{\left ( w+1\right ) \cos [ \left ( w-1\right ) t]}{% 2w\left ( w-1\right ) } + \frac{\left ( w-1\right ) \cos [ \left ( w+1\right ) t]}{% 2w\left ( w+1\right ) } \\ \gamma _ { \mu } ^{3}(s ) & = & -\frac{\cos \left ( s\right ) } { \mu w}.\end{aligned}\]]and @xmath130 @xmath131 by using the theorem [ c5 ] , we can obtained the general equation of general helix family in euclidean 3-space according to the non - zero constant @xmath132 as follows@xmath133where@xmath134+\left ( w-1\right ) \cos [ \left ( w+1\right ) t]\right\ } \\ & & + \frac{1}{2w\left ( w^{2}-1\right ) } \left\ { \left ( w+1\right ) ^{2}\sin [ \left ( w-1\right ) t]+\left ( w-1\right ) ^{2}\sin [ \left ( w+1\right ) t]\right\ } \\ \widetilde{\gamma } _ { \mu } ^{2}(s ) & = & \frac{-\left ( c - s\right ) } { 2w}\left\ { \left ( w+1\right ) \sin [ \left ( w-1\right ) t]+\left ( w-1\right ) \sin [ \left ( w+1\right ) t]\right\ } \\ & & + \frac{1}{2w\left ( w^{2}-1\right ) } \left\ { \left ( w+1\right ) ^{2}\cos [ \left ( w-1\right ) t]+\left ( w-1\right ) ^{2}\cos [ \left ( w+1\right ) t]\right\ } \\ \widetilde{\gamma } _ { \mu }
^{3}(s ) & = & \frac{\left ( c - s\right ) \sin \left ( s\right ) -\cos \left ( s\right ) } { \mu w}\end{aligned}\]]and @xmath130 @xmath135 with the parametrization@xmath136and the curvature and torsion of curve @xmath137 is @xmath138 | in this work , we studied the properties of the spherical indicatrices of involute curve of a space curve and presented some characteristic properties in the cases that involute curve and evolute curve are slant helices and helices , spherical indicatrices are slant helices and helices and we introduced new representations of spherical indicatrices . |
in the context of disk galaxies , a nuclear or circumnuclear ring is a region of large gas surface density and enhanced star formation , located within 2 kpc of the galactic nucleus . @xcite
distinguish between nuclear disks and nuclear rings , by setting the maximum width of the ring to half the ring radius .
the most compact rings ( ultra - compact nuclear rings , ucnrs ) have radii smaller than 200 pc @xcite .
the atlas of images of nuclear rings @xcite is the most complete atlas of galactic nuclear rings to date and includes 101 barred and unbarred disk galaxies with rings .
the dust nuclear rings around elliptical and s0 galaxies constitute a different class of galactic ring with no associated star formation .
the present work is not concerned with the latter rings .
circumnuclear rings in disk galaxies give rise to starbursts @xcite , which observationally are ring- or arc - shaped .
note that the term `` starburst '' does not have a general definition @xcite .
a `` circumnuclear starburst '' is a region composed of star condensations with individual masses ranging from a few@xmath6 to greater than @xmath7m@xmath2 , bright in the ultraviolet ( uv ) because they have recently emerged from their birth clouds ( the nuclear galactic ring ) , and contain hot and massive o and/or b stars
. proposed scenarios for the origin of star - forming galactic rings are galaxy collisions or mergers , accretion of intergalactic gas , and resonances caused by the actions of rotating bars or other non - axisymmetric disturbances on the motions of disk gas clouds @xcite .
note that in their analysis of h@xmath8 and continuum images of 327 local disk galaxies , @xcite found no significant increase in the central concentration of star formation as a result of the presence of a close companion .
on the other hand , there is strong observational evidence for a connection between bars and circumnuclear starbursts @xcite .
in addition , non - starburst barred galaxies may eventually become starburst barred galaxies @xcite .
the location of nuclear rings in barred galaxies may be set by inner lindblad resonances ( ilrs ) , which can prevent gas from flowing further in .
such resonances can cause gas to accumulate in circumnuclear rings where clusters can form @xcite .
alternative models for the location of starbursts in barred galaxies can be found in @xcite and @xcite , while @xcite have a model for the migration of circumnuclear star clusters and nuclear rings .
circumnuclear starbursts in barred galaxies are perhaps the most numerous class of nearby starburst regions @xcite
. their detailed study can provide a stepping stone for studies of starbursts at higher redshifts . indeed ,
local starburst galaxies ( including circumnuclear starbursts ) , show star formation rate densities @xcite , uv colors @xcite , and spectral morphologies @xcite , similar to those of high redshift lyman - break galaxies ( also see @xcite ) .
determining the ages , masses , and stellar initial mass functions ( imfs ) of individual star clusters within circumnuclear starbursts is important for understanding the relations between ( 1 ) galaxy dynamics and interactions and the origin of circumnuclear star clusters ; ( 2 ) massive star feedback and the fueling of active galactic nuclei ( agn , @xcite ) ; and ( 3 ) young massive star clusters and globular clusters @xcite . in this study
, we determine the ages and masses , and constrain the imfs of 13 star clusters in messier 83 ( m83 , ngc 5236 ) , which hosts the nearest example of a circumnuclear starburst in a barred galaxy .
m83 is a nearby ( [email protected] , @xcite ) , southern ( dec[j2000]@xmath10 ) , nearly face - on ( @xmath11 , @xcite ) , sab(s)c grand - design spiral galaxy @xcite , with an optical disk spanning @xmath12 ( @xmath13kpc@xmath14kpc ) . @xcite
have estimated that at the distance of m83 , a central black hole of moderate size and luminosity would be detectable .
however , the galaxy shows no indication from radio or x - ray observations of hosting an agn @xcite .
m83 has a close dynamical companion in the dwarf irregular galaxy ngc 5253 @xcite , which contains a starburst nucleus of @xmath15pc in size ( @xcite , @xcite , @xcite ) .
m83 s arc - shaped circumnuclear starburst has been observed at wavelengths ranging from the radio to the x - rays ( see @xcite for references ) .
it spans about 200pc in length and 35pc in thickness , and it is located between two rings of dusty gaseous material which may be coincident with two inner lindblad resonances @xcite .
the starburst qualifies as an ucnr ( ring radius of @xmath16 pc ) , and as shown in this work , it breaks up into about 20 compact fuv bright star clusters .
two proposed scenarios for its origin are , the interaction of m83 with its neighbor ngc 5253 about 1gyr ago @xcite , and/or the merging of m83 with a dwarf satellite in the past .
the former scenario is supported by the `` finger '' of h i gas projecting outward from the halo of m83 toward ngc 5253 @xcite , and by the presence of off - disk h i clouds in m83 @xcite .
the merger scenario is based on kinematic studies using r band @xcite , near - infrared @xcite , and submillimeter @xcite spectroscopy , which have revealed the presence of a controversial second nucleus , hidden by dust , more massive than the optical nucleus , and located @xmath17 northwest of the latter .
recent n - body simulations by @xcite , which account for this second nucleus , predict the disruption of the circumnuclear starburst in less than an orbital time . however @xcite have concluded that there is no second nucleus in m83 , since they can not confirm the local maximum in the stellar velocity dispersion at the location of the `` second nucleus '' found by @xcite . a third scenario for the origin of the starburst could be internal galaxy dynamics associated with the bar .
the ages of circumnuclear star clusters in m83 have been previously estimated .
based on the analysis of co absorption and br@xmath18 emission - line data @xcite found that the northwestern end of m83 s starburst is the youngest , while the southeastern end is the oldest .
@xcite compared _
hst_/wfpc2 photometry with theoretical population synthesis models and found that most of m83 s circumnuclear star clusters are less than 10 myr old , with clusters younger than 5 myr preferentially located along the northwestern end .
the statistical analysis of the photometry of @xcite by @xcite confirms the age gradient along the starburst . in their analysis ,
@xcite divided the @xmath19 starburst region into 25@xmath20 angular sectors , and considered the mean cluster age as well as the oldest age within each sector , finding 5myr for the oldest cluster in the northwest and 25myr for the oldest cluster in the southeast .
the above age gradient suggests that star formation occurs in an ordered manner , and supports the idea that the starburst is fed by the inflow of bar - driven material .
if real , this gradient has profound implications for the formation and evolution of circumnuclear starbursts .
for this reason , in this work , we age - date 13 clusters within the northern portion of m83 s starburst , using three different techniques , i.e. , by comparing archival _
fuv spectroscopy of the clusters with semi - empirical and new fully theoretical model spectra , and by deriving the ages from high spatial resolution _
hst_/wfc3 multi - band optical photometry .
although our study does not cover the full extent of the starburst s arc , this does not affect our results .
in addition , we analyze the massive star population in regions between the clusters .
another goal of this work is to constrain the upper imfs of our clusters .
this is particularly interesting in the metal rich circumnuclear environment of m83 .
m83 s circumnuclear h ii regions have an oxygen to hydrogen number density ratio of 12+log(o / h)[email protected] @xcite , and a corresponding fraction of all metals to all elements by mass ( hereafter , metallicity ) that is intermediate between z=0.020 and z=0.040 , and supersolar ( 12+log[o / h]@[email protected] , z@xmath2=0.013 , @xcite ) .
we estimate the cluster masses from spectroscopy and photometry , and check which clusters have formed stars more massive than @xmath22m@xmath2 .
the random sampling of the stellar imf may prevent this from happening in low mass clusters @xcite .
this paper is organized as follows : [ observations ] describes the observations and data reduction , [ models ] presents the models used in the estimate of the spectroscopic ages and masses of the clusters , [ analysis ] explains our procedure for determining the spectroscopic and photometric properties of the clusters , in [ discussion ] we present and discuss our results , finally , we conclude in [ conclusion ] .
we analyzed archival fuv imaging and spectroscopy of compact star clusters and field regions located between about @xmath23 ( @xmath24110pc ) and @xmath25 ( @xmath24198pc ) of m83 s optical center , and spanning an arc of @xmath26200pc in size .
the data were taken with the stis onboard the _ hst _ as part of programs gto-7577 and gto-8785 ( pi : sara heap ) . we complemented these data with optical images of the same clusters taken with _
s wfc3 as part of program go / dd-11360 ( pi : oconnell ) . @xcite
discuss properties derived from optical photometry of star clusters within the larger field of view covered by the wfc3 images .
our main focus is on the fuv data , but we also briefly discuss the optical data .
figure [ fig_composite ] shows a false - color logarithmic - scale image of m83 s arc - shaped circumnuclear starburst . the image is a composite of the stis image taken with the f25qtz ( fuv ) filter , and wfc3 images taken with the f657n ( h@xmath8 ) and f555w ( v ) filters .
north is at the top and east is to the left .
the optical nucleus of m83 is the large yellow - green dot in the upper left .
the starburst starts northwest of the nucleus , wraps about the nucleus towards the southeast , and is punctuated by about 20 uv bright clusters .
the fuv bright clusters appear white , while fainter ones appear blue . the location of sn1968l , which was identified by @xcite , is indicated . at m83 s distance , 1
represents about 22 pc .
the two dusty rings discovered by @xcite and embracing the starburst are clearly seen .
the clusters with available stis fuv spectroscopy are labeled .
note that although clusters 3 , 4 and 7 seem to be composed of several bright knots clustered together , one knot always dominates the brightness .
figure [ fig_composite ] was kindly produced for us by w. p. blair .
the fuv data set consists of a direct image and four two - dimensional ( 2-d ) spectrograms .
the direct image is projected onto the mama detector , which has a field of view of 25x25 and a plate scale of 0@xmath27024 pixel@xmath28 .
the spectrograms , one long slit ( 25x2 ) and three multi - object ( slitless ) , were taken with the g140l grating , which has an average dispersion of 0.584 pixel@xmath28 .
table [ tab_data ] gives the observation date ( col .
2 ) , exposure time ( col .
3 ) , aperture position angle in degrees east of north ( col .
4 ) , filter name ( col .
5 ) , wavelength coverage for the detector@xmath29filter combination ( col .
6 ) , and extension of the most processed data file available from the hubble data archive of the fuv observations .
the data were partially processed through the on - the - fly reprocessing ( otfr ) pipeline of the archive , at retrieval time .
this included data linearization , dark subtraction , flat fielding , and summing of the repeated observations .
in addition , the pipeline corrected the direct image for geometric distortion .
after processing through the pipeline , in order to increase the signal - to - noise ( s / n ) , we added together exposures taken with the same instrument configuration , i.e. , the two slitless exposures taken with the f25sfr2 aperture , which we refer to as the f25sfr2 exposure .
because the spectrograms were taken in either a slitless or a wide slit mode with the g140l grating , the spectral traces have large offsets in the dispersion direction .
therefore , the otfr pipeline did not wavelength calibrate or convert these spectra to absolute flux . in order to extract wavelength - calibrated one - dimensional ( 1-d ) spectra from the 2-d spectra , we first located the position of each spectral trace along the cross dispersion direction , and then used the x1d module ( mcgrath , busko , & hodge , stis isr 99 - 03 ) available in the image reduction and analysis facility ( iraf , @xcite ) for this task .
several iterations were required in order to find the optimal value of xoffset , which is the parameter used by x1d for shifting a spectrum along the dispersion direction .
we used the absorption feature c ii @xmath30 ( hereafter c ii 1335 ) , for finding the optimal value of xoffset . in our spectra , this feature is unresolved and is a blend of milky way interstellar and m83 stellar components . for the width of the extraction box we used the value recommended in the stis manual for point sources , i.e. , 11 pixels @xcite .
background spectra extracted from outer portions of the 2-d spectrograms were assumed to represent purely geocoronal emission ( air glow ) from h i lyman-@xmath8 ( ly@xmath8 ) at @xmath31 and/or , [ o i ] @xmath32 , depending on the aperture , and were subtracted from the cluster spectra .
we did not subtract the underlying diffuse light because of its large variations as a function of position .
the direct image and the 2-d spectrograms from the stis are shown in the four panels of fig .
[ fig_fuvset ] .
panel ( a ) features the 25x25 direct image in logarithmic scale , with the location of m83 s optical center marked with a cross , and the spatial scale marked with a vector .
panels ( b ) and ( c ) show the long - slit and the multi - object spectrograms respectively , both taken with the same aperture ( 25mama ) .
the vertical band in panel ( b ) is due to ly@xmath8 airglow filling the 2-wide slit ( it is seen more clearly in the electronic version ) .
the ly@xmath8 airglow contaminates much of the slitless spectrogram shown in panel ( c ) .
finally , panel ( d ) features the sum of the two slitless spectrograms taken with the f25sfr2 aperture , which rejects the ly@xmath8 airglow .
no contamination from the [ o i ] air glow is apparent in panel ( d ) because the emission brightness geocoronal [ o i ] is typically less than @xmath33 of that of geocoronal ly@xmath8 , depending on the time of observation and the position of the target relative to the earth @xcite .
we used the slitless exposures for extracting spectra of those clusters falling outside the slit , as well as for studying the effect of rejecting the ly@xmath8 air glow on our results .
the spectral resolution , depends on the spatial extent of the star clusters . in order to estimate the spectral resolution of our spectra , we determined the full width at half maximum ( fwhm ) of the cross dispersion profiles of the spectral traces .
for our point - like representative , gd 153 , the fwhm is 0.09 . due to blending with the profiles of close neighbours
, we were only able to determine the fwhm of four clusters by our method . for clusters 1 , 4 , 10 , and 11 + 12 we obtained fwhm of 0.17 , 0.30 , 0.19 , and 0.23 , respectively , which suggests that the clusters are partially resolved .
since these are neither point sources nor fully extended sources , we expected the spectral resolution of our long - slit spectra to be intermediate between the values given in the stis instrument handbook , i.e. , 0.9 ( point source ) and 48 ( extended source ) .
multi - component gaussian fitting of the c ii feature at @xmath30 , carried out with the peak analysis code ( available at ftp://ftp.ncnr.nist.gov/pub/staff/dimeo/pan.zip ) reveals that the fwhm of this feature ranges from 2.5 - 7 .
our field spectra are expected to have the resolution of a fully extended source .
we corrected the cluster spectra for reddening due to the presence of dust in the galaxy and in m83 .
for the galaxy s contribution to the reddening , we adopt @xmath34 @xcite , and we assume the galactic extinction curve of @xcite . for m83 s contribution ,
we ( i ) fitted a power law of the form @xmath35 to the continuum in the range @xmath36 , ( ii ) assumed that any deviation of @xmath37 from @xmath38 , which is the expected value for a dust free starburst , is due to extinction by local dust , and ( iii ) dereddened the observed spectrum until @xmath38 was reached , using the starburst obscuration law of @xcite . in order to match each cluster with its fuv spectral trace , we started by directly comparing the positions of the clusters in the stis direct image , with the positions of the spectral traces in the 2-d spectrograms .
for this , we rotated the image and vertically shifted the spectrograms until the orientations matched , as shown in fig .
[ fig_fuvmatch ] , which shows enlarged portions of the stis image , side by side with two spectrograms .
note that the slitless spectrograms all have the same orientation , while the orientation of the long - slit spectrogram is different .
therefore , in fig . [ fig_fuvmatch ]
, we show only the slitless spectrogram taken with the f25sfr2 aperture . in panel
( a ) of fig .
[ fig_fuvmatch ] , we indicate the position of the long - slit , the locations where we extracted the sky and field region spectra , and the correspondence between clusters within the slit and their spectral traces . panel ( c ) , which shows the same image as panel ( a ) , but with a slightly different orientation , shows only one sky region due to the fact that the second falls outside of the field displayed in the panel . unfortunately , and as can be seen in fig .
[ fig_fuvmatch ] , in spite of the different orientations of the spectrograms , the spectral traces of the two brightest clusters , 11 and 12 , always overlap .
furthermore , as seen in panel ( d ) , the faint spectral traces of clusters 4 and 5 also overlap , as do the traces of cluster 14 and a source not in our sample .
fortunately , we can use the long - slit spectra for clusters 4 and 14 .
however , cluster 5 was rejected from our analysis due to its poor spectral trace .
the spatial correspondence between star clusters and their spectral traces can be checked in fig .
[ fig_fuvprof ] , where we compare the cluster radial profiles derived from the stis direct image ( bottom curves in left and right panels ) , with the cross dispersion profiles derived from the long - slit ( top curve , left panel ) and the slitless ( top curve , right panel ) 25mama spectrograms . also shown in fig .
[ fig_fuvprof ] is the cross - dispersion profile of a point - like source , represented by white dwarf gd 153 ( top dotted curve in left and right panels ) , observed as part of program 7097 , in the same configuration as the long - slit spectrogram . for convenience
, we divided each curve in fig .
[ fig_fuvprof ] by the amplitude of the tallest peak , which for the clusters is the amplitude of target 11@xmath3912 ( which overlap ) , and for the white dwarf is its own amplitude . for clarity , the top curves in fig . [ fig_fuvprof ] ( cross dispersion profiles ) were all shifted by 0.2 .
the clusters and regions used for the background subtraction are labeled .
for the radial profiles , we first rotated the image to match the orientation of each spectrogram , and we then obtained the count rate at a given vertical position by integrating over the horizontal direction .
the radial profiles in the left and right panels look slightly different because of the different angles by which we rotated the image .
we note that although the peaks for clusters in our sample are dominated by emission from the clusters themselves , diffuse light and fainter sources also contribute to their shape and amplitude .
this however , does not affect the identification of the spectrum of each cluster . for the spectral profiles ,
the count rate at a given position in the cross dispersion direction was obtained by integrating over the dispersion axis .
note the higher background of the top right curve with respect to the top left curve , which is due to contamination with ly@xmath8 air glow over a larger portion of the detector in the slitless spectrogram ( right panel of fig .
[ fig_fuvprof ] ) .
circular aperture photometry was performed on the clusters using the iraf task phot , using an aperture radius of 3 pixels and a background annulus with radii of 10 and 13 pixels .
aperture corrections , based on the estimated size of each cluster , were applied to convert the fixed aperture magnitudes to total magnitudes .
the instrumental magnitudes in the f336w , f438w , f555w , f658n , and f814w filters were converted to the vegamag photometric system by applying zeropoints available at the following url : http://www.stsci.edu/hst/wfc3/phot_zp_lbn .
we refer to the broad - band filters as the `` @xmath40 '' , `` @xmath41 '' , `` @xmath42 '' , and `` @xmath43 '' magnitudes for simplicity , although they were not transformed to the johnson - cousins system . for more details ,
refer to @xcite .
we compared our observed reddening corrected fuv spectra with synthetic dust free spectra generated with the windows version @xcite of the widely used package starburst99 ( s99 , @xcite ) .
the main input parameters in s99 are 1 ) the star - formation law , 2 ) the imf , 3 ) the metallicity and stellar evolution tracks , and 4 ) the stellar spectral library .
we tried the following parameters .
1 ) a single burst of star formation , since @xcite ruled out the possibility that star formation within individual clusters proceeded continuously . indeed , in m83 s circumnuclear region , the he i 5876 absorption line is detected in h ii regions ionized by the youngest clusters , but not in those ionized by clusters older than about @xmath44myr . if star formation was continuous , the line strength ratio of he i 5876 to h@xmath37 would be the equilibrium value in the older clusters .
2 ) imfs with high mass limits of 10 , 30 , 50 , and 100@xmath45 .
3 ) the stellar evolution tracks for non - rotating stars of @xcite for masses @xmath46 , and of @xcite for masses @xmath47 , at metallicities @xmath0 and @xmath1 . note that from h ii region measurements , @xcite found that the metallicity of m83 s nuclear region is @xmath48 , which is in between the above two metallicities .
4 ) an empirical stellar library of galactic o and b stars , as well as a new uv high resolution theoretical library for massive stars of metallicities @xmath49 , and @xmath50 .
the difference in metallicity between the stellar evolutionary tracks and the theoretical stellar library is due to the use of different solar compositions .
indeed , the wm - basic models use the revised solar composition of asplund et al .
( 2005 ) , which has a lesser content in metals than the solar composition of the stellar evolution tracks .
note that there are no empirical uv stellar libraries for @xmath51 .
we describe the empirical and the new fully theoretical stellar spectral libraries in [ libraries ] .
we discuss the advantages and disadvantages of using the theoretical library in [ theo_lib ] .
hereafter , we refer to single stellar population ( ssp ) spectra generated with the empirical stellar library as the semi - empirical models , and to ssp spectra computed with the theoretical stellar library , as the theoretical models .
we compare theoretical models at @xmath0 and @xmath1 in [ solar_vs_supersolar ] .
we compare theoretical models corresponding to different high mass limits of the imf in [ upimf ] . finally , we compare the two sets of models in [ old_vs_new ] .
the semi - empirical models were built from spectral libraries of galactic o and b stars observed in the @xmath52 range at 0.12 resolution with the far ultraviolet spectroscopic explorer @xcite , and in the @xmath53 range at 0.75 resolution with the international ultraviolet explorer @xcite .
comparisons of the semi - empirical models with observations of local and distant galaxies generally yield quite good agreement @xcite .
the theoretical models were built from a spectral library of hot massive stars ( @xmath54k , @xmath55 ) that was computed as described in leitherer et al .
( 2010 ) , using the stellar atmosphere code wm - basic @xcite , following the approach pioneered by @xcite .
the models span the spectral range @xmath56 at high ( 0.5 ) resolution .
this resolution is sufficient to resolve the main uv stellar wind lines , which have widths in excess of 1000kms@xmath28 .
the strengths and weaknesses of the theoretical stellar library are the following .
1 ) the two main publicly available codes optimized for modeling the uv spectra of ob stars with winds are cmfgen @xcite and wm - basic .
wm - basic , which we utilized , solves the radiative transfer and the hydrodynamics self - consistently to derive the density structure of the stellar wind , while cmfgen assumes an analytical approximation for the wind structure a priori .
2 ) wolf - rayet ( w - r ) stars , whose stellar evolution is quite uncertain , are modeled using wm - basic atmospheres .
this ensures that their continuum fluxes are accounted for , but w - r - specific lines are not reproduced correctly .
fortunately , w - r stars in the local universe do not contribute significantly to the uv spectrum of ob stellar populations , except for he ii @xmath571640 , which should be viewed with care . in individual w - r stars , the equivalent width ( ew ) of the he ii 1640 emission component ranges from 3 to 100 @xcite , with a preference for smaller values ( 3 - 30 or so ) for late - type wn stars , which tend to dominate observed statistics due to their high absolute visual magnitude . in starburst clusters ,
the typical ew of the 1640 bump is about 2 - 3 @xcite .
if ew=15 is the typical value for late - type wn stars , this suggests a continuum diluting about 80% of the light at 1640 , and w - r stars can not contribute more than 20% to the uv continuum .
the dilution is caused by the strong continuum of o main sequence and o supergiant stars .
3 ) we omit the geometrical effect of clumps in the stellar wind , but the associated x - ray emission from shocks propagating in the wind is included .
clumping affects the lines from trace ions , e.g. , o vi @xmath571035 and o v @xmath571371 @xcite .
in addition , the strength of the wind lines depends on the assumed ionization balance . while photo - ionization is the primary mechanism determining the ionization state of the wind , x - rays arising from shocked material can be important as well ( e.g. , @xcite ) .
4 ) we omit stark broadening of the hydrogen lowest lyman lines , which is computationally expensive . a pure doppler profile of hydrogen
is very similar to one with stark broadening included , as long as the upper principal quantum of the transition is small @xcite .
this is the result of the stark width s dependence on the fourth power of the upper principal quantum number . in practice
, this means stark broadening becomes negligible in hydrogen lines at higher energies and originating at lower densities .
lyman-@xmath8 , owing to its large optical depth , is formed far out in the wind even at relatively low mass - loss rates . therefore ,
stark broadening is not important at ages younger than 10 myr .
5 ) the evolution of massive stars in binary systems and of rotating massive stars requires further study before it can be included in the models @xcite .
6 ) the models can be used for interpreting the rest - frame uv spectra of objects with metallicities other than near - solar or those of the large ( @xmath58 ) and small ( @xmath59 ) magellanic clouds @xcite . because there are no empirical uv spectral stellar libraries at @xmath51 , we have to rely on fully theoretical star cluster spectra , at @xmath1 . here
we discuss differences in the theoretical spectra of single stellar populations that were computed with stellar evolution tracks corresponding to metallicities of @xmath0 and @xmath1 .
line - driven wind theory predicts lower mass - loss rates and lower wind terminal velocities for o stars of lower metallicity @xcite , and therefore , weaker wind features in the spectra of these stars .
observations of ms o stars in the magellanic clouds and the galaxy support the theory .
for instance , see @xcite , where empirical stellar wind results for the milky way , the lmc and the smc are compared with the ob star theoretical wind models of @xcite , which are an extension of line driven wind theory .
wm - basic models do a good job of reproducing lower wind terminal velocities at lower metallicity @xcite , as can be seen in fig .
[ fig_solar_vs_supersolar ] , which shows a comparison between the @xmath0 and the @xmath1 models , for ages from one to 20 myr ( as labeled on the right ) , in the wavelength range @xmath62 , at 0.75 resolution .
for instance , compare the absorption components of c iv 1550 at @xmath0 and @xmath1 .
in addition , the photospheric lines are deeper at the higher metallicity , as expected .
figure [ fig_upimf ] shows semi - empirical and theoretical models for ssps with different high mass limits of the stellar imf .
the figures show ages 2 , 3 , 5 , and 10myr , and focus on c iv 1550 and si iv 1400 , whose profiles are sensitive to the distributions of o main sequence and o supergiant stars , respectively .
all models in fig .
[ fig_upimf ] have a lower mass limit of @xmath63m@xmath2 .
the thin black curves use a kroupa imf from 0.1 to 100m@xmath2 , while the thick grey curves ( red in the electronic version ) use imf high mass limits ( @xmath64 ) of 10 , 30 , and 50m@xmath2 . lowering @xmath64 ,
i.e. , gradually removing the o and the most luminous b stars , has the following effects .
( i ) as @xmath64 decreases , the p - cygni profiles of si iv 1400 and of c iv 1550 become weaker . in particular , the emission component of these profiles is non - existent at @xmath65 for the semi - empirical models , and drops in strength more gradually in the theoretical models . also the strong absorption component of these profiles
is purely photospheric at @xmath66 , for both models .
( ii ) for the theoretical models at @xmath67 , the absorption lines are dominated by photospheric lines from b stars . here
we discuss differences between the semi - empirical and the theoretical models at @xmath0 .
figure [ fig_old_vs_new ] shows the theoretical and semi - empirical models , using the same format of fig .
[ fig_solar_vs_supersolar ] .
there are several noteworthy features .
1 ) in the region below 1240 the semi - empirical spectra suffer from ly@xmath8 absorption in the galactic plane .
2 ) the p - cygni profiles of the resonant doublets of n v at 1238.8 + 1242.8 ( hereafter n v 1240 ) and of c iv at 1548.2 + 1550.8 ( hereafter , c iv 1550 ) , which originate in the stellar wind , decrease in strength as the o stars evolve and disappear .
3 ) the resonant doublet of si iv at 1393.8 + 1402.8 ( hereafter , si iv 1400 ) is absent in main sequence ( ms ) o stars , and develops p - cygni profiles in giant and supergiant o stars .
this is why the p - cygni profile of si iv 1400 is strong at ages @xmath68myr , when o supergiants are present .
note that the lack of strong n v 1240 , c iv 1550 , or si iv 1400 p - cygni profiles , in the spectrum of a star cluster , indicates either that the cluster did not form very massive o stars , or that the cluster is too old to show the signatures of such stars .
4 ) at 3 myr , the semi - empirical model has somewhat stronger n v , si iv , and c iv lines than the theoretical model . at this particular time step , stars with stronger winds make a stronger contribution to the semi - empirical than to the theoretical spectrum .
this difference serves as a reminder of a crucial assumption made when linking model spectra to stellar evolution models . in the case of the theoretical spectra ,
one simply adopts the effective temperature of the atmospheres for matching the positions of the evolutionary tracks .
in contrast , for the empirical spectra , we obtain the effective temperatures from spectral types via a temperature scale .
starburst99 assumes the relation between spectral type and effective temperature of o stars derived by @xcite .
we note that switching between the relations of martins et al .
and the earlier work of @xcite or @xcite modifies the peak of the si iv emission by almost 10% .
careful comparisons with observations should provide guidance for choosing the preferred approach in the synthesis models .
in addition , the wm - basic models use the revised solar composition of asplund et al .
( 2005 ) , which has a lesser content in metals than the solar composition of the stellar evolution tracks .
while the revised solar abundances have a small effect on the evolutionary tracks of massive stars , they do affect the computed uv spectra via changed opacities and wind properties .
5 ) many of the narrow absorption lines in the semi - empirical models are of interstellar origin . for the first four myr , the absorption line si ii @xmath571264 is stronger in the semi - empirical models because of the interstellar contribution .
6 ) o v @xmath571371 , which is strong at ages @xmath69myr , i.e. , in the hottest stars , is stronger in the theoretical models due to the neglect of wind clumping @xcite .
table [ tab_observed ] allows an assessment of the uncertainties in the reddened flux at 1500 , @xmath70 , and in the power law index of the reddened fuv continuum , @xmath71 , for the 13 clusters in our final sample . for each spectral exposure and each cluster , columns ( 2)-(12 ) in the table give ( 1 ) the cluster i d ; ( 2 ) and ( 3 ) the right ascension and declination measured from the stis direct image ; ( 4)-(6 ) the median signal to noise ( s / n ) of the reddened spectrum ; ( 7)-(9 ) @xmath70 ; and ( 10)-(12 ) @xmath71 .
the faint clusters 4 and 14 have the lowest s / n values .
the rest of the clusters have reasonably good s / n ( @xmath72 ) .
the @xmath70 fluxes from the slitless exposures taken with the 25mama and f25sfr2 apertures agree within @xmath33 .
therefore , the ly@xmath8 airglow has little effect on @xmath70
. the slit blocks the light from nearby sources and lowers @xmath70 by @xmath73 .
the value of @xmath71 ( i ) increases by @xmath74 when the wavelength range for fitting the continuum is extended ( compare the two values in column 10 ) ; ( ii ) varies by at most @xmath33 , due to geocoronal ly@xmath8 contamination ( compare columns 11 and 12 ) ; and ( iii ) increases by at most @xmath33 when the slit is removed ( compare columns 10 and 11 or 12 ) .
the properties of clusters @xmath75 derived from the slitless spectra are less reliable than those derived from the long - slit spectra , and are only shown for comparison . since the two slitless exposures yield similar values in tab .
[ tab_observed ] , we decided to discard the exposure spanning the shortest wavelength range , i.e. , the one taken with the f25sfr2 aperture .
for the nine clusters within the slit , i.e. , 1 , 2 , 3 , 4 , 10 , 11 , 12 , 13 , and 14 , the ` good ' data span the wavelength range @xmath76 , while for the rest of clusters , i.e. , 6 , 7 , 8 , and 9 , they span the range @xmath77 .
we adopted the algorithm developed by @xcite for deriving the age and mass of each cluster from the best fit model to the data .
the algorithm takes as input the de - reddened cluster spectrum and a grid of starburst99 models spanning ages from 1 to 20myr in step sizes of 0.1myr .
the goodness of the fit is characterized by @xmath78 , where @xmath79 , and where @xmath80 represents the observed data for the @xmath81th pixel , @xmath82 the model data , @xmath83 the error in the observed spectrum , and @xmath84 the assigned weight .
the weighting scheme eliminates from consideration the interstellar lines , and gives the stellar wind lines a weight 1.5 times greater than that of the continuum .
prior to fitting , the resolution of the models was degraded to match the approximate resolution of the observed spectra .
the spectroscopic masses were derived by ( i ) fitting power laws to the continua of the observed and the model spectra of each cluster , where the model corresponds to the best fit age and a stellar mass of 10@xmath85m@xmath2 ; ( ii ) computing the ratio of the observed to the model power law at each wavelength point ; ( iii ) taking the average of the sum of the ratios obtained in ( ii ) ; and ( iv ) multiplying the result by 10@xmath85m@xmath2 .
the metallicity of m83 s starburst is @xmath48 with a large error bar ( see @xcite ) .
we considered theoretical models with metallicities of @xmath0 and @xmath1 .
@xcite derived the age and extinction for each cluster by finding the minimum @xmath78 when comparing observed ubvih@xmath8 magnitudes with those predicted by the charlot & bruzual ( 2009 , private communication ) stellar population synthesis models , assuming a milky way - type extinction curve @xcite , metallicities of @xmath86 and @xmath87 , and a chabrier imf from @xmath5m@xmath2 @xcite . here
, the narrow band filter contains both stellar continuum and nebular line emission , i.e. no continuum subtraction was performed for the narrow band image , and the predictions were modeled as line plus continuum .
the mass of each cluster was estimated from its extinction - corrected @xmath42 band luminosity and the age - dependent mass - to - light ratio from the stellar population models , assuming a chabrier stellar imf and a distance of 4.5 mpc .
the ages and masses derived using this technique are compared to the spectroscopic values below .
on average , the theoretical spectra fit the data better for @xmath1 than for @xmath0 ( on average , the reduced @xmath78 is 1.3 times larger at @xmath0 than at @xmath1 , with a standard deviation of 0.3 ) . note that some of the differences between the data and the best - fit models are not due to the metallicity choice , but to the quality of the data , which suffers from crowding , as well as from worse spectral resolution than when a narrow slit is used .
in any case , on average , the spectroscopic ages and masses at the two metallicities are identical .
hereafter , we only consider spectroscopic results at @xmath0 , which is the metallicity of the semi - empirical models .
cluster 14 is the youngest ( @xmath263 myr , average of spectroscopic and photometric ages ) and the least massive ( photometric mass of about @xmath88m@xmath2 ) in our sample . according to @xcite cluster 14
should have a maximum stellar mass of about 80 m@xmath2 ( i.e. below 100 m@xmath2 ) .
unfortunately , the spectrum of this cluster is too noisy to reliably pin down its maximum stellar mass .
cluster 1 is the next best cluster in our sample to test the upper end of the stellar imf , and has a mean age of @xmath264myr and a photometric mass of @xmath263@xmath6m@xmath2 . as shown in fig .
[ fig_1and10 ] , left panel , the strong si iv 1400 and c iv 1550 p - cygni profiles of cluster 1 can not be reproduced with models having maximum stellar masses of 30m@xmath2 or less . models with maximum stellar masses of 50 and 100m@xmath2 fit cluster 1 equally well .
note that @xcite found optical signatures of the presence of wn stars in the region of our clusters @xmath75 .
the p - cygni profile of he ii at 1640 , which is characteristic of wn stars , is barely detected in the fuv spectra of clusters @xmath75 , but given the point that we make in
[ theo_lib ] , this is not surprising .
the rest of clusters with strong p - cygni profiles of n v , si iv and c iv are consistent with having formed stars more massive than 30m@xmath2 . the right panel of fig .
[ fig_1and10 ] shows that the upper mass limit of the imf is harder to constrain for cluster 10 , which is without p - cygni profiles
. clusters with strong absorption profiles of si iv and c iv are consistent with having formed at least some b stars , if we assume that the absorptions are not of interstellar origin .
figures [ fig_fits_1 - 6 ] to [ fig_fits_12 - 14 ] show the background and reddening corrected rest - frame spectra of the clusters in our sample , along with best fit semi - empirical and theoretical models corresponding to a metallicity of @xmath0 and a kroupa imf from @xmath5m@xmath2 .
we used the latter models to determine the spectroscopic ages and masses of all clusters , including those without strong p - cygni profiles of n v , si iv , and c iv in their spectra .
therefore , we are assuming that clusters without strong p - cygni profiles formed massive o stars but are older than @xmath2612 myr .
another possibility for the latter clusters is that they are younger than @xmath2612 myr and did not form very massive o stars . here
, we compare our ages derived using spectroscopic and photometric techniques .
we also compare our ages with estimates from the literature .
our spectroscopic and photometric ages are compiled in columns ( 1)-(9 ) of tab .
[ tab_age ] , which give ( 1 ) the cluster i d ; ( 2 ) and ( 4 ) , the spectroscopic ages from the semi - empirical and theoretical models ; ( 3 ) and ( 5 ) , the reduced @xmath78 value of the best fits to the observed spectra ; ( 6 ) and ( 7 ) , the photometric ages from the @xmath86 and @xmath87 models ; ( 8)-(13 ) ratios of the age estimates derived from the different techniques .
the last four rows of tab .
[ tab_age ] give the minimum , maximum , mean , and standard deviation values for relevant columns .
figure [ fig_photsed ] shows the photometric age determinations for cluster 1 ( mean photometric age @xmath264myr ) and cluster 10 ( mean photometric age @xmath2619myr ) .
they were derived using the technique described in @xcite .
figure [ fig_photsed ] also shows the measured v - b and v - i colors for clusters 1 and 10 .
considering the mean of the spectroscopic and photometric ages , clusters 1 and 10 are representative of young age ( @xmath89myr ) and older age ( @xmath90myr ) clusters , respectively .
the ages estimated from the four different techniques are in excellent agreement .
ages derived by comparing uv spectroscopy with semi - empirical and theoretical predictions are within a factor of 1.5 .
this is similar to the @xmath91 uncertainties of @xmath261myr found previously for young clusters from uv spectroscopy @xcite .
ages determined by comparing the _ hst_/wfc3 optical colors of the clusters with predictions from stellar population synthesis models at the two metallicities agree within a factor of 2 . finally , the spectroscopic and photometric ages are in excellent agreement ( within a factor of 1.2 on average ) .
table [ tab_age ] shows that most clusters are quite young , with ages @xmath92myr .
clusters 6 , 7 , and 10 , which have the oldest spectroscopic ages , also have the oldest photometric ages . the worst agreement between the spectroscopic and photometric ages
is for cluster 8 ( difference of a factor of 4 ) .
the fuv spectrum of this cluster lacks the strong p - cygni profiles found in young star clusters , yet the cluster has very blue colors , resulting in a young ( @xmath264myr ) photometric age . comparing with previous work ,
our ages agree with the photometric ages derived by @xcite based on _
hst_/wfpc2 photometry , except for clusters 6 , 7 , and 10 , which are older than 6myr in our case .
we found 3.6 , 3.4 , and 4.1myr , for the ages of clusters 1 , 2 , and 3 respectively ( means of spectroscopic and photometric ages ) , which is in excellent agreement with the ages found by @xcite for region a ( all @xmath264myr ) , using the same spectroscopic data as here . on the other hand
, @xcite determined an age of 7myr for their region b , from the equivalent width of h@xmath37 .
region b corresponds to the location of our clusters 7 - 9 .
we found mean ages of 15.5myr for cluster 7 and 3.4myr for cluster 9 .
as explained in the previous paragraph , we were unable to pin down the age of cluster 8 , which is also in region b. finally , region d of @xcite corresponds to a cluster that we did not include in our sample , due to its truncated spectral trace .
we conclude that our spectroscopic and photometric age estimates are within the expected uncertainties for 12 out of 13 of the clusters .
the clusters are @xmath263 - 20myr old and were not all formed at the same time .
in addition , we do not find any compelling evidence for an age gradient along m83 s arc - shaped starburst , but rather see clumping of clusters of similar age with very young clusters dominating the northern - most and southern - most portion of the starburst that we studied , and somewhat older clusters found between .
this is in disagreement with @xcite and @xcite , who have suggested that there is an age gradient along the starburst , as described in the introduction . here
, we compare the masses estimated for our clusters using spectroscopic and photometric techniques .
these are compiled in columns ( 1)-(7 ) of tab .
[ tab_mass ] , which give ( 1 ) the cluster i d ; ( 2 ) and ( 3 ) , the masses derived by fitting the semi - empirical and the theoretical models to the fuv spectroscopy ; ( 4 ) and ( 5 ) the masses derived from the optical photometry assuming metallicities of @xmath49 and @xmath86 ; and ( 6)-(11 ) ratios of masses obtained from the different techniques .
column ( 6 ) of tab .
[ tab_mass ] shows that the spectroscopic masses from the semi - empirical models are in very good agreement with those from the theoretical models , and that the semi - empirical masses are systematically larger by a factor of 1.5 on average than the theoretical masses .
the latter is because the semi - empirical models have systematically larger values of the mean ratio described in ( ii ) of the last paragraph of [ ana_spec ] .
column ( 7 ) of tab .
[ tab_mass ] shows that the photometric masses are not significantly affected by metallicity .
finally , columns ( 8) to ( 11 ) show that the spectroscopic masses are systematically larger than the photometric masses by a factor of 4 - 6 on average .
we suggest that this is the result of the high background contamination of our wide slit and slitless crowded spectra .
optical photometry provides more leverage for determining the cluster mass .
this is because photometry captures light from low and intermediate mass stars , which contribute more by mass to the imf than massive stars .
in addition , the total light from the cluster is estimated better from the photometry .
therefore , we consider our photometric masses ( m@xmath93 ) as more reliable than our spectroscopic masses .
our most massive cluster has m@xmath93 between 2 and @xmath94m@xmath2 , comparable to the virial mass of the ionizing cluster of 30 doradus , ngc 2070 ( @xmath95m@xmath2 , @xcite ) , in the large magellanic cloud .
according to @xcite , young star clusters with masses larger than 10@xmath96m@xmath2 can last an age comparable to or exceeding the age of the universe .
therefore , this massive cluster in m83 could be a globular cluster progenitor , while two other clusters , which have m@xmath97m@xmath2 , may also survive .
cluster 14 has m@xmath98m@xmath2 and appears to be the most compact in size in the fuv image .
finally , the rest of the clusters have m@xmath93 of a few@xmath3m@xmath2 . note that our older clusters tend to be more massive than the younger ones .
this is probably an observational effect .
older clusters at lower masses are harder to detect in the fuv .
columns ( 1)-(5 ) of tab .
[ tab_reddening ] give ( 1 ) the cluster s i d , ( 2 ) @xmath99 , the power - law index of the continuum over the wavelength range @xmath77 , corrected for a foreground milky way reddening of @xmath34 ; ( 3 ) the intrinsic reddening of the cluster derived from the fuv spectra ; ( 4 ) the intrinsic reddening derived from the optical photometry ; and ( 5 ) @xmath100 , the intrinsic luminosity of the cluster at 1500 , derived by adopting a distance of 4.5 mpc to m83 .
overall , the agreement between the reddening values derived from the spectroscopic and the photometric analysis is fair . for clusters 3 and 14 , which have dusty local environments ( see fig .
[ fig_composite ] ) , the agreement is good .
indeed , both clusters have large intrinsic spectroscopic and photometric reddenings .
the agreement is also good for clusters 11 and 12 , which appear to be the least reddened by both measures of the intrinsic extinction .
however , for cluster 1 , e(b - v)=0.18 from the spectroscopy , while e(b - v)@xmath1010.06 from the photometry .
cluster 3 has the largest intrinsic value of @xmath100 , while clusters 4 , 13 , and 14 , which appear faint in the optical images , have the lowest values of @xmath100 .
in addition to compact star clusters , there is uv emission from the diffuse field star population .
we obtained the spectrum representing the diffuse stellar field by summing spectra extracted from the long slit exposure , at the locations indicated by the dashed lines in fig .
[ fig_fuvmatch ] .
we performed no background or reddening correction for the field spectrum .
in addition , we did not shift the spectral traces in the dispersion direction , due to the lack of narrow interstellar lines that could be used for reference .
figure [ fig_fieldsp ] shows the result .
the spectrum lacks the p - cygni profiles of n v , si iv , and c iv , which are signatures of o stars , but it shows strong absorption components in these features , which indicates the presence of b stars .
this is consistent with a picture where ( massive ) o stars form mostly in clusters but not in the field , and where the field is dominated by b stars from clusters which disrupt rapidly , on timescales of @xmath2610myr @xcite .
we estimated the ages , reddenings , and masses of 13 clusters within the circumnuclear starburst of m83 , the nearest such example , based on _
hst_/stis / fuv - mama imaging and spectroscopy , and _
hst_/wfc3/uvis photometry .
we derived four ages for each cluster .
the first two ages were obtained by fitting the uv spectra with semi - empirical @xcite and theoretical @xcite models corresponding to a metallicity of @xmath0 , giving the most weight to the spectral features n v 1240 , si iv 1400 , and c iv 1500 , which are expected to be prominent in the observed spectra of clusters with ages of less than 20myr .
note that theoretical models with @xmath1 yield , on average , ages and masses identical to those at @xmath0 .
we compared our spectroscopic ages with the photometric ages derived by @xcite using charlot and bruzual stellar population synthesis models corresponding to metallicities of @xmath86 and @xmath87 .
the spectroscopic ages derived with the semi - empirical and the theoretical predictions are within a factor of 1.2 on average .
the spectroscopic and photometric ages agree at a similar level .
our ages agree with those derived from _
hst_/wfpc2 photometry by @xcite , except for clusters 6 , 7 , and 10 , which are older than 6myr in our case .
our ages for clusters 1 - 3 agree with the ages of region a derived from stis fuv spectroscopy by @xcite .
the clusters are @xmath263 - 20myr old and were not all formed at the same time .
we found no age gradient along m83 s starburst , in disagreement with @xcite and @xcite .
our clusters with strong p - cygni profiles have photometric masses of at least a few@xmath3m@xmath2 , seem to have formed stars more massive than @xmath4m@xmath2 , and are consistent with a kroupa imf from @xmath102m@xmath2 .
the rest of the clusters are consistent with having formed at least some b stars .
b stars dominate the field population .
this work was supported by nasa grant n1317 .
r. c. is grateful for support from nsf through career award 0847467 .
we would like to thank william p. blair for producing fig .
[ fig_composite ] .
in addition , we would like to thank brad whitmore and our referee for comments that greatly improved the quality of this paper . , l. , jogee , s. , barazza , f. d. , marinova , i. , & shen , j. 2009 , in astronomical society of the pacific conference series , vol .
419 , astronomical society of the pacific conference series , ed .
s. jogee , i. marinova , l. hao , & g. a. blanc , 402+ , d. 1986 , in presented at the society of photo - optical instrumentation engineers ( spie ) conference , vol .
627 , society of photo - optical instrumentation engineers ( spie ) conference series , ed .
d. l. crawford , 733+ , t. , lundgren - andersson , a. , & olofsson , h. 2004 , in iau symposium , vol .
222 , the interplay among black holes , stars and ism in galactic nuclei , ed .
t. storchi - bergmann , l. c. ho , & h. r. schmitt , 219220 lccccccc slitless spectra & 1998 - 05 - 02 & 42 & -24.9 & f25sfr2 & 1275 - 1700 & sfl + slitless spectra & 1998 - 05 - 03 & 45 & -24.9 & f25sfr2 & 1275 - 1700 & sfl + slitless spectra & 1998 - 05 - 03 & 39 & -24.9 & 25mama & 1150 - 1700 & sfl + direct image & 1999 - 05 - 19 & 23 & 32.9 & f25qtz & 1475 - 1700 & x2d + long slit spectra & 2000 - 04 - 23 & 42 & -9.5 & 25mama & 1150 - 1700 & flt [ tab_data ] lccccccccccc 1 & 13:37:00.49 & -29:51:54.93 & 18 & 17 & 27 & 6.0e-15 & 8.9e-15 & 8.0e-15 & -1.3 ( -1.1 ) & -0.7 & -0.5 + 2 & 13:37:00.45 & -29:51:55.40 & 16 & 14 & 24 & 4.8e-15 & 6.8e-15 & 6.4e-15 & -1.5 ( -1.2 ) & -0.9 & -0.9 + 3 & 13:37:00.54 & -29:51:55.61 & 14 & 10 & 19 & 3.8e-15 & 4.4e-15 & 4.4e-15 & -0.6 ( -0.4 ) & 0.2 & 0.8 + 4 & 13:37:00.42 & -29:51:57.52 & 6 & & & 9.2e-16 & & & -0.6 ( -0.7 ) & & + 5 & 13:37:00.34 & -29:51:58.31 & & & & & & & & & + 6 & 13:37:00.37 & -29:51:59.48 & & 16 & 23 & & 5.3e-15 & 5.4e-15 & & -1.4 & -1.5 + 7 & 13:37:00.42 & -29:51:59.83 & & 24 & 35 & & 1.1e-14 & 1.1e-14 & & -1.4 & -1.5 + 8 & 13:37:00.44 & -29:51:59.99 & & 24 & 36 & & 1.2e-14 & 1.2e-14 & & -1.6 & -1.6 + 9 & 13:37:00.43 & -29:52:00.20 & & 22 & 33 & & 9.5e-15 & 1.0e-14 & & -1.6 & -1.7 + 10 & 13:37:01.00 & -29:52:00.71 & 21 & 22 & 34 & 7.7e-15 & 1.1e-14 & 1.2e-14 & -1.5 ( -1.3 ) & -1.4 & -1.3 + 11 & 13:37:01.00 & -29:52:01.45 & 34 & 35 & 50 & 2.1e-14 & 2.4e-14 & 2.6e-14 & -2.2 ( -1.9 ) & -2.1 & -2.2 + 12 & 13:37:00.54 & -29:52:01.62 & 34 & 35 & 50 & 2.1e-14 & 2.5e-14 & 2.6e-14 & -2.3 ( -2.0 ) & -2.1 & -2.2 + 13 & 13:37:00.56 & -29:52:01.91 & 22 & 23 & 35 & 9.0e-15 & 1.2e-14 & 1.2e-14 & -2.3 ( -2.0 ) & -2.0 & -2.1 + 14 & 13:37:00.53 & -29:52:02.62 & 7 & & & 1.5e-15 & & & -0.7 ( -0.7 ) & & [ tab_observed ] lcccccccccccc 1 & 3.9 & 3.3 & 2.6 & 5.4 & 4.2 & 3.8 & 1.5 & 1.1 & 0.9 & 0.6 & 1.0 & 0.7 + 2 & 3.4 & 4.2 & 2.4 & 7.4 & 4.2 & 3.8 & 1.4 & 1.1 & 0.8 & 0.6 & 0.9 & 0.6 + 3 & 3.5 & 4.8 & 3.5 & 4.0 & 4.0 & 5.5 & 1.0 & 0.7 & 0.9 & 0.9 & 0.6 & 0.6 + 4 & 3.5 & 1.5 & 2.7 & 1.9 & 4.2 & 5.0 & 1.3 & 0.8 & 0.8 & 0.7 & 0.7 & 0.5 + 6 & 19.8 & 1.9 & 16.8 & 1.7 & 10.0 & 24.0 & 1.2 & 0.4 & 2.0 & 1.7 & 0.8 & 0.7 + 7 & 14.2 & 2.3 & 14.3 & 2.9 & 9.5 & 24.0 & 1.0 & 0.4 & 1.5 & 1.5 & 0.6 & 0.6 + 8 & 15.6 & 2.4 & 14.3 & 2.4 & 4.0 & 3.6 & 1.1 & 1.1 & 3.9 & 3.6 & 4.3 & 3.9 + 9 & 4.5 & 5.3 & 3.7 & 2.3 & 3.3 & 2.2 & 1.2 & 1.5 & 1.4 & 1.1 & 2.1 & 1.7 + 10 & 11.8 & 2.9 & 14.1 & 4.7 & 13.2 & 24.0 & 0.8 & 0.5 & 0.9 & 1.1 & 0.5 & 0.6 + 11 & 3.4 & 6.2 & 2.6 & 8.8 & 6.3 & 4.6 & 1.3 & 1.4 & 0.5 & 0.4 & 0.7 & 0.6 + 12 & 3.4 & 7.1 & 2.6 & 10.4 & 6.3 & 4.4 & 1.3 & 1.4 & 0.5 & 0.4 & 0.8 & 0.6 + 13 & 4.2 & 2.6 & 2.8 & 4.9 & 9.1 & 5.0 & 1.5 & 1.8 & 0.5 & 0.3 & 0.8 & 0.6 + 14 & 2.2 & 2.4 & 2.6 & 4.1 & 3.3 & 2.2 & 0.8 & 1.5 & 0.7 & 0.8 & 1.0 & 1.2 + min & 2.2 & 1.5 & 2.4 & 1.7 & 3.3 & 2.2 & 0.8 & 0.4 & 0.5 & 0.3 & 0.5 & 0.5 + max & 19.8 & 7.1 & 16.8 & 10.4 & 13.2 & 24.0 & 1.5 & 1.8 & 3.9 & 3.6 & 4.3 & 3.9 + mean & 7.2 & 3.6 & 6.5 & 4.7 & 6.3 & 8.6 & 1.2 & 1.1 & 1.2 & 1.0 & 1.1 & 1.0 + stdev & 5.9 & 1.8 & 5.8 & 2.7 & 3.2 & 8.8 & 0.2 & 0.5 & 0.9 & 0.9 & 1.0 & 0.9 [ tab_age ] lcccccccccc 1 & 1.1e+05 & 6.9e+04 & 2.5e+04 & 2.7e+04 & 1.5 & 0.9 & 4.3 & 2.8 & 4.0 & 2.6 + 2 & 6.0e+04 & 5.2e+04 & 1.9e+04 & 2.1e+04 & 1.2 & 0.9 & 3.1 & 2.7 & 2.9 & 2.5 + 3 & 1.4e+05 & 1.2e+05 & 4.4e+04 & 4.7e+04 & 1.2 & 0.9 & 3.2 & 2.8 & 3.0 & 2.6 + 4 & 3.1e+04 & 2.3e+04 & 1.7e+04 & 1.8e+04 & 1.3 & 0.9 & 1.9 & 1.4 & 1.7 & 1.3 + 6 & 7.4e+05 & 3.8e+05 & 6.4e+04 & 9.9e+04 & 2.0 & 0.6 & 11.6 & 5.9 & 7.5 & 3.8 + 7 & 1.0e+06 & 6.5e+05 & 1.9e+05 & 3.1e+05 & 1.6 & 0.6 & 5.5 & 3.4 & 3.4 & 2.1 + 8 & 9.3e+05 & 5.2e+05 & 4.1e+04 & 3.6e+04 & 1.8 & 1.1 & 22.8 & 12.7 & 25.6 & 14.2 + 9 & 1.2e+05 & 7.7e+04 & 1.8e+04 & 2.0e+04 & 1.5 & 0.9 & 6.4 & 4.2 & 6.0 & 3.9 + 10 & 5.7e+05 & 4.5e+05 & 6.8e+04 & 1.3e+05 & 1.3 & 0.5 & 8.3 & 6.6 & 4.4 & 3.5 + 11 & 9.1e+04 & 7.3e+04 & 5.0e+04 & 5.8e+04 & 1.3 & 0.9 & 1.8 & 1.4 & 1.6 & 1.2 + 12 & 7.9e+04 & 6.3e+04 & 5.5e+04 & 4.8e+04 & 1.3 & 1.1 & 1.4 & 1.1 & 1.6 & 1.3 + 13 & 4.3e+04 & 2.3e+04 & 2.6e+04 & 1.6e+04 & 1.8 & 1.6 & 1.7 & 0.9 & 2.7 & 1.4 + 14 & 3.0e+04 & 2.8e+04 & 3.9e+03 & 4.2e+03 & 1.1 & 0.9 & 7.6 & 7.2 & 7.1 & 6.7 + min & 3.1e+04 & 2.3e+04 & 1.7e+04 & 1.6e+04 & 1.2 & 0.5 & 1.4 & 0.9 & 1.6 & 1.2 + max & 1.0e+06 & 6.5e+05 & 1.9e+05 & 3.1e+05 & 2.0 & 1.6 & 22.8 & 12.7 & 25.6 & 14.2 + mean & 3.3e+05 & 2.1e+05 & 5.1e+04 & 6.9e+04 & 1.5 & 0.9 & 6.1 & 4.1 & 5.5 & 3.6 + stdev & 3.8e+05 & 2.2e+05 & 4.7e+04 & 8.3e+04 & 0.3 & 0.3 & 5.9 & 3.3 & 6.4 & 3.5 + [ tab_mass ] lccccc 1 & -1.7 & 0.18 & 0.02 & 0.06 & 1.4e+38 + 2 & -1.9 & 0.16 & 0.06 & 0.10 & 9.4e+37 + 3 & -0.9 & 0.28 & 0.28 & 0.20 & 2.0e+38 + 4 & -1.0 & 0.23 & 0.24 & 0.20 & 5.0e+37 + 6 & -1.8 & 0.13 & 0.20 & 0.04 & 8.8e+37 + 7 & -1.7 & 0.13 & 0.22 & 0.06 & 1.9e+38 + 8 & -1.9 & 0.10 & 0.16 & 0.16 & 1.5e+38 + 9 & -1.9 & 0.10 & 0.14 & 0.18 & 1.3e+38 + 10 & -1.9 & 0.14 & 0.02 & 0.00 & 1.3e+38 + 11 & -2.5 & 0.05 & 0.00 & 0.08 & 1.5e+38 + 12 & -2.6 & 0.04 & 0.00 & 0.00 & 1.3e+38 + 13 & -2.6 & 0.03 & 0.00 & 0.02 & 4.8e+37 + 14 & -1.1 & 0.23 & 0.18 & 0.22 & 5.7e+37 [ tab_reddening ] | we analyze archival _ hst_/stis / fuv - mama imaging and spectroscopy of 13 compact star clusters within the circumnuclear starburst region of m83 , the closest such example .
we compare the observed spectra with semi - empirical models , which are based on an empirical library of galactic o and b stars observed with _ iue _ , and with theoretical models , which are based on a new theoretical uv library of hot massive stars computed with wm - basic .
the models were generated with starburst99 for metallicities of @xmath0 and @xmath1 , and for stellar imfs with upper mass limits of 10 , 30 , 50 , and 100m@xmath2 .
we estimate the ages and masses of the clusters from the best fit model spectra , and find that the ages derived from the semi - empirical and theoretical models agree within a factor of 1.2 on average .
a comparison of the spectroscopic age estimates with values derived from _
multi - band photometry shows a similar level of agreement for all but one cluster .
the clusters have a range of ages from about 3 to 20myr , and do not appear to have an age gradient along m83 s starburst . clusters with strong p - cygni profiles have masses of a few@xmath3m@xmath2 , seem to have formed stars more massive than @xmath4m@xmath2 , and are consistent with a kroupa imf from @xmath5m@xmath2 .
field regions in the starburst lack p - cygni profiles and are dominated by b stars . |
practical solvability of performance analysis and control design problems for stochastic systems often depends on tractability of relevant quantities , such as moments of the system variables . for example , the kalman filtering and linear quadratic gaussian control theories @xcite substantially employ the preservation of gaussian nature of probability distributions of the state variables governed by linear sdes . under the linear dynamics , the first and second order moments of the variables ( and more complicated functionals of gaussian distributions )
are amenable to a complete analysis .
it is the convenience of linear gaussian models that makes them so popular in filtering / control with quadratic and related ( for example , risk - sensitive ) performance criteria .
these advantages motivate the approximation of a nonlinear stochastic system by an effective linear model which underlies the stochastic linearization techniques .
the latter date back to @xcite and have recently been extended to quantum stochastic systems @xcite .
a different approach to computing the statistical characteristics of a nonlinear stochastic system ( oriented at approximating probability distributions rather than system dynamics ) consists , for example , in representing the probability distribution of its state variables as a mixture of gaussian distributions whose parameters evolve in time .
in fact , mixed gaussian distributions arise as exact posterior probability distributions in the lainiotis multimodel filter @xcite , where the conditional gaussian distributions from partial kalman filters are weighted by recursively updated posterior probabilities of the corresponding linear models conditioned on the observations .
this combination of a bank of kalman filters with a `` mixing '' block is a recursive implementation of the bayesian approach .
the important property that the resulting mixture of gaussian pdfs is a legitimate pdf , which is nonnegative everywhere and satisfies the normalization condition , does not always come with pdf approximations , in general .
for example , the truncated gram - charlier and edgeworth expansions @xcite , based on the hermite polynomials , are not equipped with this feature , although they provide control over moments or cumulants up to an arbitrary given order .
the aim of the present paper is to outline an approach to the approximation of pdfs by using quadratic forms of weighted complex - valued orthonormal basis functions with positive semi - definite hermitian matrices of unit trace .
these matrices are called stochastic density matrices ( sdm ) in order to emphasize an analogy ( and , at the same time , avoid confusion ) with the quantum mechanical density matrices @xcite .
the sdm approximation leads to a legitimate pdf which is nonnegative everywhere and satisfies the normalization condition .
furthermore , it retains the possibility to control the moments of the pdf for orthonormal bases with an algebraic structure , such as the hermite polynomial and fourier bases .
the sdm approximation can be optimized by using a proximity criterion for pdfs based on the second - order relative renyi entropy @xcite , which leads to a quadratic minimization problem .
this allows the sdm approach to be applied to pdfs of markov diffusion processes , governed by nonlinear sdes , by reducing the approximate numerical integration of the fokker - planck - kolmogorov equation ( fpke ) @xcite to the solution of an ode for the sdm , which resembles the galerkin approximations for parabolic pdes @xcite . as an illustration
, we consider a smoluchowski sde @xcite on a multidimensional torus , which provides an example of a nonlinear stochastic system with rotational degrees of freedom .
the sdm approach admits a real version in the case of real - valued basis functions , with an appropriate reformulation of the results .
it is relevant to mention a connection of this approach with the methods using the sum of squares ( sos ) of polynomials for lyapunov stability analysis and global optimization @xcite .
however , the sdm approach , proposed in the present paper , serves a different purpose here and is not restricted to polynomials .
the paper is organised as follows .
section [ sec : sdm ] describes the class of pdfs generated by an sdm and a set of orthonormal functions .
section [ sec : alg ] relates the algebraic structure of the orthonormal basis to the moments of such pdfs .
section [ sec : eff_par ] discusses effective parameters of the sdm which control the pdf .
section [ sec : herm ] specifies this class of pdfs for the multivariate hermite polynomial and fourier bases .
section [ sec : sdmapp ] describes the sdm approximation of a given pdf using a quadratic criterion .
section [ sec : sdmdyn ] extends the sdm approximation to pdf dynamics of markov processes .
section [ sec : ssde ] considers the smoluchowski sde on a multidimensional torus .
section [ sec : spat ] reformulates the corresponding fpke in the spatial frequency domain .
section [ sec : smolsdm ] describes the sdm approximation of the pdf dynamics using the fourier basis and provides numerical results .
section [ sec : conc ] makes concluding remarks .
suppose @xmath0 is a finite - dimensional state space of a dynamical system . to be specific , we assume that @xmath0 is a domain in @xmath1 , or an @xmath2-dimensional torus @xmath3 . in
what follows , we use a complex hilbert space @xmath4 of square integrable complex - valued functions on the set @xmath0 with a weight @xmath5 .
the norm @xmath6 in the space @xmath7 is generated by the inner product @xmath8 where the integral is over the standard @xmath2-variate lebesgue measure , and @xmath9 is the complex conjugate .
furthermore , we denote by @xmath10 the frobenius inner product @xcite of complex matrices @xmath11 and @xmath12 , which reduces to @xmath13 for complex hermitian matrices ( the real hilbert space of such matrices of order @xmath14 is denoted by @xmath15 , with @xmath16 the corresponding set of positive semi - definite matrices ) . also , @xmath17 is the complex conjugate transpose , and vectors are organised as columns unless specified otherwise .
suppose @xmath18 is a fixed but otherwise arbitrary orthonormal basis in the hilbert space ( [ ch ] ) , whose elements are indexed by a denumerable set @xmath19 .
the latter is assumed to be a subset of a multidimensional integer lattice which contains the origin : @xmath20 .
therefore , @xmath21 where @xmath22 is the kronecker delta . now , suppose @xmath23 is a given nonempty finite subset of @xmath19 consisting of @xmath24 elements . also , let @xmath25 be a matrix of unit trace @xmath26 .
such matrices form a convex subset of the space @xmath27 which we denote by @xmath28 with the orthonormal basis @xmath29 in @xmath7 and the set @xmath23 being fixed , we associate with @xmath30 a real - valued function @xmath31 defined as a quadratic form @xmath32 where @xmath33 is a vector - valued map formed from the basis functions as @xmath34 [ lem : ps ] for any matrix @xmath35 from the set ( [ fs ] ) , the corresponding function @xmath36 , defined by ( [ p ] ) and ( [ phi ] ) , is a pdf on the set @xmath0 .
the positive semi - definiteness @xmath37 and the positiveness @xmath38 in ( [ p ] ) imply that @xmath39 for all @xmath40 .
furthermore , the orthonormality ( [ ortho ] ) and the assumption @xmath41 imply that the function @xmath36 satisfies the normalization condition @xmath42 , and hence , @xmath36 is indeed a pdf . due to its properties and the role in the construction ( [ p ] ) of a legitimate pdf
, the matrix @xmath43 resembles the density matrices in quantum mechanics @xcite . in order to reflect this analogy ( and , at the same time ,
avoid confusion ) with the quantum mechanical density matrices and the stochastic matrices of transition probabilities of classical markov chains @xcite , we will refer to @xmath43 as a _ stochastic density matrix _ ( sdm ) . since the set @xmath44 of sdms @xmath43 in ( [ fs ] ) is convex and the pdf @xmath36 in ( [ p ] ) depends linearly on @xmath43 , the set of such pdfs is also convex .
if the hilbert space in ( [ ch ] ) is restricted to real - valued functions , then the sdm becomes a real symmetric matrix , and the results that follow can be appropriately reformulated for this case .
for what follows , we assume that @xmath45 , and hence , in view of ( [ ortho ] ) , the weighting function @xmath46 satisfies @xmath47 and is itself a pdf .
this allows the inner product ( [ braket ] ) to be represented as @xmath48 , where @xmath49 denotes the expectation over the pdf @xmath46 .
also , suppose the basis functions are algebraically closed in the sense that there exist complex constants @xmath50 such that for any @xmath51 , the representation @xmath52 holds for all @xmath40 and contains only a finite number of terms with nonzero coefficients @xmath50 .
the coefficients @xmath50 in ( [ tri ] ) , which quantify the algebraic structure of the orthonormal basis @xmath29 , can be computed as @xmath53 and are symmetric with respect to the subscripts @xmath54 , @xmath55 .
furthermore , @xmath50 is real and invariant under arbitrary permutations of its subscripts in the real case mentioned at the end of section [ sec : sdm ] . for any @xmath56
, we define the @xmath55th _ structure matrix _ @xmath57 by @xmath58 where the expectation applies entrywise , and the map @xmath59 is given by ( [ phi ] ) .
in particular , since @xmath45 , then ( [ e ] ) implies that @xmath60 for all @xmath61 , and hence , @xmath62 is the identity matrix of order @xmath63 .
since the left - hand side of ( [ tri ] ) is the @xmath64th entry of the matrix @xmath65 , the algebraic closedness property is representable in a vector - matrix form : @xmath66 this series contains only a finite number of terms since the matrix @xmath67 in ( [ e ] ) vanishes for all but finitely many indices @xmath56 which form the set @xmath68 this set , consisting of @xmath69 elements , depends on the set @xmath23 and contains @xmath70 ( since @xmath62 ) .
substitution of ( [ phiphiphi ] ) into the definition of the pdf @xmath36 in ( [ p ] ) leads to @xmath71 here , @xmath72 is a linear vector - valued map defined in terms of the nonzero structure matrices by @xmath73 and , similarly to ( [ phi ] ) , the map @xmath74 is formed from the basis functions as @xmath75 the second equality in ( [ pes ] ) follows from @xmath36 and @xmath46 being real - valued . in view of this equality and ( [ ortho ] ) , the expectations of the basis functions over the pdf @xmath36 ( which play the role of generalized moments of this pdf ) are computed as @xmath76 for any @xmath77 and vanish for all @xmath78 in view of ( [ mho ] ) . therefore , ( [ pes ] ) describes the series expansion of the pdf @xmath36 from ( [ p ] ) over the orthonormal basis in @xmath7 , and the map @xmath79 in ( [ c ] ) encodes the dependence of all the nonzero moments in ( [ mom ] ) on the sdm @xmath43 . the fact that the moments depend linearly on @xmath43 allows the sdm to be chosen so as to satisfy given moment specifications for the corresponding pdf @xmath36 in ( [ p ] ) .
note that the moments @xmath80 , with @xmath81 , are responsible for the deviation of the pdf @xmath36 from the weighting function @xmath46 .
in particular , ( [ pes ] ) implies that the second - order relative renyi entropy @xcite of @xmath36 with respect to @xmath46 ( with the latter playing the role of a reference pdf ) can be expressed in terms of the moments in ( [ mom ] ) as @xmath82 here , the orthonormality condition ( [ ortho ] ) is used in combination with the property that @xmath83 .
note that the pdf @xmath36 , defined by ( [ p ] ) , can also be represented in the form @xmath84 which resembles the sos structures @xcite .
here , @xmath85 denote the eigenvalues of the sdm @xmath43 ( they are all real and nonnegative and satisfy @xmath86 ) , and the functions @xmath87 are obtained by a unitary transformation of the functions @xmath88 as @xmath89 also , @xmath90 is a unitary matrix whose columns are the eigenvectors of @xmath43 , and hence , @xmath91 , where @xmath92 is a diagonal matrix formed from the eigenvalues of @xmath43 .
the functions @xmath93 , defined by ( [ theta ] ) , are also orthonormal elements of the hilbert space @xmath7 in ( [ ch ] ) , that is , @xmath94 .
the representation ( [ psi2 ] ) of the pdf @xmath36 as a convex combination of the squares of the functions @xmath95 involves not only the freedom of varying the coefficients @xmath85 , but also the possibility of `` mixing '' the @xmath63 elements @xmath88 of the given orthonormal basis of @xmath7 in an arbitrary unitary fashion ( [ theta ] ) . since the sdm @xmath43 is hermitian and has unit trace , then the total number of real parameters which enter the pdf @xmath36 through @xmath43 is @xmath96 .
however , not all of these parameters are active , in general . from the decomposition @xmath97 of the matrix @xmath43 into the orthogonal projections @xmath98 and @xmath99 onto the linear subspace @xmath100 spanned by the nonzero structure matrices in ( [ e ] ) , and its orthogonal complement @xmath101 , respectively , it follows that @xmath102 does not influence the pdf @xmath36 in ( [ pes ] ) . here
, @xmath103 is a finite subset of @xmath19 given by ( [ mho ] ) .
hence , the number of effective parameters of @xmath36 does not exceed the dimension @xmath104 of the subspace ( [ ce ] ) , that is , the rank of an appropriate gram matrix : @xmath105 the following lemma provides a condition for the nonsingularity of the gram matrix , that is , the linear independence of the structure matrices @xmath106 , with @xmath107 , in the hilbert space @xmath108 . to this end , for any finite subset @xmath109 , we denote by @xmath110 the subspace of @xmath7 spanned by the corresponding set of basis functions .
also , let @xmath111 be a set of functions on @xmath0 associated with the subspace @xmath112 as @xmath113 from the algebraic closedness ( [ tri ] ) of the basis functions , it follows that @xmath114 indeed , in view of the notation ( [ chm ] ) , any two functions @xmath115 are representable as @xmath116 and @xmath117 for some vectors
@xmath118 , where @xmath59 is given by ( [ phi ] ) .
therefore , ( [ phiphiphi ] ) implies that @xmath119 , whence the inclusion ( [ hhh ] ) follows in view of the arbitrariness of the functions @xmath115 .
[ lem : hh ] the linear subspace @xmath120 , defined by ( [ ce ] ) , has full dimension @xmath121 if and only if the orthogonal complement of the set @xmath111 , given by ( [ hh ] ) , to the subspace @xmath122 consists of the zero function : @xmath123 as mentioned above in regard to ( [ dim ] ) , the property @xmath121 is equivalent to the linear independence of the structure matrices @xmath106 , with @xmath107 , in the space @xmath108 .
now , a linear combination @xmath124 of these matrices with complex coefficients @xmath125 vanishes if and only if so does the quantity @xmath126 for all @xmath127
. here , @xmath128 , @xmath129 , and @xmath130 are auxiliary functions .
hence , the matrices @xmath67 , with @xmath107 , are linearly dependent if and only if there exists a function @xmath131 such that @xmath132 for all @xmath133 , that is , @xmath134 .
this establishes the equivalence between the linear independence of these matrices and the condition ( [ qh ] ) .
for completeness , we will now specify the pdfs ( [ p ] ) for two classical hilbert spaces of functions . assuming that @xmath135 , consider the functions @xmath136 which are labeled by @xmath2-indices @xmath137 ( with @xmath138 the set of nonnegative integers ) and are obtained from the @xmath2-variate hermite polynomials @xmath139 here
, use is made of the standard multiindex notation @xmath140 , @xmath141 , and @xmath142 , so that @xmath143 is the tensor product of the corresponding univariate hermite polynomials .
the functions ( [ h ] ) form an orthonormal basis in the real hilbert space @xmath144 of real - valued square integrable functions , where the weight @xmath46 is the @xmath2-variate standard normal pdf @xmath145 the orthonormal basis @xmath146 , given by ( [ h ] ) and ( [ h ] ) , satisfies a real - valued version of ( [ tri ] ) . in view of (
* eq . ( 3.13 ) on p. 28 ) and ( * ? ? ?
* theorem 3.2.1 on p. 13 ) , the structure coefficients ( [ e ] ) are computed as @xmath147 for all @xmath2-indices @xmath148 whose entries do not exceed the corresponding entries of @xmath149 and such that @xmath150 consists of even integers , with @xmath151 otherwise . in this case , the pdf @xmath36 in ( [ p ] ) is specified by a real symmetric sdm @xmath43 , and its representation ( [ pes ] ) in terms of the structure matrices from ( [ e ] ) takes the form @xmath152 in view of ( [ nun ] ) . here
, @xmath23 is a finite subset of @xmath153 , and , in view of ( [ eh ] ) , the corresponding set @xmath103 in ( [ mho ] ) is the minkowski sum of the set @xmath23 with itself : @xmath154 .
although ( [ pherm ] ) is organised as a truncated hermite polynomial expansion , its coefficients @xmath155 are parameterized by the sdm @xmath43 in a specific way , which , according to lemma [ lem : ps ] , guarantees that @xmath36 is a legitimate pdf . as another example , consider the fourier basis on the @xmath2-dimensional torus @xmath3 ( where @xmath156 is realised as a half - open interval @xmath157 ) consisting of the functions @xmath158 these functions are indexed using the @xmath2-dimensional integer lattice @xmath159 and form an orthonormal basis in the complex hilbert space @xmath160 with a constant weight @xmath161 .
the latter is the pdf of the uniform distribution over the torus .
since the functions ( [ fourier ] ) satisfy @xmath162 , the algebraic structure coefficients in ( [ tri ] ) are given by @xmath163 in this case , in view of ( [ pes ] ) , the pdf @xmath36 in ( [ p ] ) takes the form of a trigonometric polynomial @xmath164 where the set @xmath165 ( which is the minkowski difference of the set @xmath23 with itself ) is symmetric about the origin in @xmath159 .
again , lemma [ lem : ps ] ensures that ( [ pfourier ] ) describes a legitimate pdf on the torus for any sdm @xmath43 . in the multivariate case being considered , the trigonometric polynomial @xmath36 in ( [ pfourier ] ) is not necessarily reducible to the squared absolute value of a single trigonometric polynomial .
the fejer - riesz theorem @xcite guarantees such a spectral factorization of a nonnegative trigonometric polynomial only in the univariate case . from the representation ( [ psi2 ] ) , with @xmath166
, it follows that ( [ pfourier ] ) is a mixture of @xmath63 such factorizations : @xmath167 where , in accordance with ( [ theta ] ) , @xmath168 are trigonometric polynomials .
the pdf @xmath36 in ( [ psi2fourier ] ) reduces to a single fejer - riesz spectral factorization in the case of a rank - one sdm @xmath169 , where @xmath170 is a unit complex vector .
recalling the analogy with the quantum mechanical density matrices mentioned above , such sdms correspond to pure quantum states @xcite which are extreme points of the convex set of mixed states .
therefore , this multivariate fourier basis example shows that the sdm approach is , in principle , able to produce a wider class of pdfs than those obtained by squaring a single linear combination of elementary functions .
we will now consider a problem of approximating a given pdf @xmath171 by the pdf @xmath36 from ( [ p ] ) .
more precisely , assuming that the orthonormal basis in @xmath7 and the set @xmath23 are fixed , the sdm @xmath43 is varied over the set ( [ fs ] ) so as to minimize a discrepancy between the actual pdf @xmath172 and the approximating pdf @xmath173 ( which is parameterized by @xmath43 ) : @xmath174 for example , one of such proximity criteria is described by a quadratic functional @xmath175 where use is made of the renyi entropy from ( [ br ] ) , provided @xmath176 and @xmath177 are finite , and the @xmath178 factor is introduced for further convenience . a similar , though different , ``
mean integrated squared error '' criterion is employed in the kernel density estimation ; see , for example , @xcite and the references therein .
in view of the linear dependence of @xmath179 on @xmath43 , the quantity @xmath180 , given by ( [ d ] ) , is a convex ( but not necessarily strictly convex ) quadratic function of @xmath43 , which makes ( [ dmin ] ) a convex quadratic minimization problem over the convex set of sdms @xmath44
. we will therefore consider a `` regularised '' version of this problem : @xmath181 where @xmath182 is an additional parameter which weights the strictly convex function @xmath183 , with the latter also playing the role of a barrier function for the sdm @xmath43 to avoid singularity and hence , to remain positive definite .
the resulting strictly convex minimization problem ( [ dminmu ] ) has a unique solution which is described below .
the following theorem employs a positive semi - definite self - adjoint operator @xmath184 , acting on the hilbert space @xmath27 and completely specified by the structure matrices ( [ e ] ) as @xmath185 we will also need a linear operator @xmath186 which maps a pdf @xmath187 to a complex hermitian matrix @xmath188 with the latter being related to the generalized moments of @xmath172 : @xmath189 the finiteness of these moments is guaranteed by the assumption @xmath190 in view of the orthonormality ( [ ortho ] ) and the cauchy - bunyakovsky - schwarz inequality . [
th : sopt ] suppose @xmath187 is a given pdf with finite renyi entropy in ( [ br ] ) : @xmath191
. then for any given value of the barrier parameter @xmath182 , the optimization problem ( [ dminmu ] ) , with the proximity criterion ( [ d ] ) , has a unique solution @xmath192 which satisfies @xmath193 here , @xmath194 is found from the normalization condition @xmath41 , and the operators @xmath184 and @xmath186 are defined by ( [ ca])([bef ] ) .
application of the method of lagrange multipliers to the strictly convex minimization problem ( [ dminmu ] ) yields the following condition of optimality : @xmath195 here , @xmath194 is the lagrange multiplier associated with the trace constraint in ( [ fs ] ) , and @xmath196 is the frechet derivative of @xmath180 in ( [ d ] ) as a composite function of the sdm @xmath43 , where use is made of the pointwise frechet derivative @xmath197 of the pdf @xmath179 in ( [ p ] ) . by combining ( [ phiphiphi ] ) with the moments of the pdf @xmath179 in ( [ mom ] ) , it follows that @xmath198 here , the operator @xmath184 , defined by ( [ ca ] ) , is the second - order frechet derivative @xmath199 , and use is made of ( [ cb ] ) , ( [ bef ] ) .
substitution of ( [ dd])([fmom ] ) into ( [ dl ] ) leads to ( [ sopt ] ) .
the uniqueness of the pair @xmath200 , satisfying ( [ sopt ] ) together with @xmath192 and @xmath41 , is ensured by strict monotonicity of @xmath201 and @xmath202 with respect to @xmath43 in the sense of the partial ordering on @xmath27 induced by positive semi - definiteness .
indeed , ( [ pmom ] ) implies that the operator @xmath184 is nondecreasing since @xmath203 for any @xmath204 , while the map @xmath205 is strictly increasing @xcite .
the barrier parameter @xmath182 in ( [ dminmu ] ) controls sensitivity of the optimal sdm @xmath43 to the pdf @xmath172 being approximated . in particular , @xmath206 . at the other extreme , for small values of @xmath207 ,
the sdm @xmath43 can become nearly singular and highly sensitive to @xmath172 .
the pdf @xmath179 in ( [ p ] ) , with the sdm @xmath43 computed according to theorem [ th : sopt ] , can be used as a legitimate approximation for pdfs of random processes .
more precisely , suppose @xmath208 is the time - varying pdf of a @xmath0-valued markov diffusion process @xmath209 with an infinitesimal generator @xmath210 ( so that @xmath211 for any smooth test function @xmath212 ; see , for example , @xcite ) .
for what follows , it is assumed that the renyi entropy @xmath176 in ( [ br ] ) remains finite .
this can be studied by using the integro - differential relations @xmath213 which follow ( provided @xmath172 and @xmath46 are smooth enough ) from the pdf dynamics governed by the fpke @xmath214 where the adjoint @xmath215 is understood in the sense of the standard @xmath216-space with the unit ( or constant ) weight .
then the sdm @xmath43 of the corresponding approximation @xmath179 of @xmath172 acquires dependence on time , which can be represented in the form of an ode .
the theorem below employs a self - adjoint operator @xmath217 on the space @xmath27 , which is the frechet derivative of the function of @xmath43 on the left - hand side of ( [ sopt ] ) : @xmath218 due to the assumption that @xmath182 , the operator @xmath217 is positive definite and strictly increasing for any given @xmath192 .
these properties of @xmath217 are inherited by its inverse @xmath219 .
also , let @xmath220 be a linear operator which maps the pdf @xmath172 to the complex hermitian matrix @xmath221 where @xmath210 is the generator of the markov diffusion process @xmath222 , and use is made of the structure matrices ( [ e ] ) .
[ th : sdot ] suppose the pdf @xmath223 of the underlying markov diffusion process has finite renyi entropy @xmath176 in ( [ br ] ) at every moment of time @xmath224 .
then the sdm @xmath43 , associated with @xmath172 according to theorem [ th : sopt ] , satisfies the ode @xmath225 where the operators @xmath217 and @xmath220 are defined by ( [ f ] ) and ( [ k ] ) . by taking the time derivative on both sides of ( [ sopt ] ) , it follows that @xmath226 where use is made of ( [ f ] ) and a combination of ( [ cb ] ) with ( [ k ] ) .
the invertibility of the operator @xmath217 allows ( [ sdot ] ) to be solved for @xmath227 as @xmath228 since @xmath217 is strictly increasing , then @xmath229 .
hence , @xmath230 , and @xmath231 in ( [ sdot1 ] ) can be uniquely found so as to satisfy the condition @xmath232 ( which comes from the preservation of @xmath41 in time ) : @xmath233 by substituting ( [ lamdot ] ) back into ( [ sdot1 ] ) , it follows that the sdm @xmath43 is governed by ( [ sdot2 ] ) .
the right - hand side of ( [ sdot2 ] ) is linear with respect to the pdf @xmath172 and depends on @xmath43 in a rational fashion .
the latter follows from the representation @xmath234 obtained by applying the vectorization @xmath235 of matrices @xcite to ( [ ca ] ) and ( [ f ] ) and using the relation @xmath236 in view of @xmath43 being hermitian , with @xmath237 the kronecker product .
if @xmath172 is an invariant pdf of the markov process @xmath222 , then ( [ k ] ) implies that @xmath238 , and ( [ sdot2 ] ) yields @xmath239 in accordance with the static setting considered in section [ sec : sdmapp ] .
now , for nonlinear stochastic systems , the solution @xmath172 of the fpke ( [ fpke ] ) is usually not available in a closed form . in this case
( when approximations of @xmath172 are particularly important ) , the right - hand side of ( [ sdot2 ] ) can be evaluated by replacing the unknown pdf @xmath172 with its sdm approximation @xmath179 , which leads to @xmath240 here , @xmath241 is a linear operator on the space @xmath27 , obtained by substituting the pdf @xmath179 from ( [ pes ] ) into the operator @xmath220 in ( [ k ] ) : @xmath242 the right - hand side of the ode ( [ sode ] ) is a rational function of the entries of @xmath43 ( thus resembling the riccati equations ) , which can be represented in the vectorised form by combining ( [ finvvec ] ) with @xmath243 following from ( [ q ] ) .
the constant matrices in ( [ finvvec ] ) and ( [ qvec ] ) can be precomputed using the structure matrices ( [ e ] ) and the matrix elements of the generator @xmath210 over the basis . by construction
, the ode ( [ sode ] ) has a positive definite solution whose trace is preserved in time . therefore , being a closure of ( [ sdot2 ] ) , the sdm dynamics produce a legitimate approximate solution @xmath179 of the fpke ( [ fpke ] ) .
we will now consider an application of the sdm approach to the approximation of pdfs for random processes governed by nonlinear sdes . as an illustrative example
, we will use a version of the smoluchowski sde @xcite on the @xmath2-dimensional torus @xmath3 : @xmath244 here , @xmath245 is a @xmath3-valued markov diffusion process , and @xmath246 is a twice continuously differentiable function with the gradient @xmath247 , where @xmath248 is the partial derivative with respect to the @xmath54th angular coordinate . also , @xmath249 is a scalar parameter , and @xmath250 is an @xmath2-dimensional standard wiener process .
the state space in the form of a multidimensional torus corresponds to systems with rotational degrees of freedom .
the sde ( [ ssde ] ) ( which is understood in the ito sense ) is a noisy version of the gradient descent for the function @xmath251 and is employed , for example , in the simulated annealing algorithm of stochastic optimization @xcite .
the function @xmath251 can be interpreted as the potential energy of a dissipative dynamical system in contact with a heat bath at temperature @xmath252 which specifies the noise level @xmath253 in ( [ ssde ] ) as @xmath254 , with @xmath255 the boltzmann constant .
this interpretation is motivated by the fact @xcite that the invariant probability measure for the smoluchowski sde ( [ ssde ] ) is absolutely continuous with the gibbs - boltzmann pdf @xmath256 here , @xmath257 is the statistical mechanical partition function @xcite of the auxiliary parameter @xmath258 , associated with the potential @xmath251 .
the invariant pdf @xmath259 in ( [ inv ] ) is a unique equilibrium point of the fpke ( [ fpke ] ) for the pdf @xmath260 of the random vector @xmath261 , which takes the form @xmath262 here , @xmath263 and @xmath264 are the divergence and laplace operators over the spatial variables .
the corresponding generator @xmath210 acts on a smooth test function @xmath265 as @xmath266
although the invariant pdf @xmath259 admits a closed - form representation ( [ inv ] ) , the time evolution ( [ sfpke ] ) of @xmath172 towards the equilibrium can be complicated , and it is especially so for multiextremum potentials @xmath251 . for example ,
in the context of molecular dynamics simulation , such @xmath251 can represent fractal - like energy landscapes ( as functions of the dihedral angles in protein macromolecules ) which are considered to play an important role in relaxation phenomena associated with protein folding @xcite .
this motivates investigation of the pdf dynamics for such fpkes .
we will now reformulate the fpke ( [ sfpke ] ) in the spatial frequency domain using the fact that all functions on the torus @xmath3 are @xmath267-periodic with respect to the angular variables and ( under the assumption of square integrability underlying the hilbert space @xmath268 ) can be represented by fourier series .
let @xmath269 be the fourier series for the potential @xmath251 and the pdf @xmath223 over the corresponding basis ( [ fourier ] ) .
since the functions @xmath251 and @xmath172 are real - valued , their fourier coefficients satisfy the hermitian property @xmath270 and @xmath271 for all @xmath272 and any time @xmath273 . the normalization condition for the pdf
@xmath223 is equivalent to @xmath274 since the potential @xmath251 is assumed to be twice continuously differentiable , @xmath275 .
furthermore , if @xmath251 is infinitely differentiable , it can be shown that the fpke ( [ sfpke ] ) has a smooth fundamental solution @xcite . similarly to the energy estimates for parabolic pdes @xcite , the norm @xmath276 at any time @xmath277 is amenable to an upper bound in terms of @xmath278 .
this is obtained by applying the gronwall - bellman lemma to the following differential inequality which is given here for completeness .
[ lem : diss ] the pdf @xmath172 , governed by the fpke ( [ sfpke ] ) for the smoluchowski sde ( [ ssde ] ) with an infinitely differentiable potential @xmath251 , satisfies @xmath279 a combination of ( [ sfpke ] ) , ( [ cg ] ) with the identities @xmath280 and @xmath281 ( and the property that @xmath172 and @xmath251 are real - valued ) leads to @xmath282 here , the first two equalities are , in fact , a particular case of ( [ brdot ] ) since the hilbert space @xmath7 being considered employs a constant weight .
also , use is made of the poincare - wirtinger inequality @xmath283 applied to the function @xmath284 which satisfies @xmath285 in view of ( [ f0 ] ) .
substitution of the last equality from ( [ fff ] ) into ( [ diss ] ) leads to ( [ diss0 ] ) .
another upper bound for the renyi entropy @xmath176 is provided by the inequalities @xmath286 where both denominators are finite and strictly positive due to ( [ inv ] ) and the continuity of @xmath251 over the torus .
the second of these inequalities follows from the dissipation relation @xmath287 whereby @xmath288 is a nonincreasing function of time @xmath289 . here
, @xmath290 is the pdf of the markov process @xmath222 with respect to the invariant measure ( so that the first equality in ( [ brff ] ) is similar to ( [ brdot ] ) ) , and use is made of a change of measure together with the relations @xmath291 and @xmath292 , with the latter following from ( [ cg ] ) .
now , ( [ vf ] ) allows the matrix elements of the generator @xmath210 over the fourier basis ( [ fourier ] ) to be computed as @xmath293 for all @xmath294 .
hence , in addition to ( [ f0 ] ) , the fourier coefficients of the pdf @xmath172 satisfy a denumerable set of odes @xmath295 which represent the fpke ( [ sfpke ] ) in the spatial frequency domain .
here , the convolution sum comes from the term @xmath296 and is responsible for the coupling of these odes ( the trivial case of a constant potential @xmath251 is not considered ) .
the truncation of the set of odes ( [ freqode ] ) ( for example , in accordance with galerkin s method for parabolic pdes @xcite ) leads to an approximate system of equations whose solution is not necessarily nonnegative and does not correspond to a legitimate pdf on the torus .
this issue can be overcome by using the sdm approximation of pdf dynamics described in section [ sec : sdmdyn ] .
in application to the smoluchowski sde ( [ ssde ] ) being considered , the sdm dynamics ( [ sode ] ) can be implemented in the vectorised form @xmath297 using ( [ finvvec ] ) , ( [ qvec ] ) together with the structure coefficients ( [ efourier ] ) and the matrix elements ( [ cgmat ] ) over the fourier basis ( [ fourier ] ) . the corresponding sdm approximation ( [ pfourier ] ) of the pdf @xmath172 in ( [ sfpke ] ) is @xmath298 where the maps @xmath79 in ( [ c ] ) and @xmath299 in ( [ psi ] ) are given by @xmath300 here , @xmath301 and @xmath302 in the case when the set @xmath23 and the associated set @xmath103 in ( [ mho ] ) are the discrete cubes @xmath303\bigcap \mz)^n , \qquad \mho= ( [ -2r,2r]\bigcap \mz)^n,\ ] ] with @xmath14 a positive integer .
we will now provide results of a numerical experiment on the sdm approximation of pdfs for the smoluchowski sde ( [ ssde ] ) with @xmath304 in the two - dimensional case @xmath305 .
the potential @xmath251 was generated as a trigonometric polynomial @xmath306 of @xmath307 with @xmath308 ( that is , @xmath309 independent harmonics ) and exponentially distributed random amplitudes @xmath310 and initial phases @xmath311 , with the latter being uniformly distributed over the interval @xmath157 for any @xmath312 ; see fig . [
fig : v ] . generated as a trigonometric polynomial of the spatial variables @xmath313 with random coefficients.,width=377,height=226 ]
the sdm approximation ( [ pfour ] ) was produced using ( [ sodevec ] ) and ( [ cpsi ] ) with the sets ( [ sets ] ) with @xmath314 , @xmath315 , @xmath316 and @xmath317 .
the initial conditions for the actual pdf @xmath172 and the sdm @xmath43 were @xmath318 and @xmath319 , with the latter corresponding to the same uniform distribution over the torus with the pdf @xmath320 .
the actual pdf @xmath172 was obtained through the standard finite - difference numerical solution of the fpke on a mesh of @xmath321 points with time step @xmath322 . the evolution of the relative values @xmath323 of the quadratic proximity criterion ( [ d ] ) ( with @xmath324 being the squared @xmath216-norm of @xmath172 ) is presented in fig . [ fig : dd ] which shows that the relative error remained within @xmath325 . of the sdm approximation in the course of evolution of the actual pdf @xmath172.,width=377,height=226 ]
that the time interval captures the essential part of the movement of the system towards equilibrium is seen from fig .
[ fig : fff ] , whereby both the actual pdf @xmath172 and its sdm approximation @xmath179 at the end of simulation were close to the invariant pdf @xmath259 .
( semitranslucent surface ) , the actual pdf @xmath172 ( black wireframe ) and its sdm approximation @xmath179 ( red wireframe ) after 2000 steps of evolution.,width=377,height=226 ]
we have outlined an approach to the approximation of pdfs by quadratic forms of weighted orthonormal basis functions parameterised by sdms .
the sdm approximation produces a legitimate pdf which satisfies the normalization condition and is nonnegative everywhere . for orthonormal bases with an algebraic structure , we have provided an optimization procedure for the sdm approximation using a quadratic criterion .
the optimal sdm approximation has been extended to pdfs of markov diffusion processes , and an ode has been obtained for the sdm dynamics which yields a legitimate approximate solution of the fpke .
this has been demonstrated for the smoluchowski sde with a multiextremum potential on a torus .
the sdm approach is also applicable ( in the spirit of projective filtering @xcite ) to the approximation of posterior pdfs governed by the kushner - stratonovich equations @xcite . in this application ,
the conditional sdm would play a part similar to that of the covariance matrix computed in the kalman filters .
however , unlike the covariance matrices , the sdm is not restricted to the second moments and involves higher order moments of the system variables .
numerical integration of the sdm dynamics can be implemented in a square - root form using the smoothness of the cholesky factorization of positive definite matrices @xcite .
error analysis of the sdm approximation is another line of research to be tackled in future publications .
b.d.o.anderson , and j.b.moore , _ optimal control : linear quadratic methods _
, prentice hall , london , 1989 .
o.e.barndorff-nielsen , and d.r.cox , _ asymptotic techniques for use in statistics _ , chapman & hall , london , 1989 .
r.c.booton , the analysis of nonlinear control systems with random inputs , proc .
nonlinear circuit analysis , polytechnic inst .
brooklyn , new york , vol . 2 , 1953 .
z.i.botev , j.f.grotowski , and d.p.kroese , kernel density estimation via diffusion , _ annals statist .
38 , no . 5 , 2010 , pp .
t.k.caughey , equivalent linearization technique , _ j. acoust .
35 , no . 11 , 1963 , pp .
t.duong , ks : kernel density estimation and kernel discriminant analysis for multivariate data in r , _ j. statist . soft .
21 , no . 7 , 2007 , pp .
116 .
s.janson , _ gaussian hilbert spaces _ , cambridge university press , cambridge , 1997 .
i.karatzas , and s.e.shreve , _ brownian motion and stochastic calculus _ , 2nd ed .
, springer - verlag , new york , 1991 .
i.e.kazakov , approximate probabilistic analysis of the accuracy of performance of essentially nonlinear control systems , _ automat .
remote contr .
17 , 1956 , pp .
423450 .
s.kirkpatrick , c.d.gelatt , and m.p.vecchi , optimization by simulated annealing , _ science _ ,
4598 , 1983 , pp .
h.j.kushner , on the differential equations satisfed by conditional probability densities of markov processes , with applications , _ j. siam control , ser .
a _ , vol .
2 , no . 1 , 1964 , pp . 106119 .
d.g.lainiotis , partitioned linear estimation algorithms : discrete case , _ ieee trans .
20 , no . 2 , 1975 , pp .
j.b.lasserre , global optimization with polynomials and the problem of moments , _ siam j. optim .
11 , no . 3 , 2001 , pp .
796817 .
g.lois , j.blawzdziewicz , and c.s.ohern , protein folding on rugged energy landscapes : conformational diffusion on fractal networks , _ phys .
e _ , vol .
81 , 2010 , 051907 .
j.r.magnus , _ linear structures _ , oxford university press , new york , 1988 .
p.a.parrilo , structured semidefinite programs and semialgebraic geometry methods in robustness and optimization , phd thesis , california institute of technology , pasadena , california , 2000 .
a.renyi , on measures of entropy and information , proc .
4th berkeley sympos .
i , 1961 , 547561 .
f.riesz , and b.sz.-nagy , _ functional analysis _ , f.ungar publishing , new york , 1955 .
j.j.sakurai , _ modern quantum mechanics _ , addison - wesley , reading , mass . , 1994 .
a.n.shiryaev , _ probability _ , 2nd ed . ,
springer , new york , 1996 .
s.p.smith , differentiation of the cholesky algorithm , _ j. comput .
4 , no . 2 , 1995 , pp . 134147 .
d.w.stroock , _ partial differential equations for probabilists _ , cambridge university press , cambridge , 2008 .
r.van handel , and h.mabuchi , quantum projection filter for a highly nonlinear model in cavity qed , _ j. opt .
b : quantum semiclass . opt .
_ , vol . 7 , 2005 , pp . s226s236
i.g.vladimirov , and i.r.petersen , gaussian stochastic linearization for open quantum systems using quadratic approximation of hamiltonians , proc .
mtns 2012 , melbourne , victoria , 913 july 2012 ( preprint : arxiv:1202.0946v1 [ quant - ph ] , 5 february 2012 ) .
r.zwanzig , _ nonequilibrium statistical mechanics _ , oxford university press , new york , 2001 . | this paper outlines an approach to the approximation of probability density functions by quadratic forms of weighted orthonormal basis functions with positive semi - definite hermitian matrices of unit trace .
such matrices are called stochastic density matrices in order to reflect an analogy with the quantum mechanical density matrices .
the sdm approximation of a pdf satisfies the normalization condition and is nonnegative everywhere in contrast to the truncated gram - charlier and edgeworth expansions . for bases with an algebraic structure , such as the hermite polynomial and fourier bases , the sdm approximation can be chosen so as to satisfy given moment specifications and can be optimized using a quadratic proximity criterion .
we apply the sdm approach to the fokker - planck - kolmogorov pdf dynamics of markov diffusion processes governed by nonlinear stochastic differential equations .
this leads to an ordinary differential equation for the sdm dynamics of the approximating pdf . as an example
, we consider the smoluchowski sde on a multidimensional torus . |
lunar occultations ( los ) can efficiently yield high angular resolution observations from the analysis of the diffraction light curves generated when background sources are covered by the lunar limb .
the technique has been employed to measure hundreds of stellar angular diameters , binary stars , and sources with extended circumstellar emission ( see charm2 catalog , * ? ? ?
in the past few years , a program to observe los in the near - infrared at the eso very large telescope ( vlt ) has been very successful both in quantity , with over a thousand events recorded , and in quality with a combination of angular resolution far exceeding the diffraction limit of a single telescope ( @xmath4 ) and a sensitivity significantly better than that currently achieved by long - baseline interferometry ( k @xmath3mag ) .
the drawbacks are that los are fixed - time events , yielding mainly a one - dimensional scan of the source , and that the source can not be chosen at will .
details on the los program at the vlt can be found in @xcite , and references therein . here , we report on 25 sources discovered to be binary with projected separations below one arcsecond , and in fact mostly below the 57 milliarcseconds ( mas ) diffraction limit of the telescope at the given wavelength .
we also report on one previously known system . in sect .
[ section : data ] we describe the observational procedure , the sample composition , and the data analysis . in sect .
[ section : results ] we report the individual results , and provide some context from previous bibliography when available .
some considerations on the statistics of binary detections from our vlt lo program and on the prospects of follow - up of selected systems are given in sect .
[ sec : conclusions ] .
the observations were carried out between april 2010 and october 2011 , using the 8.2-m ut3 melipal telescope of the vlt and the isaac instrument operated in burst mode .
most of the observations were carried out in service mode , based on a strategy of profiting from short slots that might become available depending on the atmospheric conditions and execution status of other service programs .
consequently , the targets were inherently random .
a few observations were parts of isolated nights dedicated to lo observations in visitor mode , and in this case the sources were selected on the basis of their colors and brightness in very extincted regions of the galactic bulge .
the sources observed in service mode have the field `` our i d '' in table [ tab : list ] beginning with p85 to p88 , which are the eso periods under consideration .
the sources without this prefix were observed in visitor mode .
table [ tab : list ] provides a list of the observations and of the characteristics of the sources , ordered by time .
a sequential number is included , for ease of cross - reference .
our predictions were generated from the 2mass catalogue , and so is the listed near - infrared photometry . we did not attempt to derive proper k - band photometry from our light curves , due to the lack of calibration sources .
however we notice that in some cases differences between our counts and the 2mass magnitudes of up to a factor 2 are present , pointing to possible variability ( fig .
[ fig : counts ] ) .
further identifications in table [ tab : list ] , as well as visual photometry and spectral types , are extracted from the _ simbad _ database .
each observation consisted of 7500 or 5000 frames ( in service or visitor runs , respectively ) in a 32x32-pixel ( @xmath5 ) sub - window , with a time sampling of 3.2ms .
this was also the effective integration time .
a broad - band k@xmath6 filter was employed for all events , except in the case of the brighter sources stars 4 and 22 , for which a narrow - band filter centered at 2.07@xmath7 m was employed to avoid possible non - linearities .
the events were disappearances , with lunar phases ranging from 38% to 96% ( median 66% ) .
airmass ranged from 1.1 to 2.0 , while seeing ranged from @xmath8 to @xmath9 ( median @xmath10 ) .
the lo light curves are generated in vacuum at the lunar limb and the diffraction fringes span a range of few 0.1s , so that the technique is in any case largely insensitive to atmospheric perturbations . the data cubes were converted to light curves by extracting the signal at each frame within a mask tailored to the seeing and image quality .
the light curves were then analyzed using both a model - dependent @xcite and a model - independent method ( cal , @xcite ) .
this latter is well suited to derive brightness profiles in the case of faint binaries , from which the initial values of the model can be inferred .
the least - squares fits are driven by changes in normalized @xmath11 , with a noise model defined for each light curve from the data before and after the occultation .
more details on the instrumentation and the method can be found in @xcite and references therein .
it should be noted that only restricted portions of the light curves around the main disappearance , corresponding to angular extensions of @xmath2 , were considered . in general , companions with projected separations larger than this would not appear in our list .
rllllrrrrrrrl 1 & p85 - 06 & 07283985 + 2151062 & bd+22 1693 & 21-apr-10 & 00:03:25 & 9.52 & 9.12 & 8.50 & 8.36 & 8.30 & f2 + 2 & p85 - 23 & 19224512 - 2046033 & bd-21 5366 & 21-aug-10 & 05:39:03 & 11.18 & 10.07 & 8.09 & 7.59 & 7.44 & + 3 & p85 - 26 & 19240606 - 2103008 & bd-21 5373 & 21-aug-10 & 05:59:25 & 11.63 & 10.28 & 6.73 & 5.93 & 5.65 & + 4 & s033 & 18115684 - 2330488 & akari - irc - v1 & 16-sep-10 & 03:08:18 & 16.50 & 15.60@xmath12 & 5.70 & 4.51 & 4.02 & + 5 & p86 - 02 & 22275268 - 0418586 & hd 212913 & 15-nov-10 & 00:52:55 & 10.61 & 8.92 & 5.42 & 4.28 & 4.07 & m ... + 6 & p86 - 06 & 22282898 - 0417248 & & 15-nov-10 & 01:33:46 & & & 10.32 & 9.76 & 9.63 & + 7 & p86 - 13 & 22304733 - 0348326 & hd 213344 & 15-nov-10 & 03:24:31 & 10.82 & 9.75 & 7.77 & 7.23 & 7.11 & k0 + 8 & p86 - 21 & 23551978 + 0532182 & tyc 593 - 1067 - 1 & 17-nov-10 & 00:35:44 & 10.44 & 9.92 & 8.94 & 8.74 & 8.66 & + 9 & p86 - 27 & 23581573 + 0612341 & tyc 593 - 1360 - 1 & 17-nov-10 & 03:37:07 & 10.85 & 9.88 & 8.44 & 8.04 & 7.95 & + 10 & p86 - 31 & 23592250 + 0614580 & tyc 593 - 1337 - 1 & 17-nov-10 & 04:20:09 & 11.62 & 11.11 & 9.98 & 9.72 & 9.66 & + 11 & p87 - 014 & 09463209 + 0913340 & & 13-apr-11 & 23:11:43 & & & 10.21 & 9.67 & 9.55 & + 12 & p87 - 026 & 10220366 + 0515554 & tyc 252 - 1217 - 1 & 11-may-11 & 23:03:59 & 12.03 & 11.33 & 9.86 & 9.48 & 9.42 & + 13 & p87 - 041 & 10070867 + 0628201 & tyc 250 - 82 - 1 & 07-jun-11 & 22:55:04 & 11.23 & 10.55 & 9.53 & 9.27 & 9.15 & + 14 & vmj-023 & 17350423 - 2319491 & & 13-jul-11 & 04:49:05 & & & 9.75 & 8.60 & 8.08 & + 15 & vmj-056 & 17395562 - 2303408 & hd 160257 & 13-jul-11 & 07:26:09 & 9.18 & 8.58 & 7.47 & 7.20 & 7.13 & g2v + 16 & vma-049 & 18142147 - 2207047 & iras 18113 - 2208 & 10-aug-11 & 03:14:03 & & & 7.48 & 6.20 & 5.65 & + 17 & vma-055 & 18151114 - 2213294 & & 10-aug-11 & 03:33:19 & & & 7.80 & 6.55 & 5.85 & + 18 & vma-062 & 18154557 - 2220045 & & 10-aug-11 & 03:57:43 & & & 8.45 & 7.32 & 6.71 & + 19 & vma-083 & 18172695 - 2147313 & & 10-aug-11 & 05:27:34 & & & 9.25 & 8.05 & 7.37 & + 20 & p88 - 003 & 17370333 - 2239165 & hd 159700 & 02-oct-11 & 23:33:40 & 11.28 & 9.67 & 5.81 & 5.04 & 4.60 & k7 + 21 & p88 - 008 & 17371972 - 2215045 & iras 17343 - 2213 & 03-oct-11 & 00:02:24 & & & 8.89 & 7.86 & 7.10 & + 22 & p88 - 021 & 17401076 - 2208280 & iras
17371 - 2207 & 03-oct-11 & 01:30:13 & 12.16 & 10.67 & 5.32 & 4.11 & 3.75 & + 23 & p88 - 046 & 18395035 - 2052114 & bd-20 5222 & 04-oct-11 & 02:02:01 & 11.00 & 9.79 & 7.30 & 6.70 & 6.53 & + 24 & p88 - 050 & 18402840 - 2040072 & ko sgr & 04-oct-11 & 02:31:16 & 12.70 & 14.50 & 6.84 & 5.91 & 5.36 & + 25 & p88 - 052 & 18403433 - 2034059 & iras 18375 - 2036 & 04-oct-11 & 02:45:15 & & & 7.89 & 6.85 & 6.31 & +
table [ tab : results ] lists our results , following closely the format already used in previous papers . the same sequential number used in table [ tab : list ] is included , followed by the 2mass identification .
the next two columns are the observed rate of the event and its deviation from the predicted value .
the difference is due to the local limb slope @xmath13 , from which the actual observed position angle pa and contact angle ca are derived .
we then list the binary fit results , namely the signal - to - noise ratio ( snr ) of the light curve , the separation and brightness ratio , and the two individual magnitudes based on the total magnitude listed in the 2mass . as mentioned in sect .
[ section : data ] , for stars 4 and 22 narrow - band filters were used but we do not expect significant effects on k@xmath14 , k@xmath15 .
rlrrrrrrccrr 1 & 07283985 + 2151062 & 0.6326 & -12.0% & -26 & 94 & -23 & 26.5 & 190.29 @xmath16 0.06 & 2.335 @xmath16 0.004 & 8.69 & 9.61 + 2 & 19224512 - 2046033 & 0.4199 & -8.5% & -4 & 7 & -58 & 64.2 & 55.5 @xmath16 0.4 & 7.98 @xmath16 0.02 & 7.56 & 9.82 + 3 & 19240606 - 2103008 & 0.4715 & -12.2% & -6 & 107 & 43 & 91.6 & 8.4 @xmath16 0.2 & 12.06 @xmath16 0.04 & 5.74 & 8.44 + 4 & 18115684 - 2330488 & 0.6350 & -5.8% & -5 & 33 & -38 & 70.3 & 8.8 @xmath16 0.3 & 18.9 @xmath16 0.2 & 4.07 & 7.27 + 5 & 22275268 - 0418586 & 0.4576 & -7.9% & -6 & 83 & 33 & 116.8 & 6.29 @xmath16 0.05 & 7.53 @xmath16 0.01 & 4.20 & 6.39 + 6 & 22282898 - 0417248 & 0.2254 & -21.5% & -6 & 110 & 58 & 10.4 & 6.9 @xmath16 2.6 & 4.5 @xmath16 0.1 & 9.85 & 11.48 + 7 & 22304733 - 0348326 & 0.6890 & 10.9% & 9 & 283 & 46 & 47.7 & 42.5 @xmath16 0.4 & 29.2 @xmath16 0.3 & 7.15 & 10.81 + 8 & 23551978 + 0532182 & 0.5415 & -7.0% & -10 & 58 & 8 & 39.4 & 15.3 @xmath16 0.7 & 12.7 @xmath16 0.2 & 8.74 & 11.50 + 9 & 23581573 + 0612341 & 0.6639 & 0.4% & 1 & 47 & -10 & 46.1 & 19.6 @xmath16 1.0 & 45.1 @xmath16 1.6 & 7.97 & 12.11 + 10 & 23592250 + 0614580 & 0.6767 & -4.2% & -9 & 62 & 2 & 2.5 & 40.1 @xmath16 0.3 & 1.03 @xmath16 0.01 & 10.40 & 10.43 + 11 & 09463209 + 0913340 & 0.4553 & -22.3% & -16 & 141 & 20 & 7.5 & 47.9 @xmath16 0.2 & 1.50 @xmath16 0.01 & 10.11 & 10.54 + 12 & 10220366 + 0515554 & 0.6121 & -3.2% & -4 & 146 & 20 & 9.0 & 29.4 @xmath16 3.5 & 2.34 @xmath16 0.02 & 9.81 & 10.73 + 13 & 10070867 + 0628201 & 0.4769 & 5.3% & 3 & 179 & 53 & 10.3 & 371.7 @xmath16 4.1 & 1.447 @xmath16 0.007 & 9.72 & 10.12 + 14 & 17350423 - 2319491 & 0.3648 & -0.3% & 0 & 137 & 58 & 31.6 & 8.3 @xmath16 0.8 & 9.7 @xmath16 0.1 & 8.19 & 10.66 + 15 & 17395562 - 2303408 & 0.6512 & 26.4% & 12 & 324 & 67 & 29.3 & 41.96 @xmath16 0.07 & 1.137 @xmath16 0.003 & 7.81 & 7.95 + 16 & 18142147 - 2207047 & 0.5564 & 0.1% & 0 & 222 & -32 & 202.4 & 32.4 @xmath16 0.3 & 144.4 @xmath16 1.6 & 5.65 & 11.05 + 17 & 18151114 - 2213294 & 0.6585 & -1.7% & -7 & 71 & -3 & 209.2 & 15.1 @xmath16 0.2 & 100.1 @xmath16 1.3 & 5.86 & 10.86 + 18 & 18154557 - 2220045 & 0.3918 & -28.0% & -17 & 95 & 22 & 72.9 & 9.9 @xmath16 0.4 & 24.7 @xmath16 0.3 & 6.75 & 10.24 + 19 & 18172695 - 2147313 & 0.4303 & 9.8% & 3 & 194 & -58 & 22.5 & 27.8 @xmath16 0.9 & 12.2 @xmath16 0.1 & 7.45 & 10.17 + 20 & 17370333 - 2239165 & 0.3548 & -33.8% & -20 & 98 & 19 & 99.8 & 8.8 @xmath16 0.1 & 25.60 @xmath16 0.09 & 4.64 & 8.16 + 21 & 17371972 - 2215045 & 0.4863 & 10.0% & 5 & 31 & -47 & 124.9 & 7.30 @xmath16 0.07 & 18.9 @xmath16 0.1 & 7.16 & 10.35 + 22 & 17401076 - 2208280 & 0.6415 & -1.0% & -1 & 36 & -40 & 135.5 & 6.2 @xmath16 0.3 & 36.5 @xmath16 0.5 & 3.78 & 7.69 + 23 & 18395035 - 2052114 & 0.7922 & 0.4% & 1 & 262 & 12 & 109.3 & 748.4 @xmath16 0.2 & 23.13 @xmath16 0.04 & 6.58 & 9.99 + 24 & 18402840 - 2040072 & 0.7780 & -2.1% & -3 & 227 & -23 & 217.6 & 18.1 @xmath16 0.2 & 78.3 @xmath16 0.6 & 5.37 & 10.11 + 25 & 18403433 - 2034059 & 0.5391 & -15.4% & -9 & 19 & -52 & 111.9 & 8.1 @xmath16 0.3 & 18.5 @xmath16 0.1 & 6.37 & 9.54 + many of our sources are in the direction of the galactic bulge , having 17@xmath17 @xmath18 ra @xmath18 19@xmath17 and dec @xmath19 @xmath20 .
they have generally very red colors ; however a color - color diagram shows that these are mostly consistent with significant amounts of interstellar extinction ( fig .
[ fig : colors ] ) , with a possible notable exception to be discussed later . in line with this extinction ,
many of our sources have relatively faint optical counterparts , or none .
few of them have been studied in detail previously , and spectral information is correspondingly scarce . in the reminder of this section we discuss individual cases in the context of available literature , when possible .
star 5 : is and , for which no literature exists except a generic late m spectral type .
the tycho-2 parallax values seem to place this star at @xmath19300pc , thus hinting at a giant star .
the star is not in the bulge and is @xmath21 below the galactic disk , yet its location in fig .
[ fig : colors ] is peculiar and indicative of substantial reddening .
we attempted fits to our light curve with a binary , a resolved diameter , and a diameter plus companion models .
the first one gave the best normalized @xmath11 and is the one we adopt in our results .
star 12 : was included in tycho-2 as .
however , no binarity was detected , including in the subsequent dedicated re - analysis of the tycho double star catalogue .
no literature references were found for this source .
star 13 : is . in spite of
the relatively large measured projected separation , no binarity was detected in tycho 2 .
no literature references are present for this source .
star 15 : is , a bright , nearby , g2v - type star .
it was classified as a pre - main sequence star by , on the basis of x - ray emission and high resolution spectra .
we find it to to be binary with a rather large projected separation ( see fig .
[ fig : vmj-056 ] ) .
it was observed by hipparcos as , but no binarity was found . as for the recent case of new binaries in the pleiades
, this shows the potential of lo to further extend the statistics of binarity among young stars in the context of multiple star formation mechanisms .
star 16 : is the infrared source .
no literature references were found for this source .
it is also the binary with the largest magnitude difference in our sample , @xmath22k=5.4mag against a dynamic range of 5.8mag .
star 21 : is the infrared source , also measured by the akari satellite
. there are no bibliographical references .
it is one of the reddest sources in our sample , having j - k=1.8mag from 2mass .
our recorded flux shows an increase of @xmath1980% above the 2mass k@xmath6 magnitude ( cfr .
[ fig : counts ] ) , pointing to possible variability .
the detection of the 3.2mag fainter companion with 7mas projected separation is shown in fig .
[ fig : p88 - 08 ] . in the lower panel of the figure , note that the cal algorithm preserves the integrated flux ratios , not the peaks in the brightness profile .
star 23 : is the star with the largest projected separation in our list , @xmath23 , however it does not seem to have been detected before in spite of being relatively bright
. the 1:23 brightness ratio in the k - band , and possibly more at visual wavelengths , could be one reason .
concerning projection effects of a nearby field star , we have investigated images from dss , hst and 2mass without evidence of other stars within the isaac subwindow .
star 24 : , also identified as , was found to have a regular period of 312 days by @xcite and thus tentatively classified as a mira - type star . the same author , however , noted the peculiarity of rather steep increases in luminosity , and the relatively short - lived maxima .
she suggested the possibility of a companion in a cataclysmic system , or just a visual association . in this scenario ,
the light at minimum would come mostly from the secondary .
the photographic magnitude difference between minima and maxima is about 3.1mag . using the hoffleit photometric period ,
our observation would have occurred at phase 0.8 or just before the onset of the outburst - like maximum .
uncertainties are possible due to the @xmath19100 cycles intervened since hoffleit s first determination , but are not easy to estimate .
data in the aavso database are also not very complete . under the above assumptions ,
the @xmath22k=4.7mag would point to a primary much redder than the secondary , consistent with the scenario outlined in @xcite .
we also recorded a lo light curve of = on april 13 , 2011 ( star p87 - 010 in our database ) , detecting a well - separated companion with @xmath24mas projected separation along pa=@xmath25 .
the 2mass - based magnitudes of the two components are k=9.04 and k=10.31 .
this is listed as the wide double @xcite , without orbital parameters and with a separation of @xmath26 .
given the wide nature of this binary , we did not include it in our tables and figures , and we mention our results only as a complement to the previous observations .
@xcite list the sources observed in the same program and found to be unresolved , covering also the period under consideration here . from the first to the last night considered in table [ tab : list ] , a total of 403 lo light curves were observed at the vlt with isaac , therefore our serendipitous binary detection fraction is ( 26/403 ) @xmath27% . restricting ourselves to the purely random service mode observations ,
the number is ( 19/231 ) @xmath28% .
we note that the visitor mode targets were in crowded and extincted regions in , and consequently mostly deep into , the galactic bulge , with a median k=6.4mag .
by contrast , the service mode targets were randomly scattered and with median k=8.0mag .
one possible reason for the higher incidence of binaries among the service mode targets is that they are statistically closer to us , therefore providing a better spatial scale for the same angular resolution . in a previous work based on a comparable volume of lo
observed also in the near - ir in subarray fast mode from calar alto , it was found that the serendipitous binary detection frequency was significantly smaller than the present one , @xmath29% .
the difference can be justified in principle with the lower sensitivity ( limiting k @xmath30mag then , and @xmath3mag here ) and the slower time sampling ( @xmath31ms then , against 3.2ms here ) . however , both cases appear to have an inferior yield of binary stars than observations by a fast photometer : e.g. , @xcite using the tirgo 1.5 m telescope found ( 26/157 ) @xmath32% binary detection frequency .
the explanation in this case is that the targets were brighter , being selected prior to the availability of the 2mass catalog , with average k @xmath29mag .
hence , the sources were generally significantly closer to us than in the present study
. the flux ratios of the binaries in our list range from @xmath22k=0.0 to 5.4mag , with a median of 2.8mag . from fig .
[ fig : k1k2 ] it can be seen that half of the sample falls in the range 2@xmath334mag . as for the separations , lo can measure only projected values , and the median for our sample is 18mas . considering a random projection correction of 2/@xmath34 , about half of our binaries would remain inaccessible to techniques with less than @xmath35 resolution .
this corresponds , e.g. , to the k - band diffraction limit of a 12 m telescope .
long - baseline interferometry at some of the largest facilities can provide the necessary angular resolution , but it suffers from a sensitivity which is significantly more limited than for lo and in general from a reduced dynamic range .
we outline schematically the situation in fig . [
fig : brsep ] . from the figure
, it can be estimated that only about 2/3 of the sources in our list might be effectively followed up by independent methods . in conclusion
, we find from the present work as well as from previous similar research that a routine program of random lo with an ir detector operated in subarray mode at an 8m - class telescope with time resolution of order 3ms can detect companions to stars in the general field down to a sensitivity k @xmath3mag , with separations as small as a few milliarcseconds ( at least 10 times better than the diffraction limit of the telescope ) .
the expected detection frequency can range from @xmath36% in distant regions such as deep in the galactic bulge , to @xmath37% in areas closer to the solar neighbourhood .
most of these binaries or multiples turn out to be new detections , with limited cross - identifications , spectral determinations and previous literature . due to the small separations and high brightness ratios , only up to @xmath38% of these lo - detected binaries
can be followed up by other more flexible methods such as adaptive optics and interferometry .
a significant fraction of such new binary detections will remain isolated until significantly larger telescopes or more sensitive interferometers become available .
we are grateful to the eso staff in europe and chile for their continued support especially of the service observations . of acknowledges financial support from mineco through a _
juan de la cierva _ fellowship and from _ mcyt - sepcyt plan nacional i+d+i aya#2008 - 01225_. this research made use of the simbad database , operated at the cds , strasbourg , france , and of data products from the two micron all sky survey ( 2mass ) , which is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by the national aeronautics and space administration and the national science foundation .
this research has made use of the washington double star catalog maintained at the u.s .
naval observatory .
bessell , m. s. , & brett , j. m. 1988 , , 100 , 1134 fabricius , c. , hg , e. , makarov , v. v. , et al . 2002 , , 384 , 180 hoffleit , d. 1960 , , 65 , 100 hg , e. , fabricius , c. , makarov , v. v. , et al .
2000 , , 357 , 367 mason , b. d. , wycoff , g. l. , hartkopf , w. i. , douglass , g. g. , & worley , c. e. 2001 , , 122 , 3466 richichi , a. 1989 , , 226 , 366 richichi , a. , baffa , c. , calamai , g. , & lisi , f. 1996 , , 112 , 2786 richichi , a. , percheron , i. , & khristoforova , m. 2005 , , 431 , 773 richichi , a. , fors , o. , merino , m. , et al . 2006 , ,
445 , 1081 richichi , a. , chen , w. p. , fors , o. , & wang , p. f. 2011 , , 532 , a101 richichi , a. , chen , w. p. , cusano , f. , et al . 2012a , , 541 , a96 richichi , a. , cusano , f. , fors , o. , moerchen , m.2012b , , 203 , 33 rieke , g. h. , & lebofsky , m. j. 1985 , , 288 , 618 torres , c. a. o. , quast , g. r. , da silva , l. , et al .
2006 , , 460 , 695 | we report on 25 sub - arcsecond binaries , detected for the first time by means of lunar occultations in the near - infrared as part of a long - term program using the isaac instrument at the eso very large telescope .
the primaries have magnitudes in the range k=3.8 to 10.4 , and the companions in the range k=6.4 to 12.1 .
the magnitude differences have a median value of 2.8 , with the largest being 5.4 .
the projected separations are in the range 6 to 748 milliarcseconds and with a median of 18 milliarcseconds , or about 3 times less than the diffraction limit of the telescope . among our binary detections
are a pre - main sequence star and an enigmatic mira - like variable previously suspected to have a companion .
additionally , we quote an accurate first - time near - ir detection of a previously known wider binary .
we discuss our findings on an individual basis as far as made possible by the available literature , and we examine them from a statistical point of view .
we derive a typical frequency of binarity among field stars of @xmath0% , in the resolution and sensitivity range afforded by the technique ( @xmath1 to @xmath2 , and k@xmath3mag , respectively ) .
this is in line with previous results by the same technique but we point out interesting differences that we can trace up to sensitivity , time sampling , and average distance of the targets .
finally , we discuss the prospects for further follow - up studies . |
the discovery of a light higgs boson at the lhc @xcite completes beautifully the standard model ( sm ) description of fundamental interactions . nonetheless , a number of questions about the theoretical consistency of the theory remain unanswered , and point to the existence of some new physics around or not far from the tev scale . in the lack of direct detection of any new particle , a worthy and rather model - independent tool for the study of beyond - standard model ( bsm ) physics is provided by lorentz and gauge - invariant effective lagrangians , which respect a given set of symmetries , including the low - energy established ones . in particular , with a light higgs observed , two main classes of effective lagrangians are pertinent , depending on how the electroweak ( ew ) symmetry breaking ( ewsb ) is assumed to be realized : linearly for elementary higgs particles or non - linearly for dynamical " -typically composite- ones . in elementary higgs scenarios , the higgs particle belongs to an @xmath0 doublet @xmath1 . if the new physics scale is @xmath2 , where @xmath3 is the ew vacuum expectation value ( vev ) , the corresponding linear effective lagrangian contains operators weighted by inverse powers of the cutoff scale @xmath4 : the leading corrections to the sm lagrangian have then canonical mass dimension @xmath5 @xcite .
a typical example of possible underlying physics are supersymmetric models . in dynamical higgs scenarios , on the other hand ,
the higgs particle is a composite field which happens to be a pseudo - goldstone boson ( gb ) of a global symmetry @xmath6 , that is broken down spontaneously to a subgroup @xmath7 at a scale @xmath8 , corresponding to the masses of the lightest strong resonances .
the higgs mass is protected by the gb shift symmetry , thus avoiding the electroweak hierarchy problem .
explicit realizations include composite higgs models @xcite .
the most suitable effective lagrangian for this scenario is a non - linear @xcite or
chiral " one : a derivative expansion as befits goldstone boson dynamics .
remarkably , the effective linear and chiral lagrangians with a light higgs are in general different .
there is not a one - to - one correspondence of the leading corrections of both expansions , and one expansion is not the limit of the other unless specific constraints are imposed by hand @xcite or follow from particular dynamics at high energies @xcite . finding signals which differentiate between these two categories may therefore provide a powerful insight to the origin of the ewsb mechanism .
a few relevant examples of such discriminating signatures will be presented in this talk , together with a brief discussion of the results obtained for specific composite higgs models ( _ i.e. _ specific choices of the coset @xmath9 ) .
the effective low - energy chiral lagrangian for a light higgs is entirely written in terms of the sm fermions and gauge bosons and of the physical higgs @xmath10 .
the sm gbs can be described by a dimensionless unitary matrix @xcite : @xmath11 , @xmath12 with @xmath13 denoting respectively the @xmath14 global transformations of the scalar potential .
higgs couplings are now ( model - dependent ) generic functions @xmath15 the @xmath0 structure is absent in them and , as often pointed out ( e.g. refs .
@xcite ) , the resulting effective lagrangian can describe many setups including that for a light sm singlet isoscalar .
the light dynamical scalar particle @xmath10 can be described by the phenomenological lagrangian @xmath16 where the leading order @xmath17 is the sm lagrangian , and @xmath18 describes any deviation from the sm due to strong - interacting new physics present at scales above the ew one .
the former term reads then @xmath19 + i\bar{q}\slashed{d}q+i\bar{l}\slashed{d}l\\ & -\frac{v+ h}{\sqrt2}\left(\bar{q}_l{\mathbf{u}}\mathbf{y}_q q_r+{\text{h.c.}}\right)-\frac{v+ h}{\sqrt2}\left(\bar{l}_l{\mathbf{u}}\mathbf{y}_l l_r+{\text{h.c.}}\right)\ , , \end{split}\ ] ] where ( @xmath20 ) is the vector ( scalar ) chiral field transforming in the adjoint of @xmath0 .
the covariant derivative is @xmath21 with @xmath22 and @xmath23 denoting the @xmath0 and @xmath24 gauge bosons , respectively . in eq .
, the first line describes the @xmath10 and gauge boson kinetic terms , and the effective scalar potential @xmath25 .
the second line describes the @xmath26 and @xmath27 masses and their interactions with @xmath10 , as well as the kinetic terms for gbs and fermions .
finally , the third line corresponds to the yukawa - like interactions written in the fermionic mass eigenstate basis .
a compact notation for the right - handed fields has been adopted , gathering them into doublets @xmath28 and @xmath29 .
@xmath30 and @xmath31 are two @xmath32 block - diagonal matrices containing the usual yukawa couplings .
the term @xmath18 includes all the effective operators with up to four derivatives allowed by lorentz and gauge symmetries . in the bosonic ( pure gauge , pure higgs and gauge-@xmath10 operators ) , cp even sector ,
to which we restrict in this talk , it can be decomposed as @xmath33 where @xmath34 are model - dependent coefficients , and the operators @xmath35 are defined by @xcite : @xmath36 ) { \mathcal{f}}_{2}(h){\nonumber}\\ { \mathcal{p}}_{3}(h)&= ig { \operatorname{tr}}({w_{\mu\nu}}[{\mathbf{v}}^\mu,{\mathbf{v}}^\nu ] ) { \mathcal{f}}_{3}(h){\nonumber}\\ { \mathcal{p}}_{4}(h ) & = ig ' { b_{\mu\nu}}{\operatorname{tr}}({\mathbf{t}}{\mathbf{v}}^\mu ) { \partial}^\nu{\mathcal{f}}_{4}(h){\nonumber}\\ { \mathcal{p}}_{5}(h)&= ig { \operatorname{tr}}({w_{\mu\nu}}{\mathbf{v}}^\mu ) { \partial}^\nu{\mathcal{f}}_{5}(h){\nonumber}\\ { \mathcal{p}}_{6}(h ) & = ( { \operatorname{tr}}({\mathbf{v}}_\mu{\mathbf{v}}^\mu))^2 { \mathcal{f}}_{6}(h){\nonumber}\\ { \mathcal{p}}_{7}(h ) & = { \operatorname{tr}}({\mathbf{v}}_\mu{\mathbf{v}}^\mu ) { \partial}_\nu{\partial}^\nu{\mathcal{f}}_{7}(h ) { \nonumber}\\ { \mathcal{p}}_{8}(h ) & = { \operatorname{tr}}({\mathbf{v}}_\mu{\mathbf{v}}_\nu ) { \partial}^\mu{\mathcal{f}}_{8}(h){\partial}^\nu{\mathcal{f}}_{8}'(h){\nonumber}\end{aligned}\ ] ] @xmath37 ) { \mathcal{f}}_{13}(h ) \\ { \mathcal{p}}_{14}(h ) & = g { \varepsilon}^{\mu\nu\rho\lambda } { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\mu ) { \operatorname{tr}}({\mathbf{v}}_\nu w_{\rho\lambda } ) { \mathcal{f}}_{14}(h)\\ { \mathcal{p}}_{15}(h ) & = { \operatorname{tr}}({\mathbf{t}}{\mathcal{d}}_\mu{\mathbf{v}}^\mu ) { \operatorname{tr}}({\mathbf{t}}{\mathcal{d}}_\nu{\mathbf{v}}^\nu ) { \mathcal{f}}_{15}(h ) \\ { \mathcal{p}}_{16}(h ) & = { \operatorname{tr}}([{\mathbf{t}},{\mathbf{v}}_\nu]{\mathcal{d}}_\mu{\mathbf{v}}^\mu ) { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}^\nu ) { \mathcal{f}}_{16}(h ) \\ { \mathcal{p}}_{17}(h ) & = ig { \operatorname{tr}}({\mathbf{t}}{w_{\mu\nu } } ) { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}^\mu ) { \partial}^\nu{\mathcal{f}}_{17}(h)\\ { \mathcal{p}}_{18}(h ) & = { \operatorname{tr}}({\mathbf{t}}[{\mathbf{v}}_\mu,{\mathbf{v}}_\nu]){\operatorname{tr}}({\mathbf{t}}{\mathbf{v}}^\mu ) { \partial}^\nu{\mathcal{f}}_{18}(h)\\ { \mathcal{p}}_{19}(h ) & = { \operatorname{tr}}({\mathbf{t}}{\mathcal{d}}_\mu{\mathbf{v}}^\mu){\operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\nu ) { \partial}^\nu{\mathcal{f}}_{19}(h)\\ { \mathcal{p}}_{20}(h ) & = { \operatorname{tr}}({\mathbf{v}}_\mu{\mathbf{v}}^\mu ) { \partial}_\nu{\mathcal{f}}_{20}(h){\partial}^\nu{\mathcal{f}}_{20}'(h ) \\ { \mathcal{p}}_{21}(h ) & = ( { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\mu))^2 { \partial}_\nu{\mathcal{f}}_{21}(h){\partial}^\nu{\mathcal{f}}_{21}'(h ) \\ { \mathcal{p}}_{22}(h ) & = { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\mu){\operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\nu ) { \partial}^\mu{\mathcal{f}}_{22}(h){\partial}^\nu{\mathcal{f}}_{22}'(h)\\ { \mathcal{p}}_{23}(h ) & = { \operatorname{tr}}({\mathbf{v}}_\mu{\mathbf{v}}^\mu ) ( { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\nu))^2 { \mathcal{f}}_{23}(h)\\ { \mathcal{p}}_{24}(h ) & = { \operatorname{tr}}({\mathbf{v}}_\mu{\mathbf{v}}_\nu){\operatorname{tr}}({\mathbf{t}}{\mathbf{v}}^\mu){\operatorname{tr}}({\mathbf{t}}{\mathbf{v}}^\nu ) { \mathcal{f}}_{24}(h)\\ { \mathcal{p}}_{25}(h ) & = ( { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\mu))^2 { \partial}_\nu{\partial}^\nu{\mathcal{f}}_{25}(h)\\ { \mathcal{p}}_{26}(h ) & = ( { \operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\mu){\operatorname{tr}}({\mathbf{t}}{\mathbf{v}}_\nu))^2 { \mathcal{f}}_{26}(h ) \end{split}\ ] ]
some of the chiral effective operators listed in the previous section contribute to quantities that have been well measured in current or past experiments , including triple ( tgc ) and quartic ( qgc ) gauge vertices , higgs couplings to two gauge bosons ( hvv ) and the ew parameters @xmath38 , @xmath39 , @xmath40 .
therefore , it is possible to set bounds on the coefficients @xmath34 with which they enter the lagrangian and on some of the parameters @xmath41 appearing inside the associated functions @xmath42 ( see eq . ) .
the results of the global fit are reported in ref .
@xcite , together with all the details of the renormalization procedure and of the numerical analysis .
the linear and non - linear efts are essentially different from each other in general , and it is of the utmost interest to investigate what are the signatures , at the phenomenological level , that would allow to disentangle which of them describes best nature . to this aim , it is illuminating to compare the basis of eq . with a complete set of independent linear operators of dimension six in the bosonic cp even sector , exploiting the correspondence @xmath43 .
such an analysis is performed in refs .
@xcite for the so - called hisz linear basis @xcite . in what follows
we implicitly use that linear basis , but limiting the discussion to a few relevant examples .
two main categories of effects stem from the phenomenological comparison of the effective lagrangians : 1 .
couplings that are predicted to be correlated in the linear parameterization , but that receive contributions from independent operators in the non - linear description .
this is the case of the non - linear operators @xmath44 and @xmath45 ( see eq . ) , that are in correspondence with the unique linear term @xmath46\,,\ ] ] where @xmath44 contributes to the tgcs usually dubbed @xmath47 and @xmath48 , while @xmath45 introduces the anomalous hvv vertices @xmath49 and @xmath50 .
+ in a linear scenario any departure of one of these couplings from its sm value is expected to be correlated with effects in the other three , since they all receive a contribution from @xmath51 , obviating for the time being all the other possible operators . moreover , the relative magnitude of such contributions is fixed by the structure of the covariant derivative @xmath52 . in the most general non - linear framework , instead
, no such correlation is present : deviations in @xmath53 are parameterized in terms of the coefficient @xmath54 , while those in the two anomalous hvv vertices are proportional to @xmath55 .
this effect is due to the different gauge representation chosen in the two theories for the higgs field : in the chiral formalism the higgs particle @xmath10 is treated as a gauge singlet , independent of the three sm gbs . as a consequence ,
this framework lacks the strong link between the couplings of the higgs and those of the longitudinal gauge bosons , which in the linear realization is imposed by the doublet structure of the field @xmath1 .
a completely analogous analysis holds for another pair of chiral operators , @xmath56 and @xmath57 , that contribute to the same tgv and hvv vertices as @xmath58 and correspond to the linear operator @xmath59\,.\ ] ] + in the event of some anomalous observation in either of the couplings mentioned above , the presence or absence of correlations would allow for direct testing of the nature of the higgs boson .
this is illustrated in fig .
[ fig : nl1 ] , where the results of the combined analysis of the tgv and hvv data are projected into combinations of the coefficients @xmath60 defined such that at order @xmath5 in the linear regime @xmath61 , @xmath62 , while @xmath63 .
notice that in the numerical analysis the complete ensemble of chiral operators is taken into account simultaneously . in the linear framework ( at order @xmath5 )
there is no equivalent of the variables @xmath64 : therefore the @xmath65 coordinate in the right panel of fig . [
fig : nl1 ] corresponds to the linear regime ( at order @xmath5 ) , and any evidence for a departure from this point would represent a smoking gun of a non - linear realization of the ewsb . would future data point to a departure from the origin of the first figure , instead , it would indicate bsm physics irrespective of the linear or non - linear character of the underlying dynamics .
+ analogous ( de)correlation effects between couplings with different number of higgs legs have been discussed in refs .
a more complex example , that involves the six chiral operators @xmath66 is analyzed in ref .
@xcite . + and @xmath67 defined in eq . , which converge to @xmath68 and @xmath69 in the linear @xmath5 limit . *
right * : a non - linear versus linear discriminator : constraints on the combinations @xmath70 and @xmath71 . in this case
the dot at @xmath65 indicates the linear limit .
both figures show the allowed regions at 68% , 90% , 95% , and 99% cl after marginalization over to the other parameters .
the star corresponds to the best fit point of the analysis .
_ , title="fig : " ] and @xmath67 defined in eq . , which converge to @xmath68 and @xmath69 in the linear @xmath5 limit .
* right * : a non - linear versus linear discriminator : constraints on the combinations @xmath70 and @xmath71 . in this case
the dot at @xmath65 indicates the linear limit .
both figures show the allowed regions at 68% , 90% , 95% , and 99% cl after marginalization over to the other parameters .
the star corresponds to the best fit point of the analysis .
_ , title="fig : " ] 2 .
couplings that appear only at higher order in the linear expansion , i.e. in linear operators of dimension @xmath72 , but are allowed as first - order corrections to the sm ( i.e. at the four - derivatives level ) in the non - linear description .
a striking example is that of the operator @xmath73 defined in eq .
, whose linear sibling is an operator of dimension 8 : @xmath74 the operator @xmath73 contains the anomalous tgc @xmath75 , called @xmath76 in the parameterization of @xcite .
any effect on this coupling , if found to be comparable in size to other leading corrections to the sm , would represent a smoking gun of non - linearity in the ewsb sector .
+ the key to understand how this kind of effect arises is the adimensionality of the @xmath77 matrix , which ensures that the gb contributions do not exhibit any scale suppression .
this is in contrast with the linear description , where the light @xmath10 and the three sm gbs are encoded into the scalar doublet @xmath1 , with mass dimension one : in that case any insertion of @xmath1 pays the price of a suppression factor @xmath78 .
+ current limits on the coupling @xmath76 originate from its impact on radiative corrections to @xmath27 physics , studied at lep .
however , the lhc has the potential to improve these bounds : the study presented in @xcite , based on the kinematical analysis of the process @xmath79 , shows that with a luminosity of 300 fb@xmath80 at a c.o.m .
energy of 14 tev it is possible to measure @xmath76 at a level of precision comparable to that of the current constraints on dimension 6 linear operators .
we now particularize the discussion of the previous sections to the case in which a specific composite higgs model is assumed , _
i.e. _ in which two specific groups @xmath6 and @xmath7 are chosen , such that the higgs particle and the three gbs of the sm arise as the gbs of the spontaneous breaking @xmath81 . in this kind of scenario ,
the gbs have a characteristic scale usually denoted by @xmath82 that satisfies @xmath83 @xcite . at scales above @xmath82
, the model can be described by a high - energy effective lagrangian @xmath84 , manifestly invariant under the complete group @xmath6 , while the low - energy ( below @xmath82 ) effects can always be expressed in terms of the effective operators defined in eq .
, whose coefficients shall satisfy some model - dependent constraints .
studying the connection between @xmath84 and @xmath85 , it is possible to identify which is , for each composite higgs realization , the predicted pattern in the coefficients of the low - energy effective chiral operators , an information that may indeed be very valuable when trying to unveil the origin of ewsb .
the construction of the high - energy effective lagrangian for a completely generic symmetric and @xmath39 any generator of the coset and of the preserved subgroup respectively , it is defined by the schematic constraints @xmath86\propto t , \quad[t , t]\propto x , \quad[x , t]\propto x$ ] . ] coset is reported in detail in ref .
@xcite , together with the discussion of three specific composite higgs realizations : the original @xmath87 georgi - kaplan model @xcite , the minimal intrinsically custodial - preserving @xmath88 model @xcite and the minimal intrinsically custodial - breaking @xmath89 model , where by custodial breaking we mean sources of breaking other than those resulting from gauging the sm subgroup .
here below we summarize the main results . following the general ccwz construction @xcite and under the assumption of a symmetric coset @xmath9 , the gbs @xmath90 arising from the spontaneous breaking @xmath81
can be collectively described by the field @xmath91 , being @xmath92 the generators of the coset @xmath9 .
the gauge fields can be introduced as @xmath93 where @xmath94 and @xmath95 denote the embedding in @xmath6 of the @xmath96 generators .
the most general lagrangian describing the interactions of the gbs of a non - linear realization of the symmetric coset @xmath9 and of the sm gauge fields can then be written as @xmath97 where the sum extends over a complete basis of bosonic , cp even , operators @xmath98 with at most four derivatives , constructed out of the structures @xmath99 and @xmath100 . such a basis is reported in eq . ( 3.23 ) of ref .
@xcite .
remarkably , the effective lagrangian at high energy contains at most ten arbitrary coefficients @xmath101 , which govern the projection of @xmath84 into @xmath102 defined in eq . .
strong relations are therefore predicted among the many low - energy coefficients of @xmath85 . upon specifying the choice of the coset @xmath9
, the number of free parameters can be further reduced , due to peculiar relations among the generators of @xmath6 : as an example , the model @xmath88 allows for only eight independent high - energy operators . for the complete result of the matching , inclusive of the three models considered , we refer the reader to tables 1 and 2 of ref .
here we only give the illustrative example of the operator @xmath103\right)\,,\ ] ] that for both the @xmath87 and @xmath88 models decomposes as @xmath104 , with the dependence on the higgs field specified by the functions @xmath105=\\
= & \frac{\xi}{4}\left[1 + 2\sqrt{1-\frac{\xi}{4}}\frac{h}{v}+\left(1-\frac{\xi}{2}\right)\frac{h^2}{v^2}+{\mathcal{o}}(h^3/v^3)\right]\ , , \end{split}\ ] ] where @xmath106 is a model - dependent parameter that quantifies the degree of non - linearity of the theory : for @xmath107 the construction converges to the linear realization . as can be easily seen from the second line of eq . , the dependence on the @xmath10 field does not respect the linear pattern @xmath108 , which is only recovered in the limit @xmath109 .
conversely , the gauge couplings contained in @xmath110 and @xmath111 combine with the same relative weight as in the linear @xmath5 description ( given by @xmath51 , see eq . ) .
this is a general result of the study performed in @xcite that holds for all the three models considered .
a further tantalising outcome of this analysis is the fact that the @xmath112 functions turn out to be the same , up to rescalings of @xmath82 , for all the three models , suggesting that they may be universal to any realistic composite higgs model .
this feature may prove very relevant for the analysis of experimental data , as it provides a precise and almost model - independent signature of composite higgs models .
more recently , the same subject was explored by ref .
@xcite , that drew similar conclusions .
my work is supported by an esr contract of the european union network fp7 itn invisibles ( marie curie actions , pitn - ga-2011 - 289442 ) .
i also acknowledge partial support of the spanish mineco s `` centro de excelencia severo ochoa '' programme under grant sev-2012 - 0249 .
and i thank the organizers of the la thuile conference for the kind invitation and for their efforts in organizing this enjoyable meeting . | depending on whether electroweak physics beyond the standard model is based on a linear or on a non - linear implementation of the electroweak symmetry breaking , a linear or a chiral effective lagrangian is more appropriate . in this talk ,
the main low - energy signals that allow to recognize whether the observed `` higgs '' scalar is a dynamical ( composite ) particle or rather an elementary one are presented , in a model - independent way .
the patterns of effective couplings produced upon the assumption of specific composite higgs models are also discussed . |
hierarchical models and observations suggest that galaxy mergers and interactions play a key role in galaxy assembly and star formation , but to what extent is still unclear .
studies of gas - rich mergers in the local universe ( e.g. , antennae ; see * ? ? ?
* ) and n - body simulations @xcite have revealed fundamental signatures of the galaxy merger process , including tidal tails , multiple nuclei , and violent bursts of star formation .
while interaction - induced star formation is thought to be primarily responsible for ultra luminous infrared galaxies ( ulirgs , which have @xmath13 ) both locally and at high redshift @xcite , luminous infrared galaxies ( lirgs , @xmath14 ) appear to have multiple driving mechanisms , merger - induced star formation being only one .
luminous infrared ( ir ) galaxies are thought to be the dominant producers of the cosmic infrared background ( cirb ) , and major contributors to the evolution of the cosmic star formation rate ( csfr ) of galaxies , especially at @xmath15 @xcite .
the rapid decline from @xmath8 of the csfr density has been linked to a decline in the merger rate .
however , recent close pair studies have suggested that the merger rate has remained fairly constant from @xmath8 @xcite , and at @xmath15 the ir population is dominated by morphologically normal galaxies @xcite .
the combination of these two results suggest that the bulk of star formation at @xmath8 is not driven by major mergers .
however it must be noted that different merger selection criteria probe different stages of the merger process .
quantitative measurements of galaxy asymmetry ( @xcite ) are more likely to probe later stages , while early stage mergers can be identified by carefully searching for close companions .
there should be some overlap between these techniques if galaxy pairs are close enough to have induced strong tidal interactions , but galaxies in pairs could also have normal morphologies , hence if early stage mergers are not considered , the impact interactions / merging have will be underestimated .
traditionally , close pair studies have been carried out in the optical / near - ir @xcite .
however recent investigations have begun to explore the mid - ir properties ( star formation ) of galaxy pairs , finding a mid - ir enhancement in pairs separated by less then ten s of kpc s @xcite .
the amount of ir luminosity stemming from individual processes ( star formation or fueling an agn ) in interacting pairs and mergers still remains open . to investigate this question
we have conducted a study of the frequency of mips 24 detected , and undetected close optical galaxy pairs and morphologically defined mergers in the _ spitzer _ first look survey ( fls ) .
we find that the fraction of 24 detected , optically selected close pairs and mergers increases with redshift , and are important contributors to the ir luminosity and star formation rate density at @xmath8 . in the discussion that follows
, any calculation requiring cosmology assumes @xmath16=0.3 , @xmath17=0.70 , and h@xmath18=70kms@xmath19mpc@xmath19 .
the _ spitzer _ extragalatic component of the fls is a 3.7 @xmath20 region centered around r.a.=@xmath21 , decl.=@xmath22 .
observations of this field were taken using all four infrared array camera ( irac ) channels ( fazio et al .
2004 ) and three multiband imaging photometer ( mips ) bands ( rieke et al . 2004 ) .
additional ground base images in u*,g from cfht s megacam @xcite , g , i data from palomar 200 " lfc and noao 4-m r and k band ( fadda et al .
2004 ; glassman et al .
2006 in prep ) have also been obtained .
this work focuses on the 0.12 @xmath20 acs - hst f814w imaging of the verification strip , which has 3@xmath23 depths in mips 24 of 0.1mjy .
object detection and photometry were performed using sextractor ( bertin & arnouts 1996 ) .
particular care was taken to ensure accurate de - blending of galaxies in close proximity to one another , while avoiding detections of substructure within a single galaxy , consistent with other reductions of hst imaging with close galaxy pairs in mind ( patton et al . 2005 ) .
there were @xmath359,000 sources extracted within the @xmath24 band ( hereafter extracted magnitudes referred to as @xmath25 ) .
we compared our number counts to those from the hubble deep field ( hdf ) north and south and determined a limiting magnitude of @xmath2627.4 . using the full mips catalog from the fls we selected 24sources within the area covered by the acs imaging ( @xmath30.12 @xmath20 ) . in order to correlate the mips objects with those identified in the optical we first cross - identified sources from the mips 24 sample to the irac catalog using a tolerance radius of 2.0@xmath27
this choice was primarily motivated by the fwhm of the mips 24 ( psf@xmath3@xmath28 ) and confirmed by visual inspection .
we then cross - correlated the irac / mips catalog to the acs sample which we band merged with u * , g and @xmath29 requiring a positional agreement of @xmath30@xmath31 .
when multiple counterparts were identified , we selected the closest object . ultimately we found 1155 acs sources also detected by irac and mips at 24 .
the redshifts used in this study were determined exclusively from optical spectroscopy .
they were obtained by cross - correlating the acs sample , limited to @xmath3226.5 ( @xmath33@xmath329,000 ) with various fls spectroscopic datasets .
the vast majority of the included redshifts ( @xmath3497% ) were obtained with the deep imaging multi - object spectrograph ( deimos ) on the w.m .
keck ii 10-m telescope ; however , the final sample also included a few redshifts based on sloan digitized sky survey ( sdss ) and wiyn hydra / mos ( marleau et al .
2006 in prep ) spectra .
galaxies in the fls verification region were targeted for spectroscopic follow - up during two deimos campaigns that bracketed spitzer s launch .
the selection criteria for these campaigns are summarized below .
for the 2003 pre - launch campaign , targets were selected based on nir ( @xmath35 ) and optical ( @xmath36 ) colors .
the primary sample included sources with @[email protected] , @xmath3819.0 and a @xmath36 color selection that restricted the numbers of low redshift ( @[email protected] ) sources .
for the 2004 post - launch campaign , a purely 24 selected sample ( @xmath40@xmath41120@xmath42jy ) was targeted for follow - up .
the combined @xmath25 distribution of targeted and detected sources is shown in figure [ fig : magdist ] ( _ top _ ) along with the cumulative redshift identification efficiency ( _ bottom _ ) . the overall spectroscopic completeness ( defined here as the fraction of targeted sources with high quality redshifts )
is @xmath370% for the full sample and @xmath380% for sources with @xmath4325.0 . for a more detailed description of the observing strategy , primary selection criteria and the overall flux and redshift distributions
see @xcite .
since we were exploring the mid - ir properties of galaxies no optical limit was imposed , instead an ir luminosity cut ( @xmath44 or @xmath45 ) was used , so that a fair comparison could be made at different redshifts . the absolute b magnitude ( @xmath46 )
distribution of the mips spectroscopic sample between 0.5@xmath37@[email protected] probes -21@xmath30@xmath46@xmath30 - 19 fairly uniformly and we are not strongly biased at higher redshifts .
cross - correlation of the band - merged photometric catalogs with the redshift samples results in a data set of 476 sources with @xmath47 and @xmath0@xmath39@xmath1 . of those , 245 ( 51% ) are mips 24-detected with a measured @xmath48 .
the remaining 231 ( 49% ) are non - detected at 24 . despite the fact that the mips and non - mips galaxies were selected slightly differently , the resultant colors of objects with spectroscopic redshifts have uniform color properties ( figure [ fig : colors ] ) .
to properly constrain the role interactions and mergers play in galaxy evolution , all stages of the process must be considered .
typically , merger history analyses utilize either pair or structural methods .
galaxies in pairs are pre - mergers , or systems undergoing interactions , while morphological or structural methods find galaxies that have already undergone a merger and are dynamically relaxing .
when discussing the pair fraction , and inferred merger rates ( as defined in section 4 ) , it must be noted that these measurements are highly dependent on the techniques and selection criteria used to identify ongoing mergers , especially for galaxies at high redshifts .
the first method , which we will call the close pair method " , is to count the number of galaxy pairs within some projected separation , @xmath49 , and magnitude difference ( @xmath50 ) .
if we assume that these systems will merge within a given time - scale due to dynamical friction , we can determine the merger rate .
although not all pairs will merge , they can potentially trigger star formation through gravitational interactions .
an alternative is to select merging systems based on morphological indicators either by overall appearance @xcite or computational measurements such as asymmetry ( a ) , and clumpiness ( s ) of a system @xcite , or gini coefficient ( g ) , and @xmath51 parameters @xcite . due to the comparatively limited spatial resolution of _ spitzer _ ( compared with optical imaging ) ,
seaching for close galaxy pairs or morphological signatures of interaction at mid - ir wavelengths is currently restricted to the nearby universe .
however , we can correlate optically - selected pairs / mergers with global mid - ir properties and investigate the ir activity in these systems out to high redshifts .
we applied the close pairs technique to identify the average number of close companions per galaxy , hereafter @xmath52 .
this measurement is similar in nature to the pair fraction when there are infrequent triples or higher order n - tuples .
since this is the case here , @xmath52 will be occasionally referred to as the pair fraction .
companions were selected using a standard operational close pair definition of 5@xmath53kpc@xmath54kpc , and an optical magnitude difference ( @xmath50 ) @xmath55 compared to the host galaxy , to select nearly equal mass major mergers . the term host " or
primary " galaxy are both used to reference the pair member with a measured redshift .
the inner radius is applied to avoid detection of substructure within a galaxy , while the outer @xmath56kpc limit represents the radius within which satellites are expected to strongly interact with the halo of the host and merge within 0.5 - 0.9 gyrs ( patton et al .
1997 ; conselice et al 2003 ) .
we find 87 close pairs out of 476 galaxies which fulfill these criteria ( see table 1 ) . to study the fraction of ir - bright galaxies in pairs , we split the pair sample into two sub - sets : those which were detected and those undetected with mips at 24 down to the flux limits of our survey ( 0.1 mjy ) .
figure [ fig : stamps ] shows a subset of close pairs ( both detected and undetected at 24 ) with mips contours . due to the small separations of close pairs ( 20@xmath53kpc corresponds to 3.6 at @xmath8 ) relative to the beam of the mips 24 images ( fwhm @xmath36 ) , there are a few instances ( 5 ) where only a single 24 detection is found centered between the pair members ( see middle left image in figure [ fig : stamps ] ) . in these cases
we assume all 24 flux is coming from the primary galaxy .
since we have redshift information for only the primary galaxy and not the companions we need to consider what fraction of these close pairs are a result of random projection effects .
a field correction was determined using two separate methods to account for these close pairs .
the first assumed the same optical magnitude and redshift distributions independently for both the detected and undetected 24 samples , while the positions were randomized .
the close pair algorithm was applied to 50 realizations of these mock catalogs and the average @xmath52 for each redshift bin was taken to be the pair fraction expected from random .
this assumes the absence of clustering .
we investigated the environments of 24 detected and undetected objects on scales of @xmath57@xmath5820@xmath53kpc , and found them to be comparable , confirming that the increase in pair - fraction of the 24 detected pairs is not because they preferentially lie in clusters . on the other hand
, there is a weak indication that galaxies detected at 24 are more likely to lie in small groups . since such groups may , in some cases , be physical associations , we count such cases as separate pairs . however , the number of these cases is small , and does not influence our results in any significant way .
the second method utilizes the @xmath25 magnitude distribution of the full photometric catalog ( @xmath359,000 sources ) , and determines the average number of companions , within 1.5 mags ( @xmath25 ) , normalized to the area covered by 5@xmath53kpc@xmath54kpc .
the results obtained from the two field correction methods agreed within @xmath59 , which is negligible compared to the uncertainly in @xmath52 .
the average of the two methods was taken to be the final field correction .
both the pair catalog and randomly generated catalogs were visually inspected for false pairs due to single galaxies being broken up into multiple components in the source extraction phase , or contaminating stars in the photometric catalog , and were removed . the field - corrected optical pair fractions for the 24 detected and undetected sub - samples are presented in figure [ fig : pfrac ] and table 1 .
errors are computed using the jackknife technique @xcite , e.g. given a sample of @xmath33 galaxies the variance is given by [ ( @xmath60@xmath61 . the partial standard deviations , @xmath62 , are computed for each object by taking the difference between @xmath52 , the quantity being measured and the same quantity with the ith galaxy removed , @xmath63 , such that @xmath64 . to allow a more direct comparison to be made between the generally lower - luminosity low - z pairs , and those at higher redshift
, we derived pair fractions for mips detected galaxies with an @xmath65 ( approximately the ir luminosity of the famous antenna galaxies ) . in this way
we ensure that the sub - luminous galaxies do not strongly influence the pair fractions in the lowest redshift bin .
the derived @xmath52 for 24 detected close pairs is @xmath66 at @xmath67 and increases to @xmath68 at z@xmath31 .
in contrast , close pairs with no 24@xmath42 m detection show no increase with redshift and have pair fractions consistent with zero at all redshifts .
the higher pair fraction of mips bright sources is marginally significant due to the small number of sources in the highest redshift bin , more mir selected samples between @xmath691 - 1.5 are required to strengthen our findings .
cccccc & & 24 detected & & + 0.2 - 0.5 & 32 & 0.188 ( 6 ) & 0.078 ( 2.5 ) & 0.110 @xmath70 & 0.83 + 0.5 - 0.80 & 82 & 0.171 ( 14 ) & 0.057 ( 4.7 ) & 0.114 @xmath71 & 0.93 + 0.80 - 1.0 & 82 & 0.122 ( 10 ) & 0.029 ( 2.4 ) & 0.093 @xmath72 & 0.90 + 1.0 - 1.3 & 49 & 0.429 ( 21 ) & 0.182 ( 8.9 ) & 0.247 @xmath73 & 0.67 + & & 24 undetected & & + 0.2 - 0.5 & 44 & 0.136 ( 6 ) & 0.102 ( 4.5 ) & 0.034@xmath74 & 1.00 + 0.5 - 0.80 & 76 & 0.132 ( 10 ) & 0.134 ( 10.2 ) & 0@xmath75 & 1.00 + 0.80 - 1.0 & 56 & 0.214 ( 12 ) & 0.193 ( 10.8 ) & 0.021@xmath76 & 0.83 + 1.0 - 1.3 & 55 & 0.145 ( 8) & 0.180 ( 9.9 ) & 0 @xmath77 & 0.75 + we would like to be able to rule out the possibility that @xmath52 is biased by the brightest ir sources at @xmath78 , since merger fractions change as a function of luminosity and mass @xcite . to address this we placed a higher ir luminosity limit ( @xmath79 ) on the sample , so that at @xmath15 the same populations were being probed ( optically we are probing -22@xmath58@xmath46@xmath58 - 19 ) .
we still find an increase in @xmath52 from the lower ( @xmath80 ) to the higher ( @xmath78 ) redshift bins of similar magnitude compared to when the lower ir limit ( @xmath44 ) was used .
therefore the increase in @xmath52 found at @xmath78 is likely not a result of merely probing brighter ir systems but rather due to a physical increase in the merge rate for the 24 population , however deeper 24 imaging and spectroscopy are required to confirm this . when we consider the averaged pair fraction over @xmath81 for the 24 detected sample we find that galaxies above a flux limit of @xmath82mjy are five times more likely to be in a close galaxy pair , than those below this limit . to explore the structural components of galaxies in our sample we used the cas ( concentration , asymmetry , clumpiness ) quantitative classification system @xcite , and visual classifications . to measure the merger fraction using structural classifications we visually inspected the full 24 detected spectroscopic catalog with the following groupings : early type ( e , s0 ) , mid - types ( sa - sb ) , late - types ( sc - irr ) , compact systems , disturbed disks , and mergers .
the methodology for carrying out this classification is described in detail in @xcite .
basically , each galaxy was viewed on a computer screen and classified into one of our types .
overall we find that @xmath83 of 24 detected galaxies are disks , which is consistent with @xcite , while @xmath84 are merging systems and @xmath85 were classified as disturbed disks and are possible minor mergers .
a fraction of the disk - dominated objects do show some visual signs of a morphological disturbance , or are in a pair , as we will discuss later in this paper .
galaxies undergoing a major merger event can also generally be identified by their large asymmetries in the rest frame optical @xcite .
we defined a major merger as a galaxy having an asymmetry ( a ) @xmath86 and @xmath87 ( see figure [ fig : stamps2 ] for examples ) .
this limit has been shown to be a clean way to find galaxy mergers , without significant contamination from non - merging galaxies @xcite .
figure [ fig : mfract ] shows how the merger fraction for cas defined mergers evolves as a function of redshift for both 24 detected objects ( top panel ) and lirg / ulirg galaxies ( bottom panel ) . as with
the 24 detected close pair sample there is an elevated merger fraction compared to other works @xcite in which no 24 flux limit was imposed , and a slight indication of evolution with redshift , but it is statistically consistent with @xmath88 ( dashed line ) , where @xmath89 is the slope of a power - law of form @xmath90 later discussed in @xmath91 .
we also performed a cas analysis of our close pairs sample which revealed that 24 detected pairs are notably more asymmetric than the undetected - mips close pairs ( figure [ fig : cas ] ) , suggesting that interactions and collisions may play a role in their ir activity .
if the 24 detected close pairs were generally of a different morphological classification than those pairs undetected at 24 the discrepancy in the asymmetries could be explained . to address this issue each close pair
was visually inspected and classified by four of the authors to be either disk or bulge - dominated .
we find that 81% of the 24 pairs have disk morphologies while @xmath92 of the undetected 24 hosts were also disk dominated , hence the discrepancy between the asymmetries of the two groups is not caused by classification differences , but rather is a physical effect .
one of the goals of studying mergers and interactions is to determine how the galaxy merger rate evolves with redshift .
most studies of galaxy mergers involve determining the merger fraction , yet the merger rate , which is defined as the number of galaxies merging per unit time per unit volume , is a more physical quantity that can be used to determine the full merger history .
the rate in which galaxies merge also affects the mass function of galaxies , and is likely linked to the cosmic star formation rate . since we are considering a very broad range in the merger process , from early - stage or pre - mergers selected via close galaxy pairs , and later - stage mergers chosen based on morphological criteria , we must be careful when determining their respective merger rates , as the time - scales for these processes are all different .
there are two variations of the merger rate definition .
the first is the number of mergers that a galaxy will undergo per unit time ( @xmath93 ) , and the second is the total number of mergers taking place per unit time per unit co - moving volume ( @xmath94 )
. since we are primarily interested in mergers which are also mid - ir bright systems we will have to restrict ourselves to measuring @xmath93 because the evolution of the 24 luminosity function with redshift is currently not well constrained , and our redshifts are not complete enough to reconstruct this evolution . in order to determine @xmath93 we need to identify systems which are destined to merge .
we have approached this measurement from three different perspectives , close pairs to select pre - mergers or interactions , visual inspection to select interactions after the first passage , and late stage mergers , as well as cas criteria which quantitatively selects for later stage mergers . by combining information about the number of ongoing mergers ( @xmath95 ) and the time - scales , ( @xmath96 ) on which they will undergo said merger
, one can estimate an overall merger rate @xmath97 .
each method of identifying mergers / interactions is capturing a different snapshot of the merger process , each with different merger timescales .
the value of @xmath52 is directly proportional to the number of mergers per galaxy ( @xmath95 ) , such that @xmath98 ( @xmath99 is a constant relating to the number of mergers per galaxy ) .
hence , the merger rate detemined using close galaxy pairs is given by @xmath100 .
the value of @xmath99 depends on the nature of the merging systems under consideration . if one were to identify a pure set of galaxy pairs each consisting of one companion undergoing a merger , then @xmath101 . in our case
it exclusively accounts for close pairs which are in doubles and perhaps higher order n - tuples .
our definition of @xmath99 differs by a factor two from @xcite which in this instance would have @xmath102 since they have redshifts for both pair members and one merger is made up of two companions .
we have redshift information for only one pair member , therefore one merger is made up of a primary and one companion . the merger rate equation for merging galaxies selected by visual classification and cas parameters
is simply @xmath103 , where @xmath104 is the galaxy merger fraction . before the merger fraction can be used to calculate the merger rate we need to understand the time - scale in which a merger occurs .
each technique of identifying mergers has a different time - scale since each is sensitive to a different interval of the merger process .
there are two main methods that have been used to estimate the time - scale of a merger : dynamical friction arguments , and n - body models .
the details of these methods are beyond the scope of this paper but see @xcite and @xcite for a review .
we take the average merger time - scale for a set of close companions of roughly equal mass to merge as @xmath30.5 gyr@xmath105 , derived from dynamical arguments @xcite .
@xcite showed through n - body simulations that visual classification selects on - going mergers over a longer time - scale ( 1.0 gyr@xmath105 ) since the human eye detects both early and later stage mergers , while the asymmetry of a galaxy is sensitive to 0.41 gyr@xmath106 @xcite of the merger sequence .
within the past two decades numerous studies have been performed to estimate the evolution of the galaxy merger fraction , using both the close pair technique @xcite and morphological parameters @xcite .
evolution in the galaxy merger fraction is often parameterized by a power - law of form @xmath90 , and has yielded a wide range of results , spanning 0@xmath107@xmath89@xmath1075 .
the large spread in values is in part due to the different selection criteria used to identify merging systems and biases from optical contamination or redshift completeness .
@xcite considered these biases and demonstrated that most results to that date were consistent with their estimate of @xmath108 .
recently , optical and near - ir close pair studies @xcite have derived merger fractions with little redshift evolution ( @xmath89@xmath31 ) , as have some morphological studies using ( @xmath109 ) , and @xmath51 @xcite .
when we consider all the close pairs identified in our sample , both those detected at 24 and not , we find a merger fraction and rate consistent with recent studies showing little redshift evolution .
however , when we separate the pair sample into systems with a 24 detection above 0.1 mjy , and those below it , we do see a stronger evolution of @xmath52 with redshift , ( recall that @xmath110 ) and therefore also in the merger fraction and rate ( figure [ fig : mr ] ) .
similarly , visually classified mergers and those identified via asymmetry levels ( @xmath111 ) using the cas parameters , also show redshift evolution in the merger fraction and rate . the merger fraction computed using the different methods are in good agreement when normalized by their respective time - scales , reinforcing the idea that we are probing different phases of the merger process . considering all three merger selection techniques we find the best fit of the merger rate parameterized by @xmath112 to be @xmath113 , 2.12 @xmath114 , with a reduced @xmath115=0.39 .
this result suggests that when one considers a sample of close galaxy pairs solely on their optical fluxes , brighter than @xmath116 - 19 , little evolution of the merger rate with redshift is found .
however close pairs emitting 24 flux exhibit an increase in the merger rate with redshift .
the infrared luminosity limit ( @xmath117 ) imposed on the close pairs and mergers allows us to primarily probe systems in a lirg / ulirg phase at @xmath118 ( see next section for details ) .
the increase of the merger fraction and rate of this population of galaxies coupled with the fact that lirg / ulirg galaxies dominate the sfr density at @xmath15 @xcite suggests that merging does in fact play an increasingly important role in star formation out to @xmath8 .
one way to quantify the role merging galaxies play in triggering star formation is to investigate their contribution to ir luminosity densities . infrared luminosities ( 8 - 1000 ) were calculated utilizing the 24 fluxes and two different template methods : @xcite in a similar manner as @xcite for the full mips 24 spectroscopic sample ( figure [ fig : lirall ] ) .
mips pairs and mergers share a similar luminosity distribution to 24 bright field galaxies , although red - agn seem generally more luminous which is in part due to template mismatches @xcite .
the @xmath119 of a galaxy is a combined measure of the reprocessed uv photons intercepted by dust from massive young stars and agn .
therefore to investigate the contribution an interacting or merging galaxy makes towards the total @xmath119 density from star formation alone we must first remove agn from our sample . due to the nonuniform rest - frame spectral coverage of our sample we rely on the four - band irac color selection used by @xcite to identify and remove agn candidates ( figure [ fig : agn ] ) . over the modest redshift range of our sample , this method is still effective at separating ir - warm agn from starburst systems .
we find an agn contamination rate of @xmath120 for the full 24 sample , while @xmath121 of the hosts in a pair or merger were characterized as agn .
with agn candidate objects removed we can infer the contribution to the @xmath119 density from star formation coming from 24 galaxies in a interaction / merger as a function of redshift .
we derive the number of statistically real " galaxy pairs from our pair fraction result at each redshift interval and determine the total @xmath119 density from close pairs which is in turn divided by the @xmath119 density from the whole sample .
we find that paired galaxies ( @xmath117 ) are responsible for @xmath122 of the ir background stemming from star formation at @xmath8 .
since we only know the redshift of the host galaxy we select real " close pairs in a statistical sense , and derive error bars for the close pairs contribution by the spread of 50 realizations of the @xmath119 density from different combinations of 24 galaxy pairs .
we also applied this analysis to cas and visually classified mergers , which make up an additional @xmath312% , and @xmath322% of the ir luminosity density respectively .
naturally there is a some overlap in mergers identified through close pair criteria and morphological parameters , since interacting pairs can exhibit tidal tails and asymmetric structures , causing them to also be identified morphologically as mergers .
we found that @xmath123 of cas defined mergers were also in a close pair , and 31% of visually identified mergers were also classified by cas as merging . in cases where a merging system was identified using multiple techniques
it s contribution was only counted once .
for example , if a merger identified morphologically ( either through cas or visual inspection ) is also in a close pair it is removed from the morphological merger catalog , or if a cas merger is also identified visually the merger is removed from the visual merger catalog .
this insures that no close pair or merger is counted more than once when deriving the contribution from interactions and mergers to the ir luminosity density .
the combination of these three merger selection techniques identifies a large range in the merger process , from pre - merger to late stage mergers , implying that @xmath360% of the infrared luminosity density at @xmath39@xmath31 can be attributed to galaxies involved in some stage of a major merger ( figure [ fig : lumden ] ) .
the remaining @xmath340% of the ir background from lirg / ulirgs is likely to predominately come from active , isolated gas - rich star - forming spirals , with some contribution from minor mergers .
if we exclude visually classified mergers the close pair / merger contribution to the ir density is @xmath338% , in good agreement with @xcite who estimate a moderate contribution from interacting and merging systems of @xmath5836% .
it must be noted however that neither @xcite or this work have considered the contribution from minor mergers and are therefore lower limits .
an important and highly debated question is : how important are galaxy mergers in understanding the dramatic decline of the cosmic sfr density from @xmath8 to the present day ?
it has been well established that mergers and interactions can induce violent bursts of star formation @xcite .
so to investigate this contribution we derived the sfr for our 24 detected close pairs and mergers , using their @xmath119 .
the infrared luminosity of a galaxy is a star formation rate tracer which is unaffected by the extinction of dust .
the dominant heat sources of most dusty , high - opacity systems such as lirgs and starbursts is stellar radiation from young stars . in these types of systems
the @xmath119 can be converted into a sfr using the calibration of @xcite , @xmath124 , where @xmath119 is the integrated luminosity from 8 - 1000 as determined in section 6.0 .
we estimated the contribution mergers and interactions above @xmath117 make to the sfr density at @xmath51 in two ways .
the first is simply to consider their contribution to the @xmath119 density which is a star formation tracer .
section 6.0 determined that mergers and interactions at z@xmath31 ( above @xmath117 ) are responsible for 40 - 60% of the ir luminosity density . using the results of @xcite which showed that @xmath1250.7 lirgs produce @xmath126 of the star formation rate density
, we can infer that mergers and interactions in lirg / ulirg phase would be responsible for @xmath127 ( @xmath128 ) of the sfr density at @xmath51 , since ir activity traces dusty star formation .
the second more detailed approach utilizes the sfr density directly arising from our sample of mergers and interactions . at 1.0@xmath30@[email protected]
we find that 59% ( 12 close pairs , and 17 later stage mergers ) of galaxies detected at 24 are involved in some stage of an interaction or merger . in this same redshift
range our sample is insensitive to galaxies with ir luminosities @xmath30@xmath129 . to correct for this we derived a scaling factor ( @xmath37 ) simply by comparing the number of observed objects of a given @xmath119 in a specific redshift range to the number expected from models @xcite .
however to go any further we must assume that our spectroscopic sample is representative of this population at @xmath31 , and by all accounts this appears to be true .
using the derived pair fraction we can then infer the total number of major mergers and interactions occurring ( fulfilling our criteria ) in a given volume and @xmath119 limit .
the lower limit of the sfr density at @xmath51 from merging and interacting galaxies is found to be 0.066 @xmath130 . using the extinction corrected
lilly - madau " plot ( 0.1585 @xmath130 at @xmath691 ) @xcite we find that mergers and interactions are responsible for at least @xmath131 of the sfr density at @xmath51 ( assuming mergers contribute 60% of the ir density ) .
both approaches are in good agreement , and are only a lower limit , since objects flagged as agn were not considered even though some of their @xmath119 is a result of star formation , and minor mergers which have been shown to also induce bursts of star formation were not included .
these results have interesting implications for the physical mechanisms that drive the decline in the cosmic sfr ( csfr ) density from @xmath8 to present day .
they suggest that when all stages of the merger process are considered ( pre - merger to later stage merger ) major interactions and mergers contribute close to half of the @xmath8 sfr density , and the decline in the number of 24 detected mergers / inteactions is a significant , but perhaps not the primary driver for the decline in the cosmic sfr .
this conclusion differs in interpretation from @xcite , which generally suggest that the evolution of the merger rate is not a significant underlying cause of the decline in the cosmic sfr , but rather a strong decrease in the sfr of morphologically undisturbed spiral galaxies is the dominant mechanism .
their results do not preclude the possibility that their star forming ( undisturbed ) disks " could be in widely separated pairs , and when we only consider quantitatively defined morphological mergers our results are consistent with theirs stressing the importance of considering the merger process in its entirety .
it must also be mentioned that we are probing to higher redshifts than @xcite , which found that major galaxy mergers account for @xmath3030% of the ir luminosity density at @[email protected] , consistent with our findings of 35% at that redshift .
our results also agree that at @xmath50.7 isolated undisturbed spiral galaxies are a primary contributor , however , the influence shifts to interactions and mergers at @[email protected] .
our findings point to an increased importance of mips bright interactions and mergers to the ir luminosity density and sfr density at @[email protected] .
this conclusion is not hampered by the small statistics of the @xmath121 bin .
figure [ fig : lumden ] shows the ir luminosity density contribution from interactions / mergers at @[email protected] to be @xmath337% and 52% at @[email protected] , reinforcing this increasing trend .
using a spectroscopic sample of field galaxies from the acs component of the fls and dividing it into two subsets , those with a 24 detection ( above @xmath82mjy ) and those without ( or below ) we identified optically merging / interacting systems via close pair statistics and morphological methods .
we find that roughly 25% of galaxies emitting at 24 have a close companion at @xmath51 while at @xmath50.5 only @xmath132 are in pairs .
in contrast , those undetected at mips 24 showed a pair fraction consistent with zero at all redshifts ( 0.2@xmath30@[email protected] ) . on average mips
24 galaxies are five times more likely than non - mips sources to have a close companion over 0.2@xmath30@[email protected] . when the samples are combined ( regardless of 24 flux )
we find pair fractions consistent with previous studies @xcite showing little evolution with redshift .
an important and open question is the cause of star formation in lirg galaxies at high - z .
some morphological studies have suggested that since at least half of the lirg galaxies exhibit disk dominated morphologies @xcite at @xmath1250.7 and low non - evolving merger fractions @xcite that the driver of ir activity in high - z lirgs is from on - going star - formation from isolated gas - rich spirals and not merger or interaction induced .
one bias of @xmath133 morphological studies involving the identification of merging / intereacting systems is the limitation of detecting low surface brightness features such as tidal tails caused by close interactions , which can lead to an underestimate of the importance of mergers in the evolution of galaxies at @xmath1341 .
ultimately both close pair and morphological techniques must be applied and considered , to obtain a complete major merger timeline .
our analysis is the first to probe merger rate evolution combining close pairs and later stage mergers while considering the ir activity of these systems .
we find that close pair statistics , visually classified mergers , and those identified via quantitative cas parameters all showed similar evolution in their merger rates .
fitting the merger rate evolution function @xmath135 for 24 detected mergers above @xmath82mjy , we find @xmath136 .
this result agrees with previous claims of an increase ( @xmath137 ) of the merger rate out to @xmath8 @xcite .
however this evolution is not seen when ir faint ( @xmath138mjy ) mergers are included , suggesting that it is the lirg - merger population that is evolving with redshift .
the mid - ir emission of lirgs is indicative of dust enshrouded star formation ( and some agn activity ) , and at @xmath15 they dominate the ir luminosity density and in turn the volume - averaged star formation rate density at @xmath8 .
we estimate that close galaxy pairs are responsible for @xmath139 of the ir luminosity density resulting from star formation at @xmath8 , while later stage mergers contribute @xmath140 .
this implies that 40 - 60% of the infrared luminosity density at @xmath8 can be attributed to galaxies involved in some stage of a major merger , indicating that merger - driven star formation is responsible for 30 - 40% of the star formation density at @xmath8 .
this value is a lower limit since minor mergers and interactions / mergers with an agn were not considered .
ultimately , our findings suggest that interactions and mergers of lirg phase galaxies play an increasingly important role in both the ir luminosity and sfr density from @xmath15 out to @xmath141 , and are vital to our understanding of the evolution and mass assembly of luminous ir galaxies .
we would like to thank h. shim , m. i m , r. chary , and c. borys for their contributions to this work , v. charmandaris for useful suggestions , and the anonymous referee for valuable comments that improved the clarity of the paper .
this work is based on observations made with the _ spitzer observatory _ , which is operated by the jet propulsion laboratory , california institute of technology , under nasa contract 107 .
support for this work was provided in part by the _ spitzer graduate student fellowship _ program and an ontario graduate scholarship in science and technology .
the authors wish to recognize and acknowledge the very significant cultural role and reverence that the summit of mauna kea has always had within the indigenous hawaiian community .
we are most fortunate to have the opportunity to conduct observations from this mountain .
abraham , r. g. , van den bergh , s. , glazebrook , k. , ellis , r. s. , santiago , b. x. , surma , p. , & griffiths , r. e. 1996 , , 107 , 1 abraham , r. g. , tanvir , n. r. , santiago , b. x. , ellis , r. s. , glazebrook , k. , & van den bergh , s. 1996 , , 279 , l47 abraham , r. g. , van den bergh , s. , & nair , p. 2003
, , 588 , 218 barnes , j. e. 2004 , , 350 , 798 barton , e. j. , geller , m. j. , & kenyon , s. j. 2000 , , 530 , 660 bell , e. f. , et al .
2005 , , 625 , 23 bertin , e. , & arnouts , s. 1996 , , 117 , 393 bundy , k. , fukugita , m. , ellis , r. s. , kodama , t. , & conselice , c. j. 2004 , , 601 , l123 burkey , j. m. , keel , w. c. , windhorst , r. a. , & franklin , b. e. 1994 , , 429 , l13 carlberg , r. g. , pritchet , c. j. , & infante , l. 1994 , , 435 , 540 carlberg , r. g. , et al .
2000 , , 532 , l1 cassata , p. , et al . 2005 , , 357 , 903 chary , r. , & elbaz , d. 2001 , , 556 , 562 choi , p. i. , et al .
2006 , , 637 , 227 conselice , c. j. 1997 , , 109 , 1251 conselice , c. j. , bershady , m. a. , & jangren , a. 2000 , , 529 , 886 conselice , c. j. , gallagher , j. s. , iii , & wyse , r. f. g. 2002 , , 123 , 2246 conselice , c. j. 2003 , , 147 , 1 conselice , c. j. , bershady , m. a. , dickinson , m. , & papovich , c. 2003 , , 126 , 1183 conselice , c. j. , chapman , s. c. , & windhorst , r. a. 2003 , , 596 , l5 conselice , c. j. , blackburne , j.a . , &
papovich , c. 2005 , apj , 620 , 564 conselice , c. j. 2006 , , 638 , 686 cox , t.j . , et al .
2006 , , submitted , astro - ph/0503201 dale , d. a. , helou , g. , contursi , a. , silbermann , n. a. , & kolhatkar , s. 2001 , , 549 , 215 dasyra , k. m. , et al . 2006 , , 638 , 745 efron , b. 1981 , biometrika,68 , 589 efron , b. , & tibshirani , r. 1986 , stat .
, 1,54 elbaz , d. , cesarsky , c. j. , chanial , p. , aussel , h. , franceschini , a. , fadda , d. , & chary , r. r. 2002 , , 384 , 848 fadda , d. , jannuzi , b. t. , ford , a. , & storrie - lombardi , l. j. 2004 , , 128 , 1 fadda , d. , et al .
2006 , , 131 , 2859 fazio , g. g. , et al .
2004 , , 154 , 10 kennicutt , r. c. , jr .
1998 , , 36 , 189 lacy , m. , et al .
2004 , , 154 , 166 lacy , m. , et al .
2005 , , 161 , 41 lagache , g. , et al .
2004 , , 154 , 112 lavery , r. j. , remijan , a. , charmandaris , v. , hayes , r. d. , & ring , a. a. 2004 , , 612 , 679 le fvre , o. , et al .
2000 , , 311 , 565 le floch , e. , et al .
2005 , , 632 , 169 lilly , s. j. , le fevre , o. , hammer , f. , & crampton , d. 1996 , , 460 , l1 lin , l. , et al .
2004 , , 617 , l9 lin , l. , et al .
2006 , , accepted , astro - ph/0607272 lotz , j , m. , davis , m. , faber , s. m. , guhathakurta , p. , gwyn , s. , huang , j. , koo , d. c. , le floch , e. et al .
2006 , , submitted .
madau , p. , pozzetti , l. , & dickinson , m. 1998 , , 498 , 106 melbourne , j. , koo , d. c. , & le floch , e. 2005 , , 632 , l65 mihos , j. c. , & hernquist , l. 1996 , , 464 , 641 patton , d. r. , pritchet , c. j. , yee , h. k. c. , ellingson , e. , & carlberg , r. g. 1997 , , 475 , 29 patton , d. r. , carlberg , r. g. , marzke , r. o. , pritchet , c. j. , da costa , l. n. , & pellegrini , p. s. 2000 , , 536 , 153 patton , d. r. , et al .
2002 , , 565 , 208 patton , d. r. , grant , j. k. , simard , l. , pritchet , c. j. , carlberg , r. g. , & borne , k. d. 2005 , , 130 , 2043 rieke , g. h. , et al .
2004 , , 154 , 25 sanders , d. b. , soifer , b. t. , elias , j. h. , madore , b. f. , matthews , k. , neugebauer , g. , & scoville , n. z. 1988 , , 325 , 74 schweizer , f. 1982 , , 252 , 455 thompson , r. i. , weymann , r. j. , & storrie - lombardi , l. j. 2001 , , 546 , 694 shim , h. , i m , m. , pak , s. , choi , p. , fadda , d. , helou , g. , & storrie - lombardi , l. 2006 , , 164 , 435 wolf , c. , et al .
2005 , , 630 , 771 xu , c. k. , sun , y. c. , & he , x. t. 2004 , , 603 , l73 yee , h. k. c. , & ellingson , e. 1995 , , 445 , 37 zepf , s. e. , & koo , d. c. 1989 , , 337 , 34 | by combining the 0.12 square degree f814w _ hubble space telescope _ ( _ hst _ ) and _ spitzer _ mips 24 imaging in the first look survey ( fls ) , we investigate the properties of interacting and merging mid - infrared bright and faint sources at @xmath0z@xmath1 .
we find a marginally significant increase in the pair fraction for mips 24 detected , optically selected close pairs , pair fraction@xmath2 at z@xmath31 , in contrast to @xmath4 at @xmath50.4 , while galaxies below our 24 mips detection limit show a pair fraction consistent with zero at all redshifts . additionally , 24 detected galaxies with fluxes @xmath6mjy are on average five times more likely to be in a close galaxy pair between @xmath0z@xmath1 than galaxies below this flux limit . using the 24 flux to derive the total far - ir luminosity we find that paired galaxies ( early stage mergers ) are responsible for @xmath7 of the ir luminosity density resulting from star formation at @xmath8 while morphologically classified ( late stage ) mergers make up @xmath9
this implies that @xmath10 of the infrared luminosity density and in turn @xmath11 of the star formation rate density at @xmath8 can be attributed to galaxies at some stage of a major merger or interaction .
we argue that , close pairs / mergers in a lirg / ulirg phase become increasingly important contributers to the ir luminosity and star formation rate density of the universe at @xmath120.7 . |
the interference of electron waves scattered by single defects results in an oscillatory dependence of the point contact conductance @xmath0 on the applied voltage @xmath1 this effect originates from quantum interference between the principal wave that is directly transmitted through the contact and the partial wave that is scattered by the contact and the defect or several defects .
such conductance oscillations have been observed in quantum point contacts @xcite and investigated theoretically in the papers ludoph , namir , avotina2,avotina3 . in our previous paper
@xcite the oscillatory voltage dependence of the conductance of a tunnel point contact in the presence of a single point - like defect has been analyzed theoretically and it has been shown that this dependence can be used for the determination of defect positions below a metal surface by means of a scanning tunnelling microscope ( stm ) . in the model of a spherical fermi surface ( fs )
the amplitude of the conductance oscillations is maximal when the contact is placed directly above the defect .
the oscillatory part of the conductance @xmath2 for this situation is proportional to @xmath3where @xmath4 the depth of the defect and @xmath5 and @xmath6 are the fermi wave vector and effective mass of the electrons @xcite
. materials with an almost spherical fs are most suitable for this model . in most metals
the dispersion relation for the charge carriers is a complicated anisotropic function of the momentum .
this leads to anisotropy of the various kinetic characteristics @xcite . particularly , as shown in ref .
kosevich , the current spreading may be strongly anisotropic in the vicinity of a point - contact
. this effect influences the way the point - contact conductance depends on the position of the defect .
for example , in the case of a au(111 ) surface the ` necks ' in the fs should cause a defect to be invisible when probed exactly from above .
qualitatively , the wave function of electrons injected by a point contact for arbitrary fs @xmath7 has been analyzed by a. kosevich @xcite .
he noted that at large distances from the contact the electron wave function for a certain direction @xmath8 is defined by those points on the fs for which the electron group velocity is parallel to @xmath9 unless the entire fs is convex there are several such points .
the amplitude of the wave function depends on the gaussian curvature @xmath10 in these points , which can be convex @xmath11 or concave @xmath12 .
the parts of the fs having different signs of curvature are separated by lines of @xmath13 ( inflection lines ) . in general
there is a continuous set of electron wave vectors for which @xmath13 .
the electron flux in the directions having zero gaussian curvature exceeds the flux in other directions @xcite . electron scattering by defects in metals with an arbitrary fs
can be strongly anisotropic @xcite . generally , the wave function of the electrons scattered by the defect consists of several superimposed waves , which travel with different velocities . in the case of an open constant - energy surface
there are directions along which the electrons can not move at all .
scattering events along those directions occur only if the electron is transferred to a different sheet of the fs @xcite . in this paper
we analyze the effect of anisotropy of the fs to the possibility of determination of the position of a defect below a metal surface by use of a stm .
we show that the amplitude and the period of the conductance oscillations are defined by the local geometry of the fs , namely by those points for which the electron group velocity is directed along the radius vector from the contact to the defect .
general formulas for the wave function and point contact conductance are obtained in sections ii and iii . in sec . iv the asymptotic forms of the wave function and the point contact conductance for large distances of the defect from the contact are found .
the general results are illustrated for two specific models of the fs : an ellipsoid ( sec .
v ) and a corrugated cylinder ( sec .
vi ) . using these models , for which analytical dependencies of the conductance on voltage and defect position can be found , we describe the manifestation of common features of fs geometries to the conductance oscillations : anisotropy of a convex part ( ` bellies ' ) , changing of the curvature ( inflection lines ) and presence of open directions ( ` necks ' ) .
let us consider as a model for our system a nontransparent interface located at @xmath14 separating two metal half - spaces , in which there is an orifice ( contact ) of radius @xmath15 centered at the point @xmath16 .
the potential barrier in the plane @xmath14 is taken to be a delta function , @xmath17 where @xmath18 is a two dimensional vector in the plane of the interface , with @xmath19 .
the function @xmath20 in all points of the plane @xmath14 except in the contact , where @xmath21 .
at the point @xmath22 near the contact in the upper half - space , @xmath23 , a point - like defect is placed . the electron interaction with the defect
is described by the potential @xmath24 which is confined to a small region with a characteristic radius @xmath25 around the point @xmath26 .
it is known that one can obtain an effective schrdinger equation for quasiparticles in a metal from the dispersion relation @xmath27 ( the band structure ) by replacement of the quasimomentum @xmath28 ( below for short we write momentum ) in the function @xmath29 with the momentum operator @xmath30 @xcite .
here we do not specify the specific form of the dependence @xmath31 , except that it satisfies the general condition of point symmetry @xmath32 . for simplicity
we assume that fs has only one sheet ; there is only one zone described by the function @xmath33 in the reduced zone scheme a given momentum @xmath28 identifies a single point within the first brillouin zone . the wave function @xmath34 satisfies the schrdinger equation with an effective hamiltonian @xmath35 , @xmath36 \psi \left ( \mathbf{r}\right ) = d\left ( \mathbf{r}-\mathbf{r}_{0}\right ) \psi \left ( \mathbf{r}\right ) , \label{schrod}\ ] ] where @xmath37 is defined by eq .
( [ u(z ) ] ) , @xmath38 is the applied electrical potential , and @xmath39 is the electron energy .
we consider a large barrier potential @xmath40 in this case the amplitude @xmath41 of the electron wave function passing through the barrier is @xmath42 where @xmath43 and @xmath44 are the @xmath45-components of the velocity @xmath46 of incident electrons ( in ) and electrons specularly reflected by the barrier ( ref ) , respectively . under condition of specular reflection
the energy @xmath47 and the component of the momentum tangential to the interface , @xmath48 at @xmath14 are conserved .
the components of the electron momentum perpendicular to the interface , @xmath49 and @xmath50 are related by the equations , @xmath51 the velocities @xmath43 and @xmath44 have the opposite sign @xmath52 where @xmath53 is a unit vector normal to the interface laying in the half - space of the electron wave under consideration ( @xmath54 for @xmath23 and @xmath55 for @xmath56 ) .
we will assume that the crystallographic axes in half - spaces @xmath57 are identical .
in this case the momenta and velocities for electrons incident on the barrier and for those transmitted through the barrier are equal . in general eq .
( [ pz_ref ] ) may have several solutions , i.e. several specularly reflected states may correspond to an incident state with momentum @xmath58 such reflection is called multichannel specular reflection @xcite .
below we assume that there is only one reflected electron state .
in the limit of a small probability of electron tunnelling through the barrier , @xmath59 the applied voltage drops entirely over the barrier and we take the electric potential to be a step function @xmath60 the reference point of zero electron energy is the bottom of the conduction band in the upper half - space , @xmath61 the conduction band in the lower half - space @xmath56 is shifted by a value @xmath62 we also assume that the applied bias @xmath63 is much smaller than the fermi energy and in solving the schrdinger equation ( [ schrod ] ) we neglect the electric potential @xmath64 equation ( [ schrod ] ) can be solved by using perturbation theory with the small parameter @xmath65 @xcite . in the zeroth approximation in this parameter
we have the problem of an impenetrable partition between two metal half - spaces .
we start by solving for the wave function @xmath66 for a tunnelling point contact of low transparency , @xmath67 without defects ( @xmath68 ) .
the wave function @xmath69 , in zeroth order in the parameter @xmath65 , satisfies the boundary condition @xmath70 at the interface , @xmath71 let us consider an electron wave @xmath72 incident on the junction from the lower half - space @xmath56 , so that @xmath73 . in this half - space
to first approximation in the parameter @xmath41 the solution @xmath74 of the homogeneous schrdinger equation can be written in the form @xcite : @xmath75where the second term , @xmath76 , describes the changes in the reflected wave as a result of transmission through the contact .
the wave function transmitted into the half - space @xmath23 is proportional to the amplitude @xmath41 , @xmath77the function @xmath78 satisfies the condition of continuity and the condition of conservation of probability flow at @xmath79 for small @xmath80 these conditions reduce to @xmath81 in the absence of defects @xmath82 the solution of the schrdinger equation is given by @xcite , @xmath83 where @xmath84@xmath85 are roots of the equation @xmath86 corresponding to waves with velocities @xmath87 and @xmath88 let @xmath89 be a spherically symmetric scattering potential for a point - like defect , with a range @xmath25 that is order of the fermi wave length @xmath90 ( @xmath91 , the maximal value of @xmath92 , when @xmath93 ) . for a point - like defect ( @xmath94 ) , the right hand side in the schrdinger equation ( [ schrod ] ) can be rewritten as @xmath95 @xcite .
this makes it possible to find a solution to eq.([schrod ] ) by means of the green function @xmath96 of the homogeneous equation ( at @xmath97 ) .
the wave function scattered from the defect , @xmath98 , can be expressed in terms of the wave function @xmath99 transmitted into the upper metal half - space , @xmath100where @xmath101because the green function has a singularity at @xmath102 eq .
( [ psi_scat ] ) is correct if the integral @xmath103 ( [ j_p ] ) converges in the point @xmath104 . to proceed with further calculations we assume that the scattering potential is small and use perturbation theory in the interaction with the defect
this implies that we take @xmath105 the wave function solution , that is linear in @xmath80 and with a contribution to first order in @xmath92 , is @xmath106the green function @xmath107 in zeroth approximation in the parameters @xmath92 and @xmath80 should be calculated from the wave functions ( [ psi_0 ] ) , @xmath108 this is the green function of the schrdinger equation in the half - space @xmath23 with hard - wall boundary conditions at @xmath14 . substituting the wave function @xmath109 , eq .
( [ psi_0 ] ) , for @xmath23 in eq.([g_funct ] ) we find , @xmath110for @xmath111 one should make the replacements @xmath112 and @xmath113 in eq .
( [ gz > z0 ] ) ; @xmath114 are given by eq .
( [ pz+/- ] ) . the main contribution to the integral in eq .
( [ psi_approx ] ) comes from a small region near the point @xmath115 .
far from the point @xmath22 ( @xmath116 the solution ( [ psi_approx ] ) takes the form : @xmath117 where @xmath118 is the constant of electron - impurity interaction .
the electrical current @xmath119 can be evaluated from the electron wave functions , @xmath120 , of the system through @xcite , @xmath121 .
\label{current}\]]here@xmath122is the density of probability flow in the @xmath45 direction for the momentum @xmath123 integrated over a plane @xmath124 xconst , @xmath125 is the fermi distribution function , and @xmath126 is the velocity operator , @xmath127 for the definiteness we choose in eq.([current ] ) @xmath128 at low temperatures the tunnel current is due to those electrons in the half - space @xmath129 having an energy between the fermi energy , @xmath130 and @xmath131 , because on the other side of the barrier @xmath23 only states with @xmath132 are available . after performing the integration over a plane at @xmath133 where the wave function ( [ psi0+g ] ) can be used , we find the density flow ( den_flow ) becomes , @xmath134where the mean value at energy @xmath39 is obtained by integration over the surfaces of constant energy in momentum space , @xmath135scaled by the velocity , @xmath136 . in eq .
( flow_int ) , @xmath137 and @xmath138 are given by eq .
( pz+/- ) and @xmath139 is the electron density of states per unit volume . taking into account eq .
( [ flow_int ] ) we can calculate the current - voltage characteristics @xmath119 .
the conductance @xmath0 is the first derivative of the current @xmath119 with @xmath140 and in the limit @xmath141 we obtain , @xmath142after integration over @xmath143 ( [ int_en ] ) , eq.([conduct ] ) should be expanded in the parameter @xmath144
in this section we find the wave function at a large distances from the contact , @xmath145 , and an asymptotic expression for the conductance in the limit of a large distance between the defect and the contact , @xmath146 and a small contact radius , @xmath147 , where @xmath90 is the characteristic electron fermi wave length .
for @xmath148 the function @xmath149 in eq .
( [ f ] ) takes the form @xcite , @xmath150and the wave function ( [ psi+00 ] ) can be written as @xmath151 let us consider the integral @xmath152where @xmath153 is the phase accumulated over the path travelled by the electron between the contact and the point @xmath8 , @xmath154this kind of integrals appear in the expressions for the wave function ( eqs .
( [ phi+0 ] ) , and ( [ gz > z0]),([psi0+g ] ) ) and for the conductance ( eqs .
( [ flow_int]),([conduct ] ) ) . at a large distance , @xmath155 , the exponent under the integral in eq .
( lambda ) is a rapidly oscillating function and the integral can be calculated by the stationary phase method ( see , for example , ref .
fedoriuk ) .
the stationary phase points @xmath156 are defined by the equation , @xmath157with eq .
( [ phase ] ) we find:@xmath158 for @xmath145 the asymptotic value @xmath159 of the integral ( [ lambda ] ) is given by @xmath160 .
\label{lam}\]]here@xmath161is the phase ( [ phase ] ) in a point defined by eq .
( [ st_point2 ] ) , @xmath162 is the gaussian curvature of the surface of constant energy @xmath163 , and @xmath164 is the angle between the vector @xmath8 and the @xmath45 axis . at the stationary phase
points the curvature @xmath165 can be written as @xmath166 _ { \mathbf { p}_{t}=\mathbf{p}_{t}^{\left ( \mathrm{st}\right ) } } , \label{gaus0}\ ] ] where @xmath167 is the algebraic adjunct of the element @xmath168of the inverse mass matrix @xmath169 @xcite ; @xmath170 are components of the unit vector @xmath171 note that for an arbitrary fs @xmath172 in the point @xmath173 depends on the direction of vector @xmath174 .
it is follows from eq .
( [ st_point2 ] ) that the velocity at the stationary phase point @xmath175is parallel to the radius - vector @xmath19 . if the curvature of the fs changes sign , eq .
( [ st_point2 ] ) has more then one solution @xmath176 ( @xmath177 ) . in that case
the value of the integral ( [ lambda ] ) is replaced by a sum over all points @xmath178 , which in the limit of large distances is , @xmath179with @xmath180 given by eq .
( [ lam ] ) for each stationary phase point @xmath181 .
it may also occur that eq .
( [ st_point2 ] ) does not have any solution for given directions of the vector @xmath8 , and the electron can not propagate along these directions .
these two energy surface properties result in complicated patterns of the distribution of the modulus of the wave function : 1 ) for directions for which eq .
( [ st_point2 ] ) has several solutions a quantum interference pattern of the electron waves with different velocities should be observed .
2 ) when eq .
( [ st_point2 ] ) has no solution for the selected direction of the vector @xmath8 classical motion in this direction is forbidden and the wave function is exponentially small . for large values
@xmath182 the asymptotic behaviors of the green function , eq .
( [ gz > z0 ] ) , and of the conductance , eq . ( [ conduct ] ) at @xmath146 , can be found analogous to the evaluation of the integral in eq .
( [ lambda ] ) .
the slowly varying functions of the momentum must be taken in the stationary phase point .
for the partial wave scattered by the defect that is moving towards the interface @xmath137 in eq .
( [ phase ] ) must be replaced by @xmath183 . in this case
the stationary points have a group velocity @xmath184 directed from the point @xmath185 towards the contact . from the central symmetry of the fs , @xmath186
, it follows that the two stationary phase points for the function @xmath187 are antiparallel , @xmath188 . next , we derive an asymptotic expression for the wave function ( [ psi0+g ] ) , with @xmath189 , for a symmetric orientation of the fs with respect to the interface , so that @xmath190 , @xmath191 and @xmath192 , if @xmath193 . under these conditions
the wave function ( 19 ) takes the form , @xmath194with@xmath195 in eq .
( [ psi_asym ] ) the velocity @xmath196 is taken in the stationary phase points @xmath197 corresponding to the directions of the vector with coordinates @xmath198 . we have assumed for the gaussian curvature , eq .
( [ gaus0 ] ) , that @xmath199 in the stationary phase points @xmath200 .
for those points at which @xmath201 the integral ( [ lambda ] ) diverges .
this means that the third derivative of the phase @xmath187 ( [ phase ] ) with respect to @xmath202 must be taken into account . in sec .
v this is done for a model fs having cylindrical symmetry with respect to an open direction .
here we only note that the amplitude of the wave function @xmath203 ( [ psi0_asym ] ) in a direction of zero gaussian curvature is larger than for other directions , and decreases more slowly as compared to the @xmath204 dependence of eq.([lam ] ) .
this results in an enhanced current flow near the cone surface defined by the condition @xmath201 @xcite .
the same effect also appears in the second term of the wave function ( [ psi_asym ] ) , that describes the scattered wave : the amplitude of this partial wave is maximal in the directions of zero gaussian curvature .
if the fs has a flat part , i.e. eq.([st_point2 ] ) holds at all points of a fs region of finite area @xmath205 , the associated electron waves propagate in the metal without decrease of their amplitude @xcite . for the flat part
the dispersion relation @xmath206 can be presented as @xmath207 where @xmath208 is the electron velocity . for such fs the asymptotic value of the integral ( lambda ) is@xmath209
when the distance between the contact and the defect is large , @xmath146 , we obtain the conductance of the tunnel junction , using eq .
( [ flow_int ] ) and the asymptotic expression for the wave function ( [ psi_asym ] ) , @xmath210where @xmath211 is the zero - bias ( @xmath212 ) conductance of the junction without defect , @xmath213 in deriving eq .
( [ del_g ] ) we have assumed that @xmath214 and @xmath215 therefore , all functions of the energy @xmath39 in eq .
( [ del_g ] ) can be taken at @xmath216 , except for the phase @xmath217 .
when @xmath214 , @xmath218and when the product @xmath219 clearly the conductance ( [ g_asym ] ) is an oscillatory function of the voltage @xmath140 .
note that if the inequality @xmath220 is not satisfied , the value of the conductance ( [ g0_arbit ] ) as well as the amplitude of its oscillations depend on @xmath1 the periods of oscillations are defined by the energy dependence of the function @xmath221 and they remain the same for any voltage .
below presenting formulas for the conductance in different cases we do nt expand arguments of oscillatory functions in the parameter @xmath222 the obtained results properly describe the total conductance at @xmath223 and also can be used for the analysis of periods of oscillations at @xmath224 in eq.([g_asym ] ) the @xmath225 denotes the function ( [ lam ] ) , in which @xmath226 equation ( [ g_asym ] ) shows that for a tunnel junction of small size the amplitude and period of conductance oscillations depend on the local geometry fs in those points for which the velocity is directed along the vector @xmath26 . in the case of a convex fs there is only one stationary phase point satisfying eq .
( [ st_point2 ] ) which allows simplifying the expression for the conductance ( [ g_asym ] ) .
the oscillating part of the conductance , @xmath227 , can be written as , @xmath228 _
{ \mathbf{p}_{t}=\mathbf{p}_{t}^{\left ( \mathrm{st}\right ) } } , \label{del_g}\]]where @xmath229 .
the phase of the oscillations in the conductance , @xmath230 , is determined by the phase that the electron accumulates along the trajectory from the contact to the defect and back @xmath231where @xmath232 is the chord connecting the two points on the surface of constant energy for which the velocities are antiparallel and aligned with the vector @xmath185 .
if the direction from the contact to the defect coincides with a direction of electron velocities for a flat part of the fs we should use asymptotic expression ( [ lambda_fl ] ) for the function @xmath233 when calculating the conductance ( [ g_asym ] ) .
for this case the oscillating part of the conductance , @xmath2 , is given by , @xmath234note that @xmath235 . in the case when the fs has a flat part , the amplitude of the conductance oscillations for those special directions does not depend on the distance between the contact and the defect .
in the next sections we shall consider two models of anisotropic fss . the first model , of an ellipsoidal fs , illustrates the main features of the conductance oscillations for metals with a convex fs , i.e. with positive gaussian curvature .
the fs of the second model has the shape of a corrugated cylinder , which illustrates the effects of sign inversion of the curvature and the presence of open directions of the fs .
these two models allow us to obtain dependencies of the conductance on the applied voltage and on the defect position in analytical form .
for an ellipsoidal fs the schrdinger equation ( [ schrod ] ) can , in fact , be solved exactly in the limit @xmath148 and the wave function ( [ psi0+g ] ) and the conductance ( [ conduct ] ) can be found for arbitrary distances between the contact and the defect .
for this fs the dependence of the electron energy @xmath39 on the momentum @xmath28 is given by relation , @xmath236were @xmath237 are the components of the electron momentum @xmath28 , @xmath238 are constants representing the components of the inverse effective mass tensor @xmath169 .
the tensor @xmath169 can be diagonalized to the form @xmath239 so that the momentum - space lengths @xmath240 correspond the semi - axes of the fs ellipsoid @xmath241 . for an ellipsoidal fs .
the shape of the fs is defined by the mass ratios @xmath242=1 , @xmath243 , and the long axis of the ellipsoid is rotated by @xmath244 around the @xmath245axis , away from the @xmath246axis .
the coordinates are measured in units of @xmath247 .
the position of the defect is @xmath248 .
, width=302 ] for the ellipsoidal fs in absence of a defect ( zeroth approximation ) the wave function @xmath249 ( phi+0 ) can be obtained by integration of eq .
( [ phi+0 ] ) over momentum , @xmath250where@xmath251and @xmath252here , @xmath253 are the elements of the mass tensor @xmath254 , which is the inversive to the tensor @xmath169 ( @xmath255 , with @xmath256 the unitary tensor ; @xmath257 ) . the green function ( [ gz > z0 ] ) takes the form @xmath258using eqs .
( [ phi0_ell ] ) and ( [ g_ellips ] ) we obtain the wave function ( [ psi0+g ] ) to first approximation in the strength of the impurity potential .
the modulus of this wave function is illustrated in fig .
fig - ell - abs - phi for a plane normal to the interface passing through the impurity and the contact .
the long axis of the ellipsoid is rotated by @xmath259 around the @xmath245axis , away from the @xmath246axis .
an interference pattern is visible of partial waves reflected from the impurity with those emanating from the contact that is influenced by the anisotropy of the electronic band structure .
the conductance in the limit of low temperatures , @xmath141 , is obtained from eqs .
( [ flow_int]),([conduct ] ) by integration over all directions of the momentum @xmath28 and integration over the space coordinate @xmath260 in a plane @xmath261 ( @xmath262 ) , retaining only terms to first order in @xmath263 ( i.e. ignoring multiple scattering at the impurity site ) , @xmath264 } \frac{1}{\gamma _ { 0}^{4}\left ( \varepsilon _ { \mathrm{f}},\mathbf{r}_{0}\right ) } \times \right .
\label{g(v ) } \\ \left . \left
[ \left ( 1-\frac{1}{\gamma _ { 0}^{2}\left ( \varepsilon _ { \mathrm{f% } } , \mathbf{r}_{0}\right ) } \right ) \sin 2\gamma _ { 0}\left
( \varepsilon _ { % \mathrm{f}}+ev,\mathbf{r}_{0}\right ) + \frac{1}{\gamma _ { 0}\left ( \varepsilon _ { \mathrm{f}},\mathbf{r}_{0}\right ) } \cos 2\gamma _ { 0}\left ( \varepsilon _ { % \mathrm{f}}+ev,\mathbf{r}_{0}\right ) \right ] \right\ } .
\notag\end{gathered}\]]with @xmath265 given by ( ga_el ) .
the amplitude of the conductance oscillations is maximal when @xmath266 is minimal . for a fixed depth @xmath4
this minimum occurs when the defect position @xmath267 is in the point @xmath268 with respect to the point contact at @xmath16 , where @xmath269the minimal value of the phase then becomes , @xmath270the phase @xmath271 corresponds to the extremal value of the chord of the fs in the direction normal to the interface .
@xmath272 in eq .
( [ g(v ) ] ) is the conductance in the absence of a defect @xmath273 : @xmath274 } \sqrt{m_{zz}}}. \label{g_c}\ ] ] fig .
[ fig - ell - deltag ] shows a plot of the normalized conductance , @xmath275 , eq .
( [ g(v ) ] ) , for the contact as a function of the position of the defect , @xmath267 , in the limit of low voltage , @xmath276 .
we find that @xmath277 is an oscillatory function of the defect position that reflects the anisotropy of the fs and the oscillations are largest when the defect is placed in the position @xmath268 , defined by eq .
( [ x0y0 ] ) .
, as a function of the position of the defect @xmath278 in the plane @xmath279 for the same shape and orientation of the ellipsoidal fs as in fig .
[ fig - ell - abs - phi ] .
the coordinates are measured in units @xmath280 and the defect sits at @xmath281 . , width=453 ] for the ellipsoidal model fs the wave function and conductance have been obtained exactly , within the framework of the model . for large distances @xmath282 and @xmath283 they transform into the asymptotic expressions , eqs .
( psi_asym ) , ( [ g_asym ] ) .
we do not present the asymptotic form explicitly but it agrees to within a term proportional to @xmath284 to the exact from , eq.([g(v ) ] ) . in fig .
[ fig - comparision ] we compare the results for the calculations of the conductance by using the exact ( [ g(v ) ] ) and asymptotic ( [ g_asym ] ) expressions .
the figure confirms that for relatively small distances ( fig .
[ fig - comparision](a ) ) the asymptotic formula still qualitatively describes the conductance very well and that for larger distances ( fig .
[ fig - comparision](b ) ) the two results are in a good agreement . the parameters for the fs in figs .
[ fig - ell - deltag ] and [ fig - comparision ] are the same as those for fig .
[ fig - ell - abs - phi ] .
( [ x0y0 ] ) of the maximum amplitude by using the asymptotic ( dashed curve ) and exact ( solid curve ) formulas .
the depths of the defect are @xmath285 ( a ) and @xmath286 ( b ) in units of @xmath287 . , width=302 ]
the second model fs we want to discuss has the form of a corrugated cylinder ( fig .
[ fig - openfs ] ) , which is open along the direction @xmath288 , @xmath289where @xmath290 is the size of the brillouin zone , and @xmath6 is an effective mass .
we further impose that the momentum perpendicular to the symmetry axis of the fs remains finite , @xmath291where@xmath292are the maximal and minimal radii of the cylindrical surface , respectively . as a consequence of rotational symmetry the gaussian curvature @xmath10 ( gaus0 ) of the surface depends only on @xmath293 .
the central part of the surface ( ` belly ' ) has a positive curvature @xmath294 while the ends near the brillouin zone boundary ( ` necks ' ) have negative curvature . in the direction perpendicular to the symmetry axis
there are two partial waves propagating with different parallel velocities , @xmath295 and @xmath296 , belonging to the parts of fs having opposite sign of @xmath10 .
rotating away from the perpendicular direction towards the axis the two solutions persist but the two corresponding points on the fs move closer together until they merge at the curve defined by @xmath13 , the inflection line . for directions beyond this angle ( i.e. for @xmath297 in fig .
[ fig - openfs ] ) no propagating wave solutions exist . on the inflection line a unique solution with velocity @xmath298
is found .
for @xmath299 there are two stationary phase points @xmath300 @xmath301 that satisfy eq .
( [ st_point2 ] ) , corresponding two different velocities @xmath302 directed along the radius vector @xmath8 .
the larger value , @xmath303 , belongs to the belly of the fs @xmath304 and the smaller one , @xmath305 , belongs to the neck @xmath306 . at the inflection line of the surface
we have , @xmath307which defines the value of perpendicular momentum @xmath308 . from this condition
we obtain @xmath309the cone inside of which no propagating states exist is defined by the condition @xmath310where the components of the velocity @xmath311 and @xmath312 are given by , @xmath313 in spite of the simplicity of the model fs ( [ e_open1 ] ) , the integrals in eqs .
( [ psi0+g]),([conduct ] ) can not be evaluated analytically .
we can only discuss the asymptotic behavior for @xmath146 .
qualitatively this result should also be valid for @xmath314 .
for the directions that have two stationary phase points , having opposite signs of the gaussian curvature , eq .
( [ g_asym ] ) acquires the form @xmath315 where @xmath316 is the conductance of the contact without defect given by eq .
( [ g0_arbit ] ) , @xmath317 , and @xmath318 .
the appearance of the conductance oscillations depends strongly on the orientation of the fs with respect to the interface . below we will consider two specific orientations , having the axis of the fs either perpendicular or parallel to the interface
. when the iso - energy surface is open along the contact axis @xmath45 the components of the momenta in eq .
( [ e_open1 ] ) are @xmath319 and @xmath320 . in this case
the conductance of the clean contact ( without defect ) becomes , @xmath321from eq .
( [ st_point2 ] ) the stationary phase points for the iso - energy surface are , @xmath322the angle @xmath323 is defined by the direction of the radius vector @xmath8 .
the gaussian curvature @xmath324 and the phase @xmath325 in the points ( [ p012 ] ) are given by relations , @xmath326@xmath327the angle @xmath328 in eqs .
( [ p012]),([k0_open1 ] ) is contained within the interval @xmath329 where the @xmath330 is given by eq .
( [ x ] ) .
the modulus of the wave function is plotted in fig .
[ fig - modulus - openfs ] . for the calculation of the wave function ,
( [ psi_asym ] ) , we used formulas ( [ k0_open1 ] ) for the curvature and ( [ ga0_open1 ] ) for the phase in the asymptotic expression for the integral @xmath331 ( [ lam ] ) .
although , strictly speaking eq .
( [ psi_asym ] ) is not applicable in the vicinity of the contact and near the defect , nor inside the classically inaccessible region , fig .
[ fig - modulus - openfs ] illustrates the main features of this problem .
one observes the interference of the two partial waves with different velocities , the existence of a forbidden cone , the anisotropy of the waves scattered by the defect , and the enhanced wave function amplitude near the edge of the forbidden cone . for a warped cylindrical fs having the open direction along the contact axis @xmath45 .
the coordinates are measured in units of @xmath332 .
the parameters used in the model are @xmath333 , @xmath334 , and a defect sits at @xmath335,width=377 ] at the inflection lines , where @xmath13 and @xmath336 is given by eq .
( [ x ] ) , the square root in eq .
( [ p012 ] ) is equal to zero . for this direction two
stationary phase points merge and the electron velocity is directed along the cone of the classically forbidden region .
the asymptotic expression for the conductance ( [ g_op ] ) diverges at these points , which implies that the third derivative of the phase , eq .
( [ phase ] ) , with respect to @xmath293 must be taken into account . when the vector @xmath337 connecting the point contact to the defect lies along the cone of the forbidden region
the conductance oscillations have maximal amplitude and the conductance takes the form , @xmath338where@xmath339and @xmath340 is a numerical constant , @xmath341 the energy dependencies of @xmath342 and @xmath343 are given by eq .
( [ p(e ) ] ) .
[ fig - deltag - open - perp ] shows a plot of the oscillatory part @xmath344 of the conductance @xmath345 , eq .
( [ g_op ] ) , as a function of the lateral position of the defect @xmath267 for a fixed distance @xmath4 from the interface . of the conductance , as a function of the lateral position of the defect @xmath346 in the plane @xmath347 the open direction of the fs is oriented perpendicular to the interface .
the coordinates are measured in units of @xmath348 .
the parameters used for the model fs ( [ e_open1 ] ) are @xmath349 , @xmath350 , and a defect sits at a depth of @xmath351,width=377 ] the oscillation pattern has a ` dead ' region in the center , corresponding to defect positions inside the classically inaccessible part of the metal for electrons injected by the point contact .
this region is defined by the cone ( [ x ] ) and its radius @xmath352 depends on the depth of the defect under surface .
the oscillations in the conductance are largest when the defect is placed at the edge of the cone @xmath353 . in this case
the defect is positioned in a direction of velocity belonging to the inflection line of the fs and the electron flux in this direction is maximal .
the second orientation we want to discuss is that with the fs ( [ e_open1 ] ) having its open direction ( the axis ) parallel to interface , with @xmath354 and @xmath355 .
the existence of a classically inaccessible region for this geometry leads to a strongly anisotropic current density in the @xmath356 - plane .
the expression for the point contact conductance without defect @xmath316 becomes , @xmath357the expressions for the phase , eq .
( [ ga_el ] ) , and the gaussian curvature , eq .
( [ gaus0 ] ) , now read , @xmath358and @xmath359respectively . for this geometry
we use spherical coordinates , with @xmath336 the angle between the vector @xmath8 and the @xmath360-axis .
the variables @xmath361 have been obtained from eq .
( [ st_point2 ] ) , @xmath362the first stationary phase point , @xmath363 , corresponds to a positive gaussian curvature and the second one , @xmath364 , to a negative curvature . for @xmath365 , eq.([x ] ) , they become equal , @xmath366and the curvature ( [ gaus2 ] ) vanishes . figure [ fig - mod - openfs - par ] , acquired by using eqs .
( [ psi_asym ] ) , ( [ phase3 ] ) and ( [ gaus2 ] ) , illustrates the interference of waves with different velocities the electrons emerging from the contact and the interference with the waves scattered by the defect . for this geometry the classically inaccessible region
is found near the interface to both sides of the contact . for the warped cylindrical fermi surface with the open direction along the @xmath360-axis and parallel to the plane of interface .
the coordinates are measured in units of @xmath367 .
the fermi surface parameters are @xmath368 , @xmath369 , and the defect position is @xmath370.,width=377 ] of the conductance , as a function of the lateral position of the defect @xmath346 in the plane @xmath279 .
the open direction of the fs is oriented parallel to the interface along the @xmath246direction .
the coordinates are measured in units of @xmath371 .
the parameters used for the model fs ( [ e_open1 ] ) are @xmath372 , @xmath373 , and a defect sits at a depth of @xmath374 .
, width=377 ] figure [ fig - conductance - openfs - par ] shows a plot of the oscillatory part @xmath2 of the conductance @xmath375 as a function of the lateral position of the defect @xmath267 for a fixed distance @xmath4 from the interface . in this case two ` dead ' regions appear symmetrically with respect to the center of the oscillation pattern along the open direction of the fs .
the center of the pattern corresponds to a defect sitting on the axis of the contact , @xmath376 , for which @xmath377 , @xmath378 , @xmath379 , and @xmath380 .
at this point eq .
( [ g_asym ] ) takes the form @xmath381 \right\ } .
\notag\end{aligned}\ ] ]
we have analyzed the oscillatory voltage dependence of the conductance of a tunnel junction in the presence of an elastic scattering center located inside the bulk for metals with an anisotropic fs .
these oscillations result from electron waves being scattered by the defect and reflected back by the contact , interfering with electrons that are directly transmitted through the contact .
the introduction of anisotropic electron movement beams the following implication : several points on the fs may share the same direction of the group velocity vector @xmath382 whereas other directions for @xmath383 can be absent .
two non - spherical shapes for the fs have been investigated : the ellipsoid and the corrugated cylinder ( open surface ) .
contrary to the case of a spherical fs @xcite in the ellipsoidal model ( [ energy ] ) the center of the conductance oscillation pattern does not need to coincide with the actual position of the defect but is displaced over a vector @xmath384 .
when the stm tip is placed at this point the oscillatory part of the conductance is given by , @xmath385the oscillation period depends on @xmath386 , the component of the tensor of inversive mass ( [ m^(-1 ) ] ) for motion in the @xmath45-direction , and on the depth @xmath4 of the defect .
this allows us in principle to map out the positions of defects , as long as the shape of the fs is known . apart from the period of the oscillations , there is also information in the amplitude . since short periods
should correspond to small amplitudes this may be used for a test of consistency .
however , quantitatively the amplitude is also influenced by unknown factors such as the defect scattering efficiency .
the ellipsoidal fs is exceptional in that the problem can be solved exactly .
this allows us to compare the calculation with the asymptotic approximation , and this shows that the approximation works very well for distances larger than @xmath387 . in the case of the corrugated cylinder ( [ e_open1 ] ) the open necks cause cones with opening angle @xmath388 ( defined by the inflection line of the fs ) to be classically inaccessible .
if the orientation is such that the open direction is orthogonal to the surface this will result in a ` dead ' region with radius @xmath389 ( [ x ] ) where no conductance fluctuations can be observed .
thus , by measuring the size of this dead region we directly obtain the position of the defect .
the oscillation amplitude will be maximal at the border of the dead region , since the current density will be highest in the direction of the group velocity at the inflection line .
in analogy to a hurricane the ` eye ' is surrounded by a ring of intense currents .
such rings of high amplitude oscillations have already been reported very recently in experiments on ag and cu(111 ) surfaces @xcite . for our model
fs , along this border the oscillating part of the conductance is , apart from a phase factor , described by , @xmath390 where @xmath342 and @xmath343 are the maximal and minimal radii of the surface of constant energy in the direction perpendicular to the axis of the cylinder ( [ p(e ) ] ) , @xmath391 is the amplitude of corrugation of the fs , and @xmath392 is the size of the brillouin zone .
again we find that the depth of the defect is determining the oscillation period , so that for given fs parameters this information can be exploited to investigate the structure of the metal below the surface .
if the open direction is parallel with the surface the highest amplitude will be found with the stm tip straight above the defect . for @xmath393
the conductance oscillations are described by eq .
( [ gopen2 ] ) . clearly , the oscillation pattern is more complicated than that from eq .
g_el_osc , since there are contributions from the belly as well as from the neck parts of the fs , plus a sum frequency . for small necks
the signal will be dominated by the oscillation due to the belly .
although the two models presented in this paper are still rather artificial , they provide insights that are quite valuable for experimental work in this field .
the most prominent conclusion is that the regular oscillations due to convex parts of the fs , that behave as for the isotropic fs discussed previously @xcite , will often be dominated by signals due to special directions . on any surface that features regions of zero curvature
, the strongest conductance fluctuations will come from electrons travelling with the group velocity of that region .
not only does this hold for the inflection lines around the necks in the ( 111 ) direction of e.g. cu , ag or au , it is also true for the almost flat facets in the ( 110 ) direction of the same metals . in the case of an inflection line
the signal will decay as @xmath394 , whereas for a flat facet the signal does not decay at all .
this effect can be exploited for imaging defects up to much larger depths than previously estimated .
the particular shape of the fs for nearly all metals contain many detailed features that will allow us to check the validity of the conclusions drawn from the measured data .
we thank m. wenderoth for communicating his unpublished results .
. is supported by a grant of the european intas young scientists program ( no 04 - 83 - 3750 ) and yu.a.k .
was supported by the european erasmus mundus program on nanoscience . | in a previous paper ( avotina _ et al . _ phys .
rev .
b * 71 * , 115430 ( 2005 ) ) we have shown that in principle it is possible to image the defect positions below a metal surface by means of a scanning tunnelling microscope .
the principle relies on the interference of electron waves scattered on the defects , which give rise to small but measurable conductance fluctuations .
whereas in that work the band structure was assumed to be free - electron like , here we investigate the effects of fermi surface anisotropy .
we demonstrate that the amplitude and period of the conductance oscillations are determined by the local geometry of the fermi surface .
the signal results from those points for which the electron velocity is directed along the vector connecting the point contact to the defect . for a general fermi surface geometry
the position of the maximum amplitude of the conductance oscillations is not found for the tip directly above the defect .
we have determined optimal conditions for determination of defect positions in metals with closed and open fermi surfaces . |
the gravitational lensing effect of an ensemble of point masses has long been a subject of investigation ( see , e.g. , schneider , ehlers , & falco 1992 for a summary of earlier work ) .
most of the studies have focused on the situation where the spatial distribution of the point mass lenses is `` compact , '' i.e. , the size of the region within which the lenses are distributed is much smaller than the other distances ( those between the source and the lenses and between the observer and the lenses ) in the problem . for gravitational lensing calculations , a compact distribution of lenses
is usually approximated by its projection onto a single lens plane .
the important case of low optical depth , single plane microlensing has been investigated in great detail ( kaiser 1992 ; mao 1992 ; rauch 1992 ; kofman 1996 , hereafter paper i ; see also schneider 1992 and references therein ) , and we now understand reasonably well the caustic structure , the illumination pattern , the probability distribution of macroimage magnification , and the flux variability of macroimages in this case . there are , however , situations in which the lensing objects are modeled by an `` extended '' ( or three - dimensional ) distribution of point masses , i.e. , by a point mass distribution that is comparable in scale to the other distances in the problem .
a well - known example is the microlensing of stars in the large magellanic cloud and the galactic bulge by foreground stars and possibly massive compact halo objects ( machos ) ( alcock 1993 ; aubourg 1993 ; udalski 1993 ) .
the theoretical analysis of this problem is relatively simple because the optical depth to microlensing is very small ( @xmath9 ) and the point mass lenses are essentially acting independently ( paczyski 1986 , 1991 ; griest 1991 ) .
( note , however , that microlensing by physical binary systems is also possible ; mao & paczyski 1991 ; udalski 1994 ; alard , mao , & guibert 1995 . )
a more complicated problem is the possibility that a substantial fraction of the dark matter in the universe is in the form of compact objects ( e.g. , carr 1994 ) .
then the universe as a whole has a significant optical depth to microlensing ( press & gunn 1973 ; dalcanton 1994 ) .
this latter problem has been studied by several authors using numerical simulations ( refsdal 1970 ; schneider & weiss 1988a , b ; rauch 1991 ) , and it was found that there are qualitative differences between two- and three - dimensional microlensing .
for example , for @xmath5 , the three - dimensional macroimage magnification distribution does not show the prominent caustic - induced feature produced by a planar distribution of point masses ( rauch 1992 ; but see 4 below ) .
the usual approach to gravitational lensing by a three - dimensional distribution of lenses is the multiple lens - plane approximation ( blandford & narayan 1986 ; kovner 1987 ) .
the properties of multiplane gravitational lensing and , in particular , the reasons for many of the differences between two- and three - dimensional microlensing have not been investigated in detail , and they are not yet well understood .
besides the statistical microlensing studies mentioned above , there have been few studies of the basic mathematical properties of multiplane gravitational lensing ( levine & petters 1993 ; levine , petters , & wambsganss 1993 ) or the properties of specific multiplane configurations ( kochanek & apostolakis 1988 ; lee & paczyski 1990 ; erdl & schneider 1993 , hereafter es ) .
this was perhaps due to the lack of lensing candidates with lenses at different redshifts .
however , recent observations of the lens systems 2345@xmath10007 ( pell 1996 ) , 2016@xmath10112 ( lawrence , neugebauer , & matthews 1993 ) , and 0957@xmath10561 ( bernstein , tyson , & kochanek 1993 ; but see dahle , maddox , & lilje 1994 ) suggest that these systems may involve lenses at different redshifts . as we shall see in 3 , to understand some of the differences between two- and three - dimensional microlensing , it is useful to study in detail specific cases of simple three - dimensional lens configurations
the study by erdl & schneider ( es ) of the lens system consisting of two point masses at different distances from the observer is a step in this direction .
they considered primarily the critical curves and the caustics and were able to obtain a complete classification of their topological properties with respect to the model parameters in this particular case .
the main motivation of this study is to give a systematic theory of gravitational microlensing by a random distribution of point masses . in the accompanying paper i we refined the theory of gravitational microlensing for a planar distribution of point masses . in particular
, we obtained a practical semi - analytic expression for the macroimage magnification distribution , @xmath4 , for low optical depth @xmath11 , by allowing for the shear perturbations of neighboring lenses . in this paper
we study the properties of gravitational microlensing by a three - dimensional distribution of point masses .
we are interested in the critical curves , the caustics , the illumination pattern , and especially the macroimage magnification distribution .
we shall not consider other interesting aspects such as time delays and the temporal behavior of microlensing events ( note , however , that the latter is given by the magnification along tracks through the illumination pattern ) . in
2 we write down the lens equations . in 3 we study in detail multiplane configurations with two , three , and four point masses to find out what new features are expected in three - dimensional microlensing .
we also explain qualitatively why these new features arise .
the case of low optical depth three - dimensional microlensing is studied in
4 , using an approach similar to that used in paper i to study the caustic - induced feature in @xmath4 for the planar case .
the resulting semi - analytic @xmath4 is compared to the numerical results of rauch ( 1991 ) , and the differences between the two- and three - dimensional results are discussed .
our conclusions are summarized in
we begin by establishing the relevant equations for gravitational lensing by a three - dimensional lens distribution . as in paper
i , we shall only consider gravitational lensing in an einstein - de sitter cosmological background weakly perturbed by the gravitational field of the lenses .
our notation is identical to that used in paper i , and the readers are referred to 2 of paper
i for the definitions of terms not explicitly defined here .
let us consider an isotropic point source at the origin of our coordinates and an observer plane at a comoving distance @xmath12 from the source ( see fig . 1 ) .
in the absence of perturbations , the observer plane would be uniformly illuminated and a light ray which leaves the source with angle @xmath13 would pierce the observer plane at @xmath14 .
this defines our planar lagrangian coordinates @xmath15 . in the presence of density inhomogeneities ,
the light ray would suffer deflections and pierce the observer plane at the eulerian coordinates @xmath16 .
for a three - dimensional ( or extended ) lens distribution , gravitational lensing is usually studied within the framework of the multiple lens - plane approximation ( blandford & narayan 1986 ; kovner 1987 ) .
it is assumed that the lens distribution can be approximated by a series of @xmath17 geometrically thin configurations , i.e. , lens planes , and that the separations between the lens planes are sufficiently large .
then the deflection of a light ray by one of the lens planes is not influenced by the others and a light ray is deflected @xmath17 times on the way from the source to the observer ( see fig . 1 ) .
after @xmath18 deflections , the location at which the light ray pierces the @xmath19th lens plane , projected onto the observer plane , is @xmath20 where @xmath21 is the deflection by the @xmath22th plane ( which can be expressed as the gradient with respect to @xmath23 of an effective surface gravitational potential on plane @xmath22 ) , @xmath24 is a dimensionless distance parameter defined by @xmath25 and @xmath26 is the comoving distance between planes @xmath22 and @xmath19 . note that @xmath24 can take on values between 0 ( in the limit @xmath27 ) and 1 ( in the limit @xmath28 or @xmath29 ) .
if we denote the observer plane as plane @xmath30 , then the lagrangian to eulerian mapping is @xmath31 where @xmath32 unlike the single plane lens mapping , this multiplane lens mapping is not in general a gradient mapping . for the particular case where the lens planes are populated by point masses , @xmath33 where @xmath34 is the cosmological scale factor at the redshift of lens plane @xmath22 and @xmath35 is the location of the @xmath36th lens ( of mass @xmath37 ) on lens plane @xmath22 .
note that @xmath35 is the position projected onto the observer plane , i.e. , @xmath38 where @xmath39 is the angular position of the lens from the source .
henceforth , we shall assume for simplicity that all the point masses have the _ same _ mass @xmath40 .
then @xmath41 where @xmath42 is the ( lagrangian ) einstein radius of a point mass on the @xmath22th lens plane .
for a random distribution of point masses , the optical depth to microlensing is @xmath43 , where @xmath44 and @xmath45 are , respectively , the projected ( onto the observer plane ) surface number density and the physical surface mass density of point masses on lens plane @xmath22 .
when we consider the situation where a significant fraction of the dark matter in the universe is in the form of cosmologically distributed compact objects , since the mean of the density inhomogeneities should be zero , we should compensate the randomly distributed point masses on each plane by a _
negative _ smooth surface mass density .
then a term of the form @xmath46 should be added to the right hand side of equation ( [ eq5 ] ) .
as we mentioned in 1 , numerical simulations of gravitational microlensing by a three - dimensional random distribution of point masses have revealed differences between two- and three - dimensional microlensing ( refsdal 1970 ; schneider & weiss 1988a , b ; rauch 1991 ) . to find out and understand qualitatively some of these differences , it is useful to study in detail specific cases of simple lens configurations . in this section
we study numerically the properties of multiplane gravitational lensing by a few point masses . to reduce the number of parameters , we restrict our study to cases where the distance between the source and the observer is much smaller than the horizon scale , i.e. , @xmath47 .
if there are @xmath48 point masses on lens plane @xmath22 ( we shall denote the total number of point masses as @xmath49 ) , the sum of the squared ( lagrangian ) einstein radii is @xmath50 for the numerical computations , it is convenient to express all positions in units of @xmath51 and to define a dimensionless mass parameter @xmath52 so that @xmath53 . then the multiplane lens equation with identical point masses ( eqs.[[eq1 ] ] and [ [ eq5 ] ] ) can be written as @xmath54 first let us consider in detail configurations with two point masses . without any loss of generality , we can choose a coordinate system such that @xmath55 and @xmath56 .
since we assume that @xmath47 , the two remaining parameters , @xmath57 and @xmath58 , are related by @xmath59 .
the topology of the critical curves ( and the caustics ) as a function of the parameters @xmath60 has been classified by es ( see also schneider & weiss 1986 for the @xmath61 case ) . in general , for given @xmath57 and @xmath58 , there are five transitions in the topology of the critical curves as @xmath62 increases from @xmath63 ( see fig . 6 of es ) .
however , in the limit @xmath64 ( i.e. , in the limit of two point masses on the same plane ) , three of the transitions merge at @xmath65 and the number of transitions is reduced to two . for comparison ,
we show in figures 27 two sets of models with @xmath61 and @xmath66 .
( we have also computed models with @xmath67 , but the results are qualitatively similar to those with @xmath66 and are not shown here . ) the configurations shown in figures 27 were chosen to sample regions of the parameter space with different critical curve topologies .
the parameters @xmath68 and the critical curve topology types ( in the notation of es ) of these configurations are listed in table 1 .
( note that the topology type @xmath69 is not represented ; with @xmath59 , type @xmath69 only occurs at very small @xmath62 . ) in figures 2 and 5 we show the critical curves of the configurations with @xmath61 and @xmath70 respectively .
the corresponding caustics and iso - magnification contours ( on the observer plane ) are shown in the left and right panels of figures 3 and 6 . in figures 4 and 7
we plot the `` normalized '' differential cross - section @xmath71 , where @xmath0 is the differential cross - section for magnification @xmath72 and @xmath73 is the differential cross - section of a single isolated point mass lens with einstein radius @xmath51 as defined in equation ( [ eq7 ] ) .
as we saw in paper i ( see also 4 below ) , the function @xmath74 is a useful measure of the deviations of the cross - section of a more complex lens configuration from that of the point mass lens .
the cross - sections and the iso - magnification contours shown in these ( and all subsequent ) figures were obtained using the ray - shooting method ( see , e.g. , kayser , refsdal , & stabell 1986 ; schneider & weiss 1986 , 1988b ) . since the magnification factor is calculated in this method by collecting rays into a grid of square pixels , it is the magnification for an observing device ( or equivalently a source ) of the size of one pixel . to assess the effects of finite pixel size , we performed two sets of calculations , with the linear pixel sizes in one set a factor @xmath75@xmath76 smaller ( see table 1 ) .
the pixel sizes were chosen so that all ( except the very small ) caustics are resolved .
we show in figures 3 and 6 the contours from the calculations with the smaller pixels and in figures 4 and 7 the function @xmath74 from both sets of calculations .
as expected , the finite pixel size introduces an upper limit to the magnification factor and produces extra `` bumps '' in @xmath74 ( rauch 1992 ) , but it does not have significant effects on the position or the strength of the real caustic - induced features in @xmath74 .
the @xmath61 ( and @xmath77 ) models have already been studied in detail by schneider & weiss ( 1986 ) ; hence , we only give a brief description of these models here . for @xmath78 , the region near each of the point masses
is perturbed by the weak shear from the other point mass .
so we have two critical lines that are slightly flattened ellipses and two caustics that have the familiar `` astroid '' shape ( see 3 and fig . 2 of paper i ) .
as @xmath79 , the critical curves and the caustics become asymmetric ( figs .
2_a _ and 3_a _ ) . the corresponding differential cross - section ( fig .
4_a _ ) shows caustic - induced features that are similar to those found in the case of a point mass perturbed by shear ( see fig . 3 of paper i ) .
there is , in addition , a bump ( at @xmath80 ) which is associated with the last iso - magnification contour that encloses both of the caustics . as @xmath62 decreases below @xmath76 , the number of critical curves ( and caustics )
changes from two to one to finally three , and the topology of the critical curves ( and the caustics ) changes from @xmath81 to @xmath82 to finally @xmath83 ( see table 1 ) . in all cases , there are caustic - induced features in the differential cross - section at moderate to high @xmath72 and the cross - section at small @xmath72 is reduced .
each of these features is associated with either a local minimum or saddle point in the magnification pattern or an iso - magnification contour that touches the caustics at one or more points . in the latter case ,
the contours of slightly larger @xmath72 are broken into disconnected shorter segments by the caustics and the loss of the area inside the caustics results in a drop in @xmath74 .
( an example that has been discussed in detail in 3 of paper i is the feature associated with the contour osculating the astroid - shaped caustic of the point mass plus weak shear lens . ) when the two point masses are not on the same lens plane , there are new features and transitions in the critical curves and the caustics ( see figs . 5 and 6 ) . at large
lagrangian separation @xmath62 ( fig .
5_a _ ) , besides the two critical curves that are similar to the ones seen in the @xmath61 case ( see 4 for an explanation ) , there is an extra critical curve inside the critical curve that encloses point mass @xmath84 ( i.e. , the point mass closer to the source ) .
this critical curve is offset from point mass @xmath84 in the direction opposite point mass @xmath76 and both its size and its distance from point mass @xmath84 increase with decreasing @xmath62 .
the corresponding caustic is located at a large distance along the positive @xmath85-axis ( fig .
this caustic has the ( asymmetric ) astroid shape of the two caustics that are known from the @xmath61 case , and it increases in size and moves toward the other two caustics as @xmath62 decreases . the critical curve ( and caustic ) configuration at large @xmath62 that we have just described is denoted as topology type @xmath72 by es . except for the presence of the new critical curve and caustic , the topological transitions shown in figures 5_a__c _ and 6_a__c _ are similar to those shown in figures 2 and 3 .
the remaining transitions do not exist in the single lens plane case .
however , as es pointed out , except for the elliptic umbilic transition between @xmath86 and @xmath87 , all transitions are the beak - to - beak type .
we can explain the new critical line ( and caustic ) and its qualitative behaviors if we focus on the configurations of topology types @xmath72@xmath86 . for these configurations ,
it is clear that the new critical line is composed of the set of rays that are strongly deflected by point mass @xmath84 ( on lens plane @xmath84 ) to the vicinity of point mass @xmath76 ( on lens plane @xmath76 ) and are then infinitely magnified by point mass @xmath76 .
obviously this process can only occur when the point masses are not on the same lens plane .
the new critical line is closer to point mass @xmath84 for larger @xmath62 because a light ray must be more strongly deflected by point mass @xmath84 to reach the vicinity of point mass @xmath76 . in the magnification patterns ( fig .
6 ) , there are local minima and saddle points that are associated with the new caustic ; there are also iso - magnification contours that touch the new caustic at two points ( e.g. , the second outermost contour in fig .
these new features in the magnification patterns are expected to produce corresponding features in the cross - sections .
however , as we can see in figure 7 , the `` normalized '' differential cross - sections @xmath74 are actually quite similar to those shown in figure 4 . features associated with the new caustic are not noticeable until @xmath88 , and even in cases where the new caustic ( which is transformed at small @xmath62 ) encloses a large area of the observer plane , the new features are typically weak and at @xmath89 .
( they are stronger and at higher @xmath72 for the @xmath67 models . )
if we examine the magnification patterns ( and fig .
8) more carefully , we see that the new caustic is usually located in regions where the background magnification is small . in addition
, the magnification throughout most of the region inside the new caustic and , in particular , the minimum magnification inside the new caustic are only slightly higher than the low background magnification .
hence the new features in the magnification patterns and the corresponding features in @xmath74 have small @xmath72 values .
next , let us consider specifically the feature associated with the iso - magnification contour that touches the new caustic at two points ( e.g. , the second outermost contour in fig .
this contour is long and , except for the distortion near the new caustic , nearly circular .
contours of slightly higher magnification are broken by the new caustic into a short arc around one of the cusps and a long arc that encloses the other caustic(s ) .
this results in a decrease in the cross - section @xmath0 because an area that would otherwise be enclosed by two of these contours ( of magnification @xmath72 and @xmath90 ) is lost to the interior of the caustic .
although the area lost to the caustic can be large , it is typically much smaller than the area enclosed by the long ( and the short ) arcs .
consequently , the decrease in @xmath91 , which is approximately the fractional decrease in @xmath0 , is small .
similarly , the increase in @xmath74 at @xmath92 , where @xmath93 is the minimum magnification inside the new caustic , is small because @xmath93 is typically small and the area enclosed by the contours of magnification @xmath93 and @xmath94 outside the new caustic ( mainly two long arcs ) is much larger than the area enclosed by the contour @xmath94 inside the new caustic .
the relatively weak influence of the new caustic on the magnification pattern is demonstrated more clearly in the sample of magnification profiles plotted in figure 8 .
although the magnification must be infinite on the caustic , only a very small region near the caustic has @xmath95 .
in fact , the numerically obtained @xmath72 , which is finite at a caustic because of the finite pixel size of our calculations , shows a maximum at the new caustic that is quite small . this behavior of the magnification pattern near the new caustic matches the description of the `` thin caustics '' found in the microlensing simulations of schneider & weiss ( 1988b ) . we have therefore found in the simple two - point - mass case a clear demonstration of the occurrence of thin caustics . as we mentioned earlier , the new caustic consists of light rays that are strongly deflected by the first point mass to the vicinity of the second point mass and
are then infinitely magnified by the second point mass . in the simulations of schneider & weiss , which have a large number of point masses distributed on several lens planes ,
a light ray can be strongly deflected on more than one lens plane .
it is likely that most of the caustics that are strongly deflected on one or more lens planes ( i.e. , the secondary caustics defined in 3.2 below ) are thin caustics .
this identification for the thin caustics is consistent with the conjecture by schneider & weiss that the thin caustics are caused by steep deflection gradients because the regions ( in lagrangian space ) with strong deflections are also regions with steep deflection gradients .
now let us consider briefly configurations with three and four point masses .
two configurations with @xmath96 point masses and one with @xmath97 are shown in figures 911 .
these configurations have one point mass on each lens plane ( i.e. , @xmath98 ) and their parameters are listed in table 2 . the configuration _ a _ with @xmath99 was chosen because it has the critical curve topology of point masses widely separated in lagrangian space while the smaller critical curves are clearly resolved .
the other configurations are `` typical '' configurations resulting from randomly distributing the point masses within a cone of constant lagrangian radius .
we have ( arbitrarily ) chosen a cone radius such that a light ray passing within the cone would be infinitely magnified if the mass of the lenses were uniformly distributed within the cone . a striking feature of figure 9 ( and fig .
5 ) is the rapid increase in the number of critical curves with @xmath1 . for a configuration with @xmath1 point masses well separated in lagrangian space and with each point mass on its own lens plane ( i.e. , @xmath98 )
, we can calculate the number of critical curves by extending the discussion given in 3.1 for the origin of the new critical curve .
light rays from the source can reach the vicinity of the point mass on the @xmath22th lens plane after being strongly deflected by any combination ( or none ) of the previous @xmath100 point masses . for each combination , there is a set of rays that are infinitely magnified by the @xmath22th point mass ( and the weak deflections on the remaining planes ) .
therefore , there are @xmath101 critical curves associated with the @xmath22th point mass and the total number of critical curves is @xmath102 .
the critical curves shown in figures 5_a _ and 9_a _ are consistent with these numbers .
( the total number of critical curves is smaller for configurations with @xmath103 and approaches @xmath1 in the limit @xmath104 , i.e. , in the single lens plane case . )
hereafter , we shall refer to the caustics that are not strongly deflected as the _ primary _ caustics and those that are strongly deflected on at least one lens plane as the _ secondary _ caustics .
although the number of critical curves increases exponentially with @xmath1 , we can see in figure 9 that the size of a critical curve decreases rapidly as the number of strong deflections associated with the critical curve ( or equivalently the number of critical curves within which it is nested ) increases . on the other hand , as in the two - point - mass case ,
a caustic formed after a larger number of strong deflections is not necessarily smaller .
for example , in figure 10_c _ , the caustic located near @xmath105 [ which corresponds to the critical curve located near @xmath106 and the caustic with six cusps [ which corresponds to the largest critical curve ] are comparable in size . in the two - point - mass case , due to the absence of overlapping or self - intersecting caustics ,
a complete classification of the critical curve topology also provides a complete classification of the caustic topology ( es ) .
this is not true when @xmath107 . in particular , there are overlaps of caustics that belong to different critical curves
. examples of such overlapping caustics are the caustics located near @xmath108 in figure 10_a _ and those in the lower left corner of figure 10_c_. ( there are no self - intersecting caustics in the configurations shown in fig . 10 , but they may be present in other configurations . ) finally , we discuss the normalized differential cross - sections plotted in figure 11 . for the `` typical '' configurations _ b _ and _ c _ , the qualitative behaviors of @xmath74 agree with what one might expect from the two - point - mass configurations studied in 3.1 .
the cross - section at small @xmath72 is smaller than that for a single point mass .
there are prominent features at moderate to high @xmath72 that are associated with the primary caustics .
features associated with the secondary caustics are typically weak , but some of them are noticeable at @xmath89 . for the configuration _ a _ ,
the features associated with the primary caustics , i.e. , those at @xmath109 , are similar to the features in figures 4_a _ and 7_a_. this is expected since configuration _ a _ has the critical curve topology of widely separated point masses .
however , the features associated with the secondary caustics are at unusually high @xmath72 : the `` plateau '' between @xmath110 and @xmath111 consists of a series of such features .
this is due to the fact that the parameters of configuration _ a _ were specially chosen so that the smaller critical curves are clearly resolved .
for these parameters , the caustics are unusually close to one another ( see fig .
10_a _ and notice the change in scale between figs .
10_a _ and 10_b _ ) . as a result
, the secondary caustics are located in regions of relatively high background magnification and the associated features in @xmath74 are at unusually high @xmath72 .
the study of simple lens configurations in the previous section shows that microlensing by a three - dimensional distribution of point masses is , in general , much more complicated than microlensing by lenses confined to a single lens plane .
but there are also indications that , in the low optical depth limit , some aspects of the three - dimensional problem can be reduced to the planar one . for example , the multiplane configurations with point masses widely separated in lagrangian space show primary caustics and corresponding features in @xmath74 that are similar to those found in the single plane configurations . in 4.1
we confirm this analytically for the limiting case of a point mass lens perturbed by the weak shear from the other lenses .
another useful result from the study in
3 is that the significant features in @xmath74 at moderate to high @xmath72 are associated with the primary caustics .
therefore , we can neglect the features associated with the secondary caustics when we analyze the macroimage magnification distribution , @xmath4 , at high @xmath72 . in 4.2
we derive the feature induced by the primary caustics at the high magnification end of @xmath4 for a three - dimensional low optical depth lens distribution .
the resulting semi - analytic @xmath4 is compared to previous numerical results in 4.3 , and
the differences between the two- and three - dimensional results are discussed in 4.4 .
in the low optical depth limit , lensing ( other than low magnification events ) by a three - dimensional distribution of point masses is , as in the planar case , typically dominated by a single point mass . without loss of generality
, we shall designate one of the point masses residing on the @xmath112th lens plane as the dominant one and locate this point mass at the origin of the lens screen .
on all of the other screens , the light ray is only weakly perturbed , and the perturbations are expanded to the first order in @xmath113 . before the light ray passes the @xmath112th screen with the dominant lens ( i.e. , for @xmath114 ) , the lens mapping ( eqs.[[eq1 ] ] and [ [ eq5 ] ] ) is simplified to @xmath115 where the constant deflection @xmath116 and the shear matrix @xmath117 ( similar to @xmath118 and @xmath119 defined in eq.[13 ] of paper i ) represent the perturbative influence of the lenses on the @xmath22th plane .
equation ( [ eq11 ] ) can also be written as @xmath120 where @xmath121 and @xmath122 is the identity matrix . once the light ray has traversed the @xmath112th lens screen , the location at which the light ray intersects subsequent ( i.e. , @xmath123 ) screens is given by @xmath124 where @xmath125 denotes the influence of the dominant lens and @xmath126 note that , in the absence of the dominant lens , @xmath127 is the path followed by a light ray with @xmath128 . on the observer plane [ which is the @xmath129th screen , but we also denote it with the subscript `` o '' ] , the eulerian position is given by @xmath130 if we use the fact that @xmath131 and transform the eulerian variable according to @xmath132 , the above equation can be cast into the form @xmath133 where @xmath134 , @xmath135 although there is only shear on each lens screen , the matrix @xmath136 is in general a combination of shear and convergence ( i.e. , it is symmetric but not traceless )
. thus equation ( 19 ) is formally equivalent to the single plane lens equation for the point mass lens plus shear and smooth surface density ( compare with eq.[15 ] of paper i for the point mass plus shear lens ) .
this is a specific example of the `` non - linear telescope '' ( or generalized quadrupole lens ) discussed by kovner ( 1987 ; see also schneider 1992 ) .
we are interested in the case where the shear perturbations on all the lens screens are weak .
to first order in @xmath137 , the expressions for @xmath138 , @xmath139 , and @xmath136 are @xmath140 and @xmath141 respectively .
hence @xmath136 is reduced to a shear matrix : @xmath142 , where @xmath143 is the magnitude of the `` effective '' shear and @xmath144 then equation ( 19 ) reduces to the single plane lens equation for the point mass plus weak shear lens ( eq.[15 ] of paper i ) . in 3.12 of paper i , we studied in detail the properties of single plane lensing by the point mass plus weak shear lens .
the results are applicable to the mapping from @xmath145 to @xmath146 ( eq.[19 ] ) .
we can then infer the properties of the lens mapping from @xmath15 to @xmath16 ( eq.[[eq18 ] ] ) by taking into account the effects of the transformations @xmath147 and @xmath148 .
first , it is easy to see that , to first order in @xmath137 , both transformations are a combination of translation and shear distortion and are one - to - one .
second , since the deformation tensor @xmath149 there is a simple correspondence between the caustics ( and the critical curves ) of the mappings ( [ eq18 ] ) and ( 19 ) . in addition
, because the determinants @xmath150 and @xmath151 are @xmath152 , the magnification @xmath153 at a point @xmath16 is , to first order in @xmath143 , identical to the `` magnification '' @xmath154 at the corresponding point @xmath155
. similarly , the correction factors for the differential cross - section are @xmath152 .
all of the corrections that we have just considered are of the same order as the terms already ignored in the weak shear analysis in paper i. therefore , the effects of the transformations are negligible , and the properties derived in 3.1 - 2 of paper
i are also applicable to the lens mapping ( [ eq18 ] ) .
briefly , for a point mass lens perturbed by weak shear in three dimensions , the critical line is a slightly flattened ellipse ( of size @xmath156 ) ; the caustic has the shape of an astroid , with the four cusp catastrophes ( at a distance @xmath157 from the center of the astroid ) connected to each other by four fold catastrophes ( see fig . 2 of paper i )
; the differential cross - section has the scaling property ( see eq.[23 ] of paper i ) @xmath158 ) = { 2 \pi x_{e , l}^2 \over ( a^2 -1)^{3/2 } } \ , \varphi(a s'[\chi_{lo } ] ) , \label{eq23}\ ] ] where the normalized differential cross - section @xmath159 depends on @xmath72 and @xmath143 only through the combination @xmath160 ; and @xmath159 , which is shown in figure 3 of paper i , shows strong caustic - induced features at the critical values @xmath161 and @xmath162 . as we mentioned in 2 ,
when we consider the situation where a significant fraction of the dark matter in the universe is in the form of cosmologically distributed compact objects , we should compensate the randomly distributed point masses on each lens plane by a negative smooth surface mass density .
the derivation in this section can be generalized to this case if we replace @xmath137 in all the equations ( except eq.[[eq12b ] ] ) by @xmath163 .
the results remain valid , but the magnification @xmath72 should be interpreted as the magnification with respect to an `` empty beam '' ( i.e. , a light beam that is not affected by the point masses ) .
as we discussed earlier , in the case of low optical depth microlensing by a three - dimensional distribution of point masses , features associated with the secondary caustics are negligible at the high magnification end of the macroimage magnification distribution @xmath4 .
therefore , as in the two - dimensional case ( see 3.3 of paper i ) , @xmath4 at high @xmath72 can be approximated as a superposition of the cross - sections of the individual point mass lenses , with the shear perturbation due to the other point masses varying from lens to lens .
let us assume that the point masses are distributed randomly on @xmath17 lens planes , each with projected ( onto the observer plane ) surface number density @xmath164 .
then @xmath4 can be written as @xmath165 where @xmath166 is the cross - section given in equation ( [ eq23 ] ) , @xmath167 is the probability distribution of the effective shear @xmath143 for a dominant lens at @xmath168 , @xmath169 is the probability that the dominant lens is located on the @xmath112th screen , and @xmath170 is the total projected surface number density . since @xmath169 is simply @xmath171 we have @xmath172 where @xmath173 the function @xmath174 has the same structure as @xmath4 of the two - dimensional case ( eq.[29 ] of paper i ) , and it is the contribution to the macroimage magnification distribution by the lenses on the @xmath112th lens screen as the dominant lenses . the effective shear @xmath143 that enters into the cross - section @xmath166 is a weighted superposition of the shear induced by all except the dominant lens ( eq.[[eq20b ] ] ) . in the appendix
we show that @xmath175^{3/2 } } , \label{eq28}\ ] ] where @xmath176 is the `` effective '' optical depth at @xmath168 .
[ note that the effective optical depth @xmath177 is smaller than the ( usual ) optical depth @xmath178 because the extra factors @xmath179 ( for @xmath180 ) and @xmath181 ( for @xmath182 ) are less than unity . ] substituting equations ( [ eq23 ] ) and ( [ eq28 ] ) into equation ( [ eq27 ] ) , we find that @xmath174 is similar to @xmath4 for a single lens plane ( compare with eq.[31 ] of paper i ) : @xmath183,\ ] ] where @xmath184 is the function introduced in 3.3 of paper i to describe the caustic - induced feature in @xmath4 in the single lens plane case
( see fig . 4 of paper i ; also fig .
12 of this paper ) .
thus @xmath174 exhibits the same caustic - induced bump ( with a maximum enhancement of @xmath185 at @xmath186 ) and the same scaling property ( with @xmath187 being a function of the combination @xmath188 ) as @xmath4 of the single lens plane case .
there is , however , a significant difference between the two- and three - dimensional cases . to obtain the macroimage magnification distribution in the three - dimensional case , we must sum @xmath174 over all the lens planes ( eq.[[eq26 ] ] ) : @xmath189 = { 2 \tau \over ( a^2 - 1)^{3/2 } } f_n ( a ; \tau ) , \label{eq31}\ ] ] where @xmath190 , \label{eq32}\ ] ] and @xmath191 is the ( usual ) optical depth .
although we denote the caustic - induced feature in the @xmath17-plane case as @xmath192 for simplicity , it should be clear from equation ( [ eq32 ] ) that @xmath192 depends on the location and the surface density of the lens planes in a non - trivial manner and does not have any simple scaling properties ( when @xmath193 ) .
note also that the differences between @xmath194 and @xmath192 are due to the differences between the optical depths @xmath11 and @xmath177 . to proceed further , we need to assume a model for the distribution of the point mass lenses . as in the numerical simulations reported by several authors ( refsdal 1970 ; schneider & weiss 1988a , b ; rauch 1991 ) , we consider a lens distribution with _ constant _ comoving mass density @xmath195 .
we divide the volume between the source and the observer into @xmath17 shells of equal comoving thickness ( @xmath196 ) and place the point masses in each shell on a lens plane at the middle of the shell . in figure 12
we plot the numerically evaluated @xmath192 for @xmath104 , @xmath76 , @xmath197 , @xmath198 , and @xmath199 , and for source redshift @xmath200 and @xmath201 .
it is clear that @xmath202 has converged by @xmath203 .
this agrees with the conclusion from a study of lensing by smoothly variable mass distribution that a fully three - dimensional mass distribution is well approximated by as few as eight or @xmath199 screens ( lee & paczyski 1990 ) .
it is also clear that @xmath204(@xmath205 ) is very insensitive to the source redshift , at least for @xmath206 . in the limit @xmath207 ,
the summations in the expressions for @xmath11 , @xmath177 , and @xmath192 are replaced by integrals : @xmath208 ; \\
f_\infty ( a ; \tau ) & = & { 4 \pi g \rho_l a_0 ^ 3 \over \tau } \int_0^{\chi_{so } } d\chi { \chi \left(\chi_{so}-\chi\right ) \over
a(\chi ) \chi_{so } } f_1 \left[a \tau { \tau ' \over \tau}\right ] .\end{aligned}\ ] ] equations ( 33a b ) can be evaluated analytically , and they are particularly simple in the limit @xmath200 ( i.e. , in the limit @xmath209 ) : @xmath210 .\end{aligned}\ ] ] thus the contribution to @xmath204 is dominated by the lens planes at @xmath211 with @xmath212 .
this is consistent with the numerical results shown in figure 12 .
the most significant change from @xmath187 to @xmath204 is the shift in the location of the bump from @xmath213 to @xmath214 .
otherwise , the peak value ( @xmath215 enhancement ) and the width of @xmath204 are only slightly smaller and wider .
several authors ( refsdal 1970 ; schneider & weiss 1988a , b ; rauch 1991 ) have used numerical simulations to calculate the macroimage magnification distribution , @xmath4 , produced by a three - dimensional distribution of point masses . in the most recent study by rauch
, distributions with high resolution in @xmath72 were obtained from monte carlo simulations for three cases with @xmath216 , @xmath217 , and @xmath218 .
the `` equivalent @xmath219 value '' is related to the average magnification ( with respect to an empty beam ) on the observer plane , @xmath220 , by @xmath221 , i.e. , @xmath219 is the optical depth @xmath11 of a _ two - dimensional _ lens distribution with the same @xmath220 . for a three - dimensional lens distribution , @xmath219 and @xmath11
are not identical in general , but @xmath222 in the limit @xmath223 .
we wish to compare the semi - analytic @xmath4 derived in 4.2 to the distributions obtained by rauch .
( we shall focus on his results for an einstein - de sitter cosmological background , but rauch also found that the distributions for the same @xmath219 but different background cosmology are nearly identical . ) since our derivation assumes low optical depth , the best case for comparison should be the @xmath216 case . unfortunately ,
even with @xmath224 simulations , the monte carlo simulations only provide useful information about @xmath4 for @xmath225 ( see fig . 2 of rauch 1991 ) .
this is not sufficient for detecting the caustic - induced feature because significant deviation from @xmath226 is not expected for @xmath227 ( @xmath228 for @xmath229 ) ( see @xmath204 in fig .
we shall not show the comparison for this case and simply note that the numerically obtained @xmath4 ( for @xmath225 ) agrees with the low optical depth semi - analytic result , with or without the caustic - induced feature . in figure 13
we show the distributions obtained by rauch for @xmath230 and @xmath218 ( histograms ) .
the data are those shown in figures 3 and 4 of rauch ( 1991 ) , but they are plotted in the form @xmath231 and we have increased the bin sizes at large @xmath72 to reduce statistical fluctuations . as rauch ( 1992 ) pointed out , the distributions in the three - dimensional case do not show the prominent bump at @xmath232 that was found in the planar case
. however , the distributions are not featureless .
this can be seen most clearly in the @xmath233 case ( note , e.g. , the inflection point at @xmath234 ) . in a plot of @xmath231 ,
the low optical depth distribution without the caustic - induced feature is a horizontal line of amplitude @xmath235 , while the semi - analytic distribution with the caustic - induced feature is @xmath236 ( eq.[[eq31 ] ] ; recall that @xmath204 is very insensitive to the source redshift ) .
since @xmath230 and @xmath218 are not much less than unity , two corrections to the low optical depth semi - analytic results are necessary when we compare them to the numerical ones .
first , it is no longer accurate to assume that @xmath237 . for these particular models , @xmath230 and @xmath218
correspond to @xmath238 and @xmath239 , respectively .
second , the normalization @xmath240 can be significantly different from the low optical depth result @xmath241 . in
5.1 of paper i , we discussed the renormalization of @xmath4 due to the net convergence of light rays and multiple streaming for the single lens plane case . from an analytic result for @xmath4 in the high magnification limit , which is exact even for finite optical depth @xmath11
( see also schneider 1987 ) , we adopted the normalization @xmath242 ( for @xmath223 , @xmath243 and @xmath244 , and the correction to the normalization @xmath241 is small . )
the analysis that led to equation ( [ eq35 ] ) can not be easily generalized to the three - dimensional case ( it would require the statistical properties of the matrix @xmath245 which depends on the shear matrices @xmath137 _ non - linearly _ ; see eq.[14 ] ) , and an equivalent expression for @xmath240 for a three - dimensional lens distribution is not known . nevertheless , in the range of @xmath11 of interest here ( i.e. , for @xmath246 ) , it is likely that equation ( [ eq35 ] ) is a good approximation for @xmath240 in the three - dimensional case . in equation ( [ eq35 ] ) , the dominant correction to the normalization @xmath241 is the factor @xmath220 .
( the correction due to the factor @xmath247 is second order in @xmath11 and less than @xmath248 for @xmath246 . )
the factor @xmath220 arises from converting a probability distribution in lagrangian space to a distribution in eulerian space ( see 5.1 of paper i ) .
the conversion ( and hence the factor @xmath220 ) should remain in any generalization of the analysis to the three - dimensional case .
furthermore , as in the planar case ( see 5.1 of paper i ) , we can derive the factor @xmath220 using a different argument : in 4.2 we assumed that the surface density of the astroid - shaped caustics on the observer plane is the same as the projected surface density of point masses @xmath249 ( see eq.[[eq24 ] ] ) ; there is , however , a net convergence of the light rays by the overall lens distribution ; this increases the density of astroids by a factor @xmath220 , and introduces an additional factor @xmath220 in the normalization of @xmath4 in equation ( [ eq31 ] ) .
therefore , we shall adopt equation ( [ eq35 ] ) for the normalization @xmath240 in the three - dimensional case . in figure 13
we plot @xmath240 ( dashed lines ) and @xmath236 ( solid lines ) for @xmath230 and @xmath218 .
they should be compared to the numerical results at @xmath250 ( see 4.5 for a brief discussion of the deviations at small @xmath72 ) . as we can see , the adopted normalization @xmath240 is in good agreement with the amplitudes of the numerical results .
in addition , the numerical results show deviations from the horizontal dashed lines that are similar in shape to the semi - analytic caustic - induced feature @xmath204 ( note , e.g. , that the inflection point in the numerical result for @xmath233 coincides with the location @xmath251 where @xmath204 starts to deviate significantly from unity ) . in the @xmath230 case ,
there is little , if any , difference between the numerical and semi - analytic results ; the only possibly statistically significant deviation is that the numerical distribution may be higher near @xmath252 . in the @xmath233 case ,
the difference between the numerical and semi - analytic results is more pronounced , with the numerical distribution slightly higher than the semi - analytic distribution over the range @xmath253 .
note , however , that the number of events in the histograms at @xmath254 is small ( see the fluctuations amongst the same bins in the @xmath230 case ) , and the true distribution may not be as flat as the histograms shown . the comparison in
4.3 shows that , in the case of a three - dimensional lens distribution , the function @xmath204 , which was derived in the low optical depth limit , provides a reasonably good fit to the caustic - induced feature in @xmath4 for @xmath11 as large as @xmath239 .
this is very different from the two - dimensional case studied in paper i. for a two - dimensional lens distribution , the caustic - induced feature in the numerically obtained @xmath4 is more prominent than the low optical depth semi - analytic feature @xmath187 for @xmath11 as small as @xmath217 .
this was shown in figure 6 of paper i where we compared the numerical and semi - analytic distributions for @xmath255 and @xmath218 in a plot of @xmath256 .
the numerical distributions show a clear minimum at @xmath257 and are significantly higher than the semi - analytic results in the range @xmath258 .
for both the two- and three - dimensional lens distributions , the low optical depth analysis assumes that the point mass lenses are well separated in lagrangian space .
then the deflection near each point mass is due to that point mass and the shear perturbation from the other lenses , and an astroid - shaped ( primary ) caustic is associated with each point mass .
the low optical depth analysis is not valid if the lens density is high enough that a significant fraction of the lenses have close neighbor(s ) , and their caustics are complex structures produced by the collective effect of two ( or more ) point masses . in
5.3 of paper i , we estimated for the two - dimensional case the fraction of point masses whose caustics are not isolated astroids , @xmath259 , by considering a point mass and its nearest neighbor as a two - point - mass lens : @xmath260 . for @xmath6 , @xmath261 .
this explains the relatively strong contribution to @xmath4 by the collective effect of two ( or more ) point masses ( which results in the enhanced caustic - induced feature at moderate @xmath72 ) for @xmath11 as small as @xmath217 .
to understand the difference between the two- and three - dimensional cases , we need to estimate @xmath259 for the three - dimensional case .
we now extend the method used in paper i to estimate @xmath259 to the three - dimensional case .
we begin by considering lensing configurations with two point masses on different lens planes . since our results in 4.2 and the numerical results of rauch ( 1991 ) show that @xmath4 is not sensitive to either the source redshift or the background cosmology , we shall simplify the analysis by assuming that @xmath209 .
then the two point mass configurations of interest are identical to those studied in 3.1 and
are specified by two parameters : the dimensionless distance @xmath24 between the lens planes and the dimensionless lagrangian separation @xmath62 between the lenses [ the actual lagrangian separation is @xmath262 ; see eq.([eq7 ] ) ] . note that we have changed the notation slightly and the point masses are on lens planes @xmath22 and @xmath19 instead of lens planes @xmath84 and @xmath76 . as we saw in 3.1 ,
the topology of the caustics ( and the critical curves ) changes with decreasing @xmath62 from type @xmath72 to @xmath69 .
the only topology with two astroid - shaped primary caustics is type @xmath72 ; the other topology types are produced by the strong effect of both point masses ( see fig . 6 ) . in figure 14
we plot the separation @xmath263 at which the topology changes from type @xmath72 to type @xmath264 as a function of @xmath24 .
( the data were obtained using the numerical code that generated the caustics and critical curves shown in figs .
the separation @xmath263 decreases with increasing @xmath24 . in the limit @xmath265 , the two point masses are on the same plane and we have @xmath266 ( schneider & weiss 1986 ) .
in the limit @xmath267 ( i.e. , in the limit @xmath268 or @xmath269 ; see eq.[[eq2 ] ] ) , @xmath270 ; this is due to the focusing of the light rays by the point mass on lens plane @xmath22 and the fact that @xmath271 .
let us consider first a simple extension of the method used in paper i to estimate @xmath259 .
it is convenient to consider the projected distribution ( i.e. , the distribution in lagrangian space ) of the point mass lenses .
if we select randomly one of the point masses , the probability that the point mass is on lens plane @xmath22 is @xmath272 , where @xmath170 is the total projected surface number density of point masses .
this point mass has an absolute nearest neighbor in lagrangian space .
since the point masses are distributed randomly , the probability that the absolute nearest neighbor is on lens plane @xmath19 and at a lagrangian distance less than @xmath273 is @xmath274 $ ] . if we ignore the deflection due the other lenses and consider the point mass and its nearest neighbor as a two - point - mass lens , the discussion in the previous paragraph tells us that the primary caustic associated with the point mass is not an isolated astroid if @xmath275 .
therefore , an estimate of the fraction of point masses whose caustics are not isolated astroids is @xmath276 \right\ } .
\label{eq36}\ ] ] in the three - dimensional case , since the ( lagrangian ) einstein radius @xmath156 increases from @xmath63 to @xmath277 as @xmath168 increases ( see eq.[[eq6 ] ] ) , it is possible , e.g. , for the primary caustic ( and critical curve ) of a point mass to merge with that of its second nearest neighbor ( from plane @xmath36 ) , but not with that of its absolute nearest neighbor ( from plane @xmath19 ) , if @xmath278 .
thus the function @xmath279 , which takes into account the absolute nearest neighbor only , probably underestimates the fraction of point masses whose caustics are not isolated astroids .
we can derive a better estimate by taking into account the nearest neighbors from all of the lens planes .
for a randomly selected point mass ( on lens plane @xmath22 with probability @xmath272 ) , the probability that the nearest neighbor from lens plane @xmath19 is at lagrangian distance greater than @xmath273 is @xmath280 . if we ignore the deflection due the other lenses and consider the point mass and its nearest neighbor from lens plane @xmath19 as a two - point - mass lens , the two - point - mass lens produces two astroid - shaped primary caustics if @xmath281 .
thus the probability that the primary caustic associated with the point mass is an isolated astroid is @xmath282 $ ] .
therefore , a better estimate of the fraction of point masses whose caustics are not isolated astroids is @xmath283 \right\ } .
\label{eq37}\ ] ] as with @xmath192 , although @xmath284 and @xmath285 are denoted as functions of @xmath11 , they do depend on the location and the surface density of the lens planes .
for @xmath104 , i.e. , for a single lens plane , both @xmath286 and @xmath279 reduce correctly to @xmath287 . in the limit @xmath223 , @xmath288 in figure 15
we show @xmath279 for @xmath104 , @xmath76 , @xmath197 , @xmath198 , and @xmath277 ( dashed lines ) and @xmath289 ( solid line ) , using the constant comoving density model adopted in 4.2 .
both @xmath284 and @xmath285 have converged by @xmath290 .
the difference between @xmath291 and @xmath292 is small for @xmath293 , but it increases to @xmath294 at @xmath295 . it is clear from figure 15 that @xmath289 is significantly smaller than @xmath296 ( the highest dashed curve in fig .
15 ) . in the limit @xmath223 , @xmath297 while @xmath298 . for @xmath6 , @xmath299 and @xmath261
; even for @xmath300 , @xmath301 only .
thus , in the three - dimensional case , the fraction of point masses whose ( primary ) caustics are not isolated astroids and the contribution to @xmath4 by configurations of two ( or more ) point masses are much smaller .
this explains why the low optical depth semi - analytic feature @xmath204 provides a reasonably good fit to the caustic - induced feature in @xmath4 for @xmath11 as large as @xmath239 . at sufficiently large @xmath11
( @xmath302 ) , @xmath303 and @xmath4 should deviate significantly from the semi - analytic result .
finally , we discuss briefly the deviations of @xmath4 from the semi - analytic distribution @xmath304 at low magnification ( see fig .
13 ) . as in the planar case ,
the superposition of the cross - sections of individual lenses that results in the caustic - induced feature @xmath204 is only valid for the magnifications seen by the observers close to one of the astroid - shaped primary caustics .
the semi - analytic distribution also requires modification at low magnification as it diverges as @xmath305 at @xmath306 and is not normalizable . in the planar case
, an observer not particularly well aligned with any single lens sees a primary image close to the unperturbed position of the source and a large number of faint secondary images ( one close to each lens ) produced by strongly deflected rays . in paper
i , we took into account the correlations between the magnifications of the primary and the secondary images and derived the modification of @xmath4 at low magnification .
we have not attempted a similar calculation in the three - dimensional case because the problem is significantly more complicated .
first , for an observer that is not particularly close to either a primary or a secondary caustic , there are now @xmath307 secondary images ( where @xmath1 is the number of lenses ) because a light ray can reached the observer after being strongly deflected by one or more lenses .
although the secondary images that have been strongly deflected by more than one lens are typically much fainter than those that have been strongly deflected by only one lens , they are far more numerous and we may have to account for them . a more serious problem is the presence of weak caustic - induced features at low magnification that are associated with the secondary caustics ( see 3 ) . unlike the features associated with the primary caustics at high magnification , there are no simple semi - analytic descriptions of the features associated with the secondary caustics .
in this paper and paper i , we have sought to develop a systematic theory of gravitational microlensing .
we focused on microlensing by a planar distribution of point masses in paper i. in this paper we have studied gravitational microlensing by a three - dimensional distribution of point masses , focusing on the nature of the critical curves , the caustics , the illumination pattern , and especially the macroimage magnification distribution @xmath4 . in the process , we have explained many of the differences between two- and three - dimensional microlensing that have been seen in previous numerical simulations . in 3 we studied the lensing properties of three - dimensional lens distributions by considering in detail multiplane configurations with two , three , and four point masses .
the lensing properties of a three - dimensional lens distribution is fundamentally different from that of a planar distribution in that a light ray could have arrived in the vicinity of a lens either having suffered only weak deflections along the way from the source , or after having been strongly deflected one or more times by the intervening lens .
the latter situation can not arise in a purely planar distribution of lenses , and it gives rise to secondary caustics and critical lines that have no analogues in the planar case .
in addition , in the case of more than two point masses , the caustic topology is no longer simply related to the critical curve topology because caustics corresponding to different critical lines may overlap each other . for @xmath1 point masses that are widely separated in lagrangian space ,
there are @xmath3 primary critical curves ( i.e. , critical curves that have not suffered any strong deflections ) and @xmath2 critical curves in total .
features in the `` normalized '' differential cross - section , @xmath74 , that are associated with the secondary caustics are relatively weak and at low magnifications . significant features occuring at moderate to high magnifications are invariably associated with the primary caustics . in
4 we derived a semi - analytic expression for the caustic - induced feature at the high magnification end of @xmath4 for a low optical depth lens distribution . in the low optical depth limit
, we showed that the multiplane lens equation near a point mass is formally equivalent to the single plane equation of a point mass perturbed by shear .
consequently , the ( primary ) caustic has the familiar astroid shape , and the differential cross - section exhibits the same scaling behavior found in paper i ( i.e. , @xmath308 depends on magnification @xmath72 and the effective shear @xmath309 through the combination @xmath310 ) . by modeling the illumination pattern as a superposition of the patterns due to individual `` point mass plus weak shear '' lenses
, we found that the caustic - induced feature in the macroimage magnification distribution @xmath4 can be written as a weighted sum of the semi - analytic feature derived in paper i for a planar lens distribution ( eq.[[eq32 ] ] ) .
the resulting semi - analytic caustic - induced feature is similar to the feature in the planar case , but it is shifted to higher magnification . in the moderate to high @xmath72 regime , our semi -
analytic @xmath4 agrees remarkably well with the magnification distribution obtained from numerical simulations by rauch ( 1991 ) , even for optical depth as high as @xmath311 .
this is very different from the two - dimensional case studied in paper i. in the two - dimensional case , we have found that the caustic - induced feature in @xmath4 tends to be more pronounced than the low optical depth result , even for optical depth as low as @xmath5 .
we argued that this particular difference between two- and three - dimensional microlensing is due to the fact that for a given value of optical depth , the fraction of point masses whose ( primary ) caustics are not simple astroids is significantly higher in the planar case than in the three - dimensional case .
finally , we discussed briefly the deviations of the numerical @xmath4 from the semi - analytic distribution at low magnification .
we thank k. rauch for providing us with the results of rauch ( 1991 ) .
we also thank s. mao and k. rauch for helpful discussions .
a.b . and l.k
acknowledge the hospitality shown to them at cita during their visits in 1995 .
this work was supported in part by nserc ( canada ) , the ciar cosmology program , cita , the institute for astronomy ( l.k . ) , the dudley observatory ( a.b . ) , and a cita national fellowship ( m.h.l . ) .
in this appendix we evaluate @xmath167 , the probability distribution of the effective shear @xmath143 at a given @xmath168 , due to a low optical depth three - dimensional distribution of point masses .
it is assumed that the point masses are distributed randomly on @xmath17 lens planes , each with projected ( onto the observer plane ) surface number density @xmath164 .
specifically , on each lens plane , @xmath312 point masses are distributed randomly within a circle of lagrangian radius @xmath313 . at a given @xmath168 , the effective shear ( in matrix form ) is ( eq.[[eq20b ] ] ) @xmath314 where ( eq.[[eq12b ] ] ) @xmath315 it is convenient to represent the two distinct components of the shear matrix @xmath137 as a vector : @xmath316 where @xmath317 . then the effective shear @xmath136 can be written as an appropriately weighted vector superposition : @xmath318 where we have defined @xmath319 to simplify the notations . since all the @xmath320 are statistically independent and have similar statistical properties , the formal expression for the probability distribution of @xmath321 is @xmath322 where @xmath323 is the two - dimensional delta function .
to evaluate equation ( a5 ) , it is convenient to express the delta function and @xmath324 as fourier transforms : @xmath325 @xmath326 where the characteristic function @xmath327 is ( nityananda & ostriker 1984 ) @xmath328 substituting equations ( a6 ) and ( a7 ) into equation ( a5 ) , we can perform the integration with respect to @xmath320 : @xmath329 next , we integrate with respect to @xmath330 , grouping together the @xmath48 integrals associated with the @xmath22th screen : @xmath331 in the limit @xmath332 .
hence , @xmath333 where the `` effective '' optical depth @xmath177 is @xmath334 performing the integration with respect to @xmath335 in equation ( a11 ) , we obtain @xmath336 since @xmath337 depends only on the magnitude of the effective shear @xmath143 , it is convenient to consider the probability distribution for the latter : @xmath175^{3/2}}.\ ] ] we have expressed the probability distribution explicitly as a conditional probability that depends on the value of @xmath168 .
it should be noted that @xmath167 has the same functional form as the shear distribution due to a planar random distribution of point masses ( eq.[30 ] of paper i ) .
cccc 0.0 & 2.4 & @xmath81 & 0.020 , 0.0320.0 & 1.0 & @xmath82 & 0.016 , 0.0240.0 & 0.6 & @xmath83 & 0.016 , 0.0240.3 & 2.0 & @xmath72 & 0.020 , 0.0400.3 & 1.0 & @xmath264 & 0.018 , 0.0360.3 & 0.6 & @xmath86 & 0.016 , 0.0320.3 & 0.5 & @xmath87 & 0.016 , 0.0320.3 & 0.4 & @xmath338 & 0.016 , 0.0320.3 & 0.2 & @xmath338 & 0.020 , 0.040 ccclrc a & 3 & 1 & 0.1 & @xmath339 & 0.010 , 0.024 & & 2 & 0.5 & @xmath340 & & & 3 & 0.9 & @xmath341 & b & 3 & 1 & 0.55032 & @xmath342 & 0.012 , 0.024 & & 2 & 0.79370 & @xmath343 & & & 3 & 0.94104 & @xmath344 & c & 4 & 1 & 0.5 & @xmath345 & 0.012 , 0.024 & & 2 & 0.72112 & @xmath346 & & & 3 & 0.85499 & @xmath347 & & & 4 & 0.95647 & @xmath348 & | in paper i we studied the theory of gravitational microlensing for a planar distribution of point masses . in this second paper , we extend the analysis to a three - dimensional lens distribution .
first we study the lensing properties of three - dimensional lens distributions by considering in detail the critical curves , the caustics , the illumination patterns , and the magnification cross - sections @xmath0 of multiplane configurations with two , three , and four point masses . for @xmath1 point masses that are widely separated in lagrangian space ,
we find that there are @xmath2 critical curves in total , but that only @xmath3 of these produce prominent caustic - induced features at the high magnification end of @xmath0 . in the case of a low optical depth random distribution of point masses , we show that the multiplane lens equation near a point mass can be reduced to the single plane equation of a point mass perturbed by weak shear .
this allows us to calculate the caustic - induced feature in the macroimage magnification distribution @xmath4 as a weighted sum of the semi - analytic feature derived in paper i for a planar lens distribution .
the resulting semi - analytic caustic - induced feature is similar to the feature in the planar case , but it does not have any simple scaling properties , and it is shifted to higher magnification . the semi - analytic distribution is compared to the results of previous numerical simulations for optical depth @xmath5 , and they are in better agreement than a similar comparison in the planar case .
we explain this by estimating the fraction of caustics of individual lenses that merge with those of their neighbors . for @xmath6 ,
the fraction is @xmath7 , much less than the @xmath8 for the planar case .
= cmmib10 scaled 1200 = cmmib10 = cmbsy10 scaled 1200 = cmbsy10 = = = = = = = `` 710b = ' ' 710c = `` 710d = ' ' 710e = `` 710f = ' ' 7110 = `` 7111 = ' ' 7112 = `` 7113 = ' ' 7114 = `` 7115 = ' ' 7116 = `` 7117 = ' ' 7118 = `` 7119 = ' ' 711a = `` 711b = ' ' 711c = `` 711d = ' ' 711e = `` 711f = ' ' 7120 = `` 7121 = ' ' 7122 = `` 7123 = ' ' 7124 = `` 7125 = ' ' 7126 = `` 7127 = ' ' 7272 = " 7201 # 1#1 # 1 |
the popular hierarchical model of galaxy formation predicts that elliptical galaxies found today in the cores of rich clusters formed at high redshifts and at rare high - density peaks of the dark matter distribution . within these regions ,
gas - rich proto - galaxies merge together rapidly , forming stars at a high rate .
the same reservoir of gas used to build the stellar mass can also fuel the growth of the supermassive black holes ( smbh ) found dormant in the centre of the galaxies at low redshift ( e.g. kauffmann & haehnelt 2000 ; page et al .
2001 ; hopkins et al . 2008 ) . since massive star formation is known to be a dusty phenomenon , the optical light from many young galaxies should be highly obscured .
dust enshrouded star formation is , however , a luminous phenomenon in the millimetre through far - infrared wavebands where the star - light absorbed by dust grains is re - emitted .
indeed , in the last decade , deep surveys at submillimetre ( e.g. smail , ivison & blain 1997 ; hughes et al . 1998 ; eales et al .
1999 ; scott et al .
2002 ; coppin et al . 2006 ) and millimetre ( e.g. greve et al . 2004
; bertoldi et al . 2007 )
wavelengths have identified a high - redshift population of massive , dusty galaxies that are undergoing extreme starbursts ( e.g. smail et al .
. likewise , hotter dust heated in the circumnuclear environment of an agn will emit at mid - infrared wavelengths .
therefore , submillimetre and mid - infrared observations can be combined to study activity in forming galaxies due to dust obscured starbursts and agn . to - date
there has been no published survey at ( sub)millimetre or far - infrared wavelengths of a randomly selected region of sky over an area sufficiently large to ensure that very rare structures in the early universe are contained within it .
although large surveys of this type will be available soon from instruments such as scuba-2 on the jcmt and spire / pacs on - board the _ herschel space observatory _ , published surveys have so far been limited to an area of @xmath9 deg@xmath10 ( coppin et al . 2006 ; austermann et al .
2010 ; wei et al .
one of the largest published submillimetre surveys is the scuba half - degree extragalactic survey ( shades ; mortier et al .
2005 ; coppin et al .
2006 ) which , due to technical problems , mapped only 0.2 deg@xmath4 discovering @xmath11 submillimetre galaxies ( smgs ) . whilst this survey provided one of the largest and most uniformly selected samples of smgs ever assembled ,
the area of sky surveyed is extremely small by most standards .
for example , @xmath12cdm simulations predict that a dark matter halo of mass greater than a few times @xmath13 m@xmath14 at @xmath15 will evolve to become a very rich cluster of galaxies by @xmath16 ( e.g. farrah et al .
2006 ) . however , since the surface density of such massive haloes at @xmath17 is only of order @xmath18 deg@xmath10 ( evrard et al .
2002 ) , a wide - area survey would be required in order to contain one by chance
. one promising method of locating rare structures is to target the fields of high - redshift active galactic nuclei ( agn ) . given their huge luminosities , such objects must contain a smbh even at very high redshifts , and therefore must represent some of the most massive objects in existence at their epochs .
they should thus act as signposts to rare high - density regions of the early universe , and can be observed efficiently with existing technology .
several groups have exploited this technique to search for proto - clusters forming around high - redshift radio galaxies ( e.g. kurk et al .
2000 ; pentericci et al .
2000 ; ivison et al .
2000 ; stevens et al . 2003 ; smail et al . 2003 ; de breuck et al .
2004 ; greve et al . 2007 ; venemans et al . 2007 ) and quasi - stellar objects ( qsos ; stevens et al . 2004 ; priddey , ivison & isaak 2008 ) .
most of these studies have concentrated on extreme objects , those with the highest luminosities and at the highest redshifts , and all have found over - densities of star - forming galaxies ( typically ly@xmath19 emitters or smgs ) in the fields of the agn . here
we extend this work to investigate the environments of typical qsos at submillimetre and mid - infrared wavelengths .
the qsos are selected over the redshift range @xmath0 where the star - formation rate density and accretion luminosity density of the universe peak .
this work is motivated by the desire to study the coupled formation of stellar mass and black holes in galaxies that should evolve to become typical ellipticals in moderately rich clusters by the present day . since the bulk of the field smg population
are also found at these redshifts ( chapman et al .
2003 , 2005 ) we can also compare the properties of galaxies found in different environments .
this paper presents submillimetre and mid - infrared imaging data for 5 qso fields .
we extract submillimetre catalogues and integral number counts , and compare these with results from blank field surveys .
we then characterise the mid - infrared properties of the qsos and the smgs discovered in their fields .
a hubble constant @xmath20 kms@xmath21mpc@xmath21 and density parameters @xmath22 and @xmath23 are assumed throughout .
[ cols="<,^,^,^,^,^ " , ] ( 8.0,3.0 ) ( 7.0,3.0 )
likely mid - infrared counterparts to the submillimetre galaxies are given in table [ table : spit ] .
for each field we searched within fixed radii of the extracted submillimetre positions .
these radii of 5 arcsec at @xmath24 and @xmath25 m , and 10 arcsec at @xmath26 m were chosen to approximately match the theoretical error circles on the submillimetre positions added in quadrature with the uncertainty on the mid - infrared positions , which are significant at @xmath26 m ( see ivison et al .
2007 for a full discussion ) .
the corrected @xmath27 statistic ( downes et al . 1986 ) is given for each source found within the search radii .
this quantity is given by @xmath28 $ ] where @xmath29 , @xmath30 , @xmath31 is the search radius , @xmath32 is the surface density of objects at the limiting flux density of the catalogue , @xmath33 is the surface density of objects brighter than @xmath34 where @xmath34 is the flux density of a potential counterpart galaxy located a distance , @xmath35 , away from the submillimetre position . a value of @xmath36 is typically used as a threshold to select robust radio ids ( e.g. ivison et al . 2002 , 2007 ) to submillimetre galaxies .
we select all mid - infrared sources found within our search radii as potential counterparts and use the @xmath27 statistic to distinguish between competing ids when more than one galaxy is found .
we note that almost all smgs listed in table [ table : spit ] have mid - infrared counterparts for which the computed @xmath27 value is @xmath37 in at least one waveband , and can therefore be considered robust based on this criterion .
the 4 qsos detected at submillimetre wavelengths have counterparts in all 3 _ spitzer _ bands .
in addition , we find counterparts in at least one waveband for @xmath38 per cent of the smgs .
postage stamp images of the mid - infrared counterparts are shown in figs [ fig : post1 ] and [ fig : post2 ] which can be found in appendix a. lcccccccclfield & i d & @xmath39 & cat # & @xmath40 & ra & dec & @xmath41 & @xmath27 & note + & & ( @xmath2 m ) & & ( @xmath2jy ) & ( j2000.0 ) & ( j2000.0 ) & ( arcsec ) & + rxj@xmath42 & @xmath43 & @xmath24 & @xmath44 & @xmath45 & @xmath46 @xmath47 @xmath48 & @xmath49 @xmath50 @xmath51 & @xmath52 & @xmath53 & qso + [ 1pt ] & @xmath43 & @xmath54 & @xmath55 & @xmath56 & @xmath46 @xmath47 @xmath57 & @xmath49 @xmath50 @xmath51 & @xmath58 & @xmath59 & qso + [ 1pt ] & @xmath43 & @xmath60 & @xmath61 & @xmath62 & @xmath46 @xmath47 @xmath63 & @xmath49 @xmath50 @xmath64 & @xmath65 & @xmath66 & qso + [ 1pt ] & @xmath67 & @xmath24 & @xmath68 & @xmath69 & @xmath46 @xmath47 @xmath70 & @xmath49 @xmath50 @xmath71 & @xmath72 & @xmath73 & + [ 1pt ] & @xmath67 & @xmath54 & @xmath74 & @xmath75 & @xmath46 @xmath47 @xmath76 & @xmath49 @xmath50 @xmath77 & @xmath78 & @xmath79 + [ 5pt ] rxj@xmath80 & @xmath43 & @xmath24 & @xmath81 & @xmath82 & @xmath83 @xmath84 @xmath85 & @xmath86 @xmath87 @xmath88 & @xmath89 & @xmath59 & qso + [ 1pt ] & @xmath43 & @xmath54 & @xmath74 & @xmath90 & @xmath83 @xmath84 @xmath91 & @xmath86 @xmath87 @xmath92 & @xmath93 & @xmath94 & qso + [ 1pt ] & @xmath43 & @xmath60 & @xmath95 & @xmath96 & @xmath83 @xmath84 @xmath97 & @xmath86 @xmath87 @xmath98 & @xmath99 & @xmath94 & qso + [ 1pt ] & @xmath100 & @xmath24 & @xmath101 & @xmath102 & @xmath83 @xmath84 @xmath103 & @xmath86 @xmath104 @xmath105 & @xmath106 & @xmath107 & + [ 1pt ] & @xmath100 & @xmath54 & @xmath108 & @xmath109 & @xmath83 @xmath84 @xmath110 & @xmath86 @xmath104 @xmath111 & @xmath112 & @xmath113 & + [ 1pt ] & @xmath100 & @xmath60 & @xmath114 & @xmath115 & @xmath83 @xmath84 @xmath116 & @xmath86 @xmath104 @xmath117 & @xmath118 & @xmath119 & + [ 1pt ] & @xmath120 & @xmath24 & @xmath121 & @xmath122 & @xmath83 @xmath84 @xmath123 & @xmath86 @xmath124 @xmath125 & @xmath126 & @xmath127 & + [ 1pt ] & @xmath120 & @xmath54 & @xmath128 & @xmath129 & @xmath83 @xmath84 @xmath130 & @xmath86 @xmath124 @xmath131 & @xmath132 & @xmath73 & + [ 1pt ] & @xmath133 & @xmath24 & @xmath134 & @xmath135 & @xmath83 @xmath84 @xmath136 & @xmath86 @xmath87 @xmath137 & @xmath138 & @xmath139 & + [ 1pt ] & @xmath133 & @xmath24 & @xmath140 & @xmath141 & @xmath83 @xmath84 @xmath142 & @xmath86 @xmath87 @xmath143 & @xmath144 & @xmath145 & + [ 1pt ] & @xmath133 & @xmath54 & @xmath146 & @xmath147 & @xmath83 @xmath84 @xmath148 & @xmath86 @xmath87 @xmath149 & @xmath150 & @xmath151 + [ 1pt ] & @xmath133 & @xmath54 & @xmath152 & @xmath153 & @xmath83 @xmath84 @xmath154 & @xmath86 @xmath87 @xmath143 & @xmath144 & @xmath155 + [ 5pt ] rxj@xmath156 & @xmath100 & @xmath24 & @xmath157 & @xmath158 & @xmath159 @xmath160 @xmath161 & @xmath162 @xmath163 @xmath164 & @xmath165 & @xmath113 & + [ 1pt ] & @xmath100 & @xmath54 & @xmath166 & @xmath167 & @xmath159 @xmath160 @xmath161 & @xmath162 @xmath163 @xmath168 & @xmath169 & @xmath119 & + [ 1pt ] & @xmath100 & @xmath60 & @xmath114 & @xmath170 & @xmath159 @xmath160 @xmath171 & @xmath162 @xmath163 @xmath172 & @xmath173 & @xmath94 & + [ 1pt ] & @xmath174 & @xmath24 & @xmath175 & @xmath176 & @xmath159 @xmath160 @xmath177 & @xmath162 @xmath163 @xmath178 & @xmath179 & @xmath180 & + [ 1pt ] & @xmath174 & @xmath54 & @xmath181 & @xmath182 & @xmath159 @xmath160 @xmath183 & @xmath162 @xmath163 @xmath184 & @xmath185 & @xmath186 + [ 1pt ] & @xmath187 & @xmath188 & @xmath1 & @xmath189 & @xmath159 @xmath160 @xmath190 & @xmath162 @xmath163 @xmath191 & @xmath192 & @xmath193 & likely false i d + [ 5pt ] rxj@xmath194 & @xmath43 & @xmath24 & @xmath195 & @xmath196 & @xmath197 @xmath198 @xmath199 & @xmath200 @xmath201 @xmath202 & @xmath203 & @xmath204 & qso + [ 1pt ] & @xmath43 & @xmath54 & @xmath205 & @xmath206 & @xmath159 @xmath198 @xmath207 & @xmath200 @xmath201 @xmath208 & @xmath209 & @xmath204 & qso + [ 1pt ] & @xmath43 & @xmath60 & @xmath1 & @xmath210 & @xmath159 @xmath198 @xmath211 & @xmath200 @xmath201 @xmath212 & @xmath213 & @xmath94 & qso + [ 1pt ] & @xmath174 & @xmath24 & @xmath214 & @xmath215 & @xmath159 @xmath198 @xmath216 & @xmath200 @xmath217 @xmath218 & @xmath219 & @xmath220 & + [ 1pt ] & @xmath174 & @xmath24 & @xmath221 & @xmath222 & @xmath159 @xmath198 @xmath223 & @xmath200 @xmath217 @xmath224 & @xmath225 & @xmath226 & + [ 1pt ] & @xmath174 & @xmath54 & @xmath227 & @xmath228 & @xmath159 @xmath198 @xmath223 & @xmath200 @xmath217 @xmath229 & @xmath230 & @xmath231 & + [ 1pt ] & @xmath187 & @xmath24 & @xmath232 & @xmath233 & @xmath159 @xmath198 @xmath234 & @xmath200 @xmath201 @xmath235 & @xmath118 & @xmath236 & + [ 1pt ] & @xmath187 & @xmath54 & @xmath237 & @xmath238 & @xmath159 @xmath198 @xmath239 & @xmath200 @xmath201 @xmath240 & @xmath241 & @xmath242 & + [ 1pt ] & @xmath67 & @xmath24 & @xmath243 & @xmath244 & @xmath159 @xmath198 @xmath245 & @xmath200 @xmath217 @xmath246 & @xmath247 & @xmath248 & + [ 5pt ] rxj@xmath249 & @xmath43 & @xmath24 & @xmath250 & @xmath251 & @xmath252 @xmath253 @xmath254 & @xmath255 @xmath256 @xmath257 & @xmath258 & @xmath259 & + [ 1pt ] & @xmath43 & @xmath54 & @xmath260 & @xmath261 & @xmath252 @xmath253 @xmath254 & @xmath255 @xmath256 @xmath262 & @xmath263 & @xmath264 & + [ 1pt ] & @xmath43 & @xmath60 & @xmath265 & @xmath266 & @xmath252 @xmath253 @xmath267 & @xmath255 @xmath256 @xmath268 & @xmath269 & @xmath270 & + [ 1pt ] & @xmath100 & @xmath24 & @xmath81 & @xmath271 & @xmath252 @xmath272 @xmath273 & @xmath255 @xmath274 @xmath275 & @xmath276 & @xmath186 & qso + [ 1pt ] & @xmath100 & @xmath54 & @xmath277 & @xmath278 & @xmath252 @xmath272 @xmath279 & @xmath255 @xmath274 @xmath280 & @xmath281 & @xmath242 & qso + [ 1pt ] & @xmath100 & @xmath60 & @xmath282 & @xmath283 & @xmath252 @xmath272 @xmath284 & @xmath255 @xmath274 @xmath285 & @xmath286 & @xmath119 & qso + [ 1pt ] & @xmath187 & @xmath24 & @xmath287 & @xmath288 & @xmath252 @xmath272 @xmath289 & @xmath255 @xmath274 @xmath290 & @xmath291 & @xmath204 & + [ 1pt ] & @xmath187 & @xmath54 & @xmath292 & @xmath293 & @xmath252 @xmath272 @xmath294 & @xmath255 @xmath274 @xmath295 & @xmath296 & @xmath204 & + [ 1pt ] & @xmath187 & @xmath60 & @xmath297 & @xmath298 & @xmath252 @xmath272 @xmath299 & @xmath255 @xmath274 @xmath300 & @xmath301 & @xmath66 & + [ 1pt ] & @xmath302 & @xmath24 & @xmath303 & @xmath304 & @xmath252 @xmath272 @xmath305 & @xmath255 @xmath274 @xmath218 & @xmath306 & @xmath307 & + [ 1pt ] & @xmath308 & @xmath24 & @xmath309 & @xmath310 & @xmath252 @xmath272 @xmath311 & @xmath255 @xmath312 @xmath313 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath24 & @xmath314 & @xmath315 & @xmath252 @xmath272 @xmath316 & @xmath255 @xmath312 @xmath317 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath24 & @xmath318 & @xmath319 & @xmath252 @xmath272 @xmath320 & @xmath255 @xmath312 @xmath321 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath54 & @xmath322 & @xmath323 & @xmath252 @xmath272 @xmath324 & @xmath255 @xmath312 @xmath325 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath54 & @xmath326 & @xmath327 & @xmath252 @xmath272 @xmath328 & @xmath255 @xmath312 @xmath329 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath54 & @xmath330 & @xmath331 & @xmath252 @xmath272 @xmath332 & @xmath255 @xmath312 @xmath333 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath54 & @xmath334 & @xmath335 & @xmath252 @xmath272 @xmath336 & @xmath255 @xmath312 @xmath337 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath54 & @xmath338 & @xmath339 & @xmath252 @xmath272 @xmath340 & @xmath255 @xmath312 @xmath341 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath60 & @xmath342 & @xmath343 & @xmath252 @xmath272 @xmath344 & @xmath255 @xmath312 @xmath345 & & & see appendix a + [ 1pt ] & @xmath308 & @xmath60 & @xmath61 & @xmath346 & @xmath252 @xmath272 @xmath347 & @xmath255 @xmath312 @xmath348 & & & see appendix a +
( 8.0,4.3 ) mid - infrared colours provide a powerful means of distinguishing between galaxies which contain agn and those that do not ( e.g. ivison et al .
2004 , 2007 ; egami et al .
luminous galaxies at @xmath8 hosting agn have power - law mid - infrared spectra whereas those without agn are dominated by starlight in the irac bands and dust / pah emission at 24 @xmath2 m . in fig .
[ fig : colours ] we plot the ratio of 24 and 8 @xmath2 m flux density against the ratio of 8 and 4.5 @xmath2 m flux density for those objects detected in at least two _ spitzer _ bands .
we superimpose redshift tracks for the local ulirgs mrk 231 ( agn - like sed ) and arp 220 and ngc 6240 ( starburst - like seds ) .
the qsos and smgs clearly inhabit a different regions of this diagram .
as expected the qsos have colours more similar to mrk 231 than to the starburst galaxies .
three qsos have remarkably similar mid - infrared colours given that they were selected on their x - ray properties .
the other qso , rx j163308 , has significantly different irac colours but we note from the postage stamp image ( fig .
[ fig : post1 ] ) that it has a close ` companion ' galaxy which is blended with the qso in our extracted catalogue and will thus contaminate the colours .
while the other three qsos have @xmath349 m flux density ratios roughly consistent with the mrk 231 template at their known redshifts we note that their 24/8.0 @xmath2 m flux density ratios appear to be lower than that of mrk 231 by a factor of @xmath350 which might be related to the relative strength of emission / absorption features .
all but one of the smgs on the other hand have mid - infrared colours consistent with arp 220 and/or ngc 6240 at @xmath8 .
the one exception has colours consistent with arp 220 at @xmath351 but its faintness and non - detection at 24 @xmath2 m are more consistent with a mrk 231-like object at @xmath8 .
it is thus the only smg in our sample for which mid - infrared colours indicate significant emission from an agn .
we thus find no strong evidence for buried agn in the smgs .
having said this , we note that deep x - ray ( e.g. vignati et al . 1999 ; netzer et al .
2005 ) and infrared spectroscopy ( e.g. lutz et al .
2003 ; armus et al .
2006 ) observations of ngc 6240 show that this galaxy does in fact host a highly obscured agn but that the energetics are dominated by the starburst .
this leaves open the possibility that many of the smgs contain buried agn which are not powerful enough to dominate over the starburst in the mid - infrared .
we have shown that the number counts of smgs with flux densities @xmath6 mjy in the @xmath352 arcmin diameter fields surrounding our x - ray absorbed qsos is higher than that found for a blank field .
most of these galaxies have mid - infrared colours very similar to the local ulirg arp 220 if it were placed at @xmath8 .
fitting the millimetre to far - infrared spectrum of arp 220 with a greybody and then integrating to determine the far - infrared luminosity we find l@xmath353 l@xmath14 which corresponds to a star - formation rate ( sfr ) of @xmath354 m@xmath14 yr@xmath21 after applying the usual scaling - law ( kennicutt 1998 ) . using the same greybody fit and applying the appropriate k - correction leads to predicted 850 @xmath2 m flux densities of 1.73 , 1.44 and 1.42 mjy for arp 220 at z=1 , 2 and 3 respectively .
the smgs in our sample have 850 @xmath2 m flux densities ranging from about @xmath355 mjy implying that their sfrs range between @xmath356 m@xmath14 yr@xmath21 if arp 220 is indeed a good analogue . therefore , if the starburst can be sustained for a few hundred million years or so then a stellar mass equivalent to that of the bulge of a large galaxy will be assembled .
it is natural to speculate that an over - dense region of space at high redshift containing galaxies that are capable of building large , homogeneous populations of stars will evolve over cosmic time to become clusters of galaxies containing a population of massive elliptical galaxies . since it is well established that elliptical galaxies contain dormant smbhs ( e.g. magorrian et al .
1998 ) , and that these must have assembled at high redshifts where the accretion luminosity density peaks , we expect that our smgs should contain growing black holes , i.e. agn .
how the major growth phase of the black hole coincides temporally with the luminous starburst that builds up the stellar mass is a current topic of debate .
indeed , models predict a range of possibilities ( e.g. archibald et al 2002 ; di matteo , springel & hernquist 2005 ) but there must be a link between the formation of the stars in the bulge and the smbh because of the tight correlation between their masses found for local galaxies ( ferrarese & merritt 2000 ) . after a number of early null results using moderate depth x - ray data ( e.g. fabian et al .
2000 ; waskett et al .
2003 ) deep x - ray ( alexander et al . 2005a , b , 2008 ) and mid - infrared spectroscopy ( menndez - delmestre et al .
2007 ; pope et al . 2008 ) observations show that smgs found in the field often do contain agn but that the masses of their black holes may be much smaller than those found in high redshift qsos , and consequently that the bulk of the far - infrared emission is due to the starburst .
moreover , estimates of the host galaxy masses suggest that the growth of the black hole lags the growth of the spheroid ( borys et al . 2005 ; alexander et al .
2008 ; see also coppin et al .
2009 ) . while this may be the case for smgs , analysis of massive galaxies selected on other criteria find evidence to the contrary , suggesting that more than one scenario is possible for the formation of a massive galaxy and its smbh .
for example , studies of powerful radio sources ( mclure et al .
2006 ) and luminous qsos ( peng et al .
2006 ) suggest that the black - hole mass to bulge luminosity ratio increases towards high redshifts thus indicating that the black hole reaches its critical mass before the galaxy bulge is fully formed .
the mid - infrared colours of our smgs also argue against the presence of a powerful agn in these objects unless they are very highly obscured .
they are thus either intrinsically different objects to the qso or they are caught at a different time in their evolution .
page et al .
( 2010 , in prep . ) argue that the x - ray absorption in the qsos has its origin in an ionized absorber associated with a powerful wind capable of terminating the star formation .
such feedback is a popular ingredient of models since it leads naturally to the relationship between smbh mass and spheroid mass discussed above ( e.g. fabian 1999 ) .
if the smgs in the same field evolve to form a cluster of ellipticals then they would presumably go through a similar evolutionary sequence to the qso but are caught at an earlier stage of their life cycle at @xmath8 .
estimates of the stellar masses would be useful to put further constraints on their evolutionary status .
to this end we are currently fitting models to optical through infrared data to determine photometric redshifts and restframe @xmath357-band luminosities ; an analysis of the rxj 094144 field is currently in progress ( carrera et al .
2010 , in prep . ) .
one difference worthy of note between the findings of this paper and those for high - redshift ( typically @xmath358 ) radio galaxies ( hzrgs ) concerns very bright smgs in their fields .
two of the seven hzrg fields imaged with scuba contain smgs with @xmath359 mjy ( stevens et al . 2003 ) whereas the brightest object found around the qsos has @xmath360 mjy .
although the samples are too small to give a statistically significant result it appears that the smgs detected around the lower redshift , less powerful , radio - quiet agn have lower @xmath361 .
a direct comparison is not informative because the sources were extracted with different signal - to - noise criteria and by different methods but comparing flux densities significantly above the catalogue cut - offs yields @xmath342 smgs with @xmath362 mjy in the 7 hzrg fields while only @xmath363 are detected around the 5 qsos .
given the effectively flat relationship between @xmath361 and redshift for @xmath364 this difference in flux density leads to a proportional difference in star - formation rates ( all other things being equal ) .
there is thus reasonably good evidence for a difference in star - formation activity in individual galaxies in these fields indicating that the galaxy formation process might be dependent on the magnitude of the density peak and/or its redshift .
for example , in the case of hzrgs it was found that the detection rate increased dramatically at @xmath365 ( archibald et al .
2001 ; reuland et al .
2004 ) which is significantly higher than the average redshift of our qso sample .
another possibility is that the radio properties of the central agn are in some way linked to environmental density ( e.g. kauffmann , heckman & best 2008 ) .
indeed , recent work by falder et al .
( 2010 ) finds evidence for larger over - densities of galaxies around radio - loud objects ( based on work carried out at 3.6 @xmath2 m ) .
perhaps radio synchrotron luminosity is enhanced in regions where the inter - galactic medium is more dense .
an obvious next step will be to conduct a statistically meaningful survey of fields around agn over the full range of redshifts in order to track star - formation activity as a function of cosmic time .
the next generation bolometer camera on the jcmt ( scuba-2 ) and spire / pacs on - board the _ herschel space observatory _ are instruments with good enough sensitivity to make such a campaign practical .
it would also allow a bigger area to be mapped around each agn , better matching the size of proto - cluster regions predicted by numerical simulations .
we thank ian smail for extensive comments on the draft manuscript and mark thompson for useful discussions .
the james clerk maxwell telescope is operated by the joint astronomy centre on behalf of the science and technology facilities council of the united kingdom , the netherlands organisation for scientific research , and the national research council of canada .
jcmt data were taken under project ids m03au46 , m03bu32 and m04bu14 .
this work is based [ in part ] on observations made with the spitzer space telescope , which is operated by the jet propulsion laboratory , california institute of technology under a contract with nasa .
support for this work was provided by nasa through an award issued by jpl / caltech .
, m.j.p . and f.j.c .
acknowledge support from the royal society . f.j.c .
acknowledges further support from the spanish ministerio de educacin y ciencia under project esp2006 - 13608 .
alexander d.m . , smail i. , bauer f.e . ,
chapman s.c . ,
blain a.w . , brandt w.n . ,
ivison r.j . , 2005a , nat , 434 , 738 alexander d.m . , bauer f.e
. , chapman s.c . , smail i. , blain a.w . , brandt w.n .
, ivison r.j . , 2005b , apj , 632 , 736 alexander d.m .
et al . , 2008 ,
aj , 135 , 1968 archibald e.n . ,
dunlop j.s . , hughes d.h . , rawlings s. , eales s.a . , ivison r.j .
, 2001 , mnras , 323 , 417 archibald e.n . , dunlop j.s . , jimenez r. , friaa a.c.s .
, mclure r.j . ,
hughes d.h . , 2002 , mnras , 336 , 353 armus l. et al . , 2006 ,
apj , 640 , 204 austermann j.e .
et al . , 2010 ,
mnras , 401 , 160 bertin e. , arnouts s. , 1996 , a&as , 117 , 393 bertoldi f. et al . , 2007 , apjs , 172 , 132 borys c. , smail i. , chapman s.c .
, blain a.w.,alexander d.m . ,
ivison r.j . , 2005 , apj , 635 , 853 chapman , s.c . ,
blain a.w .
, ivison r.j . ,
smail i.r . , 2003 , nat , 422 , 695 chapman , s.c . ,
blain a.w . , smail i. , ivison r.j . , 2005 ,
apj , 622 , 772 coppin k. , halpern m. , scott d. , borys c. , chapman s. , 2005 , mnras , 357 , 1022 coppin k. et al . , 2006 , mnras , 372 , 1621 coppin k. et al . , 2009 , mnras , 389 , 45 de breuck c. et al . , 2004 , a&a , 424 , 1 di matteo t. , springel v. , hernquist l. , 2005 , nat , 433 , 604 downes a.j.b .
, peacock j.a . , savage a. , carrie d. r. , 1986 , mnras , 218 , 31 eales s.a . ,
lilly s. , gear w. , dunne l. , bond j.r .
, hammer f. , le fvre o. , crampton d. , 1999 , apj , 515 , 518 egami e. et al . , 2004 , apjs , 154 , 130 evrard a.e .
et al . , 2002 ,
apj , 573 , 7 fabian a.c .
, 1999 , mnras , 308 , l39 fabian a.c .
2000 , mnras , 315 , l8 falder j.t .
2010 , mnras , in press ( arxiv:1001.5419 ) farrah d. et al . , 2006 ,
apj , 643 , l139 ferrarese l. , merritt d. , 2000 , apj , 539 , l9 greve t.r . ,
ivison r.j . , bertoldi f. , stevens j.a . , dunlop j.s . , lutz d. , carilli c.l . , 2004 , mnras , 354 , 779 greve t.r
, stern d. , ivison r.j . , de breuck c. , kovcs a. , bertoldi f. , 2007 , mnras , 382 , 48 holland w.s .
et al . , 1999 , mnras , 303 , 659 hopkins p.f . , hernquist l. , cox t.j . , kere d. , 2008 , apjs , 175 , 356 hughes d.h .
et al . , 1998 , nat , 394 , 241 ivison r.j .
et al . , 2002 ,
mnras , 337 , 1 ivison r.j .
et al . , 2004 ,
apjs , 154 , 124 ivison r.j .
et al . , 2007 , mnras , 380 , 199 jenness t. , stevens j.a . ,
archibald e.n . , economou f. , jessop n.e .
, robson e.i .
, 2002 , mnras , 336 , 14 kauffmann g. , haehnelt m. , 2000 , mnras , 311 , 576 kauffmann g. , heckman t.m . ,
best p.n . , 2008 , mnras , 384 , 953 kennicutt r.c . , 1998 , apj , 498 , 541 kurk j.d .
et al . , 2000 , a&a , 358 , l1 lacy m. et al . , 2005 , apjs , 161 , 41 lutz d. , sturm e. , genzel r. , spoon h.w.w .
, moorwood a.f.m . , netzer h. , sternberg a. , 2003 , a&a , 409 , 867 magorrian j. et al . , 1998 ,
aj , 115 , 2285 menndez - delmestre k. et al . , 2007 , apj , 655 , l65 mclure r.j . , jarvis m.j . , targett t.a . , dunlop j.s
, best p.n . , 2006 , mnras , 368 , 1395 mortier a.m.j .
et al . , 2005 ,
mnras , 363 , 563 netzer h. , lemze d. , kaspi s. , george i.m .
, turner t.j .
, lutz d. , boller t. , chelouche d. , 2005 , apj , 629 , 739 page m.j .
, mittaz j.p.d . ,
carrera f.j . , 2001 , mnras , 325 , 575 page m.j .
, stevens j.a .
, mittaz j.p.d .
, carrera f.j . , 2001 , science , 294 , 2516 peng c.y .
, impey c.d .
, rix h .- w .
, kochanek c.s . ,
keeton c.r . ,
falco e.e . , lehr j. , mcleod b.a . , 2006 ,
apj , 649 , 616 pentericci l. et al . , 2000 ,
a&a , 361 , l25 pope a. et al .
, 2008 , apj , 675 , 1171 priddey r.s .
, ivison r.j . ,
isaak k.g . , 2008 , mnras , 383 , 289 reuland m. , rttgering r.
, van breugel w. , de breuck , c. , 2004 , mnras , 353 , 377 scott s.e et al . , 2002 ,
mnras , 331 , 817 seymour n. et al .
, 2007 , apjs , 171 , 353 smail i. , ivison r.j . , blain a.w . , 1997 ,
apj , 490 , l5 smail i. , ivison r.j . ,
blain a.w .
, kneib j .-
, 2002 , mnras , 331 , 495 smail i. , scharf c.a .
, ivison r.j .
, stevens j.a . , bower r.g . ,
dunlop j.s . , 2003 , apj , 599 , 86 stevens j.a .
et al . , 2003 ,
nat , 425 , 264 stevens j.a .
, page m.j .
, ivison r.j .
, smail i. , carrera f.j . , 2004 , apj , 604 , l17 stevens j.a .
, page m.j . ,
ivison r.j . ,
carrera f.j .
, mittaz j.p.d . , smail i. , mchardy i.m . , 2005 , mnras , 360 , 610 venemans b.p .
et al . , 2007 , a&a , 461 , 823 vignati p. et al . , 1999 ,
a&a , 349 , l57 waskett t.j .
et al . , 2003 ,
mnras , 341 , 1217 wei a. et al . , 2009 ,
apj , 707 , 1201
in this section we present and discuss postage stamp images of the qsos and smgs . for the rx j@xmath366 field
we compare the mid - infrared counterparts to the smgs with those identified previously at optical and near - infrared wavelengths .
we also discuss cases where likely mid - infrared counterparts were not catalogued because they fell just outside our formal search radii , and one case where an obvious smg was not found by our extraction algorithm ; these sources were not included in the analysis described in the main part of this paper .
the scuba data for this field were published previously by stevens et al .
( 2004 ) . in the present paper
we have treated the data slightly differently in the final reduction stages ( rebinning and smoothing ) .
we also apply a formal source extraction algorithm in this work resulting in only 2 sources at 850 @xmath2 m and 4 at 450 @xmath2 m whereas stevens et al .
( 2004 ) simply identified peaks on the maps above a 3-@xmath367 signal - to - noise threshold , identifying 6 sources .
as already noted in section [ section : se ] the source extraction algorithm fails to detect sources that are blended which has undoubtedly affected the 850 @xmath2 m extraction . at 450 @xmath2 m we can see that sources 450#1 and 450#3 appear more elongated than the other two objects hinting that they may be resolved .
it was these objects that stevens et al .
( 2004 ) identified as blends .
inspection of the postage stamps shows that source * 850#2/450#2 * is coincident with a bright mid - infrared source detected in all 3 _ spitzer _ channels .
this object was one of a pair of sources detected in the @xmath368- and @xmath357-bands by stevens et al .
spitzer _ data show that the correct identification is the object with the redder @xmath369 colour since this object becomes progressively more prominent with increasing wavelength compared to its neighbour and is detected at 24 @xmath2 m .
source * 450#3 * falls in a crowded field .
we find 2 possible identifications in the irac channels .
of these source 206/160 lies closer to the 450 @xmath2 m source centroid and is also coincident with faint ( formally undetected ) emission at 24 @xmath2 m .
source 206/160 was not one of the possible identifications discussed by stevens et al .
( 2004 ) since it is very faint / undetected in the @xmath357-band .
we conclude that the likely identification for this smg is the very red object 206/160 .
source * 450#4 * was matched with an extremely red object ( @xmath370 ) by stevens et al .
this source is not formally recovered in this work because it lies outside our 5 arcsec irac search radius but is visible in the postage stamp to the west of the 450 @xmath2 m centroid as a bright source at 4.5 and 8.0 @xmath2 m .
given its red colours we consider this object as a likely identification .
measured flux densities are @xmath371 and @xmath372 @xmath2jy at 4.5 and 8.0 @xmath2 m respectively giving a colour similar to arp 220 at the redshift of the qso .
the 4.5-@xmath2 m centroid is 6.6 arcsec from the extracted 450 @xmath2 m position at @xmath373 , @xmath374 while the 8.0-@xmath2 m centroid is 6.4 arcsec from the extracted 450 @xmath2 m position at @xmath375 , @xmath376 .
the calculated p - statistic is 0.089 and 0.064 at 4.5 and 8.0 @xmath2 m respectively . in section [ section : catmem ] we noted that the qso in this field was not found by our extraction algorithm but was detected by page et al .
the latter observations used scuba in ` photometry ' mode where the central bolometer of each array is centred on the target but no image is made .
inspection of fig .
[ fig : allmaps ] shows that while there is submillimetre emission at the location of the qso ( located below source 5 on the 850 @xmath2 m image ) the surrounding structure is complex .
it is thus possible that the source extraction algorithm failed to locate a significant peak because of blending .
it is also likely that the photometry observation of this object reported by page et al .
( 2001 ) is contaminated by emission from the smgs close to the qso . in table
[ table : spit ] we noted that object * 850#5 * has a likely false identification at 24 @xmath2 m .
this can also be attributed to the crowded field close to the qso since it is the qso itself that has fallen within the search radius at 24 @xmath2 m while it was not picked up in the smaller search radii adopted for the irac channels .
source * 850#2 * has no listed identification in table [ table : spit ] but the postage stamp images show bright 4.5 and 8.0 @xmath2 m sources close to the 850 @xmath2 m centroid .
the closest 4.5 @xmath2 m source is at @xmath378 , @xmath379 which at 5.3 arcsec from the smg is only just outside our formal search radius .
this object has a flux density of @xmath380 @xmath2jy and is coincident with an 8.0 @xmath2 m source although the latter is not detected by sextractor .
inspection of the 8.0 @xmath2 m image reveals that the diffraction spike from the bright qso at the field centre falls on top of this object probably explaining the non - detection .
although the p - statistic for the 4.5 @xmath2 m source is high at 0.259 we consider this object as a likely identification because of its detection in both irac bands .
source * 850#4 * has no listed identification in table [ table : spit ] but the postage stamp images show a bright 4.5 and 8.0 @xmath2 m source close to the 850 @xmath2 m centroid along with faint ( formally undetected ) emission at 24 @xmath2 m .
the 4.5 @xmath2 m source is located at @xmath381 , @xmath382 only 5.2 arcsec from the smg , has a flux density of @xmath383 and a p - statistic of 0.266 .
the 8.0 @xmath2 m source is located at @xmath384 , @xmath385 only 5.9 arcsec from the smg , has a flux density of @xmath386 and a p - statistic of 0.198 .
this object has irac colours similar to arp 220 at the redshift of the qso .
the proximity of this object to the smg together with its detection in in both irac bands make it a likely identification .
inspection of the 850 @xmath2 m map for this field ( fig .
[ fig : allmaps ] ) shows submillimetre emission north of the qso at approximately @xmath387 , @xmath388 . there is a negative region to the east of this position
which is consistent with a negative image of the source located at one of the off - positions .
the western off - position would fall off the map .
we believe that this is a real smg ( probably two sources ) which was missed by the extraction algorithm because of its proximity to the edge of the map .
inspection of the postage stamp images reinforces this belief since we find two 24 @xmath2 m sources coincident with the smg position and at the same approximate position angle as the 850 @xmath2 m emission .
the irac data reveal 3 and 5 sources at 4.5 and 8.0 @xmath2 m respectively .
the southern 24/850 @xmath2 m source has two possible irac identifications .
of these , source 260/232 is the brightest and reddest object ; its @xmath349 m flux density ratio is @xmath389 . if source 10 is the 24 @xmath2 m counterpart then the ratio of flux densities at 24 and 8.0 @xmath2 m is @xmath390 which would place this object on the mrk 231 curve shown in fig [ fig : colours ] indicating that it hosts a buried agn , possibly at a lower redshift than rx j163308 .
the northern 24/850 @xmath2 m source has 1 counterpart at 4.5 @xmath2 m and 3 at 8.0 @xmath2 m although the 4.5 @xmath2 m emission is extended in the same manner as that at 8.0 @xmath2 m implying that this smg could have multiple mid - infrared counterparts .
inspection of the flux densities presented in table [ table : spit ] shows that the irac colours are similar to those of arp 220 at the redshift of the qso so this source is unlikely to host an agn .
since this system is one of the most interesting in our sample we include it in table [ table : spit ] as source * 850#7 * although it is not included in the 850 @xmath2 m number counts analysis because it was not found by the extraction algorithm . | we present submillimetre and mid - infrared imaging observations of five fields centred on quasi - stellar objects ( qsos ) at @xmath0 .
all @xmath1 qsos were detected previously at submillimetre wavelengths . at 850 ( 450 )
@xmath2 m we detect 17 ( 11 ) submillimetre galaxies ( smgs ) in addition to the qsos .
the total area mapped at 850 @xmath2 m is @xmath3 arcmin@xmath4 down to rms noise levels of @xmath5 mjy / beam , depending on the field .
integral number counts are computed from the 850 @xmath2 m data using the same analytical techniques adopted by ` blank - field ' submillimetre surveys .
we find that the ` qso - field ' counts show a clear excess over the blank - field counts at deboosted flux densities of @xmath6 mjy ; at higher flux densities the counts are consistent with the blank - field counts .
robust mid - infrared counterparts are identified for all four submillimetre detected qsos and @xmath7 per cent of the smgs .
the mid - infrared colours of the qsos are similar to those of the local ulirg / agn mrk 231 if placed at @xmath8 whilst most of the smgs have colours very similar to those of the local ulirg arp 220 at @xmath8 .
mid - infrared diagnostics therefore find no strong evidence that the smgs host buried agn although we can not rule out such a possibility .
taken together our results suggest that the qsos sit in regions of the early universe which are undergoing an enhanced level of major star - formation activity , and should evolve to become similarly dense regions containing massive galaxies at the present epoch .
finally , we find evidence that the level of star - formation activity in individual galaxies appears to be lower around the qsos than it is around more powerful radio - loud agn at higher redshifts .
galaxies : formation - galaxies : starburst - galaxies : evolution - galaxies : high redshift - submillimetre - qso : general |
a bright soft x - ray enhancement in the monoceros and gemini constellations was first detected in rocket experiments over thirty years ago @xcite , and resolved into a shell - like structure by _
heao-2 _ a decade later @xcite .
spectral studies with _ rosat _ confirmed that this so - called ` monogem ring ' is a supernova remnant , probably in the adiabatic expansion phase @xcite .
the distance is poorly constrained by evolutionary arguments self - consistent sedov - taylor models were found by @xcite at all distances between 100 and 1300 pc but distances around 300 pc are preferred on grounds of supernova energetics . at this distance , the model age is 86,000 years , explosion energy is @xmath1 erg , and current radius is 66 pc .
the inferred interstellar medium density is @xmath2 , typical of the hot interstellar medium .
the radio pulsar psr b0656 + 14 lies very close in projection to the center of the monogem ring , and an association between the objects seems natural @xcite .
the pulsar is young , with a characteristic spin - down timescale of 110 kyr @xcite , pulsed non - thermal optical and x - ray emission , and unpulsed thermal x - ray emission consistent with a @xmath3 year - old cooling neutron star @xcite .
but because the pulsar distance was estimated from interstellar dispersion measurements @xcite to be 760 pc more than twice the best estimate for the remnant and a proper motion measurement appeared to show the pulsar moving towards the center of the remnant @xcite , a physical association has been widely regarded as unlikely ( * ? ? ?
* for example ) .
new very long baseline interferometric measurements ( brisken et al .
2003 ) lead us to reconsider this negative conclusion .
the distance , now accurately known from parallax to be @xmath4 pc , is much lower than previously believed , and the proper motion , 44 mas / yr , is about 40% smaller . as we will show , these measurements , together with a re - examination of the x - ray data , convincingly demonstrate that the pulsar and remnant were born in a single supernova explosion about @xmath3 yrs ago .
the monogem ring ( fig .
1 ) shows significant deviations from circular symmetry , with what may be a blow - out to the west , away from the galactic plane , and a missing quadrant in the northwest , where foreground absorption is relatively high @xcite . because of the lack of symmetry , the morphological center of the ring is somewhat uncertain .
for example , at low energy ( e.g. , the _ rosat _ r1 band image displayed as fig . 6 in plucinsky et al .
1996 ) , the portion of the ring in the galactic plane is not readily visible , making the remnant appear smaller and offset slightly away from the plane ; the arc is evident in higher energy observations ( fig .
1 , and also note fig . 8 of plucinsky et al .
1996 ) , suggesting the importance of greater absorption on lines of sight near the plane . in projection , the pulsar , at @xmath5 , lies well within the ring . to test the consistency of its position with the morphological center of the ring
, we have fitted ` by eye ' a circle to the incomplete ring , centered on the pulsar .
this fit , shown in fig . 1 , demonstrates that the pulsar s position is not inconsistent with the morphological center of the ring , within admittedly large uncertainties . with the measured proper motion , 44 mas / yr
, the pulsar has moved about @xmath6 in @xmath3 yrs , from roughly @xmath7 .
this inferred birth position appears slightly further from the current center of the ring , but not unacceptably so .
it is also important to distinguish the morphological center of the remnant from its true dynamical center : we should expect an old remnant expanding into a medium with density decreasing away from the galactic plane to have its apparent center offset from its explosion center in the direction away from the plane , just as observed @xcite . in the end
, we conclude that the pulsar position at birth was within a few degrees of the current morphological center of the ring .
we note that the _ a priori _ likelihood of an unrelated background source projected within the ring falling this close to the center is about 5% .
however , on its own the positional coincidence is a strong argument neither for nor against an association .
the strongest argument against a physical association between the pulsar and remnant has been the pulsar distance , estimated at 760 pc from the measured column density of free electrons .
sedov modeling formally allows a wide range of distances for the monogem ring , but a distance of about 300 pc has been preferred .
although estimates of pulsar distances through interstellar dispersion are fairly crude ( the quoted uncertainty was 25% @xcite ) the distance discrepancy appeared significant . the small parallax distance , @xmath4 pc , thus came as a surprise .
( we note , though , that the greater distance implied an unacceptably large thermal x - ray luminosity : an attempt to directly estimate the distance from x - ray modeling yielded @xmath8 pc or less @xcite . ) with this precise pulsar distance available , we consider whether the distance estimate for the remnant can also be improved .
a strong interstellar o vi absorption feature at @xmath9 km / s has been detected in the star 15 mon , in the southern region of the ring @xcite .
the parallax distance of 15 mon is @xmath10 pc @xcite .
because o vi at such high velocity is rarely found in the disk , and the velocity is similar to that expected from a remnant at this stage , we consider this to be a secure upper limit to the distance of the monogem ring .
other evidence suggests the ring is no farther than the pulsar .
the absorption is low , @xmath11 @xcite , compared to @xmath12 for the pulsar @xcite .
the pulsar dispersion measure is much higher than expected at a distance of 290 pc , suggesting a line of sight through a highly ionized region .
it is also unlikely that the remnant is significantly closer than the pulsar . at an assumed distance of 300 pc ,
the inferred explosion energy is already low : @xmath1 erg .
a distance below 230 pc would imply an implausibly low energy , below @xmath13 erg .
we conclude that the pulsar and remnant distances agree to within @xmath14 pc .
indeed , since the shell radius is @xmath15 pc , the evidence is strong that the pulsar is currently within the expanding supernova shell .
finally , we must consider the relative ages of the pulsar and remnant .
the remnant age , from sedov modeling , has been estimated at 86,000 yrs for a distance of 300 pc .
two lines of evidence suggest a comparable age for the neutron star .
first , its current temperature is in good agreement with standard cooling models for a @xmath3 yr old neutron star @xcite .
second , the characteristic spin - down timescale for the pulsar is @xmath16 yrs .
the second argument deserves more comment .
the age of a pulsar slowing with a torque proportional to a constant power @xmath17 of its frequency , @xmath18 , can be expressed as @xmath19,\ ] ] where @xmath20 is the initial spin period . for magnetic - dipole braking ,
the `` braking index '' @xmath21 , so if the pulsar is born spinning much faster than its current period , @xmath22 , while the age is overestimated if the pulsar were born near its current period . for a handful of very young pulsars , where the braking index can be measured directly ,
@xmath23 @xcite . for middle - aged pulsars , the good agreement between timing ages and so - called kinetic ages ( @xmath24 , where @xmath25 is the height above the galactic plane ) requires a mean braking index near 3 @xcite .
we acknowledge that the timing age can be misleading .
several putative pulsar - snr associations suggest large discrepancies between timing and true ages : for example , psr j1012@xmath265226 , in the @xmath27 kyr old remnant pks 1209@xmath2651/52 , has a spin - down age of @xmath28 kyr @xcite , and psr j1811@xmath261925 , with a spin - down age of 24 kyr , has been associated with the snr of ad 386 @xcite .
nevertheless , a recent comparison of kinetic and timing ages for 21 pulsars under 10 myr in age found a typical discrepancy of only @xmath29 @xcite . we conclude that both the timing and cooling ages of the pulsar are consistent with the remnant age .
the angular position , distance , and age of the pulsar and remnant are all in excellent agreement with the hypothesis that both were born at the same location , @xmath30 pc away , @xmath3 yrs ago . the likelihood that this is merely a positional and temporal coincidence is extremely low . crudely estimating the galactic type ii supernova rate at one per century and the region within 300 pc of earth as @xmath31 of the area of the galactic disk , the mean supernova interval in this region is about 100 kyr .
the chance that two unrelated supernovae occur within @xmath32 kyr , within @xmath33 on the sky , and within 50 pc in radial distance is vanishingly small .
perhaps the most persuasive argument for an association comes from assuming the opposite . if this @xmath3 yr old remnant is not that of the supernova that formed the @xmath3 yr old pulsar , then where is the remnant of that supernova ?
many old remnants are of course invisible because of distance or environment , but if the monogem ring and pulsar were formed in different supernovae then they occurred in very close physical and temporal proximity .
the monogem ring itself is evidence that a remnant of age @xmath3 years expanding into this particular environment is visible at 300 pc .
we conclude that a single supernova , 300 pc away and a hundred thousand years ago , formed both psr b0656 + 14 and the monogem ring .
supernova remnants are generally believed to be acceleration sites for cosmic rays @xcite , and in this case it is intriguing to note that the monogem ring may hold a clue to a long - standing puzzle : the origin of the steepening in the primary cosmic - ray spectrum at about @xmath34 ev ( 3 pev ) , called the ` knee . ' between about @xmath35 and a few times @xmath36 ev , the differential cosmic ray energy spectrum is well - described by a broken power law , proportional to below the knee and above the knee .
( a recent review can be found in @xcite . )
possible explanations for the knee are extremely varied , including loss of the most energetic particles from the galaxy , unknown physical processes in the development of the atmospheric shower through which the cosmic rays are detected , or , most likely , a termination in the acceleration process .
a typical cut - off energy for acceleration in supernova remnants is @xmath37 ev , where @xmath38 is the nuclear charge , though the exact cut - off will vary by perhaps an order of magnitude with variations in the explosion energy , magnetic field , interstellar medium density , and age @xcite .
recently , erlykin and wolfendale ( ew ) have drawn attention @xcite to the sharpness of the knee feature , which has been a challenge for models in which the break arises from propagation effects or from a stochastic superposition of multiple sources with varying high - energy cut - offs .
ew propose that around the knee a single nearby source dominates the cosmic ray spectrum , which is otherwise a smooth superposition of the contributions from many supernova remnants throughout the galaxy and from whatever sources provide the higher energy cosmic rays , up to @xmath39 ev and beyond . at energies of a few pev ,
this single source alone produces 60% of the flux at earth .
( this possibility has also recently been considered by other authors ( * ? ? ? * for example ) . )
although the data are not yet conclusive , ew have identified the knee as most likely due to oxygen nuclei , with a smaller second break at about 10 pev due to iron nuclei .
the shape and amplitude of the knee feature have led ew to predict @xcite that the single source is a 90100 kyr old supernova remnant between 300 and 350 pc from earth , expanding in an under - dense medium .
the match with the properties of the monogem ring is striking , though almost certainly in part coincidental considering the large remaining uncertainties in cosmic ray acceleration and diffusion models .
nevertheless , an important objection to the single source model for the knee feature has been the lack of a suitable source @xcite .
that objection now appears to have been removed .
is supported by the nsf under grant ast-0098343 , r.a.b .
is supported by nasa atp grant nag5 - 12128 , and a.g .
is supported by enterprise ireland under grant sc/2001/322 .
this research has made use of the _ rosat _ all - sky survey data , which have been processed at mpe , and of the telescopes of the national radio astronomy observatory , a facility of the nsf operated under cooperative agreement by associated universities , inc .
we thank an anonymous referee for suggestions that improved the manuscript . , m. a. c. , lindegren , l. , kovalevsky , j. , hoeg , e. , bastian , u. , bernacca , p. l. , crz , m. , donati , f. , grenon , m. , van leeuwen , f. , van der marel , h. , mignard , f. , murray , c. a. , le poole , r. s. , schrijver , h. , turon , c. , arenou , f. , froeschl , m. , & petersen , c. s. 1997 , a&a , 323 , l49 | the monogem ring is a bright , diffuse , @xmath0-diameter supernova remnant easily visible in soft x - ray images of the sky .
projected within the ring is a young radio pulsar , psr b0656 + 14 .
an association between the remnant and pulsar has been considered , but was seemingly ruled out by the direction and magnitude of the pulsar proper motion and by a distance estimate that placed the pulsar twice as far from earth as the remnant . here we show that in fact the pulsar was born very close to the center of the expanding remnant , both in distance and projection .
the inferred pulsar and remnant ages are in good agreement .
the conclusion that the pulsar and remnant were born in the same supernova explosion is nearly inescapable .
the remnant distance and age are in remarkable concordance with the predictions of a model for the primary cosmic ray energy spectrum in which the ` knee ' feature is produced by a single dominant source . |
the existence of a low - lying state in @xmath0th , with an estimated natural linewidth , @xmath7 , in the range @xmath8 mhz @xcite , was indirectly established over 3 decades ago @xcite .
however , direct searches for emission from this m1 transition have so far been unsuccessful @xcite .
recently , a new indirect measurement has established that the transition energy is 7.8 @xmath9 0.5 ev @xcite , rather than 3.5 ev as orginally thought .
although well into the vacuum ultraviolet ( vuv ) , this energy is amenable to study by laser spectroscopy , opening the door for use of this transition as a highly stable oscillator in next generation optical clocks . the transition is also predicted to show a 10@xmath10-fold enhancement in sensitivity to the time variation of several fundamental constants@xcite , _
e.g. _ the fine structure constant . while the current measurement uncertainty of the nuclear transition energy is very good by nuclear physics standards ,
it must be drastically improved in order to realize the exciting applications of this transition .
we have recently proposed a novel approach to _ directly _ measure the transition energy to a much higher precision by interrogating @xmath0th that has been doped into a vuv transparent crystal @xcite . using this approach more than 10@xmath11
th nuclei can be interrogated , which allows for fluorescence rates sufficient to directly measure the transition energy using a broadband synchrotron light source . in this article , we present measurements of inherent vuv - induced background signals which could mask fluorescence from the isomer state in @xmath0th for several candidate host crystals . while the data shows that many of the crystals display some spurious fluorescence , the fluorescence is in general of a low level and short - lived compared to that expected from the isomeric transition .
of all the crystals measured , we have determined licaf ( licaalf@xmath5 ) to be the most suitable as it yields very little background fluorescence when it is directly illuminated with vuv light .
additonally , it shows resilience to vuv - induced radiation damage and has been grown with @xmath6th as a dopant at the 0.05% level .
it also has several key structural properties that mitigate broadenings of the isomeric transition energy in a crystal environment .
the first steps towards realizing the exciting applications of this transition are identifying suitable host crystals and determining the transition frequency to a greater precision .
the key requirements for the host crystal are that it be reasonably transparent in the vuv , have a pure crystalline structure , and chemically accept th@xmath12 ions .
numerous candidate hosts can be identified in the extensive family of fluoride crystals .
potential crystals include na@xmath4thf@xmath5 , licaf ( licaalf@xmath5 ) , lisaf ( lisralf@xmath5 ) , ylf ( ylif@xmath13 ) , and caf@xmath4 , which can be grown by various techniques , _
e.g. _ czochralski , bridgman , etc .
while these crystals have band - gaps of @xmath1410ev , impurities and color centers can cause residual absorption and may interfere with the @xmath0th optical spectroscopy .
various metal ions as well as oxygen - based impurities at the part - per - million level can lead to absorption coefficients of up to 0.1@xmath15 @xcite . to ensure crystal transparency , purification methods
must be employed to reduce these impurities to part - per - billion levels @xcite .
we have applied these methods to the preparation of high purity metal fluoride precursors for crystal growth .
so far , we have studied crystals that were obtained commercially from ac materials . these include na@xmath4thf@xmath5 ,
th : licaf , th : ylf , and th : nayf which were grown using the less expensive , but chemically identical , @xmath6th isotope .
the th : nayf and th : ylf were grown with a target doping of 0.1% molar while the th : licaf was grown with a target doping of 1% molar .
these crystals are good control crystals as the isomeric transition is not present in @xmath6th , but are chemically identical and thus the doping process as well as all vuv induced backgrounds are the same .
they therefore allow us to study which crystals are most suitable for higher precision measurements of the isomeric transition energy .
as previously mentioned , a major advantage to our approach for measuring the isomer energy is the large number of @xmath0th nuclei that can be addressed .
this allows for fluorescence rates sufficient for the use of standard spectroscopic techniques to determine the transition energy to a higher precision .
we plan to use a multi - phase experimental approach @xcite to investigate the ulimate precision to which this transition energy can be measured . the first phase will be a low resolution study using the 9.0.2.1 beam line at the advanced light source ( als ) at lawrence berkeley labs .
the als provides a tunable vuv light source ( 5 ev-30 ev ) with a linewidth ( @xmath16 ) of 0.175 ev , a photon flux ( @xmath17 ) of @xmath1810@xmath19 photons/@xmath20s , and a 170 @xmath21 m @xmath22 50 @xmath21 m spot size . assuming both the light ( @xmath16 ) and broadened transition ( @xmath23 ) linewidths to be lorentzian , then far from saturation , the excitation rate , @xmath24 , is given by @xmath25 where @xmath26 ( @xmath27 ) is the transition frequency ( wavelength ) , @xmath28 is the center frequency of the als beam , and @xmath29 is the total number of ground state @xmath0th nuclei addressed by the als beam . to locate the transition the als center frequency
will be scanned across the interesting range of ( 7.8 @xmath9 0.5 ev ) in steps of 0.05 ev with 200 s of illumination followed by 100 s of light collection at each point . even in the lowest fluorescence case ( @xmath30 @xmath21hz )
, we expect a minimum total fluorescence rate of @xmath1815 khz when the als is on resonance .
once the resonance is found , longer illumination will lead to sufficient fluorescence rates for a vuv spectrometer measurement .
we expect that the transition frequency can be determined to @xmath1850 @xmath15 ( 0.1 nm ) in these experiments , but this of course relies on sufficiently low background fluorescence rates . as a result
, we have performed test runs on our control crystals using the vuv light beam at the als .
the setup used is shown in fig .
[ als_setup ] .
each control crystal ( 3 mm @xmath22 3 mm @xmath22 10 mm ) was illuminated along its long axis with the als beam and any resultant fluorescence from the crystal was collected in a direction orthogonal to the beam via a 2-lens system .
both lenses were fabricated from lif and held in vacuum .
the fluorescence was imaged onto the photocathode of a photomultiplier tube ( pmt ) ( hamamatsu r7639 ) .
the pmt was either under a constant argon purge or held in vacuum to mitigate vuv absorption by air . as shown in fig .
[ als_setup ] , two pneumatic shutters were mounted inside the vacuum chamber .
one shutter was used to prevent vuv scattered light from hitting the pmt during crystal illumination , and the second shutter was used to block the als beam before photon collection with the pmt began .
this introduced a small delay between the point when illumination ended and photon collection began , but in all cases the delay was less than 28 ms .
as the shortest prediction for the @xmath0th isomer lifetime is @xmath18100 s this is sufficiently fast to investigate any spurious long - lived background fluorescence signals in the host crystal that might interfere with measuring fluorescence from the isomeric transition . in what follows ,
we present the data for all crystals that we have measured to date . the data for th : nayf is shown in fig .
[ thnayf_excitation_spectrum ] .
the excitation spectrum was obtained by illuminating the crystal for 5 s with the als beam at a fixed energy and then collecting photons from the crystal over the next 10.7 s using the full bandwidth of the pmt ( 120 nm - 220 nm ) . due to limitations in the operation of the beam and pmt shutters , there was a delay of 28 ms between the end of the illumination period and photon collection period . at each excitation energy , the process was repeated 5 times .
the data points and their error bars in the figure are the average and standard deviation of the results .
the data shown takes into account the photon detection efficiency and the y - axis is therefore the total average fluorescence rate from the crystal over the 10.7 s of photon collection .
the background fluorescence rate , measured by leaving the vuv light beam blocked blocked during the illumination stage , was consistent with pmt dark counts ( @xmath181 hz ) and has been subtracted in the plot .
there is clearly a very large fluorescence peak at a beam energy @xmath317.4 ev , however , because the als was in high energy mode ( electron storage ring at 1.9 gev instead of 1.5 gev ) for this run , we were unable to tune below 7.4 ev .
nonetheless , we measured the lifetime of fluorescence arising from the 7.4 ev excitation light to be @xmath18320 ms .
we also measured the saturation of this fluorescence signal as a function of illumination time . due to the finite closing speed of the pneumatic shutter used to block the als beam , the data of fig .
[ thnayf_signal_vs_illumination ] can only be used to place an upper limit of @xmath18250 ms on the saturation rate of the fluorescence .
this indicates that longer illumination times will not yield higher vuv - induced background rates .
finally , we note that upon inspection of the crystal after illumination , the crystal showed some radiation damage due to the vuv exciation beam .
total fluorescence rate of th : nayf versus illumination time using an excitation energy of 7.4 ev . due to
the finite closing speed of the pneumatic shutter used to block the als beam , the lifetime of the saturation can only be taken as an upper limit . ]
total fluorescence rate of th : nayf versus illumination time using an excitation energy of 7.4 ev . due to
the finite closing speed of the pneumatic shutter used to block the als beam , the lifetime of the saturation can only be taken as an upper limit . ]
data for this crystal was taken in exactly the same way as the th : nayf crystal , and the excitation spectrum is shown in fig .
[ thylf_excitation_spectrum ] . here
again the data points are the average of 5 repetitions at each energy , and the error bars are given by the standard deviation of those points from the average .
the fluorescence lifetimes at the peaks of 8 ev , 8.9 ev , and 10 ev were measured to be @xmath1865 ms , @xmath1870 ms and @xmath1870 ms , respectively ( table [ ylflifetimes ] ) .
@*7l peak & lifetime + 8.0 ev & @xmath1865 ms + 8.9 ev & @xmath1870 ms + 10.0 ev & @xmath1870 ms + excitation spectrum for th : ylf . ] the excitation spectrum for th : licaf is shown in fig . [ licafexcitationspectrum ] and was obtained in a similar way to the above crystals though the experimental details are slightly different . here
the crystal was illuminated for 5 s , but photons were only collected for 5.4 s. due to improved timing controls , the delay time between illumination and photon collection was also reduced to 5 ms .
each data point is the average of 3 repetitions for each beam energy , and its error bar is the standard deviation of the 3 points from the average .
again , the als was in high energy mode for this data run , which prevented us from obtaining data below 7.4 ev . nevertheless , we measured the fluorescence lifetime at 7.4 ev and that data is shown in fig .
[ licafbackgrounddecay ] . from the fit to the data it is found that the lifetime is 12@xmath91 ms .
decay of the fluorescence signal from th : licaf after illuminating it at 7.4 ev .
the red line is an exponential decay fit to the data and gives 12@xmath91 ms for the fluorescence lifetime . ]
decay of the fluorescence signal from th : licaf after illuminating it at 7.4 ev .
the red line is an exponential decay fit to the data and gives 12@xmath91 ms for the fluorescence lifetime . ]
this crystal is an interesting option for studies of the isomeric transition because thorium atoms are a primary constituent and thus questions about doping efficiency are removed .
furthermore , the shifts and broadenings of the transition in the crystal due to imperfections of the doping process may be reduced .
data for this crystal were taken in the same manner as the previous crystal and the results are shown in fig .
[ na2thf6excitationspectrum ] .
the fluorescence peaks observed at an excitation envery of 7.4 ev and 9.6 ev , have both a short and a long timescale component as summarized in table [ lifetimestable ] .
the signal versus illumination time for each peak was also measured and is shown in fig .
[ na2thf6_signal_vs_illumination_time ] .
although the fluorescence rates are not large enough to pose a concern for using this as a host crystal in these studies , this crystal also exhibited some radiation damage due to the vuv excitation light .
@*7l peak & @xmath32 & @xmath33 + 7.4 ev & @xmath1835 ms & @xmath180.5 s + 9.6 ev & @xmath1835 ms & @xmath181 s + the signal versus illumination time for both fluorescence peaks observed in the na@xmath4thf@xmath5 crystal . ] the signal versus illumination time for both fluorescence peaks observed in the na@xmath4thf@xmath5 crystal . ] the data obtained for this crystal was identical to the runs performed on the th : licaf and na@xmath4thf@xmath5 crystals in the previous sections . the excitation spectrum is shown in fig . [ lisaf_excitation_spectrum ] . near the expected isomeric transition energy ,
this crystal exhibits very little fluorescence , and we therefore have plans to explore it further by growing a @xmath6th doped version .
the largest peak observed occurs at 10.5 ev and the lifetime of the fluorescence was measured to be @xmath18500 ms .
excitation spectrum of an undoped lisaf crystal .
] based on the synchrotron data we have obtained so far , we have decided to focus on licaf as a host crystal for future studies using @xmath0th . in addition to its low vuv - induced background fluorescence , licaf has several other important properties .
it is transparent down to 110 nm and has all electron spins paired .
additionally , @xmath0th@xmath12 ions are expected to substitute at the ca@xmath34 site in licaf along with two f@xmath35 interstitials for charge compensation @xcite .
the octahedral symmetry of the site minimizes electric field gradient effects .
results of rutherford backscattering measurements of the th : licaf crystal for an unpolished portion of material cut out of the bulk . the red boxed region indicates the amount of material that is removed in polishing . ]
we have also measured a density in excess of 10@xmath36 @xmath6th/@xmath37 in our th : licaf control crystal using the technique of rutherford backscattering ( rbs ) to interrogate the first @xmath3820 @xmath21 m of the crystal .
the results of these measurements are shown in fig .
[ rbs_results ] .
the rbs spectra shown are for an unpolished facet that we assume was cut from the bulk ( textiti.e . the facet was not the top of the boule ) .
this facet has a thicker region ( modeled as two layers ) of higher concentration of th doping .
the first layer has a th concentration of 1.8% that is 250 nm thick .
the layer below that has a th concentration of 0.3% that is 650 nm thick .
the bulk of the crystal has a concentration of @xmath180.07% th which is similiar to the polished facet ( 0.05% ) .
the th doping of the polished facet showed a similar inhomogeneity of th doping .
we have also performed measurements of the fluorescence backgrounds which might arise from the @xmath39-decay of @xmath0th ( 4.8mev , @xmath40 = 7880yr ) by bombarding this crystal with a 30 na , 4.8 mev beam of @xmath39-particles at the ion beam materials laboratory at los alamos national laboratory .
a total fluorescence rate corresponding to 0.3 photons per @xmath39-particle was observed over the unfiltered pmt bandwidth ( hamamatsu r7639 ) .
thus , a crystal containing 10@xmath36 @xmath0th nulcei yields a background fluorescence rate of 2 mhz which , given the fluorescence rates observed at the als , is expected to be the dominant background . finally , we were able to perform a separate run at the als when the storage ring was operated in the low energy 1.5 gev mode .
as a result , we were able to measure the crystal using the exact experimental protocol we plan to use for measuring the isomeric transition .
the excitation spectrum , which was extended down to 6.0 ev , is shown in fig . [ thlicaf_low_energy_excitation_spectrum ] . here the illumination time was 200 s , the photon collection time was 100 s , and the delay between the two was 5 ms .
only one data point was taken for each energy .
sensitivity to fluorescence from the @xmath41th isomeric transition .
the shaded box indicates the region of interest for the search given the current measurement uncertainty and the expected range for the isomer lifetime . ]
sensitivity to fluorescence from the @xmath41th isomeric transition .
the shaded box indicates the region of interest for the search given the current measurement uncertainty and the expected range for the isomer lifetime . ] with this data , the fluorescence rate measured in our @xmath39-particle bombardment measurements , and the expected fluorescence rate from the isomeric transition at the als , we have calculated the ability of our setup to measure the isomeric transition energy depending on the natural lifetime of the transition .
the results are plotted in fig .
[ sensitivity ] .
this graph shows that if the predicted range of 10 - 0.1 mhz is correct , we should be able to measure the transition with a snr @xmath14 30:1 during a single 8 hour shift at the als .
to summarize , we have characterized vuv - induced fluorescence backgrounds in several candidate host crystals that could be used in a direct search for the @xmath0th isomeric transition .
while none of the crystals displayed problematic backgrounds , these measurements have shown licaf to have the lowest fluorescence rate and shortest fluorescence lifetime overall , as well as a high resilence to damage by the vuv beam .
subsequent measurements have also shown it to have a low @xmath39 particle induced background in the vuv that might arise from the alpha decay of @xmath0th inside a licaf host crystal . additionally , a density in excess of 10@xmath36 @xmath6th nuclei/@xmath37 was measured in our @xmath6th doped licaf crystal using rbs .
thus , licaf appears to be a viable host crystal for @xmath0th and a direct search for the isomeric transition .
11 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 | we have recently described a novel method for the construction of a solid - state optical frequency reference based on doping @xmath0th into high energy band - gap crystals @xcite . since nuclear transitions are far less sensitive to environmental conditions than atomic transitions , we have argued that the @xmath0th optical nuclear transition may be driven inside a host crystal resulting in an optical frequency reference with a short - term stability of @xmath1 at 1 s and a systematic - limited repeatability of @xmath2 .
improvement by @xmath3 of the constraints on the variability of several important fundamental constants also appears possible .
here we present the results of the first phase of these experiments .
specifically , we have evaluated several high energy band - gap crystals ( th : nayf , th : ylf , th : licaf , na@xmath4thf@xmath5 , lisaf ) for their suitability as a crystal host by a combination of electron beam microprobe measurements , rutherford backscattering , and synchrotron excitation / fluorescence measurements .
these measurements have shown licaf to be the most promising host crystal , and using a @xmath6th doped licaf crystal , we have performed a mock run of the actual experiment that will be used to search for the isomeric transition in @xmath0th .
this data indicates that a measurement of the transition energy with a signal to noise ratio ( snr ) greater than 30:1 can be achieved at the lowest expected fluorescence rate . |
apart from providing independent supporting evidence of the big bang , the detection and subsequent study of cosmological chemical evolution provides the empirical details of galaxy formation and evolution .
how primordial and processed gas is consumed by star formation , the dominant feedback processes and merging scenarios , may all contribute to the overall evolution of chemical abundances .
one high - precision probe of this evolution is the spectroscopic study of damped lyman-@xmath0 systems ( dlas ) : absorbers with neutral hydrogen column densities @xmath2 .
although these observations demonstrate that dlas arise along lines of sight through distant galaxies , they do not directly disclose details such as the galaxy s morphology , luminosity , mass or age .
there exists substantial evidence that dlas arise in a variety of galaxy types . at low-@xmath3 ( @xmath4 ) ,
kinematic and hi 21-cm absorption studies ( e.g. @xcite ; @xcite ) suggest a significant contribution from spiral galaxies , a view supported at high-@xmath3 by kinematic modelling and abundance studies @xcite .
however , direct imaging at low-@xmath3 ( e.g. * ? ? ?
* ) reveals that dla hosts are a mix of irregulars , spirals and low surface - brightness galaxies ( lsbs ) .
a recent @xmath5 hi 21-cm emission study @xcite supports this .
@xcite argue that the number of dlas per redshift interval and the @xmath6 distribution imply that dlas at @xmath7 are a mix of spirals and lsbs whereas , at higher @xmath3 , they are more likely to be dwarfs .
this is supported by fitting of chemical evolution models to dla metal abundances ( e.g. * ? ? ?
* ) and by recent 21-cm absorption searches at high redshift @xcite .
though further work is clearly needed , the direct and indirect evidence for a ` mixed bag ' of dla hosts is already compelling .
evidence for an increase in dla metallicities , [ m / h ] , with cosmic time has emerged only gradually @xcite , the latter reference using over 100 dlas to provide the strongest statistical evidence so far .
such large samples are required due to the huge scatter ( @xmath8 ) in [ m / h ] at a given epoch ( see fig .
[ fig:1 ] ) , a diversity expected given the variety of dla hosts discussed above .
however , the diversity in [ m / h ] could also significantly _ bias _ any estimate of chemical evolution , as could several observational selection effects ( e.g. * ? ? ? * ) . in @xcite
we suggest that by selecting those dlas in which h@xmath1 absorption is detected , one may reduce or possibly avoid some of the biases besetting the general dla population .
there are currently 9 confirmed h@xmath1-bearing dlas ( see below ) and , typically , h@xmath1 is detected in only a few velocity components .
these h@xmath1-bearing components seem distinct from the others , showing lower temperatures and higher dust depletion factors [ m / fe ] @xcite .
therefore , h@xmath1-bearing dlas might be a less biased tracer of chemical evolution than the general dla population since they may allow one to focus on a narrower range of physical conditions throughout cosmic time .
, dust - depletion [ m / fe ] , and molecular fraction ( @xmath9 ) evolution for confirmed ( filled circles ) and tentative ( hollow circle ) h@xmath1 detections compared with values from the @xcite dlas ( small diamonds ) .
the values adopted are discussed in the text .
linear fits to the h@xmath1-bearing dlas ( long dashes ) and general dlas ( short dashes ) were obtained by quadrature addition of a constant to the individual error bars such that @xmath10 . due to its much lower @xmath11 and potentially high photo - dissociation rate @xcite ,
we do not include 0515@xmath124414 in these fits ( but see table [ tab:1]).,scaledwidth=70.0% ] h@xmath1 is detected in dlas via the lyman and werner - band uv absorption lines which generally lie in the hi lyman-@xmath0 forest .
a compilation of results from h@xmath1 searches in dlas is given in table 8 of ( * ? ? ?
* , hereafter l03 ) , for which 7 dlas have confirmed h@xmath1 detections and metallicity measurements : 0013@xmath12004 ( @xmath13 ) , 0347@xmath123819 , 0405@xmath12443 ( @xmath14 ) , 0528@xmath122505 , 0551@xmath12366 , 1232@xmath15082 and 1444@xmath15014 .
the dla towards 0013@xmath12004 @xcite comprises several absorbing components , of which 2 are dominant ( their components @xmath16 and @xmath17 ) : @xmath6 is measured at the mean redshift and we use the mean @xmath18h@xmath19 with error given by the range of @xmath18h@xmath19 in l03 . for the dla towards
1232@xmath15082 we use the @xmath18h@xmath19 value and error from @xcite .
we include two further dlas : ( i ) the recent detection towards 0515@xmath124414 @xcite and ( ii ) 0000@xmath122620 , regarded as only a tentative detection by l03 .
however , the h@xmath1 identification has been carefully scrutinised by @xcite and relies on two h@xmath1 absorption features , the l(4 - 0)r1 and w(2 - 0)q(1 ) lines , the former appearing relatively clean from lyman-@xmath0 forest blending .
a tentative detection of h@xmath1 in 0841@xmath151256 has also been reported @xcite , though confirmation requires future data and analyses .
lcccccc & & & + sample & @xmath20 & @xmath21 & @xmath20 & @xmath21 & @xmath22 & @xmath23 + + exc .
0515 & 0.08 & @xmath24 & 0.14 & @xmath25 & 2.2 & 0.28 + inc .
0515 & 0.04 & @xmath26 & 0.03 & @xmath27 & 1.9 & 0.33 + + exc .
0515 & 0.05 & @xmath28 & 0.24 & @xmath29 & 1.7 & 0.47 + inc .
0515 & 0.06 & @xmath30 & 0.25 & @xmath31 & 1.0 & 0.95 + + exc .
0515 & 0.04 & @xmath32 & & & & + inc .
0515 & 0.06 & @xmath33 & & & & + [ tab:1 ] in fig .
[ fig:1 ] we plot against @xmath11 the metallicity [ m / h ] , dust - depletion factor [ m / fe ] and molecular fraction @xmath34 $ ] for the 9 h@xmath1-bearing dlas , comparing the former two quantities with dlas in @xcite with zn , si , s or o metallicity over the relevant redshift range .
[ m / h ] for the h@xmath1-bearing dlas is [ zn / h ] , except for 0347@xmath123819 and 1232@xmath15082 where it is [ s / h ] and [ si / h ] . for 0515@xmath124414 , [ zn / fe ]
is from @xcite . the main results from fig .
[ fig:1 ] are summarized by the statistics in table [ tab:1 ] : ( i ) [ m / h ] , [ m / fe ] and @xmath9 for the h@xmath1-bearing dlas are anti - correlated with @xmath11 at the 95% confidence level ( i.e. more significant than for the general dla population ) , ( ii ) [ m / h ] shows a steeper evolution with @xmath11 and a smaller scatter about the slope than the general dla population , ( iii ) [ m / fe ] in h@xmath1-bearing dlas shows strong evolution with @xmath11 while the general dlas show no evidence for evolution , and ( iv ) @xmath9 ranges over @xmath35 and shows a very steep evolution with @xmath11 .
the new results ( i)(iii ) support our hypothesis that h@xmath1-bearing dlas form a chemically distinct sub - class and may trace chemical evolution more reliably .
the @xmath9@xmath11 correlation was studied by l03 and is discussed further below .
the [ m / h]s and @xmath9s in fig .
[ fig:1 ] are measured using the total @xmath6 across the dla profile and are not specific to the h@xmath1-bearing components .
[ m / fe ] is generally found to be uniform across most dla profiles @xcite , indicating that [ m / h ] should be uniform .
however , the h@xmath1-bearing components typically have much higher [ m / fe ] values than other components in the same dla ( e.g. l03 and 0347@xmath12383 s [ si / fe ] profile in @xcite ) .
these components usually dominate the non - refractory metal - line profiles and so , although [ m / h ] and @xmath9 will be systematically underestimated , the effect will not be large .
the fitted slopes in fig .
[ fig:1 ] are likely to be reasonably robust against this effect , but a larger sample and more detailed study is clearly required . what observational selection effects and biases could contribute to the steep @xmath9@xmath11 evolution ?
firstly , the sample is inhomogeneous since the spectra do not all have similar s / n and since the h@xmath1 detection methods and criteria were not uniform .
indeed , the weak h@xmath1 lines detected towards 0000@xmath122620 are at the typical non - detection level ( see fig . 16 in l03 ) .
the @xmath9@xmath11 correlation is therefore only tentative .
secondly , the h@xmath1 detection limit will alter with @xmath3 : equivalent widths increase but lyman-@xmath0 forest blending worsens with increasing @xmath3 .
though this is an unlikely culprit for the @xmath35 evolution observed , precise quantification of these competing effects requires numerical simulations .
dlas containing large amounts of dust could suppress detection of their background quasars and may therefore be ` missing ' from our sample @xcite .
however , since [ m / h ] , [ m / fe ] and @xmath9 are positively correlated with each other ( e.g. l03 ) , this effect is likely to suppress , rather than create , the correlations in fig .
[ fig:1 ] .
a recent survey for dlas towards radio - selected quasars @xcite also indicates that the number of such ` missing ' quasars is likely to be small .
selecting those dlas which exhibit h@xmath1 absorption may focus on systems with a narrower range of physical conditions than the dla population as a whole .
tentative support for this conjecture lies in the steeper , tighter [ m / h]@xmath11 anti - correlation observed for the h@xmath1 systems studied here .
@xcite recently presented detailed chemical evolution models which give a slope for the [ m / h]@xmath11 relation of @xmath36 .
they correct this result for various observational biases to match the shallower slope observed for the general dla population .
however , the steep [ m / h]@xmath11 slope observed for h@xmath1-bearing systems could be less affected by these biases and may avoid those introduced by sampling many different ism gas phases .
it might therefore be more comparable to the uncorrected slopes in the models .
h@xmath1-selected dlas are therefore a candidate for a less biased probe of chemical evolution .
that there exists such a large range ( @xmath37 ) in the values of @xmath9 in fig .
[ fig:1 ] may not be surprising : @xcite describes a @xmath38}=-1 $ ] photo - ionization model for clouds in local hydrostatic equilibrium . for a representative incident uv background flux and dust - to - metals ratio , the molecular fraction in this model shows a sudden increase of @xmath39 for only a small increase in the total hydrogen density .
therefore , the very steep @xmath9@xmath11 correlation could be achieved with a modest increase in dust content at lower redshifts , consistent with the observed [ m / fe]@xmath11 anti - correlation .
l03 also discuss how @xmath9 might be very sensitive to local physical conditions .
for example , within the schaye model , one expects an anti - correlation between @xmath9 and the intensity of the uv background .
however , the behaviour of the uv background flux with redshift over the range @xmath403 is still a matter of considerable uncertainty .
the strong decrease in @xmath9 at high redshift may also be consistent with recent hi 21-cm absorption measurements in dlas @xcite , where a generally higher spin / excitation temperature is found at @xmath41 .
with an increased sample size and more detailed analyses , the @xmath9@xmath11 anti - correlation , if real , may provide complementary constraints on these problems . , a. c. , mathlin , g. p. , churches , d. k. , & edmunds , m. g. 2000 , in esa sp , vol .
445 , star formation from the small to the large scale , ed .
f. favata , a. kaas , & a. wilson ( noordwijk , the netherlands : european space agency ) , 21 , s. a. , molaro , p. , centuri ' on , m. , dodorico , s. , bonifacio , p. , & vladilo , g. 2001 , in deep fields , ed .
s. cristiani , a. renzini , & r. williams , eso astrophysics symposia series ( berlin , germany : springer ) , 334 | the chemical abundances in damped lyman-@xmath0 systems ( dlas ) show more than 2 orders of magnitude variation at a given epoch , possibly because dlas arise in a wide variety of host galaxies .
this could significantly bias estimates of chemical evolution .
we explore the possibility that dlas in which h@xmath1 absorption is detected may trace cosmological chemical evolution more reliably since they may comprise a narrower set of physical conditions .
the 9 known h@xmath1 absorption systems support this hypothesis : metallicity exhibits a faster , more well - defined evolution with redshift than in the general dla population . the dust - depletion factor and ,
particularly , h@xmath1 molecular fraction also show rapid increases with decreasing redshift .
we comment on possible observational selection effects which may bias this evolution .
larger samples of h@xmath1-bearing dlas are clearly required and may constrain evolution of the uv background and dla galaxy host type with redshift .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
in classical data transmission , repeaters are used to amplify data signals ( bits ) when they become weaker during their propagation through the transmission channel .
in contrast to classical information , the above mechanism is impossible to realize when the transmitted data signals are the carriers of quantum information ( qubits ) , which can not be amplified or cloned without destroying the encoded information @xcite , @xcite .
therefore , the carrier has to propagate along the entire length of the transmission channel which , due to various losses , leads to an exponentially decreasing probability to receive it intact at the end of the channel . to circumvent this problem ,
the quantum repeater was proposed in the seminal ref .
@xcite that encapsulates three building blocks ( i ) entanglement distribution , ( ii ) entanglement purification , and ( iii ) entanglement swapping , which have to be applied sequentially . using the first building block , a large set of low - fidelity entangled qubit pairs
is generated between all repeater nodes , which becomes distilled by the second block into a smaller set of high - fidelity entangled pairs .
entanglement swapping , finally , combines two purified entangled pairs distributed between the neighboring repeater nodes into one entangled pair leading to a gradually increasing distance of shared entanglement .
obviously , quantum repeater schemes are not straightforward .
the above mentioned building blocks , for instance , require a feasible and reliable quantum logic , while low - fidelity entangled pairs have to be distributed over reasonably long distances . up to now , only a few schemes , which distribute entanglement over the distance of 200 km using fiber - optic @xcite and 100 km using free - space channel @xcite between the nodes have been experimentally demonstrated .
nevertheless , motivated by a rapidly growing experimental and theoretical progress in the field of quantum communication , various quantum repeater schemes and single building blocks have been proposed @xcite , @xcite , @xcite , @xcite , @xcite , @xcite . in our previous paper @xcite
, we already proposed a dynamical quantum repeater scheme in which the entanglement between the two neighboring repeater nodes was distributed using controlled displacements of input coherent light , while the generated low - fidelity entangled pairs were purified using ancillary ( four - partite ) entangled states .
this scheme exploited solely the evolution of short chains of atoms coupled to optical cavities located in each repeater node , such that any explicit usage of quantum logical gates has been avoided . in the present paper
, we propose an improved quantum repeater scheme that ( as well ) avoids two - qubit quantum logical gates and where the entanglement is distributed using a sequence of controlled phase shifts and displacements of input coherent light , while entanglement purification is free of ancillary entangled resources .
this repeater scheme exhibits reasonable fidelities and repeater rates providing an attractive platform for long - distance quantum communication .
in particular , our entanglement distribution becomes very efficient , simple , and practical , as it only requires discriminating two optical coherent states with opposite signs , where each allows for conditionally preparing a certain two - qubit bell state over the distance in two neighboring repeater nodes .
this way , in the ideal loss - free case , maximum two - qubit entanglement can be deterministically generated with unit fidelity in the limit of large coherent - state amplitudes .
note that in the present work we focus on a standard ( original ) quantum repeater scheme as described above and based on the use of quantum memories and two - way classical communication .
other more recent approaches that make use of quantum error correction codes reduce or completely eliminate the need for storing quantum information and for employing two - way communication . as a result ,
extremely high rates are achievable in this new generation of repeaters ( for a classification into three generations of quantum repeaters , see ref .
@xcite ) .
( @xmath0 ) . in the upper region ( framed by a dashed rectangle ) , the atomic pairs become ( one - by - one ) entangled providing the entanglement resources for the next framed region , where the entanglement purification takes place .
a successful purification round leads to an increase of fidelity associated with a stationary atomic pair , while the other pairs get projected once they leave the cavities @xmath1 ( @xmath2 ) . in the final ( lower ) framed region ,
the atoms ( associated with the purified atomic pair ) enter the cavities @xmath3 ( @xmath0 ) , where they get swapped with atoms ( associated with other two purified atomic pairs ) leading to a triple distance enlargement of shared entanglement .
( b ) the coherent state discrimination ( csd ) setup . the coherent states @xmath4 encapsulated in eq .
( [ state0 ] ) are guided into the upper input port of the beam - splitter , while the ancilla coherent state @xmath5 into the lower input port .
( c ) structure of a three - level atom in the @xmath6-configuration subjected to the cavity and laser fields .
see text for description.,title="fig:",scaledwidth=95.0% ] + the paper is organized as follows . in the next section ,
we describe in detail our practical quantum repeater scheme .
we introduce and discuss the entanglement distribution , purification , and swapping protocols in secs .
ii.a , ii.b , and ii.c , respectively . in sec .
ii.d , we discuss a few relevant issues related to the implementation of our repeater scheme , while a brief analysis of final fidelities and repeater rates is given in sec .
iii along with the summary and outlook .
the main physical resources of our repeater scheme are ( i ) three - level atoms , ( ii ) high - finesse optical cavities , ( iii ) continuous and pulsed laser beams , ( iv ) balanced beam splitters , and ( v ) photon detectors . in fig .
[ fig1](a ) , we display schematic view of our experimental setup that encapsulates two neighboring repeater nodes ( b and c ) and includes the entanglement distribution , purification , and swapping protocols in a single setup . in this setup , each repeater node encapsulates single - mode cavities @xmath7 , @xmath1 , and @xmath3 ( @xmath8 , @xmath2 , and @xmath0 ) , single atoms conveyed along the setup with the help of two vertical optical lattices , two sources of weak coherent - state pulses @xmath9 and @xmath10 ( @xmath11 and @xmath12 ) , detectors @xmath13 and @xmath14 ( @xmath15 and @xmath16 ) connected to the neighboring node via a classical communication channel , and a magneto - optical trap ( mot ) that provides atomic chains to be conveyed . the alignment of vertical lattices is such that the conveyed atoms cross cavities at their anti - nodes ensuring , therefore , a strong atom - cavity coupling regime . in addition , node b includes a source of weak coherent - state pulses @xmath17 , while node c includes a coherent state discrimination ( csd ) setup displayed in fig .
[ fig1](b ) .
finally , each repeater node encapsulates two optical lattices denoted as s and v ( conveyors ) , where the atoms inserted into the v - conveyor are transported with a constant velocity through all three cavities , while the atoms in the s - conveyor can be slowed down or accelerated by demand . for convenience ,
our setup is divided in three ( framed by dashed rectangles ) parts corresponding to three building blocks of a quantum repeater as mentioned in the introduction .
below , we clarify each part of our setup and explain all the manipulations and elements . ) , we display the evolution of the input coherent state @xmath18 due to the sequence of three operations , namely , ( i ) controlled phase shift @xmath19 in node b , ( ii ) controlled displacement @xmath20 in node c , and ( iii ) controlled phase shift @xmath19 in node c. in this figure , for simplicity , we set @xmath21 ( lossless case).,title="fig:",scaledwidth=95.0% ] + in this part of setup , atoms @xmath22 and @xmath23 are extracted from the mots of each repeater node and inserted into the s - conveyors .
both atoms are initialized in the ground state @xmath24 and arrive simultaneously in the cavities @xmath7 and @xmath8 .
each atom encodes a qubit by means of a three - level atom in the @xmath6 configuration as displayed in fig .
[ fig1](c ) . in order to protect this qubit against the decoherence , the ( qubit )
states @xmath24 and @xmath25 are chosen as the stable ground and long - living metastable atomic states or as the two hyperfine levels of the ground state . once conveyed into the cavity @xmath7 ( @xmath8 ) , atom @xmath22 ( @xmath23 ) couples simultaneously to the photon field of cavity and two continuous laser beams as displayed in fig .
[ fig1](c ) .
the evolution of the coupled atom - cavity - laser system in the node b is governed by the hamiltonian @xmath26 where @xmath27 is the respective pauli operator in the computational basis @xmath28 , while @xmath29 is the effective coupling ( see appendix , where we show that the above hamiltonian is produced deterministically in our setup ) .
the evolution governed by @xmath30 reads @xmath31 where @xmath32 .
this evolution implies a phase shift of the cavity field by the angle @xmath33 or @xmath34 conditioned upon the atomic state , where @xmath33 is proportional to the atom - cavity - laser evolution time that , in turn , is inversely proportional to the velocity of conveyed atom .
in contrast to node b , the evolution of the coupled atom - cavity - laser system in the node c is governed by a sequence of two evolutions . while the second evolution is governed by the hamiltonian ( [ ham - rot ] ) , the first evolution is governed by the hamiltonian @xmath35 where @xmath36 is the atom - field coupling ( see appendix , where we show that the above hamiltonian is produced deterministically in our setup ) .
the evolution governed by @xmath37 reads @xmath38 where @xmath39 .
this evolution implies a displacement of the cavity field mode by the amount @xmath40 conditioned upon the atomic state , where @xmath40 is proportional to the atom - cavity - laser evolution time that , in turn , is inversely proportional to the velocity of conveyed atom .
we remark that both hamiltonians imply that the ( fast - decaying ) excited state @xmath41 remains almost unpopulated during the respective evolutions .
first , a pulse of a weak coherent light produced by the source @xmath17 and characterized by a real amplitude @xmath42 interacts with the atom - cavity system in node b , where the cavity is prepared in the vacuum state .
the evolution ( [ evol - rot ] ) leads to the atom - light entangled state @xmath43 where @xmath44 and @xmath45 .
the resulting coherent state from the cavity is outputted into the transmission channel between the nodes .
since we are dealing with a high - finesse cavity and since the fast - decaying atomic state @xmath41 remains almost unpopulated during the evolution , the dominant photon loss occurs in the optical fiber connecting the cavities @xmath7 and @xmath1 that plays the role of a transmission channel in our setup .
since photon loss increases with the length of the fiber , to a good approximation , we describe the loss using a beam - splitter model that transmits only a part of the coherent light pulse through the channel , @xmath46 where the subscript @xmath47 refers to an environmental light mode responsible for the fiber relaxation . in this expression
, @xmath48 describes the attenuation of the transmitted light through the fiber , where @xmath49 is the distance between the repeater nodes , while @xmath50 is the attenuation length that can reach almost @xmath51 km for fused - silica fibers at telecommunication wavelengths .
below we set @xmath52 km .
next , the attenuated light pulse interacts with the atom - cavity system in node c , where ( as in node b ) the cavity is prepared in the vacuum state . by tracing over the environmental degrees of freedom ( mode with the subscript @xmath47 ) , the evolution @xmath53 in node b along with the sequence of evolutions @xmath54 and @xmath53 in node c followed by an unconditional displacement @xmath55 leads to the entangled state between the atoms and optical modes ( see fig .
[ fig5 ] ) @xmath56 where @xmath57 such that @xmath58 , while @xmath59 .
the ( entangled with both atoms ) light pulse is guided into the balanced beam splitter displayed in fig .
( [ fig1])(b ) and that is characterized by the transmissivity and reflectivity parameters @xmath60 and @xmath61 , respectively . according to this coherent state discrimination ( csd ) setup , the input pulse interferes with the ( ancilla ) coherent state @xmath62 sent into another input port of the beam splitter .
it can be readily checked that the coherent states @xmath63 encapsulated in eq .
( [ state0 ] ) and supplemented by the ancilla state are transformed due to the beam splitter as follows @xmath64 on the right hand side of above equations , each ket vector in the expression corresponds to an output port of the beam splitter . having two detectors resolving _ click _ or
_ no click _ detection events , there are three possible detection patterns , since the pattern @xmath65click , click@xmath66 can not happen . among these patterns , the pattern @xmath65no click
, no click@xmath66 is inconclusive , while the remaining two patterns @xmath65click , no click@xmath66 and @xmath65no click , click@xmath66 lead to the desired state discrimination .
we compute first the respective probabilities of success @xmath67 where @xmath68 denotes the state @xmath69 being transformed in concordance with eq .
( [ transf ] ) .
conditioned upon above detection patterns , the density function @xmath68 reduces to the one of two ( atom - atom ) entangled states @xmath70 or @xmath71 where the entanglement fidelity is given by the expression @xmath72 if the output of photon detection yields the inconclusive pattern @xmath65no click , no click@xmath66 , then the atoms @xmath22 and @xmath23 should be discarded and the entanglement distribution protocol repeated using the next atomic pair conveyed by means of s - conveyors along the setup in both repeater nodes . in this section , we developed a feasible scheme for distribution of atomic entanglement that is conditioned upon two detection patterns indicated above , where the total success probability associated with these patterns is @xmath73 , while the fidelity is given by the above expression .
we exploit the controlled phase shift ( [ evol - rot ] ) in both repeater nodes supplemented by a controlled displacement in the second node .
in contrast to ref .
@xcite and our previous paper @xcite , this combined approach greatly simplifies the tripartite entangled state of light and two atoms [ compare eq .
( [ state0 ] ) in this paper to eqs .
( 14 ) and ( 4 ) in the above mentioned references ] . more specifically , eq . ( [ state0 ] )
encapsulates only two optical coherent states , which differ by a minus sign and are , therefore , much easier to discriminate .
this simplification , in turn , implies a rather uncomplicated discrimination setup ( csd ) displayed in fig .
[ fig1](b ) consisting of a single beam - splitter and two photon detectors to resolve click and no - click events .
we emphasize that in the above cited references , the entangled state of light and two atoms involve three different coherent states leading to a more demanding discrimination scheme .
specifically , the discrimination scheme in ref .
@xcite requires two beam splitters and three unconditional displacements together with three photon detectors capable to resolve click and no click events .
although the respective csd scheme in our previous paper @xcite requires a single beam splitter and one unconditional displacement , a notable difficulty poses the requirement to use photon number resolving detectors . as a consequence of simplified csd in this paper
, we obtain a much higher success probability ( compare eq .
( [ prob - succ1 ] ) in this paper to @xmath74 in ref .
@xcite and to @xmath75 and @xmath76 in ref .
@xcite ) leading , in turn , to much higher repeater rates .
we remark that , in contrast to the controlled phase shift used in ref .
@xcite , in this paper , we exploit the evolution ( [ evol - rot ] ) based on @xmath27 instead of @xmath77 and , moreover , we avoid manipulations of atomic states associated with the produced entangled state . finally , we note that eq .
( [ fidelity ] ) coincides with the respective fidelity obtained in our previous paper , in which we have mentioned that this fidelity is close to one when @xmath78 .
since we considered one single purification round , this restriction led us to the regime @xmath79 that we considered through the paper .
due to four purification rounds considered in this paper , we succeed to relax this restriction and consider @xmath80 ( see below ) . assuming that the entanglement distribution was successful , atoms @xmath22 and @xmath23 are conveyed into the area , in which they are subjected to the ( single - qubit ) hadamard transformation . due to this unitary transformation , the states ( [ state1a ] ) and ( [ state1b ] )
take the form @xmath81 once atoms @xmath22 and @xmath23 enter the cavities @xmath1 and @xmath2 , their ( conveyed ) velocities decrease gradually until zero , such that atoms remain trapped right inside the respective cavities .
at the same time , atoms @xmath82 and @xmath83 are inserted from mots into v - conveyors of both repeater nodes and transported with a constant velocity along the setup .
similar to atoms @xmath22 and @xmath23 , this atomic pair is entangled and subjected then to the hadamard gate .
assuming that the second entanglement distribution was successful , atoms @xmath82 and @xmath83 are now described by the state @xmath84 or @xmath85 having the structure of eqs .
( [ state2a ] ) or ( [ state2b ] ) , respectively .
atoms @xmath82 and @xmath83 are conveyed along the setup until they couple ( simultaneously ) to cavities @xmath1 and @xmath2 prepared both in the vacuum state .
at this point , each of these cavities shares a pair of atoms @xmath22 , @xmath82 , and @xmath23 , @xmath83 , respectively . inside the cavities @xmath1 and @xmath2 ,
these atomic pairs evolve due to the interaction governed by the heisenberg xx hamiltonians @xmath86 over the time period @xmath87 , where @xmath88 is the effective coupling .
the resulting density functions @xmath89 describe four - partite entangled states .
we remark that the above heisenberg xx hamiltonian is produced deterministically in our setup by coupling atoms ( simultaneously ) to the same cavity mode and two laser beams in the strong driving regime ( see appendix a of ref .
@xcite ) . , and @xmath90 along with the swapped fidelity @xmath91 as functions of the initial fidelity ( [ fidelity ] ) .
( b ) success probability @xmath92 associated with the entanglement distribution and purification as a function of elementary repeater distance @xmath49 displayed for @xmath93 ( solid curve ) , @xmath94 ( dotted curve ) , and @xmath95 ( dashed curve).,title="fig:",scaledwidth=95.0% ] + being further conveyed along the setup , atoms @xmath82 and @xmath83 decouple from cavities @xmath1 and @xmath2 , and get projected outside in the computational basis .
purification protocol is successful if the outcome of atomic projection reads @xmath96 in this case , above detection patterns lead to the reduced density functions @xmath97 where the purified fidelity and the success probability take the form @xmath98 the above purified states preserves the rank 2 form and are characterized by the fidelity ( [ fidelity1 ] ) displayed in fig .
[ fig2](a ) by a dotted curve .
we remark that if the outcome of atomic projection disagrees with the patterns ( [ pattern2a ] ) and ( [ pattern2b ] ) , the entanglement purification is considered unsuccessful . in this case , we discard all atoms and repeat both distribution and purification protocols using fresh atomic pairs conveyed from mots in each repeater node . assuming that the entanglement purification is successful , atoms @xmath99 and @xmath100 are inserted from mots into v - conveyors and transported with a constant velocity along the setup . similar to atoms @xmath82 and @xmath83 , this atomic pair gets entangled , subjected to the hadamard gate , and conveyed further into the cavities @xmath1 and @xmath2 , such that each of these cavities shares a pair of atoms @xmath22 , @xmath99 , and @xmath23 , @xmath100 , respectively . in these cavities , atoms evolve due to the heisenberg xx interaction over the time period @xmath101 .
being further conveyed along the setup , atoms @xmath99 and @xmath100 are then projected in the computational basis . as before ,
the entanglement purification is successful if the outcome of atomic projection coincides with ( [ pattern2a ] ) and ( [ pattern2b ] ) . in this case , we get the reduced density functions @xmath102 where the purified fidelity @xmath103 and the success probability @xmath104 are computed using the iterative formulas @xmath105 where @xmath106 , while @xmath103 is displayed in fig .
[ fig2](a ) by a dashed curve .
assuming that the third purification ( using atoms @xmath107 , @xmath108 ) is successful , atoms @xmath109 , @xmath110 are conveyed along the setup and coupled then to the cavities @xmath1 , @xmath2 in order to realize the last ( fourth ) purification round .
in contrast to the sequence we described above , atoms @xmath109 , @xmath110 are not projected after they leave the respective cavities .
instead , atoms @xmath22 and @xmath23 are projected in the computational basis using the ( nondestructive ) projective measurements directly inside the cavities ( see sec .
ii.d below ) .
in contrast to the patterns ( [ pattern2a ] ) and ( [ pattern2b ] ) , the entanglement purification is successful if the outcome of atomic projection ( atoms @xmath22 and @xmath23 ) reads @xmath111 with the success probability @xmath112 , above patterns yield the reduced density functions @xmath113 completely characterized by the fidelity @xmath90 displayed in fig .
[ fig2](a ) by a dot - dashed curve .
it is readily seen that we obtain an almost - unit purified fidelity for @xmath114 using four ( successful ) purification iterations . in this section
, we developed a high - fidelity scheme for entanglement purification that leads ( iteratively ) to a gradual growth of the purified fidelity , and where the total probability of success @xmath115 is associated with four purification rounds which , in turn , encapsulates success probabilities of five entanglement distributions . by inserting eq .
( [ fidelity ] ) in the above expression , we display @xmath92 in fig .
[ fig2](b ) as a function of @xmath49 ( encoded by @xmath116 ) for @xmath93 , @xmath23 , and @xmath82 .
as expected , the total success probability decreases with the growth of the distance between the nodes .
we emphasize that the sequence of steps utilized in this section can be related to the entanglement purification scheme c ( so - called entanglement pumping ) suggested in ref .
@xcite , in which the fidelity of a single low - fidelity entangled pair becomes gradually increased at the cost of all other low - fidelity entangled pairs available .
indeed , in our approach we also produced first the entangled pair ( [ state2a ] ) or ( [ state2b ] ) described by the fidelity @xmath117 that became gradually increased up to @xmath90 at the cost of entangled pairs associated with the atoms @xmath118 and each described by the fidelity @xmath117 as well . in each node of proposed purification protocol
we exploit one evolution governed by the heisenberg xx hamiltonian ( 17 ) followed by a projection of a single atom in the computation basis .
in contrast to our previous paper @xcite , in which we have utilized ( i ) a four - partite entangled state [ see eq .
( 8) in the above reference ] generated in a probabilistic fashion through the cat state discrimination , ( ii ) evolution governed by the heisenberg xy hamiltonian , and ( iii ) projection of two atoms in each repeater node , the proposed approach greatly simplifies our protocol since we avoid any ancilla state and we postselect only the atomic state detection events associated with two atoms ( and not four as in the previous paper ) . due to these improvements , in this paper , we succeed to increase significantly the success probability @xmath92 associated with the entire purification protocol [ compare eq .
( [ prob - fin ] ) with eq .
( 26 ) in our previous paper ] .
we emphasize that a high purified fidelity obtained in this paper is the consequence of multiple purification rounds realized as an inherent part of our repeater scheme in this paper .
although we considered just four such rounds , our setup is capable to perform an arbitrary number of rounds at the cost of reduced success probability .
the setup displayed in fig .
1(a ) of our previous paper , in contrast , is by design limited to a single purification round , such that an extension to four such rounds would require three additional cavities and six additional atoms in each repeater node along with three additional communication channels . using this extended setup in the framework of approach proposed in our previous paper
, we have checked that the purified fidelity obtained using four successive purification rounds is slightly smaller as the fidelity @xmath90 obtained in this paper . however , since one successful purification round is conditioned upon the generation of the ancilla state ( 8) and atomic postselection ( 15 ) , we obtained a much lower probability of success if compared with @xmath92 derived above .
this result leads , in turn , to a dramatically low success probability associated with the entire scheme and almost vanishing repeater rates . assuming that the entanglement purification was successful , the ( high - fidelity ) entangled atoms @xmath109 and @xmath110 are conveyed along the setup to the swapping region displayed in the bottom rectangle of fig .
[ fig1](a ) . here , atoms couple the cavities @xmath3 and @xmath0 , both prepared in vacuum state .
the conveyed atoms together with the trapped atoms form two atomic pairs @xmath109 , @xmath119 , and @xmath110 , @xmath120 .
we recall that atoms @xmath119 and @xmath120 are entangled with atoms @xmath121 and @xmath122 , respectively , where each pair is described by the mixed states @xmath123 and @xmath124 having the structure of eqs .
( [ state6a ] ) and ( [ state6b ] ) . according to the entanglement swapping protocol we proposed in ref .
@xcite , each atomic pair evolves in cavities @xmath3 and @xmath0 due to the heisenberg xx interaction utilized in the previous section .
we remark that being applied on the separate states of an atomic pair ( in the computational basis ) over the time period @xmath125 , this evolution implies @xmath126 such that the resulting states form the modified bell basis @xmath127 where @xmath128 .
c | c c c c & 1 & 2 & 3 & 4 + 1 & @xmath129 & @xmath130 & @xmath131 & @xmath132 + 2 & @xmath130 & @xmath129 & @xmath132 & @xmath131 + 3 & @xmath132 & @xmath131 & @xmath130 & @xmath129 + 4 & @xmath131 & @xmath132 & @xmath129 & @xmath130 + this observation suggests a deterministic realization of entanglement swapping using just two consecutive steps ( i ) atomic pairs @xmath109 , @xmath119 , and @xmath110 , @xmath120 are subjected to the evolutions @xmath133 and @xmath134 , respectively , that is followed by ( ii ) projection of each atomic pair in the computational basis
. obviously , these two steps are equivalent with the atomic projection in the modified bell basis ( [ mbasis ] ) , where the swapped state is given by the expression @xmath135 taken here , for simplicity , for a particular case when all three entangled pairs have the structure of eq . ( [ state6b ] ) . in this expression , @xmath136 is the normalization factor , the states @xmath137 and @xmath138 are displayed in table [ tab1 ] , while the swapped fidelity takes the form @xmath139 as expected , the swapped density function ( [ state6 ] ) is diagonal in the standard bell basis and is completely characterized by @xmath91 displayed in fig .
[ fig2](a ) by a solid curve . in this section
, we employed the swapping protocol introduced in our previous paper .
this protocol is deterministic and exploits the same evolution as utilized in the previous section along with atomic projections .
in contrast to the entanglement distribution and purification protocols , however , the swapping protocol is entirely characterized by the success probability @xmath140 that is affected mainly by the efficiency of atomic projective measurements and can be close to one ( see below ) .
we conclude , therefore , that the total probability of success associated with the entanglement distribution , purification , and two swappings is given by the expression @xmath141 + for simplicity , in the setup displayed in fig .
[ fig1](a ) , we considered just two repeater nodes ( b and c ) , where the atomic pairs @xmath121 , @xmath119 and @xmath120 , @xmath122 were initially entangled and given both by eq .
( [ state4 ] ) .
we are ready now to introduce the experimental setup that includes explicitly nodes a , b , c , and d. this setup is displayed in fig .
[ fig3 ] and , in contrast to fig .
[ fig1](a ) , includes entanglement distribution and purification protocols associated with the ( initially disentangled ) atomic pairs @xmath121 , @xmath119 and @xmath120 , @xmath122 .
simultaneously with the atomic pair @xmath109 , @xmath110 , these pairs are conveyed along the setup ( but in the opposite direction ) and follow the same sequence of interactions and projective measurements .
we recall that all three building blocks of our repeater require projective measurements of atoms which are located inside and outside the cavity .
the method of atomic non - destructive measurements demonstrated in refs .
@xcite , @xcite enables projective measurements of atoms coupled ( strongly ) to a cavity field that fits perfectly in our experimental setup .
the physical mechanism behind these measurements exploits the suppression of cavity transmission due to the strong atom - cavity coupling . recall that each atom in our scheme is a three - level atom in the @xmath6-configuration [ see fig .
[ fig1](c ) ] , where only the states @xmath24 and @xmath41 are coupled to the cavity field .
if one such atom couples the cavity and is prepared in the @xmath24 state , such that the cavity resonance is sufficiently detuned from the atomic @xmath142 transition frequency , then the cavity transmission drops according to the atom - cavity detuning and atom - cavity coupling . on the other hand ,
the cavity transmission remains unaffected if the atom was prepared in the state @xmath25 .
once sufficiently many readouts of cavity transmission are recorded , this technique enables us to determine the state of a single atom with a high efficiency @xcite .
since the atom - cavity coupling increases proportionally with the number of loaded ( into cavity ) atoms , the same technique enables us to distinguish the following three separate states of two atoms ( i ) @xmath143 , ( ii ) @xmath144 or @xmath145 , and ( iii ) @xmath146 ( see @xcite ) .
we remark , however , that this approach can not distinguish between the states @xmath144 and @xmath145 leading to an incompleteness of information about the swapped state ( [ state6 ] ) [ see table [ tab1 ] ] . in order to avoid this problem ,
one of atoms in @xmath3 ( @xmath0 ) has to leave the cavity right after the inconclusive event occurs and be again projected outside the cavity .
the atomic projection outside the cavity is realized using a laser beam and a ccd camera located inside the oval regions displayed in fig .
[ fig1](a ) below the cavities @xmath1 and @xmath2 .
while the laser beam removes atoms in a given quantum state from the conveyor without affecting atoms in the other state ( so - called push - out technique @xcite ) , the ccd camera is used to detect the presence of remaining atoms via fluorescence imaging and determine , therefore , the state of each atom in question .
in contrast to the previously discussed atomic projection , this technique is destructive and , thus , the projected atoms can not be further utilized .
( a ) , @xmath23 ( b ) , and @xmath82 ( c ) , we display the overall repeater distance @xmath147 as a function of @xmath148 ( number of segments ) plotted for the final fidelities @xmath149 ( solid curve ) , @xmath150 ( dotted curve ) , @xmath151 ( dashed curve ) , and @xmath152 ( dot - dashed curve ) . for the average photon numbers @xmath153 ( d ) , @xmath23 ( e ) , and @xmath82 ( f ) , we display re - scaled repeater rates ( [ rate2 ] ) as functions of @xmath49 ( elementary length ) plotted for the final fidelities @xmath149 ( solid curve ) , @xmath150 ( dotted curve ) , @xmath151 ( dashed curve ) , and @xmath152 ( dot - dashed curve ) .
we remark using plots ( d)-(f ) along with plots ( a)-(c ) , one can easily calculate re - scaled repeater rates as functions of entire distance @xmath154.,title="fig:",scaledwidth=95.0% ] +
in the previous section , we have introduced our repeater scheme encapsulating three segments ( four nodes ) that corresponds to the overall distance of @xmath155 .
the extension of the sketch shown in fig .
[ fig3 ] to an arbitrary number @xmath148 of segments is straightforward . with no loss of generality
, we consider odd values of @xmath148 corresponding to @xmath156 repeater nodes or @xmath157 swappings .
setting @xmath158 , and @xmath82 , we display in figs .
[ fig4](a ) , ( b ) , and ( c ) the overall distance as a function of @xmath148 taken for the final fidelities @xmath159 , which are identified with the fidelity obtained after @xmath157 swapping operations .
we infer from this figure that smaller values of @xmath160 lead to larger overall distances , while the segment length @xmath49 decreases with the increasing of @xmath148 .
we remark that this drawback is originated to the lack of repurification in our scheme that is supposed to act after each ( or a few ) swapping operation(s ) .
we also emphasize that the obtained final fidelity is higher by about one order of magnitude as the respective fidelity obtained in our previous paper ( compare , for instance , fig .
[ fig4](a ) to fig .
5(a ) in ref . @xcite ) . besides final fidelities
, we calculate the repeater rates which , together with the success probabilities displayed in fig . [ fig2](b ) , provide the main set of characteristics associated with a quantum repeater . since the atomic ( fast - decaying ) excited states remain almost unpopulated and each atomic qubit is encoded by other two ( long - living ) states , we assume that the atomic coherence times exceeds the overall time required to complete entanglement distribution , purification , and swapping in all repeater nodes .
this assumption corresponds to a repeater with an ideal memory and implies that the only source of decoherence is the photon loss in the transmission channel .
we compute repeater rates ( in units of pairs per second ) using the following expression @xcite @xmath161 where @xmath162 is the total time required to distribute and purify an entangled pair over one repeater segment followed by two swappings , @xmath148 is the number of segments , while @xmath163 according to the experimental setup displayed in fig .
[ fig1](a ) , the total time is composed of ( i ) interaction times required for entanglement distribution , purification , and swapping protocols , ( ii ) time required for both destructive and non - destructive atomic projections , and ( iii ) time required to communicate between the repeater nodes by means of coherent - state light and classical communication .
in contrast to the cases ( i ) and ( iii ) , the time ( iii ) is readily computed using the relation @xmath164 , where @xmath165 m / s is the velocity of light in the optical fiber . without loss of generality , therefore , we express the total time as @xmath166 , where @xmath167 is a real number encoding the times ( i ) and ( iii ) in terms of ( iii ) . by inserting this expression into eq .
( [ rate1 ] ) , we consider the re - scaled repeater rate @xmath168 this re - scaled repeater rate is determined by the triplet @xmath169 , which we extract from fig .
[ fig4 ] for a given value of @xmath170 . setting @xmath171 , and @xmath82
, we display in figs .
[ fig4](d ) , ( e ) , and ( f ) the re - scaled repeater rates as functions of @xmath49 taken for the final fidelities @xmath172 , and @xmath173 .
we infer from this figure that smaller values of @xmath160 lead to smaller repeater rates @xmath174 .
we remark that this behavior was expected since small @xmath160 imply a larger total distance @xmath154 , which in turn , leads to a reduction of the amount of produced entangled pairs ( per second ) . in this paper , a complete quantum repeater scheme that encapsulates
entanglement distribution , purification , and swapping protocols was proposed .
in contrast to conventional repeater schemes , we completely avoid two - qubit logical gates by exploiting cavity qed evolution along atomic projective measurements .
our repeater scheme has a conveyor like structure , in which single atoms are inserted into one of two optical lattices and conveyed along the entire repeater node . at the same time
, another chain of atoms is conveyed along the neighboring repeater node in a synchronous fashion .
these two nodes form together a repeater segment , while the entire set of segments form the quantum repeater itself . in figs .
[ fig1](a ) and [ fig3 ] , the sketch of experimental setup was displayed and a detailed description of all necessary steps and manipulations were provided .
a comprehensive analysis of success probabilities and final fidelities obtained after multiple purification and swapping operations was given .
in addition , the comparison with regard to the results obtained in our previous paper has been provided .
in particular , our entanglement distribution here is based on a simpler coherent - state discrimination measurement distinguishing only two coherent states with opposite signs , which can be easily and optimally done by linear optics and on - off detections . following recent developments in cavity qed ,
finally , we briefly pointed to and discussed a few practical issues related to the implementation of our repeater scheme .
we stress that although the physical resources utilized in our repeater are experimentally feasible , its explicit realization for a long - distance quantum communication still poses a serious challenge . *
acknowledgments * + we thank the bmbf for support through the q.com ( former quorep ) program .
in this appendix , we show that the evolutions ( [ evol - rot ] ) and ( [ evol - displ ] ) governed by the hamiltonians ( [ ham - rot ] ) and ( [ ham - displ ] ) are produced deterministically in our framework .
namely , we consider an atom coupled strongly to the cavity and subjected to the detuned laser fields as displayed in fig .
[ fig1](c ) .
the evolution of this interacting system is governed by the hamiltonian @xmath175 \nonumber \\ & + & \hbar \left ( \omega_1 { | 1 \rangle } { \langle 1 | } + \omega_e { | e \rangle } { \langle e | } + \omega_0 { | 0 \rangle } { \langle 0 | } \right ) \ , , \end{aligned}\ ] ] where @xmath176 denotes the coupling strength of atoms to the cavity mode , while @xmath177 denotes the coupling strengths of atoms to both laser fields .
we switch to the interaction picture using the unitary transformation @xmath178}. \nonumber\ ] ] in this picture , the hamiltonian ( [ ham1 ] ) takes the form @xmath179 where the notation @xmath180 , @xmath181 , and @xmath182 has been introduced .
we require that @xmath183 and @xmath184 are sufficiently far detuned , such that no atomic @xmath185 or @xmath186 transitions can occur .
we expand the evolution governed by the hamiltonian ( [ ham2 ] ) in series and keep the terms up to the second order , that is , @xmath187 by performing integration and retaining only linear - in - time contributions , we express this evolution in the form @xmath188,\ ] ] where the effective hamiltonian is given by @xmath189 after removing constant contributions .
we switch to the interaction picture with respect to the first term of @xmath190 . in this picture
, we obtain @xmath191 we switch now from the atomic basis @xmath192 to the basis @xmath193 , where @xmath194 in this basis , the hamiltonian ( [ ham4 ] ) takes the form @xmath195 , \qquad\end{aligned}\ ] ] where @xmath196 and @xmath197 , and where we removed all the constant contributions .
we switch one more time to the interaction picture with respect to the first term of ( [ ham5 ] ) . in this picture , the hamiltonian ( [ ham5 ] ) becomes @xmath198 .
\nonumber\end{aligned}\ ] ] in the strong driving regime , i.e. , for @xmath199 , we eliminate the last ( fast oscillating ) term using the same arguments as for the rotating wave approximation . using the identity @xmath200
, the hamiltonian ( [ ham6 ] ) reduces to @xmath201 in the case of vanishing @xmath202 ( equivalently @xmath203 ) , the above hamiltonian takes the simplified form @xmath204 which , under the notation @xmath205 , coincides with the hamiltonian ( [ ham - displ ] ) . in the case of non - vanishing @xmath202
, we introduce @xmath206 , such that @xmath207 . due to this assumption ,
we eliminate in eq .
( [ ham6 ] ) the first term along with the terms @xmath208 and @xmath209 due to the rotating wave approximation , by which these fast oscillating terms play a minor role in the evolution . without these terms
, the hamiltonian ( [ ham6 ] ) takes the usual jaynes - cummings form @xmath210 expressed in the ( effective ) atomic basis @xmath211 .
we require , finally , that @xmath212 is sufficiently far detuned , such that the transition @xmath213 can not occur .
as above , we expand again the evolution governed by ( [ ham9 ] ) in series up to the second order and retain only linear - in - time contributions after the integration .
this procedure leads to the effective hamiltonian , @xmath214 where we removed all constant contributions and used the identity @xmath200 .
since the second term commutes with the first term , we eliminate the second term by means of an appropriate interaction picture .
the resulting hamiltonian [ that is , the first term of ( [ ham10 ] ) ] , coincides with ( [ ham - rot ] ) under the notation @xmath215 . | in the framework of cavity qed , we propose a quantum repeater scheme that uses coherent light and chains of atoms coupled to optical cavities .
in contrast to conventional repeater schemes , we avoid the usage of two - qubit quantum logical gates by exploiting solely the cavity qed evolution . in our previous paper [ d. gonta and p. van loock : phys .
rev .
a * 88 * , 052308 ( 2013 ) ] , we already proposed a quantum repeater in which the entanglement between two neighboring repeater nodes was distributed using controlled displacements of input coherent light , while the produced low - fidelity entangled pairs were purified using ancillary ( four - partite ) entangled states . in this paper ,
the entanglement distribution is realized using a sequence of controlled phase shifts and displacements of input coherent light .
compared to previous coherent - state - based distribution schemes for two - qubit entanglement , the present scheme relies only upon a simple discrimination of two coherent states with opposite signs , which can be performed in a quantum mechanically optimal fashion via a beam splitter and two on - off detectors .
the entanglement purification utilizes a scheme that avoids the usage of ancillary entangled resources .
our repeater scheme exhibits reasonable fidelities and repeater rates providing an attractive platform for long - distance quantum communication . |
recently , there has been an increasing interest for the nonperturbative quantities of quantum chromodynamics ( qcd ) , observed in hard scattering processes , together with the experimental and theoretical progresses . among those physical quantities ,
fragmentation and parton - distribution functions reveal important information on the internal structures of the hadrons .
especially , the fragmentation function plays an crucial role in describing semi - inclusive hadron productions in electron annihilation and deeply inelastic scattering ( dis ) processes . in ref .
@xcite , various fragmentation functions are intensively investigated using the @xmath0 evolution equation , such as the dokshitzer - gribov - lipatov - altarelli - parisi ( dglap ) equation , resulting in empirical parametrizations for those fragmentation functions .
more practical approaches have been done for the fragmentation functions , taking into account the effective chiral models for qcd , i.e. liner ( pseudoscalar , ps ) and nonlinear ( pseudovector , pv ) @xmath1 models ( or quark - pseudoscalar ( ps ) meson coupling model ) @xcite , and nambu - jona - lasinio ( njl ) model @xcite .
it is worth noting that , in refs .
@xcite , it was not uniquely determined which coupling schemes , i.e. ps and pv ones , are more reliable to describe the fragmentation function . comparing with the njl model calculations @xcite , the ps scheme results looks similar to that from the njl model . in ref .
@xcite , the authors performed the dglap evolution using the numerical code qcdnum for the dgalp evolution @xcite , in addition to the contributions from the resonances and jets .
in the present work , we employ the nonlocal chiral quark model ( nlchqm ) defined in minkowski space .
a unique feature of this model , which is not observed in other models , is that quarks and ps mesons are interacting each other nonlocally : this interesting nonlocal interaction can be regarded as the consequence of the quark and nontrivial qcd vacuum interactions in terms of the the semiclassical nontrivial solution for the yang - mills equation , i.e. instanton @xcite . from the numerical calculations , we observe that the present model provides quite distinctive results in comparison to those from other models .
especially , the bump shown in the fragmentation function curve appears at @xmath2 , whereas the ps - scheme and njl results produce it at larger @xmath3 values @xmath4
. this tendency observed in the present framework , saying , the bump appears in the lower @xmath3 regions , may give more comparable results with the empirical data at higher @xmath0 after the dglap evolution , in comparison to other models .
the present report is structured as follows : in sections ii and iii , we briefly introduce the unpolarized fragmentation function and the nlchqm which is based on the instanton vacuum configuration in principle .
the numerical results are given in section iv with related discussions .
final section is devoted to summary , conclusion , and future plans .
the unpolarized fragmentation function @xmath5 for @xmath6 , in which @xmath7 and @xmath8 indicate the quark with the flavor @xmath9 and ps meson respectively , is defined in the light - cone coordinate , assuming the light - cone gauge , as follows @xcite : @xmath10|_{zk^-=p^-},\ ] ] where @xmath11 and @xmath3 stands for the on - shell four momentum of the meson and its momentum faction possessed by a quark fragmented , whereas @xmath12 for the time- and space - momentum components of the quark on the light cone , respectively . @xmath13 and @xmath14 denotes the transverse momentum of the quark and renormalization scale at which the fragmentation process evolved . for the later convenience ,
we assign the fragmentation function in eq .
( [ eq : frag ] ) as the _ elementary _ fragmentation function , since it represents an elementary fragmentation process from a single quark to a ps meson , and in order to distinguish it from the _ renormalized _ one as discussed in ref .
@xmath15 corresponds to the relevant correlation for the fragmentation process .
in order to investigate this nonperturbative quantity , defined in eq .
( [ eq : frag ] ) , we will use nlchqm , which is motivated from the instanton - liquid model @xcite .
we note that , to date , various nonperturbative qcd properties have been well studied in terms of the instanton vacuum configuration , being comparable to experiments as well as lattice qcd simulations @xcite . in the instanton model ,
nonperturbative qcd effects are deciphered by the nontrivial quark - instanton interactions in the dilute instanton ensemble , interpreting the spontaneous breakdown of chiral symmetry ( sbcs ) . considering these interesting features , below , we write the effective chiral action ( echa ) of nlchqm in minkowski space as follows : @xmath16=-\mathrm{sp}\ln\left[i\rlap{/}{\partial } -\hat{m}_f-\sqrt{m(\partial^2)}u^{\gamma_5}\sqrt{m(\partial^2)}\right],\ ] ] where @xmath17 and @xmath18 denote the functional trace @xmath19 over all the relevant spin spaces and su(3 ) current - quark mass matrix @xmath20 , respectively .
note that , in deriving echa in eq .
( [ eq : eca ] ) , we have simply changed the euclidean metric in the instanton - induced effective action into that for minkowski space , assuming an appropriate analytic continuation between those spaces @xcite .
although this simple change of the metric is not fully understood , in practical applications , it works qualitatively very well @xcite .
the momentum - dependent effective quark mass , which is generated from the nontrivial interactions between the quarks and nonperturbative qcd vacuum , can be written in a simple @xmath21-pole type form factor as follows : @xmath22^{n}.\ ] ] here , @xmath21 stands for an arbitrary but finite positive integer
. we will take @xmath23 with @xmath24 mev , as usual in the instanton - motivated models @xcite , throughout the present work .
note that @xmath25 stands for the nonlinear ps - meson field . for more details on this notation ,
one refers refs .
@xcite . by expanding it up to @xmath26 from echa in eq .
( [ eq : eca ] ) , one can derive the following effective interaction lagrangian density for the nonlocal quark - quark - ps meson vertex which is relevant for the fragmentation function in hand : @xmath27 where @xmath28 denotes the weak - decay constant for the ps meson @xmath8 . finally , after performing the trace over the relevant spin spaces in eq .
( [ eq : frag ] ) using eq .
( [ eq : lag ] ) , we arrive at a neat expression for the elementary fragmentation function in terms of the nlchqm as follows : @xmath29 ^ 2\ } } { [ z^2k^2_t+z(z-1)\bar{m}^2_f+z\bar{m}^2_{f'}+(1-z)m^2_\phi]^2},\ ] ] where we have used the notation @xmath30 , where @xmath31 indicates the current - quark mass with the flavor @xmath9 .
we note that , if we turn off the nonlocal interaction , i.e. taking @xmath32 , the expression in eq .
( [ eq : ddddd ] ) turns into that for the local interaction models , such as refs .
@xcite , with a proper prefactor .
in this section , we present numerical results for the elementary fragmentation function for the pion and kaon via the nlchqm . in the left panel of fig .
[ fig1 ] , we depict the curves for the favored elementary fragmentation functions as functions of @xmath3 at the renormalization scale @xmath33 gev : @xmath34 ( solid ) , @xmath35 ( dot ) , and @xmath36 ( dash ) . as indicated in the left panel of fig .
[ fig1 ] , the kaon fragmentation functions are much smaller than that for the pion , since the pion fragmentation must be much populated at low energy scale @xmath37 . in the right panel of fig .
[ fig1 ] , we show the numerical results from refs . @xcite for comparison .
we note that the ps scheme result is almost the same with that from the njl model calculation @xcite .
if we compare ours with the ps - scheme result , which is comparable to the njl one , it is found that the nonlocal quark - ps meson interaction makes considerable differences , i.e. the bump moves to smaller-@xmath3 regions , almost locating at the middle of the range @xmath38 . [ cols="^,^ " , ] although , in this report , we do not perform the dglap evolution for higher @xmath0 to compare with the empirical data for the fragmentation functions @xcite , taking into account that the dglap evolution enhances the curve far more in the vicinity of @xmath39 , the present numerical results with the bump in the middle are expected to give more reliable dglap evolution results , in comparison to those for the ps - scheme ( or njl model ) results @xcite . moreover , as expected , we can conclude that the present nonlocal results behave like the average between the ps- and pv - scheme ones qualitatively as shown in fig .
more detailed analyses on these observations will be discussed in separate larger volumes .
in this report , we have studied the unpolarized - fragmentation function , employing the nonlocal chiral quark model . from the numerical results
, we observed that the nonlocal quark - ps meson interaction , which is a unique feature of the present theoretical framework , plays an important and distinctive role to produce the curves for the fragmentation function , in comparison to other effective model calculations .
it was also viewed that calculated curves for the fragmentation function in the present work can provide more realistic results , being comparable to the empirical data after the dglap evolution to higher @xmath0 values . more detailed analyses , including the meson - cloud effects , parton - distribution function using the drell - levi - yan ( dly ) relation , and dglap evolutions to higher @xmath40 , will appear soon elsewhere .
this talk was presented at the international conference _
the fifth asia - pacific conference on few - body systems in physics 2011 _ ( apfb2011 ) , seoul , republic of korea , @xmath41 august 2011 .
the authors thank p. ko , j. w. chen , h. kohyama for fruitful discussions .
s.i.n . is grateful to the hospitality during his staying at national taiwan university ( ntu ) with the financial support from ncts ( north ) in taiwan , where the present work was initiated .
the works of s.i.n . and b.g.y
. were partially supported by the grant nrf-2010 - 0013279 from national research foundation ( nrf ) of korea .
the work of c.w.k . was supported by the grant nsc 99 - 2112-m-033 - 004-my3 from national science council ( nsc ) of taiwan .
3 m. hirai , s. kumano , t. h. nagai and k. sudoh , phys .
d * 75 * , 094009 ( 2007 ) .
a. bacchetta , r. kundu , a. metz and p. j. mulders , phys .
d * 65 * , 094021 ( 2002 ) .
d. amrath , a. bacchetta and a. metz , phys .
d * 71 * , 114018 ( 2005 ) .
t. ito , w. bentz , i. -ch .
cloet , a. w. thomas and k. yazaki , phys .
d * 80 * , 074008 ( 2009 ) .
m. botje , comput .
commun . * 182 * , 490 ( 2011 ) .
qcdnum : dglap evolution program by m. botje ( http://www.nikhef.nl/user/h24/qcdnum ) .
d. diakonov and v. y. petrov , nucl .
b * 272 * , 457 ( 1986 ) .
e. v. shuryak , nucl .
b * 203 * , 93 ( 1982 ) .
d. diakonov and v. y. petrov , nucl .
b * 245 * , 259 ( 1984 ) .
d. diakonov , prog .
* 51 * , 173 ( 2003 ) .
t. schfer and e. v. shuryak , rev .
* 70 * , 323 ( 1998 ) .
j. mulders and r. d. tangerman , nucl .
b * 461 * , 197 ( 1996 ) [ erratum - ibid .
b * 484 * , 538 ( 1997 ) ] .
s. i. nam and h. -ch .
kim , phys . rev .
d * 77 * , 094014 ( 2008 ) .
s. i. nam and h. -ch .
kim , phys .
b * 700 * , 305 ( 2011 ) .
m. musakhanov , eur .
j. * c9 * , 235 ( 1999 ) .
m. musakhanov , nucl .
a * 699 * , 340 ( 2002 ) .
s. i. nam , h. -ch .
kim , a. hosaka and m. m. musakhanov , phys .
d * 74 * , 014019 ( 2006 ) .
s. i. nam and h. -ch .
kim , phys .
d * 74 * , 076005 ( 2006 ) .
m. praszalowicz and a. rostworowski , phys .
d * 64 * , 074003 ( 2001 ) .
m. v. polyakov and c. weiss , phys .
d * 59 * , 091502 ( 1999 ) . | in this talk we present our recent studies for the unpolarized fragmentation functions for the pion and kaon , employing the nonlocal chiral quark model , which manifests the nonlocal interaction between the quarks and pseudoscalar mesons , in the light - cone frame . it turns out that the nonlocal interaction produces considerable differences in comparison to typical local - interaction models . |
in early 1970 s , efimov pointed out a possibility of having an infinite number of three - body bound states even when none exists in the separate two - body subsystems @xcite .
this occurs when the two - body scattering length is much larger than the range of the two - body interaction . as a candidate of such three - body states ,
efimov discussed about the famous hoyle state @xcite ( the second @xmath8 state at 7.65mev in the @xmath9c nucleus ) taking a model of three @xmath10 particles ( clusters of three @xmath0he nuclei ) as well as about the three - nucleon bound state ( @xmath11h nuclei ) . in nuclear systems , the borromean states , weakly bound three - body states though having no bound two - body subsystems , are familiar but not classified as efimov states . in atomic systems , triatomic @xmath0he ( trimer ) have been expected to have bound states of efimov type since the realistic @xmath0he-@xmath0he interactions @xcite give a large @xmath0he-@xmath0he scattering length ( @xmath12 ) , much greater than the potential range ( @xmath13 ) , and a very small @xmath0he dimer binding energy ( @xmath14 mk ) .
( experimentally , ref .
@xcite evaluated a scattering length of @xmath15 @xmath16 and a binding energy of @xmath17 mk ) . as is mentioned in recent reviews about the @xmath0he trimer @xcite ( further references therein ) , i ) a lot of three - body calculations using the realistic pair potentials have shown that the @xmath0he trimer possesses two bound states with binding energies of nearly @xmath18 mk and @xmath19 mk , rather well established that , if the @xmath0he trimer excited state exist
, it should be efimov nature , and iii ) it is suggested that the @xmath0he trimer ground state may be considered as an efimov state since the ground- and excited - state binding energies move along the same universal scaling curve under any small deformation of the two - body potential ( for details , see , e.g. , sec.iii of ref .
experimentally , the @xmath0he trimer ground state has been observed in ref .
@xcite to have the @xmath0he-@xmath0he bond length of @xmath20 @xmath16 in agreement with theoretical predictions , whereas a reliable experimental evidence for the @xmath0he trimer excited state is still missing . only very recently , experimental evidences of efimov trimer states have been reported in the work using the ultracold gases of cesium atoms @xcite , potasium atoms @xcite , lithium-7 atoms @xcite , and atoms @xcite , in which the two - body interaction between those alkali atoms was manipulated so as to tune the scattering length to values significantly greater than the potential range .
these experiments have been accessing the study of a wide variety of interesting physical systems in the atomic and nuclear fields .
recently , the study extends to the efimov physics and universality of four - atomic systems ( tetramers ) .
though the interactions between @xmath0he atoms can not be manipulated , the study of @xmath0he trimer using the realistic pair potentials has been providing fundamental information to the efimov physics .
now it is one of the challenging subjects to precisely investigate the structure of @xmath0he tetramer using the realistic @xmath0he-@xmath0he potential .
so far there exist in the literature a large number of @xmath0he trimer calculations @xcite giving well converged results with the realistic @xmath0he-@xmath0he interactions .
however , calculations of the tetramer remain limited @xcite ; in those papers , the binding energy of the tetramer ground state agrees well with each other , while that of the loosely bound excited state differs significantly from one another .
thus the main purpose of the present paper is to perform accurate calculations of the @xmath0he tetramer ground and excited states using a realistic @xmath0he-@xmath0he interaction , the lm2m2 potential @xcite .
we employ the gaussian expansion method ( gem ) for _ ab
initio _ variational calculations of few - body systems @xcite .
the method has been proposed and developed by the present authors and collaborators and applied to various types of three- , four- and five - body systems in nuclear physics and exotic atomic / molecular physics ( cf .
review papers @xcite ) .
advantage of using the gem for the @xmath0he tetramer calculation in the presence of the strong short - range repulsive potential is as follows : some 30000 symmetrized four - body gaussian basis functions , ranging from very compact to very diffuse , are constructed on the full 18 sets of jacobi coordinates without assuming any pair correlation function .
they forms a nearly complete set in the finite coordinate space concerned , so that one can describe accurately both the short - range structure and the long - range asymptotic behavior ( up to @xmath2 ) of the four - body wave function , which makes it possible to find new facets of @xmath0he clusters .
we thus find that precisely the same shape of the short - range correlation ( @xmath21 ) in dimer appears in the ground and excited states of trimer and tetramer .
this gives a foundation to an _ a priori _ assumption that a two - particle correlation function ( such as the jastrow s ) so as to simulate the short - range part of the dimer wave function is incorporated in the trimer and tetramer wave functions from the beginning . by illustrating the asymptotic behavior of the @xmath0he trimer and tetramer , we discuss about an interesting relation between their excited - state wave functions and the dimer wave function .
we then come to propose a dimerlike - pair model that predicts the binding energy of the first excited state of the @xmath22-cluster system , @xmath0he@xmath3 , measured from the ground state of @xmath0he@xmath5 to be approximately @xmath6 using the dimer binding energy @xmath7 .
we explicitly write the asymptotic form of the total wave function of @xmath0he trimer ( tetramer ) .
the asymptotic normalization coefficient ( anc ) @xcite , namely the amplitude of tail function of the ( trimer - atom ) relative motion in the present case , is a quantity to reflect the internal structure of trimer ( tetramer ) . therefore , attention to the anc might be useful when one intends to reproduce the non - universal variation of the @xmath0he trimer ( tetramer ) states by means of parametrizing models beyond efimov s universal theory .
the paper is organized as follows : in sec .
ii , we apply the gem to the three - body calculation of the @xmath0he trimer ground and excited states showing that the calculated results agree excellently with the literature . in sec .
iii , the four - body calculation of the @xmath0he tetramer ground and excited states is presented .
summary is given in sec .
the @xmath0he trimer bound states have extensively been studied in many theoretical work using realistic potentials .
monte - carlo , hyperspherical , variational and faddeev techniques were used to calculate accurately the binding energies of the ground and excited states @xcite ( see also recent reviews @xcite ) .
nevertheless , in this section , we explain our gaussian expansion method ( gem ) and present the calculated result for the @xmath0he trimer in order to demonstrate high accuracy of our calculation before we report our investigation of the @xmath0he tetramer in the next section .
we take all the three sets of jacobi coordinates ( fig .
[ fig:3bodyjacobi ] ) , @xmath23 and @xmath24 and cyclically for @xmath25 and @xmath26 , @xmath27 being the position vector of @xmath28th particle .
hamiltonian of the system is expressed as @xmath29 where @xmath30 and @xmath31 , @xmath32 being mass of a @xmath0he atom .
@xmath33 is the two - body @xmath0he-@xmath0he potential as a function of the pair distance @xmath34 .
we calculate the three - body bound - state wave function , @xmath35 , which satisfies the schrdinger equation @xmath36 since we consider the @xmath0he atom as a spinless boson , we expand the wave function of three identical spinless bosons in terms of @xmath37-integrable , fully symmetric three - body basis functions : @xmath38 it is of importance that those basis functions \{@xmath39 } , which are nonorthogonal to each other , are constructed on the full three sets of jacobi coordinates ; this makes the function space of @xmath40 } quite wide .
the eigenenergies @xmath41 and amplitudes @xmath42 of the ground and excited states are determined by the rayleigh - ritz variational principle : @xmath43 where @xmath44 .
eqs.(2.5 ) results in a generalized eigenvalue problem : @xmath45 \,a_{\alpha ' } = 0 .
\qquad\ ] ] the matrix elements are written as @xmath46 the lowest - lying two @xmath47-wave eigenstates , @xmath48 , will be identified as the trimer ground @xmath49 and excited ( @xmath50 ) states .
we express each basis function @xmath51 as a product of a function of @xmath52 and that of @xmath53 : @xmath54_{j m},\ ] ] where @xmath10 specifies a set of quantum numbers @xmath55 commonly for the components @xmath56 .
@xmath57 is the total angular momentum and @xmath58 is its @xmath59-component . in this paper
, we consider the trimer bound states with @xmath60 .
then , the totally symmetric three - body wave function requires @xmath61 .
one of the most important issues of the present variational calculation is what type of radial shape we use for @xmath62 and @xmath63 .
the basis functions should be capable of precisely describing the strong short - range correlation ( without assuming any correlation function _ a priori _ ) and the long - range asymptotic behavior of very loosely bound states .
the gem recommends two types of functions which are tractable in few - body calculations and work acculately .
one is the gaussian function and the other , more powerful one , is the complex - range gaussian function @xcite . in the next subsection
, we introduce the former that was successfully used in our previous study ( sec . 3.1 of ref .
@xcite ) of the @xmath0he trimer ground and excited state with the use of the hfdhe2 potential @xcite .
the latter function is introduced in sec.ii.c . the radial function @xmath64 in ( 2.9 ) is taken to be a gaussian multiplied by @xmath65 ( similarly for @xmath66 ): @xmath67 where normalization constants are omitted for simplicity .
setting of the ranges by stochastic or random choice does not seem suitable for describing the strong short - range correlation and the long - range asymptotic behavior of the wave function .
any intended choice of the ranges is necessary .
the gem recommends to set them in a _
geometric _ progression : @xmath68 with common ratios @xmath69 and @xmath70 .
this greatly reduces the nonlinear parameters to be optimized .
we designate a set of the geometric sequence by @xmath71 instead of @xmath72 and similarly for @xmath73 , which is more convenient to consider the spatial distribution of the basis set .
optimization of the nonlinear range parameters is in principle by trial and error procedure but much of experiences and systematics have been accumulated in many studies using the gem . the basis functions \{@xmath74 } have the following properties : i ) they range from very compact to very diffuse , more densely in the inner region than in the outer one .
while the basis functions with small ranges are responsible for describing the short - range structure of the system , the basis with longest - range parameters are for the asymptotic behavior .
ii ) they , being multiplied by normalization constants for @xmath75 , have a relation @xmath76 which tells that the overlap with the @xmath77-th neighbor is _ independent _ of @xmath78 , decreasing gradually with increasing @xmath77 .
we then expect that the coupling among the whole basis functions take place smoothly and coherently so as to describe properly both the short - range structure and long - range asymptotic behavior simultaneously .
we note that a single gaussian decays quickly as @xmath79 increases , but appropriate superposition of many gaussians can decay even exponentially up to a sufficiently large distance .
a good example is shown in fig . 3 of ref .
@xcite for the @xmath0he dimer wave function ( with the hfdhe2 potential ) that is accurate up to @xmath80 @xmath16 with the use of the nonlinear parameters \{@xmath81 , @xmath82
and @xmath83 @xmath16 } ( the same - quality dimer wave function is seen in fig . 2 below in sec.ii.d using the complex - range gaussians with the lm2m2 potential ) . a lot of successful examples of the three- and four - body gem calculations are shown in review papers @xcite and in papers of five - body calculations @xcite .
the examples includes our previous calculation of the ground and excited states of @xmath0he trimer using the hfdhe2 potential ; the binding energies were in good agreement with those given by a feddeev - equation calculation @xcite .
as for the trimer wave function , we showed , in figs . 3 , 13 and 14 in ref.@xcite , that the strong short - range correlation ( @xmath84 ) and asymptotic behavior ( up to @xmath85 ) of the trimer ground and excited states are simultaneously well described .
also , the three - body basis functions ( 2.9)(2.13 ) together with the lm2m2 potential were used recently by naidon , ueda and one of the present authors ( e. h. ) @xcite to study the universality and the three - body parameter of @xmath0he trimers . before we proceed to the calculation of the @xmath0he tetramer ground and excited states
, we improve the gaussian shape of the basis functions so as to have more sophisticated ( but still tractable ) radial dependence .
we then test the new basis in the calculation of the trimer states below . in ref .
@xcite , we proposed to improve the gaussian shape by introducing _ complex _ range instead of the real one : @xmath86 where @xmath87 and @xmath88 are given by ( 2.12 ) . using @xmath89 , we construct two kinds of _ real _ basis functions : @xmath90/2 , \\
\phi_{nl}^{({\rm sin})}(x ) & = & x^l\:e^{-(x / x_n)^2}\ , { \rm sin}\ , \omega ( x / x_n)^2 \nonumber\\ & = & [ \phi_{nl}^{(-\omega)}(x)-\phi_{nl}^{(\omega)}(x ) ] /2i,\end{aligned}\ ] ] where we usually take @xmath91 .
the three - body basis function @xmath92 in ( 2.9 ) is replaced by @xmath93_{j m},\ ] ] where @xmath10 specifies a set @xmath94 the new basis \{@xmath95 } apparently extend the function space from the old ones ( 2.10 ) since they have the oscillating components ; see sec.2.4 and sec.2.5 of ref .
@xcite for some examples taking this advantage in calculations of highly vibrational excited states ( with @xmath96 25 nodes ) and scattering states .
the _ sin_-type basis ( 2.17 ) particularly work when the wave function is extremely suppressed at @xmath97 due to the strongly repulsive short - range potential . in the following calculations ,
we employ the new basis ( 2.16 ) and ( 2.17 ) for the @xmath79-space instead of ( 2.10 ) , but keep ( 2.11 ) for the @xmath98-space .
note that , when calculating the matrix elements ( 2.7 ) and ( 2.8 ) using @xmath99 and @xmath100 , we explicitly take ( 2.15 ) and the right - most expression of ( 2.16 ) and ( 2.17 ) since the computation programming is almost the same as that for ( 2.10 ) though some of real variables are changed to complex ones .
a great advantage of the real- and complex - range gaussian basis functions is that the calculation of matrix elements ( 2.7 ) and ( 2.8 ) is easily performed . as for the overlap and kinetic - energy matrix elements of the trimer ( tetramer ) ,
all the six-(nine-)dimensional integrals give analytical expression . in the case of the potential matrix
, we have analytical expression except for the one - dimensional numerical integral having the final form @xmath101 we explained , in ref .
@xcite , various techniques to perform the three- and four - body matrix - element calculations as easily , accurately and rapidly as possible .
it is to be emphasized that the gem few - body calculations need neither introduction of any _ a priori _ pair correlation function ( such as the jastrow function ) nor separation of the coordinate space by @xmath102 and @xmath103 , @xmath104 being the radius of a strongly repulsive core potential .
proper short - range correlation and asymptotic behavior of the total wave function are _ automatically _ obtained by solving the schrdinger equation ( 2.2 ) using the above basis functions for _ ab initio _ calculations . to describe the interaction between the @xmath0he atoms ,
we employ one of the most widely used @xmath0he-@xmath0he interactions , the lm2m2 potential by aziz and slaman @xcite .
use is made of @xmath105 k@xmath106 as the input mass of @xmath0he atom .
we can then precisely compare calculated results for the tetramer ground and excited states with those obtained by lazauskas and carbonell @xcite who made a faddeev - yakubovsky ( fy ) equation calculation taking the same potential and @xmath0he mass as above . recently ,
the authors of ref .
@xcite claim that a more precise value of @xmath107 k@xmath106 should be employed .
we shall additionally show the trimer and tetramer binding energies in the case of using this value .
we calculated the @xmath0he dimer binding energy , say @xmath7 , and the wave function , @xmath108 , using the same prescription as described above .
we expanded @xmath109 with 100 basis functions of ( 2.16 ) and ( 2.17 ) as @xmath110\ ] ] with a parameter set @xmath111 we obtained @xmath112 mk , @xmath113 , and @xmath114 @xmath16which are the same as those obtained in the literature .
experimentally , @xmath115 @xcite from which @xmath116 mk was estimated . as shown in fig . 2 , both the strong short - range correlation ( @xmath117 ) and the asymptotic behavior of the dimer are well described . in the lower panel
, @xmath118 precisely reproduces the exact asymptotic shape @xmath119 ( @xmath120 ) with @xmath121 @xmath122 up to @xmath123 @xmath16 which is large enough for our discussions .
there are 30 basis functions whose gaussian ranges @xmath124 , which is sufficiently dense to describe the short - range structure of the wave function precisely .
an interesting issue is whether the same shape of the short - range correlation in fig .
[ fig : dimer ] appear also in the trimer and tetramer ground and excited states without assuming any two - body correlation function .
we calculated the wave functions of the trimer ground state , @xmath125 , and the excited state , @xmath126 , and their binding energies , @xmath127 and @xmath128 , respectively , as well as some mean values with the @xmath129 @xmath130 .
some of results are summarized in table i together with those obtained in the literature .
our results excellently agree with those by refs.@xcite .
the @xmath0he-@xmath0he bond length in the trimer ground state was measured as @xmath131 @xcite , which is well explained by the calculations , @xmath132 .
crrrr trimer & + & present & ref.@xcite & ref.@xcite & ref.@xcite + @xmath127 ( mk ) & 126.40 & 126.39 & 126.4 & 126.40 + @xmath133 ( mk ) & 1660.4 & 1658 & 1660 & + @xmath134 ( mk ) & @xmath135 & @xmath136 & @xmath137 & + @xmath138 ( ) & @xmath139 & 10.95 & 10.96 & + @xmath140 ( ) & @xmath141 & 9.612 & 9.610 & + @xmath142 ( @xmath143 & @xmath144 & 0.135 & & + @xmath145 ( @xmath146 & @xmath147 & 0.0230 & & + @xmath148 ( ) & 6.326 & & 6.49 & 6.32 + @xmath149(@xmath150 & 0.562 & & 0.567 & + trimer & + & present & ref.@xcite & ref.@xcite & ref.@xcite + @xmath128 ( mk ) & 2.2706 & 2.268 & 2.265 & 2.2707 + @xmath133 ( mk ) & 122.15 & 122.1 & 121.9 & + @xmath134 ( mk ) & @xmath151 & @xmath152 & @xmath153 & + @xmath138 ( ) & @xmath154 & 104.3 & 101.9 & + @xmath140 ( ) & 84.51 & 83.53 & 83.08 & + @xmath142 ( @xmath143 & 0.0265 & 0.0267 & & + @xmath145 ( @xmath146 & 0.00216 & 0.00218 & & + @xmath148 ( ) & 60.33 & & 58.8 & 59.3 + @xmath155(@xmath150 & 0.179 & & 0.178 & + [ table : mean - trimer ] those converged results were given by taking the symmetric three - body basis function \{@xmath39 } with @xmath156 , in which the shortest - range set is @xmath157,@xmath158 ) and the longest - range one is @xmath159,@xmath160 ) .
all the nonlinear parameters of the gaussian basis set are listed in .
ccccccccccc & @xmath161 & & + 5 & & 4 & + @xmath162 & @xmath163 & @xmath164 & @xmath165 & @xmath166 & & @xmath167 & @xmath168 & @xmath169 & @xmath170 & @xmath171 number + & & [ ] & [ ] & & & & & [ ] & [ ] & @xmath171 of basis + + 0 & 22 & 0.3 & 150.0 & 1.0 & & 0 & 50 & 0.4 & 600.0 & 2200 + 2 & 17 & 0.6 & 150.0 & 1.0 & & 2 & 40 & 0.8 & 400.0 & 1360 + 4 & 14 & 0.8 & 130.0 & 1.0 & & 4 & 30 & 1.0 & 200.0 & @xmath172840 + + there are neither additional parameter nor assumptions .
the present calculation is so transparent that it is possible for the readers to repeat the calculation and check the results reported here .
the parameters for the gaussian ranges are in round numbers but further optimization of them do not improve the binding energies ( @xmath173 mk and @xmath174 mk ) as long as we calculate them with _ five _ significant figures ( cf .
another check in sec.ii.h about the accuracy of the calculation ) .
convergence of the binding energies @xmath127 and @xmath128 with respect to increasing partial waves @xmath175 is shown in table iii in comparison with the faddeev calculation by lazauskas and carbonell @xcite .
the case @xmath176 is sufficient in the present work as long as the accuracy of five significant digits is required .
the convergence of the present result is more rapid than that of the faddeev solution ( the same will be seen in the tetramer calculation in sec.iii ) .
the reason is that both the interaction and the wave function are truncated in the angular - momentum space @xmath177 in the faddeev calculations , but the full interaction is included in the present calculation ( with no partial - wave decomposition ) though the wave function is truncated @xmath177 .
the difference of the convergence in the two calculation methods was precisely discussed in the case of the three nucleon bound states ( @xmath11h and @xmath11he nuclei ) in our gem calculation @xcite and in a faddeev calculation @xcite ; for an illustration of the difference , see fig .
15 in ref . @xcite . in this context , it is worth pointing out that , in table [ table : mean - trimer ] , our result precisely agrees with another faddeev calculation by ref.@xcite with no the partial wave decomposition .
cccccc trimer & & @xmath171 & + & 2 & & 2 + @xmath178 & @xmath127 ( mk ) & @xmath128 ( mk ) & & @xmath127 ( mk ) & @xmath128 ( mk ) + 0 & 121.00 & 2.2397 & & 89.01 & 2.0093 + 2 & 126.39 & 2.2705 & & 120.67 & 2.2298 + 4 & 126.40 & 2.2706 & & 125.48 & 2.2622 + 8 & & & & 126.34 & 2.2677 + 12 & & & & 126.39 & 2.2680 + 14 & & & & 126.39 & 2.2680 + [ table : resonance ] use of the value @xmath107
k@xmath106 @xcite results in @xmath179 mk and @xmath180 mk , while ref .
@xcite gives @xmath181 mk and @xmath182 mk , respectively .
calculation of the binding energy was also made perturbatively with @xmath183 , where @xmath184 and @xmath185 are those obtained with @xmath105 k@xmath186 ; this gives @xmath187 mk and @xmath180 mk .
the calculations below in sec.ii.f-h take @xmath105 k@xmath186 . in order to see how the present method describes the short - range structure of trimer
, we calculated the pair correlation function ( pair distribution function or two - body density ) @xmath188 defined by @xmath189 where the symbol @xmath190 means the integration over @xmath191 only .
this integration gives an analytical expression owing to the use of the gaussian basis functions ; here , we explicitly rewrite @xmath129 as a function of @xmath192 by transforming the other coordinates @xmath25 and @xmath26 into @xmath192 .
@xmath193 is independent of @xmath194 and is apparently normalized as @xmath195 .
it presents the probability of finding two particles at an interparticle distance @xmath79 . in fig .
[ fig : trimer - den - short ] , short - range structure of @xmath196 is illustrated together with @xmath197 for the @xmath0he dimer .
the dashed line is for the trimer ground state @xmath49 .
the solid line for the excited state @xmath198 and the dotted line for the dimer have been multiplied by factors 14.5 and 6.0 , respectively .
it is of interest that precisely the same shape of the short - range correlation @xmath199 ) as seen in the dimer appears both in the trimer ground and excited states ( the same will be seen in the tetramer ground and excited states in the next section ) .
this gives a foundation to an _ a priori _ assumption that a two - particle correlation function ( such as the jastrow function ) to simulate the short - range part of the dimer radial wave function @xmath109 is incorporated in the three - body wave function from the beginning . to investigate the trimer configuration in the asymptotic region where one atom is far from the other two ,
we calculate the overlap function @xmath200 @xcite to describe the overlap between the trimer wave function @xmath201 and the dimer one @xmath202 : @xmath203 in fig .
[ fig : trimer - red - long ] , we plot @xmath204 for the ground and excited states
. they should asymptotically satisfy @xmath205 where @xmath206 is the binding wave number given by @xmath207 ( @xmath208 @xmath122 , @xmath209 @xmath122 ) . the amplitude @xmath210 is called the asymptotic normalzation coefficient ( anc ) @xcite defining the amplitude of the tail of the radial overlap function . the asymptotic functions ( 2.25 ) with @xmath211@xmath212 and @xmath213@xmath212 ( see the open circles ) are precisely reproduced by the dashed line ( @xmath214 ) and the solid line ( @xmath50 ) , respectively , up to @xmath215 @xmath16 , which demonstrates the accuracy of our wave functions in the asymptotic region .
the values of @xmath210 agree with those given by barletta and kievsky @xcite using a variational method with correlated hyperspherical harmonics functions ( see table i ) .
the total three - body wave function @xmath201 is represented asymptotically as @xmath216 the anc , @xmath210 , is a quantity to convey the interior structural information of the trimer to the asymptotic behavior .
it is known , in the nuclear peripheral reactions where only the asymptotic tails of the wave functions of reacting particles contribute to the reaction process , the cross section is proportional to the squared anc which can be measured in some specific systems @xcite .
the idea of anc might be available to the calculation of @xmath0he atoms reactions such as @xmath217 .
also , attention to the anc might be useful when one intends to reproduce the non - universal variations of the trimer states by parametrizing effective models . in fig .
[ fig : trimer - red - long ] , we find that the solid line ( @xmath50 ) is parallel to the dotted line ( dimer ) ; namely , @xmath218 @xmath122 ) is very close to @xmath219 @xmath122 ) .
this agreement is not accidental , but is understandable from a model , which we refer to as a dimerlike - pair model , for the asymptotic behavior of the trimer excited state ( fig .
[ fig : dimer - model]a ) .
the model tells that i ) particle @xmath220 , located far from @xmath221 and @xmath222 which are _ loosely _ bound ( dimer ) , is little affected by the interaction between @xmath221 and @xmath222 , ii ) therefore , the pair @xmath220 and @xmath221 at a distance @xmath79 is asymptotically dimerlike , iii ) since asymptotically , the amplitude of particle @xmath220 along * y * is dimerlike , namely @xmath223 .
if this model is acceptable , we can predict that , in the asymptotic region , the pair correlation function of the trimer excited state , @xmath224 , should decay exponentially with the same rate as that in the dimer ( @xmath225 ) .
this is clearly seen in fig .
[ fig : trimer - den - long ] ; the solid and dotted lines have almost the same exponentially - decaying rate of @xmath226 ) .
the same evidence is seen in figs . 3 and 14 of our previous calculation of the @xmath0he trimer using the hfdhe2 potential reported in ref .
@xcite .
once we accept the dimerlike - pair model ( @xmath223 ) , we can estimate @xmath128 , the trimer excited - state binding energy , using @xmath7 . with the use of the definitions of the binding wave numbers : @xmath227 where @xmath228 and @xmath31 .
taking @xmath223 , we can then predict @xmath229 which is close to 2.2706 mk by the present three - body calculation using the lm2m2 potential for which we have the ratio @xmath230 . in order to see a deviation of the ratio from 7/4 depending on the realistic potentials in the literature
, we refer to @xmath231 1.59 @xcite ( sapt2 @xcite ) , 1.65 @xcite ( sapt2007 @xcite ) , 1.74 @xcite ( tty @xcite ) , 1.74 ( lm2m2 ) , 2.01 @xcite ( hfdhe2 ) .
the dimerlike - pair model provides a reason why the ratio @xmath232 is located around 7/4 in a narrow region of 1.62.0 .
we note that this model should be considered under the condition that the @xmath0he atoms are interacting with a realistic pair potential and should not be discussed in any situation where a large deformation of the strength is posed to the potential ( cf . a discussion in sec.iii of ref .
@xcite on the efimov states in @xmath0he trimer ) .
we try to apply the same model to the tetramer excited state ( fig .
[ fig : dimer - model]b ) and predict its binding energy @xmath233 .
asymptotically , particle @xmath220 decays from the trimer @xmath234 ) as @xmath235 with @xmath236 where @xmath237 is the reduce atom - trimer mass .
taking @xmath238 , we predict @xmath239 as @xmath240 when employing the calculated values of @xmath241 and @xmath7 with lm2m2 . in sec.iii
, we make a four - body calculation of the tetramer with lm2m2 and check the above prediction of @xmath233 . in this subsection
, we discuss about a technical subject on a numerical trouble which arises when solving the generalized eigenvalue problem ( 2.6 ) .
this is due to the fact that the overlap matrix @xmath242 becomes almost singular when a very large number of nonorthogonal basis functions @xmath243 } employed . in this case , because of the non - negligible round - off error in double - precision computation ( @xmath24416 decimal digits ) , we may obtain no solution of ( 2.6 ) or a solution that includes some unphysically too deep erroneous bound state . in order to overcome this trouble
, we took the following two steps : _ step _ i ) : we first diagonalize the overlap matrix @xmath242 : @xmath245 where @xmath246 .
the eigenvalues @xmath247 are positive definite since @xmath248 .
we then define a new , symmetrized _ orthonormal _ basis set : @xmath249 where @xmath250 .
the generalized eigenvalue problem ( 2.6 ) are then equivalently converted into a standard eigenvalue problem : @xmath251 \ , { \widehat a}_{n ' } = 0,\ ] ] where @xmath252 , and @xmath253 here , we arrange @xmath254 in the decreasing order of @xmath247 : @xmath255 when the nonorthogonality among the basis functions \{@xmath256 } is very large , some of @xmath247 become extremely small and therefore the large factor @xmath257 may cause a serious cancellation in the summation in ( 2.33 ) .
since the present calculation is performed by double - precision computation , such a large cancellation may generate a substantial round - off error in ( 2.33 ) and hence in the matrix elements ( 2.36 ) .
this may give rise to some erroneous eigenstates in ( 2.35 ) that have unphysically huge binding energies .
cccc + @xmath258 & @xmath259 & @xmath127 ( mk ) & @xmath128 ( mk ) + + + 3250 & @xmath260 & @xmath261 & @xmath262 + 3240 & @xmath263 & @xmath264 & @xmath265 + 3200 & @xmath266 & @xmath267 & @xmath268 + 3150 & @xmath269 & @xmath270 & @xmath271 + 2950 & @xmath272 & @xmath273 & @xmath274 + 2750 & @xmath275 & @xmath276 & @xmath277 + 2250 & @xmath278 & @xmath279 & @xmath280 + + _ step _ ii ) : we therefore omit such members of \{@xmath281 } that have too small @xmath247 .
the binding energies of such unphysical states decreases quickly as the basis size is reduced .
finally , we reach an appropriate size , say @xmath282 , of the basis \{@xmath281 } for which those unphysical states have disappeared from the low - energy region , and energies of the lowest - lying ( deepest ) states take physically reasonable values .
it is to be emphasized that the binding energies of so - obtained lowest - lying physical states are stable against further reduction of @xmath282 .
table iv explicitly demonstrates _ step _ ii ) .
we start with @xmath156 basis functions \{@xmath256 } whose parameters are given in table ii . when the size of the new basis \{@xmath281 } is reduced from @xmath283 to @xmath284 according to ( 2.37 ) , the solution of ( 2.35 ) has come to include no unphysical state and give the binging energies @xmath285 mk and @xmath286 mk for the lowest two states . by checking the stability of the energy values against further decreasing @xmath282 , we verify the values of @xmath173 mk and @xmath174 mk in table i.
calculation of the @xmath0he tetramer using realistic potentials has been performed in refs . @xcite .
although binding energy of the ground state obtained in the papers agrees well with each other ( @xmath287 mk ) , that of the loosely bound excited state differs significantly from each other ; namely , the binding energy with respect to the trimer ground state ( 126.4 mk ) is given as 1.1 mk by the faddeev - yakubovski ( fy ) equations method @xcite , 6.6 mk by monte carlo methods combined with the adiabatic hyperspherical approximation @xcite and 52 mk recently by using a method of the correlated potential harmonic basis functions @xcite .
though the faddeev result ( 1.1 mk ) seems to the present authors the most accurate , the excited state was not solved as a bound - state problem in ref . @xcite but the result was extrapolated from the atom - trimer scattering phase shifts .
thus the purpose of this section is to perform , using the same lm2m2 potential as in ref .
@xcite , accurate bound - state calculation of the tetramer excited state , not only giving a precise binding energy but also describing the short - range correlation and the asymptotic behavior of the wave function properly .
the gem has extensively been employed in bound - state calculations of various four - body systems in nuclear and hypernuclear physics ( cf .
review papers @xcite ) .
extension from three - body gem calculations to four - body ones in the presence of strong short - range repulsion is a familiar subject in nuclear physics .
for example , the study of three - nucleon bound states ( @xmath11h and @xmath11he nuclei ) in ref .
@xcite was extended to that of four - nucleon ground state ( @xmath0he nucleus , @xmath288 ) @xcite and the first excited , very diffuse state ( @xmath288 ) @xcite .
the study of the three-@xmath10-particle system ( @xmath9c nucleus ) @xcite was extended to that of the four-@xmath10-particle system ( @xmath289o nucleus ) @xcite with the strongly repulsive pauli - blocking projection operator on the @xmath10-@xmath10 motion .
therefore , extension of the @xmath0he trimer calculation to the tetramer one is straightforward on account of those experiences .
we take two types of jacobi coordinate sets , k - type and h - type ( fig .
[ fig:4bodyjacobi ] ) .
namely , for k - type , @xmath290 , @xmath291 and @xmath292 and cyclically for @xmath293 by the symmetrization between the four particles . for h - type ,
@xmath294 , @xmath295 , @xmath296 and cyclically for @xmath297 .
an explicit illustration of the totally 18 sets of the rearrangement jacobi coordinates of four - body systems is seen in fig .
18 of ref .
@xcite .
the total four - body wave function @xmath298 is to be obtained by solving the schdinger equation @xmath299 with the hamiltonian @xmath300 where @xmath30 , @xmath301 and @xmath302 on the k - type coordinates , and @xmath303 and @xmath304 on the h - type ones .
@xmath298 is expanded in terms of the symmetrized @xmath37-integrable k - type and h - type four - body basis functions : @xmath305 with @xmath306 in which @xmath307 is a function of @xmath28-th set of jacobi coordinates .
it is of importance that @xmath308 and @xmath309 are constructed on the full 18 sets of jacobi coordinates ; this makes the function space of the basis quite wide .
the eigenenergies @xmath41 and amplitudes @xmath310 are determined by the rayleigh - ritz variational principle : @xmath311 where @xmath312 and @xmath313 .
this set of equations results in a generalized eigenvalue problem : @xmath314 \:a_{\alpha_{\rm c'}}^{\rm ( c ' ) } = 0,\ ] ] where @xmath315 and @xmath316 .
the matrix elements are given by @xmath317 up to here is the most general way of variational calculations for bound states of identical spinless four particles .
we describe the basis function @xmath318 in the form @xmath319_\lambda % \otimes y_{l_z}({\widehat { \bf z}_i } ) \big]_{j m } , \nonumber \\
( i=1 , ... , 12)\end{aligned}\ ] ] @xmath320_\lambda % \otimes y_{l_z}({\widehat { \bf z}_i } ) \big]_{j m } , \nonumber \\ ( i=13 , ... , 18)\end{aligned}\ ] ] where @xmath321 specifies a set @xmath322 which is commonly for the components @xmath323 ; and similarly for @xmath324 commonly for @xmath325 .
since we consider the case of @xmath60 in this paper , the totally symmetric four - body wave function requires i ) @xmath326 , @xmath327 and @xmath328 for the k - type basis and ii ) @xmath326 , @xmath329 and @xmath330 for the h - type basis . in ( 3.11 ) and ( 3.12 ) ,
the radial functions are assumed , as in sec.ii , to be @xmath331 with geometric sequences of the gaussian ranges : @xmath332 in ( 3.15 ) , the cos(sin)-type function is not adopted for @xmath66 of the h - type basis though @xmath98 is the distance between two particles .
this is because the cos(sin)-type basis for the @xmath79-coordinate are applied to all the pairs of h - type by the symmetrization of the four particles . in the tetramer calculation ,
the total number , @xmath333 , of the symmetrized four - body basis functions ( 3.4 ) and ( 3.5 ) amounts to @xmath334 30000 , ranging from very compact to very diffuse , to obtain a well converged solution .
since the nonorthogonality among those basis functions is too large to solve directly the generalized eigenvalue problem ( 3.8 ) , we take the same two - step method as described in sec.ii.h in the trimer calculation .
we finally solve the same type of standard eigenvalue problem as ( 2.35 ) using the symmetric orthonormal four - body basis functions , @xmath335 , in which the basis with too small @xmath247 have been omitted . in the calculation of the @xmath0he - tetramer ground state @xmath336 and the excited state @xmath337 ,
the converged result was obtained by employing the symmetric four - body basis functions of @xmath338 with @xmath339 .
table v shows the convergence of the binding energies of the tetramer ground @xmath340 and excited @xmath341 states with respect to increasing @xmath178 .
column k+h is the result with both the k - type and h - type basis functions in ( 3.3 ) , and column (k) is that with the k - type basis only .
contribution from the k - type basis is dominant , but that from the h - type is sizable . without the latter
the excited state does not become bound ( @xmath342 mk ) even for @xmath339 . since both type bases
are not orthogonal to each other , the role of h - type one can be substituted in principle by the k - type one if a very large @xmath178 is employed .
but , this is not practical ; use of both types of bases is essentially important . as seen in table v , convegence of the binding energies with increasing @xmath178
is more rapid in our calculation than that in the fy - equations calculation @xcite .
the reason of this difference in the conversion is the same as that mentioned in the trimer calculation ( sec.ii.e ) .
cccccccc tetramer & & @xmath171 & + & 5 & & 1 & & & & & @xmath343 ( mk ) + & 2 & & 2 & & + @xmath178 & k+h & ( k ) & & k+h & ( k ) & & + 0 & 500.71 & @xmath16 ( 185.96 ) & @xmath344 & @xmath345 & @xmath16 ( @xmath345 ) & & 348.8 + 2 & 558.29 & @xmath16 ( 508.62 ) & & 127.24 & @xmath16 ( @xmath345 ) & & 505.9 + 4 & 558.98 & @xmath16 ( 532.56 ) & & 127.33 & @xmath16 ( @xmath345 ) & & 548.6 + 6 & & & & & & & 556.0 + 8 & & & & & & & 557.7 + [ table : b(4 ) ] [ h ] ccccccccccccccc & & & & & + & 4 & & 4 & & 4 + & @xmath162 & @xmath346 & @xmath164 & @xmath165 & & @xmath167 & @xmath347 & @xmath169 & @xmath170 & & @xmath348 & @xmath349 & @xmath350 & @xmath351 + & & & [ ] & [ ] & & & & [ ] & [ ] & & & & [ ] & [ ] + + k & 0 & 14 & 0.2 & 20.0 & & 0 & 15 & 0.8 & 50.0 & & 0 & 20 & 0.8 & 400.0 + k & 2 & 12 & 0.4 & 20.0 & & 2 & 14 & 0.8 & 40.0 & & 0 & 16 & 0.8 & 300.0 + k & 0 & 8 & 0.3 & 6.0 & & 2 & 8 & 0.8 & 6.0 & & 2 & 8 & 1.0 & 6.0 + k & 2 & 6 & 0.3 & 6.0 & & 0 & 8 & 0.8 & 6.0 & & 2 & 8 & 1.0 & 6.0 + k & 0 & 8 & 0.3 & 6.0 & & 1 & 8 & 0.8 & 6.0 & & 1 & 8 & 1.0 & 6.0 + k & 2 & 6 & 0.4 & 6.0 & & 1 & 6 & 0.8 & 6.0 & & 1 & 8 & 1.0 & 6.0 + k & 2 & 6 & 0.5 & 6.0 & & 2 & 8 & 0.8 & 6.0 & & 2 & 8 & 1.0 & 6.0 + h & 0 & 12 & 0.3 & 20.0 & & 0 & 12 & 0.4 & 16.0 & & 0 & 14 & 0.8 & 25.0 + h & 2 & 6 & 0.6 & 6.0 & & 2 & 6 & 0.8 & 6.0 & & 0 & 8 & 1.0 & 6.0 + h & 0 & 6 & 0.3 & 6.0 & & 2 & 6 & 0.8 & 6.0 & & 2 & 8 & 1.0 & 6.0 + h & 2 & 6 & 0.6 & 6.0 & & 0 & 6 & 0.8 & 6.0 & & 2 & 8 & 1.0 & 6.0 + h & 2 & 6 & 0.6 & 6.0 & & 2 & 6 & 0.8 & 6.0 & & 2 & 8 & 1.0 & 6.0 + + cccc + @xmath258 & @xmath259 & @xmath343 ( mk ) & @xmath233 ( mk ) + + + 24800 & @xmath260 & @xmath352 & @xmath353 + 24790 & @xmath263 & @xmath354 & @xmath353 + 24780 & @xmath355 & @xmath356 & @xmath353 + 24750 & @xmath266 & @xmath357 & @xmath353 + 24700 & @xmath358 & @xmath359 & @xmath360 + 24300 & @xmath361 & @xmath362 & @xmath363 + 23800 & @xmath278 & @xmath364 & @xmath365 + + in table vi we list the nonlinear parameters of the four - body gaussian basis functions , ( 3.11)(3.19 ) , in the case of @xmath366 with 23504 basis functions ( cf .
table v ) to avoid too long listing for @xmath339 .
the range parameters are given in round numbers but further optimization of them do not improve the binding energies as long as we calculate them with five significant figures . when solving the generalized eigenvalue problem ( 3.8 ) , we take the same two - step method as mentioned in sec.ii.h .
table vii shows stability of the binding energes of the ground and excited states @xmath367 against the decreasing number @xmath282 of the symmetrized orthonormal four - body basis functions corresponding to \{@xmath281 } in eq.(2.33 ) .
calculated binding energies and some of mean values of the tetramer ground and excited states are summerized in table viii(a ) in comparison with those obtained by lazauskas and carbonell @xcite with the fy - equations method in which the excited state was not obtained by a direct bound - state calculation but the binding energy ( 127.5 mk ) was extrapolated from the atom - trimer scattering calculations .
our result of @xmath368 mk , which is very closed to @xmath233 in ref.@xcite , confirms the existence of the very shallow bound excited state ( @xmath60 ) of the @xmath0he tetramer .
the tetramer excited state is located only by 0.93 mk below the trimer ground state ( @xmath369 mk ) .
this is analogous to that the trimer excited state lies by 0.967 mk below the dimer ; a reason was explained in sec.ii.g by taking the dimerlike - pair model .
table viii(b ) lists @xmath343 and @xmath233 obtained in other literature papers by the monte carlo methods @xcite and by using the correlated potential harmonic basis @xcite .
all @xmath343 values agree well with the results by the present and fy - equations calculations , but @xmath233 by refs .
@xcite deviate significantly from our and fy - equations results .
use of @xmath107 k@xmath106 @xcite results in @xmath370 mk and @xmath371 mk
. the same perturbative treatment for the small difference of @xmath372 as used in sec.ii.e gives @xmath373 mk and @xmath371 mk .
the calculations below in sec.iii.c take @xmath105 k@xmath186 .
( a ) mean values of the @xmath0he tetramer ground and excited states calculated by the present work and the fy - equations method @xcite with the use of the lm2m2 potential .
@xmath374 is the binding energy , @xmath375 stands for interparticle distance and @xmath376 is the distance of a particle from the center - of - mass of the tetramer .
( b ) the other literature work @xcite on the binding energies ( see text ) .
the values originally given in units of cm@xmath122 are transformed in units of mk in the parentheses . [ cols="^,>,>,>,>,>",options="header " , ] ccccc ( b ) + tetramer & ref .
@xcite & ref .
@xcite & ref .
@xcite & ref .
@xcite + @xmath343 ( cm@xmath122 ) & @xmath16 0.388(1 ) @xmath16 & @xmath16 0.3886(1 ) @xmath16 & @xmath16 0.387(1 ) @xmath16 & @xmath16 0.388 @xmath16 + @xmath378 ( mk ) & ( 558 ) & ( 559.1 ) & ( 557 ) & ( 558 ) + @xmath233 ( cm@xmath122 ) & & & 0.0922 & 0.124 + @xmath378 ( mk ) & & & ( 133 ) & ( 178 ) + definition of the pair correlation function ( 2.23 ) for the trimer is extended to the tetramer states @xmath379 ) : @xmath380 where @xmath381 means the integration over @xmath191 and @xmath382 .
the overlap function between a tetramer state @xmath383 and a trimer one @xmath384 ) is defined as a function of the atom - trimer distance @xmath59 as an extension from ( 2.24 ) : @xmath385 all the integrals in ( 3.20 ) and ( 3.21 ) give the analytical expression owing to the use of gaussian basis functions ; @xmath383 is to be transformed to a function of @xmath386 , and @xmath387 to that of @xmath192 . in fig .
[ fig : tetra - den - short ] , we illustrate the short - range structure of the pair correlation functions @xmath388 of the tetramer ground @xmath49 and excited @xmath198 states together with those of the trimer states .
it is to be emphasized that the same shape of short - range correlation ( @xmath389 ) appears in all the states ( cf .
fig . 3 for the dimer ) without introducing any pair correlation function .
the pair correlation functions @xmath388 of the tetramer @xmath130 take very small values in the strongly - repulsive potential region ( @xmath390 ) in fig .
[ fig : tetra - den - short ] ; relative ratio of the values to the peak value is @xmath391 .
this ratio is to be compared with @xmath392 in the case of the four - nucleon bound state ( @xmath0he nucleus ) calculated with a realistic nucleon - nucleon interaction with a strong short - range repulsion ( the ratio is seen in fig . 1 of ref .
@xcite for the calculated pair correlation function of the @xmath0he nucleus).@xmath171 we understand that so strong is the repulsive core of the atom - atom potential . in fig .
[ fig : tetra - red - short ] , we plot the overlap function @xmath393 , multiplied by @xmath59 , between the tetramer states ( @xmath394 ) and the trimer states ( @xmath395 ) in the region @xmath396 .
the two lines of @xmath397 are to be compared with the result in fig . 4 of ref.@xcite of the fy - equations calculation ; the latter result represents the k - type fy components as a function of atom - trimer distance @xmath59 .
the two kinds of the results are resemble to each other though they do not stand for the same quantity .
as for the excited state , the fy component is derived approximately by modifying the k - type fy amplitude of the zero - energy scattering @xcite ; the resulting amplitude is slightly more enhanced in the inner region than our overlap function of @xmath398 .
this is reflected in the r.m.s distance @xmath138 in table viii(a ) . in the plot of the overlap functions between the trimer
_ excited _ state @xmath399 and the tetramer states , it is reasonably seen that @xmath400 is much smaller than @xmath401 @xmath402 and decreases more rapidly along @xmath59 . in fig .
[ fig : tetra - red - long ] , @xmath403 are illustrated in the asymptotic region .
they should satisfy @xmath404 with @xmath405 ( @xmath406 @xmath122 and @xmath407 @xmath122 ) . the dashed line ( @xmath408 ) and solid line ( @xmath409 ) reproduces the asymptotic functions ( 3.22 ) with the anc @xmath410 @xmath212 and @xmath411 @xmath212
( see the open circles ) , respectively , up to @xmath215 .
asymptotically the tetramer is dissociated into the trimer ground state and a distant atom in the symmetric way between the four atoms ( with negligible amount of the trimer excited state ) .
therefore , the tetramer wave function @xmath412 is represented asymptotically as @xmath413 where the summation over @xmath78 symmetrizes the four atoms ; namely , the @xmath78th atom is isolated at @xmath414 from the trimer @xmath415 in which the @xmath78th atom is absent and the other three atoms are symmetrized . as for the tetramer s anc
, we can make the same comments as those for the trimer s anc below eq.(2.26 ) . in fig .
[ fig : tetra - red - long ] , we note that the exponentially - decaying slope of the solid line ( @xmath416 ) is very close to that in the trimer excited state ( @xmath417 ) and that in the dimer ( @xmath418 ) .
this gives a support to the dimerlike - pair model for the tetramer excited state in the asymptotic region ( fig .
[ fig : dimer - model]b ) .
we can therefore predict that , in the asymptotic region , the pair correlation function @xmath419 should be proportional to @xmath420 .
this is seen in fig .
[ fig : tetra - den - long ] though the solid line ( @xmath50 ) is not so excellently straight as in the trimer excited state ( dot - dashed line ) due to the complexity of the four - body calculation .
as mentioned in sec.ii.g , the dimerlike - pair model predicted @xmath421 mk . in the present calculation , we obtained @xmath422 mk .
the prediction is very good as long as the lm2m2 potential is employed .
generally , in the case of @xmath0he@xmath3 , the excited - state binding energy calculated using realistic potentials may be expected as @xmath423 where the factor multiplied to @xmath7 comes from the ratio of the reduced mass of the dimerlike pair ( @xmath424 ) to that of the @xmath0he@xmath16-@xmath0he@xmath5 system ( @xmath425 ) .
but , as discussed in sec.ii.g , the factor might depend on the realistic potentials with a small deviation ( roughly @xmath426 ) .
we have calculated the ground and excited states of @xmath0he trimer and tetramer using the lm2m2 potential which has a strong short - range repulsive potential and is one of the most widely used @xmath0he-@xmath0he interactions .
we employed the gaussian expansion method ( gem ) for _ ab initio _ variational calculations of few - body systems .
the symmetrized three-(four-)body gaussian basis functions , ranging from very compact to very diffuse , are constructed on the full sets of possible jacobi coordinates .
therefore , the basis set , spanning a wide function space , is suitable for describing both the short - range correlation ( without assuming any pair correlation function ) and the long - range asymptotic behavior of the trimer ( tetramer ) wave function as well as suitable for obtaining accurate binding energies .
the main conclusions are summarized as follows :
\i ) calculated binding energies of the trimer ground and excited states , @xmath127 and @xmath128 , respectively , agree excellently with the literature ( table i ) ; we have @xmath173 mk and @xmath174 mk .
\ii ) as for the binding energies of the tetramer ground and excited states , we obtained @xmath427 mk and @xmath422 mk ( situated only 0.93 mk below the atom - trimer threshold ) .
the former is in good agreement with the literature calculations , while the latter supports the result of 127.5 mk by ref .
@xcite differently from the other literature results ( table viii ) .
\iii ) we found that the strong short - range correlation ( @xmath21 ) seen in the dimer appears also in the ground and excited states of the trimer and tetramer precisely in the same shape ( figs . 3 and 8) .
this gives a foundation to an _ a priori _ assumption that a pair correlation function to simulate the short - range part of the dimer wave function is incorporated in the three-(four-)body wave function from the beginning .
\iv ) illustrating the overlap function between the trimer excited state and the dimer ( @xmath428 ) and that between the tetramer excited state and the trimer ground state ( @xmath429 ) , we found that those overlap functions are almost proportional to the dimer wave function in the asymptotic region up to @xmath2 @xmath16 ( figs . 4 and 10 ) .
also it was found that the pair correlation functions of trimer and tetramer excited states ( @xmath430 and @xmath431 , respectively ) are almost proportional to the squared dimer wave function in the asymptotic region ( figs . 6 and 11 ) .
we then came to propose a dimerlike pair model ( fig .
5 ) that predicts the excited - state binding energy of @xmath0he@xmath3 ( @xmath4 ) using eq . ( 3.24 ) .
it will be of interest to examine this model in the case of @xmath432 .
a five - body calculation of the pentamer , @xmath0he@xmath433 , is in progress .
\v ) we calculated the asymptotic normalization coefficient ( anc ) of the tail function of the relative motion in trimer ( tetramer ) .
this result may be available in peripheral reactions ( insensitive to the interior of the system ) including the @xmath0he trimers and tetramers , in which the reaction cross section will be proportional to the squared anc .
the anc is a quantity to convey the interior structural information to the asymptotic behavior .
therefore , attention to this quantity might be helpful when one tries to reproduce the non - universal variation of the @xmath0he trimer ( tetramer ) states by means of parametrizing models beyond efimov s universal theory .
the numerical calculations were performed on at kek and yifp .
h. kamada , a. nogga , w. glockle , e. hiyama , m. kamimura , k. varga , y. suzuki , m. viviani , a. kievsky , s. rosati , j. carlson , steven c. pieper , r. b. wiringa , p. navratil , b. r. barrett , n. barnea , w. leidemann , and g. orlandini , phys .
c * 64 * , 044001 ( 2001 ) . | we calculated the @xmath0he trimer and tetramer ground and excited states with the lm2m2 potential using our gaussian expansion method ( gem ) for _ ab initio _ variational calculations of few - body systems .
the method has extensively been used for a variety of three- , four- and five - body systems in nuclear physics and exotic atomic / molecular physics .
the trimer ( tetramer ) wave function is expanded in terms of symmetric three-(four-)body gaussian basis functions , ranging from very compact to very diffuse , without assuming any pair correlation function .
calculated results of the trimer ground and excited states are in excellent agreement with the literature .
binding energies of the tetramer ground and excited states are obtained to be 558.98 mk and 127.33 mk ( 0.93 mk below the trimer ground state ) , respectively .
we found that precisely the same shape of the short - range correlation ( @xmath1 ) in the dimer appear in the ground and excited states of trimer and tetramer .
the overlap function between the trimer excited state and the dimer and that between the tetramer excited state and the trimer ground state are almost proportional to the dimer wave function in the asymptotic region ( up to @xmath2 )
. also the pair correlation functions of trimer and tetramer excited states are almost proportional to the squared dimer wave function .
we then come to propose a model which predicts the binding energy of the first excited state of @xmath0he@xmath3 ( @xmath4 ) measured from the @xmath0he@xmath5 ground state to be nearly @xmath6 using the dimer binding energy @xmath7 . |
the physics of suppressed electroweak ( ew ) processes by quantum loops figure as the best window for search into the new physics territory .
more than three decades of the theoretical work strengthens our ability to predict their amplitudes with high precision .
while electroweak and hard qcd calculations rely on the perturbative framework and the renormalization theory , the hadron matrix elements are most accurately evaluated within the lattice qcd .
the extensive search for new physics at the lhc reveals a new boson particle @xcite with a mass around 125 gev .
since the standard model ( sm ) does not contain massive neutrinos , lepton or baryon number violations and a candidate particle for the cold dark matter , it is very unlikely that the new particle is the sm higgs boson .
however , any extension of the sm that has to fulfil above improvements , such as susy or gut extension , implicates the existence of more scalar particles .
it is well known that the higgs mechanism does not solve the problem of particle masses but only breaks the gauge symmetry , i.e. the fermion masses are free parameters .
the alternative explanation of the 125 gev resonance is its identification as the scalar or pseudoscalar meson , which is a mixture of the ( pseudo)scalar glueball and toponium , suggested by p. cea @xcite , thus leaving the problem of the symmetry - breaking mechanism unsolved .
anyhow , the new lhc data with the total and partial decay widths of the new particle , as well as its mass , spin and parity , should resolve this puzzle .
the first lhc results @xcite point to a surpuls of photons in the decay of the 125 gev resonance assumed to be a sm higgs scalar , as well as to the absence of any firm decay signal into fermions . in this paper
, we present complete one loop results for the gauge invariant functions appearing in the suppressed ew processes within the sm and the by theory @xcite , the theory without the higgs scalar but with the lorentz and gauge invariant uv cut - off .
the next chapter is devoted to the explicit deccription of the calculations with details in the appendix , while the concluding section provides numerical estimates and points out the difference between the sm and the by theory .
in this paper , we shall omit a detailed description of the by theory with its mathematical @xcite and phenomenological advantages in particle physics and cosmology @xcite .
the lorentz and gauge invariant cut - off of the by theory is fixed by the wick s theorem and trace anomaly @xcite through weak coupling and the weak gauge boson mass @xmath0 where g = weak coupling , @xmath1=weak gauge boson mass .
the noncontractible space can not essentially affect the strong coupling below the scale @xmath2 @xcite , hence the impact of @xmath3 on the hard qcd corrections to the electroweak processes @xcite is negligible .
thus , we focus onto the dominant one loop electroweak contributions written in the form of the gauge invariant functions @xmath4 @xcite .
these functions appear in the meson mixings @xmath5 and meson decays @xmath6 @xcite . the inami - lim functions @xmath7 and @xmath8 are treated and compared between the sm and by theory in refs . @xcite and @xcite .
we have to reevaluate the self - energy , vertex and box type diagrams for z bosons , photons and gluons in figs .
1 - 4 of ref .
@xcite in both the sm and the by theory to get the explicit forms for the gauge invariant @xmath9 functions .
we list now our results obtained in the t hooft - feynman gauge for the @xmath10 functions @xcite referring to figures 1 - 4 in ref .
@xcite : @xmath11 box diagram as a sum of four graphs of fig .
2 ( a),(b),(c),(d ) of ref .
@xcite : @xmath12 , \end{aligned}\ ] ] @xmath13 is the lepton mass , @xmath14 is the quark mass and @xmath1 the weak boson mass . for the definitions and analytic expressions of @xmath15 and @xmath16 functions ,
see the appendix .
@xmath17 box diagram as a graph of fig .
2 ( a ) of ref .
@xcite : @xmath18 for the definition and analytic expression of the l function , see the appendix .
@xmath19 vertex is given in fig .
3 ( a - h ) of ref .
@xcite and is reevaluated in ref .
@xcite : @xmath20 where @xmath21 and @xmath22 green functions are defined and evaluated in the appendix ( @xmath23 denotes the quark mass and @xmath24 ) .
@xmath25 denotes the contribution to the induced quarks - photon vertex to the second order in the external momentum @xmath26 from graphs of fig .
3 ( a - h ) of the ref . @xcite ( eq .
( b.1 ) ) .
the form factors @xmath27 extracted from the two different lorentz structures , @xmath28 and @xmath29 , must be equal and this is a check on the validity of our calculation : @xmath30 @xmath31 , \\ ( d)&=&-\frac{2}{3}m_{\alpha}^{2}[m_{\alpha}^{2}\tilde{c}'_{0}(0;m_{\alpha},m_{w } ) -2 \tilde{c}'_{3}(0;m_{\alpha},m_{w})-\tilde{c}_{1}(0;m_{\alpha},m_{w } ) -\tilde{c}_{2}(0;m_{\alpha},m_{w } ) ] , \\
( e)&=&m_{w}^{2}[-4 \tilde{c}'_{3}(0;m_{w},m_{\alpha } ) -\frac{1}{2}\tilde{b}'_{0}(0;m_{\alpha},m_{w } ) -2 m_{w}^{2 } \tilde{c}'_{0}(0;m_{w},m_{\alpha } ) \\ &
-&\frac{1}{2 } \tilde{c}_{0}(0;m_{w},m_{\alpha } ) -2 \tilde{c}_{1}(0;m_{w},m_{\alpha } ) ] , \\ ( f+g)&=&2 m_{\alpha}^{2}m_{w}^{2}\tilde{c}'_{0}(0;m_{w},m_{\alpha } ) , \\ ( h)&=&-2 m_{\alpha}^{2 } \tilde{c}'_{3}(0;m_{w},m_{\alpha } ) , \end{aligned}\ ] ] @xmath32 @xmath33 , \hspace{80 mm } \\
\{d\ } & = & \frac{4}{3}m_{\alpha}^{2}[\tilde{c}_{1}(0;m_{\alpha},m_{w } ) + \tilde{c}_{2}(0;m_{\alpha},m_{w } ) ] , \\ \{e\ } & = & m_{w}^{2}[\tilde{c}_{0}(0;m_{w},m_{\alpha } ) + 4 \tilde{c}_{2}(0;m_{w},m_{\alpha } ) + 4 \tilde{c}_{1}(0;m_{w},m_{\alpha } ) ] , \\ \{h\ } & = & m_{\alpha}^{2}[2 \tilde{c}_{2}(0;m_{w},m_{\alpha } ) + 2 \tilde{c}_{1}(0;m_{w},m_{\alpha } ) + \frac{1}{2 } \tilde{c}_{0}(0;m_{w},m_{\alpha } ) ] , \\\end{aligned}\ ] ] the definitions and the explicit forms of the green functions @xmath34 can be found in the appendix .
only ( a - d ) graphs of the fig . 3 of ref .
@xcite contribute to the gauge invariant quarks - gluon vertex in the form factor @xmath35 : @xmath36 @xmath37\},\end{aligned}\ ] ] @xmath38.\end{aligned}\ ] ] knowing @xmath39 functions , we construct the gauge invariant linear combinations @xmath40 @xcite : @xmath41 which appear in the amplitudes of the suppressed electroweak processes @xcite .
the @xmath11 and @xmath17 functions are almost equal for small lepton masses .
numerical evaluations and conclusions are given in the next chapter .
we can accomplish the aim of this paper to verify our sm results with the standard formulas ( see the appendix ) and then to compare all the results with the functions of the by theory containing the lorentz and gauge invariant cut - off @xmath3 .
the following definition is introduced in tables i and ii : @xmath56 , @xmath57 and similarly for other functions because these combinations appear in the amplitudes after applying the unitarity of the quark mixing matrix @xcite .
the functions without superscripts are the standard ones @xcite , and those with @xmath58 superscripts are the sm functions evaluated in this paper , whereas those with @xmath59 superscripts are the functions attributed to the @xmath59 theory with @xmath3 .
one can notice that the functions for the c - quark agree for our sm outputs and the standard functions @xcite ( table i ) , but there are disagreements for the t - quark for @xmath60 and @xmath61 functions ( table ii ) .
we check our sm formulas for @xmath60 and @xmath61 through current conservation , hence these standard functions @xcite are not accurate for the heaviest t - quark .
the vertex contributions for @xmath62 agree perfectly graph by graph with the inami - lim @xcite ( a.1)-(a.3 ) results , but their sum is an approximation @xcite .
as expected , a comparison between the sm ( @xmath63 ) and the by theory ( @xmath64 ) functions reveals a difference of roughly 10 - 30 % , thus opening the possibility to discover the deviations from the sm for the suppressed electroweak processes and meson mixings , provided that the hadron matrix elements can be calculated with a sufficient precision @xcite and the measurements have small systematic and statistical errors @xcite .
the most recent result of the lhcb @xcite , measuring the branching fraction of @xmath65 , appears to be lower than the sm prediction , thus favouring our theory of noncontractible space @xcite .
@xmath66 , \\ c(x)&=&\frac{x}{8}[\frac{x-6}{x-1}+\frac{3x+2}{(x-1)^2}\ln x ] , \\ d(x)&=&-\frac{4}{9}\ln x+\frac{-19x^{3}+25x^{2}}{36(x-1)^{3 } } + \frac{x^{2}(5x^{2}-2x-6)}{18(x-1)^4}\ln x , \\
e(x)&=&-\frac{2}{3}\ln x+\frac{x^{2}(15 - 16x+4x^{2})}{6(1-x)^{4 } } \ln x+\frac{x(18 - 11x - x^{2})}{12(1-x)^{3}}.\end{aligned}\ ] ] @xmath68^{-1}[x+m_{2}^{2}]^{-1}[x+m_{3}^{2}]^{-2 } , \\
f_{1}^{sm;by}(m_{1},m_{2},m_{3 } ) & \equiv & \int^{\infty;\lambda}_{0 } dx x^{2 } [ x+m_{1}^{2}]^{-1}[x+m_{2}^{2}]^{-1}[x+m_{3}^{2}]^{-2 } , \\
l^{sm;by}(m_{1},m_{2 } ) & \equiv & -2 \int^{\infty;\lambda}_{0 } dq q^{3 } [ q^{2}+m_{1}^{2}]^{-2}[q^{2}+m_{2}^{2}]^{-1},\end{aligned}\ ] ] @xmath78^{-1}[q^{2}-m_{1}^{2}]^{-1 } [ ( q+k/2)^{2}-m_{2}^{2}]^{-1 } , \\
\frac{\imath}{16\pi^{2}}k_{\mu}\tilde{c}_{1}(k^{2};m_{1},m_{2})\equiv ( 2\pi)^{-4}\int d^{4}q q_{\mu}[(q+k)^{2}-m_{1}^{2}]^{-1}[q^{2}-m_{1}^{2}]^{-1 } [ ( q+k/2)^{2}-m_{2}^{2}]^{-1 } , \\ \frac{\imath}{16\pi^{2}}[k_{\mu}k_{\nu}\tilde{c}_{2}(k^{2};m_{1},m_{2 } ) + g_{\mu\nu}\tilde{c}_{3}(k^{2};m_{1},m_{2 } ) ] \hspace{60 mm } \\ = ( 2\pi)^{-4}\int d^{4}q q_{\mu}q_{\nu } [ ( q+k)^{2}-m_{1}^{2}]^{-1}[q^{2}-m_{1}^{2}]^{-1 } [ ( q+k/2)^{2}-m_{2}^{2}]^{-1}.\end{aligned}\ ] ] @xmath80 , \\ \tilde{c}_{3}(k^{2};m_{1},m_{2})=\frac{1}{3 } [ \frac{1}{2}\tilde{b}_{0}(k^{2}/4;m_{1},m_{2 } ) -\frac{1}{2}\tilde{b}_{1}(k^{2}/4;m_{1},m_{2 } ) + ( m_{1}^{2}-k^{2}/4 ) \tilde{c}_{0}(k^{2};m_{1},m_{2})].\end{aligned}\ ] ] @xmath85/[24 m_{1}^{2}(m_{1}^{2 } -m_{2}^{2})^{4 } ] , \\ \tilde{c}_{0}^{by}(0;m_{1},m_{2 } ) & = & - [ ( m_{1}^{2}-m_{2}^{2}+ m_{2}^{2 } \ln \frac{m_{2}^{2}}{m_{1}^{2}})/(m_{1}^{2}-m_{2}^{2})^{2 } + ( -m_{1}^{4}+m_{1}^{2}m_{2}^{2}+(\lambda^{2}+m_{1}^{2})m_{2}^{2 } \\ & \times&\ln ( \lambda^{2}+m_{1}^{2})-(\lambda^{2}+m_{1}^{2})m_{2}^{2 } \ln ( \lambda^{2}+m_{2}^{2}))/((\lambda^{2}+m_{1}^{2 } ) ( m_{1}^{2}-m_{2}^{2})^{2})].\end{aligned}\ ] ] | the quantum loop suppressed electroweak processes appear to be very sensitive probes for the symmetry - breaking mechanisms .
since the standard model does not involve massive neutrinos , baryon or lepton number violations and the cold dark matter particle , there are great expectations in search for physics beyond the standard model .
the lhc experiments reported no new physics , except for the appearance of the 125 gev heavy resonance .
if this resonance turns out to be a higgs boson , we shall be faced with the paradox that the standard model , as a theoretically and experimentally incomplete and defective theory , is confirmed .
if , on the other hand , this resonance is a scalar or pseudoscalar meson - a mixture of a ( pseudo)scalar glueball and toponium , one has to study alternative symmetry - breaking mechanisms .
this paper is devoted to the evaluations of the gauge invariant electroweak functions that are relevant for the suppressed electroweak processes within the sm or the uv finite theory in which the noncontractible space breaks the gauge symmetry .
deviations from the sm amplitudes range from 10% to 30% , thus being measurable at the lhcb and future superb factories .
the most recent result of the lhcb , measuring certain b meson branching fraction , appears to be lower than the sm prediction , thus favouring our theory of noncontractible space .
+ pacs numbers : 12.60.-i ; 12.15.-y ; 11.15.ex |
since the groundbreaking work of kenneth wilson @xcite , a proof of confinement for quantum chromodynamics in the continuum limit is lacking .
this is expected to be one of the major goals for our understanding of low - energy behavior of hadronic matter . since then , a lot of fundamental work has been pursued reaching a deeper comprehension of the problem @xcite .
stated in its simplest way , the confinement problem can be expressed through the behavior of the expectation value of the wilson loop that , for a free theory , is just reduced to the computation of a double path integral on the propagator of the theory .
when this expectation takes the simple form @xmath0 , being @xmath1 the area spanned by the loop and @xmath2 a constant called string tension , one has that the _ area law _ holds and the potential rises linearly with the distance granting completely bounded states .
this result has been confirmed by computations on lattice @xcite and is generally stated as `` cornell potential '' being the sum of coulombian and linearly rising terms .
the main problem to face to understand confinement is that the theory at low energies is strongly coupled .
this means that perturbation theory is no more reliable to perform computations .
besides , very few techniques are available to manage the theory in this limit and , generally , one resorts to numerical studies on a lattice .
indeed , recent studies of the propagators of a pure yang - mills theory produced some striking results . these results on huge volumes , arriving to such a significant value as @xmath3 , were presented at the lattice 2007 conference in regensburg @xcite .
the shocking conclusion was that the scenario devised since then @xcite , generally accepted as correct , was not describing the situation seen on the lattice in three and four dimensions : the gluon propagator was reaching a plateau at lower momenta with a finite non - zero value in zero , rather than being zero there , and the ghost propagator was behaving as that of a free massless particle while an enhanced one was expected .
all in all , the running coupling was seen to bend clearly toward zero @xcite without evidence of a non - trivial fixed point as was expected instead .
anyhow , there is not a widespread agreement on the definition of the running coupling in the infrared limit .
the scenario that emerged has an interesting interpretation . assuming that an infrared trivial fixed point for a pure yang - mills theory exists @xcite , the theory can be easily managed through a perturbation expansion .
the propagator we obtain can be used to get several properties of the theory and qcd as well @xcite . in this paper
, we will show that this propagator , by itself , implies a screened potential and so , it needs a next - to - leading order correction to recover a cornell potential and confinement .
it is important to point out that , while yang - mills theory has an infrared trivial fixed point , this is no more true for qcd due to the presence of quarks .
besides , what makes small a running coupling is just the decreasing of momenta that are expected to go to zero like the fourth power . as we will see , the excellent agreement with numerical data , till a perfect coincidence for numerical solution of dyson - schwinger equations @xcite , is strongly supporting this view .
this implies that the definition of the running coupling discussed in these papers is acceptable in the infrared limit and we are exploiting this possibility .
finally , we emphasize that a decoupling solution for the gluon propagator does not grant a potential behaving as the cornell potential with a linearly rising term but rather a screened potential . this has been recently shown , numerically solving dyson - schwinger equations , by gonzalez , vento and mathieu @xcite . as we will show in this paper
, the reason for this is that we need loop corrections to the propagator in the infrared limit . in this way , a quartic term in the momenta
is seen to appear , in agreement with a gribov - like propagator , providing the needed linearly rising term in the potential .
the paper is structured as follows . in sec.[sec2 ] , we present the formalism applied to scalar field and yang - mills theories . in sec.[sec3 ] , we show a comparison with numerical data strongly supporting the formalism . in sec.[sec4 ] , we give the next - to - leading order correction to the propagator . in sec.[sec5 ] , we discuss the renormalized gluon propagator we obtained . in sec.[sec6 ] , we compute the interquark potential and the string tension .
finally , in sec.[sec7 ] , we present our conclusions .
in the eighties , carl bender and others @xcite proposed a new approach to cope with a strong coupled quantum field theory .
this pioneering view was too radical and so , produced too singular results to be useful .
but one can improve on it as we showed recently @xcite , assuming a bare coupling formally going to infinity .
as we will show , immediate conclusions can be obtained both for scalar and yang - mills quantum field theories .
we consider a massless neutral scalar field with the following generating functional @xmath4=n\int [ d\phi]\exp\left\{i\int d^4x\left[\frac{1}{2}(\partial\phi)^2-\frac{\lambda}{4}\phi^4+j\phi\right]\right\}.\ ] ] in order to get a strong coupling expansion , we formally rescale space - time coordinates as @xmath5 . then , the functional takes the form @xmath4=n\int [ d\phi]\exp\left\{\frac{i}{\lambda}\int d^4x\left[\frac{1}{2}(\partial\phi)^2-\frac{1}{4}\phi^4+\frac{1}{\lambda}j\phi\right]\right\}.\ ] ] but @xmath6 is an arbitrary function and so we rescale it as @xmath7 .
now , a perturbative expansion can be written down as @xmath8 that yields the following terms into the action @xmath9 \label{eq : ord0 } \\
s_1&=&\int d^4x\left[\partial\phi_0\partial\phi_1-\phi_0 ^ 3\phi_1+j\phi_1\right ] \label{eq : ord1 } \\
s_2&=&\int d^4x\left[\frac{1}{2}(\partial\phi_1)^2-\frac{3}{2}\phi_0 ^ 2\phi_1 ^ 2 + \partial\phi_0\partial\phi_2-\phi_0 ^ 3\phi_2+j\phi_2\right ] .
\label{eq : ord2}\end{aligned}\ ] ] we note that this expansion requires that we are able to solve the classical equation of motion @xmath10 in the limit @xmath11 . indeed , making use of this equation the following functional
is obtained @xmath4\approx e^{i\int d^4x\left[\frac{1}{2}(\partial\phi_0)^2-\frac{\lambda}{4}\phi_0 ^ 4+j\phi_0\right ] } \int[d\phi_1 ] e^{i\frac{1}{\lambda}\int d^4x\left[\frac{1}{2}(\partial\phi_1)^2-\frac{3}{2}\lambda\phi_0 ^ 2\phi_1 ^ 2\right]}.\ ] ] here @xmath12 is a functional of the current @xmath6 . in this form
, we are not able to manage this expression .
so , we make the ansatz @xcite @xmath13 and we need to solve the following equation for the green function @xmath14 ^ 3=\frac{1}{\mu}\delta^4(x - x').\ ] ] here @xmath15 is an arbitrary constant with the dimension of energy arising in the solution of the classical theory .
then , using an iterative procedure , the next - to - leading order correction is @xmath16 ^ 3j(x'')\right .
\nonumber \\ & -&\left.\mu^3\left[\int d^4x''\delta(x'-x'')j(x'')\right]^3\right\}+\ldots.\end{aligned}\ ] ] higher order corrections can be similarly obtained .
finally , the propagator can be obtained explicitly @xcite .
one has @xmath17 being @xmath18 with @xmath19 and a mass spectrum @xmath20 we recognize that the theory develops a mass gap due to the finiteness of the coupling .
this result would be very difficult to obtain with a weak coupling expansion .
but , another striking result is that this theory is infrared trivial as the generating functional in this limit is just gaussian @xmath21=n\exp\left[\frac{i}{2}\int d^4xd^4yj(x)\delta(x - y)j(y)\right].\ ] ] this can be also evinced from the form of the propagator ( [ eq : prop ] ) that presents just simple poles and can be ascribed to a free theory of massive particles with a superimposed spectrum of a harmonic oscillator .
finally , the next to leading order term takes the form @xmath22\approx z_0[j]\int [ d\phi]\exp\left\{\frac{i}{\lambda}\int d^4x\left[\frac{1}{2}(\partial\phi)^2-\frac{3}{2}\lambda\left(\int d^4x_1\delta(x - x_1)j(x_1)\right)\phi^2\right]\right\}.\ ] ] the foundations for a possible understanding of qcd at low - energies were posed on the eighties @xcite .
people realized that , with a current expansion , a sensible low - energy limit was obtained .
but not a great step forward was possible due to the stumbling block of an unknown gluon propagator .
this was a major obstacle to complete this program .
we will now see how a consistent scenarion can indeed be built on this basis .
we have the generating functional @xmath23&=&n\exp\left\{-i\int d^4x\left[\frac{1}{4}{\rm tr}f^2+\frac{1}{2\xi}(\partial\cdot a)^2 + ( \bar c^a\partial_\mu\partial^\mu c^a+g\bar c^a f^{abc}\partial_\mu a^{b\mu}c^c)\right]\right\}\times \nonumber \\ & & \exp\left[i\int d^4x\left(j^a_\mu a^{\mu a}+\bar\epsilon^a c^a+\bar c^a\epsilon^a\right)\right]\end{aligned}\ ] ] being @xmath24 the field strength and @xmath25 the vector potential .
exactly in the same way as for the scalar field , we operate a rescaling of space - time variable through the t hooft coupling @xmath26 for a su(n ) gauge group .
so , again , we have to solve the classical equations of motion @xmath27 for an arbitrary gauge fixed through the parameter @xmath28 .
now , we would like to study quantum field theory for yang - mills equations assuming that a trivial infrared fixed point exists . in order to get such a theory
, we have to show that a set of classical solutions exist to build up the corresponding quantum field theory .
indeed , this is the case and such solutions are instantons @xcite .
this can be proved quite easily by putting @xmath29 being @xmath30 some constant coefficients .
then , yang - mills equations collapse to @xmath31 in the lorenz ( landau ) gauge , this equation is further simplified and we can use the fact that @xmath32 and so we set @xmath33 . in this case , the theory reduces to that of the scalar field and we can easily do quantum field theory for a trivial infrared fixed point .
quantum field theory will be build up on the instanton field @xmath34 that is replicated for each component of the yang - mills field .
now , we are able to give explicitly the gluon propagator in the landau gauge as @xmath35 with @xmath36 given by eq.([eq : prop ] ) . here @xmath37 and we interchange the constants @xmath15 and @xmath38 .
so , we can write the spectrum for su(n ) as @xmath39 this can be proven very easily from the definition of the two - point function . we have @xmath40 using our instanton solutions , one sees that @xmath41 giving our result . a possible choice for components of the gauge field is @xmath42 being @xmath43 the minkowski metric .
we are now able to develop a quantum field theory for the yang - mills field supporting a trivial infrared fixed point . for the sake of completeness ,
we give here the ghost propagator .
ghost field just decouples in the infrared limit due to the mapping theorem and so , @xmath44 then , the generating functional for the gauge field in the infrared limit takes the following gaussian form @xmath45&=&n'\exp\left[\frac{i}{2}\int d^4xd^4y j^{\mu a}(x)d_{\mu\nu}^{ab}(x - y)j^{\nu b}(y)\right ] \nonumber \\ & + & o\left(\frac{1}{\sqrt{n}g}\right)\end{aligned}\ ] ] entailing triviality .
the spectrum has a mass gap with a superimposed spectrum of a harmonic oscillator as if such particles would have a structure .
we note that infrared triviality does not apply to qcd due to the presence of quarks .
a priori , it is not clear why our classical solutions should be selected to build up a quantum field theory . even if our approach is consistent , we can not be sure that something should enter , and it is instead neglected , into the solutions needed for the theory .
so , we need to compare our conclusions with numerical data from lattice and numerical dyson - schwinger equations .
these studies have been pursued for some decades and reached their maturity with a significant breakthrough on 2007 where the gluon propagator was seen to reach a finite non - zero value lowering momenta @xcite .
this was also confirmed by numerical studies of dyson - schwinger equations @xcite .
firstly , the most important point to be assessed is the existence of the trivial fixed point at infrared ( momenta going to zero ) .
as already said , this can be seen from recent lattice computations @xcite and numerical solution of dyson - schwinger equations @xcite provided one agrees on the definition of the running coupling at lower momenta . assuming this , we will show below the behavior of the running coupling for the latter case ( see fig . [
fig : rc ] ) .
we note also that , given a different definition of running coupling , a similar result has been given in @xcite .
[ fig : rc ] , @xmath46 and @xmath47.,title="fig : " ] the definition of the running coupling is the one considered in @xcite and proposed in @xcite .
it is given by @xmath48 having introduced the dressing functions for the gluon and ghost propagators respectively as @xmath49 and @xmath50 . in our fit from numerical solution to dyson - schwinger equations , it is seen to be very near the zero value as momenta goes to zero .
this result is identical to the one presented in @xcite .
we note that , with this definition , the running coupling goes to zero like @xmath51 while , with the definition given in @xcite , it runs to zero like @xmath52 .
below we give the comparison with numerical results for the propagators .
data for @xmath53 and dyson - schwinger equations were given directly from measurement while measurements at @xmath54 recovered from fig . 2 in @xcite .
one sees that , increasing volumes , our propagator tends to the measured one and the agreement improves neatly in the deep infrared .
different mass gaps are due to different values of @xmath55 . for numerical dyson - schwinger equations
@xcite , one has no volume problem being already in the infinite limit and , indeed , we get perfect coincidence of our propagator with the measured one .
note that we consider a weak dependence on the gauge group as showed in @xcite that is fully consistent with our discussion above . with a mass gap of @xmath56 with a mass gap of @xmath57 ]
these numerical results give a strong support to our scenario and to the conclusion of the existence of a trivial infrared fixed point for yang - mills theory .
in sec . [ sec : ym ] , we have shown that , for a yang - mills theory with a trivial infrared fixed point , it is enough to study the corresponding quantum theory of the instanton solutions represented by a massless scalar field with a quartic self - interaction . for our aims , we need just to evaluate the next - to - leading order correction to the propagator and derive the potential from it as shown in @xcite .
this will move us away from the infrared fixed point and will provide a comprehension on the theory approaching such a fixed point .
so , from eq.([eq : delta ] ) we have to evaluate @xmath58 ^ 3\ ] ] where @xmath59 is the one given in eq.([eq : prop ] ) .
we point out here that , for a yang - mills theory , we have to set @xmath60 .
turning to momentum space this gives @xmath61 this integral is well - known in quantum field theory , it arises as a two loop sunrise diagram @xcite and , in the form given above , has not a closed form value .
but , we are in the infrared limit and we need its value just in the limit of small momenta . but before we take small momenta limit , consistently with our approximation technique , we need to take the limit @xmath11 .
this will give immediately @xmath62 \nonumber \\ & + & o\left(\lambda^{-\frac{5}{2}}\right),\end{aligned}\ ] ] where we have set , using eq.([eq : ms ] ) , @xmath63 .
these integrals are really singular but we note that we can not extend the integration range to infinity as we expect our computation to hold just for momenta @xmath64 , the cut - off of the theory that emerged by integration of the classical equations .
so , the theory has a natural cut - off . in this way we can immediately evaluate the correction with the following integrals : @xmath65 so
, we finally get @xmath66+o\left(\lambda^{-\frac{3}{2}}\right).\ ] ] we note that the cut - off disappears everywhere except as a scale for the momenta contribution as it should .
now , we can interpret this as an expansion in inverse of @xmath67 of the denominator of the propagator and this means that we have to evaluate @xmath68&= & \nonumber \\
p^2\left(1-\frac{1}{\lambda^\frac{1}{2}}\frac{27}{\pi^8 } + \frac{1}{\lambda}\frac{3.3\cdot 48}{\pi^8}\right ) + \frac{1}{\lambda}\frac{3.3\cdot 9}{\pi^8}\frac{p^4}{\mu^2}-m_n^2\left[1-\frac{1}{\lambda^\frac{1}{2}}\frac{27}{\pi^8 } + \frac{1}{\lambda}\frac{3.3\cdot 48}{\pi^8 } \left(1+\frac{3}{16}\frac{p^2}{\mu^2}\right)\right].\end{aligned}\ ] ] we get a mass renormalization factor causing the appearance of a @xmath52 term in agreement with gribov expectations and , finally , looking at the mass term we see that @xmath69=(2n+1)^2m_0 ^ 2\lambda^\frac{1}{2}\left[1-\frac{1}{\lambda^\frac{1}{2}}\frac{27}{\pi^8 } + \frac{1}{\lambda}\frac{3.3\cdot 48}{\pi^8 } \left(1+\frac{3}{16}\frac{p^2}{\mu^2}\right)\right]\ ] ] and we get a renormalized running coupling @xmath70.\ ] ] from this we get the renormalization constant for the field @xmath71
once we obtained the field renormalization constant @xmath72 , we can provide a next - to - leading order formula for the gluon propagator .
let us introduce the constant @xmath73 and @xmath60 .
we can rewrite the gluon propagator in the landau gauge in the following form @xmath74 where we have set @xmath75.\ ] ] we note that the coefficients @xmath76 are exponentially damped with @xmath77 and so , just the first few terms give the physics of infrared yang - mills theory .
this means that our propagator agrees fairly well with the one obtained in the refined gribov - zwanziger scenario @xcite .
so , we can attempt to evaluate the condensates after the correction is introduced .
as already done in @xcite , we consider just two terms in our propagator and write down @xmath78 in order to estimate the condensates , we note that @xmath79 obtained from the exact relation @xmath80 accounting for the exponential damping of higher coefficients . now , by the definition given in @xcite of the condensate , we get @xmath81.\ ] ] in our previous work @xcite we get about @xmath82 for this condensate and our result here is not so much different from it .
again , this is a proof of existence of such a condensate and our scenario agrees fairly well with that given in @xcite .
in order to compute the potential in a pure yang - mills theory at the infrared fixed point , we have to evaluate @xmath83[d\bar c][dc ] e^{-\frac{i}{4}\int d^4x{\rm tr}f^2+is_g[\bar c , c ] } { \rm tr}{\cal p}e^{ig\oint_{\cal c } dx^\mu t^aa^a_\mu(x)}}{\int [ da][d\bar c][dc ] e^{-\frac{i}{4}\int d^4x{\rm tr}f^2+is_g[\bar c , c]}}\ ] ] being @xmath84 $ ] the contribution of the ghost field , @xmath85 the anti - hermitian generators of the gauge group and @xmath86 the path ordering operator . in our case , in the infrared limit , we have a trivial fixed point .
this implies that our generating functional takes a gaussian form @xmath21=\exp\left[\frac{i}{2}\int d^4x'd^4x''j^{a\mu}(x')d_{\mu\nu}^{ab}(x'-x'')j^{b\nu}(x'')\right]\ ] ] and so , the evaluation of the wilson loop is straightforwardly obtained as @xcite @xmath87=\exp\left[-\frac{g^2}{2}c_2(r)\oint_{\cal c}dx^\mu\oint_{\cal c}dy^\nu d_{\mu\nu}(x - y)\right]\ ] ] being @xmath88 the quadratic casimir operator that for su(n ) in the fundamental representation , @xmath89 , is @xmath90 . in our case , the propagator has the form given in sec.[sec2 ] for the landau gauge .
the fall - off to large distances of this propagator grants that ordinary arguments to evaluate the integrals on the path apply . indeed , using fourier
transform one has @xmath87=\exp\left[-\frac{g^2}{2}c_2(r)\int\frac{d^4p}{(2\pi)^4}\delta(p^2)\left(\eta_{\mu\nu}-\frac{p_\mu
p_\nu}{p^2}\right)\oint_{\cal c}dx^\mu\oint_{\cal c}dy^\nu e^{-ip(x - y)}\right].\ ] ] we need to evaluate @xmath91 provided the contributions coming from taking into account the term @xmath92 run faster to zero at large distances .
this must be so also in view of the gauge invariance of wilson loop . with the choice of time component in the loop going to infinity while distance is kept finite
, we can evaluate the above integral in the form @xmath93 and we are left with @xmath87\approx \exp\left[-t\frac{g^2}{2}c_2(r)\int\frac{d^3p}{(2\pi)^3}\delta({\bm p},0)e^{-i{\bm p}\cdot{\bm x}}\right]\ ] ] this yields @xmath87=\exp\left[-tv_{ym}(r)\right]\ ] ] being @xmath94 this equation should change if we consider a running coupling but , for the moment , we will not do this as does not change too much our conclusions .
so , in our case we have to evaluate the integral @xcite @xmath95 using eq.([eq : dg ] ) .
if we neglect the contribution of the sunrise diagram we see that the propagator is a sum of yukawa propagators .
this means that the above integral will yield a sum of yukawa potentials and we get an overall screened potential that does not grant confinement as discussed in @xcite .
so , we see that in the deep infrared , one - gluon exchange does not grant a confining theory .
but , accounting also for our two - loop correction , we will have to compute instead @xmath96 we can now evaluate this integral by rewriting it in the form @xmath97 where we have considered the limit @xmath11 , @xmath98 even after a renormalization of mass @xmath99 is taken , @xmath80 and we have set @xmath100 .
this integral can be computed exactly to give @xmath101 that gives at small values of @xmath102 @xmath103.\ ] ] so , remembering that @xmath104 , being @xmath105 for su(n ) in the fundamental representation , we get a linear rising potential with @xmath106 as expected .
we can rewrite this as @xmath107 this result should be compared to the @xmath108 case @xcite that has @xmath109 the agreement is strikingly good in its functional form .
we have shown that , in the framework of our formalism , confinement is achieved for qcd satisfying the area law from the expectation value of the wilson loop once next - to - leading order correction is accounted for .
we started from the hypothesis of the existence of a trivial infrared fixed point for a yang - mills theory and this drove us to the fundamental conclusion that the scenario generally depicted by wilson and gribov is essentially correct .
as an aside , it interesting to note as our approach recovers in some limit the refined gribov - zwanziger scenario confirming the existence of a @xmath110 condensate .
this condensate is relevant for the existence of a mass gap in the theory .
the overall agreement with lattice data and numerical studies of dyson - schwinger equations gives a serious hint that this view can represent an important track toward a full understanding of quantum field theory in the infrared limit .
i would like to thank marco ruggieri for useful comments and discussions and for providing code for numerical solving dyson - schwinger equations .
i also thank orlando oliveira for sharing his numerical data with me .
this problem was pointed out to me by owe philipsen at bari conference sm&ft 2011 .
i would like to thank him for this .
g. s. bali , phys .
rept . * 343 * , 1 - 136 ( 2001 ) .
i. l. bogolubsky , e. m. ilgenfritz , m. muller - preussker , a. sternbeck , pos * lat2007 * , 290 ( 2007 ) .
a. cucchieri , t. mendes , pos * lat2007 * , 297 ( 2007 ) .
o. oliveira , p. j. silva , e. m. ilgenfritz , a. sternbeck , pos * lat2007 * , 323 ( 2007 ) .
l. von smekal , r. alkofer , a. hauck , phys .
lett . * 79 * , 3591 - 3594 ( 1997 ) .
l. von smekal , a. hauck , r. alkofer , annals phys . * 267 * , 1 ( 1998 ) .
d. zwanziger , nucl .
* b364 * , 127 - 161 ( 1991 ) .
i. l. bogolubsky , e. m. ilgenfritz , m. muller - preussker , a. sternbeck , phys .
* b676 * , 69 - 73 ( 2009 ) . m. frasca , phys .
b * 670 * , 73 ( 2008 ) m. frasca , mod .
a * 24 * ( 2009 ) 2425 m. frasca , arxiv:0802.1183 [ hep - th ] ( unpublished ) .
m. frasca , pos * facesqcd * , 039 ( 2011 ) .
m. frasca , phys .
c * 84 * , 055208 ( 2011 ) .
a. c. aguilar , a. a. natale , jhep * 0408 * , 057 ( 2004 ) .
a. c. aguilar , d. binosi , j. papavassiliou , phys .
rev . * d78 * , 025010 ( 2008 ) .
p. gonzalez , v. mathieu and v. vento , phys .
d * 84 * , 114008 ( 2011 ) . c. m. bender , f. cooper , g. s. guralnik and d. h. sharp , phys .
d * 19 * , 1865 ( 1979 ) . c. m. bender , f. cooper , g. s. guralnik , r. roskies and d. h. sharp , phys .
* 43 * , 537 ( 1979 ) .
m. frasca , phys . rev .
d * 73 * , 027701 ( 2006 ) [ erratum - ibid .
d * 73 * , 049902 ( 2006 ) ] m. frasca , mod
. phys .
a * 22 * , 1293 ( 2007 ) m. frasca , int .
j. mod .
a * 23 * , 299 ( 2008 ) m. frasca , int .
j. mod .
a * 22 * , 2433 ( 2007 ) j. t. goldman , r. w. haymaker , phys .
* d24 * , 724 ( 1981 ) . t. schafer and e. v. shuryak , rev .
70 , 323 ( 1998 ) .
p. boucaud , f. de soto , a. le yaouanc , j. p. leroy , j. micheli , h. moutarde , o. pene , j. rodriguez - quintero , jhep 0304 , 005 ( 2003 ) .
r. alkofer and l. von smekal , phys .
rept . * 353 * , 281 ( 2001 ) .
a. maas , jhep * 1102 * , 076 ( 2011 ) . m. caffo , h. czyz , s. laporta and e. remiddi , nuovo cim . a 111 , 365 ( 1998 ) d. dudal , h. verschelde , j. a. gracey , v. e. r. lemes , m. s. sarandy , r. f. sobreiro and s. p. sorella , jhep * 0401 * , 044 ( 2004 ) .
d. dudal , j. a. gracey , s. p. sorella , n. vandersickel and h. verschelde , phys . rev .
d * 78 * , 065047 ( 2008 ) .
d. dudal , s. p. sorella and n. vandersickel , phys .
d * 84 * , 065039 ( 2011 ) .
m. frasca , arxiv:1202.4105 [ hep - ph ] .
d. dudal , o. oliveira and n. vandersickel , phys .
d * 81 * , 074505 ( 2010 ) .
a. cucchieri , d. dudal , t. mendes and n. vandersickel , arxiv:1111.2327 [ hep - lat ] .
m. pak , h. reinhardt , phys .
rev . * d80 * , 125022 ( 2009 ) .
d. karabali , v. p. nair and a. yelnikov , nucl .
b * 824 * , 387 ( 2010 ) . | we analyze the wilson loop for a pure yang - mills theory , using a decoupling solution in close agreement with lattice computations . at one - gluon exchange level
it is seen that the potential can not yield a linear rising contribution as expected for a confining theory . next - to - leading order correction gives rise to a quartic term for momenta in the gluon propagator that , in agreement with gribov s view , yields a linear confining term .
this correction is due to a two - loop or sunrise integral that we need to evaluate in the low - momenta limit . in the infrared regime , the physical consistency of the theory
is determined by a natural cut - off , arising from the integration of the classical equations of the theory , fixing in this way the regularization scheme . |
liquid water behaviour as a function of temperature and pressure differs from that of most liquids . as a major example
the presence of a line of maximum density ( tmd line ) in the p - t phase diagram is most significant , as it divides the entire p - t phase diagram into two regions with remarkably different properties : the coefficient of thermal expansion is negative on the low - temperature side of the tmd line , while it is positive on the high - temperature side .
similarly , several other thermodynamic and dynamical quantities show peculiar properties .
for instance the isothermal compressibility and heat capacity exhibit a minimum as a function of temperature and show an anomalous behavior more pronounced in the supercooled liquid state @xcite . when decreasing temperature in fact the coefficient of thermal expansion , the isothermal compressibility and the constant - pressure specific heat increase rapidly .
these quantities appear to diverge at a temperature @xmath1 if they are extrapolated below the lowest temperatures at which they are measurable ( @xmath2 c at @xmath3 @xmath4 ) @xcite .
the supercooled region below @xmath2 c is experimentally unreachable due to the strong tendency of water to crystallize .
it is nonetheless believed that nucleation in supercooled water might be due to the presence of impurities that drive the liquid toward the more stable phase @xcite .
therefore the experimentally unreachable zone of supercooling is believed to be physically significant .
measurements of the rate of evaporation on amorphous water have in fact proved that the amorph can be connected with normal liquid water by a reversible thermodynamic path at atmospheric pressure @xcite .
besides at ambient pressure experiments have proven the existence of supercooled liquid water close to the glass transition temperature @xcite . in spite of all the interest driven by these anomalies a coherent theory of the thermodynamic and transport properties of supercooled water
does not yet exist , also due to the difficulties encountered in experiments .
different thermodynamic scenarios have been proposed through the years for the peculiar metastable behaviour of water @xcite . among these
three of them have received a great attention in literature . the stability limit conjecture ( slc )
is the first scenario proposed .
it attributes the anomalies of water to the presence of a continuous retracing spinodal curve , bounding the superheated and supercooled states .
the spinodal , which represents the limit of mechanical stability of a liquid , is hypothized to retrace to higher pressure values below a temperature at which it intersects the locus of the tmd @xcite .
the second critical point scenario ( scp ) , based on extrapolation of simulated data , ascribes the anomalous properties of water to the presence of a metastable , low temperature liquid - liquid critical point , associated with a phase transition between a low - density and a high density liquid phase @xcite .
experimental evidences of the existence of two liquid phases compatible with a second critical point have been presented @xcite .
the singularity free scenario ( sf ) explains the behaviour of supercooled water with the presence of anomalous fluctuations due to the hydrogen bonds and no underlying singularity is invoked @xcite .
the sf assumes that upon isobaric cooling the thermodynamic response functions go through a maximum but remain finite .
an anomalous increase in isothermal compressibility , heat capacity , and thermal expansion is explained by the existence of the density maxima locus , which is negatively sloped in the @xmath5 plane @xcite . according to this scenario , no phase transition or critical point occurs at low temperatures . a fundamental role in the clarification of the scenario of supercooled water
is played by the study of the behaviour of the spinodal line , and its relation with the line of tmd . due to the limitations of the experiments in the supercooled realm , in order to assess the different hypothesis numerical studies have become of uttermost importance .
many computer simulation studies indeed have been performed with several different water site models , st2,tip4p , spc / e and tip5p @xcite .
all the above models do not take into account explicitly the polarizability of the water molecules .
polarizable models for water have been also developed @xcite to give a more realistic description of the behaviour of the system in different thermodynamic conditions .
depending on the zone of the phase diagram investigated they have proven to be equally or more realistic to describe the features of water .
studies of the behaviour of the spinodal line for polarizable water model potential have been never carried out .
it is therefore of interest to study the behaviour of the spinodal line for potentials where polarizability is explicitly taken into account .
we present here a molecular dynamics , md , study of the spinodal and the tmd line of supercooled water performed with the polarizable potential , introduced in the literature by ruocco and sampoli @xcite and parametrized by brodholt , sampoli and vallauri @xcite , the bsv potential .
we found that the bsv model is appropriate for our study since it reproduces the thermodynamical properties of water in the region of the gas - liquid coexistence better than other polarizable models over a broad temperature range @xcite .
it has also been shown that within the bsv model the site - site radial distribution functions are in good agreement with the experimental data in a broader range of thermodynamical conditions .
@xcite in the next section we describe the computer simulation in details .
after presenting in the third section the structural properties of the system upon supercooling , the behaviour of the spinodal and of the tmd lines are reported in the fourth section .
last section is devoted to conclusions .
we performed md computer simulations of water with the model 4 of ref .
an induced polarizable dipole moment @xmath6 , located in the center - of - mass of the molecule , describes the effect of the electric field of the environment on the molecules .
the induced dipole @xmath7 is calculated from the local electric field @xmath8 with an iterative procedure by assuming an isotropic polarizability .
the value of this polarizability is fixed to the single molecule value @xmath9 @xmath10 @xcite .
each simulation is conducted in the @xmath11 ensemble with @xmath12 water molecules enclosed in a cubic box with periodic boundary conditions .
the simulations have been performed with the minimum image convention and a cut - off of the interactions at half of the box length .
the long range part of the electrostatic interactions is taken into account with the reaction field method .
details of the extension of the reaction field method to include polarization effects can be found in references @xcite .
the time step @xmath13 for the integration of the molecular trajectories is fixed at @xmath14 @xmath15 . to analyse the phase diagram and the spinodal curve inside the metastable supercooled region we carried out isothermal simulations for eight different temperatures : @xmath16 , @xmath17 , @xmath18 , @xmath19 , @xmath20 , @xmath21 , @xmath22 and @xmath23 k. for each eight different densities are simulated , namely @xmath24 , @xmath25 , @xmath26 , @xmath27 , @xmath28 , @xmath29 , @xmath30 and @xmath31 @xmath32 .
the box length @xmath33 spans from @xmath34 to @xmath35 to cover the range of investigated densities . during equilibration the temperature is controlled via a velocity rescaling procedure .
production runs are performed in the microcanonical ensemble . for the temperatures and the densities closer to the spinodal line the longest equilibration time required was of @xmath36 @xmath37 . in tables
[ table1 ] and [ table2 ] we report the temperatures , densities , pressures and internal energies of the simulations we have performed .
structural quantities have been calculated averaging from a minimum of @xmath38 configurations at @xmath39 k to a maximum of @xmath40 configurations at @xmath41 k , all equally spaced and taken every @xmath42 .
we have calculated the site - site radial distribution functions ( rdf ) @xmath43 to monitor the internal structure at every temperature .
we report @xmath44 , @xmath45 and @xmath46 for @xmath39 k in fig.[fig:1 ] and @xmath44 for @xmath41 k in fig.[fig:1bis ] calculated at several densities . at @xmath39 k the peak positions agree with the experimentally measured structure of water @xcite . upon decreasing temperature
the peaks remain at the same position and become sharper . by varying the densities
we do not observe very large changes in the @xmath43 apart for an increase in the height of the peaks , even when the system approaches the spinodal line ( see next section ) .
this is an important test to verify the absence of disomogeneities or cavitation phenomena close to the limit of mechanical stability .
a further check has been performed with the calculations of the local density fluctuations . for this sake
we have divided our simulation box into @xmath47 sub - boxes and calculated the average distribution functions of the local density ( not shown ) . in all cases
we do not find deviations from a gaussian distribution centered on the fixed density of the system .
a deeper insight in the local ordering of the atoms can be achieved by calculating the coordination numbers . in the upper inset of fig .
[ fig:1bis ] we show @xmath48 , the coordination number of the oxygens , as a function of @xmath49 : @xmath50 where @xmath51 is the value of the interatomic distance at which the first minimum in @xmath44 is located .
it is important to observe the different behaviour of @xmath48 at the two temperatures @xmath39 k and @xmath41 k on lowering the density . at @xmath39 k for the highest density @xmath48 results to be @xmath52 .
this value is the signature of an open network of water , where also interstitial molecules are present @xcite . @xmath48 then decreases until it reaches the limiting value @xmath53 characteristic of the local tetrahedral order . at the lower temperature @xmath41 k @xmath48
is found to be quite constant around the value @xmath53 .
these findings indicate that the open network of @xmath54-bonds becomes better defined as @xmath49 and/or @xmath55 decrease . in connection with the tendency of the @xmath54-bond network to become more ordered it
is expected a sharpening of peaks of the rdf .
a quantitative measure of this sharpening can be obtained looking at the value of the rdf at its first minimum , @xmath56 . as a consequence
we also expect to observe a decrease of @xmath56 as the coordination shells become better defined . in the lower inset of fig .
[ fig:1bis ] we observe instead the presence of a minimum in the behaviour of @xmath56 as function of the density .
the minimum is more evident at @xmath41 k. this effect evidences an important change in the thermodynamical properties of the system , since it is related to the behaviour of the entropy in the vicinity of the spinodal , as previously discussed in literature @xcite .
therefore we can consider that the change of slope in the @xmath56 curve is a signature of the approaching of the system to the limit of mechanical stability .
the limit of the mechanical stability of a system can be determined from a study of the isothermal compressibility . in the space of variables identifying the thermodynamic state of the system in fact ,
the condition for the mechanical stability is @xcite @xmath57 where @xmath58 is the isothermal compressibility defined as @xmath59 in the phenomenological description of the liquid - gas transition , as given for instance by mean field theories like the van der waals equation , along an isotherm the limits of the mechanical stability are marked by the changes in the slope of the @xmath60 curve .
therefore from eq .
( [ eq : kt ] ) it is found that the crossing to unstable states with @xmath61 takes place at singularity points where @xmath58 diverges .
the spinodal line is identified by these singularity points .
in particular the point where the minimum of @xmath60 is located represents the limit of stability of the superheated liquid and so it belongs to the liquid branch of the spinodal .
we are interested here at the liquid spinodal which extends in the @xmath62 plane below the liquid - gas coexistence into the region of negative pressures .
the liquid spinodal indicates also the limiting values of tension for the existence of an homogeneous fluid before phenomena like cavitation take place @xcite .
the behaviour of @xmath60 for the bsv water is shown in fig .
[ fig:3 ] for @xmath63 ranging from 1.05 to 0.83 @xmath64 . in the inset of fig .
[ fig:3 ] four isotherms are represented upon lowering the temperature from @xmath65 k to @xmath66 k. isotherms from @xmath67 k to @xmath0 k are shown in the main frame of fig .
[ fig:3 ] . from @xmath39
k to the lowest temperature investigated the @xmath60 curves exhibit a minimum in the simulated range of densities .
the isotherms corresponding to lower temperatures have larger fluctuations due to the influence of both the low temperatures and the negative pressures in order to better follow the curves for the lowest temperatures shown in fig .
[ fig:3 ] we fitted the calculated points of the isotherms with a fourth - order polynomial function .
we observe however that due to the fluctuations the fit systematically slightly overestimates the values of the minima .
therefore we will consider the minima directly observed for the plot of the spinodal line .
based on eq .
[ eq : kt>0 ] the minima of the isotherms represent the limit of stability of the superheated liquid and identify the values of the pressure at the spinodal , @xmath68 . as @xmath55 decreases @xmath68 clearly shifts to larger values of @xmath49 .
this behaviour is similar to that observed for tip5p @xcite while for st2 and tip4p this shift is less evident @xcite .
the resulting estimated liquid spinodal line @xmath68 is plotted in fig .
[ fig:5 ] . in the same figure
also the @xmath69 isochores are shown as obtained by the data set of tables [ table1 ] and [ table2 ] . at least down to the lowest investigated temperature , @xmath0
k , the spinodal line is not retracing . in the slc framework
the spinodal line is expected to become reentrant at the point where it intersects the tmd line in the phase diagram .
we recall that this behaviour is required if the tmd line has a negative slope @xcite . along the tmd line
the coefficient of thermal expansion @xmath70 goes to @xmath71 . since the thermal pressure coefficient @xmath72 @xmath73 is connected to the coefficient @xmath74 by the following relation @xmath75 the tmd points are on the line connecting the minima of the estimated @xmath69 isochore for different @xmath49 .
according to fig .
[ fig:5 ] as @xmath49 decreases , approaching the spinodal line , the isochores exhibit a minimum at smaller values of @xmath55 .
the isochore for the lowest reported density @xmath76 @xmath32 does not show any minimum in the range of @xmath55 investigated where instead the minima for the other isochores are found .
the tmd line does not approach the spinodal line and at the lower @xmath55 investigated it results to have a positive slope .
this behaviour of the tmd line prevents an intersection with the spinodal at negative p and low t , unless a further change in the slope of the tmd line is hypothized .
these results are in agreement with the findings of the studies done on non polarizable models for water @xcite .
in fact for all these models the tmd line changes slope to avoid intersection with a non reentrant spinodal .
differences among the models can be found in the values of pressures and temperatures where the curves span .
in particular the change of slope of the tmd line appears in our model at negative pressures , @xmath77 mpa , similar to spc / e @xcite , @xmath78 mpa and to st2 @xcite . for tip5p @xcite
the change of slope appears at ambient pressure instead .
the most studied alternative to the slc for explaining the anomalies of water is the presence of a second critical point .
the scp hypothesis @xcite is based on the consideration that in the glassy state two distinct types of structure are observed @xcite , the low density amorphous ( lda ) ice and the high density amorphous ( hda ) ice .
this polymorphism of water would be extended into the liquid phase and it would emerge as a coexistence line between two liquid phases culminating in a second critical point @xcite .
computer simulation of waterlike lattice models show evidence in fact of two immiscible liquid forms of water @xcite .
more recently several amorphous ice phases @xcite have been experimentally observed and signatures have been found in computer simulation of the existence of several liquid - liquid coexistences in supercooled water @xcite .
also gibbs ensemble monte carlo ( gemc ) studies on the bsv water have found two liquid - liquid coexistence curves besides the liquid - vapor equilibrium @xcite .
the liquid branch of the gas - liquid coexistence curve obtained with the gemc is reported in fig .
[ fig:7 ] , where we collect the relevant features of the behaviour of the bsv water in the @xmath79 thermodynamical plane .
the spinodal line obtained in the present work in the supercooled region is compatible with the gemc results .
it comes out that the metastability region explored here goes very close to the boundaries of the region where new phases of the liquid water have been found .
we have done molecular dynamics simulation of supercooled water described with a polarizable bsv potential .
the aim of the paper was to calculate the behaviour of the spinodal line and of the tmd line down to lowest possible temperatures and pressures .
we have performed a structural analysis of water described by the bsv polarizable potential in the region of the supercooled liquid and we have found that the system persists in an homogeneous phase down to @xmath0 k , @xmath80 mpa and this occurs for @xmath81 .
signatures of the presence of the mechanical stability limit are observed in the behaviour of the values of the rdf at the first minimum .
the boundaries of the liquid mechanical stability at low @xmath55 and negative @xmath6 have been determined by the direct calculations of the liquid spinodal trough the minima of the isotherms @xmath60 .
the line of the tmd in the same thermodynamical region has also been detected from the minima of the isochores @xmath82 .
we did not find any retracing of the spinodal line in the investigated region while we observe a change of slope of the tmd line which prevents for continuity a possible intersection of the tmd with the spinodal even down to lower @xmath55 and @xmath6 .
these results are in agreement with previous calculations done with non polarizable site potentials @xcite .
our analysis excludes indeed the possibility of the slc scenario . as a matter of fact
the no retracing of the spinodal is compatible however with both the si interpretation @xcite and the possible existence of a second critical point @xcite . on the other hand
we have shown that the liquid spinodal calculated in the present work is in agreement with the gas - liquid coexistence curve obtained with the gemc @xcite . since in the gemc
there are clear indications of the existence of two liquid - liquid coexistence regions we can infer that our results support the second critical point scenario .
the use of a polarizable potential in the present work has given an answer to two important issues .
the first is that polarizable and non polarizable potentials are equally able to describe an unique , and therefore likely , scenario for the liquid - gas spinodal of supercooled water .
this is an important achievement as the parameters of these potentials have been calculated through fit to experimental data performed at ambient conditions .
therefore the uniqueness of phase behaviour far from these conditions reinforces the possibility that these potentials are able to predict existing features of water .
the second is that the analysis performed with the polarizable model can be extended in a different zone of the phase diagram for determining the liquid - liquid spinodal and indications about the second critical point location .
99 c. a. angell in _ water : a comprehensive treatise _ , edited by f. franks ( plenum press , new york , 1981 ) , vol .
7 , p. 1 ; annu .
34 * , 593 ( 1983 ) .
debenedetti _ metastable liquids .
concepts and principles _ princeton , nj , university press .
r. speedy , p. g. debenedetti , r. s. smith , c. huang , and b. d. kay , j. chem .
phys . * 105 * , 240 ( 1996 ) .
r. s. smith , b. d. kay , nature * 398 * 788 ( 1999 ) .
debenedetti , j. phys . : condens .
matt . * 15 * , r1670 ( 2003 ) . r.j .
speedy , c.a .
angell , j. phys .
chem . * 65 * , 851 ( 1976 ) .
r. speedy , j. phys
. chem . * 86 * 982 ( 1982 ) ; _ ibid _ * 86 * , 3002 ( 1982 ) .
p.h.poole , f.sciortino , u.essmann and h.e.stanley , nature * 360 * , 324 ( 1992 ) .
o. mishima , h.e .
stanley nature * 392 * , 164 ( 1998 ) ; ibid .
* 396 * , 329 ( 1998 ) .
s. sastry , p.g .
debenedetti , f. sciortino and h. e. stanley phys .
e * 53 * 6144 ( 1996 ) .
p.h.poole , f.sciortino , u.essmann and h.e.stanley , phys . rev .
e * 48 * , 3799 ( 1993 ) f. sciortino , p.h .
poole , u. essmann and h. e. stanley , phys .
e * 55 * , 727 ( 1997 ) .
p. h. poole , i. saika - voivod and f. sciortino j. phys .
matter * 17 * , l431 ( 2005 ) .
h. tanaka , j. chem .
* 105 * , 5099 ( 1996 ) .
s. harrington , p.h .
poole , f. sciortino and h.e .
stanley , j. chem .
phys . * 107 * , 7443 ( 1997 ) .
m.yamada , s. mossa , h.e .
stanley , f.sciortino , phys .
rev . lett . *
88*,195701 ( 2002 ) m.sprik and m. l. klein , j. chem .
* 89 * , 7556 ( 1988 ) .
p. ahlstrom , a. wallqvist , s. engstrom , b. jonsson , mol .
phys . * 68 * , 563 ( 1989 ) .
d. van belle , m. froeyen , g. lippens , and s. j. wodak , mol .
phys . * 77 * , 239 ( 1992 ) .
r. e. kozack and p. c. jordan , j. chem
* 96 * , 3120 ( 1992 ) .
s. w. rick , s. j. stuart and b. j. berne , j. chem .
* 101 * , 6141 ( 1994 ) .
g. ruocco and m. sampoli , mol .
* 82 * , 875 ( 1994 ) .
brodholt j. , sampoli m. , and vallauri r. , 1995 , mol .
, * 86 * , 149 ( 1995 ) .
p. g. kusalik , f. liden , and i. m. svishchev , j. chem .
103 , 10169 ( 1995 ) .
a. a. chialvo and p. t. cummings , j. chem . phys .
* 105 * , 8274 ( 1996 ) .
l. x. dang and t. m. chang , j. chem . phys .
* 106 * , 8149 ( 1997 ) .
k. kiyohara , k. e. gubbins and a. z. panagiotopolous , mol . phys . * 94 * , 803 ( 1998 ) . b. guillot and y. guissani , j. chem
. phys . * 114 * , 6720 ( 2001 ) .
h. b. yu , t. hansson , w. f. van gunsteren , j. chem . phys . * 118 * , 221 ( 2003 ) .
p.jedlovszky and r. vallauri , j. chem .
* 122 * , 081101 ( 2005 ) .
p.jedlovszky , r.vallauri , j. chem .
* 115 * , 3750 ( 2001 ) .
p.jedlovszky , r.vallauri , phys.rev.e * 67 * 011201 ( 2003 ) .
jedlovszky p. , brodholt j. p. , bruni f. , ricci m. a. , soper a. k. , and vallauri r. , 1998 , j. chem .
, * 108 * , 8528 .
p.jedlovszky , r.vallauri , mol.phys.97(1999 ) 1157 p.jedlovszky , j.richardi , j.phys.chem .
110 , 8019 ( 1999 ) c. a. tulk , c. j. benmore , j. urquidi , d. d. klug , j. neuefeind , b. tomberli and p. a. egelstaff , science * 297 * , 1320 ( 2002 ) .
i. brovchenko , a. geiger , a. oleinikova , j. chem .
phys . * 118 * , 9473 ( 2003 ) . | we develop a series of molecular dynamics computer simulations of liquid water , performed with a polarizable potential model , to calculate the spinodal line and the curve of maximum density inside the metastable supercooled region .
after analysing the structural properties , the liquid spinodal line is followed down to @xmath0 k. a monotonic decrease is found in the explored region .
the curve of maximum density bends on approaching the spinodal line .
these results , in agreement with similar studies on non polarizable models of water , are consistent with the existence of a second critical point for water . |
the possibility of baryogenesis by means of electroweak phase transition has recently been widely examined , since the electroweak baryogenesis can in principle be tested in the future accelerator experiments [ 1 ] .
if the electroweak phase transition is strongly first order , it can fulfil the departure from thermal equilibrium which is one of the three conditions required by sakharov that are necessary for the dynamic generation of the baryon asymmetry during the evolution of the universe [ 2 ] .
it has already been observed that in the standard model ( sm ) the electroweak phase transition can not be strongly first order unless the mass of the scalar higgs boson is smaller than its lower bound set experimentally by lep [ 3 ] .
the sufficient strength of the first order electroweak phase transition is essential for preserving the generated baryon asymmetry at the electroweak scale . in the literature ,
a number of articles have been devoted to study the possibility of accommodating the strongly first order electroweak phase transition in various models beyond the sm [ 4 ] . among them
, an interesting possibility has been investigated several years ago , where an extension of the sm with a real higgs singlet field has been adopted within the context of the electroweak phase transition [ 5 ] .
we consider that model is inspiring , because adding a real higgs singlet field is the simplest extension of the higgs sector of the sm . in that model ,
the strength of the first order electroweak phase transition has been stronger than that in the case of the sm . due to the presence of cubic terms in the tree - level higgs potential ,
a strongly first order electroweak phase transition has been ensured for a scalar higgs boson mass of around 100 gev , for the top quark mass of 127 gev . in this paper
, we would like to reexamine the model in a more rigorous way , in the sense that we perform the full integration of the finite - temperature effective potential instead of approximating it up to second order in temperature , that we include the higgs contributions to the finite temperature higgs potential at the one - loop level , and that we take into account the zero - temperature one - loop higgs potential .
as the experimental lower bound on the scalar higgs boson mass by now is improved to be more than 114 gev , and the top quark mass is usually set as about 175 gev , we would like to explore the parameter space of the model , with these new numbers in mind , where the first order electroweak phase transition may be strong enough . we find that the model is able to produce strongly first order electroweak phase transition for some parameter regions where a scalar higgs boson may have a mass larger than 114 gev .
further , we calculate cross sections for scalar higgs productions in @xmath2 collisions .
let us consider the higgs potential proposed by ref .
[ 5 ] where one real higgs singlet field is introduced to the sm higgs sector .
we follow the notations of ref . [ 5 ] . thus , the higgs potential consists of one complex higgs doublet field @xmath3 and one real higgs singlet field @xmath4 .
the tree - level higgs potential is given by @xmath5 where @xmath6 , @xmath7 , @xmath8 , and @xmath9 are parameters with mass dimension , and the rest are dimensionless .
we assume that @xmath10 , and @xmath11 have a positive values .
the real part of the neutral component of the higgs doublet field is , as in the sm , going to develop the non - zero vacuum expectation value ( vev ) @xmath12 while the higgs singlet field @xmath13 .
the tree - level values for @xmath14 and @xmath15 are determined by the extremum conditions @xmath16 where @xmath17 the above extremum conditions are satisfied if @xmath18 we denote the zero temperature values of @xmath14 and @xmath15 as @xmath19 and @xmath20 , respectively .
we take @xmath21 gev whereas @xmath20 is not fixed but variable . in this model , there are two neutral scalar higgs bosons . at the tree level ,
the masses for them are given by @xmath22 \cr & & \mbox { } \pm \left [ { 1 \over 4 } \left \ { - \mu_{\phi}^2 + \mu_s^2 + ( 3 \lambda_{\phi } - \lambda ) u^2 - ( 3 \lambda_s - \lambda ) v^2 - \left ( { \omega \over 2 } - 2 \alpha \right ) v \right \}^2 \right .
\cr & & \mbox { } \left .
+ \left ( 2 \lambda v - { \omega \over 2 } \right ) ^2 u^2 \right ] ^{1/2 } \ .\end{aligned}\ ] ] we associate the lighter neutral scalar higgs boson with the real part of the neutral component of the higgs doublet @xmath3 while the heavier one with the real higgs singlet @xmath4 . at the tree level ,
the gauge bosons and the top quark mass are @xmath23 gev , @xmath24 gev , and @xmath25 gev .
they are all defined in terms of @xmath19 , independently of @xmath20 .
now , we consider the one - loop contributions . at the one loop level , the zero temperature effective potential [ 6 ] and the finite temperature one [ 7 ] are respectively given by @xmath26 \ , \ ] ] @xmath27 \ , \ ] ] where @xmath28 , @xmath29 , @xmath30 , @xmath31 account for the degrees of freedom of each participating particles .
the negative sign in @xmath32 is for bosons while the positive sign for fermions . in @xmath33
, we take the renormalization scale as @xmath34 gev .
the @xmath35 dependence of the potential would disappear if all of higher - order contributions be taken into account .
this is desirable since physical observable quantities should be independent of the renormalization scale .
as we consider only the one - loop contribution in @xmath33 , the scale dependence remains . in order to reduce the effect of the scale dependence
, one may introduce appropriate counter terms to the one - loop contribution , after regularizing the relevant fields in the effective potential . in our model , the scale dependence on the mass of top quark or on the mass of weak gauge bosons can completely be eliminated in the one - loop contribution .
this is possible because the higgs singlet in our model does not contribute to fermion masses or the weak gauge boson masses .
thus , the one - loop contributions to these masses in our model are exactly identical to those in the sm .
hence , the scale dependence of the one - loop contributions to these masses may disappear if @xmath36 in eq .
( 6 ) is substituted by @xmath37 , where we can neglect @xmath38 .
this is effectively equivalent to replacing @xmath35 by @xmath39 .
the case of the higgs boson mass is more complicated than the case of the fermion masses or the weak boson masses , since the squared mass of the higgs boson at the tree level depends not only @xmath14 but also @xmath15 , and it is given by eq .
( 5 ) which is not in the form of @xmath40 ( where @xmath41 and @xmath42 are constants ) . nevertheless ,
comparing with the one - loop contribution to the top quark mass , we might effectively set @xmath35 in the one - loop contribution to the higgs boson mass by either @xmath43 or @xmath44 , in a similar way as @xmath45 in the case of top quark mass .
if a model is related to a new physics at certain scale , the renormalization scale might be taken comparable to the energy scale of the new physics , in order to consider the higher - order effects of the model reasonably .
for example , in a supersymmetric axion model , @xmath35 is taken to be @xmath46 gev in the one - loop effective potential , where new physics is expected to emerge at that energy scale [ 8 ] . in our study , we are interested in the tree - level mass of the higgs boson at the electroweak symmetry breaking scale .
thus , in our case , we might set @xmath34 gev . for small @xmath47 , that is , at high temperature
, @xmath32 can be expanded into the well - known form for the temperature induced effective potential .
rather , we perform the integration without using the high temperature approximation in our analysis , in order to consider relatively large scalar higgs boson mass .
the full effective potential that we consider therefore consists of @xmath48 note that the effective higgs potential possesses a symmetry under interchange @xmath49 .
thus , we may confine ourselves within the region of @xmath50 . in order for the first order electroweak phase transition to take place ,
we need two separate vacua .
the vacua are defined as the points in the ( @xmath14 , @xmath15)-plane at which the higgs potential has local minima .
the two extremum conditions for the higgs potential may be expressed as @xmath51 where the first equation is decomposed such that it is explicitly satisfied by @xmath52 . in the above expressions , @xmath53 and @xmath54
are given as @xmath55 with @xmath56 @xmath57 note that the negative sign in @xmath58 is for bosons whereas the positive sign is for fermions .
since the first extremum equation is satisfied by either @xmath52 or @xmath59 , one of the two points in the ( @xmath14 , @xmath15)-plane which satisfy the two extremum equations is simply ( 0 , @xmath60 ) where @xmath60 is the solution of @xmath61 .
this is the vacuum of the unbroken phase state .
we solve @xmath61 to obtain @xmath60 numerically by the bisection method .
let the other point in the ( @xmath14 , @xmath15)-plane which satisfies the two extremum equations be denoted as ( @xmath62 , @xmath63 ) , where @xmath64 .
this is the vacuum of the broken phase state .
this point is obtained by the newtons s method . in order to check that the extremum @xmath65 thus obtained
is minimum indeed , we need a jacobian matrix @xmath66 this point is a minimum of the higgs potential if they satisfy @xmath67 and @xmath68 .
the distance between the two minima in the ( @xmath14 , @xmath15)-plane is defined as @xmath69 .
the nonzero @xmath70 ensures that the electroweak phase transition is discontinuous , that is , first order .
the strength of the first order electroweak phase transition is measured by the critical temperature @xmath71 , which is defined as the temperature at which the values of the potential at the two minima are equal : @xmath72 .
we determine @xmath71 by varying temperature @xmath73 until @xmath72 is met .
if the ratio @xmath74 is larger than 1 , the first order electroweak phase transition is defined conventionally as a strong one .
now , let us evaluate the masses of the two neutral scalar higgs bosons from the symmetric @xmath75 mass matrix at the one - loop level .
the masses are evaluated at zero temperature .
thus , @xmath32 does not contribute to the masses at the one - loop level , and we should take @xmath76 and @xmath77 in the final expressions .
the elements of the mass matrix for the neutral scalar higgs bosons are given by @xmath78 where @xmath79 and @xmath80 are obtained from @xmath81 and @xmath33 , respectively .
explicitly , they are obtained as @xmath82 and @xmath83 { \log ( m_{s^0}^2 / m_{\phi^0}^2 ) \over ( m_{s^0}^2 - m_{\phi^0}^2 ) } \ , \end{aligned}\ ] ] where @xmath84 @xmath85 in terms of these matrix elements , the neutral scalar higgs boson masses at the one - loop level are given as @xmath86 \ , \ ] ] which depend on the quartic coupling coefficients @xmath87 , @xmath88 , and @xmath11 , and three parameters @xmath20 , @xmath7 , and @xmath6 , having mass dimension . here
, it is assumed that @xmath89 .
the quartic coupling coefficients satisfy the condition of @xmath90 ( @xmath91 ) for perturbation theory to work , in principle .
thus , they are not large . in practice
, we may set their values to be order of 1 or less .
the value of @xmath20 should be comparable to that of @xmath21 gev , because it has a similar property as @xmath19 .
note that @xmath14 is the higgs field in the sm for @xmath92 and @xmath19 in the sm is the electroweak symmetry breaking scale .
likewise , we assume that the other parameters @xmath7 and @xmath6 are not so much different from @xmath19 .
since we are interested in the mass region of 200 to 400 gev for the upper bound on the higgs boson masses , we choose the values of the parameters such that they yield the higgs boson masses within this region . for numerical study , we randomly search the parameter space by varying the relevant parameters within the ranges of @xmath93 , @xmath94 gev , and @xmath95 gev .
starting with a set of relevant parameters whose values are chosen randomly within the allowed ranges , we examine if @xmath96 , which is the criterion for the first order electroweak phase transition to be strong .
if this criterion is met , we continue to calculate the masses of the neutral scalar higgs bosons .
this sequence of calculations produces a point in the ( @xmath97 , @xmath74)-plane or in the ( @xmath98 , @xmath74)-plane . in this way
, we perform calculations to obtain 5000 points . figs .
1a and 1b show the results of our numerical studies on the masses of the neutral scalar higgs bosons . in fig .
1a , the 5000 points thus obtained are plotted in the ( @xmath97 , @xmath74)-plane , and in fig .
1b in the ( @xmath98 , @xmath74)-plane .
notice that the points in the ( @xmath98 , @xmath74)-plane are scattered slightly toward to the right of the plane , as compared to the points scattered in the ( @xmath97 , @xmath74)-plane , since we assume that @xmath89 .
these figures , therefore , indicate that there are sufficiently wide regions in the parameter space where the first order electroweak phase transition is strong as well as the masses of the neutral scalar higgs bosons are relatively large .
we now examine the possibility of detecting these neutral scalar higgs bosons in the future linear @xmath0 collider .
the three main production processes for the neutral scalar higgs bosons in @xmath2 collisions are : @xmath99 where the higgs - strahlung process is dominant at the center of mass energy corresponding to that of lep2 whereas @xmath100 and @xmath101 fusion processes are comparable at much higher center of mass energy , such as that of the future linear @xmath0 collider with @xmath102 gev . the cross section for @xmath3 or @xmath4 production via the higgs - strahlung process is related to the corresponding production cross section of the sm scalar higgs boson , @xmath103 , as @xmath104 and , similarly , the cross sections for @xmath3 and @xmath4 production via the @xmath100 fusion process or the @xmath101 fusion process are related to the corresponding production cross section of the sm scalar higgs boson , @xmath103 , as @xmath105 where @xmath106 or @xmath101 , and @xmath107 is the mixing angle between the two neutral scalar higgs bosons : @xmath108 these relationships hold because the coefficients for @xmath109 and @xmath110 coupling in our model are respectively @xmath111 and @xmath112 times the @xmath113 coupling coefficient in the sm .
for a given set of parameter values in our model , the above relationships indicate that , in case of the higgs - strahlung process , although neither @xmath114 nor @xmath115 is larger than @xmath116 , they can not be simultaneously smaller than the sm value . if one becomes negligible , the other approaches to the sm value .
the same argument holds in case of the fusion processes . if @xmath117 , the neutral scalar boson @xmath3 in our model would behave like @xmath103 in the sm .
its properties would be identical to those for @xmath103 in the sm .
in particular , if @xmath118 , the lower bound on the mass of @xmath3 would be the same as the lower bound on the sm higgs boson mass , 115 gev , which has been set by the lep experiments .
if , on the other hand , @xmath119 , the neutral higgs boson @xmath4 in our model would behave like @xmath103 in the sm .
let us consider the higgs - strahlung process .
at the center of mass energy of lep2 , the higgs - strahlung process in the sm is dominant over the fusion processes .
we first calculate both @xmath114 and @xmath115 , and then choose the larger one @xmath120 since at least one of them is comparable to @xmath116 .
we calculate @xmath121 for the each of the 5000 sets of relevant parameter values which produce the points in figs
we plot the result against @xmath97 in fig .
2 . the number of points in fig . 2 is smaller than those in figs . 1 , because those points in figs .
1 yielding @xmath122 gev are kinematically excluded from fig .
2 . from fig .
2 , the lower bound on @xmath97 might be roughly estimated .
it can be seen from fig .
2 that the lowest value for @xmath97 is about 20 gev .
however , this lower bound on @xmath97 does not take into account the discovery limit . if the discovery limit of the neutral scalar higgs boson @xmath3 in our model via the higgs - strahlung process in @xmath0 collisions at the lep2 energy is , for example , 400 fb , those points above 400 fb should be excluded since lep2 had not discovered @xmath3 or @xmath4 .
then , the lower bound on @xmath97 is pushed upward to about 65 gev .
it is known that the cross section for the sm higgs production via the higgs - strahlung process in @xmath0 collisions at the lep2 energy is about 200 to 250 fb for @xmath123 gev .
assuming that the discovery limit of the @xmath3 production cross section in our model is roughly the same order of magnitude , namely , a few hundred fb , we might estimate that @xmath124 gev .
similar argument would apply to the lower bound on the mass of the neutral scalar boson @xmath4 , if we plot @xmath121 against @xmath98 .
the lep2 experiments have provided detailed , mass dependent , limits on the higgs production and decay cross sections , yielding a limit on the higgs mass as a function of the ratio of the cross section times the branching ratio of its decay into @xmath125 pairs to the sm value .
combined with similar analyses for other decay modes , the lep2 experiments may exclude a higgs boson with mass of order 60 gev even if its production cross section is a tenth of the sm value . in ref .
[ 9 ] , the same analysis has been done independently of the specific hadronic decay mode of the higgs boson . comparing the result of ref .
[ 9 ] with our values , one can conclude that the bound on @xmath97 is indeed of order 60 gev . among the points in fig .
2 , those points with @xmath124 gev and small cross section , lower than 100 fb , are consistent with lep data . however , the key point here is that the lep2 experiments can not exclude the whole points in fig .
this implies that there are parameter regions in our model , where both the electroweak phase transition is strongly first order and the productions of the neutral scalar higgs bosons via the higgs - strahlung process in @xmath0 collisions are consistent with the lep2 experiments .
now , we increase @xmath126 of the @xmath0 collisions up to the proposed ilc center of mass energy of 500 gev . at this energy , depending on the scalar higgs boson mass , the @xmath100 and @xmath101 fusion processes are comparable to the higgs - strahlung process . at this energy ,
the search for the sm higgs boson via the higgs - strahlung process in @xmath0 collision may extend to the mass region up to about 400 gev .
the discovery limit of the cross section for the sm higgs production via the higgs - strahlung process at the ilc of @xmath127 gev is about 3 fb . for our model ,
taking the mixing angle factor @xmath128 or @xmath129 into account , the discovery limit would slightly be larger than the sm one .
it is thus possible to assume that the discovery limit of the cross section for the higgs production via the higgs - strahlung process in our model is less than 10 fb . at @xmath127 gev , all of the 5000 sets of parameter values we select in figs .
1 are kinematically allowed to produce @xmath3 or @xmath4 in @xmath0 collisions .
we calculate the cross sections for @xmath3 and @xmath4 production via the higgs - strahlung process , the @xmath100 process , and the @xmath101 process , namely , @xmath114 , @xmath115 , @xmath130 , @xmath131 , @xmath132 , and @xmath133 .
we also choose in each channel the larger one between the @xmath3 production cross section and the @xmath4 production cross section : @xmath134 , @xmath135 , and @xmath136 .
our results are plotted in figs .
3 - 5 . figs .
3 show the results for the higgs - strahlung process , fig .
3a for @xmath114 , fig .
3b for @xmath115 , and fig .
3c for @xmath121 . for simplicity ,
all of these figures are plotted against @xmath97 .
3c suggests that , at @xmath127 gev , the production cross sections via the higgs - strahlung process are perceivably higher than the discovery limit of 10 fb , for 5000 sets of parameter values in our model .
therefore , these sets of parameter values in our model may be explored at the ilc with @xmath127 gev . figs .
4 show the results for the @xmath100 fusion process , and figs . 5 for the @xmath101 fusion process .
the production cross sections via the @xmath101 fusion process are generally smaller than those via the other channels .
the production cross sections via the @xmath100 fusion process are considerably large for smaller scalar mass , but rapidly decrease as the scalar mass increases .
these behaviors are expectably consistent with the sm predictions .
successful baryogenesis requires the presence of cp violating sources beyond the ckm phase in the sm .
this is because the ckm phase in the sm can not produce sufficient cp violation for the required amount of the electroweak baryogenesis .
the higgs sector of our model can not accommodate cp violation since it consists of the sm higgs doublet and a real higgs singlet .
thus , we need sources of cp violation other than its higgs sector . as possible sources of cp violation , scenarios for new physics are good candidate .
assuming the existence of new physics have usually been induced to solve the hierarchy problem . among various scenarios for new physics ,
two of them can be incorporated with our model [ 10 , 11 ] . in one scenario
[ 10 ] , higher - dimensional interaction terms associated with the higgs doublet may be introduced at the electroweak energy scale . in this case
, the cp violation occurs at the electroweak scale , and the effective yukawa coupling for top quark is given by @xmath137 $ ] , where @xmath138 is a cp violating phase , @xmath139 is a real parameter , and @xmath140 is a cut off energy scale related to the new physics which is much larger than the electroweak energy scale .
[ 10 ] suggests that the energy scale for the new physics might be the majorana mass scale of the right - handed neutrino in left - right symmetric models .
another scenario is more realistic and suitable for our model [ 11 ] , where energy scale for the electroweak phase transition and the energy scale for the cp violation can be made different .
the higgs sector of the model studied in ref .
[ 11 ] has a conventional complex higgs doublet and an extra complex higgs singlet .
the conventional complex higgs doublet play the role of the sm higgs doublet , introducing the sm higgs boson , and the extra complex higgs singlet introduces two additional neutral higgs bosons .
both of the additional neutral higgs bosons are heavier than the sm higgs boson ; one of them stays at the electroweak scale whereas the other has a mass of high energy scale .
thus , among the three neutral higgs bosons of the model , there are two neutral higgs bosons at the electroweak scale and the third one is decoupled from them .
the model considered in ref . [ 11 ] introduces not only the extra complex higgs singlet but also extra vector - like down quark .
the introduction of a vector - like quarks to the sm is attractive since they naturally arise in grand unified theories such as @xmath141 , noted ref . [ 11 ] . in the fermion sector of the model considered in ref .
[ 11 ] , the sm fermions stay at the electroweak scale whereas the vector - like down quark stays at high energy scale
. consequently , the electroweak phase transition occurs at the electroweak scale , whereas the cp violation occurs at high energy scale . in this case
, therefore , the contributions of the heavy third neutral higgs boson and the vector - like down quark in the finite - temperature effective potential are boltzmann suppressed and thus negligible when the electroweak phase transition are considered at the electroweak scale . in other words , the model considered in ref . [ 11 ] becomes effectively identical to our model at the electroweak scale .
actually , by suitably redefining the tree - level higgs potential , and decoupling the third neutral higgs boson , in the model considered in ref . [ 11 ] , it becomes exactly the tree - level higgs potential of our model .
thus , our model for the electroweak phase transition may be extended naturally into the model considered in ref . [ 10 ] in order to accommodate the cp violation at high energy scale , without modifying our study at the electroweak scale .
the terms proportional to @xmath6 and @xmath7 in the tree - level higgs potential come from @xmath142 symmetry breaking terms .
if @xmath143 , these two terms are absent , and the parameter region allowing strongly first order electroweak phase transition disappears .
this can be seen , for example , in fig . 2 of ref . [
11 ] , where @xmath6 is the same and @xmath7 is @xmath144 .
thus , the @xmath142 breaking terms with @xmath145 are crucial for the strongly first order electroweak phase transition .
actually , there should be a potential barrier between two vacua in the tree - level higgs potential in order for the strongly first order electroweak phase transition to occur .
as an example , in a nonminimal supersymmetric model , a trilinear term may be introduced in its tree - level higgs potential to replace the @xmath146-term .
this term allows strongly first order electroweak phase transition in the nonminimal supersymmetric model . also , in ref .
[ 5 ] , one can see that there are a global minimum and a local minimum in the tree - level higgs potential , which become two separate global minima by the contribution of the temperature dependent effective potential .
the sm with an additional real higgs singlet in its higgs sector is studied within the context of the electroweak phase transition .
radiative corrections by the one - loop contribution from gauge bosons , top quark , and the higgs bosons are included , as well as the finite temperature effects .
the one - loop finite - temperature corrections are integrated numerically , instead of employing the high temperature approximation . by randomly choosing a large number of sets of the relevant parameter values within the ranges of @xmath93 , @xmath94 gev , and @xmath95 gev , we select those sets of parameter values that satisfy @xmath96 . for those particular sets of parameter values ,
we calculate the masses of the two neutral scalar higgs bosons of the model and the production cross sections for them via three dominant processes in @xmath0 collisions .
we find that there are as much as 5000 sets of parameter values of this model that may accommodate the strongly first order electroweak phase transition , which yield reasonably large masses for the neutral scalar higgs bosons and large production cross sections in @xmath0 collisions with @xmath126 = 500 gev .
thus , for certain parameter values , the prediction on the masses of the neutral scalar higgs bosons are consistent with the present experimental lower bound .
the results of our calculations suggest that this model may be tested in the near future at the proposed ilc
. 99 v.a .
kuzmin , v.a .
rubakov , and m.e .
shaposhnikov , phys .
b * 155 * , 36 ( 1985 ) ; m.e .
shaposhnikov , jetp lett .
* 44 * , 465 ( 1986 ) ; nucl .
b * 287 * , 757 ( 1987 ) ; nucl .
b * 299 * , 797 ( 1988 ) ; l. mclerran , phys .
62 * , 1075 ( 1989 ) ; n. turok and j. zadrozny , phys .
* 65 * , 2331 ( 1990 ) ; nucl
b * 358 * , 471 ( 1991 ) ; l.d .
mclerran , m.e .
shaposhnikov , n. turok , and m.b .
voloshin , phys .
b * 256 * , 451 ( 1991 ) ; m. dine , p. huet , r.j . singleton , and l. susskind , phys .
b * 257 * , 351 ( 1991 ) ; a.g .
cohen , d.b . kaplan , and a.e .
nelson , annu .
nucl . part .
* 43 * , 27 ( 1993 ) ; m. trodden , rev .
* 71 * , 1463 ( 1999 ) ; a. riotto and m. trodden , annu .
nucl . part .
sci . * 49 * , 35 ( 1999 ) .
sakharov , jetp lett .
* 5 * , 24 ( 1967 ) .
bochkarev , s.v .
kuzmin , and m.e .
shaposhnikov , mod . phys
. lett . a * 2 * , 417 ( 1987 ) ; phys .
b * 244 * , 275 ( 1990 ) ; m. dine , r.g .
leigh , p. huet , a. linde , and d. linde , phys .
d * 46 * , 550 ( 1992 ) ; p. arnold and o. espinosa , phys .
d * 47 * , 3546 ( 1993 ) ; z. fodor and a. hebecker , nucl .
b * 432 * , 127 ( 1994 ) ; k. kajantie , m. laine , k. rummukainen , and m. shaposhnikov , phys . rev .
* 77 * , 2887 ( 1996 ) ; f. csikor , z. fodor , and j. heitger , phys .
lett . * 82 * , 21 ( 1999 ) .
bochkarev , s.v .
kuzmin and m.e .
shaposhnikov , phys .
d * 43 * , 369 ( 1991 ) ; g.w . anderson and l.j .
hall , phys .
d * 45 * , 2685 ( 1992 ) ; n. turok and j. zadrozny , nucl . phys .
b * 369 * , 729 ( 1992 ) ; g.f .
giudice , phys .
d * 45 * , 3177 ( 1992 ) ; j.r .
espinosa , m. quirs , and f. zwirner , phys .
b * 307 * , 106 ( 1993 ) ; k. enqvist , k. kainulainen , and i. vilja , nucl .
b * 403 * , 749 ( 1993 ) ; m. pietroni , nucl .
b * 402 * , 27 ( 1993 ) ; i. vilja , phys . lett .
b * 324 * , 197 ( 1994 ) ; a.t .
davies , c.d .
froggatt , g. jenkins , and r.g .
moorhouse , phys .
b * 336 * , 464 ( 1994 ) ; a.t .
davies , c.d .
froggatt , and r.g .
moorhouse , phys .
b * 372 * , 88 ( 1996 ) ; m. carena , m. quirs and c.e.m .
wagner , phys .
b * 380 * , 81 ( 1996 ) ; d. delepine , j.m .
gerard , r.g .
felipe , and j. weyers , phys .
b * 386 * , 183 ( 1996 ) ; b. de carlos and j.r .
espinosa , nucl .
b * 503 * , 24 ( 1997 ) .
j.m . cline and p.a .
lemieux , phys .
d * 55 * , 3873 ( 1997 ) ; m. carena , m. quirs and c.e.m .
wagner , nucl . phys .
b * 524 * , 3 ( 1998 ) ; m. laine and k. rummukainen , phys .
lett . * 80 * , 5259 ( 1998 ) ; nucl . phys .
b * 535 * , 423 ( 1998 ) ; j.m . cline and g.d .
moore , phys .
lett . * 81 * , 3315 ( 1998 ) ; k. funakubo , a. kakuto , s. otsuki , and f. toyoda , prog .
phys . * 99 * , 1045 ( 1998 ) ; s.j .
huber and m.g .
schmidt , eur .
j. c * 10 * , 473 ( 1999 ) ; k. funakubo , prog .
theor . phys .
* 101 * , 415 ( 1999 ) ; m. losada , nucl .
b * 537 * , 3 ( 1999 ) ; m. bastero - gil , c. hugonie , s.f .
king , d.p .
roy , and s. vempati , phys .
b * 489 * , 359 ( 2000 ) ; a. menon , d.e .
morrissey , and c.e.m .
wagner , phys .
d * 70 * , 035005 ( 2004 ) ; s.w .
ham , s.k .
kim , e.j .
yoo , and d. son , phys .
d * 70 * , 075001 ( 2004 ) ; j. kang , p. langacker , t. li , and t. liu , phys .
lett . * 94 * , 061801 ( 2005 ) ; m. carena , a. megevand , m. quirs , and c.e.m .
wagner , hep - ph/0410352 .
j. choi and r.r .
volkas , phys .
b * 317 * , 385 ( 1993 ) ; j. choi , phys .
b * 345 * , 253 ( 1995 ) .
s. coleman and e. weinberg , phys .
d * 7 * , 1888 ( 1973 ) .
l. dolan and r. jackiw , phys .
d * 9 * , 3320 ( 1974 ) . v. barger , c.w .
chiang , j. jiang , and t. li , nucl .
b 705 , 71 ( 2005 ) .
lep higgs working group for higgs boson searches , hep - ex/0107034 .
x. zhang , s.k .
lee , k. whisnant , and b.l .
yong , phys .
d 50 , 7042 ( 1993 ) .
branco , d. delepine , d. emmanuel - costa , r. gonzalez felipe , phys .
b 442 , 229 ( 1998 ) . 0.3 in fig .
2 : for the same 5000 sets of parameter values as in fig .
1a , the larger one of the production cross sections , @xmath148 , for the neutral scalar higgs bosons via higgs - strahlung process in @xmath0 collisions with @xmath149 gev is plotted against @xmath97 . 0.3 in fig .
3a : for the same 5000 sets of parameter values as in fig . 1a , the cross section for @xmath3 production via higgs - strahlung process in @xmath0 collisions with @xmath127 gev is plotted against @xmath97 . 0.3 in fig .
3b : for the same 5000 sets of parameter values as in fig .
1a , the cross section for @xmath4 production via higgs - strahlung process in @xmath0 collisions with @xmath127 gev is plotted against @xmath97 . 0.3 in fig .
3c : for the same 5000 sets of parameter values as in fig .
1a , the larger one between the cross section for @xmath3 production and the cross section for @xmath4 production via higgs - strahlung process in @xmath0 collisions with @xmath127 gev is plotted against @xmath97 . | the higgs potential of the standard model with an additional real higgs singlet is studied in order to examine if it may allow the strongly first order electroweak phase transition .
it is found that there are parameter values for which this model at the one - loop level with finite temperature effect may allow the desired phase transition .
those parameter values also predict that the masses of the neutral scalar higgs bosons of the model are consistent with the present experimental bound , and that their productions in @xmath0 collisions may be searched at the proposed ilc with @xmath1 = 500 gev in the near future . |
single phase models of bulk melting usually treat this process as a destabilization of the solid due to proliferation of crystalline defects or excitations . among these
, lindemann suggested that melting occurs when the amplitude of atomic vibrations exceeds a critical value@xcite .
born proposed that melting is connected with the vanishing of the shear elastic coefficients @xcite .
other models suggest that melting is a consequence of the proliferation of intrinsic crystalline defects such as vacancies @xcite .
however , there is no clear experimental evidence supporting any one of these models , all of which rely on a single type of defect or excitation .
a model of melting involving interstitials coupled to shear strains proposed by granato @xcite is an exception as it involves coupling of defects and excitations .
recently , a dramatic decrease of the shear resistance was observed near the bcc - hcp transition of solid @xmath1@xcite .
these observations could have been interpreted as resulting from a coupling of point defects with a phonon which softens near the transition .
were it not for the bcc - hcp transition , one could envisage a continuation of this process until the crystal would melt .
thus , we have experimental evidence of a microscopic process which can potentially lead to melting
. however , this scenario requires a coupling of phonons and point defects , and therefore supports a picture similar to granato s , in which more than one type of excitation is involved . to understand this further
, it is necessary to investigate how point defects couple to phonons , and how this influences the shear resistance . in the following , we report results of molecular dynamics ( md ) simulations of the influence of point defects on the elastic properties of copper .
copper was chosen because of the large body of experimental data to which the results of the simulations can be compared .
we focus on the shear elastic coefficients , with the intention to test the born model in the presence of point defects .
we performed molecular dynamics simulations on samples containing up to 2048 atoms using the `` tight - binding ( tb ) potential '' with parameters calculated by cleri and rosato @xcite appropriate to copper .
newton s equations were solved using gear s predictor - corrector algorithm @xcite .
the time - step used was about 1 fsec . for temperature control
we used the method known as the nose - hoover thermostat @xcite .
the method of berendsen @xcite was used to calculate the volume of the samples .
this method does not generate trajectories in the constant npt ( here , n denotes the constant number of atoms , p is the constant pressure , and t is the temperature ) ensemble , but it sets the mean value of the pressure to the desired one while predicting the correct volume of the system . for ensemble average calculations we used the parrinello and rahman ( pr ) method @xcite . the statistical ensemble created by the parrinello rahman combined with nose s thermostat was identified as the isothermal - isotension ( ntt ) ensemble of statistical mechanics ( here , t is the constant tension ) .
we calculated the elastic constants using a fluctuation formula in both the canonical and in the ntt ensembles . in the ntt ensemble the formula for the isothermal elastic constants @xmath2 is @xmath3 where @xmath4 is the strain and @xmath5 is the volume of the sample at zero strain . calculating elastic constants using this formula involves two steps .
first we have to find the reference matrix , @xmath6 , defining the shape of the simulation box .
the condition which determines the shape of the box is that the sample is at zero stress . after running the simulation to calculate @xmath6 ,
the program should be used again to calculate the fluctuations .
not many studies have been made using this method because of the very slow convergence of the strain fluctuations @xcite . in the constant nvt ensemble ( here ,
n denotes the constant number of atoms , v is the constant volume , and t is the temperature ) , md can determine the elastic constants by using fluctuation formulas derived by ray and rahman : @xmath7 where @xmath8 is the kronecker delta and @xmath9 is the ensemble average of the born term , defined as @xmath10 where @xmath11 are the indices of the elastic constants tensor .
this formula for calculating the elastic constants is much less compact than the ntt formula , mainly because it contains the explicit second derivative of the inter atomic potential .
the main advantage of this method , and the reason for it being used more than the ntt method , is the faster convergence of the stress - stress fluctuation term relative to the convergence of the strain - strain fluctuation term .
two steps are required to calculate the elastic constants using this formula . as in the ntt formalism first one has to find the zero stress reference matrix and volume @xmath6 and @xmath5 and then run the nvt md program to calculate the elastic constants using equation ( [ nvt ] ) .
in order to validate the programs and to check the potential at various conditions we performed a series of calculations of the physical properties of copper using our programs and compared the results with experimental data .
first , we calculated the thermal expansion up to the melting point at zero external pressure .
our results are compared with neutron scattering data @xcite in fig .
[ alpha ] .
as expected , the calculated values are close to the experimental values at low temperatures ( we must bear in mind that the lattice constant at zero temperature was used as input in the fitting of the potential ) .
the agreement remains good even at elevated temperatures , with a maximal difference of about 10% .
secondly , the phonon spectrum , @xmath12 , was determined by calculating the time fourier - transform of the following quantities@xcite : @xmath13 where , @xmath14 and @xmath15 are the time dependent positions and velocities of the @xmath16th atom . the polarization vectors are defined as @xmath17 in the longitudinal case , while @xmath18 and @xmath19 , the transverse polarizations , are perpendicular to @xmath20 and to each other .
we then calculated the phonon spectrum along the @xmath21 and @xmath22 directions . due to the importance of long - time correlations
the calculations were made under conditions of constant energy without temperature control .
we ran the md program for @xmath23 time steps with temperature and pressure control , @xmath24 time steps with temperature control only , and then @xmath25 was saved to disk during @xmath26 time steps under energy conservation conditions .
the resulting spectrum , shown in fig .
[ wvsk ] , is in good agreement with neutron scattering data@xcite .
thirdly , to test the algorithms used to obtain the elastic constants , we calculated their variation with temperature using both approaches described in section ii .
the system was initially equilibrated for @xmath27 steps , and @xmath28 subsequent steps were used to calculate the fluctuations .
we found that the convergence of the pr method was slower than that of the nvt method . to monitor the convergence of the algorithm , we compared values of the elastic coefficients calculated along directions which are symmetrically equivalent . in the final result , after @xmath28 time steps , in both methods , the three equivalent elastic constants differ by no more than @xmath29 . from this variance
, one can estimate the accuracy of the calculation .
comparison with the experimental values measured by neutron scattering @xcite , shows that the shear moduli of the model - lattice and of real copper soften in the same way within the estimated accuracy of the simulations ( e.g. 10% ) .
the results are shown in fig .
[ shear_vs_temp ] .
the typical computer time needed to calculate the elastic constants varies between 4 days for a calculation in the nvt ensemble to one week in the ntt ensemble on a pentium - pro computer .
after successful validation of the pure system , point defects were introduced into the lattice by either removing an atom ( vacancy ) or adding one ( interstitial ) .
the simulations were done using a box having a fixed number of sites .
periodic boundary conditions were used throughout . in order to prevent mutual annihilation of the defects , we used defects of only one species in each run .
we first used simulated annealing@xcite to find the most stable configuration of a vacancy and of an interstitial in the model .
the formation energy of a vacancy was found to be @xmath30 .
the lowest energy configuration of an interstitial atom was found to be a ( 100)-split interstitial with a formation energy of @xmath31 .
these formation energies agree well with experimental data @xcite , and with previous simulations @xcite .
formation energies of other possible configurations of interstitials were determined using the conjugate gradient method , and are summarized in table [ ein_shem ] .
we point out that the difference between these energies is on the order of 0.1 ev ( @xmath32 1000k ) , which means that interstitials of various configurations may be present at high temperatures .
the influence of point defects was determined by calculating the shear elastic coefficients in samples containing different concentration of defects using md . for any concentration of defects ,
the number of atoms and unit cells in the sample is kept constant , so that the number of externally introduced defects is conserved . under this constraint ,
atoms and defects are free to diffuse , agglomerate etc .
we first discuss vacancies .
we found that the influence of vacancies on the shear elastic coefficients is too small to be resolved within our stated accuracy .
for example fig .
[ c_vs_vac_and_int_1400 ] shows that @xmath33 is independent of the concentration of vacancies .
this is correct both at low temperatures and near the melting point .
previously , influence of vacancies on the elastic properties at zero temperature was treated analytically by dederichs , et al .
@xcite who also reviewed earlier work .
the two body potentials used by these authors are less accurate than the one used here .
they found a reduction of 2 - 3 % in the values of @xmath33 and 1.5 - 3 % in values of @xmath34 per percent of vacancies .
the effect is indeed small , falling within the uncertainty limits of our calculation .
we turn now to the influence of interstitials .
in contrast to the vacancies , interstitials seem to cause a large softening of of the shear moduli , ( see fig . [ c_vs_vac_and_int_1400 ] ) . at low temperatures , we can qualitatively compare the results with the analytical calculation of dederichs _ et al _ , made at t=0k @xcite , and with the irradiation experiment of holder _
et al_@xcite , made at t=4.2k . in the latter ,
the defects produced by the irradiation are frenkel pairs and not interstitials .
the common trend is that defects induce a large softening of the shear moduli .
the softening is anisotropic , meaning that @xmath33 softens more than @xmath34 .
bearing in mind the different conditions under which the different studies were made , the qualitative agreement can be regarded as good .
the main new thrust of this work regards studying the influence of point defects on the shear moduli at increasingly higher temperatures , up to the melting temperature @xmath35 .
we emphasize that due to the use of periodic boundary conditions , the crystal can be super - heated and @xmath35 is higher than the thermodynamic melting temperature . in general , our results
can be explained by the combined effect of volume change and the dia - elastic effect@xcite .
volume changes resulting from introduction of defects arise due to the specific volume per defect being larger than that of an atom residing at a normal lattice site .
we can compare the change of the elastic coefficients caused by defects with that induced by heating the crystal up to a temperature where thermal expansion increases the volume by the same amount .
if the dependence of the elastic coefficients on the volume is the same , it would mean that effect of the defects is simply to expand the lattice .
we demonstrate such a comparison in fig .
[ cvsv400 ] for data obtained at t= 400k .
as one can see in this figure , the change of @xmath34 due to thermal expansion ( without defects ) is identical with that induced by increasing density of defects at a constant temperature when plotted against the lattice constant .
this proves that @xmath34 is affected only by volume changes . on the other hand ,
the change of @xmath33 due to thermal expansion is very different from that induced by increasing the density of defects , and can not be explained by the volume increase alone .
we attribute this behavior to the dia - elastic effect .
briefly , the dia - elastic effect describes the coupling of the internal vibration modes of the interstitial with the phonons . some of the internal vibration modes , the so called resonance modes , have frequencies within the band of the usual lattice spectrum .
mode coupling will lower the long wavelength part of the phonon branch related to the elastic coefficients .
for example , the ( 100 ) split interstitial is coupled to ( 100 ) shear strain . if the lattice is deformed in this way the response of the atoms forming the defect would be larger than the response of atoms occupying normal lattice sites .
the large response leads to a reduction of @xmath33 elastic coefficient@xcite . on the other hand ,
@xmath34 is associated with ( 110 ) shear strain , which does not couple to the ( 100 ) split interstitial .
consequently , the decrease of @xmath34 is induced only by volume expansion , as indeed seen in fig.[cvsv400 ] .
our results indicate that at high temperatures we can make one of the shear moduli vanish by adding a sufficient number of self - interstitials to the sample .
the number of defects needed for this purpose is temperature dependent .
the absolute change of the shear moduli per % of interstitials is larger for @xmath33 at all temperatures , as summarized in table [ sum1400 ] .
however , the _ relative _ sensitivity of @xmath34 becomes larger at high temperatures as melting is approached . because its initial value at elevated temperatures is already very small .
this is plotted in fig .
[ delta_c_rel ] .
consequently,@xmath34 is the first one to vanish .
this result is quite surprising , as it is opposite to what one would expect based on the low temperature behaviour@xcite .
we determined the melting temperature within our model at different temperatures and defect concentration , by monitoring the structural order parameter after @xmath26 time steps .
the results are summarized in fig .
[ phase_dia_int ] .
each point in the figure represents the results of several runs made at constant t and with the same defect density , but at a different random initial configuration .
the melting process found is mechanical melting , and the sample collapses due to the vanishing of @xmath34 .
this was found by correlating the temperatures at which @xmath34 vanishes with the melting temperature .
the melting process is rather fast , its duration being on the order of a typical vibration time of the crystal .
thus , this is a bulk melting mechanism .
as can be seen from our results the interstitials reduce the mechanical melting point , and a concentration of about @xmath36 lowered the melting point by about @xmath37 .
the same can not be said about vacancies , which did not reduce the mechanical melting point even at very high concentration ( up to 4% ) .
it turns out that the specific volume per atom at the phase boundary in fig .
[ phase_dia_int ] is fixed at 13.31@xmath38 .
the same specific volume is found by extrapolating the dependence of @xmath34 on the volume to the value at which @xmath34=0 .
this specific value is consistent with the specific volume of the liquid phase at t=1356k , the thermodynamic melting temperature , as suggested previously@xcite .
the way one reaches that specific volume , either by heating or by adding defects at a constant temperature does not seem to matter . in summary , we have simulated the influence of point defects on the shear moduli of cu at temperatures up to @xmath39 .
mechanical melting occurs due to the fullfillment of the born criterion , where one of the shear moduli of the solid vanishes .
we found that mechanical melting of the crystal is brought about by the vanishing of @xmath34 , rather than that of @xmath33 .
melting occurs once the specific volume reaches a critical value .
this value is independent of the path by which it is reached , either by thermal expansion alone , or by any combination of thermal expansion and of expansion due to interstitials introduced into the sample .
since the specific volume per vacancy is smaller than the atomic volume , vacancies do not expand the crystal .
hence , vacancies play no role in bulk melting .
we are grateful to david saada , adham hashibon , zaher salman for their contribution to this project , and iucc for the use of the computer facilities .
we thank r. w. cahn for his comments .
this work was supported in part by the israel science foundation and by the vpr technion fund for the promotion of research .
ubbelohde a.r .
, , clarendon press , oxford 1965 . born m. , , * 7 * , 591 ( 1939 ) .
frenkel d. , , dover publications , inc .
new - york 1955 .
granato a.v .
, , * 68 * , 974 ( 1992 ) ; granato a.v . , metallurgical and materials trans . *
29a * , 1837 ( 1998 ) .
berent i. and polturak e. , , * 81 * , 846 ( 1998 ) .
ducastelle f. , , edited by m. meyer and v. pontikis , kluwer , dordrecht , 1991 .
cleri f. and rosato v. , , * b48 * , 22 ( 1993 ) .
allen m. p. and tildesley d. j. , , clarendon press , oxford , 1987 .
hoover w.g . , , * a31 * , 1695 ( 1985 ) . berendsen h.j.c . , postma p.m. , van gunsteren w.f . , dinola a. and haak j.r . , ,
* 81 * , 3684 ( 1984 ) .
parrinello m. and rahman a. , , * 45 * , 1196 ( 1980 ) .
ray j.r . and rahman a. , , * 80 * , 4423 ( 1984 ) . ray j.r . and rahman a. , , * 82 * , 4243 ( 1985 ) . ray j.r . , ,
* 8 * , 109 ( 1988 ) .
, mansour k. a. , lee m. w. , and ray j. r. , , * b46 * , 8027 ( 1992 ) .
larose a. and brockhouse b. n. , , * 54 * , 1990 ( 1976 ) .
caprion d. , jund p. , jullien r. , , * 77 * , 675 ( 1996 ) .
nilsson g. , and rolandson s. , , * b7 * , 2393 ( 1973 ) .
amihai silverman and joan adler , , * 6 * , 277 ( 1992 ) .
wollenberger j.j .
, , edited by cahn r.w . and haasen p. , north - holland , amsterdam 1983 .
dederichs p.h . ,
lehmann c. and scholz a. , , * b20 * , 155 ( 1975 ) .
holder j. , granato a. v. , and rehn l. e. , , * b10 * , 349 , ( 1974 ) , ibid .
, 363 ( 1974 ) .
dienes g.j .
, , * 86 * , 228 ( 1952 ) .
dederichs p.h . , lehmann c. , schober h.r . , scholz a. and zeller r. , , * 69 @xmath40 70 * , 176 ( 1978 ) .
dederichs p.h . ,
lehmann c. and scholz a. , * 31 * , 1130 ( 1973 ) .
wolf d. , okamoto p. r. , yip s. , lutsko j. f. , and kluge m. , , * 5 * , 286 ( 1990 ) .
tallon j. l. , , * a39 * , 151 ( 1979 ) . | we present molecular dynamics simulations of the influence of point defects on the shear elastic coefficients of copper .
we find that vacancies do not influence these coefficients at all , while the introduction of interstitials causes a large reduction in the elastic coefficients .
the simulations establish a phase diagram of the melting temperature as a function of the density of interstitials . a crystal having no free surface undergoes bulk mechanical melting as a result of the vanishing of @xmath0 once the specific volume reaches a critical value , equal to the experimental volume of liquid phase .
this critical volume is history independent , in the sense that it can be reached either by heating the crystal or by adding defects at a constant temperature .
these results generalize the born model of melting for the case where point defects are present . ] |
the notion of magnetic charge has intrigued physicists since dirac @xcite showed that it was consistent with quantum mechanics provided a suitable quantization condition was satisfied : for a monopole of magnetic charge @xmath0 in the presence of an electric charge @xmath1 , that quantization condition is ( in this paper we use rationalized units ) @xmath2 where @xmath3 is an integer . for a pair of dyons , that is , particles carrying both electric and magnetic charge ,
the quantization condition is replaced by @xcite @xmath4 where @xmath5 and @xmath6 are the charges of the two dyons . ) below . ] with the advent of `` more unified '' non - abelian theories , classical composite monopole solutions were discovered @xcite .
the mass of these monopoles would be of the order of the relevant gauge - symmetry breaking scale , which for grand unified theories is of order @xmath7 gev or higher .
but there are models where the electroweak symmetry breaking can give rise to monopoles of mass @xmath8 tev @xcite .
even the latter are not yet accessible to accelerator experiments , so limits on heavy monopoles depend either on cosmological considerations @xcite , or detection of cosmologically produced ( relic ) monopoles impinging upon the earth or moon @xcite .
however , _ a priori _ , there is no reason that dirac / schwinger monopoles or dyons of arbitrary mass might not exist : in this respect , it is important to set limits below the 1 tev scale . such an experiment is currently in progress at the university of oklahoma @xcite , where we expect to be able to set limits on _ direct _ monopole production at fermilab up to several hundred gev .
this will be a substantial improvement over previous limits @xcite .
but _ indirect _ searches have been proposed and carried out as well .
de rjula @xcite proposed looking at the three - photon decay of the @xmath9 boson , where the process proceeds through a virtual monopole loop .
if we use his formula @xcite for the branching ratio for the @xmath10 process , compared to the current experimental upper limit @xcite for the branching ratio of @xmath11 , we can rule out monopole masses lower than about 400 gev , rather than the 600 gev quoted in ref . @xcite .
similarly , ginzburg and panfil @xcite and more recently ginzburg and schiller @xcite considered the production of two photons with high transverse momenta by the collision of two photons produced either from @xmath12 or quark-(anti-)quark collisions . again the final photons are produced through a virtual monopole loop .
based on this theoretical scheme , an experimental limit has appeared by the d0 collaboration @xcite , which sets the following bounds on the monopole mass @xmath13 : @xmath14 where @xmath15 is the spin of the monopole .
it is worth noting that a mass limit of 120 gev for a dirac monopole has been set by graf , schfer , and greiner @xcite , based on the monopole contribution to the vacuum polarization correction to the muon anomalous magnetic moment .
( actually , we believe that the correct limit , obtained from the well - known textbook formula @xcite for the @xmath0-factor correction due to a massive dirac particle is 60 gev . )
the purpose of the present paper is to , first , describe the key theoretical elements necessary for the establishment of a direct limit for production of monopoles at fermilab : an estimate of the production cross section for monopole - antimonopole pairs at the collider must be made , and then an estimate of the binding probability of such produced monopoles with matter must be given , so that we can predict how many monopoles would be bound to the detector elements that are run through our induction detector .
such estimates were given in our proposal to fermilab ; the complete analysis will be given in the experimental papers to follow . here
the emphasis will be on the elementary processes involved .
a secondary purpose of this paper is to critique the theory of refs .
@xcite , @xcite , @xcite , and @xcite , and thereby demonstrate that experimental limits , such as that of ref .
@xcite , based on virtual processes are unreliable .
we will show that the theory is based on a naive application of electromagnetic duality ; the resulting cross section can not be valid because unitarity is violated for monopole masses as low as the quoted limits , and the process is subject to enormous , uncontrollable radiative corrections .
it is not correct , in any sense , as refs . @xcite and @xcite state ,
that the effective expansion parameter is @xmath16 , where @xmath17 is some external photon energy ; rather , the factors of @xmath18 emerge kinematically from the requirements of gauge invariance at the one - loop level .
if , in fact , a correct calculation introduced such additional factors of @xmath18 , arising from the complicated coupling of magnetic charge to photons , we argue that no limit could be deduced for monopole masses from the current experiments .
it may even be the case , based on preliminary field - theoretic calculations , that processes involving the production of real photons vanish .
it is envisaged that if monopoles are sufficiently light , they would be produced by a drell - yan type of process occurring in @xmath19 collisions at the tevatron .
the difficulty is to make a believable estimate of the elementary process @xmath20 , where @xmath21 stands for quark and @xmath13 for magnetic monopole .
it is not known how to calculate such a process using perturbation theory ; indeed , perturbation theory is inapplicable to monopole processes because of the quantization condition ( [ quant ] ) .
it is only because of that consistency condition that the dirac string , for example , disappears from the result .
only formally has it been shown that the quantum field theory of electric and magnetic charges is independent of the string orientation , or , more generally , is gauge and rotationally invariant @xcite .
it has not yet proved possible to develop generally consistent schemes for calculating processes involving real or virtual magnetically charged particles .
partly this is because a sufficiently general field theoretic formulation has not yet been given ; this defect will be addressed elsewhere @xcite .
however , the nonrelativistic scattering of magnetically charged particles is well understood @xcite . thus it should not be surprising that an eikonal approximation gives a string - independent result for electron - monopole scattering provided the condition ( [ quant ] ) is satisfied .
more than two decades ago schwinger proposed @xcite and urrutia carried out such a calculation @xcite .
indeed , this subject has had arrested development .
since this is the only successful field - theoretic calculation yet presented , it may be useful to review it here .
( a more detailed discussion will be presented in ref .
@xcite . )
the interaction between electric ( @xmath22 ) and magnetic ( @xmath23 ) currents is given by @xmath24 here @xmath25 is the usual photon propagator , and the arbitrary `` string '' function @xmath26 satisfies @xmath27 it turns out to be convenient for this calculation to choose a symmetrical string , which satisfies @xmath28 in the following we choose a string lying along the straight line @xmath29 , in which case the function may be written as a fourier transform @xmath30 in the high - energy , low - momentum - transfer regime , the scattering amplitude between electron and monopole is obtained from eq .
( [ weg ] ) by inserting the classical currents , @xmath31 where @xmath32 and @xmath13 are the masses of the electron and monopole , respectively .
let us choose a coordinate system such that the incident momenta of the two particles have spatial components along the @xmath33-axis : @xmath34 and the impact parameter lies in the @xmath35 plane : @xmath36 apart from kinematical factors , the scattering amplitude is simply the transverse fourier transform of the eikonal phase , @xmath37 where @xmath38 is simply @xmath39 with the classical currents substituted , and @xmath40 is the momentum transfer .
first we calculate @xmath38 ; it is immediately seen to be , if @xmath29 has no time component , @xmath41 where @xmath42 is the component of the photon momentum perpendicular to the @xmath33 axis . from this expression
we see that the result is independent of the angle @xmath43 makes with the @xmath33 axis .
we next use proper - time representations for the denominators in eq .
( [ chi ] ) , @xmath44e^{i\lambda { \bf \hat n\cdot k_\perp } } e^{-|\lambda|\epsilon}.\end{aligned}\ ] ] we then complete the square in the exponential and perform the gaussian integration to obtain @xmath45 or @xmath46 because @xmath47 must be continuous when @xmath48 and @xmath49 lie in the same direction , we must have the schwinger quantization condition for an infinite string , @xmath50 where @xmath3 is an integer .
to carry out the integration in eq .
( [ eikonal ] ) , choose @xmath49 to make an angle @xmath51 with @xmath40 , and the projection of @xmath48 in the @xmath35 plane to make an angle @xmath52 with @xmath40 ; then @xmath53 to avoid the appearance of a bessel function , we first integrate over @xmath54 , and then over @xmath51 : @xmath55 where @xmath56 is a unit circle about the origin , and where again the quantization condition ( [ quant2 ] ) has been used . squaring this and putting in the kinematical factors we obtain urrutia s result @xcite @xmath57 which is exactly the same as the nonrelativistic , small angle result found , for example , in ref . @xcite .
this calculation , however , points the way toward a proper relativistic treatment , and will be extended to the crossed process , quark - antiquark production of monopole - antimonopole pairs elsewhere .
once the monopoles are produced in a collision at the tevatron , they travel through the detector , losing energy in a well - known manner ( see , e.g. , ref .
@xcite ) , presumably ranging out , and eventually binding to matter in the detector ( be , al , pb , for example ) . the purpose of this section is to review the theory of the binding of magnetic charges to matter .
we consider the binding of a monopole of magnetic charge @xmath0 to a nucleus of charge @xmath58 , mass @xmath59 , and magnetic moment @xmath60 @xmath61 being the spin of the nucleus .
( we will assume here that the monopole mass @xmath62 , which restriction could be easily removed . ) other notations for the magnetic moment are @xmath63 the charge quantization condition is given by eq .
( [ quant ] ) . because the nuclear charge is @xmath58 , the relevant angular momentum quantum number is [ recall @xmath3 is the magnetic charge quantization number in eq .
( [ quant ] ) ] @xmath64 we do not address the issue of dyons @xcite , which for the correct sign of the electric charge will always bind electrically to nuclei . in this subsection we follow the early work of malkus @xcite and the more recent paper of bracci and fiorentini @xcite .
( there are also the results given in ref .
@xcite , but this reference seems to contain errors . ) the neutron ( @xmath66 ) is a special case .
binding will occur in the lowest angular momentum state , @xmath67 , if @xmath68 since @xmath69 , this condition is satisfied for all @xmath3 . in general , it is convenient to define a reduced gyromagnetic ratio , @xmath70 this expresses the magnetic moment in terms of the mass and charge of the nucleus .
binding will occur in the special lowest angular momentum state @xmath71 if @xmath72 thus binding can occur here only if the anomalous magnetic moment @xmath73 .
the proton , with @xmath74 , will bind .
binding can occur in higher angular momentum states @xmath75 if and only if @xmath76 for example , for @xmath77 , @xmath78 , and for @xmath79 , @xmath80 .
thus @xmath81he , which is spin 1/2 , will bind in the first excited angular momentum state because @xmath82 . unfortunately , to calculate the binding energy , one must regulate the potential at @xmath83 .
the results shown in table 1 assume a hard core .
the reference here is @xcite .
the assumption made here is that @xmath84 .
( there are only 3 exceptions , apparently : @xmath85h , @xmath86li , and @xmath87b . ) binding in the lowest angular momentum state @xmath88 is given by the same criterion ( [ lowest ] ) as in spin 1/2 .
binding in the next state , with @xmath89 occurs if @xmath90 where @xmath91 the previous result for @xmath65 is recovered , of course .
@xmath92 is a special case : then @xmath93 is always negative , while @xmath94 if @xmath95 , where @xmath96 for higher spins , both @xmath97 can exceed @xmath98 : @xmath99 where for @xmath100 @xmath101 for @xmath102be , for which @xmath103 , we can not have binding because @xmath104 , @xmath105 , where the ranges come from considering different values of @xmath3 from 1 to @xmath106 . for @xmath107 ,
@xmath108 so @xmath109al will bind in either of these states , or the lowest angular momentum state , because @xmath110 , and @xmath111 , @xmath112 .
kazama and yang treated the dirac equation @xcite .
see also @xcite and @xcite .
in addition to the bound states found nonrelativistically , deeply bound states , with @xmath113 are found .
these states always exist for @xmath114 . for @xmath115 ,
these ( relativistic ) @xmath116 bound states exist only if @xmath117 .
thus ( modulo the question of form factors ) kazama and yang @xcite expect that electrons can bind to monopoles .
( we suspect that one must take the existence of these deeply bound states with a fair degree of skepticism .
see also @xcite . )
as expected , for @xmath115 we have weakly bound states only for @xmath118 , which is the same as the nonrelativistic condition ( [ lowest ] ) , and for @xmath114 , only if @xmath119 , where @xmath120 is given in eq .
( [ kappac ] ) .
olsen , osland , and wu considered this situation @xcite . in this case , no bound states exist , unless an additional interaction is introduced ( this is similar to what happens nonrelativistically , because of the bad behavior of the hamiltonian at the origin ) .
bound states are found if an `` induced magnetization '' interaction ( quadratic in the magnetic field ) is introduced .
binding is then found for the lowest angular momentum state @xmath121 again if @xmath73 . for the higher angular momentum states ,
the situation is more complicated : * for @xmath122 : bound states require @xmath123 , and * for @xmath124 : bound states require @xmath125 .
but these results are probably highly dependent on the form of the additional interaction .
the binding energies found are inversely proportional to the strength @xmath126 of this extra interaction . .weakly bound states of nuclei to a magnetic monopole .
the angular momentum quantum number @xmath75 of the lowest bound state is indicated . in notes ,
nr means nonrelativistic and r relativistic calculations ; hc indicates an additional hard core interaction is assumed , while ff signifies use of a form factor .
im = induced magnetization , the additional interaction employed for the relativistic spin-1 calculation .
we use @xmath127 except for the deuteron , where @xmath128 is required for binding . [
cols="^,^,^,^,^,^,^,^",options="header " , ] clearly , this summary indicates that the theory of monopole binding to nuclear magnetic dipole moments is rather primitive .
the angular momentum criteria for binding is straightforward ; but in general ( except for relativistic spin 1/2 ) additional interactions have to be inserted by hand to regulate the potential at @xmath83 .
the results for binding energies clearly are very sensitive to the nature of that additional interaction
. it can not even be certain that binding occurs in the allowed states .
in fact , however , it seems nearly certain that monopoles will bind to all nuclei , even , for example , be , because the magnetic field in the vicinity of the monopole is so strong that the monopole will disrupt the nucleus and will bind to the nuclear , or even the subnuclear , constituents .
now the question arises : is the bound complex of nucleus and monopole rigidly attached to the crystalline lattice of the material ?
this is a simple tunneling situation .
the decay rate is estimated by the wkb formula @xmath129 where the potential is crudely @xmath130 @xmath131 is the nuclear mass @xmath132 monopole mass , and the inner and outer turning points , @xmath133 and @xmath134 are the zeroes of @xmath135 . provided the following equality holds , @xmath136 which should be very well satisfied , since the right hand side equals @xmath137 mev@xmath138 , we can write the decay rate as @xmath139,\ ] ] where the characteristic field , defined by @xmath140 , is @xmath141 t. if we put in @xmath142 t , and @xmath143 , @xmath144mev , appropriate for @xmath145al , we have for the exponent , for @xmath127 , @xmath146 , corresponding to a rather long time ! to get a 10 yr lifetime ,
the binding energy would have to be only of the order of 1ev .
monopoles bound with kilovolt or more energies will stay around forever .
then the issue is whether the entire al atom can be extracted with the 1.5 t magnetic field present in cdf .
the answer seems to be unequivocally no .
the point is that the atoms are rigidly bound in a lattice , with no nearby site into which they can jump .
a major disruption of the lattice would be required to dislodge the atoms , which would probably require kilovolts of energy @xcite .
some such disruption was made by the monopole when it came to rest and was bound in the material , but that disruption would be very unlikely to be in the direction of the accelerating magnetic field . again , a simple boltzmann argument shows that any effective binding slightly bigger than 1 ev will result in monopole trapping `` forever . ''
this argument applies equally well to binding of monopoles in ferromagnets .
if monopoles bind strongly to nuclei there , they will not be extracted by 5 t fields , contrary to the arguments of goto et al .
@xcite the corresponding limits on monopoles from ferromagnetic samples of carrigan et al .
@xcite are suspect .
finally , let us consider the process contemplated in refs . @xcite and @xcite , that is @xmath147 where the photon scattering process is given by the one - loop light - by - light scattering graph shown in fig . [ fig1 ] .
if the particle in the loop is an ordinary electrically charged electron , this process is well - known @xcite . if , further , the photons involved are of very low momentum compared the the mass of the electron , then the result may be simply derived from the well - known euler - heisenberg lagrangian @xcite , which for a spin 1/2 charged - particle loop in the presence of homogeneous electric and magnetic fields is ) is only valid when @xmath148 . ]
@xmath149 . \label{ehlagrangian}\ ] ] here the invariant field strength combinations are @xmath150 @xmath151 being the dual field strength tensor , and the argument of the hyperbolic cosine in eq .
( [ ehlagrangian ] ) is given in terms of @xmath152^{1/2}=[({\bf h}+i{\bf e})^2]^{1/2}.\ ] ] if we pick out those terms quadratic , quartic and sextic in the field strengths , we obtain @xmath153\nonumber\\ & & \mbox{}-{\pi\alpha^3\over630}{1\over m^8}f^2[8(f^2)^2 + 13 ( f { } ^*f)^2]+\dots . \label{ehlag}\end{aligned}\ ] ] the lagrangian for a spin-0 and spin-1 charged particle in the loop is given by similar formulas which are derived in ref .
@xcite and ( implicitly ) in ref .
@xcite , respectively . given this homogeneous - field effective lagrangian
, it is a simple matter to derive the cross section for the @xmath154 process in the low energy limit .
( these results can , of course , be directly calculated from the corresponding one - loop feynman graph with on - mass - shell photons .
see refs .
explicit results for the differential cross section are given by ref .
@xcite : @xmath155 and the total cross section for a spin-1/2 charged particle in the loop is and @xmath156 , respectively
. numerically the coefficients are @xmath157 , @xmath158 , and @xmath159 for spin 0 , spin 1/2 , and spin 1 , respectively . ]
@xmath160 here , @xmath17 is the energy of the photon in the center of mass frame , @xmath161 .
this result is valid provided @xmath162 .
the dependence on @xmath32 and @xmath17 is evident from the lagrangian ( [ ehlag ] ) , the @xmath17 dependence coming from the field strength tensor . further note that perturbative quantum corrections are small , because they are of relative order @xmath163 @xcite .
processes in which four final - state photons are produced , which may be easily calculated from the last displayed term in eq .
( [ ehlag ] ) , are even smaller , being of relative order @xmath164 .
so light - by - light scattering , which has been indirectly observed through its contribution to the anomalous magnetic moment of the electron @xcite , is completely under control for electron loops . how is this applicable to photon scattering through a monopole loop ? at first blush
this calculation seems formidable .
the interaction of a magnetically charged particle with a photon involves a `` string , '' as described by the function @xmath165 given in eq .
( [ string ] ) .
the interaction between electric and magnetic charges is given by the complicated expression ( [ weg ] ) .
this coupling is equivalent to the interaction between the magnetic current @xmath23 and the electromagnetic field , @xmath166 from eqs .
( [ weg ] ) and ( [ string ] ) one obtains the relevant string - dependent monopole - photon coupling vertex in momentum space , @xmath167 where we have , for variety s sake , chosen a semi - infinite string . as we have noted , the choice of the string is arbitrary ; reorienting the string is a kind of gauge transformation .
in fact , it is this requirement that leads to the quantization conditions ( [ quant ] ) and ( [ squant ] ) . the authors of refs .
@xcite , @xcite , and @xcite do not attempt a calculation of the `` box '' diagram with the interaction ( [ mmint ] ) .
rather , they ( explicitly or implicitly ) appeal to duality , that is , the symmetry that the introduction of magnetic charge brings to maxwell s equations : @xmath168 and similarly for charges and currents .
thus the argument is that for low energy photon processes it suffices to compute the fermion loop graph in the presence of zero - energy photons , that is , in the presence of static , constant fields .
the box diagram shown in fig .
[ fig1 ] with a spin-1/2 monopole running around the loop in the presence of a homogeneous @xmath169 field is then obtained from that analogous process with an electron in the loop in the presence of a homogeneous @xmath170 field , with the substitution @xmath171 .
since the euler - heisenberg lagrangian ( [ ehlag ] ) is invariant under the substitution ( [ duality ] ) on the fields alone , this means we obtain the low energy cross section @xmath172 through the monopole loop from eq .
( [ llcs ] ) by the substitution @xmath171 , or @xmath173 it is critical to emphasize that the euler - heisenberg lagrangian is an effective lagrangian for calculations at the _ one fermion loop level _ for low energy , i.e. , @xmath174 .
it is commonly asserted that the euler - heisenberg lagrangian is an _ effective lagrangian _ in the sense used in chiral perturbation theory @xcite .
this is not true .
the qed expansion generates derivative terms which do not arise in the effective lagrangian expansion of the euler - heisenberg lagrangian @xcite .
one can only say that the euler - heisenberg lagrangian is a good approximation for light - by - light scattering ( without monopoles ) at low energy because radiative corrections are down by factors of @xmath175 .
however , it becomes unreliable if radiative corrections are large . in this regard , both the ginzburg @xcite and the de rjula @xcite articles , particularly ref .
@xcite , are rather misleading as to the validity of the approximation sketched in the previous section .
they state that the expansion parameter is not @xmath0 but @xmath16 , @xmath13 being the monopole mass , so that the perturbation expansion may be valid for large @xmath0 if @xmath17 is small enough .
but this is an invalid argument .
it is only when external photon lines are attached that extra factors of @xmath18 occur , due to the appearance of the field strength tensor in the euler - heisenberg lagrangian .
moreover , the powers of @xmath0 and @xmath18 are the same only for the @xmath176 process . the expansion parameter is @xmath177 , which is huge . instead of radiative corrections being of the order of @xmath175 for the electron - loop process , these corrections will be of order @xmath177 , which implies an uncontrollable sequence of corrections .
for example , the internal radiative correction to the box diagram in fig .
[ fig1 ] have been computed by ritus @xcite and by reuter , schmidt , and schubert @xcite in qed . in the @xmath178 term in eq .
( [ ehlag ] ) the coefficients of the @xmath179 and the @xmath180 terms are multiplied by @xmath181 and @xmath182 , respectively .
the corrections become meaningless when we _ replace _
@xmath183 .
this would seem to be a devastating objection to the results quoted in ref .
@xcite and used in ref .
but even if one closes one s eyes to higher order effects , it seems clear that the mass limits quoted are inconsistent . if we take the cross section given by eq .
( [ llcs ] ) and make the substitution ( [ subs ] ) , we obtain for the low energy light - by - light scattering cross section in the presence of a monopole loop @xmath184 if the cross section were dominated by a single partial wave of angular momentum @xmath75 , the cross section would be bounded by @xmath185 if we take @xmath186 as a typical partial wave . comparing this with the cross section given in eq .
( [ monocs ] ) , we obtain the following inequality for the cross section to be consistent with unitarity , @xmath187 but the limits quoted @xcite for the monopole mass are less than this : @xmath188 because , at best , a minimum @xmath189 gev ; the theory can not sensibly be applied below a monopole mass of about 1 tev .
( note that changing the value of @xmath75 in the unitarity limits has very little effect on the bound ( [ unitbd ] ) since an 8th root is taken : replacing @xmath75 by 50 reduces the limit ( [ unitbd ] ) only by 50% . ) similar remarks can be directed toward the de rjula limits @xcite .
that author , however , notes the `` perilous use of a perturbative expansion in @xmath0 . ''
however , although he writes down the correct vertex , eq .
( [ mmvertex ] ) , he does not , in fact , use it , instead appealing to duality , and even so he admittedly omits enormous radiative corrections of @xmath190 without any justification other than what we believe is a specious reference to the use of effective lagrangian techniques for these processes .
apparently , then , the formal small @xmath17 result obtained from the euler - heisenberg lagrangian can not be valid beyond a photon energy @xmath191 .
the reader might ask why one can not use duality to convert the monopole coupling with an arbitrary photon to the ordinary vector coupling .
the answer is that little is thereby gained , because the coupling of the photon to ordinary charged particles is then converted into a complicated form analogous to eq .
( [ mmint ] ) .
this point is stated and then ignored in ref .
@xcite in the calculation of @xmath10 .
there is , in general , no way of avoiding the complication of including the string .
we are currently undertaking realistic calculations of virtual ( monopole loop ) and real ( monopole production ) magnetic monopole processes .
these calculations are , as the reader may infer , somewhat difficult and involve subtle issues of principle involving the string , and it will be some time before we have results to present . therefore , here we wish to offer plausible qualitative considerations , which we believe suggest bounds that call into question the results of ginzburg et al .
@xcite .
our point is very simple .
the interaction ( [ mmint ] ) couples the magnetic current to the dual field strength .
this corresponds to the velocity suppression in the interaction of magnetic fields with electrically charged particles , or to the velocity suppression in the interaction of electric fields with magnetically charged particles , as most simply seen in the magnetic analog of the lorentz force , @xmath192 that is , the force between an electric charge @xmath1 and magnetic charge @xmath0 , moving with relative velocity @xmath193 and with relative separation @xmath194 is @xmath195 this velocity suppression is reflected in nonrelativistic calculations . for example
, the energy loss in matter of a magnetically charge particle is approximately obtained from that of a particle with charge @xmath58 by the substitution @xcite @xmath196 and the classical nonrelativistic dyon - dyon scattering cross section near the forward direction is @xcite @xmath197 { 1\over(\theta/2)^4},\quad \theta\ll1,\ ] ] the expected generalization of the rutherford scattering cross section at small angles .
of course , the true structure of the magnetic interaction and the resulting scattering cross section is much more complicated .
for example , classical electron - monopole or dyon - dyon scattering exhibits rainbows and glories , and the quantum scattering exhibits a complicated oscillatory behavior in the backward direction @xcite .
these reflect the complexities of the magnetic interaction between electrically and magnetically charged particles , which can be represented as a kind of angular momentum @xcite
. nevertheless , for the purpose of extracting qualitative information , the naive substitution , @xmath198 seems a reasonable first step . ] which reduces the quoted mass limits of ref .
@xcite a bit . ]
indeed , such a substitution was used in the proposal @xcite to estimate production rates of monopoles at fermilab .
the situation is somewhat less clear for the virtual processes considered here .
nevertheless , the interaction ( [ mmint ] ) suggests there should , in general , be a softening of the vertex . in the current absence of a valid calculational scheme
, we will merely suggest two plausible alternatives to the mere replacement procedure adopted in refs .
@xcite .
we first suggest , as seemingly ref .
@xcite does , that the approximate effective vertex incorporates an additional factor of @xmath18 .
thus we propose the following estimate for the @xmath199 cross section in place of eq .
( [ monocs ] ) , @xmath200 since there are four suppression factors in the amplitude .
now a considerably larger value of @xmath17 is consistent with unitarity , @xmath201 if we take @xmath186 again .
we now must re - examine the @xmath202 cross section . in the model given in ref .
@xcite , where the photon energy distribution is given in terms of the functions @xmath203 , @xmath204 , the physical cross section is given by @xmath205 where now ( _ cf . _
( 25 ) of in the first reference in ref .
@xcite ) @xmath206 where , for spin 1/2 , ( up to factors of order unity ) @xmath207 the result in ( [ newdsigma ] ) differs from that in ref .
@xcite by a factor of @xmath208 .
the photon distribution function @xmath209 used is rather strongly peaked at @xmath210 .
( this peaking is necessary to have any chance of satisfying the low - frequency criterion . )
when we multiply by @xmath211 , the amplitude is greatly reduced and the peak is shifted above @xmath212 , violating even the naive criterion for the validity of perturbation theory .
nevertheless , the integral of the distribution function is reduced by two orders of magnitude , that is , @xmath213 this reduces the mass limit quoted in @xcite by a factor of @xmath214 , to about 500 gev , where @xmath215 .
this dubious result makes us conclude that it is impossible to derive any limit for the monopole mass from the present data . as for the de rjula limit , @xmath216 , and @xmath217 , proportional to @xmath218 in each case , once again ignoring both the string and the radiative correction problem .
he assumes that the monopole is a heavy vector - like fermion , and obtains a limit of @xmath219 .
our ansatz changes @xmath218 to @xmath220 , so that @xmath221 gev , a substantial reduction . ] from the @xmath10 process , if we insert a suppression factor of @xmath18 at each vertex and integrate over the final state photon distributions , given by eq . (
18 ) of ref .
@xcite , the mass limit is reduced to @xmath222 gev , again grossly violating the low energy criterion . and the limit deduced from the vacuum polarization correction to the anomalous magnetic moment of the muon due to virtual monopole pairs @xcite is reduced to about 2 gev .
the reader might object that this @xmath18 softening of the vertex has little field - theoretic basis .
therefore , we propose a second possibility that does have such a basis .
the vertex ( [ mmvertex ] ) suggests , and detailed calculation supports ( based on the tensor structure of the photon amplitudes , when @xmath29 points in the time direction . apart from this different tensor structure , the vacuum polarization is given by exactly the usual formula , found , for example in ref .
details of this and related calculations will be given elsewhere .
] ) the introduction of the string - dependent factor @xmath223 at each vertex , where @xmath21 is the photon momentum .
such a factor is devastating to the indirect monopole searches for any process involving a real photon , such as that of the d0 experiment @xcite or for @xmath10 discussed in @xcite , the amplitude vanishes . because such factors can and do appear in full monopole calculations , it is clearly premature to claim any limits based on virtual processes involving real final - state photons .
the field theory of magnetic charge is still in a rather primitive state .
indeed , it has been in an arrested state of development for the past two decades . with serious limits
now being announced based on laboratory measurements , it is crucial that the theory be raised to a useful level . at the present time , we believe that theoretical estimates for the production of real monopoles are more reliable than are those for virtual processes .
this is because , in effect , the former are dominated by tree - level processes .
we have indicated why the indirect limits can not be taken seriously at present ; and of course only the real production processes offer the potential of discovery . perhaps the arguments here will stimulate readers to contribute to the further development of the theory , for it remains an embarrassment that there is no well - defined quantum field theory of magnetic charge .
we are very pleased to dedicate this paper to kurt haller on the occasion of his 70th birthday , in view of his many contributions to non - abelian gauge theory , of which dual qed is a disguised variant .
we thank igor solovtsov for helpful conversations and we are grateful to the u.s .
department of energy for financial support .
t. t. wu and c. n. yang in _ properties of matter under unusual conditions _ , ed . by h. mark and s. fernbach ( wiley , new york , 1969 ) , p. 349 ; phys .
d * 12 * , 3843 ( 1975 ) ; a. m. polyakov , jetp lett . *
20 * , 194 ( 1974 ) ; y. nambu , phys . rev .
d * 10 * , 4262 ( 1974 ) ; g. t hooft , nucl . phys . * b79 * , 276 ( 1974 ) ; b. julia and a. zee , phys . rev .
d * 11 * , 2227 ( 1975 ) .
p. h. eberhard , et al .
d * 4 * , 3260 ( 1971 ) ; r. r. ross , et al .
d * 8 * , 698 ( 1973 ) ; p. b. price , shi - lun guo , s. p. ahlen , and r. l. fleischer , phys .
lett . * 52 * , 1264 ( 1984 ) ; b. cabrera , phys .
lett . * 48 * , 1378 ( 1982 ) ; * 51 * , 1933 ( 1983 ) ; * 55 * , 25 ( 1985 ) ; h. jeon and m. j. longo , phys . rev . lett . * 75 * , 1443 ( 1995 ) , erratum * 76 * , 159 ( 1996 ) ; m. ambrosio et al .
, phys . lett . *
b406 * , 249 ( 1997 ) .
p. osland and t. t. wu , nucl .
* b247 * , 421 , 450 ( 1984 ) ; * b256 * , 13 , 32 ( 1985 ) ; p. osland , c. l. schultz , and t. t. wu , nucl . phys .
* b256 * , 449 ( 1985 ) ; p. osland and t. t. wu , nucl .
* b261 * , 687 ( 1985 ) .
actually , only the result for the low energy cross section for a @xmath224 loop is given in f .- x .
dong , x .- d .
jiang , and x .- j .
zhou , phys .
d * 47 * , 5169 ( 1993 ) ; g. jikia and a. tkabladze , phys .
b323 * , 453 ( 1994 ) .
see also references therein . | recently , there has been renewed interest in the search for low - mass magnetic monopoles . at the university of oklahoma we are performing an experiment ( fermilab e882 ) using material from the old d0 and cdf detectors to set limits on the existence of dirac monopoles of masses of the order of 500 gev . to set such limits ,
estimates must be made of the production rate of such monopoles at the tevatron collider , and of the binding strength of any such produced monopoles to matter .
here we sketch the still primitive theory of such interactions , and indicate why we believe a credible limit may still be obtained . on the other hand ,
there have been proposals that the classic euler - heisenberg lagrangian together with duality could be employed to set limits on magnetic monopoles having masses less than 1 tev , based on virtual , rather than real processes .
the d0 collaboration at fermilab has used such a proposal to set mass limits based on the nonobservation of pairs of photons each with high transverse momentum .
we critique the underlying theory , by showing that the cross section violates unitarity at the quoted limits and is unstable with respect to radiative corrections .
we therefore believe that no significant limit can be obtained from the current experiments , based on virtual monopole processes . |
in higher dimensional supergravity and string theories compactifications of spacetimes are an essential tool to make contact with the four - dimensional phenomenology . in the past
the majority of research was focused on the investigation of compactifications to four - dimensional minkowski or anti - de sitter spacetime .
klein spectrum of such compactifications was studied in detail to analyze the properties of the resulting supergravity theories @xcite . in contrary
, cosmology motivates the study of effective four - dimensional geometries that correspond to an expanding universe .
the successful cosmological model with quasi - de sitter epochs during inflation and dark energy domination today gradually shifts the interest towards higher dimensional models that provide possibilities for incorporating effective four - dimensional cosmology .
so far , realistic embeddings of cosmology / de sitter geometry into fundamental superstring / m - theories require tools beyond simple low - energy supergravity solutions @xcite .
higher rank form fields are natural ingredients of higher dimensional supergravity theories . in particular , in the attempt to construct cosmological solutions , form fields play a crucial role .
they are utilized to stabilize the moduli fields associated with the shape of the extra dimensions @xcite .
furthermore , non - trivial background fluxes in the extra dimensional space generically induce exponential potentials for the volume modulus or radion .
the applicability of such potentials for cosmological quintessence and inflaton fields was recently investigated in @xcite .
commonly the dynamics are analyzed within the effective four - dimensional theory , which has obvious limitations .
most importantly , it requires a stable higher dimensional background configuration .
the stability of background solutions is conveniently investigated by the study of linear perturbations .
the analysis of perturbations in compactifications with effective four - dimensional de sitter geometry shows that it is not at all generic for finding stable background configurations .
in particular , realistic scenarios with a large number of extra dimensions are plagued by two possible channels of instabilities in the scalar sector of perturbations .
\(i ) the nature and dynamics of the tachyonic instability of the lowest kaluza klein state in the scalar spectrum i.e .
, the volume modulus were investigated in @xcite .
the instability was recognized previously in @xcite . the generic contribution that arises from a four - dimensional inflationary geometry with expansion rate @xmath0
is given by @xmath1 where @xmath2 is the number of extra dimensions .
it is possible to compensate this term with positive contributions from the curvature of the internal space ( if positive ) and from stabilizing bulk fields , e.g. scalar fields or form fluxes .
if not stabilized this tachyon indicates a non - linear reconfiguration of the compactification @xcite .
generic late - time attractors are universal kasner - like solutions , where the internal dimensions shrink to zero size , or complete decompactifications to a higher dimensional de sitter space , where all dimensions expand isotropically , or transitions towards regimes where the de sitter curvature is small and the tachyonic mode disappears .
\(ii ) it turns out that a second instability can arise from the quadrupole and higher moments of kaluza
klein excitations @xmath3 in the scalar sector .
contrary to the instability of the volume modulus that can be stabilized due to the interaction with matter fields , this instability arises from the presence of non - trivial background configurations of matter fields .
whereas the instability of the volume modulus ( that is , the kaluza klein zero - mode ) preserves the spherical symmetry of the extra space , the second instability indicates a deformation of the internal geometry .
although the tachyonic mode was noticed in @xcite , its nature and consequences have not been explored so far and non - perturbative examples for this instability have not been studied yet .
the range of values of the flux @xmath4 that allows for stable compactifications shrinks with the number of extra dimensions
. for more than four extra dimensions , stable compactifications can not be found at all for this model . on the other hand
, the mass spectrum of vector and tensor perturbations does not reveal additional channels of instabilities . for the vector sector ,
the lowest lying mass states are massless and independent of the matter fields .
they transform in the adjoint representation of the isometry group of the internal manifold and their mass is protected by this local symmetry . unlike the scalar sector of perturbations , the coupling between matter and metric perturbations of the higher kaluza
klein vector modes does not lead to additional instabilities in the sector of vector perturbations .
the lowest excitation of the tensor spectrum is the massless four - dimensional graviton , whose mass is protected by four - dimensional diffeomorphism invariance . besides the massless graviton a tower of positive kaluza
klein modes appears in the effective theory , whose scaling is only sensitive to the properties of the internal manifold .
although the tachyonic instabilities render the discussion within the effective four - dimensional theory invalid , they may trigger interesting dynamics . the tachyonic scalars couple to the matter fields in the effective four - dimensional theory . in kaluza
klein compactifications the standard model fields of the effective four - dimensional theory are considered to be zero - modes of the higher dimensional theory .
the unstable volume modulus couples gravitationally to such fields , which might lead to interesting phenomenological consequences such as tachyonic preheating of the standard model fields . in simple braneworld models
the preheating of brane fields from the decaying radion was investigated in @xcite . in this paper , we present the complete analysis of the mass spectrum of perturbations around de sitter vacua in a unified way .
we systematically treat scalar , vector and tensor perturbations .
we choose the model of freund - rubin compactifications , where all spectra are obtained analytically .
nevertheless , the qualitative results are expected to be valid for general de sitter compactifications with flux stabilization that are frequently discussed in the literature . in particular , we investigate the resulting spectrum in the context of inflation .
light scalar modes ( with masses @xmath5 smaller than the inflation scale @xmath0 ) acquire a scale invariant spectrum of perturbations after inflation and therefore provide additional sources for cosmological perturbations .
the mass scale of kaluza
klein excitations of vector and tensor modes is always larger than @xmath0 , so that no contribution to post - inflationary cosmology is expected .
the analysis follows closely the calculations done in @xcite but goes substantially beyond this for the case of vector perturbations and shape moduli fields , which have not been addressed in the framework of de sitter compactifications before .
the calculation of the complete spectrum allows us to conclude that the tachyonic instability of de sitter compactifications is located in the scalar sector only .
the paper is structured as follows . in sec .
[ sec : background ] , we set up the background solution for the de sitter compactification .
the perturbations are introduced in sec .
[ sec : perturbations ] . they are classified according to their transformation properties with respect to the de sitter isometry group .
the equations of motion for the perturbations are obtained and their spectra calculated . in section [ sec : applications ] , we discuss possible implications of the presence of light mass states for cosmological models . some technical details are shown in [ sec : notation][sec : residual ] .
we consider a @xmath6-dimensional product spacetime @xmath7 . like in freund - rubin compactification @xcite , we allow for a @xmath2-form flux field that spontaneously compactifies @xmath2 of the spatial dimensions . for simplicity ,
we assume that the compact space is a @xmath2-sphere i.e . ,
@xmath8 with radius @xmath9 .
the generalization to any homogeneous space is straightforward .
the @xmath10-dimensional part of product space is a de sitter space @xmath11 with a curvature scale @xmath12 . for cosmological applications @xmath13 .
to compensate for the positive curvature of @xmath14 , a cosmological constant @xmath15 is introduced .
we do not discuss the microscopical origin of the cosmological constant , but we assume that it is common to all dimensions . in principle , considering quantum effects , one can expect additional contributions from the casimir energy to the cosmological constant of the compact space .
its contribution however strongly depends on the choice and the dimension of the compact manifold and we will assume that it can be neglected in comparison to the overall cosmological constant .
the theory we study is described by the action @xmath16 from which the general equations of motion for the metric and the @xmath2-form field are derived @xmath17 the planar coordinate patch for @xmath10-dimensional de sitter geometry is parameterized by the coordinates @xmath18 and the @xmath2-dimensional sphere with radius @xmath9 is labeled by the coordinates @xmath19 @xmath20 throughout this work we use capital italic indices @xmath21 to address all spacetime dimensions .
greek indices @xmath22 take values in @xmath23 and correspondingly @xmath24 are used to label the compactified dimensions . the static @xmath2-form flux that supports this compactification
is given by @xmath25 where @xmath26 is the completely antisymmetric volume form of the @xmath2-sphere .
the background is characterized by four parameters : the de sitter scale @xmath0 , the radius of compactification @xmath9 , the flux strength @xmath4 and the cosmological constant @xmath15 .
the equations of motions ( [ eq : eom1 ] ) and ( [ eq : eom2 ] ) reduce to algebraic constraints in this space of parameters @xmath27 the last relation shows the requirement of a positive cosmological constant that compensates for the curvatures of the spacetime .
the form field flux is an additional parameter that enriches the dynamics twofold . unlike the cosmological constant it is associated with a field degree of freedom .
it is natural to consider perturbations of this field .
secondly , it allows one to create a hierarchy between the hubble scale @xmath0 and the size of the internal manifold @xmath9 , which is needed if one wants to apply this background as approximate solution of the late - time universe , where @xmath28 .
dirac s quantization condition requires @xmath4 to be quantized .
although we treat @xmath4 effectively as a continuous parameter , its discrete nature is implied . in the limit @xmath29 ,
when the flux disappears , one recovers the known results for standard kaluza
klein compactifications .
the range of values for the flux strength @xmath4 is limited by the physical restriction of @xmath30 . from the first of the equations ( [ eq : param_back ] ) follows that @xmath31 the limit @xmath32 corresponds to four - dimensional effective minkowski space geometry with @xmath33 .
in this section we analyze the dynamics of perturbations in the background of freund - rubin compactifications introduced in the previous section . first the metric perturbations are grouped into scalar , vector and tensor perturbations .
the gauge is fixed to eliminate gauge degrees of freedom that correspond to the infinitesimal coordinate transformations of the form @xmath34 similarly perturbations for the matter fields are introduced , grouped into scalars and vectors .
gauge degrees of freedom associated with the transformation properties of the @xmath35-form gauge potential @xmath36 are fixed .
the most general set of perturbations that respects the product geometry of the chosen background spacetime is parameterized by the following line element @xmath37 dx^\mu dx^\nu+\l[(1 + 2\phi)g_{mn}+h_{(mn)}\r]dy^mdy^n+v_{\mu n}dx^\mu dy^n\ , \end{aligned}\ ] ] that contains the scalars @xmath38 , @xmath39 and @xmath40 , vectors @xmath41 and the @xmath2-dimensional traceless graviton @xmath42 . throughout this paper , indices in parentheses
indicate the symmetrized and traceless components of the tensorial object i.e .
, @xmath43 .
it is customary to further decompose the vectors and tensor into longitudinal and transverse parts , e.g. @xmath44 and @xmath45 .
however , this decomposition is only relevant for the zero - modes that is , for the @xmath46-independent sector .
the gauge choice that we impose below in equations ( [ eq : dedonder ] ) has a residual gauge freedom that allows us to impose @xmath47 , @xmath48 and @xmath49 for the @xmath46-independent perturbations , [ cf.equations ( [ eq : res_constraint ] ) ] .
for more details on the residual gauge freedom we refer to the calculations in [ sec : residual ] . in the massive i.e . , @xmath46-dependent sector these fields become the longitudinal components of the vectors @xmath41 and tensor @xmath42 .
after compactification and integrating out the extra dimensions , the graviton of the effective four - dimensional theory is not canonically normalized unless the metric @xmath50 is rescaled by an appropriate weyl transformation , cf .
[ sec : weyl ] . on the linear level
this amounts to the redefinition of the scalar perturbation @xmath38 by the so - called weyl shift @xmath51 to fix the gauge and to obtain equations of motion in a suitable form the de donder gauge conditions are imposed @xmath52 the de donder conditions are trivially satisfied for the @xmath46-independent perturbations .
therefore , @xmath41 consists of @xmath2 homogeneous vectors and @xmath40 contains @xmath53 homogeneous scalars . for inhomogeneous perturbations ,
the de donder gauge imposes one condition on the vector fields @xmath41 and @xmath2 conditions among the scalars @xmath40 .
therefore , @xmath41 represents @xmath54 massive vectors and @xmath40 contains @xmath55 scalars .
the missing vector and scalar modes are reshuffled to the longitudinal polarizations of the graviton and the remaining vectors .
this has been shown explicitly for compactifications on a @xmath2-dimensional torus in @xcite .
the gauge choice of equations ( [ eq : dedonder ] ) does not fix the gauge freedom @xmath56 completely . the residual gauge freedom
consists of functions that satisfy the additional constraints @xmath57 we fixed the residual gauge that corresponds to the @xmath46-independent coordinate transformations .
the remaining residual gauge transformations will be discussed in [ sec : residual ] .
next we provide the notation for the form field perturbations and the gauge choice analogous to the notation of @xcite .
the perturbations of the form field strength are denoted with @xmath58 . unlike the background , which only has non - vanishing components in the directions of the compact space
, the perturbations can fluctuate freely in all directions .
locally , the perturbations of the @xmath2-form field can be represented by a @xmath35-from potential i.e .
, @xmath59 . depending on the index structure of the potential @xmath60
, it either transforms as a scalar @xmath61 , vector @xmath62 , or higher rank anti - symmetric tensor @xmath63 .
in particular , interesting are scalars and vectors , since they mix with the metric scalars and vectors .
the decomposition of the field strength into gauge field scalars and vectors is given by @xmath64}\ , \nonumber\\ \fl f_{\mu m_2\cdots m_q}=&q\,a_{[m_2\cdots m_q;\mu]}\,\,=&(-)^{q-1}(q-1)a_{\mu[m_2\cdots m_{q-1}|m_q]}+a_{m_2\cdots m_q;\mu}\ , \\
\fl f_{\mu\nu m_3\cdots m_q}=&q\ , a_{[\nu m_3\cdots
m_q;\mu]}=&a_{\nu m_3\cdots m_q;\mu}-a_{\mu m_3\cdots
m_q;\nu}+(-)^{q-1}(q-2 ) a_{\mu\nu[m_3\cdots m_{q-1}|m_q]}\ .\nonumber\end{aligned}\ ] ] like the electromagnetic gauge potential , the @xmath35-potential @xmath60 consists of gauge degrees of freedom , represented by the transformation @xmath65 , where @xmath66 is an arbitrary @xmath67-form . to fix this gauge freedom a lorentz - like gauge condition is imposed @xmath68 a similar condition holds for the potentials with one or more indices in the direction of the de sitter space . on a compact @xmath2-dimensional manifold without boundary any @xmath69-form @xmath70
can be decomposed into an exact , a co - exact and a harmonic @xmath69-form i.e . , @xmath71 where the lower index indicates the rank of the form .
the gauge choice in equation ( [ eq : lorentz ] ) is equivalent to the statement that @xmath72 is co - closed that is , @xmath73 .
therefore only the co - exact and harmonic contribution survive in the general decomposition .
the three - dimensional analogue of the above - described decomposition is representing a vector @xmath74 whose divergence vanishes
i.e . , @xmath75as the curl of another vector , @xmath76 .
generalized to higher rank tensor objects the decomposition can be written as @xmath77}(y)+\sum\limits_{h=1}^{b_2 } \beta^h_\mu(x){\epsilon^{mn}}_{m_3\cdots m_q } y^h_{[mn]}(y)\ , \\ & \spc\vdots\nonumber\end{aligned}\ ] ] where the first term in each line represents the generalization of the curl and the second term summarizes the contribution from harmonic forms on the compact space -forms @xmath78}(y)$ ] on a compact manifold depends on its @xmath69th betti number i.e . , the number of non - trivial @xmath69-cycles . on a @xmath2-sphere the only harmonic forms are the constant zero - forms and the volume form .
the existence of additional scalars @xmath79 and vector fields @xmath80 requires harmonic one - forms @xmath81 and two - forms @xmath82}$ ] as exist for example on a torus . ] .
the @xmath83 , @xmath84,@xmath85 , @xmath86}(y)$ ] are the scalar , vector and tensor eigenforms of the laplace - beltrami operator on the compact space i.e . , @xmath87 .
this completes the classification and gauge fixing of the perturbations . in the following subsections
the equations of motion and spectra for the perturbations are derived . in the scalar sector , the metric fluctuations @xmath39 and @xmath88 mix with the scalar from the flux fluctuations @xmath89 .
similarly , the fluctuations @xmath41 couple to the flux perturbations @xmath90 in the sector of vector fluctuations . the sector of tensor modes
is determined by the metric fluctuations @xmath42 alone .
in what follows we ignore the harmonic sector , @xmath91 , since it gives rise to massless fields that are related to special topological properties of the internal space .
for the simple example of a @xmath2-sphere , the only non - vanishing harmonic forms are the constant scalar functions and the volume form , corresponding to the betti numbers @xmath92 and @xmath93 .
all other betti numbers vanish .
one set of dynamical equations for the perturbations are obtained from the linearized version of the equations ( [ eq : eom1 ] ) .
the expression of the ricci tensor is readily obtained from the general formula for metric perturbations @xmath94
: @xmath95 , \end{aligned}\ ] ] where @xmath96 , @xmath97 and @xmath98 , respectively . the expression of the linearized ricci tensor
is given in the set of equations ( [ eq : ein1])([eq : ein3 ] ) , where the gauge conditions ( [ eq : dedonder ] ) were used for simplification .
the right - hand side of the field equations is expressed through a linear combination @xmath99 involving the energy
momentum tensor @xmath100 : @xmath101 the above expression ( [ eq : rhs ] ) is expanded up to first order in the perturbations .
the particular form ( [ eq : flux_back ] ) of the background flux is used explicitly .
this gives rise to various contractions of the epsilon tensor @xmath26 with the perturbations of the form flux @xmath102 .
these contractions can be reduced to expressions in the perturbations @xmath89 and @xmath103 only , that were introduced in the equations ( [ eq : hodgedecomp ] ) .
useful relations that simplify the right - hand side of the linearized einstein equations to the form presented in the equations ( [ eq : ein4])([eq : ein6 ] ) are collected in [ sec : formequations ] . @xmath104 \label{eq : ein1}\\ \fl \delta r_{\mu n}=\l[\frac{p+q-2}{p-2}\phi_{;\mu}-(p-1)\psi_{;\mu}+\frac{1}{2 } { h_{(\lambda\mu)}}^{;\lambda}\r]_{|n}\!+\frac{1}{2}\l[v^\lambda_{\,\,n;\mu\lambda}-(\box+\delta)v_{\mu n}+v^k_\mu\,r_{kn } \r ] \nonumber\\\label{eq : ein2}\\ \fl \delta r_{m n}=\l[\frac{2(p+q-2)}{p-2}\phi - p\psi\r]_{|(mn)}+\l[\frac{2(p+q-2)}{q(p-2)}\delta\phi-(\box+\delta)\phi-\frac{p}{q}\delta\psi\r ] g_{mn}\nonumber\\ + \frac{1}{2}\l[v^\lambda_{m;\lambda|n}+v^\lambda_{n;\lambda|m}\r]\nonumber\\+\frac{1}{2}\l[-(\box+\delta ) h_{(mn)}+r^l_m\ , h_{(ln)}+r^l_n h_{(lm)}+2r^{k\spc\spc\spc l}_{\spc mn}h_{(kl)}\r]\ , \label{eq : ein3}\\ \fl \delta s_{\mu\nu}=\l[2q\l(\frac{q-1}{p+q-2}c^2\!-\!\frac{p\!-\!1}{p\!-\!2}h^2\r)\phi\!+\!2(p\!-\!1)h^2\psi\!-\!\frac{2(q-1)}{p+q-2}c\delta b\r]\!\gamma_{\mu\nu}\nonumber\\+(p-1)h^2h_{(\mu\nu)}\label{eq : ein4}\\ \fl \delta s_{\mu n}=
c\l(b_{;\mu|n}+\delta b_{\mu n}- b^l_{\mu|nl}\r)+(p-1)h^2v_{\mu n } \label{eq : ein5}\\ \fl \delta s_{m n}=\l\{2\frac{p-1}{p+q-2}\delta cb+2\l[(p-1)h^2-\frac{(q-1)(p-2)}{p+q-2}c^2\r]\phi\r\}g_{mn}\nonumber\\+\l[c^2+(p-1)h^2\r]h_{(mn)}.\label{eq : ein6}\end{aligned}\ ] ] the linearized equations of motion of the form field are obtained from the expansion of equation ( [ eq : eom2 ] ) to first order in the perturbations .
the perturbed christoffel symbols can be calculated from the general formula @xmath105 .
again , the covariant derivative is either constructed from @xmath50 or @xmath106 depending on the coordinate that is differentiated with respect to .
one finds @xmath107\epsilon^{p_1\cdots p_q}\ , \label{eq : max1 } \\ \fl 0=\l({f_{\lambda\mu p_3\cdots p_q}}^{;\lambda}-{f_{\mu l p_3\cdots p_q}}^{|l}-c g^{mn}\delta\gamma^l_{\mu n}\epsilon_{m l p_3\cdots p_q}\r)\epsilon^{p_1\cdots p_q}\ , \label{eq : max2}\\ \fl\spc\spc\vdots\nonumber\\ \fl 0={f_{\lambda\mu_2\cdots\mu_q}}^{;\lambda}+{f_{l\mu_2\cdots \mu_q}}^{|l}\ .\end{aligned}\ ] ] it is convenient to factorize the @xmath46-dependence of the perturbations in various tensor harmonics of the sphere .
they satisfy orthogonality relations and therefore break the system of equations ( [ eq : ein1])([eq : ein6 ] ) into equations for each representation @xmath108 . like the factorization in the equations ( [ eq : hodgedecomp ] ) , we introduce @xmath109 where @xmath110 collectively stands for the perturbations @xmath111 and @xmath112 is a collective index over the eigenvalues of the corresponding tensor representation .
we note again that a pair of indices in parentheses indicates the symmetrized and traceless components of the corresponding second - rank tensor i.e . , @xmath113 .
additionally , the condition @xmath114 holds , which is compatible with the de donder gauge condition ( [ eq : dedonder ] ) . in the following ,
we suppress the summation symbol and the index @xmath112 .
first we calculate the mass spectrum for the scalar metric perturbations @xmath39 and @xmath88 that couple to the scalar form field perturbation @xmath89 . to find their equations of motion , we compare the coefficients in front of the spherical harmonics @xmath83 in the einstein equations ( [ eq : ein1])([eq : ein6 ] ) and the maxwell equations ( [ eq : max1])([eq : max2 ] ) .
the first of these equations is further simplified by taking the trace over @xmath115 and @xmath116 .
one obtains the following set of equations : @xmath117\phi+2{\frac{q-1}{p+q-2}}c\delta b\r.&\nonumber\\ \l.-\l[2{\frac{p-1}{p}}\box+\delta+2(p-1)h^2\r]\psi\r\}&\hspace{-3.5em}y\,\,\spc\spc\,\,=0\label{eq : scalar1}\\ \l\{{\frac{1}{2 } } h_{(\lambda\mu)}^{\spc\spc\spc;\lambda}+{\frac{p+q-2}{p-2}}\phi_{;\mu}-(p-1)\psi_{;\mu}-cb_{;\mu}\r\}&\hspace{-3.5em}y_{|n}\,\,\spc\,\,=0\label{eq : scalar2}\\ \l\{2\frac{p+q-2}{p-2}\phi - p\psi\r\}&\hspace{-3.5em}y_{|(mn)}=0\label{eq : scalar3}\\ \fl \l\ { \l(\box+\delta\r)\phi + 2\l[(p-1)h^2-{\frac{(p-2)(q-1)}{p+q-2}}c^2\r]\phi+2{\frac{p-1}{p+q-2}}c\delta b\r.&\nonumber\\ \l.+\frac{1}{q}\l[p\delta\psi-\frac{2(p+q-2)}{p-2}\delta\phi\r]\r\}&\hspace{-3.5em}y\,\spc\spc\,\,\,=0\label{eq : scalar4}\\ \l\{c^2\l[p\psi-2{\frac{q(p-1)}{p-2}}\phi\r]+c(\box+\delta)b\r\ } & \hspace{-3.5em}y_{|n}\,\,\spc\,\,=0\,.\label{eq : scalar5}\end{aligned}\ ] ] in what follows , the eigenvalues of the laplace operator @xmath118 are denoted by @xmath119 . for a @xmath2-sphere @xmath119 takes the values @xmath120 for @xmath121 [ i.e. , @xmath122 .
the set of equations ( [ eq : scalar1])([eq : scalar5 ] ) has to be analyzed for three different cases .
( i ) the general case , where @xmath123 and @xmath124 , in which all five equations ( [ eq : scalar1])([eq : scalar5 ] ) have to be satisfied .
( ii ) the second case is a peculiarity of highly symmetric compact manifolds , where @xmath123 but @xmath125 . for the @xmath2-sphere ,
the spherical harmonics with angular momentum @xmath126 ( i.e. , the first massive kaluza klein excitations ) satisfy this property .
( iii ) the last case deals with the homogeneous perturbations , for which both @xmath127 and @xmath128 .
let us first consider the general ( @xmath46-dependent ) case .
equation ( [ eq : scalar3 ] ) can be used to eliminate @xmath88 .
additionally , @xmath131 in equation ( [ eq : scalar1 ] ) is eliminated with equation ( [ eq : scalar2 ] ) .
then it is straightforward to show that the first equation is a linear combination of equations ( [ eq : scalar4 ] ) and ( [ eq : scalar5 ] ) , which are the dynamical equations for the perturbations @xmath89 and @xmath39 .
we define the following dimensionless quantities to shorten the notation : @xmath132=\l[{\frac{q(p-1)}{p+q-2}}c^2\rho^2\!-\!(q-1)\r],\nonumber\end{aligned}\ ] ] and denote the eigenvalues of the laplacian on the @xmath2-sphere with @xmath119 , @xmath133 , with @xmath134 .
finally one obtains the dynamical system @xmath135 next the system ( [ eq : scalar_eom ] ) is diagonalized .
the mass spectrum and mass eigenstates are given by @xmath136 \phi\ , \nonumber\\ x_-&\propto\,\,\,\,2{\frac{p-1}{p+q-2}}\lambda cb-\l[\sqrt{b^2+{\frac{4q(p-1)}{p+q-2}}a\lambda}-b\r ] \phi\ .\nonumber\end{aligned}\ ] ] for this case , the algebraic relation ( [ eq : scalar3 ] ) between @xmath39 and @xmath88 can not be imposed anymore .
again , @xmath140 is eliminated from the equations ( [ eq : scalar1 ] ) and ( [ eq : scalar2 ] ) .
@xmath88 is determined from the last equation ( [ eq : scalar5 ] ) .
the remaining two equations lead to identical equations of motion for the variable @xmath141 @xmath142 if the general spectrum ( [ eq : scalar_mass ] ) is specialized to the case @xmath143 , one finds that @xmath144 and @xmath145 . as argued in @xcite , the second mode @xmath146 that corresponds to the negative branch of ( [ eq : scalar_mass ] ) is gauged away by the enlarged residual gauge symmetry for the case @xmath143 ( @xmath126 ) , cf . [ sec : residual ] . in the homogeneous case ,
the residual gauge freedom is enhanced by the @xmath10-dimensional diffeomorphism invariance , @xmath149 .
this allows us to impose the transverse condition on @xmath42i.e . , @xmath150which is not valid for inhomogeneous modes as will be shown in sec .
[ sec : tensors ] .
this difference in the properties of the homogeneous and inhomogeneous graviton can be understood as follows : the homogeneous ( massless ) graviton has @xmath151 polarizations and its transverse and traceless polarizations completely decouple from the scalar and vector sector of perturbations .
the inhomogeneous ( massive ) graviton acquires @xmath152 longitudinal polarizations from the massive scalar and vector sector .
therefore , the physical massive graviton is a linear combination of @xmath42 and the scalar and vector modes .
the explicit nature of this combination depends on the choice of gauge . for our gauge
the physical massive graviton is given by @xmath153 that is introduced in equation ( [ eq : phys_graviton ] ) .
additionally , for the case @xmath154 , the scalar maxwell equation ( [ eq : scalar5 ] ) is a trivial identity .
therefore , the homogeneous component of @xmath89 is unphysical and can be set to zero .
the absence of a zero - mode perturbation of the form flux respects the quantized nature of @xmath4 .
higher multipole perturbations average to zero when integrated over the entire sphere ; therefore the total flux @xmath4 remains unchanged .
similarly , the equations ( [ eq : scalar2 ] ) and ( [ eq : scalar3 ] ) are trivial identities .
one additional constraint arises from the off - diagonal @xmath155 einstein equations .
the massless graviton only decouples from the scalar sector if @xmath156 . in the notation of the paper @xcite
, this constraint translates into @xmath157 , where the index ckp denotes the variables for scalar perturbations used in @xcite .
the remaining equation for the variable @xmath39 reduces to @xmath158\phi\ .\ ] ] this corresponds to the positive branch in the general formula ( [ eq : scalar_mass ] ) if @xmath159 and to the negative branch in the case when @xmath160 , giving a mass for the homogeneous mode @xmath39 : @xmath161 in the limit where the flux is turned off , we reproduce the result found in @xcite of a tachyonic mode @xmath162 for @xmath13 . remarkably , the mass is independent of the number of internal dimensions .
the appearance of a tachyonic mode in the spectrum of perturbations signals the instability of the background configuration . in what follows
, we collect all possible channels of instabilities .
it turns out that the tachyonic modes all reside in the scalar sector of the perturbations . in equation ( [ eq : radion ] ) , we have seen already the tachyonic nature of the homogeneous mode @xmath39 that is generic if the flux is below a threshold value @xmath164 or in other words if the inflation scale @xmath0 is higher than the flux @xmath4 that stabilizes the compactification . however , as the authors of @xcite showed , the homogeneous mode is not the only source for instabilities . if the negative branch of the general formula ( [ eq : scalar_mass ] ) is analyzed for the higher kaluza
klein excitations i.e .
, @xmath165one finds additional tachyons above a flux critical @xmath166 . in tab . [ tab : compare ] the threshold flux for the volume modulus , @xmath167 and the critical flux for the negative branch of @xmath168 are calculated for various numbers @xmath2 of extra dimensions and @xmath13 .
stable compactifications only exist for fluxes higher than the threshold flux @xmath169 and lower than the critical flux @xmath166 .
consequently , stable configurations are found for @xmath170 and @xmath171 extra dimensions for all fluxes larger than @xmath169 . in @xmath172 extra dimensions ,
stable compactifications are only achieved for a range of fluxes @xmath173 and no stable compactifications are possible for more than four extra dimensions .
the mass spectrum of the scalar perturbations @xmath39 and @xmath89 is summarized in fig .
[ fig : scalar_spectrum ] for the number of extra dimensions @xmath174 and @xmath175 .
the mass of the zero - mode ( denoted by @xmath176 kk ) is given by equation ( [ eq : radion ] ) , the mass for the special mode @xmath126 ( denoted by @xmath177 kk ) is obtained from equation ( [ eq : lambdagleichq ] ) and all the other excitations are graphical representations of the two branches @xmath178 in equation ( [ eq : scalar_mass ] ) . the threshold flux @xmath169 and the critical flux @xmath179 are indicated by the vertical lines .
the dependence on @xmath46 of the metric perturbations @xmath40 is factorized according to equation ( [ eq : tensordecomp ] ) .
the dynamics of the shape moduli is obtained from the coefficient in front of @xmath180 in the @xmath181 einstein equations . using the background equations ( [ eq : param_back ] )
, the equation simplifies to @xmath182 where @xmath183 are the eigenvalues of the laplace operator on symmetric and traceless tensors on @xmath184that is , @xmath185 with @xmath186 and @xmath187 . obviously , the sphere is stable against shape moduli perturbations .
the shape moduli fields have a positive kaluza
klein spectrum , which only depends on the geometry of the internal space .
it is not affected by the details of the de sitter geometry .
it is instructive to analyze the dynamics of the homogeneous scalar mode @xmath39 from the point of view of an effective four - dimensional theory beyond the linear approximation .
the analysis of perturbations suggests the following ansatz for the scalar mode @xmath39 : @xmath188 this particular choice of parameterization is compatible with the results for the perturbations , i.e. @xmath189 , and it ensures that the effective four - dimensional theory exhibits canonical einstein gravity .
the condition of flux quantization requires that @xmath4 is no longer a constant .
it is also promoted to a four - dimensional field @xmath190 the spacetime - dependent factor takes into account the change in the volume of the internal space such that the total flux remains constant .
the background maxwell equations ( [ eq : eom2 ] ) are still trivially satisfied . with the equations ( [ eq : eff1 ] ) and ( [ eq : eff2 ] ) the action of the system ( [ eq : action ] ) takes the following form @xmath191\r.\!&\!-\frac{q(p+q-2)}{p-2}(\nabla\phi)^2\\ & \l.+r[g]e^{-2\frac{p+q-2}{p-2}\phi}\!-\!2\lambda e^{-2\frac{q}{p - q}\phi}\!-\!c_0 ^ 2e^{-2q\frac{p-1}{p-2}\phi}\r\}\ , \nonumber\end{aligned}\ ] ] where we neglected a boundary term proportional to @xmath192 . @xmath193 $ ] is the four - dimensional ricci scalar and @xmath194=q(q-1)\rho^{-2}$ ] is the ricci scalar constructed with the metric @xmath195 . after integrating out the extra dimensions and rescaling @xmath196 we obtain the effective potential @xmath197 for the canonically normalized field @xmath39 : @xmath198 the stability of the background solution is determined by the shape of this potential at @xmath199 .
the parameters @xmath15 , @xmath200 , @xmath0 and @xmath9 are related by the background equations of motion ( [ eq : param_back ] ) . if eqs .
( [ eq : param_back ] ) are taken into account then the potential has the following properties at @xmath199 @xmath201 which is equivalent to the results obtained in eq .
( [ eq : radion ] ) .
however , the effective potential ( [ eq : eff4 ] ) allows us to analyze the dynamics beyond the stability of the background solution . in fig .
[ fig : eff ] we illustrate the effective potential for various values of the flux @xmath200 .
it is clearly visible that in the case of vanishing flux , i.e. @xmath202 , the background solution corresponds to an unstable maximum .
it can either evolve towards @xmath203 , which corresponds to the collapse of the internal manifold , or towards @xmath204 , which corresponds to the decompactification of the internal space , cf .
@xcite . for flux values @xmath205 a second minimum occurs at a finite but smaller size of the internal space .
as analyzed in @xcite , this corresponds to transitions of the system towards a stable anti - de sitter or de sitter configuration . for flux values of @xmath206
a minimum at @xmath199 is formed and the background solution is classically stable .
@xmath207 , @xmath208 of the mode @xmath39 for various values of fluxes . in the absence of fluxes @xmath202
, the background solution is unstable . for fluxes
@xmath205 a minimum is formed , to which the system can evolve , and for @xmath206 the background solution is stable.,scaledwidth=50.0% ] the mass spectrum for vector perturbations of the effective @xmath10-dimensional theory is obtained from the equations of motion for the vector perturbations @xmath103 and @xmath41 .
they appear as coefficients in front of the vector harmonics @xmath209 .
the system of vector perturbations needs to be diagonalized to find the spectrum and the mass eigenstates .
they are linear combinations of the metric vector perturbations @xmath41 and the form flux vector perturbations @xmath103 .
we list the equations of motion for the vector modes obtained from the einstein and maxwell equations ( [ eq : ein1])([eq : max2 ] ) .
the form field is expressed in terms of its gauge potential ( [ eq : potential ] ) .
the gauge potential is further decomposed according to the equations ( [ eq : hodgedecomp ] ) and the relations ( [ eq : formulae ] ) are used for simplification @xmath210\delta^m_n+(v_{\mu}+2cb_\nu)\max\r\}y_{m}&=0\ , \label{eq : vector1}\\ \l\{c^2 { v^{\lambda}}_{;\lambda}\delta^m_n-\max c{b^{\lambda}}_{;\lambda}\r\}y_{m}&=0\ , \label{eq : vector2}\\ \l\{\l[-c\max b_{\mu}+c^2 v_{\nu}\r]\delta^m_n - c\max b_{\nu}\r\}y_m&=0\ , \label{eq : vector3}\end{aligned}\ ] ] where @xmath211 represents the ordinary maxwell operator and similarly @xmath212 , which up to a sign equals to the action of the laplace
beltrami operator @xmath213 on the one - forms @xmath214 .
the second equation ( [ eq : vector2 ] ) follows from the third one ( [ eq : vector3 ] ) by differentiation .
the two dynamical equations ( [ eq : vector1 ] ) and ( [ eq : vector3 ] ) for the physical vector modes @xmath41 and @xmath103 can be recast compactly @xmath215{c c } \kappa^i-2(p-1)h^2 & 2 \kappa^i\rho^{-2}\\ c^2 & \kappa^i\rho^{-2 } \end{array}\!\r)\!\!\l(\ !
\begin{array}[c]{c } v^i_\nu\\ c b^i_\nu \end{array}\!\r)\!y^i_n , \end{aligned}\ ] ] where @xmath216 are the eigenvalues of the laplace beltrami operator , @xmath217 , that take values in @xmath218 for @xmath219 .
the eigenvalues of the mass matrix are given by @xmath220 ^ 2 + 2\kappa c^2\rho^2}\ , \label{eq : vec_spectrum}\\ \fl \frac{m_{\pm}^2}{h^2}&=\frac{\kappa}{q-1}\l[\frac{c^2}{h^2}+(p-1)\r]-(p-1)\pm\sqrt{(p-1)^2 + 2\kappa\frac{p-1}{q-1}\frac{c^2}{h^2}+\frac{2\kappa}{q-1}\frac{c^4}{h^4}}\ , \nonumber\end{aligned}\ ] ] like in the case of scalar perturbations , the set of eigenstates @xmath221 that corresponds to the eigenvalues given in equation ( [ eq : vec_spectrum ] ) consists of a linear combination of the metric perturbations @xmath41 and the flux perturbation @xmath103 : @xmath222 where the equations are to be understood for each eigenfunction @xmath223 separately with corresponding values of @xmath216 and @xmath224 . it can be shown @xcite that the values of @xmath225 are bound from below by @xmath226 .
the bound is saturated when @xmath227i.e .
, for the killing vectors of the compact space . together with this bound , it follows immediately from equation ( [ eq : vec_spectrum ] ) that all the mass squares in the spectrum are non - negative . therefore , de sitter compactifications are stable with respect to vector modes , or in physical terms , the de sitter space is stable against the development of anisotropies .
the spectrum consists of two kaluza
klein towers , whose scaling depends on two quantities : the size @xmath9 of the extra dimension and the field strength of the flux @xmath4 .
it is of interest to investigate the nature of the states with the lowest eigenvalues and the spectrum in the limit of vanishing flux . in the limit when the form flux is switched off , @xmath29
, its perturbation @xmath228 also vanishes . only for the negative branch
do the coefficients of the mass eigenstates in equation ( [ eq : mass_d ] ) remain regular and the limit can be performed : @xmath229 .
the spectrum ( [ eq : vec_spectrum ] ) reduces to @xmath230 the positive branch originates from the presence of the form field fluctuations .
it introduces a new scale @xmath4 , but it does not affect the lowest mass modes , which are in either case obtained from the negative branch .
one obtains @xmath231 for @xmath232 and all values of @xmath4 .
the lowest vector modes are massless and correspond to the isometries of the compact spacetime .
the spectrum of vector modes is summarized in fig .
[ fig : vector_spectrum ] . in the absence of flux
all remaining three parameters of the background model @xmath0 , @xmath15 and @xmath9 are related to each other by factors of order unity , cf .
equations ( [ eq : param_back ] ) .
the flux introduces a new scale and enables us to create a hierarchy between @xmath0 and the other parameters .
when @xmath4 is very close to its maximal value @xmath233 , the number @xmath234 is very small .
this creates a hierarchy in a direction that is phenomenologically interesting with @xmath235 . ' '' '' + ' '' '' + [ fig : vector_tensor ] we show in this subsection that the homogeneous , @xmath46-independent component of the graviton @xmath42 reduces to the ordinary @xmath10-dimensional de sitter graviton .
the details about the wave functions and representations can be found in @xcite . for the homogeneous mode all
@xmath46-derivatives vanish . then for the weyl - shifted metric perturbations @xmath236 and @xmath237 the left - hand side of the einstein equations is obtained from equations ( [ eq : dricci ] ) @xmath238 similarly , the homogeneous contribution of the perturbations to the @xmath155-components of the energy
momentum tensor follows from equation ( [ eq : ein4 ] ) : @xmath239 the terms in @xmath39 of expressions ( [ eq : graviton_lhs_hom ] ) and ( [ eq : graviton_rhs_hom ] ) cancel by virtue of the equations of motion ( [ eq : zeromode ] ) for the zero - mode of @xmath39 .
commuting the derivatives in the expression ( [ eq : graviton_lhs_hom ] ) and imposing transverse and traceless conditions on @xmath240 , equations ( [ eq : graviton_lhs_hom ] ) and ( [ eq : graviton_rhs_hom ] ) reduce to the ordinary equation of motion for the massless graviton in de sitter space with an apparent mass of @xmath241 : @xmath242 slightly more involved is the identification of the massive gravitons .
the starting point is the traceless component of the @xmath155 einstein equations : @xmath243\!+h^2 h_{(\mu\nu)}-\frac{1}{p } h_{(\kappa\lambda)}^{\spc\spc\spc;\kappa\lambda}\gamma_{\mu\nu}-(p-2)\psi_{;(\mu\nu)}\ .\end{aligned}\ ] ] next @xmath244 , @xmath140 and @xmath88 are eliminated with the scalar equations ( [ eq : scalar1])([eq : scalar5 ] ) .
one finds @xmath245 the physical graviton is transverse and traceless .
following the ansatz of @xcite , we construct the physical graviton @xmath153 as follows : @xmath246 where @xmath247 and @xmath248 are arbitrary constants , that are determined from equation ( [ eq : reduced_tracefree ] ) and the conditions @xmath249 the value of the constants @xmath247 and @xmath248 depends on the eigenvalue of the laplacian @xmath118 acting on the scalar functions @xmath250 .
they are given by @xmath251}\ , \nonumber\\ v=&\frac{2(d-2)}{(p-1)\l[\lambda \rho^{-2}-(p-2)h^2\r]}\ , \end{aligned}\ ] ] with the eigenvalues of the laplacian @xmath119 that can take values @xmath252 for @xmath253 . from equation ( [ eq : transverse ] ) the spectrum of the graviton
is obtained . neglecting the apparent mass shift of @xmath241 from the de sitter space ,
one obtains a simple kaluza klein spectrum for the graviton modes , determined by the geometry of the internal space and unaffected by the bulk matter fields ( in our case the form flux @xmath254 ) : @xmath255 for comparison , the lowest kaluza klein excitations of the gravitational waves are plotted in fig .
[ fig : tensor ] .
in this section we analyze two possible consequences for the effective four - dimensional cosmology that result from the properties of the mass spectrum of the scalar , vector and tensor perturbations that we calculated in sec .
[ sec : perturbations ] .
( i ) during inflation an almost scale invariant spectrum of perturbations is generated for modes whose mass is smaller than the scale of inflation @xmath0 .
these perturbations contribute to the later evolution of the universe .
( ii ) the perturbations gravitationally couple to the standard model fields .
we assume that the standard model fields are realized as zero - modes of the corresponding higher dimensional fields and calculate the coupling of the scalar perturbation to the standard model fields . to estimate the dynamics of the perturbations during inflation it is important to know their masses in comparison to the expansion rate @xmath0 .
the perturbations are phenomenologically interesting when their mass is smaller than the @xmath0 .
figure [ fig : masses_h ] shows the mass spectrum of the scalar , vector and tensor perturbations in units of the expansion rate @xmath0 for a ten - dimensional spacetime compactified on a six - dimensional sphere . from the spectrum for the scalar perturbations in fig .
[ fig : masses_h ] we see immediately that , besides the volume modulus , higher kaluza
klein excitations can also have masses smaller than the expansion rate @xmath0 for certain ranges of the form field strength @xmath4 . during inflation
an almost scale invariant spectrum of perturbations will be generated for these modes .
they therefore contribute as cosmological perturbations to the evolution of the universe .
in particular , we note the different nature of the volume modulus and the higher kaluza
klein excitations with respect to their coupling to the standard model fields , which is discussed below , cf . sec .
[ sec : coupling ] . apart from the massless vector fields that are associated with the killing vector fields on the @xmath2-sphere , all vector modes have masses larger than the inflationary scale @xmath0 .
consequently , the kaluza klein modes of vector perturbations are not excited during inflation and should not play an important role in the subsequent cosmological evolution .
the massless vector fields are conformal and are not excited during inflation either . moreover
, they will disappear from the spectrum , once compactifications to more realistic internal spaces such as calabi
yau manifolds are considered .
similarly , the kaluza klein excitations of tensor modes have masses above the scale of inflation and do not contribute to the dynamics of cosmological perturbation after inflation .
the massless tensor mode corresponds to the ordinary four - dimensional graviton .
it carries no information about the extra dimensional nature of the full spacetime .
similar conclusions were derived for braneworld models in @xcite . of the inflationary geometry for @xmath13 and @xmath256 . unlike the case in figs .
[ fig : scalar_spectrum][fig : vector_tensor ] , there is no limiting flux @xmath233 in this units .
the flux @xmath4 can be arbitrarily larger than the expansions rate @xmath0 . ]
next , we briefly discuss the nature of the coupling of the scalar perturbations to the standard model fields . as a consequence of their different nature , the zero - modes and higher kaluza
klein excitations couple differently to the standard model fields .
when the standard model fields are localized on a brane , it is known that the radion universally couples to the trace of the energy
momentum tensor of the brane degrees of freedom : @xmath257 however , in kaluza
klein compactifications one does not need to assume that the standard model fields are localized . instead
, the standard model fields of the effective four - dimensional theory simply correspond to the zero - modes of the higher dimensional fields .
this point of view modifies the coupling of the volume modulus to the standard model zero - modes and equation ( [ eq : brane_coupling ] ) is not applicable .
we start with the action ( [ eq : action ] ) and add the standard model lagrangian as a collection of zero - modes to the full higher dimensional theory : @xmath258 for simplicity , we only analyze a canonical scalar field @xmath259 : @xmath260 focusing on the scalar perturbations @xmath38 and @xmath39 introduced in the line element ( [ eq : line ] ) , the action ( [ eq : sm_setup ] ) is expanded to first order : @xmath261+\cdots\r.\nonumber\\\l.-\frac{1}{2}\l(1\!-\!2\hat \psi\r)(\partial_\mu \chi)^2-\frac{1}{2}m^2\chi^2\r\ } \ , \end{aligned}\ ] ] where the dots collect the terms from the form fields , the cosmological constant and the curvature of the internal space . next the weyl shift @xmath262 , cf.equation ( [ eq : weylshift ] ) ,
is performed to obtain the linearized einstein frame and the extra dimensions are integrated out : @xmath263\l[\frac{1}{m_\text{p}^2}r-\l(\partial_\mu\chi\r)^2\r]+\cdots-\frac{1}{2}\l(1 + 4\psi - q\phi\r)m^2\chi^2\r\}\ , \end{aligned}\ ] ] where the four - dimensional planck mass @xmath264 arises from the rescaling of the fundamental scale with the volume of the internal space .
similarly the field @xmath265 is rescaled by the volume of the internal space @xmath266 .
the term @xmath267 is the reason that the coupling of the volume modulus to the standard model fields deviates from the form in equation ( [ eq : brane_coupling ] ) .
apart from this term the action ( [ eq : effective4d_general ] ) corresponds to a four - dimensional theory of gravity and a canonical scalar field @xmath265 in a de sitter geometry with metric fluctuations of the form @xmath268 .
we now discuss how the scalar perturbations interact with the standard model fields for the two different cases of a homogeneous ( volume modulus ) and inhomogeneous kaluza
klein excitations .
the result of the zero - mode for scalar perturbations derived in sec .
[ sec : scalars ] was particularly simple : @xmath269 therefore , we obtain the effective four - dimensional action for the zero - mode : @xmath270-\frac{1}{2 } ( m^2\chi^2+{\cal n}m^2_\phi \phi^2)+\frac{qm^2}{2m_\text{p}}\phi\chi^2\r\}\ , \nonumber\\\end{aligned}\ ] ] where @xmath39 is rescaled to its canonical form up to a constant @xmath271 of order unity .
the effective action shows the volume modulus as a canonical scalar field with mass @xmath272 that couples to the standard model scalar fields through the interaction term @xmath273 and the planck suppressed coupling @xmath274 .
in particular , it has no first - order coupling at all to massless ( conformal ) fields .
now one can use equation ( [ eq : scalar3 ] ) to eliminate @xmath88 .
the following four - dimensional interactions are obtained from the effective action ( [ eq : effective4d_general ] ) : @xmath275\!+\!\frac{\phi}{m_\text{p}}m^2\chi^2\r\}\ .\end{aligned}\ ] ] however , one has to take into account that @xmath39 is not the physical dynamical variable for the massive modes .
the correct kinetic terms and normalizations have to be found for the mass eigenstates calculated in equation ( [ eq : scalar_mass ] ) , which are linear combinations of the scalar perturbation @xmath39 and the scalar matter degrees of freedom , in our case the scalar component of the form flux @xmath89 .
qualitatively , the interactions in equation ( [ eq : interactions ] ) show the direct but planck mass suppressed decay channel of massive kaluza klein states into gravitons and standard model fields . alternatively ,
klein excitations of the scalar perturbation @xmath39 around its expectation value lead to variations of the planck mass and the masses of the standard model fields .
in this paper , we systematically analyzed the stability properties of de sitter compactifications with @xmath2-form fluxes .
we calculated in a unified way the complete perturbative mass spectrum of de sitter compactifications .
the most important feature of the perturbative spectrum is the appearance of tachyonic modes in the spectrum of scalar perturbations .
the remaining masses of the vector and tensor perturbations of the spectrum are non - negative and therefore do not create additional instabilities . the tachyonic mode of the volume modulus , possible ways of stabilizing it , and implications for inflation have been discussed before in @xcite .
its occurrence imposes tight constraints on the maximal scale of inflation .
it also implies that the volume of the internal space is stabilized at least at the scale of inflation to ensure a stable background configuration and a sufficiently large number of efolds during inflation .
if the stabilization of the volume modulus remains at such a high scale after inflation , the modulus is too heavy to be detected in future accelerator experiments .
the tachyonic nature of the unstabilized modulus is not related to special properties of the internal space .
it merely reflects the presence of the inflationary geometry with constant expansion rate @xmath0 in the effective four - dimensional theory .
further tachyonic modes that can arise from the quadrupole and higher kaluza
klein excitations are less explored .
non - perturbative dynamics that are triggered by these instabilities are not known .
a possible stabilization due to additional matter fields is far from obvious , since the instability originates from the presence of the form flux .
we therefore expect similar obstructions caused by this instability in more general compactifications . figure [ fig : radion_mass ] shows the contour plots for the value of the smallest mass of scalar perturbations as a function of the hubble scale @xmath0 and form flux @xmath4 .
all quantities are plotted in units of the fundamental mass scale @xmath276 . for two or three extra dimensions
the lowest mass is by and large determined by the volume modulus , cf . equation ( [ eq : radion ] ) . for four and more extra dimensions the negative branch in equation ( [ eq : scalar_mass ] ) of the higher kaluza
klein excitations becomes tachyonic for a sufficiently large contribution from the form flux . for four extra dimensions only a small range of parameters
@xmath0 and @xmath4 admits stable compactifications . for more than four extra dimensions no stable compactifications
are found at all in this model .
the solid black line in fig . [ fig : radion_mass ] encircles the region where the size of the extra dimensions @xmath277 .
it is called the sugra limit , since for smaller values of @xmath9i.e . , for parameters
@xmath0 and @xmath4 outside the encircled region
quantum gravity corrections are expected to contribute .
( in fundamental units ) .
the blue ( red ) contour lines indicate regions of ( un)stable compactifications . in the left panel
the dashed line represents the parameters for which the mass of the volume modulus vanishes . in the right panel
the upper dashed line indicates the zeros for the mass of the volume modulus and the lower dashed line shows where the mass of the second kk mode vanishes .
stable compactifications are only possible for parameters in between the two dashed lines .
the solid black line encircles the region in which the size of the extra dimensions is larger than @xmath278 fundamental units of lengths ( i.e. , the valid region for the supergravity approximation).,title="fig : " ] + ( in fundamental units ) .
the blue ( red ) contour lines indicate regions of ( un)stable compactifications . in the left panel
the dashed line represents the parameters for which the mass of the volume modulus vanishes . in the right panel
the upper dashed line indicates the zeros for the mass of the volume modulus and the lower dashed line shows where the mass of the second kk mode vanishes .
stable compactifications are only possible for parameters in between the two dashed lines .
the solid black line encircles the region in which the size of the extra dimensions is larger than @xmath278 fundamental units of lengths ( i.e. , the valid region for the supergravity approximation).,title="fig : " ] + we investigated the calculated spectrum of scalar , vector and tensor perturbations for possible phenomenological consequences . for certain ranges of the form field strength @xmath4
, the scalar sector provides light modes that generate an almost scale - invariant spectrum of fluctuations during inflation .
these perturbations decay gravitationally into standard model fields after inflation .
the details of the decay are different for the volume modulus and the higher kaluza
klein excitations .
the spectrum of vector perturbations ( [ eq : vec_spectrum ] ) does not contain tachyonic modes .
moreover , apart form the massless gauge fields that correspond to the killing vector fields on the @xmath2-sphere , all vectors have a mass larger than the scale of inflation , @xmath279 . furthermore
, the vector modes do not have a direct decay channel into standard model fields , since the only conceivable first - order coupling to the standard model fields of the form @xmath280 can be integrated by parts and vanishes due to the properties of the energy momentum tensor .
as expected , the spectrum of tensor perturbations does not depend on the details of the matter field content .
it only measures the geometry of the compactification .
the zero - mode does not feel the presence of the extra dimensions and behaves like an ordinary graviton in the de sitter geometry .
the kaluza
klein excitations of the graviton only depend on the size and the shape of the extra dimensions .
again , the massive modes have masses larger than the scale of inflation , @xmath281 . therefore , kaluza klein excitations of tensor modes will not be generated during inflation .
possible existing excitations from physics before inflation are erased during the process of inflation .
consequently , the massive gravitons are not expected to play a role for cosmological observations . in the final remark
we want to compare this spectrum to the spectrum of anti - de sitter compactifications calculated in @xcite .
the main difference in the set up between the two compactifications is the sign in the curvature of the large dimensions and the additional bulk cosmological constant .
however , many features do not change qualitatively .
most importantly , the mixing between matter and higher dimensional metric degrees of freedom in the scalar and vector sector is common to both compactifications .
it is a well known phenomenon with importance also for the ordinary theory of inflation @xcite and braneworld scenarios .
the mixing between the higher dimensional metric perturbations and matter fields in braneworlds was investigated in detail in @xcite .
even the appearance of modes with negative mass square is common to both supersymmetric and non - supersymmetric compactifications . however , in the case of anti - de sitter compactifications the stability against perturbations with tachyonic mass is ensured by the breitenlohner - freedman bound @xcite , as long as their mass is larger than this bound .
this paper is part of my ph.d .
thesis about `` cosmology with extra dimensions '' .
i am particularly grateful to my supervisor lev kofman for guidance .
additionally , i thank marco peloso , erich poppitz and pascal vaudrevange for useful discussions . support from the connaught scholarship
is acknowledged .
[ cols= " < , < " , ]
to understand the redefinition of the scalar perturbation @xmath38 in equation ( [ eq : weylshift ] ) , we consider the effective @xmath10-dimensional theory that is obtained from the compactification of a @xmath6-dimensional theory with a choice of metric of the form @xmath282 for the zero - mode of @xmath39 and this choice of metric the @xmath6-dimensional action of gravity reduces to @xmath283.\ ] ] to obtain a canonically normalized four dimensional graviton or , in other words , a four - dimensional einstein theory of gravity , the metric has to be rescaled by a weyl transformation : @xmath284 if the field @xmath39 is treated as a perturbation as in equation ( [ eq : line ] ) , the above weyl transformation ( [ eq : weyl_trafo ] ) of the metric @xmath50 amounts to the weyl shift ( [ eq : weylshift ] ) of the field @xmath285 .
in this appendix we list useful formulae that have been used to simplify the equations of motion .
they are entirely based on the properties of the antisymmetric epsilon tensor and are straightforward to derive : @xmath286
as mentioned in sec . [ sec : class_gauge ] , the de donder gauge conditions ( [ eq : dedonder ] ) do not fix the gauge freedom associated with the infinitesimal coordinate transformations ( [ eq : inf_coord ] ) completely .
consequently , there will be modes in the spectrum of perturbations that do not correspond to physical degrees of freedom . in this section
, we analyze the nature of these residual gauge degrees and impose additional constraints to eliminate them . where we split the @xmath46-independent diffeomorphisms @xmath289 into a transverse vector @xmath290 and a scalar function @xmath291 . similarly , we decompose the traceless part of homogeneous metric perturbations : @xmath292 into the transverse - traceless polarizations @xmath293 , the transverse vector @xmath294 and the scalar function @xmath295 . correspondingly , the homogeneous off - diagonal components of the metric perturbations are decomposed : @xmath296 into transverse and longitudinal polarizations . from the standard transformation of the homogeneous metric perturbations under infinitesimal transformations
@xmath297 with @xmath298 denoting the lie derivative in direction of @xmath56 , follows that the additional gauge constraints can be imposed : @xmath299 these additional constraints ensure that the homogeneous graviton @xmath42 only contains transverse - traceless polarizations and that the homogeneous metric components @xmath41 are transverse vectors . @xmath300 $ ] are another set of solutions @xmath301 , @xmath302 to the equations ( [ eq : app_residual ] ) .
this gauge symmetry is not removed by further constraints .
it remains a symmetry of the effective theory , provided that the massless vector fields @xmath41 transform in the adjoint representation of the isometry group of the @xmath2-sphere , which is generated by the killing vectors @xmath303 .
are generated by the scalar harmonics on the @xmath2-sphere @xmath83 with @xmath126 ( i.e. , the scalar harmonics that correspond to the first level of kaluza
klein excitations ) .
they satisfy the additional constraint @xmath304 .
the solution to the equations ( [ eq : app_residual ] ) is given by @xmath305\ .\ ] ] this residual symmetry is the reason why the negative branch of the spectrum ( [ eq : scalar_mass ] ) does not contribute to the physical degrees of freedom of the first excited kaluza klein modes ( i.e. , @xmath126 ) @xcite .
j. m. maldacena and c. nunez , _ supergravity description of field theories on curved manifolds and a no go theorem _ , _ int . j. mod .
phys . _ * a16 * ( 2001 ) 822855 , [ http://xxx.lanl.gov/abs/hep-th/0007018 [ hep - th/0007018 ] ] .
r. bousso , o. dewolfe , and r. c. myers , _ unbounded entropy in spacetimes with positive cosmological constant _ , _ found .
phys . _ * 33 * ( 2003 ) 297321 , [ http://xxx.lanl.gov/abs/hep-th/0205080 [ hep - th/0205080 ] ] .
l. kofman , j. martin , and m. peloso , _ exact identification of the radion and its coupling to the observable sector _ , _ phys .
_ * d70 * ( 2004 ) 085015 , [ http://xxx.lanl.gov/abs/hep-ph/0401189 [ hep - ph/0401189 ] ] .
o. dewolfe , d. z. freedman , s. s. gubser , g. t. horowitz , and i. mitra , _ stability of @xmath309 compactifications without supersymmetry _ , _ phys .
rev . _ * d65 * ( 2002 ) 064033 , [ http://xxx.lanl.gov/abs/hep-th/0105047[hep-th/0105047 ] ] .
v. f. mukhanov , h. a. feldman , and r. h. brandenberger , _ theory of cosmological perturbations .
classical perturbations .
quantum theory of perturbations .
part 3 . extensions _ , _ phys .
* 215 * ( 1992 ) 203333 . | kaluza
klein compactifications with four - dimensional inflationary geometry combine the attractive idea of higher dimensional models with an attempt to incorporate four - dimensional early - time or late - time cosmology .
we analyze the mass spectrum of cosmological perturbations around such compactifications , including the scalar , vector and tensor sector . whereas scalar perturbations were discussed before ,
the spectrum of vector and tensor perturbations is a new result of this paper .
moreover , the complete analysis shows that possible instabilities of such compactifications are restricted to the scalar sector .
the mass squares of the vector and tensor perturbations are all non - negative .
we discuss form fields with a non - trivial background flux in the extra space as matter degrees of freedom .
they provide a source of scalar and vector perturbations in the effective four - dimensional theory .
we analyze the perturbations in freund - rubin compactifications .
although it can only be considered as a toy model , we expect the results to qualitatively generalize to similar configurations .
we find that there are two possible channels of instabilities in the scalar sector of perturbations , whose stabilization has to be addressed in any cosmological model that incorporates extra dimensions and form fields .
one of the instabilities is associated with the perturbations of the form field . |
beginning with the discovery of sn 1992bi ( perlmutter et al 1995 ) , we have developed search techniques and rapid analysis methods that allow systematic discovery and follow up of `` batches '' of high - redshift supernovae . the observing strategy developed compares large numbers of galaxies in each of @xmath350 fields observed twice with a separation of @xmath33 weeks .
this search schedule makes it possible to precisely calculate the `` control time , '' the effective time during which the survey is sensitive to a type ia event . for this analysis ,
we have studied a set of 52 similar search fields observed in december 1993 and january 1994 .
the data were obtained using the `` thick '' @xmath4 eev5 camera at the 2.5 m isaac newton telescope , la palma .
the projected pixel size is @xmath5 , giving an image area of approximately @xmath6 .
exposure times were 600s in the @xmath7 filter , and the images typically reach a @xmath8 limit of @xmath9mag . seeing was typically around @xmath10 .
many of the fields were selected due to the presence of a high - redshift cluster ( @xmath11 ) .
suitable clusters and their redshifts were taken from gunn , hoessel & oke ( 1986 ) and from the rosat catalog .
the total useful area covered in this study is 1.73 sq deg .
the analysis procedure and method for finding sne can be summarized as follows . for most fields , two first - look `` reference '' images were obtained and for all fields two second look `` search '' images were obtained @xmath12 weeks after the reference images .
the images were flat - fielded and zero - points for the images were estimated by comparison with e ( red ) magnitudes of stars in the apm ( automated plate measuring facility , cambridge , uk ) possi catalog ( mcmahon & irwin 1992 ) .
the search images were combined ( after convolution to match the seeing of the worst of the four images ) and the combined reference images were subtracted from this after scaling in intensity .
the resulting difference image for each field was searched for sn candidates . in this subset of the search data
three sne were found with redshifts 0.354 , 0.375 and 0.420 determined from spectra of the host galaxies .
for the purposes of this analysis , we assume these are all type ia sne ( see perlmutter et al .
the method used to calculate the rate can be divided into two main parts : ( i ) estimation of the number of galaxies and the total stellar luminosity ( measured in @xmath13 ) to which the survey is sensitive and ( ii ) estimation of the sn detection efficiency and hence the control time .
in order to compare the distant sn rate with local equivalents , we need to know the redshifts of the galaxies we have surveyed . in this work we use the galaxy counts derived from the analysis of lilly et al ( 1995 ) to estimate the number of galaxies sampled as a function of redshift .
@xmath14 band counts as a function of redshift were calculated by lilly based on the analysis of magnitude
redshift data obtained in the canada - france redshift survey ( lilly et al , 1995 and references therein ) . to check that the assumed distribution of galaxies with @xmath14 magnitude and redshift , @xmath15 , give reasonable galaxy counts compared to our data ,
we have plotted the number of field galaxies classified by the focas software package , as a function of apparent magnitude , on one of the search images that was _ not _ targeted at a galaxy cluster .
the r - band galaxy counts given by the analysis of lilly et al ( 1995 ) integrated over the redshift range @xmath16 ( dash - dotted line ) are shown on the same scale , assuming an effective area for this image of 0.03 sq deg ( figure [ nofm ] ) .
many of our search fields were chosen specifically to target high - redshift clusters .
for each of these fields , we estimate the number of cluster galaxies by counting galaxies as a function of @xmath14 magnitude in a box of size @xmath17 pixels centered approximately on the center of the cluster as estimated by eye from the images .
the counts in a similar box in a region of the image away from the cluster were subtracted from the cluster counts to give the cluster excess counts as a function of @xmath14 mag .
typically a cluster contributes 10% of the galaxy counts on an image .
we assign these galaxies to the cluster redshift , and add the cluster contribution to the @xmath15 for that image given by the models .
values of reddening due to the galaxy were supplied by d. burstein ( derived from burstein & heiles , 1982 ) for each field and applied to the data assuming @xmath18 and @xmath19 ( cardelli et al , 1989 ) .
to calculate the control times we assumed that sn magnitude as a function of time follows the average of the best - fit , time - dilated and @xmath20-corrected leibundgut ( 1988 ) type ia template light curves ( the generalized @xmath20 correction described by kim et al .
( 1996 ) and kim , goobar & perlmutter ( 1995 ) , was used here ) .
the control time is then given by the weighted sum of days during which the sn can be detected , where the weighting is by the detection efficiency , @xmath21 . this detection efficiency is a function of the magnitude difference @xmath22 ( zero if @xmath22 is negative ) , which is itself a function of time @xmath23 relative to maximum , and @xmath24 , the time separation of the search and reference images .
it depends on the quality of the subtracted images ( seeing , transmission ) together with the detailed technique ( convolution , selection criteria ) used to extract the signal ( sne candidates ) from the background ( cosmic rays , asteroids , bad subtractions , etc ) .
in addition , there is a slight dependence on the host galaxy magnitude .
the detection efficiency was calculated using a monte - carlo method .
a synthetic image was created for every field by adding simulated supernovae to the search images using point - spread - functions drawn from each search image .
the reference images were subtracted from the synthetic search images using exactly the same software as used for the supernova search and the number of simulated sne that satisfy the selection criteria was determined .
this technique allows us to measure detection efficiencies as a function of supernovae magnitude individually for every field , thus taking into account the other parameters mentioned above .
figure [ eff](a)-(c ) shows the fractional number of simulated sn recovered , as a function of sn magnitude ( at detection ) for the three fields in which sne were found .
figure [ eff](d ) shows the efficiency as a function of relative surface brightnesses of the sn and host galaxy .
this last parameter gives an indication of the effect of sn location with respect to the host galaxy .
although this is a small effect , it was taken into account . for a typical field the detection efficiency is over 85% for any added fake stellar object brighter than @xmath25 magnitude ( note that the more recent searches of this project have worked with significantly deeper images ) .
before calculating the luminosity - weighted sn rate , we first determine a `` survey rate '' of sn discoveries that a search for ia sne can expect to obtain , per square degree .
we give the rate in a range of 1 mag in @xmath14 , centered on the mean peak @xmath14 magnitude of the 3 sne found in this search , @xmath26 .
the survey rate is given by @xmath27 where @xmath28 is the number of sne we found in the 1 magnitude range , and @xmath29 is the control time for field @xmath30 , computed for a sn with magnitude @xmath26 at maximum . for example
a value of @xmath31 days was found for the field containing sn1994h observed on 1993 december 19 and 1994 january 12 , 24 days apart .
we measure a survey rate for @xmath32 of @xmath33 sne @xmath34 ( the error quoted is statistical only ) . in practice
this translates to 1.73 sne per square degree discovered with a 3 week baseline , in data with limiting magnitude @xmath35 ( @xmath36 for @xmath14=23 ) . to calculate the rate in snu ,
we compute the expected redshift distribution of sne , @xmath37 , which is proportional to the observed sn rate , @xmath38 , where @xmath39 is the rate in the rest - frame of the supernovae .
it is given by @xmath40 where @xmath30 runs over all fields , @xmath14 is the galaxy apparent @xmath14 magnitude and @xmath41 is the galaxy rest - frame b band luminosity in units of @xmath13 .
the control times @xmath42 , in units of centuries , have been calculated for each field in bins of @xmath43 and @xmath14 ( the size of the bins is 0.5 mag in @xmath14 , 0.05 in @xmath43 ) . to compute the rest - frame b band galaxy luminosities from apparent @xmath14 magnitudes , we used @xmath44 colors and @xmath45-band @xmath20 corrections supplied by gronwall & koo .
note that the combined color , @xmath20 and evolution correction is small in the redshift range of interest ( @xmath46 ) since the observed @xmath14 band is close to the rest - frame b - band . in this calculation @xmath47 and @xmath48 were assumed .
the rest frame supernovae rate @xmath39 at @xmath0 was obtained by fitting the redshift distribution of observed sne to the expected distribution , @xmath37 using a maximum - likelihood fit using poisson statistics . the mean redshift corresponding to this rate is @xmath49 .
we derive a value for the sn rate of @xmath50 where the error is statistical only .
we have studied various sources of systematic uncertainty .
table [ systab ] summarizes their contributions .
contribution & 0.02 galaxy extinction & 0.01 apm calibration & 0.10 detection efficiency & 0.08 non - standard sne & 0.29 scanning efficiency & @xmath51 total syst .
uncertainty & @xmath52 [ systab ] for given assumed galaxy counts , the main contribution comes from the fact that type ia sne are not perfect standard candles .
we also tested the sensitivity of our result to the assumed form of galaxy evolution , @xmath15 , by recalculating the rate using the model of gronwall & koo ( 1995 ) and that used by glazebrook et al ( 1995 ) .
these give values for the rate of @xmath53 snu and @xmath54 snu respectively .
although there is a large difference in the results derived using the counts of lilly et al compared with the other models , we choose not to include this in the systematic uncertainty . for the purposes of this analysis we use the lilly et al counts ,
since these are based on data that are well - matched to our survey in magnitude and redshift range , and only small amount of extrapolation was required in converting from the @xmath55 to @xmath14 band .
host galaxy inclination and extinction effects could not be estimated following the analysis used for nearby searches because of the different detection techniques involved .
a correct estimation would require modeling of galaxy opacities , which is beyond the scope of this paper .
we therefore compare our uncorrected value with uncorrected values for nearby searches , altogether , we estimate the total systematic error to be @xmath56 and we assume the lilly et al counts for @xmath15 .
this measurement is the first direct measurement of the type ia rate at high redshift . in their pioneering work searching for high redshift supernovae , hansen et al .
( 1989 ) discovered a probable type ii event at @xmath57 and a type ia event at @xmath58 ( nrgaard - nielsen et al . , 1989 ) , however no estimates of sn ia rates were published based on this discovery .
nearby supernovae rates have been carefully reanalyzed recently ( cappellaro et al .
1993a & 1993b , turatto et al . , 1994 ,
van den bergh and mcclure , 1994 , muller et al . 1992 ) using more precise methods for calculating the control times and correcting for inclination and over - exposure of the nuclear regions of galaxies in photographic searches .
the rate obtained for type ia sne are now consistent among these groups and vary between 0.2 @xmath2 snu and 0.7 @xmath2 snu depending on the galaxy types ( e , sa etc . , higher rates are found in later type galaxies ) .
taking into account the facts that at @xmath590.4 the ratios of galaxy type are different and using the type ia rates reported in cappellaro et al ( 1993b ) , we calculate a local rate of @xmath60 @xmath2 snu for the mix of galaxies expected at @xmath0 .
our measured value of @xmath61 snu ( where statistical and systematic uncertainties have been combined ) , although slightly higher , agrees with this value within the uncertainty and indicates that type ia rates do not change dramatically out to @xmath590.4 .
note , however that correcting for host galaxy extinction and inclination may change this conclusion .
theoretical estimates of type ia sne rates have been derived from stellar and galaxy evolution models .
calculations were done mostly for elliptical galaxy type .
recent calculations , based on evolutionary models of elliptical galaxies , predict rates of @xmath62 snu ( ferrini & poggianti 1993 ) . assuming a factor of @xmath63 higher rate in non - elliptical galaxies compared to ellipticals ( capellaro et al .
1993b ) and a mix of galaxy types as above , we convert this to an overall rate of type ia sne at @xmath64 in all galaxy types , and derive a value of @xmath65 snu .
our total uncertainty of @xmath66 in the measurement presented in this paper does not allow any firm conclusion but our observed rate seems to lie above this theoretical prediction .
there may be an increase of type ia rate with redshift .
ruiz - lapuente , burkert & canal ( 1995 ) predict significant increase in rate for redshifts between 0.4 and 0.8 depending on the specific model they consider . in the near future ,
our ongoing high-@xmath43 sn search and others should provide enough data to constrain the theoretical calculations . this work was supported in part by the national science foundation ( adt-88909616 ) and the u.s .
dept . of energy ( de - ac03 - 76sf000098 ) .
we thank the la palma staff & observers for carrying out service observations .
we also thank simon lilly and caryl gronwall & david koo for providing their galaxy counts prior to publication , and richard kron for useful discussions .
i.m . hook acknowledges a nato fellowship .
r. pain thanks gerard fontaine of cnrs - in2p3 and bernard sadoulet of cfpa , berkeley for supporting this work . | we present the first measurement of the rate of type ia supernovae at high redshift .
the result is derived using a large subset of data from the supernova cosmology project as described in more detail at this meeting by perlmutter et al .
( 1996 ) .
we present our methods for estimating the numbers of galaxies and the number of solar luminosities to which the survey is sensitive , the supernova detection efficiency and hence the control time .
we derive a rest - frame type ia supernova rate at @xmath0 of @xmath1 @xmath2 snu where the first uncertainty is statistical and the second includes systematic effects . |
in paper i , by comparing the fbs ( markarian et al . 1989 ) and bqs ( green et al .
1986 ) surveys in their area in common , we derived a completeness of @xmath170% for the bqs .
a number of bright agns have since been discovered in the area which , together with our new spectroscopic observations , allowed us to refine our previous estimate of the bqs completeness .
we have obtained new spectra for 11 fbs objects .
the observations were carried out on november 25 , 1998 and january 1415 , 1999 at the byurakan astrophysical observatory ( bao ) and at the observatoire de haute - provence ( ohp ) , respectively .
the journal of observations is given in table 1 , together with relevant data .
+ .[spectra]new spectra .
1 gives the name , col .
2 the fbs number , col .
3 the original fbs classification , col .
4 the magnitude , cols . 5 and 6 the place and date of observation , col .
7 the galactic latitude , col .
8 the classification and col .
9 the redshift [ cols="<,>,<,<,<,<,>,<,<",options="header " , ] in the case of rxsj12110@xmath57005 for which the aps magnitude is not available , schwope et al . ( 2000 ) give @xmath0 17.0 , while the usno @xmath6 magnitude is 14.3 ; but this object has a moderate redshift ( @xmath7 0.127 ) ; moreover its apm @xmath6 magnitude ( irwin et al . 1994 ) is 17.66 ; we therefore adopted the schwope et al .
mag and excluded it from the complete " sample . the first bright quasar survey ( fbqs ) was built by matching the vla first survey with the cambridge automated plate measuring machine ( apm ) catalog of poss - i objects ( irwin et al .
1994 ) ; it covers an area of 2682 deg@xmath2 in the north galactic cap ; it contains 1238 objects brighter than 17.8 mag on the poss - i @xmath8 plates ( white et al .
about 1180 square degrees are within the fbs area ; they contain 38 first radio sources identified with an agn brighter than @xmath0 17.0 at @xmath9 30 ; nine are bright qsos ( @xmath10 16.0 ) , three ( cso 900 , firstj1306@xmath53915 and rxsj17102@xmath53344 ) being new .
although the numbers are small , this suggests that the complete " sample we built in paper i is only 67@xmath415% complete .
according to white et al .
( 2000 ) , qsos with radio emission above the first 1 mjy limit constitute about 25% of all qsos brighter than @xmath11 17.6 , but for qsos brighter than @xmath0 16.4 , the fbqs qso density is indistinguishable from the density of optically selected qsos .
nevertheless , of the 15 bright qsos known prior to the first survey in the area common to the first and fbs surveys , only six ( 40% ) have been detected as first radio sources ; therefore the complete identification of the first sources with bright starlike objects could not yield a complete survey of bright qsos .
+ a number of recent papers are devoted to the optical identification of rass sources ( beuermann et al .
1999 ; cao et al . 1999 ; grazian et al . 2000 ; schwope et al .
2000 ; wei et al . 1999 ; xu et al .
one of the new identifications is rxsj12043@xmath54330 , a qso at @xmath7 0.663 ( xu et al .
1999 ) ; it is also fbs1201@xmath5437 ( fbs#302 ) or pg1201@xmath5436 , which had been classified as a dc white dwarf by green et al .
its aps @xmath6 magnitude is 16.23 ; it is therefore not bright enough to be included in our complete " sample .
+ nineteen rass sources are now identified with a bright qso in the area discussed in this paper ( including the three new first qsos ) ; of the 17 fbs or bqs bright qsos in our sample ( table [ brightqso ] ) , 12 ( 70% ) are rosat all sky survey ( rass ) x - ray sources , suggesting that the total number of bright qsos is equal to 19/0.70 @xmath12 27 ( if all optically bright , x - ray sources have been discovered ) .
our complete " sample of bright qsos ( table [ brightqso ] ) contains 29 objects brighter than @xmath0 16.16 ( @xmath10 16.00 ) , three of them ( indicated by a
n " in the last column of table [ brightqso ] ) are not within the pg area .
the area common to the pg and fbs surveys at @xmath9 30 ( @xmath12250 deg@xmath2 ) contains 26 bright qsos ( 13 pg qsos and 13 others ) ( but there are 17 pg qsos with @xmath13 16.16 in the area ; this larger number is probably due to the eddington ( 1940 ) effect , the pg magnitudes being affected by relatively large errors , @xmath14 0.37 mag ) . from these data ,
we derive a surface density of 0.012 deg@xmath3 , which is to be compared with the original value of the pg survey : 0.0064 deg@xmath3 , implying a maximum completeness of 53@xmath410% for the pg survey . + grazian et al .
( 2000 ) have cross - correlated the rass with photometric databases in an 8164 deg@xmath2 area of the northern sky at @xmath15 30 , selecting all coincidences brighter than @xmath16 15.4 ; from this , they derive a surface density of bright ( @xmath17 15.5 ) qsos ( defined as agns with @xmath1823.0 ) of 10@xmath42 10@xmath19 deg@xmath3 and conclude that the true surface density of such objects is about three times larger than that derived from the pg survey .
however , they do not specify how the @xmath20 magnitude of their objects was derived .
their sample contains 46 qsos ; 15 of them have @xmath21 0.20 ; we have extracted from the aps catalogue the @xmath6 magnitudes for 12 of them ( for the three others , these magnitudes are unavailable ) ; it turns out that only one ( j172320.5@xmath5341756 ) has @xmath22 15.34 , corresponding to @xmath17 15.5 , suggesting that the @xmath6 magnitudes used by grazian et al . are underestimated and , consequently , the surface density overestimated . + lamontagne et al .
( 2000 ) claim that they found a surface density of bright qsos three times larger than the pg value .
they have searched for uv - excess stellar - like objects with @xmath17 16.5 and @xmath23 @xmath240.6 in a 840 deg@xmath2 area covering the south galactic cap ; the errors in the @xmath20 magnitudes are estimated to be 0.30 mag _ rms_.
they have found 228 such objects which have all been spectroscopically identified ; 32 are agns , out of which only eleven are brighter than @xmath0 16.16 and @xmath2524.0 ( including 0117@xmath262837 which , according to grupe et al .
( 1999 ) , has a redshift of 0.349 rather than 0.055 ) .
we derive a surface density or 0.013 deg@xmath3 , in agreement with our value and only twice the pg value .
in paper i , we compared the surface density of qsos in the bright quasar survey and in the first byurakan survey and concluded that the completeness of the bqs is of the order of 70% ; wisotzki et al .
( 2000 ) have found that the bqs is 68% complete from a comparison with the hamburg / eso survey , in agreement with our previoys estimate .
based on a number of recently published data , as well as on our own new observations , we redetermined the surface density of qsos brighter than @xmath0 16.16 in the bqs area to be @xmath10.012 deg@xmath3 , implying that the completeness of the bqs is 53@xmath410% .
it should be stressed however that the numbers involved are quite small , and that larger areas should be investigated before a definitive value of the surface density of bright qsos could be determined .
abramian g.b .
, mickaelian a.m. 1994 , astrophysics 37 , 224 bade n. , engels d. , voges w. 1998 , a&as 127 , 145 beuermann k. , thomas h .- c .
, reinsch k. et al .
1999 , a&a 347 , 47 cao l. , wei j.y . , hu j.y . 1999 , a&as 135 , 243 eddington a.s .
1940,mnras 100,354 grazian a. , cristiani s. , dodorico v. , omizzolo v. , pizella a. 2000 , aj ( astro - ph/0002183 ) green r.f . , schmidt m. , liebert j. 1986 , apjs 61 , 305 grupe d. , beuermann k. , mannheim k. , thomas h .- c . 1999 , a&a 350 , 805 irwin m. , maddox s. , mcmahon r. 1994 , spectrum 2 , 14 lamontagne r. , demers s. , wesemael f. , fontaine g. , irwin m.j . 2000 , aj 119 , 241 markarian b.e .
, lipovetsky v.a .
, stepanian j.a . ,
erastova l.k . ,
shapavalova a.i .
1989 , commun .
special astrophys . obs .
62 , 5 marzke r.o .
, huchra j.p . , geller m.j . 1996 , aj 112 , 1803 mickaelian a.m. , gonalves a.c . , vron - cetty m .-
, vron p. 1999 , astrophysics 42 , 1 ( paper i ) monet d. , bird a. , canzian b. et al .
1996 , usno - a2.0 , u.s .
naval observatory , washington d.c .
pennington r.l .
, humphreys r.m .
, odewahn s.c .
, zumach w. , thurmes p.m. 1993 , pasp 105 , 521 schwope a. , hasinger g. , lehman i. et al .
2000 , an 321 , 1 voges w. , aschenbach b. , boller t. et al . 1999 , a&a 349 , 389 wei j.y .
, dong x.y . , hu j.y . 1999 ,
a&as 139 , 575 white r.l . ,
becker r.h .
, gregg m.d .
2000 , apjs 126 , 133 wisotzki l. , christlieb n. , bade n. et al .
2000 , a&a 358 , 77 xu d.w .
, wei j.y . , dong x.y . , hu j.y .
1999 , a&as 134 , 365 | in paper i ( mickaelian et al .
1999 ) , we compared the surface density of qsos in the bright quasar survey ( bqs ) and in the first byurakan survey ( fbs ) and concluded that the completeness of the bqs is of the order of 70% rather than 3050% as suggested by several authors .
a number of new observations recently became available , allowing a re - evaluation of this completeness .
we now obtain a surface density of qsos brighter than @xmath0 16.16 in a subarea of the fbs covering @xmath12250 deg@xmath2 , equal to 0.012 deg@xmath3 ( 26 qsos ) , implying a completeness of 53@xmath410% . |
in recent years , the study of magnetohydrodynamic ( mhd ) flow of blood through artery has gained considerable interest because of its physiological applications .
investigations on mhd flow and heat transfer of non - newtonian fluids over a stretching sheet also find many important applications in engineering and industry .
for example in the extrusion of a polymer sheet from a die , the sheet is sometimes stretched .
the properties of the end product depend considerably on the rate of cooling . by drawing such a sheet in a visco - elastic electrically conducting
fluid subjected to the action of a magnetic field , the rate of cooling can be controlled and the final product can be obtained with desired characteristics . crane@xcite
investigated the problem of steady two - dimensional incompressible boundary layer flow engendered by the stretching of an elastic flat sheet which moves in its plane with a velocity varying linearly with distance from a fixed point due to application of a uniform stress .
misra et al.@xcite studied the hall effect on the steady mhd boundary layer flow of an incompressible viscous and electrically conducting fluid past a stretching surface in the presence of a uniform transverse magnetic field .
+ + [ cols= " < , < " , ] it may be noted that for values of m @xmath0 200 , the points where the reverse flow set in are quite close to each other .
the interesting observation of flow reversal owes to its origin to the stretching of the channel wall .
it is also revealed that in the vicinity of the boundary layer , the axial velocity decreases with the increase in the magnetic field strength .
this observation leads to a very important suggestion that one can avoid flow reversal ( studied here ) by applying a sufficient strong magnetic field .
8 and 9 depict the spatial variation of velocity components for different values of the viscoelastic parameter @xmath1 . from these two figures
one can a very important observation that for a purely hydrodynamic flow ( in the absence of any magnetic field ) the velocity of blood decreases with the increase in the value of the viscoelastic parameter , but under the action of a sufficient strong magnetic field , velocity is least affected by blood viscoelasticity
10 - 11 give the spatial variation of temperature for different values of the magnetic parameter m and the different values of the viscoelastic parameter @xmath1 .
from these figures we learn that the temperature increases with the increase in the magnetic field strength and the viscoelastic parameter .
it thus turns out that under the action of a magnetic field , in an electrically conducting fluid ( e.g. blood ) , there develops a resistive force ( lorentz force ) which causes impedance of flow and enhancement of temperature .
further , fig .
12 shows that under the action of a sufficient strong magnetic field the temperature of blood decreases with the increase in prandtl number .
the flow of a magnetohydrodynamic fluid in a channel with stretching walls and an associated problem of heat transfer have been the concern of this paper . the study is particularly motivated towards the flow and heat transfer of blood in a vessel having stretching wall .
it bears the potential to examine the variation of blood velocity as the magnetic field strength ( m ) , fluid ( blood ) viscoelasticity ( measured by @xmath1 ) and thermal diffusivity ( which is related to the prandtl number @xmath2 ) .
the study reveals that the flow reversal can be eliminated and the vessel wall temperature can be controlled by applying a sufficiently strong magnetic field .
+ the observation of the study that temperature increases as the strength of the magnetic field increases is likely to be useful in the development of new heating methods .
it can also have significant clinical application of hyperthemia in cancer therapy .
the objective of hyperthemia in cancer therapy is to raise the temperature of cancerous tissues above a therapeutic value @xmath3c while maintaining the surrounding normal tissue at sublethal temperature @xcite .
the data reported here through different graphs should be of interest to medical persons in the treatment of various health problems , including arrest of cancerous tumor growths and to achieve the goal for monitoring of the tumor and normal tissue temperatures . .
+ + * acknowledgements : * _ the authors wish to express their deep sense of gratitude to the reviewers of the original manuscript for their kind suggestions based upon which the present version of the paper has been prepared .
one of the authors ( j. c. misra ) is thankful to the alexander von humboldt foundation , germany for supporting this investigation .
_ + + crane lj .
flow past a stretching plate .
z angew math phys(zamp ) 1970 ; 21 : 645 - 47 .
misra jc , pal b , gupta as .
hall effects on magnetohydrodynamic flow and heat transfer over a stretching surface . in : malik
sk , editor .
mathematics and its application to idustry and new engineering area .
new delhi : insa , 2001 , p. 49 - 62 .
misra jc , chakravarty s. flow in arteries in the presence of stenosis .
j biomech 1986 ; 19 : 907 - 18 .
misra jc , patra mk , misra sc . a non - newtonian fluid model for blood flow through arteries under stenotic conditions .
j biomech 1993 ; 26 : 1129 - 41 .
pavlov kb .
magnetohydrodynamic flow of an incompressible viscous fluid caused by deformation of a plane surface .
magnitnaya gidrodinamika 1974 ; 4 : 146 - 47 .
kumari m , takhar hs , nath g. mhd flow and heat transfer over a stretching surface with prescribed wall temperature or heat flux .
warme stofubertrag 1990 ; 25 : 331 - 36 .
anderson hi .
mhd flow of viscoelastic fluid past a stretching surface .
acta mechanica 1992 ; 95 : 227 - 30 .
datti ps , prasad kv , abel ms , joshi m. mhd viscoelastic fluid flow over a non - isothermal stretching sheet .
int j engg sci 2004 ; 42 : 935 - 46 .
. unsteady mhd convective heat and mass transfer past a semi - infinite vertical permeable moving plate with heat absorption .
int j engg sci 2004 ; 42 : 217 - 30 .
rajagopal kr , na ty , gupta as . flow of a viscoelastic fluid over a stretching sheet .
rheol acta 1984 ; 23 : 213 - 15 .
rajagopal kr , na ty , gupta as .
a non - similar boundary layer on stretching sheet in a non - newtonian fluid with uniform free stream .
j math phys sci 1987 ; 21 : 189 - 200 .
anderson hi , bech kh , dandapat bs .
magnetohydrodynamic flow of a power - law fluid over a stretchinf surface .
int j non - linear mech 1992 ; 27 : 929 - 36 .
misra jc , pal b. hydromagnetic flow of a viscoelastic fluid in a parallel plate channel with stretching walls .
ind j of maths 1999 ; 41 : 231 - 47 .
misra jc , pal b , gupta as .
hydromagnetic flow of a second - grade fluid in a channel - some applications to physiological systems . math model and methods in appl sci 1998 ; 8 : 1323 - 42 .
fukada e , kaibara m. viscoelastic study of aggregation of red blood cells .
biorheology 1980 ; 17 : 177 - 82 .
thurston gb .
viscoelasticity of human blood .
biophysical j 1972 ; 12 : 1205 - 17 .
stoltz jf , lucius m. viscoelasticity and thixotropy of human blood .
biorheology 1981 ; 18 : 453 - 73 .
dewhirst mw , samulski tv .
hyperthemia in the treatment for cancer .
upjhon : kalamazoo , mi ; 1988 .
craciunescu oi , clegg st .
pulsatile blood flow effects on temperature distribution and heat transfer in rigid vessels .
j biomech engg 2001 ; 123 : 500 - 5 .
weinbaun s , jiji lm , lemons de . theory and experiment for the effect of vascular microstructure on surface tissue heat transfer - part i : anatomical foundation and model conceptualization . j biomech engg 1984 ; 106 : 321 - 30 .
jiji lm , weinbaun s , lemons de .
theory and experiment for the effect of vascular microstructure on surface tissue heat transfer - part ii : model formulation and solution .
j biomech engg 1984 ; 106 : 331 - 41 .
cavaliere r , ciocatto ec , giovannela bc , heidelberger c , hojnson ro , martotini m , rmondovi b , moricca g , rossi - fanelli a. selective heat sensitivity of cancer cells : biochemical and clinical studies .
cancer 1967 ; 20 : 1351 - 81 .
shitzer a , eberhart rc .
heat transfer in medicine and biology- analysis and applications .
new york and london : plenum press ; volume 2 , 1985 .
walters k. second - order effects in elasticity , plasticity and fluid dynamics .
pergamon ; 1964 .
heat transfer to blood vessels .
j biomech engg 1980 ; 102 : 110 - 18 .
tzirtzilakis ee , kafoussias ng .
biomagnetic fluid flow over a stretching sheet with non - linear temperature dependent magnetization .
z angew math phys(zamp ) 2003 ; 54 : 551 - 65 .
field sb , hand jw .
an introduction to the practical aspects of hyperthemia .
new york : taylor @xmath4 francis ; 1990 .
* list of figure captions * + + fig .
1 a sketch of the physical problem + + fig .
2 variation of @xmath5 with @xmath6 for m=100 and @xmath7=0.005 + + fig
. 3 variation of @xmath8 with @xmath6 for m=100 and @xmath7=0.005 + + fig
. 4 variation of @xmath5 with @xmath9 for different values of m and @xmath1=0.001 + + fig
. 5 variation of @xmath8 with @xmath9 for different values of m and @xmath1=0.001 + + fig .
6 variation of @xmath5 with @xmath9 for different values of m and @xmath1=0.005 + + fig .
7 variation of @xmath8 with @xmath9 for different values of m and @xmath1=0.005 + + fig . 8
variation of @xmath5 with @xmath9 for different values of @xmath1 + + fig .
9 variation of @xmath8 with @xmath9 for different values @xmath1
+ + fig .
10 variation of non - dimensional temperature with @xmath6 for different values of m + + fig .
11 variation of non - dimensional temperature with @xmath6 for different values of @xmath1 + + fig .
12 variation of non - dimensional temperature with @xmath6 for different values of @xmath2 + 22.0 cm | of concern in the paper is a study of steady incompressible viscoelastic and electrically conducting fluid flow and heat transfer in a parallel plate channel with stretching walls in the presence of a magnetic field applied externally .
the flow is considered to be governed by walter s liquid b fluid .
the problem is solved by developing a suitable numerical method .
the results are found to be in good agrement with those of earlier investigations reported in existing scientific literatures .
the study reveals that a back flow occurs near the central line of the channel due to the stretching walls and further that this flow reversal can be stopped by applying a strong external magnetic field .
the study also shows that with the increase in the strength of the magnetic field , the fluid velocity decreases but the temperature increases .
thus the study bears potential applications in the study of the haemodynamic flow of blood in the cardiovascular system when subjected to an external magnetic field . + * keywords : * non - newtonian fluid , mhd flow , viscoelasticity , stretching walls , + heat transfer 0.8 cm -.1 in -.1 in 0.1 cm 20.5 cm [ section ] [ theorem]definition [ theorem]observation [ theorem ] [ theorem]proposition [ theorem]rule 22.5 cm = currsize |
the aim of these lectures is to explain how to apply controlled path ideas @xcite to solve simple problems in singular pdes .
the hope is that the insight gained by doing so can inspire new applications or the construction of other more powerful tools to analyze a wider class of problems .
we discuss some problems involving singular stochastic non
linear parabolic equations from the point of view of controlled paths .
to understand the origin of such singular equations , we have chosen to present the example of a homogenization problem for a singular potential in a linear parabolic equation .
this point of view has the added benefit that it allows us to track back the renormalization needed to handle the singularities as effects living on other scales than those of interest .
the basic problem is that of having to handle effects of the microscopic scales and their interaction through non
linearities on the macroscopic behavior of the solution .
mathematically , this problem translates into the attempt of making schwartz s theory of distribution coexist with non linear operations which are notoriously not continuous in the usual topologies on distributions .
this is a very old problem of analysis and has been widely studied .
the additional input which is not present in the usual approaches is that the singularities which force us to treat the problem in the setting of schwartz s distributions are of a stochastic nature .
so we dispose of two handles on the problem : the analytical one and the probabilistic one .
the right mix of the two will provide an effective solution to a wide class of problems .
a first and deep understanding of these problems has been obtained starting from the late 90s by t. lyons @xcite , who introduced a theory of _ rough paths _ in order to settle the conflict of topology and non linearity in the context of driven differential equations , or more generally in the context of the non linear analysis of time
varying signals .
nowadays there are many expositions of this theory @xcite and we refer the reader to the literature for more details . in @xcite ,
the notion of _ controlled paths _ has been introduced in order to extend the applicability of the rough path ideas to a larger class of problems that are not necessarily related to the integration of odes but which still retain the one dimensional nature of the directions in which the irregularity manifest itself .
the controlled path approach has been used to make sense of the evolution of irregular objects such as vortex filaments and certain spdes .
later hairer understood how to apply these ideas to the long standing problem of the kardar
zhang equation @xcite , and his insights prompted the researchers to try more ambitious approaches to extend rough paths to a multidimensional setting . in @xcite , in collaboration with p. imkeller
, we introduced a notion of _ paracontrolled distributions _ which is suitable suitable to handle a wide class of spdes which were well out of reach with previously known methods .
paracontrolled distributions can be understood as an extension of controlled paths to a multidimensional setting , and they are based on new combinations of basic tools from harmonic analysis . at the same time , hairer managed to devise a vast generalization of the basic construction of controlled rough paths in the multidimensional and distributional setting , which he called the theory of _ regularity structures _
@xcite and which subsumes standard analysis based on hlder spaces and controlled rough path theory but goes well beyond that .
just few days after the lectures in mambucaba took place , it was announced that martin hairer was awarded a fields medal for his work on spdes and in particular for his theory of regularity structures @xcite as a tool for dealing with singular spdes .
this prize witnesses the exciting period we are experiencing : we now understand sound lines of attack to long standing problems , and there are countless opportunities to apply similar ideas to new problems .
the plan of the lectures is the following .
we start by explaining `` energy solutions '' @xcite , a notion of solution to ( a particular class of ) singular pdes which has the advantage of being comparably easy to handle but which also has the inconvenience of not having a comprehensive uniqueness theory so far .
this will allow us to introduce the reader to spdes in a progressive manner , and also to introduce gaussian tools ( wick products , hypercontractivity ) and to present some of the basic phenomena appearing when dealing with singular spdes .
next we set up the analytical tools we need in the rest of the lectures : besov spaces and some basic harmonic analysis based on the littlewood
paley decomposition .
next , in order to motivate the reader and to provide a physical ground for the intuition to stand on , we discuss a homogenization problem for the linear heat equation with random potential .
this will allow us to derive the need for the weak topologies we shall use and for irregular objects like white noise from first principles and `` concrete '' applications .
the homogenization problem also allows us to see the renormalization effects appear naturally and to track their mathematical meaning . starting from these problems
we introduce the two dimensional parabolic anderson model , the simplest spde in which most of the features of more difficult problems are already present , and we explain how to use paraproducts and the paracontrolled ansatz in order to keep the non linear effect of the singular data under control .
then we go on to discuss the more involved situation of the stochastic burgers equation in one dimension , which is one of the avatars of the kardar
zhang equation .
[ [ conventions - and - notations ] ] conventions and notations + + + + + + + + + + + + + + + + + + + + + + + + + we write @xmath0 if there exists a constant @xmath1 , independent of the variables under consideration , such that @xmath2 .
similarly we define @xmath3 .
we write @xmath4 if @xmath0 and @xmath5 .
if we want to emphasize the dependence of @xmath6 on the variable @xmath7 , then we write @xmath8 . if @xmath9 and @xmath10 are index variables of littlewood
paley blocks ( to be defined below ) , then @xmath11 is to be interpreted as @xmath12 , and similarly for @xmath13 and @xmath14 . in other words ,
@xmath11 means @xmath15 for some fixed @xmath16 that does not depend on @xmath9 or @xmath10 .
we use standard multi - index notation : for @xmath17 we write @xmath18 and @xmath19 , as well as @xmath20 for @xmath21 . for @xmath22
we write @xmath23 for the functions
@xmath24 which are @xmath25 times continuously differentiable with @xmath26hlder continuous derivatives of order @xmath25 , equipped with its natural norm @xmath27 .
if we write @xmath28 , then that means that @xmath29 is in @xmath30 for all @xmath31 .
the @xmath32 spaces will be defined below .
the first issue one encounters when dealing with singular spdes is the ill posed character of the equation , even in a weak sense .
typically , the non linearity does not make sense in the natural spaces where solutions live and one has to provide a suitable smaller space in which it is possible to give an appropriate interpretation to `` ambiguous quantities '' that are featured in the equation .
energy solutions @xcite are a relatively simple tool in order to come up with well defined non linearities .
the drawback is that currently the issue of uniqueness is open , at least in the most interesting cases .
it is not clear if uniqueness of energy solutions holds or even how to find conditions that ensure uniqueness . on the other hand ,
proving existence of energy solutions or even convergence to energy solutions is usually a quite simple problem , at least compared to the other approaches like paracontrolled solutions or regularity structures , where existence requires already quite a large amount of computations but where uniqueness can be established quite easily afterwards .
our aim here is to motivate the ideas leading to the notion of energy solutions .
we will not insist on a detailed formulation of all the available results .
the reader can always refer to the original paper @xcite for missing details .
applications to the large scale behavior of particle systems are studied in @xcite .
we will study energy solutions for the stochastic burgers equation on the torus @xmath33 .
the unknown @xmath34 should satisfy @xmath35 where @xmath36 is a space time white noise defined on a given probability space @xmath37 fixed once and for all .
that is , @xmath38 is a centered gaussian process indexed by @xmath39 with covariance @xmath40 = \int_{\mathbb{r}_+ \times \mathbb{t } } f ( t , x ) g ( t , x ) { \mathrm{d}}t { \mathrm{d}}x.\ ] ] the equation has to be understood as a relation for processes which are distributions in space with sufficiently regular time dependence ( for a short introduction to distributions on @xmath41 see section [ sec : preliminaries ] below ) . in particular , if we test the above relation with @xmath42 , denote with @xmath43 the pairing of the distribution @xmath44 with @xmath45 , and integrate in time over the interval @xmath46 $ ] , we get @xmath47 let us discuss the various terms in this equation . in order to make sense of @xmath43 and @xmath48 , it is enough to assume that for all @xmath49 the mapping @xmath50 is a stochastic process with continuous trajectories .
next , if we denote @xmath51 then , at least by a formal computation , we have that @xmath52 is a gaussian random field with covariance @xmath53 = ( t \wedge s ) \langle
\partial_x \varphi , \partial_x \psi \rangle_{l^2 ( \mathbb{t } ) } .\ ] ] in particular , for every @xmath49 the stochastic process @xmath54 is a brownian motion with covariance @xmath55 here we used the notation @xmath56 in order to stress the fact that @xmath57 is a martingale in its natural filtration and more generally in the filtration @xmath58 , @xmath59 .
the most difficult term is of course the nonlinear one : @xmath60 . in order to define it
, we need to square the distribution @xmath61 , an operation which in general can be quite dangerous
. a natural approach would be to define it as the limit of some regularizations .
for example , if @xmath62 is a compactly supported @xmath63 function such that @xmath64 , and we set @xmath65 , then we can set @xmath66 and define @xmath67 whenever the limit exists in @xmath68 , the space of distributions on @xmath41 . then the question arises which properties @xmath29 should have for this convergence to occur .
let us simplify the problem and start by studying the linearized equation obtained by neglecting the non linear term .
let @xmath69 be a solution to @xmath70 for all @xmath59 and @xmath49 .
this equation has _ at most _ one solution ( for fixed @xmath71 ) .
indeed , the difference @xmath72 between two solutions should satisfy @xmath73 , which means that @xmath72 is a distributional solution to the heat equation .
taking @xmath74 , where @xmath75 we get @xmath76 and then by gronwall s inequality @xmath77 for all @xmath59 .
this easily implies that @xmath78 in @xmath79 for all @xmath59 . to obtain the _ existence _ of a solution ,
observe that @xmath80 and that @xmath81 , while for all @xmath82 the process @xmath83 is a complex valued brownian motion ( i.e. real and imaginary part are independent brownian motions with the same variance ) .
the covariance of @xmath84 is given by @xmath85 = ( t \wedge s ) \delta_{k + m = 0}\ ] ] and moreover @xmath86 for all @xmath82 ( where @xmath87 denotes complex conjugation ) , as well as @xmath88 .
in other words , @xmath89 is a complex valued ornstein uhlenbeck process which solves a linear one dimensional sde and has an explicit representation given by the variation of constants formula @xmath90 this is enough to determine @xmath91 for all @xmath59 and @xmath49 .
show that @xmath92 is a complex gaussian random field , that is for all @xmath93 , for all @xmath94 , @xmath95 , the vector @xmath96 is multivariate gaussian .
show that @xmath69 has mean @xmath97 = e^{- k^2 t } x_0 ( e_k)$ ] and covariance @xmath98 ) ( x_s ( e_m ) -\mathbb{e } [ x_s ( e_m ) ] ) ] = k^2 \delta_{k + m = 0 } \int_0^{t \wedge s } e^{- k^2 ( t - r ) - k^2 ( s - r ) } { \mathrm{d}}r\ ] ] as well as @xmath98 ) ( x_s ( e_m ) -\mathbb{e } [ x_s ( e_m)])^{\ast } ] = k^2 \delta_{k = m } \int_0^{t \wedge s } e^{- k^2 ( t - r ) - k^2 ( s - r ) } { \mathrm{d}}r.\ ] ] in particular , @xmath99 |^2 ] = \frac{1 - e^{- 2 k^2 t}}{2 } .\ ] ] next we examine the sobolev regularity of @xmath69 .
for this purpose , we need the following definition .
let @xmath100 .
then the sobolev space @xmath101 is defined as @xmath102 we also write @xmath103 for the space of continuous functions from @xmath104 to @xmath101 .
[ lemma : ou - regularity]let @xmath105 and assume that @xmath106
. then almost surely @xmath107 .
let @xmath108 and consider @xmath109 let us estimate the @xmath110 norm of this quantity for @xmath111 by writing @xmath112 by cauchy
schwartz , we get @xmath113 note now that @xmath114 is a gaussian random variable , so that there exists a universal constant @xmath115 for which @xmath116 moreover , @xmath117 leading to @xmath118 for any @xmath119 $ ] and @xmath82 , we thus have @xmath120 while for @xmath121 we have @xmath122
. therefore , @xmath123^p \\ & \lesssim ( t - s)^{\kappa p } \big(\| x_0 \|^{2 p}_{h^{\alpha + \kappa } ( \mathbb{t } ) } + \big[\sum_{k \in \mathbb{z}_0 } ( 1 + | k |^2)^{\alpha } ( k^2)^{\kappa}\big]^p\big ) .
\end{aligned}\ ] ] if @xmath124 , the sum on the right hand side is finite and we obtain an estimation for the modulus of continuity of @xmath125 in @xmath126 : @xmath127 .\ ] ] now kolmogorov s continuity criterion allows us to conclude that almost surely @xmath128 whenever @xmath129 .
now note that the regularity of the ornstein uhlenbeck process does not allow us to form the quantity @xmath130 point wise in time since by fourier inversion @xmath131 , and therefore we should have @xmath132 of course , at the moment this expression is purely formal since we can not guarantee that the infinite sum converges .
a reasonable thing to try is to approximate the square by regularizing the distribution , taking the square , and then trying to remove the regularization .
let @xmath133 be the projector of a distribution onto a finite number of fourier modes : @xmath134 then @xmath135 is a smooth function of @xmath7 and we can consider @xmath136 which satisfies @xmath137 we would then like to take the limit @xmath138 . for convenience , we will perform the computations below in the limit @xmath139 , but one has to come back to the case of finite @xmath140 in order to make it rigorous .
let us introduce the notation @xmath141 .
then @xmath142 & = ( 2 \pi)^{- 1 / 2 } \delta_{k = 0 } \sum_{m \in \mathbb{z}_0 } \mathbb{e } [ x_t ( e_{- m } ) x_t ( e_m ) ] \\ & = ( 2 \pi)^{- 1 / 2 } \delta_{k = 0 } \sum_{m \in \mathbb{z}_0 } e^{- 2 m^2 t } | x_0 ( e_m ) |^2 + ( 2 \pi)^{- 1 / 2 } \delta_{k = 0 } \sum_{m \in \mathbb{z}_0 } m^2 \int_0^t e^{- 2 m^2 ( t - s ) } { \mathrm{d}}s\end{aligned}\ ] ] and @xmath143 this is not really a problem since in the equation only components of @xmath144 with @xmath82 appear .
however , @xmath145 is not even a well defined random variable . for the moment ,
let us assume that @xmath146 , which will slightly simplify the computation .
if @xmath82 , we have @xmath147 = \mathbb{e } [ x^2_t ( e_k ) x^2_t ( e_{- k } ) ] = ( 2 \pi)^{- 1 } \sum_{\ell + m = k } \sum_{\ell ' + m ' = - k } \mathbb{e } [ x_t ( e_{\ell } ) x_t ( e_m ) x_t ( e_{\ell ' } ) x_t ( e_{m ' } ) ] .\ ] ] by wick s theorem ( see @xcite , theorem 1.28 ) , the expectation can be computed in terms of the covariances of all possible pairings of the four gaussian random variables ( 3 possible combinations ) : @xmath148 & = \mathbb{e } [ x_t ( e_{\ell } ) x_t ( e_m ) ] \mathbb{e } [ x_t ( e_{\ell ' } ) x_t(e_{m ' } ) ] \\ & \quad + \mathbb{e } [ x_t ( e_{\ell } ) x_t ( e_{\ell ' } ) ] \mathbb{e } [ x_t ( e_m ) x_t(e_{m ' } ) ] \\ & \quad + \mathbb{e } [ x_t ( e_{\ell } ) x_t ( e_{m ' } ) ] \mathbb{e } [ x_t ( e_m ) x_t ( e_{\ell'})].\end{aligned}\ ] ] since @xmath82 , we have @xmath149 and @xmath150 which allows us to neglect the first term since it is zero . by symmetry of the summations ,
the two other give the same contribution and we remain with @xmath151 & = 2 \sum_{\ell + m = k } \sum_{\ell ' + m ' = k } \mathbb{e } [ x_t ( e_{\ell } ) x_t ( e_{\ell ' } ) ] \mathbb{e } [ x_t ( e_m ) x_t ( e_{m ' } ) ] \\
\nonumber & = 2 \sum_{\ell + m = k } \mathbb{e } [ x_t ( e_{\ell } ) x_t ( e_{- \ell } ) ] \mathbb{e } [ x_t ( e_m ) x_t ( e_{- m } ) ] \\
\nonumber & = \frac{1}{2 } \sum_{\ell + m = k } (
1 - e^{- 2 \ell^2 t } ) ( 1 - e^{- 2 m^2 t } ) = + \infty.\end{aligned}\ ] ] this shows that even when tested against smooth test functions , @xmath152 is not in @xmath153 .
this indicates that there are problems with @xmath152 and indeed one can show that @xmath154 does not make sense as a random variable .
to understand this better , observe that the ornstein uhlenbeck process can be decomposed as @xmath155 where we extended the brownian motions @xmath156 to two sided complex brownian motions by considering independent copies .
it is not difficult to show that the second term gives rise to a smooth function if @xmath157 , so all the irregularity of @xmath158 is described by the first term which we call @xmath159 .
note that @xmath160 for all @xmath161 and @xmath162 , where we write @xmath163 if @xmath164 , where @xmath165 and @xmath166 are independent random variables with distribution @xmath167 .
the random distribution @xmath168 then satisfies @xmath169 , or in other words it is ( @xmath170 times ) the white noise on @xmath41 .
it is also possible to deduce that the white noise on @xmath41 is indeed the invariant measure of the ornstein uhlenbeck process , that it is the only one , and that it is approached quite fast .
so we should expect that , at fixed time , the regularity of the ornstein uhlenbeck process is like that of the space white noise and this is a way of understanding our difficulties in defining @xmath130 since this will be , modulo smooth terms , the square of the space white noise .
a different matter is to make sense of the time integral of @xmath171 .
let us give it a name and call it @xmath172 . for @xmath173 , the computation of its variance gives a quite different result .
[ lem : ou square]almost surely , @xmath174 .
proceeding as in ( [ eq : ou square variance ] ) , we have now @xmath175 = 2 k^2 \int_0^t \int_0^t \sum_{\ell + m = k } \mathbb{e } [ x_s ( e_{\ell } ) x_{s ' } ( e_{- \ell } ) ] \mathbb{e } [ x_s ( e_m ) x_{s ' } ( e_{- m } ) ] { \mathrm{d}}s { \mathrm{d}}s ' .\ ] ] if @xmath176 , we have @xmath177 = \frac{1}{2 } e^{- \ell^2 ( s - s ' ) } ( 1 - e^{- 2 \ell^2 s'}),\ ] ] and therefore @xmath178 & = \frac{k^2}{2 } \int_0^t \int_0^t \sum_{\ell + m = k } e^{- ( \ell^2 + m^2 ) | s - s ' | } ( 1 - e^{- 2 \ell^2(s ' \wedge s ) } ) ( 1 - e^{- 2 m^2 ( s ' \wedge s ) } ) { \mathrm{d}}s { \mathrm{d}}s ' \\ & \leqslant \frac{k^2}{2 } \int_0^t \int_0^t \sum_{\ell + m = k } e^{-(\ell^2 + m^2 ) | s - s ' | } { \mathrm{d}}s { \mathrm{d}}s ' \leqslant k^2 t \sum_{\ell + m = k } \int_0^{\infty } e^{- ( \ell^2 + m^2 ) r } { \mathrm{d}}r \\ & = k^2 t \sum_{\ell + m = k } \frac{1}{\ell^2 + m^2}. \end{aligned}\ ] ] now for @xmath82 @xmath179
so finally @xmath180
\lesssim | k |
t$ ] . redoing a similar computation in the case @xmath181 , we obtain @xmath182 \lesssim | k | \times | t - s |$ ] . to go from this estimate to a path wise regularity result of the distribution @xmath183 , following the line of reasoning of lemma [ lemma : ou - regularity ]
, we need to estimate the @xmath184-th moment of @xmath181 .
gaussian hypercontractivity ( see theorem 3.50 of @xcite ) tells us that all @xmath185 moments of polynomials of gaussian random variables are equivalent and in particular that @xmath186 \lesssim_p ( \mathbb{e } [ | j_t ( e_k ) - j_s ( e_k ) |^2])^p .\ ] ] from here we easily derive that almost surely @xmath187 .
this shows that @xmath188 exists as a space time distribution but not as a continuous function of time with values in distributions in space . the key point in the proof of lemma [ lem : ou square ]
is the fact that the ornstein uhlenbeck process decorrelates quite rapidly in time .
the construction of the process @xmath189 does not solve our problem since we need similar properties for the full solution @xmath29 of the non linear dynamics ( or for some approximations thereof ) , and all we have done so far relies on explicit computations and the specific gaussian features of the ornstein uhlenbeck process . but at least this give us a hint that indeed there could exist a way of making sense of the term @xmath190 , even if only as a space time distribution , and that in doing so we should exploit some decorrelation properties of the dynamics .
so when dealing with the full solution @xmath29 , we need a replacement for the gaussian computations used above .
this will be provided , in the current setting , by stochastic calculus along the time direction .
indeed , note that for each @xmath49 the process @xmath191 is a semimartingale in the filtration @xmath192 . before proceeding with these computations ,
we need to develop some tools to describe the it formula for functions of the ornstein uhlenbeck process
. this will be also serve us as an opportunity to set up some analysis on gaussian spaces . for cylindrical functions @xmath193 of the form @xmath194 with @xmath195 and @xmath196 at least @xmath197 , we have by it s formula @xmath198 where @xmath199 denotes the quadratic covariation of two continuous semimartingales and where @xmath200 and @xmath201 , with @xmath202 denoting the derivative with respect to the @xmath9-th argument .
now @xmath203 and then @xmath204 where @xmath205 is the second order differential operator defined on cylindrical functions @xmath206 as @xmath207 another way to describe the generator @xmath205 is to give its value on the functions @xmath208 for @xmath209 , which is @xmath210 if @xmath211 are two cylindrical functions ( which we can take of the form @xmath194 and @xmath212 for the same @xmath213 ) , we can check that @xmath214 where the quadratic form @xmath215 is given by @xmath216 ( gaussian integration by parts ) let @xmath217 be a @xmath56-dimensional gaussian vector with zero mean and covariance @xmath218 . then for all @xmath219 we have @xmath220 = \sum_{\ell } c_{k , \ell } \mathbb{e } \left [ \frac{\partial g ( z)}{\partial z_{\ell } } \right ] .\ ] ] use that @xmath221 = e^{- \langle \lambda , c \lambda \rangle / 2}$ ] and moreover that @xmath222 = ( - i ) \frac{\partial}{\partial \lambda_k } \mathbb{e } [ e^{i \langle z , \lambda \rangle } ] = ( - i ) \frac{\partial}{\partial \lambda_k } e^{- \langle \lambda , c \lambda \rangle / 2 } = i ( c \lambda)_k e^{- \langle \lambda , c \lambda \rangle / 2}\ ] ] @xmath223 = \sum_{\ell } c_{k , \ell } \mathbb{e } [ \frac{\partial}{\partial z_{\ell } } e^{i \langle z , \lambda \rangle } ] .\ ] ] the relation is true for trigonometric functions and taking fourier transforms we see that it holds for all @xmath224 .
is then a matter of taking limits to show that we can extend it to any @xmath225 . as a first application of this formula
let us show that @xmath226 = 0 $ ] for every cylindrical function , where @xmath227 is a space white noise .
indeed , note that @xmath228 = \frac{1}{2 } \langle \varphi_i , \delta \varphi_j \rangle_{l^2 ( \mathbb{t})}$ ] , leading to @xmath229,\end{aligned}\ ] ] so that @xmath230 = 0 $ ] . in combination with it s formula
, this shows that the white noise law is indeed a stationary distribution for @xmath69 ( check it ! ) . from now on we fix the initial distribution @xmath231 , which means that @xmath232 for all @xmath59 . as another application of the gaussian integration by parts formula , we get @xmath233 & = - \frac{1}{2 } \sum_{i , j } \mathbb{e } [ f_i ( \eta ) g_j ( \eta ) ] \langle \varphi_i , \delta \varphi_j \rangle_{l^2 ( \mathbb{t } ) } .
\\ & = - \frac{1}{2 } \sum_{i , j } \mathbb{e } [ ( f ( \eta ) g_j ( \eta))_i ] \langle \varphi_i , \delta \varphi_j \rangle_{l^2 ( \mathbb{t } ) } \\ & \qquad + \frac{1}{2 } \sum_{i , j } \mathbb{e } [ f ( \eta ) g_{i j } ( \eta ) ] \langle \varphi_i , \delta \varphi_j \rangle_{l^2 ( \mathbb{t } ) } \\ & = - \sum_j \mathbb{e } [ f ( \eta ) g_j ( \eta ) \eta ( \delta \varphi_j ) ] + \frac{1}{2 } \sum_{i , j } \mathbb{e } [ f ( \eta ) g_{i j } ( \eta ) ] \langle \varphi_i , \delta \varphi_j \rangle_{l^2 ( \mathbb{t } ) } \\ & = -\mathbb{e } [ ( fl_0 g ) ( \eta)].\end{aligned}\ ] ] combining this with ( [ eq : l0 and e ] ) and with @xmath234 = 0 $ ] , we obtain @xmath235 = \mathbb{e } [ ( gl_0 f ) ( \eta)]$ ] . that is , @xmath205 is a symmetric operator with respect to the law of @xmath236 .
consider now the operator @xmath237 , defined on cylindrical functions @xmath206 by @xmath238 so that @xmath239 takes values in @xmath79 , the continuous linear functionals on @xmath240 .
show that @xmath237 is independent of the specific representation of @xmath206 , that is if @xmath241 for all @xmath242 , then @xmath243 * hint : * one possible strategy is to show that for all @xmath244 , @xmath245 we have @xmath246 + \mathbb{e } [ g ( \eta ) \langle \psi , { \mathrm{d}}f ( \eta ) \rangle ] = \sum_i \mathbb{e } [ ( f g)_i
( \eta ) \langle \psi , \varphi_i \rangle ] = 2\mathbb{e } [ \eta ( \psi ) ( fg ) ( \eta)],\ ] ] and therefore @xmath246 = \mathbb{e } [ g ( \eta )
\langle \psi , { \mathrm{d}}^{\ast } f ( \eta ) \rangle]\ ] ] with @xmath247 being the adjoint of @xmath237 for the scalar product in @xmath248 .
let @xmath249 and similarly for @xmath250 .
let @xmath251 be an orthonormal basis of @xmath252 .
show that @xmath253 the commutator between @xmath254 and @xmath255 is given by @xmath256 f ( \rho ) { : = } ( { \mathrm{d}}_{\theta } { \mathrm{d}}^{\ast}_{\psi } - { \mathrm{d}}^{\ast}_{\psi } { \mathrm{d}}_{\theta } ) f ( \rho ) = 2 \langle \psi , \theta \rangle_{l^2 ( \mathbb{t } ) } f ( \rho),\ ] ] whereas @xmath257 = 0 $ ] .
therefore , @xmath258 = \frac{1}{2 } \sum_n [ { \mathrm{d}}^{\ast}_{e_n } { \mathrm{d}}_{\delta e_n } , { \mathrm{d}}^{\ast}_{\psi } ] = \frac{1}{2 } \sum_n { \mathrm{d}}^{\ast}_{e_n } [ { \mathrm{d}}_{\delta e_n } , { \mathrm{d}}^{\ast}_{\psi } ] + \frac{1}{2 } \sum_n [ { \mathrm{d}}^{\ast}_{e_n } , { \mathrm{d}}^{\ast}_{\psi } ] { \mathrm{d}}_{\delta e_n}\ ] ] @xmath259 so if @xmath260 is an eigenvector of @xmath261 with eigenvalue @xmath262 , then @xmath263 = \lambda { \mathrm{d}}^{\ast}_{\psi}$ ] .
let now @xmath264 be an orthonormal eigenbasis for @xmath261 with eigenvalues @xmath265 and consider the functions @xmath266 then @xmath267 where we used that @xmath268 .
so these functions are eigenfunctions for @xmath205 and the eigenvalues are all the possible combinations of @xmath269 for @xmath270 .
we have immediately that for different @xmath271 these functions are orthogonal in @xmath272 .
they are actually orthogonal as soon as the indices @xmath9 differ since in that case there is an index @xmath10 which is in one but not in the other and using the fact that @xmath273 is adjoint to @xmath274 and that @xmath275 if @xmath276 does not depend on @xmath277 we get the orthogonality .
the functions @xmath278 are polynomials and they are called _ wick polynomials_. [ lem : exp - d - star]for all @xmath209 , almost surely @xmath279 if @xmath206 is a cylindrical function of the form @xmath280 with @xmath281 , then @xmath282 = \mathbb{e } [ e^{{\mathrm{d}}_{\psi } } f ( \eta ) ] = \mathbb{e } [ f ( \eta + \psi ) ] = \mathbb{e } [ f ( \eta ) e^{2 \eta ( \psi ) - \| \psi \|^2}],\ ] ] where the second step follows from the fact that if we note @xmath283 we have @xmath284 and @xmath285 so that @xmath286 for all @xmath59 and in particular for @xmath287 . the last step is simply a gaussian change of variables .
indeed if we take @xmath288 and @xmath289 for @xmath290 we have @xmath291 = \mathbb{e } [ f ( \eta ( \psi ) + \langle \psi , \psi \rangle , \eta ( \varphi_2 ) , \ldots , \eta ( \varphi_m))]\ ] ] since @xmath292 for @xmath290 .
now observe that @xmath293 is independent of @xmath294 so that @xmath295 = \int_{\mathbb{r } } \frac{e^{- z^2 / \| \psi \|^2}}{\sqrt{\pi \| \psi \|^2 } } \mathbb{e } [ f ( z + \langle \psi , \psi \rangle , \eta ( \varphi_2 ) , \ldots , \eta ( \varphi_m))]\ ] ] @xmath296 = \mathbb{e } [ f ( \eta ) e^{2 \eta ( \psi ) - \| \psi \|^2 } ] .\ ] ] the wick polynomials @xmath297 form an orthogonal basis of @xmath248 . taking @xmath298 in lemma [ lem : exp - d - star ] , we get @xmath299 which is enough to show that any random variable in @xmath300 can be expanded in a series of wick polynomials showing that the wick polynomials are an orthogonal basis of @xmath248 ( but they are still not normalized ) . indeed assume that @xmath301 but @xmath302 for all @xmath303 , @xmath304 , then @xmath305 = e^{\sum_i \sigma_i^2 \| \psi_i \|^2 } \mathbb{e } [ z e^{2 \sum_i \sigma_i \eta ( \psi_i ) - \sum_i \sigma_i^2 \| \psi_i \|^2 } ] = \mathbb{e } [ z e^{2 \sum_i \sigma_i \eta ( \psi_i ) } ] .\ ] ] since the @xmath306 are arbitrary , this means that @xmath307 is orthogonal to any polynomial in @xmath236 and then that it is orthogonal also to @xmath308 .
so let @xmath309 and @xmath310 for @xmath311 , and observe that @xmath312 = \mathbb{e } [ z f ( \eta ( \psi_1 ) , \ldots , \eta ( \psi_m))],\ ] ] which means that @xmath307 is orthogonal to all the random variables in @xmath300 which are measurable with respect to the @xmath313field generated by @xmath314 .
this implies @xmath315 .
that is , wick polynomials form a basis for @xmath248 .
the first few ( un normalized ) wick polynomials are @xmath316 @xmath317 and @xmath318 some other properties of wick polynomials can be derived using the commutation relation between @xmath237 and @xmath319 . by linearity @xmath320 , so that using the symmetry of @xmath321 we get @xmath322 then note that by lemma [ lem : exp - d - star ] we have @xmath323 expanding the exponentials , @xmath324 and identifying the terms of the same homogeneity in @xmath45 and @xmath260 respectively we get @xmath325 this gives a general formula for such products . by polarization of this multilinear form
, we can also get a general formula for the products of general wick polynomials . indeed taking @xmath326 and @xmath327 for arbitrary real coefficients @xmath328 and @xmath329 , we have @xmath330 deriving this with respect to all the @xmath331 parameters and setting them to zero , we single out the term @xmath332 where @xmath333 denotes the symmetric group on @xmath334 , and where we used the symmetry of the wick polynomials . doing the same also for the right hand side of ( [ eq : wick polynomial product ] ) we get @xmath335 where the sum over @xmath336 runs over @xmath337 permutation of @xmath338 and similarly for @xmath339 .
in particular , we obtain @xmath340 = \frac{1}{n ! } \sum_{i , j } \prod_{r = 1}^n \langle \psi_{i_r } , \psi_{j_r } \rangle = \sum_{\sigma \in s_n } \prod_{r = 1}^n \langle \psi_r , \psi_{\sigma ( r ) } \rangle .\ ] ] in our problem it will be convenient to take the fourier basis as basis in the above computations .
let @xmath341 where @xmath342 and @xmath343 form together a real valued orthonormal basis for @xmath252 .
then @xmath344 whenever @xmath345 is real valued , and we will denote @xmath346 and similarly for @xmath347 . in this way , @xmath348 is the adjoint of @xmath349 with respect to the hermitian scalar product on @xmath350 and the ornstein uhlenbeck generator takes the form @xmath351 and @xmath352 we are ready now to start our computations . recall that we want to analyse @xmath353 using it calculus with respect to the ornstein uhlenbeck process
we want to understand @xmath354 as a correction term in it s formula , so we have to find a function @xmath206 such that @xmath355 .
of course , @xmath206 will not be a cylindrical function but we only defined @xmath205 on cylindrical functions .
so to make the following calculations rigorous we would again have to replace @xmath356 by @xmath357 and then pass to the limit , see the paper @xcite for details . as before we will perform the calculations already in the limit @xmath358 , in order to simplify the computations and not to obscure the ideas through technicalities .
note that @xmath359 where @xmath360 is a second order hermite polynomial so that @xmath361 by ( [ eq : l0-eigenvalues ] ) .
therefore , it is enough to take @xmath362 this corresponds to the distribution @xmath363 ( check it ! ) .
then @xmath364 where @xmath365 is a martingale with quadratic variation @xmath366 we can estimate @xmath367 \lesssim_p \mathbb{e } [ | m_{f , t } ( \varphi ) - m_{f , s } ( \varphi ) |^{2 p } ] + \mathbb{e } [ | f ( x_t ) ( \varphi ) - f ( x_s ) ( \varphi ) |^{2 p } ] .\ ] ] to bound the martingale expectation , we will use the following burkholder inequality : let @xmath368 be a continuous local martingale with @xmath369 .
then for all @xmath370 and @xmath371 , @xmath372 \leqslant c_p \mathbb{e } [ \langle m \rangle_t^p ] .\ ] ] start by assuming that @xmath368 and @xmath373 are bounded .
it s formula yields @xmath374 and therefore @xmath375 = c_p \mathbb{e } \left [ \int_0^t | m_s |^{2 p - 2 } { \mathrm{d}}\langle m \rangle_s \right ] \leqslant c_p \mathbb{e } [ \sup_{t \leqslant t } | m_t |^{2 p - 2 } \langle m \rangle_t ] .\ ] ] by cauchy
schwartz we get @xmath375 \leqslant c_p \mathbb{e } [ \sup_{t \leqslant t } | m_t |^{2 p}]^{(2 p - 2 ) / 2 p } \mathbb{e } [ \langle m \rangle_t^p]^{1 / p } .\ ] ] but now doob s @xmath185 inequality yields @xmath376 \leqslant c'_p \mathbb{e } [ | m_t |^{2 p}]$ ] , and this implies the claim in the bounded case . the unbounded case can be treated with a localization argument .
applying burkholder s inequality , we obtain @xmath377 & \lesssim_p \mathbb{e } \left [ \left| \int_s^t \mathcal{e } ( f ( \ast ) ( \varphi ) , f ( \ast ) ( \varphi ) ) ( x_r ) { \mathrm{d}}r \right|^p \right ] + \mathbb{e } [ | f ( x_t)(\varphi ) - f ( x_s ) ( \varphi ) |^{2 p } ] \\ & \leqslant ( t - s)^{p - 1 } \int_s^t \mathbb{e } [ | \mathcal{e } ( f ( \ast)(\varphi ) , f ( \ast ) ( \varphi ) ) ( x_r ) |^p ] { \mathrm{d}}r \\ & \qquad + \mathbb{e } [ | f ( x_t ) ( \varphi ) - f ( x_s ) ( \varphi ) |^{2 p } ] \\ & = ( t - s)^p \mathbb{e } [ | \mathcal{e } ( f ( \ast ) ( \varphi ) , f ( \ast)(\varphi ) ) ( \eta ) |^p ] + \mathbb{e } [ | f ( x_t ) ( \varphi ) - f ( x_s ) ( \varphi ) |^{2 p}],\end{aligned}\ ] ] using that @xmath378 . now @xmath379 and therefore @xmath380 which implies that @xmath381 \lesssim k^2 \mathbb{e } \sum_{\ell + m = k } \frac{| \eta ( e_{\ell } ) |^2}{\ell^2 + m^2 } \lesssim k^2 \sum_{\ell + m = k } \frac{1}{\ell^2 + m^2 } \lesssim | k | .\ ] ] a similar computation gives also that @xmath382 \lesssim | k |^p .\ ] ] further , we have @xmath383 & \lesssim k^2 \sum_{\ell + m = k } \mathbb{e } \left [ \frac{(h_{\ell , m } ( x_t ) - h_{\ell , m } ( x_s))^2}{(\ell^2 + m^2)^2 } \right ] \\ &
\lesssim k^2 | t - s | \sum_{\ell + m = k } \frac{m^2}{(\ell^2 + m^2)^2 } \lesssim | k |
| t - s |.\end{aligned}\ ] ] and finally , this computation lets us recover the result that @xmath186 \lesssim_p ( t - s)^p | k |^p .\ ] ] the advantage of the it trick with respect to the explicit gaussian computation is that it goes over to the non gaussian case . indeed note that while the boundary term @xmath384 has been estimated using a lot of the gaussian information about @xmath69 , we used only the law at a fixed time to handle the term @xmath385 . in order to carry over these computation to the full process @xmath29 solution of the non
linear dynamics we need to replace the generator of @xmath69 with that of @xmath29 and to have a way to handle the boundary terms .
the idea is now to reverse the markov process @xmath29 in time , which will allow us to kill the antisymmetric part of the generator and at the same time kill the boundary terms . indeed observe that if @xmath29 solves the stochastic burgers equation , then formally we have the it formula @xmath386 where @xmath387 is now the full generator of the non linear dynamics , given by @xmath388 where @xmath389 formally , the non linear term is antisymmetric with respect to the invariant measure of @xmath205 . indeed since @xmath390 is a first order operator @xmath391 = \mathbb{e } [ ( b ( f g ) ( \eta ) ) ] -\mathbb{e } [ f ( \eta ) ( b g ( \eta ) ) ] = -\mathbb{e } [ f ( \eta ) ( b g ( \eta))]\ ] ] provided @xmath392=0 $ ] for any cylinder function @xmath206 .
let us show this .
@xmath393 & = \sum_k \mathbb{e } [ ( \partial_x \eta^2 ) ( e_k ) { \mathrm{d}}_k f ( \eta)]\\ & = \sum_k \mathbb{e } [ ( { \mathrm{d}}_k ( \partial_x \eta^2 ) ( e_k ) ) f ( \eta ) ] + \sum_k \mathbb{e } [ { \mathrm{d}}_k [ ( \partial_x \eta^2 ) ( e_k ) f ( \eta ) ] ] \end{aligned}\ ] ] but now , using that @xmath394 and that @xmath395 we have @xmath396 and using the gaussian integration by parts @xmath393 & = \sum_k \mathbb{e } [ { \mathrm{d}}_k [ ( \partial_x \eta^2 ) ( e_k ) f ( \eta ) ] ] = - \sum_k \mathbb{e } [ \eta ( e_k ) ( \partial_x \eta^2 ) ( e_k ) f ( \eta ) ] \\ & = -\mathbb{e } [ \langle \eta , \partial_x \eta^2 \rangle f ( \eta ) ] = - \frac{1}{3 } \mathbb{e } [ \langle 1 , \partial_x \eta^3 \rangle f ( \eta ) ] = 0 \end{aligned}\ ] ] since @xmath397 . the dynamics of @xmath29 backwards in time has a markovian description which is the subject of the next exercise .
let @xmath398 be a stationary markov process on a polish space , with semigroup @xmath399 and stationary distribution @xmath400 .
show that if @xmath401 is the adjoint of @xmath402 in @xmath403 , then @xmath404 is a semigroup of operators on @xmath403 ( that is @xmath405 and @xmath406 as operators on @xmath407 ) .
show that if @xmath408 , then for all @xmath409 the process @xmath410 , @xmath411 $ ] , is also markov , with semigroup @xmath412}$ ] , and that @xmath400 is also an invariant distribution for @xmath404 .
show also that if @xmath413 has generator @xmath387 then @xmath404 has generator @xmath414 which is the adjoint of @xmath387 with respect to @xmath407 .
now if we reverse the process in time letting @xmath415 , we have by stationarity @xmath416 = \mathbb{e } [ f ( u_{t - t } ) g ( u_t ) ] = \mathbb{e } [ f ( u_0 ) g ( u_t ) ] .\ ] ] so if we denote by @xmath417 the generator of @xmath418 : @xmath419 = \left .
\frac{{\mathrm{d}}}{{\mathrm{d}}t } \right|_{t = 0 } \mathbb{e } [ f ( \hat{u}_t ) g ( \hat{u}_0 ) ] = \left . \frac{{\mathrm{d}}}{{\mathrm{d}}t } \right|_{t = 0 } \mathbb{e } [ f ( u_0 ) g ( u_t ) ] = \mathbb{e } [ l g ( u_0 ) f ( u_0)],\ ] ] which means that @xmath417 is the adjoint of @xmath387 , that is @xmath420 then it s formula for the reversed process gives @xmath421 where for all test functions @xmath45 , the process @xmath422 is a martingale in the filtration of @xmath418 with covariance @xmath423 combining the it formulas for @xmath29 and @xmath418 , we get @xmath424 and @xmath425 and summing up these two equalities gives @xmath426 that is @xmath427 an added benefit of this forward
backward representation is that only term which required quite a lot of informations about @xmath69 , that is the boundary term @xmath384 does not appear at all now . as above
if @xmath428 , we end up with @xmath429 perform a similar formal calculation as in ( [ eq : antisymmetry of nonlinear ] ) to see that @xmath430 = 0 $ ] for all cylindrical functions @xmath206 , so that @xmath236 is also invariant for the stochastic burgers equation .
combine this with ( [ eq : martingale trick sbe ] ) to show that setting @xmath431 we have @xmath432 \lesssim_p ( t - s)^p | k |^p\ ] ] and letting @xmath433 we get @xmath434 \lesssim_p ( | k | / n)^{\varepsilon p } ( t - s)^p | k |^p\ ] ] for all @xmath435 .
use this to derive that @xmath436)^{1 / 2 p } \lesssim_{p , \alpha } n^{- \varepsilon / 2 } ( t - s)^{1 / 2}\ ] ] for all @xmath437 , and realize that this estimate allows you to prove compactness of the approximations @xmath438 and then convergence to a limit @xmath439 in @xmath440 .
let us cook up a definition which will allow us to rigorously perform the formal computations above in a general setting .
let @xmath441 be a couple of generalized process such that for all @xmath442 the process @xmath443 is a continuous semimartingale satisfying @xmath444 where @xmath445 is a martingale with quadratic variation @xmath446 and @xmath447 is a finite variation process with @xmath448 . for all @xmath59
the random distribution @xmath449 is a zero mean space white noise with variance @xmath450 . for any @xmath409
the reversed process @xmath415 has again properties @xmath451 with martingale @xmath452 and finite variation part @xmath453 such that @xmath454 .
any pair of processes @xmath455 satisfying these condition will be called _ controlled by the ornstein uhlenbeck process _ and we will denote the set of all such processes with @xmath456 .
assume that @xmath457 and for any @xmath458 , @xmath59 , @xmath49 let @xmath459 then for any @xmath460 @xmath461 converges in probability to a space time distribution @xmath462 .
we are now at a point were we can give a meaning to our original equation .
a pair of random distribution @xmath457 is an _ energy solution _ to the stochastic burgers equation if it satisfies @xmath463 for all @xmath59 and @xmath49 .
that is if @xmath464 .
now we are in a relatively standard setting of needing to prove existence and uniqueness of such energy solutions .
note that in general the solutions are _ pairs _ of processes @xmath455 .
the notion of energy solution has been introduced ( in a slightly different way ) in the work of gonalves and jara @xcite on macroscopic universal fluctuations of weakly asymmetric interacting particle systems . for the existence
the way to proceed is quite standard .
we approximate the equation , construct approximate solutions and then try to have enough compactness to have limiting points which then naturally will satisfy the requirements for energy solutions . for any @xmath465 consider solutions @xmath466
to @xmath467 these are generalized functions such that @xmath468 ( e_k ) { \mathrm{d}}t + ik { \mathrm{d}}\beta_t ( k)\ ] ] for @xmath469 and @xmath59 .
we take @xmath470 to be the white noise with covariance @xmath471 .
the point of our choice of the non linearity is that this ( infinite dimensional ) system of equations decomposes into a finite dimensional system for @xmath472 and an infinite number of one dimensional equations for each @xmath473 with @xmath474 .
indeed if @xmath474 we have @xmath475 ( e_k ) = 0 $ ] so @xmath476 the ornstein uhlenbeck process with initial condition @xmath477 which renders it stationary in time ( check it ) .
the equation for @xmath478 reads @xmath479 where @xmath480 this is a standard finite dimensional ode having global solutions for all initial conditions which gives rise to a nice markov process .
the fact that solutions do not blow up even if the interaction is quadratic can be seen by computing the evolution of the norm @xmath481 and by showing that @xmath482 but now @xmath483 and by symmetry of this expression it is equal to @xmath484 so @xmath485 where @xmath486 .
now @xmath487 \lesssim \int_0^t \sum_{| k | \leqslant n } k^2 | v^n_t ( k ) |^2 { \mathrm{d}}t \lesssim
n^2 \int_0^t a_t { \mathrm{d}}t\ ] ] and then by martingales inequalities @xmath488 } ( a_t)^2 ] & \leqslant 2\mathbb{e } [ a_0 ^ 2 ] + 2\mathbb{e } [ \sup_{t \in [ 0 , t ] } ( m_t)^2 ] \leqslant 2\mathbb{e } [ a_0 ^ 2]^ { } + 2\mathbb{e } [ m_t^2]^ { } \\ & \leqslant 2\mathbb{e } [ a_0 ^ 2]^ { } + cn^2 \int_0^t \mathbb{e } ( a_t ) { \mathrm{d}}t.\end{aligned}\ ] ] now gronwall s inequality gives @xmath489 } ( a_t)^2 ] \lesssim e^{cn^2 t } \mathbb{e } [ a_0 ^ 2],\ ] ] from where we can deduce ( by a continuation argument ) that almost surely there is no blowup at finite time for the dynamics . from the galerkin approximations the it trick will provide enough compactness in order to pass to the limit and build an energy solution to the stochastic burgers equation ; see @xcite for details on how to implement the it trick on the level of diffusions .
here we collect some classical results from harmonic analysis which we will need in the following .
we concentrate on distributions and spdes on the torus , but everything in this section applies mutatis mutandis on the full space @xmath490 , see @xcite .
the only problem is that then the stochastic terms will no longer be in the besov spaces @xmath32 which we encounter below but rather in weighted besov spaces .
handling spdes in weighted function spaces is more delicate and we prefer here to concentrate to the simpler situation of the torus .
the space of distributions @xmath491 is the set of linear maps @xmath492 from @xmath493 to @xmath494 , such that there exist @xmath495 and @xmath1 with @xmath496 for all @xmath49 .
clearly @xmath497 for all @xmath460 , and more generally the space of finite signed measures on @xmath498 is contained in @xmath79 .
another example of a distribution is @xmath499 for @xmath17 and @xmath500 .
in particular , the fourier transform @xmath501 , @xmath502 with @xmath503 is defined for all @xmath504 , and it satisfies @xmath505 for a suitable polynomial @xmath506 .
conversely , if @xmath507 is at most of polynomial growth , then its inverse fourier transform @xmath508 defines a distribution .
[ exo : fourier inversion]show that the fourier transform of @xmath509 decays faster than any rational function ( we say that it is of _ rapid decay _ ) .
combine this with the fact that @xmath510 defines a bijection from @xmath511 to @xmath512 with inverse @xmath513 to show that @xmath514 for all @xmath504 and @xmath515 for all @xmath516 of polynomial growth .
extend the parseval formula @xmath517 from @xmath518 to @xmath504 and @xmath509 .
[ exo : wn]fix a complete probability space @xmath519 . on that space
let @xmath38 be a spatial white noise on @xmath520 , i.e. @xmath38 is a centered gaussian process indexed by @xmath521 , with covariance @xmath40 = \int_{\mathbb{t}^d } f ( x ) g ( x ) { \mathrm{d}}x.\ ] ] show that there exists @xmath522 with @xmath523 for all @xmath524 , such that @xmath525 for all @xmath526 .
* hint : * show that
@xmath527 < \infty$ ] for some suitable @xmath528
. linear maps on @xmath79 can be defined by duality : if @xmath529 is such that for all @xmath495 there exists @xmath93 and @xmath1 with @xmath530 , then we set @xmath531 .
differential operators are defined by @xmath532 .
if @xmath533 grows at most polynomially , then it defines a _ fourier multiplier _ @xmath534 [ exercise : convolution]use the fourier inversion formula of exercise [ exo : fourier inversion ] to show that for @xmath504 , @xmath535 and for @xmath536 with @xmath29 of polynomial growth and @xmath537 of rapid decay @xmath538 we will now use littlewood paley blocks to obtain a decomposition of distributions into an infinite series of smooth functions .
of course , we have already such a decomposition at our disposal : @xmath539 .
but it turns out to be convenient not to consider each fourier coefficient separately , but to work with projections on dyadic fourier blocks .
a _ dyadic partition of unity _ consists of two nonnegative radial functions @xmath540 , where @xmath345 is supported in a ball @xmath541 and @xmath345 is supported in an annulus @xmath542 for suitable @xmath543 , such that 1 .
@xmath544 and 2 .
[ property-2-dyadic]@xmath545 for @xmath546 and @xmath547 for all @xmath548 with @xmath549 .
we will often write @xmath550 and @xmath551 for @xmath552 . dyadic partitions of unity exist , see @xcite .
the reason for considering smooth partitions rather than indicator functions is that indicator functions do not have good fourier multiplier properties .
for example we only have @xmath553 , whereas @xmath554 uniformly in @xmath10 . from now on we fix a dyadic partition of unity @xmath555 and define the dyadic blocks @xmath556 we also use the notation @xmath557 as well as @xmath558 so that @xmath559 every dyadic block has a compactly supported fourier transform and is therefore in @xmath240 .
it is easy to see that @xmath560 for all @xmath504 . for @xmath100 , the hlder - besov space @xmath32 is given by @xmath561 , where for @xmath562 $ ] we define @xmath563 with the usual interpretation as @xmath564 norm if @xmath565 .
then @xmath566 is a banach space and while the norm @xmath567 depends on @xmath555 , the space @xmath566 does not and any other dyadic partition of unity corresponds to an equivalent norm ( this follows from lemma [ lem : besov regularity of series ] below ) .
we write @xmath568 instead of @xmath569 .
let @xmath570 denote the dirac delta in 0 . show that @xmath571 . if @xmath572 , then @xmath32 is the space of @xmath25 times differentiable functions whose partial derivatives of order @xmath25 are ( @xmath573)-hlder continuous ( see page 99 of @xcite ) .
note however , that for @xmath495 the space @xmath574 is strictly larger than @xmath575 , the space of @xmath576 times continuously differentiable functions . below we will give the proof for @xmath577 , but before we still need some tools .
[ lem : poisson](poisson summation ) let @xmath578 be a schwartz function .
then @xmath579 for all @xmath580 , where @xmath581 .
let @xmath582 .
the function @xmath583 is of rapid decay since @xmath49 so the sum converges absolutely and defines a continuous function @xmath584 which is periodic of period @xmath585 in every direction .
the fourier transform over the torus @xmath520 of this function is @xmath586 since @xmath587 . by dominated convergence
the sum and the integral can be combined in a overall integration over @xmath490 : @xmath588 so we deduce that @xmath589 .
[ exercise : besov - space - inequalities]show that @xmath590 for @xmath591 , that @xmath592 for @xmath22 , that @xmath593 for @xmath594 , and that @xmath595 for @xmath596 .
* hint : * when proving @xmath597 for @xmath598 , you might need poisson s summation formula .
the following bernstein inequality is extremely useful when dealing with functions with compactly supported fourier transform .
[ lemma : bernstein](bernstein inequality ) let @xmath599 be a ball and @xmath600 .
for any @xmath601 , @xmath602 , and @xmath603 with @xmath604 we have @xmath605 let @xmath260 be a compactly supported @xmath63 function on @xmath490 such that @xmath606 on @xmath599 and write @xmath607
. then @xmath608 by young s inequality , we get @xmath609 where @xmath610 .
now it is a simple exercise to verify @xmath611 , and @xmath612 whereas @xmath613 we end up with @xmath614 it then follows immediately that for @xmath100 , @xmath615 , @xmath17 , we have @xmath616 . another simple application of the bernstein inequalities is the besov embedding theorem , the proof of which we leave as an exercise .
[ lem : besov embedding](besov embedding ) let @xmath617 and @xmath618 , and let @xmath100 . then @xmath619 is continuously embedded into @xmath620 .
[ exo : wn besov]in the setting of exercise [ exo : wn ] , use besov embedding to show that @xmath621 < \infty$ ] for all @xmath460 and @xmath31 ( in particular @xmath622 almost surely ) .
* hint : * estimate @xmath623 $ ] using gaussian hypercontractivity ( equivalence of moments ) . as another application of the bernstein inequality ,
let us show that @xmath624 for @xmath577 .
for @xmath577 we have @xmath624 , the space of @xmath625-hlder continuous functions , and @xmath626 where @xmath627 denotes the canonical distance on @xmath520 .
start by noting that for @xmath628 we have @xmath629 .
let now @xmath630 and choose @xmath631 with @xmath632 .
for @xmath633 we use bernstein s inequality to obtain @xmath634 whereas for @xmath635 we simply estimate @xmath636 summing over @xmath10 , we get @xmath637 conversely , if @xmath638 , then we estimate @xmath639 . for @xmath552 ,
the function @xmath640 satisfies @xmath641 , and therefore @xmath642 now @xmath643 , and thus we end up with @xmath644 the following lemma , a characterization of besov regularity for functions that can be decomposed into pieces which are localized in fourier space , will be immensely useful in what follows .
[ lem : besov regularity of series]@xmath645 let @xmath646 be an annulus , let @xmath647 , and let @xmath648 be a sequence of smooth functions such that @xmath649 has its support in @xmath650 , and such that @xmath651 for all @xmath10 .
then @xmath652 let @xmath599 be a ball , let @xmath22 , and let @xmath648 be a sequence of smooth functions such that @xmath649 has its support in @xmath653 , and such that @xmath654 for all @xmath10 .
then @xmath652 if @xmath649 is supported in @xmath650 , then @xmath655 only for @xmath656 .
hence , we obtain @xmath657 if @xmath649 is supported in @xmath653 , then @xmath655 only for @xmath11 .
therefore , @xmath658 using @xmath22 in the last step . when solving spdes , we will need the smoothing properties of the heat semigroup .
for that purpose we study functions of time with values in distribution spaces . if @xmath659 is a banach space with norm @xmath660 and if @xmath409 , then we define @xmath661 and @xmath662 as the spaces of continuous functions from @xmath663 respectively @xmath664 $ ] to @xmath659 , and @xmath665 is equipped with the supremum norm @xmath666 .
if @xmath667 then we write @xmath668 for the functions in @xmath661 that are @xmath625hlder continuous on every interval @xmath669 $ ] , and we write @xmath670 we then define @xmath671 for @xmath672 . for @xmath409
we set @xmath673 and we equip @xmath674 with the norm @xmath675 the notation @xmath676 is chosen to be reminiscent of the operator @xmath677 and indeed the parabolic spaces @xmath676 are adapted to @xmath678 in the sense that the temporal regularity `` counts twice '' , which is due to the fact that @xmath678 contains a first order temporal but a second order spatial derivative .
if we would replace @xmath261 by a fractional laplacian @xmath679 , then we would have to consider the space @xmath680 instead of @xmath676 .
we have the following schauder estimate on the scale of @xmath681 spaces : [ lem : schauder]let @xmath672 and let @xmath399 be the semigroup generated by the periodic laplacian , @xmath682 . for @xmath683 define @xmath684 .
then @xmath685 for all @xmath409 .
if @xmath686 , then @xmath687 and @xmath688 * bibliographic notes * for a gentle introduction to littlewood paley theory and besov spaces see the recent monograph @xcite , where most of our results are taken from . there the case of tempered distributions on @xmath490 is considered .
the theory on the torus is developed in @xcite .
the schauder estimates for the heat semigroup are classical and can be found in @xcite .
let us consider the following @xmath689-dimensional homogenization problem .
fix @xmath31 and let @xmath690 be the solution to the cauchy problem @xmath691 where @xmath692 is a random field defined on the rescaled torus @xmath693 .
this model describes the diffusion of particles in a random medium ( replacing @xmath694 by @xmath695 gives the schrdinger equation of a quantum particle evolving in a random potential ) . for
a review of related results the reader can give a look at the recent paper of bal and gu @xcite . the limit @xmath696 corresponds to looking at the large scale behavior of the model since it can be understood as the equation for the _ macroscopic _ density @xmath697 which corresponds to a _
density @xmath698 evolving according to the parabolic equation @xmath699 we assume that @xmath692 is gaussian and has mean zero and homogeneous correlation function @xmath700 given by @xmath701 = ( \varepsilon / \sqrt{2 \pi } ) ^d \sum_{k \in \varepsilon
\mathbb{z}^d } e^{i \langle x - y , k \rangle } r ( k).\ ] ] on @xmath702 we make the following hypothesis : for some @xmath703 $ ] we have @xmath704 where @xmath705 is a smooth radial function of rapid decay .
for @xmath706 it would be equivalent to require that spatial correlations ( in the limit @xmath696 ) decay as @xmath707 .
for @xmath708 this hypothesis means that spatial correlations are of rapid decay .
indeed by dominated convergence @xmath709 now @xmath710 and @xmath711 if @xmath712 ( see for example proposition 1.29 of @xcite ) , so @xmath713 for @xmath714 .
let us write @xmath715 so that @xmath716 , and let us compute the variance of the littlewood
paley blocks of @xmath717 . in order to perform more easily some computations
we can introduce a family of centered complex gaussian random variables @xmath718 such that @xmath719 and @xmath720 = \delta_{k + k ' = 0}$ ] and represent @xmath721 as @xmath722 assume @xmath723.we have for any @xmath31 and @xmath724 and any @xmath725 : @xmath726 \lesssim 2^{(2 \alpha + \kappa ) i } \varepsilon^{\kappa } .\ ] ] this estimate implies that if @xmath727 , there exists @xmath728 with @xmath729 in @xmath730 as @xmath696 .
a spectral computation gives @xmath731 so @xmath732 & = & \varepsilon^d ( \sqrt{2 \pi } ) ^d \varepsilon^{- 2 \alpha } \sum_{k \in \varepsilon \mathbb{z}^d } \rho_i ( k / \varepsilon)^2 e^{i \langle x , k / \varepsilon \rangle } r ( k)\\ & = & ( \sqrt{2 \pi } ) ^d \varepsilon^{d - 2 \alpha } \sum_{k
\in \varepsilon \mathbb{z}^d } \rho ( k / ( \varepsilon 2^i))^2 r ( k)\\ & \lesssim & \varepsilon^{d - 2 \alpha } 2^{id } \sup_{k \in \varepsilon 2^i { \mathscr{a } } } r ( k ) , \end{array } \label{eq : est - lp - v}\ ] ] where @xmath646 is the annulus in which @xmath345 is supported . now if @xmath733 we have @xmath734 \lesssim 2^{id } \varepsilon^{d - 2 \alpha } ( \varepsilon 2^i)^{\beta - d } = \varepsilon^{\beta - 2 \alpha } 2^{i \beta}$ ] .
the assumption @xmath735 then implies @xmath736 \lesssim 2^{(2 \alpha + \kappa ) i } \varepsilon^{\kappa}$ ] for any @xmath737 . in the case
@xmath738 we use that @xmath739 to estimate @xmath740 and then @xmath734 \lesssim \varepsilon^{- 2 \alpha } \lesssim 2^{2 \alpha i } ( \varepsilon 2^i)^{\kappa}$ ] for any small @xmath728 .
note that the computation carried out in equation ( [ eq : est - lp - v ] ) implies also that if @xmath741 , then essentially @xmath717 does not converge in any reasonable sense since the variance of the littlewood
paley blocks explodes .
[ rmk : different v blocks orthogonal]the same calculation as in ( [ eq : est - lp - v ] ) shows that @xmath742 = 0\ ] ] whenever @xmath743 , because in that case @xmath744 .
the previous analysis shows that it is reasonable to take @xmath745 in order to have some hope of obtaining a well defined limit as @xmath696 . in this case
@xmath717 stays bounded ( at least ) in spaces of distributions of regularity @xmath746 .
this brings us to the problem of obtaining estimates for the parabolic pde @xmath747 \times \mathbb{t}^d,\ ] ] depending only on negative regularity norms of @xmath717 .
on one side the regularity of @xmath748 is then limited by the regularity of the right hand side which can not be better than that of @xmath717 . on the other side
the product of @xmath717 with @xmath749 can cause problems since we try to multiply an ( a priori ) irregular object with one of limited regularity
. assume that @xmath750 with @xmath751 .
it is then reasonable to assume that also @xmath752 and that @xmath753 as a consequence of the regularising effect of the heat operator ( lemma [ lem : schauder ] ) .
we will see in section [ ssec : bony ] below that the product @xmath754 is under control only if @xmath755 , that is if @xmath756 . since @xmath757 in @xmath758 , it is not difficult to show that @xmath748 converges as @xmath696 to the solution @xmath29 of the linear equation @xmath759 . in this case
the random potential will not have any effect in the limit .
the interesting situation then is when @xmath760 . to understand what could happen in this case
let us use a simple transformation of the solution . write @xmath761 where @xmath762 satisfies the equation @xmath763 with initial condition @xmath764 .
then @xmath765 since @xmath766 on @xmath664 \times \mathbb{t}^d$ ] , this implies that @xmath767 satisfies @xmath768 \times \mathbb{t}^d .\ ] ] our schauder estimates imply that @xmath769 with uniform bounds in @xmath31 , so that the problematic term is @xmath770 for which this estimate does not guarantee existence .
note that @xmath771 where @xmath772 and where @xmath773}{| k / \varepsilon |^2 } \sqrt{r ( k ) } .\ ] ] assume that @xmath774 then if @xmath775 and @xmath776 we have @xmath777 = \sigma^2,\ ] ] and if @xmath778 and @xmath776 @xmath779 = 0.\ ] ] moreover @xmath780 \lesssim \varepsilon^{4 - 4 \alpha } \min(\sigma^4 , ( \varepsilon 2^q)^{\beta - 2 } \| \tilde r \|_{\infty } \sigma^2 ) .\ ] ] a computation similar to that leading to equation ( [ eq : est - lp - v ] ) gives @xmath781 & = \varepsilon^d ( \sqrt{2 \pi } ) ^d \varepsilon^{- 2 \alpha } \sum_{k \in \varepsilon \mathbb{z}^d_0 } | k / \varepsilon |^2 \big [ \int_0^t e^{-(t - s ) | k / \varepsilon |^2 } { \mathrm{d}}s \big]^2 r ( k ) \\ & = \varepsilon^d ( \sqrt{2 \pi } ) ^d \varepsilon^{2 - 2 \alpha } \sum_{k \in \varepsilon \mathbb{z}^d_0 } \frac{[1 - e^{- t ( k / \varepsilon)^2}]^2}{k^2 } r ( k ) , \end{aligned}\ ] ] which for @xmath696 and @xmath782 tends to @xmath777 = \mathbb{i}_{\alpha < 1 } ( \sqrt{2 \pi})^d \int_{\mathbb{r}^d } \frac{r ( k)}{k^2 } { \mathrm{d}}k = \mathbb{i}_{\alpha < 1 } \sigma^2 .\ ] ] let us now study the variance of @xmath783 . using equation
we have @xmath784 by wick s theorem ( @xcite , theorem 1.28 ) @xmath785 \mathbb{e } [ g ( k_2 ) g
( k_2 ' ) ] + \mathbb{e } [ g ( k_1 ) g ( k_2 ' ) ] \mathbb{e } [ g ( k_2 ) g ( k_1 ' ) ] \\ & = \mathbb{i}_{k_1 + k_1 ' = k_2 + k_2 ' = 0 } + \mathbb{i}_{k_1 + k_2 ' = k_2 + k_1 ' = 0 } , \end{aligned}\ ] ] which implies @xmath780 = \frac{\varepsilon^{2d - 4\alpha}}{(2 \pi ) ^{d } } \sum_{k_1 , k_2 \in \varepsilon \mathbb{z}^d_0 } ( \rho_q ( ( k_1 + k_2 ) / \varepsilon))^2 | g_\varepsilon ( t , k_1 ) |^2 | g_\varepsilon ( t , k_2 ) |^2 .\ ] ] for any @xmath786 ( the case @xmath787 is left to the reader ) , the variables @xmath788 and @xmath789 are bounded away from 0 and we have @xmath780 \lesssim \varepsilon^{2 d + 4 - 4 \alpha } \sum_{k_1 , k_2 \in \varepsilon \mathbb{z}^d_0 } ( \rho_q ( ( k_1 + k_2 ) / \varepsilon))^2 \frac{| r ( k_1 ) | | r ( k_2 ) |}{| k_1 |^2
| k_2 |^2 } .\ ] ] a first estimate is obtained by just dropping the factor @xmath790 and results in the bound @xmath780 \lesssim \varepsilon^{2 d + 4 - 4 \alpha } \sum_{k_1 , k_2 \in \varepsilon \mathbb{z}^d_0 } \frac{| r ( k_1 ) |
| r ( k_2 ) |}{|
k_1 |^2 | k_2 |^2 } \lesssim \varepsilon^ { 4 - 4 \alpha } \sigma^4\ ] ] another estimate proceed by taking into account the constraint given by the support of @xmath790 . in order to satisfy @xmath791
we must have @xmath792 or @xmath793 .
in the first case @xmath794 since @xmath795 .
if @xmath796 we similarly have @xmath797 this lemma shows that the interesting situation is @xmath775 .
then provided @xmath798 and @xmath799 we have @xmath800 in @xmath801 for all @xmath782 , and in fact the convergence is uniform on @xmath802 $ ] for all @xmath803 .
an easy consequence of this is that @xmath767 converges to the solution of the pde @xmath804 and since @xmath805 in @xmath806 we finally obtain the convergence of @xmath807 to the same @xmath537 . the case @xmath775 and @xmath808 remains open in the previous analysis .
when @xmath808 we can not expect @xmath809 to be finite and moreover from the above computations we see that the variance of @xmath810 remains finite and does not go to zero so the limiting object should satisfy a stochastic pde rather than a deterministic one .
if we let @xmath811 $ ] ( which depends on time but which is easily shown to be independent of @xmath812 ) , then we expect that solving the _ renormalized _ equation @xmath813 should give rise in the limit to a well defined random field @xmath814 satisfying @xmath815 , where @xmath816 and where @xmath69 is the limit of @xmath762 as @xmath696 while @xmath236 is the limit of @xmath817 .
the relation of @xmath748 with @xmath818 is easily seen to be @xmath819 .
the renormalization procedure is therefore equivalent to a time
dependent rescaling of the solution to the initial problem .
we will study the renormalization and convergence problem for a more general equation of the form @xmath820 where @xmath821 is a sufficiently smooth non linearity .
one possible motivation is that if @xmath822 solves the linear pde and we set @xmath823 for some invertible @xmath824 such that @xmath825 , then @xmath826 and thus @xmath748 satisfies the pde @xmath827 where @xmath828 and @xmath829 . in the situation
we are interested in , the second term in the right hand side is simpler to treat than the first term so , for the time being , we will drop it and we will concentrate on the equation ( [ eq : gpam ] ) in @xmath830 with @xmath775 and short ranged ( @xmath708 ) potential @xmath165 which we refer to as _
generalized parabolic anderson model _ ( gpam ) . under these conditions
@xmath717 converges to the white noise in space which we usually denote with @xmath38 and our aim will be to set up a theory in which the non linear operations involved in the definition of the dynamics of the gpam are well defined , including the possibility of the renormalization which already appears in the linear case as hinted above . while the reader should always have in mind a limiting procedure from a well defined model like the ones we were considering so far , in the following we will mostly discuss the limiting equation .
the specific phenomena appearing when trying to track the oscillations of the term @xmath831 as @xmath696 will be described by a _ renormalized product _ @xmath832 and
so we write the gpam as @xmath833 in the linear case @xmath834 , the problem of the renormalization can be solved along the lines suggested above .
another possible line of attack comes from the theory of gaussian spaces and in particular from wick products , see for example @xcite .
however , the definition of the wick product relies on the concrete chaos expansion of its factors , and since nonlinear functions change the chaos expansion in a complicated way , there is little hope of directly extending the wick product approach to the nonlinear case and moreover using these non local ( in the probability space ) objects can deliver solutions which are not physically acceptable @xcite .
equation ( [ eq : pam - intro ] ) is structurally very similar to the stochastic differential equation @xmath835 where @xmath836 denotes a fractional brownian motion with hurst index @xmath837 .
there are many ways to solve ( [ eq : sde - intro ] ) in the brownian case .
since we are interested in a way that might extend to ( [ eq : pam - intro ] ) where the irregularity appears along the two dimensional spatial variable @xmath7 , we should exclude all approaches based on information , filtrations , and a direction of time ; in particular , any approach that works for @xmath838 might seem promising .
but lyons theory of rough paths @xcite equips us exactly with the techniques we need to solve ( [ eq : sde - intro ] ) for general @xmath321 .
more precisely , if for @xmath839 we are given @xmath840 , then we can use the controlled rough path integral @xcite to make sense of @xmath841 for any @xmath492 which `` looks like '' @xmath836 , and this allows us to solve ( [ eq : sde - intro ] ) .
so the main ingredients required for controlled rough paths are the integral @xmath840 for the reference path @xmath836 , and the fact that we can describe paths which look like @xmath836 .
it is worthwhile to note that while we need probability theory to construct @xmath840 , the construction of @xmath841 is achieved using pathwise arguments and it is given as a continuous map of @xmath492 and @xmath842 . as a consequence , the solution to the sde ( [ eq : sde - intro ] ) depends pathwise continuously on @xmath843 . by the structural similarity of ( [ eq : pam - intro ] ) and ( [ eq : sde - intro ] ) , we might hope to extend the rough path approach to ( [ eq : pam - intro ] ) .
the equivalent of @xmath836 is given by the solution @xmath844 to @xmath845 , @xmath846 , so that we should assume the renormalized product @xmath847 as given .
then we might hope to define @xmath848 for all @xmath492 that `` look like @xmath844 '' , however this is to be interpreted .
of course , rough paths can only be applied to functions of a one dimensional index variable , while for ( [ eq : pam - intro ] ) the problem lies in the irregularity of @xmath38 in the spatial variable @xmath849 . in the following
we combine the ideas from controlled rough paths with bony s paraproduct , a tool from functional analysis that allows us to extend them to functions of a multidimensional parameter . using the paraproduct , we are able to make precise in a simple way what we mean by `` distributions looking like a reference distribution '' .
we can then define products of suitable distributions and solve ( [ eq : pam - intro ] ) as well as many other interesting singular spdes . keeping the homogenization problem as leitmotiv for these lectures
, we could consider also space time varying environments @xmath850 .
the scaling of the temporal variable is chosen so that it is compatible with the diffusive scaling from a microscopic description , where @xmath851 has typical variation in space and time in scales of order @xmath852 .
assume that @xmath853 , then when the random field @xmath165 is gaussian , zero mean , and with short range space time correlations , the natural choice for the magnitude of the macroscopic fluctuations is @xmath854 .
in this case @xmath717 converges as @xmath696 to a space time white noise @xmath38 . understanding the limit dynamics as @xmath696 of the solution @xmath748 to the linear equation @xmath855 represents now a more difficult problem than in the time independent situation .
a gaussian computation shows that the random field @xmath762 , solution to @xmath856 ( e.g. with zero initial condition ) , stays bounded in @xmath857 as @xmath696 .
since @xmath678 is a second order operator ( if we use an appropriate parabolic weighting of the time and space regularities ) , @xmath38 is expected to live in a space of distributions of regularity @xmath858 .
this is to be compared with the @xmath859 of the space white noise which had to be dealt with in the gpam .
renormalization effects are then expected to be stronger in this setting and the limiting object , which we denote with @xmath860 , should satisfy a ( suitably renormalized ) linear stochastic heat equation with multiplicative noise ( she ) @xmath861 as hinted by the computations in the more regular case , it is useful to consider the change of variables @xmath862 which is called cole hopf transformation . here
@xmath863 is a new unknown which satisfies now the kardar
zhang ( kpz ) equation : @xmath864 where the difficulty comes now from the squaring of the derivative but which has the nice feature to be additively perturbed by the space
time white noise , a feature which simplifies many considerations . another relevant model in applications
is obtained by taking the space derivative of kpz and letting @xmath865 in order to obtain the stochastic conservation law @xmath866 which we will refer to as the stochastic burgers equation ( sbe ) . in all these cases , @xmath867 denotes a suitably renormalized product .
the kpz equation was derived by kardar
zhang in 1986 as a universal model for the random growth of an interface @xcite . for a long time
it could not be solved due to the fact that there was no way to make sense of the nonlinearity @xmath868 in ( [ eq : kpz - intro ] ) .
the only way to make sense of kpz was to apply the cole - hopf transform @xcite : solve she ( [ eq : heat - intro ] ) ( which is accessible to it integration ) and set @xmath869 .
but there was no intrinsic interpretation of what it means to solve ( [ eq : kpz - intro ] ) . finally in 2012 , hairer @xcite used rough paths to give a meaning to the equation and to obtain solutions directly at the kpz level . in section [ sec : rbe ]
we will sketch how to recover his solution in the paracontrolled setting .
applications of the techniques used by hairer to solve the kpz problem to a more general homogenization problem with ergodic potentials ( not necessarily gaussian ) have been studied in @xcite . in @xcite
, hairer introduces a theory of regularity structures which can also be considered a generalization of the theory of controlled rough paths to functions of a multidimensional index variable .
hairer fundamentally rethinks the notion of regularity .
usually a function is called smooth if it can be approximated around every point by a polynomial of a given degree ( the taylor polynomial ) .
naturally , the solution to an spde driven by say gaussian space - time white noise is not smooth in that sense .
so in hairer s theory , a function is called smooth if locally it can be approximated by the noise ( and higher order terms constructed from the noise ) .
this induces a natural topology in which the solutions to semilinear spdes depend continuously on the driving signal . at this date
it seems that the theory of regularity structures has a wider range of applicability than the paracontrolled approach described in @xcite , but also at the expense of a very deep conceptual sophistication .
there are problems ( like the one dimensional heat equation with multiplicative noise and general nonlinearity ) that can not be solved using paracontrolled distributions , but these problems seem also quite difficult ( even if doable and work in progress ) to tackle with regularity structures .
moreover , equations of a more general kind , say dispersive equations or wave equations , are still poorly ( or not at all ) understood in both approaches .
as we have tried to motivate in the previous sections we are looking for a theory for pam which describes the possible limits of the equation @xmath870 driven by sufficiently regular @xmath236 but as @xmath236 is converging to the space white noise @xmath38 . from this point of view we are looking for a priori estimates on the solution @xmath29 to ( [ eq : pam - eta ] ) which depend only on distributional norms of @xmath236 .
so in the following we will assume that we have at hand only a uniform control of @xmath236 in @xmath871 for some @xmath751 .
for the application to the 2d space white noise we could take @xmath872 , but we will not use this specific information in order to probe the range of applicability of our approach and we will only assume that the exponent @xmath873 is such that @xmath874 . assume for a moment that we are in the simpler situation @xmath756 and @xmath875 and let us try to solve equation ( [ eq : pam - eta ] ) via picard iterations @xmath876 starting from @xmath877 . since @xmath206 preserves the @xmath878-regularity ( which can be seen by identifying @xmath878 with the classical space of bounded hlder continuous functions of space ) , the product @xmath879 is well defined as an element of @xmath880 for all @xmath59 since @xmath881 and we are in condition to apply corollary [ cor : product ] below on the product of elements in hlder besov spaces . now by lemma [ lem : schauder ] , the laplacian gains two degrees of regularity so that the solution @xmath882 to @xmath883 , @xmath884 , is in @xmath878 .
from here we obtain a contraction on @xmath806 for some small @xmath409 whose value does not depend on @xmath470 , which gives us global in time existence and uniqueness of solutions . note that in one dimension the space white noise has regularity @xmath885 ( see exercise [ exo : wn besov ] ) so taking @xmath886 we have determined that the one dimensional pam can be solved globally in time with standard techniques .
when the condition @xmath881 is not satisfied we still have that if @xmath887 then @xmath888 by the standard parabolic estimates of lemma [ lem : schauder ] .
however with the regularities at hand we can not use corollary [ cor : product ] anymore to guarantee the continuity of the operator @xmath889 . moreover , as already seen in the simpler homogenization problems above this is not a technical difficulty but a real issue of the regime @xmath760 .
we expect that controlling the model in this regime can be quite tricky since limits exists when @xmath890 but the limiting solution still feels residual order one effects from the vanishing driving signal @xmath236 .
this situation can not be improved from the point of view of standard analytic considerations .
what is needed is a finer control of the solution @xmath29 which allows to analyse in more detail the possible resonances between the fluctuations of @xmath29 and those of @xmath236 . before going on
we will revise the problem of multiplication of distributions in the scale of hlder
besov spaces , introducing the basic tool of our general analysis : bony s paraproduct .
paraproducts are bilinear operations introduced by bony @xcite to linearize a class of nonlinear hyperbolic pdes in order to analyse the regularity of their solutions . in terms of littlewood
paley blocks , a general product @xmath891 of two distributions can be ( at least formally ) decomposed as @xmath892 here @xmath893 is the part of the double sum with @xmath894 , @xmath895 is the part with @xmath896 , and @xmath897 is the `` diagonal '' part , where @xmath898 .
more precisely , we define @xmath899 of course , the decomposition depends on the dyadic partition of unity used to define the blocks @xmath900 , and also on the particular choice of the pairs @xmath901 in the diagonal part .
the choice of taking all @xmath901 with @xmath902 into the diagonal part corresponds to the fact that the partition of unity can be chosen such that @xmath903 if @xmath894 , where @xmath646 is a suitable annulus .
if @xmath904 , the only apriori information on the spectral support of the various term in the double sum is @xmath905 , that is they are supported in balls and in particular they can have non zero contributions to very low wave vectors .
we call @xmath893 and @xmath895 _ paraproducts _ , and @xmath897 the _ resonant _ term .
bony s crucial observation is that @xmath893 ( and thus @xmath895 ) is always a well - defined distribution .
heuristically , @xmath893 behaves at large frequencies like @xmath516 ( and thus retains the same regularity ) , and @xmath492 provides only a frequency modulation of @xmath516 .
the only difficulty in constructing @xmath891 for arbitrary distributions lies in handling the diagonal term @xmath897 .
the basic result about these bilinear operations is given by the following estimates .
( paraproduct estimates)[thm : paraproduct ] for any @xmath906 and @xmath907 we have @xmath908 and for @xmath596 furthermore @xmath909 for @xmath910 we have @xmath911 there exists an annulus @xmath646 such that @xmath912 has fourier transform supported in @xmath650 , and for @xmath913 we have @xmath914 by lemma [ lem : besov regularity of series ] , we thus obtain ( [ eq : para-1 ] ) .
the proof of of ( [ eq : para-2 ] ) and ( [ eq : para-3 ] ) works in the same way , where for estimating @xmath897 we need @xmath910 because the terms of the series are supported in a ball and not in an annulus .
a simple corollary is then the following : [ cor : product]let @xmath628 and @xmath915 with @xmath910 , then the product @xmath916 is a bounded bilinear map from @xmath917 to @xmath918 .
while @xmath893 , @xmath895 , and @xmath897 depend on the specific dyadic partition of unity , the product @xmath919 does not .
the independence of the product from the dyadic partition of unity easily follows by taking smooth approximations .
the ill posedness of @xmath897 for @xmath920 can be interpreted as a resonance effect since @xmath897 contains exactly those part of the double series where @xmath492 and @xmath516 are in the same frequency range .
the paraproduct @xmath893 can be interpreted as frequency modulation of @xmath516 , which should become more clear in the following example . in figure
[ fig : slow ] we see a slowly oscillating positive function @xmath29 , while figure [ fig : fast ] depicts a fast sine curve @xmath537 .
the product @xmath921 , which here equals the paraproduct @xmath922 since @xmath29 has no rapidly oscillating components , is shown in figure [ fig : modulation ] .
we see that the local fluctuations of @xmath923 are due to @xmath537 , and that @xmath923 is essentially oscillating with the same speed as @xmath537 . + if @xmath924 and @xmath925 with @xmath926 , then we can define @xmath927 , which is well defined since @xmath928 and @xmath929 , and since integration is a linear map . in this way we recover the young integral @xcite .
let @xmath836 be a fractional brownian bridge on @xmath41 ( or simply a fractional brownian motion on @xmath930 $ ] , reflected on @xmath931 $ ] ) and assume that @xmath932 .
we have @xmath933 for all lipschitz continuous @xmath45 , and @xmath934 , and in particular @xmath935 is well - defined .
this can be used to solve sdes driven by @xmath836 in a pathwise sense .
the condition @xmath910 is essentially sharp , at least at this level of generality , see @xcite for counterexamples .
it excludes of course the brownian case : if @xmath390 is a brownian motion , then almost surely @xmath936 for all @xmath937 , so that @xmath938 and thus @xmath939 fails to be well defined . see also @xcite , proposition 1.29 for an instructive example which shows that this is not a shortcoming of our description of regularity , but that it is indeed impossible to define the product @xmath940 as a continuous bilinear operation on distribution spaces .
other counterexamples are given by our discussion of the homogenization problem above .
more simply , one can consider the following situation .
consider the sequence of functions @xmath941 given by@xmath942 .
then it is easy to show that @xmath943 for all @xmath944 .
however let @xmath945 .
then @xmath946 in @xmath801 which shows that the map @xmath947 can not be continuous in @xmath948 if @xmath944 .
pictorially the situation is summarized in figure [ fig : resonances ] , where we sketched the three dimensional curve given by @xmath949 for various values of @xmath271 and in the limit .
the product @xmath950 appearing in the right hand side of pam can be decomposed via the paraproduct @xmath951 as a sum of three terms @xmath952 the first and the last of these terms are continuous in any topology we choose for @xmath953 and @xmath236 .
the _ resonant term _
@xmath954 however is problematic .
it gathers the products of the oscillations of @xmath953 and @xmath236 on comparable dyadic scales and these products can contribute to all larger scales in such a way that microscopic oscillations might build up to a macroscopic effect which does not disappear in the limit ( as we already have seen ) .
if the function @xmath206 is smooth enough , then we expect the resonances between @xmath953 and @xmath236 to correspond to the resonances between @xmath29 and @xmath236 , and as we will see this is justified .
the expected regularity of the different terms is @xmath955 but unless @xmath881 the resonant term @xmath954 can not be controlled using only the @xmath878norm of @xmath29 and the @xmath956norm of @xmath236 .
if @xmath206 is at least @xmath957 , we can use a _ paralinearization _ result ( stated precisely in lemma [ lemma : para - taylor ] below ) to rewrite this term as @xmath958 with a remainder @xmath959 provided @xmath960 . the difficulty is now localized in the linearized resonant product @xmath961 . in order to control this term
, we would like to exploit the fact that the function @xmath29 is not a generic element of @xmath878 but that it has a specific structure , since @xmath962 has to match the paraproduct decomposition given in ( [ eq : para - rhs - intro ] ) where the least regular term is expected to be @xmath963 .
in order to do so , we postulate that the solution @xmath29 is given by the following _ paracontrolled ansatz : _
@xmath964 for functions @xmath965 such that @xmath966 and the remainder @xmath967 .
this decomposition allows for a finer analysis of the resonant term @xmath968 : indeed , we have @xmath969 where the _ commutator _ is defined by @xmath970 .
observe now that the term @xmath971 does not pose any further problem , as it is bounded in @xmath972 .
the key point is now that the commutator is a bounded multilinear function of its arguments as long as the sum of their regularities is strictly positive , see lemma [ lemma : commutator ] below . by assumption , we have @xmath973 , and therefore @xmath974 . the only problematic term which remains to be handled
is thus the bilinear functional of the noise given by @xmath975 .
here we need to make the assumption that @xmath976 in order for the product @xmath977 to be well defined .
this assumption is not guaranteed by the analytical estimates at hand , and it has to be added as a further requirement to our construction . granting this last step , we have obtained that the right hand side of equation ( [ eq : pam - eta ] ) is well defined and a continuous function of @xmath978 .
it remains to check that the paracontrolled ansatz is coherent with the equation satisfied by solutions to pam .
let us first consider the linear example @xmath834 .
here we saw that the solution is of the form @xmath979 with @xmath980 where @xmath981 and @xmath982 and therefore @xmath983 . in particular @xmath984 using a paralinearization result in last step ( see lemma [ lemma : paralinearization ] below ) .
now the double paraproduct @xmath985 satisfies @xmath986 see @xcite , and therefore @xmath987 which shows that the paracontrolled ansatz is at least justified in the linear case . in the nonlinear case , the paracontrolled ansatz and the leibniz rule for the paraproduct imply that ( [ eq : pam - eta ] ) can be rewritten as @xmath988 x + { \mathscr{l}}u^{\sharp } = f ( u ) { \,\mathord{\prec}\,}\eta + f ( u ) { \,\mathord{\circ}\,}\eta + f(u ) { \,\mathord{\succ}\,}\eta,\ ] ] where we recall that @xmath989 x = { \mathscr{l}}(u^x { \,\mathord{\prec}\,}x ) - u^x { \,\mathord{\prec}\,}{\mathscr{l}}x$ ] denotes the commutator .
if we choose @xmath69 such that @xmath990 and we set @xmath991 , then we can use ( [ eq : rde expand fu ] ) and ( [ eq : rde expand resonant term ] ) to obtain the following equation for the remainder @xmath992 : @xmath993 x\\ & \hspace{2em } + f ' ( u ) c ( f ( u ) , x , \eta ) + f ' ( u ) ( u^{\sharp } { \,\mathord{\circ}\,}\eta ) + \pi_f ( u , \eta ) .
\end{array } \label{eq : u - sharp}\ ] ] lemma [ lemma : comm - l - para ] below ensures that @xmath994 x \in c { \mathscr{c}}^{2 \gamma}$ ] whenever @xmath995 , and we have already seen that all the other terms on the right hand side are in @xmath996 so that this allows us to control @xmath992 in @xmath997 .
together with @xmath998 , equation ( [ eq : u - sharp ] ) gives an equivalent description of the solution and allows us to obtain a priori estimates on @xmath29 and @xmath992 in terms of @xmath999 .
it is now easy to show that if @xmath1000 , then @xmath29 depends continuously on the data @xmath1001 , so that we have a robust strategy to pass to the limit in ( [ eq : gpam ] ) and to make sense of the solution to ( [ eq : pam - eta ] ) also for irregular @xmath1002 as long as @xmath1003 . in the remainder of this section we will prove the results ( paralinearization and various key commutators ) which we used in the discussion above , before going on to gather the consequences of our analysis in the next section . when the time dependence does not play any role
we state the results for distributions depending only on the space variable as the extension to time varying functions will not add further difficulty .
[ lemma : commutator]assume that @xmath1004 are such that @xmath1005 and @xmath1006 .
then for @xmath1007 the trilinear operator @xmath1008 allows for the bound @xmath1009 and can thus be uniquely extended to a bounded trilinear operator from @xmath1010 to @xmath1011 . for @xmath1012
this follows from the paraproduct estimates , so let @xmath1013 . by definition @xmath1014 where we used that @xmath1015 has support in an annulus @xmath1016 , so that @xmath1017 only if @xmath1018 for some fixed @xmath16 , which in combination with @xmath1019 yields @xmath1020 .
now for fixed @xmath576 , the term @xmath1021 is spectrally supported in an annulus @xmath1022 , so that @xmath1023 and we may add and subtract @xmath1024 to @xmath1025 while maintaining the bound ( [ eq : commutator bound ] ) .
it remains to treat @xmath1026 we estimate both terms on the right hand side separately .
for @xmath1027 we have @xmath1028 using @xmath1013 .
it remains to estimate the second term in ( [ eq : commutator pr1 ] ) . for @xmath1029 and @xmath1030 , any term of the form @xmath1031
is spectrally supported in an annulus @xmath1032 , and therefore @xmath1033 for @xmath1034 we can apply the commutator estimate with @xmath1035 , as long as @xmath1036 .
our next result is a simple paralinearization lemma for non linear operators .
[ lemma : paralinearization]let @xmath577 , @xmath1037 $ ] , and let @xmath1038 .
there exists a locally bounded map @xmath1039 such that @xmath1040 for all @xmath628 .
more precisely , we have @xmath1041 if @xmath1042 , then @xmath1043 is locally lipschitz continuous : @xmath1044 the difference @xmath1045 is given by @xmath1046 = \sum_{i \geqslant - 1 } u_i,\ ] ] and every @xmath1047 is spectrally supported in a ball @xmath1048 . for @xmath1049 , we simply estimate @xmath1050 .
for @xmath1051 we use the fact that @xmath492 is a bounded function to write the littlewood
paley projections as convolutions and obtain @xmath1052 { \mathrm{d}}y { \mathrm{d}}z\\ & = \int k_i ( x - y ) k _ { < i - 1 } ( x - z ) [ f ( f ( y ) ) - f ( f ( z ) ) - f ' ( f ( z ) ) ( f ( y ) - f ( z ) ) ] { \mathrm{d}}y { \mathrm{d}}z , \end{aligned}\ ] ] where @xmath1053 , @xmath1054 , and where we used that @xmath1055 for @xmath1056 and @xmath1057 for @xmath1051 . now we can apply a first order taylor expansion to @xmath206 and use the @xmath1058hlder continuity of @xmath1059 in combination with the @xmath625hlder continuity of @xmath492 , to deduce @xmath1060 therefore , the estimate for @xmath1061 follows from lemma [ lem : besov regularity of series ] .
the estimate for @xmath1062 is shown in the same way .
let @xmath516 be a distribution belonging to @xmath1063 for some @xmath1064 .
then the map @xmath1065 behaves , modulo smoother correction terms , like a derivative operator : [ lemma : para - taylor]let @xmath577 , @xmath1066 $ ] , @xmath1067 be such that @xmath1005 and @xmath1068 .
let @xmath1038 .
then there exists a locally bounded map @xmath1069 such that @xmath1070 for all @xmath628 and all smooth @xmath516 .
more precisely , we have @xmath1071 if @xmath1072
, then @xmath1073 is locally lipschitz continuous : @xmath1074 use the paralinearization and commutator lemmas above to deduce that @xmath1075 so that the claimed bounds easily follow from lemma [ lemma : commutator ] and lemma [ lemma : paralinearization ] . besides this sort of chain rule
, we also have a leibniz rule for @xmath1076 : [ lem : para - taylor product]let @xmath577 and @xmath1077 be such that @xmath1078 and @xmath1068
. then there exists a bounded trilinear operator @xmath1079 , such that @xmath1080 for all @xmath1081 and all smooth @xmath516 .
it suffices to note that @xmath1082 , which leads to @xmath1083 [ lemma : comm - l - para]let @xmath1084 , @xmath100 , and let @xmath1085 and @xmath1086 with @xmath1087 .
there exists @xmath1088 such that @xmath1089 g$ ] and @xmath1090 .
moreover @xmath1091 and for all @xmath409 @xmath1092 let @xmath409 and let @xmath1093 be a time mollification of @xmath492 such that @xmath1094 and @xmath1095 for all @xmath31 . for example we can take @xmath1096 with @xmath1097 and @xmath1098 compactly supported , smooth , and of unit integral . for @xmath1099
we have @xmath1100 + \delta_i \left [ { \mathscr{l}}(f_{\varepsilon } { \,\mathord{\prec}\,}g ) - f_{\varepsilon } { \,\mathord{\prec}\,}{\mathscr{l}}g \right],\ ] ] so that @xmath1101 + \delta_i \left [ { \mathscr{l}}(f_{\varepsilon } { \,\mathord{\prec}\,}g ) - f_{\varepsilon } { \,\mathord{\prec}\,}{\mathscr{l}}g \right ] \\ & = \delta_i \left [ ( f_{\varepsilon } - f ) { \,\mathord{\prec}\,}{\mathscr{l}}g \right ] + \delta_i \left [ { \mathscr{l}}f_{\varepsilon } { \,\mathord{\prec}\,}g - 2 \partial_x f_{\varepsilon } { \,\mathord{\prec}\,}\partial_x g \right ] , \end{aligned}\ ] ] with initial condition @xmath1102
. the schauder estimates for @xmath678 ( lemma [ lem : schauder ] ) give @xmath1103 + \delta_i \left [ \left ( { \mathscr{l}}f_{\varepsilon } \right ) { \,\mathord{\prec}\,}g + \partial_x f_{\varepsilon } { \,\mathord{\prec}\,}\partial_x g \right ] \right\|_{c_t { \mathscr{c}}^{\alpha + \beta - 2 } } \\ & \hspace{60pt } \qquad + \| ( \delta_i ( f - f_{\varepsilon } ) { \,\mathord{\prec}\,}g ) ( 0 ) \|_{\alpha + \beta } .
\end{aligned}\ ] ] choosing @xmath1104 , we have @xmath1105 and exactly the same argument also gives @xmath1106 \right\|_{c_t { \mathscr{c}}^{\alpha + \beta - 2 } } \lesssim \| f \|_{{\mathscr{l}}_t^{\beta } } \left\| { \mathscr{l}}g \right\|_{c_t { \mathscr{c}}^{\alpha - 2 } } .\ ] ] since @xmath1084 , we further get @xmath1107 \right\|_{c_t { \mathscr{c}}^{\alpha + \beta - 2 } } & \lesssim 2^{i ( \beta - 2 ) } \| \partial_t f_{\varepsilon } \|_{c_t l^{\infty } } \| g \|_{c_t { \mathscr{c}}^{\alpha } } + \| f_{\varepsilon } \|_{c_t { \mathscr{c}}^{\beta } } \| g \|_{c_t { \mathscr{c}}^{\alpha } } \\ & \lesssim \| f \|_{{\mathscr{l}}_t^{\beta } } \| g \|_{c_t { \mathscr{c}}^{\alpha } } + \| f
\|_{c_t { \mathscr{c}}^{\beta } } \| g \|_{c_t { \mathscr{c}}^{\alpha}}. \end{aligned}\ ] ] combining everything , we end up with @xmath1108 which gives the estimate for the space regularity of @xmath321 since @xmath1109 .
the time regularity of @xmath321 can be controlled similarly by noting that @xmath1110 , uniformly in @xmath1111 . here
we build a calculus of distributions satisfying a paracontrolled ansatz .
we start by defining a suitable space of such objects .
[ def : paracontrolled parabolic]let @xmath22 and @xmath1037 $ ] be such that @xmath1112 , and let @xmath1113 . a pair of distributions @xmath1114 is called _ paracontrolled _ by @xmath516 if @xmath1115 . in that case
we write @xmath1116 , and for all @xmath409 we define the norm @xmath1117 if @xmath1118 and @xmath1119 , then we also write @xmath1120 of course we should really write @xmath1121 since given @xmath492 and @xmath516 , the derivative @xmath1122 is usually not uniquely determined .
but in the applications there will always be an obvious candidate for the derivative , and no confusion will arise . the space @xmath1123 does not depend on the specific dyadic partition of unity . indeed , bony @xcite has shown that if @xmath1124 is the paraproduct constructed from another partition of unity , then @xmath1125 .
[ [ nonlinear - operations ] ] nonlinear operations + + + + + + + + + + + + + + + + + + + + as an immediate consequence of lemma [ lemma : commutator ] we can multiply paracontrolled distributions provided that we know how to multiply the reference distributions .
[ thm : paracontrolled product]let @xmath1126 , @xmath1077 , with @xmath1005 and @xmath1127 .
let @xmath1128 , @xmath1129 , and let @xmath1130 .
then @xmath1131 defines a linear operator which for all @xmath409 admits the bound @xmath1132 if there exist sequences of smooth functions @xmath1133 and @xmath1134 converging to @xmath29 and @xmath537 respectively for which @xmath1135 converges to @xmath236 , then @xmath1136 does not depend on the dyadic partition of unity used to construct it .
furthermore , there exists a quadratic polynomial @xmath506 so that if @xmath1137 satisfy the same assumptions as @xmath29 , @xmath537 , @xmath236 respectively , if @xmath1138 , and if @xmath1139 then @xmath1140 given lemma [ lemma : commutator ] ( and the paraproduct estimates theorem [ thm : paraproduct ] ) , the proof is straightforward and we leave it as an exercise . from now on we will usually write @xmath1141 rather than @xmath1136 . to solve equations involving general nonlinear functions ,
we need to examine the stability of paracontrolled distributions under smooth functions .
[ thm : paralinearization]let @xmath577 and @xmath1037 $ ] .
let @xmath1142 , @xmath1143 , and @xmath1144 .
then @xmath1145 with derivative @xmath1146 , and for all @xmath409 @xmath1147 moreover , there exists a polynomial @xmath506 which satisfies for all @xmath1148 , @xmath1149 , @xmath1150 , and @xmath1151 the bound @xmath1152 the proof is not very complicated but rather lengthy , and we do not present it here . [ [ schauder - estimate - for - paracontrolled - distributions ] ] schauder estimate for paracontrolled distributions + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the schauder estimate lemma [ lem : schauder ] is not quite sufficient : we also need to understand how the heat kernel acts on the paracontrolled structure .
[ thm : schauder paracontrolled]let @xmath577 and @xmath1153 $ ] .
let @xmath1154 and @xmath1155 with @xmath1156 .
let @xmath1157 , @xmath1158 , and @xmath1159 .
then @xmath1160 , where @xmath516 solves @xmath1161 and we have the bound @xmath1162 for all @xmath409 .
if furthermore @xmath1163 satisfy the same assumptions as @xmath1164 respectively , and if @xmath1165 , then @xmath1166 let us derive an equation for the remainder @xmath1167 .
we have @xmath1168 + f^u { \,\mathord{\prec}\,}{\mathscr{l}}u - [ f^u { \,\mathord{\prec}\,}u + f^{\sharp } ] \\ & = \left [ { \mathscr{l } } , f^u { \,\mathord{\prec}\,}\right ] u - f^{\sharp } .
\end{aligned}\ ] ] by lemma [ lemma : comm - l - para ] we know that there exists @xmath1169 such that @xmath1170 u$ ] ,
so we can apply the standard schauder estimates of lemma [ lem : schauder ] to @xmath1171 to get @xmath1172 the estimate for @xmath1173 can be derived in the same way .
* bibliographic notes * paraproducts were introduced in @xcite , for a nice introduction see @xcite .
the commutator estimate lemma [ lemma : commutator ] is from @xcite , but the proof here is new and the statement is slightly different . in @xcite , we require the additional assumption @xmath577 under which @xmath6 maps @xmath1174 to @xmath1175 and not only to @xmath1176 . theorem [ thm : paralinearization ] is from @xcite .
theorem [ thm : schauder paracontrolled ] is new , but it is implicitly used in @xcite .
the estimates presented here will only allow us to consider regular initial conditions .
more general situations can be covered by working on `` explosive spaces '' of the type @xmath1177 } \| t^{- \gamma } f ( t ) \|_{{\mathscr{c}}^{\alpha } } < \infty { \ensuremath{\operatorname{for } } } { \ensuremath{\operatorname{all } } } t > 0 \big\}\ ] ] and similar for the temporal regularity .
this is also done in @xcite .
of course it is easily possible to replace the laplacian by more general pseudo - differential operators .
we only used two properties of @xmath261 : the fact that @xmath1178 is smooth , and that the semigroup generated by @xmath261 has a sufficiently strong regularization effect .
this is also true for fractional laplacians and more generally for a wide range of pseudo - differential operators .
let us now give the details for the solution to pam in the space of paracontrolled distributions .
assume that @xmath1179 is in @xmath1180 for some @xmath31 such that @xmath1181 .
we know from theorem [ thm : paralinearization ] that if @xmath1182 and @xmath1183 , then @xmath1184 : @xmath1185 if now @xmath1186 is given , then theorem [ thm : paracontrolled product ] shows that for all @xmath1187 we have @xmath1188 with @xmath1189 it is here that we use @xmath1190 . integrating against the heat kernel and assuming that @xmath1191
, we obtain from theorem [ thm : schauder paracontrolled ] that the solution @xmath1192 to @xmath1193 , @xmath1194 , is in @xmath1195 , where @xmath69 solves @xmath1196 and @xmath1197 . in other words
, we have a map @xmath1198 and combining ( [ eq : fixpoint map1 ] ) , ( [ eq : fixpoint map2 ] ) we get @xmath1199 so that for all @xmath409 we can define @xmath1200 } .\ ] ] to set up a picard iteration domain and image space should coincide which means we should take @xmath1201 .
refining the analysis , we obtain a scaling factor @xmath1202 when estimating the @xmath1203norm of @xmath1204 .
this allows us to show that for small @xmath409 , the map @xmath1205 leaves suitable balls in @xmath1206 invariant , and therefore we obtain the ( local in time ) _ existence _ of solutions to the equation under the assumption @xmath976 . to obtain _ uniqueness
_ we need to suppose that @xmath1207 . in that case theorem [ thm : paralinearization ]
gives the local lipschitz continuity of the map @xmath1208 from @xmath1209 to @xmath1210 , while theorem [ thm : paracontrolled product ] and theorem [ thm : schauder paracontrolled ] show that @xmath1211 defines a lipschitz continuous map from @xmath1210 to @xmath1206 .
again we can obtain a scaling factor @xmath1202 , so that @xmath1205 defines a contraction on a suitable ball of @xmath1206 for some small @xmath1212 .
even better , @xmath1205 not only depends locally lipschitz continuously on @xmath29 , but also on the extended data @xmath1213 , and therefore the solution to ( [ eq : pam - eta ] ) depends locally lipschitz continuously on @xmath1214 .
so far we argued under the assumption that @xmath975 exists and has a sufficient regularity .
this should be understood via approximations as the existence of a sequence of smooth functions @xmath1215 that converges to @xmath236 , such that @xmath1216 converges to @xmath975 . however
, this hypothesis is questionable and , recalling our homogenization setting , actually not satisfied at all in the problem we are interested in .
more concretely this can be checked by considering an approximation sequence , for example if @xmath1217 is the two dimensional space white noise . indeed , if @xmath45 is a schwartz function on @xmath1218 and if @xmath1219 and @xmath1220 then we will see below that there exist constants @xmath1221 with @xmath1222 , such that @xmath1223 converges in @xmath1224 for all @xmath409 .
this is not a problem with our specific approximation .
the homogenization setting shows that even for @xmath1225 there are cases where the limiting equation is nontrivial . in the paracontrolled
setting we have continuous dependence on the data @xmath1226 , so this non triviality of the limit can only mean that it is @xmath975 which does not converge to zero .
another way to see that there is a problem is to consider the following representation of the resonant term : use @xmath1196 to write @xmath1227 integrating this equation over the torus and over @xmath411 $ ] , we get @xmath1228 but now if @xmath1229 and @xmath1230 then all the terms should be well defined and finite but this is can not be since would mean that @xmath1231 for any @xmath782 almost surely if @xmath236 is the space white noise .
note also that the problematic term @xmath1232 is exactly the correction term appearing in the analysis of the linear homogenization problem above . in order to obtain convergence of the smooth solutions in general
, we should introduce corrections to the equation to remove the divergent constant @xmath1233 .
let us see where the resonant product @xmath975 appears .
we have @xmath1234 now @xmath1235 by theorem [ thm : paralinearization ] , and if @xmath29 solves the equation , then @xmath991 by theorem [ thm : schauder paracontrolled ] .
so we should really consider the renormalized equation @xmath1236 and in the limit @xmath1237 where the paracontrolled product @xmath1238 is calculated using @xmath1239 in the place of @xmath975 .
formally , we also denote this product by @xmath1240 , so that the solution @xmath29 will satisfy @xmath1241 note that the correction term has exactly the same form as the it / stratonovich corrector for sdes and indeed we should consider ( [ eq : pam - renormalized ] ) as the `` it form '' of the equation .
the convergence properties of @xmath1216 are in stark contrast to the ode setting : if we consider the equation @xmath1242 rather than pam , then we should replace @xmath69 by @xmath307 with @xmath1243 .
but then we have in one dimension @xmath1244 , so that the convergence of @xmath1245 to @xmath1246 comes for free with the convergence of @xmath1247 to @xmath307 .
this is due to the leibniz rule for @xmath694 and it is the reason why rough path theory is trivial in one dimension .
as we have discussed , for the second order differential operator @xmath678 we have different rules and @xmath1248 so that in our setting the nontrivial term is @xmath1249 .
these considerations lead naturally to the following definition .
[ def : pam rough distribution](pam enhancement ) let @xmath1250 and let @xmath1251 be the closure of the image of the map @xmath1252 given by @xmath1253 where @xmath1254 , that is @xmath1255 and @xmath1256 .
we will call @xmath1257 the renormalized pam enhancement of the driving distribution @xmath1258 . for @xmath1259
we define @xmath1260}$ ] and we write @xmath1261 for the norm of @xmath1262 in the banach space @xmath1263 .
moreover , we define the distance @xmath1264
. it would be more elegant to renormalize @xmath1265 with a constant and not with a time - dependent function .
but since we chose @xmath1256 , we have @xmath1266 and therefore @xmath1267 diverges for any diverging sequence of constants @xmath1221 . a simple way of avoiding this problem is to consider the stationary version @xmath1268 given by @xmath1269 where @xmath1270 denotes the projection on the non - zero fourier modes , @xmath1271 .
but then @xmath1268 does not depend on time and in particular @xmath1272 , so that we have to consider irregular initial conditions in the paracontrolled approach which complicates the presentation .
alternatively , we could observe that in the white noise case there exist constants @xmath1221 so that @xmath1273 converges for all @xmath782 , and while the limit @xmath1274 diverges as @xmath1275 , it can be integrated against the heat kernel .
again , this would complicate the presentation and here we choose the simple ( and cheap ) solution of taking a time - dependent renormalization .
[ thm : pam]let @xmath1276 and @xmath31 be such that @xmath1181 .
let @xmath1277 , @xmath1278 , and @xmath1279 .
then there exists a unique solution @xmath1280 to the equation @xmath1281 up to the ( possibly finite ) explosion time @xmath1282 .
moreover , @xmath29 depends on @xmath1283 in a locally lipschitz continuous way : if @xmath1284 are such that for all @xmath1285 with @xmath1286 , the solution @xmath29 to the equation driven by @xmath1285 satisfies @xmath1287 , and if @xmath1288 is another set of data bounded in the above sense by @xmath56 , then there exists @xmath1289 for which @xmath1290 we only have to turn the formal discussion of section [ sec : fixpoint ] into rigorous mathematics .
the small factor @xmath1202 is obtained from a scaling argument and while this does not require any new insights it is somewhat lengthy and we refer to @xcite for details .
let us just indicate how to iterate the construction to obtain the existence of solutions up to the explosion time @xmath1291 .
let us assume that we constructed @xmath29 on @xmath1292 $ ] for some @xmath1293 .
now we no longer have @xmath1294 , and also the initial condition @xmath1295 is no longer in @xmath1296 .
but we only used @xmath1197 to derive the initial conditions for @xmath1297 and @xmath992 , and we only used @xmath875 to obtain @xmath1298 initial conditions for @xmath992 .
since now we already know @xmath1299 , we do not need this anymore .
as for the continuity in @xmath1285 , let @xmath1300 be another set of data also bounded by @xmath56 .
then the solutions @xmath29 and @xmath814 both are bounded in @xmath1301 by some constant @xmath1302 .
so by the continuity properties of the paracontrolled product ( and the other operations involved ) , we can estimate @xmath1303 for a polynomial @xmath506 .
the local lipschitz continuity on @xmath664 $ ] immediately follows if we choose @xmath409 small enough .
this can be iterated to obtain the local lipschitz continuity on `` macroscopic '' intervals . for the local in time existence
it is not necessary to assume @xmath1304 , it suffices if @xmath1305 .
this can be seen by considering a ball containing @xmath1306 for all @xmath1307 , a function @xmath1308 which coincides with @xmath206 on this ball , and by stopping @xmath29 upon exiting the ball .
in the linear case @xmath834 we have global in time solutions : in general we only get local in time solutions because we pick up a superlinear estimate when applying the paralinearization result theorem [ thm : paralinearization ] .
this step is not necessary if @xmath206 is linear , and all the other estimates are linear in @xmath29 . in order to apply theorem [ thm : pam ] to equation ( [ eq : pam - eta ] ) with white noise perturbation , it remains to show that if @xmath38 is a spatial white noise on @xmath1309 , then @xmath38 defines an element of @xmath1310 .
in other words , we need to construct @xmath1311 .
since @xmath1312 is a smooth function for every @xmath782 , the resonant term @xmath1313 is a smooth function , and therefore we could formally set @xmath1314 .
but we will see that this expression does not make sense .
recall that @xmath1315 is a complex valued , centered gaussian process with covariance @xmath1316 = \delta_{k + k ' = 0},\ ] ] and such that @xmath1317 .
[ l : anderson area expectation ] for any @xmath1318 and @xmath782 we have @xmath1319 = \mathbb{e } [ \delta_{- 1 } ( p_t \xi { \,\mathord{\circ}\,}\xi ) ( x ) ] = ( 2 \pi)^{- 2 } \sum_{k \in \mathbb{z}^2 } e^{- t |k|^2 } .\ ] ] in particular , @xmath1320 does not depend on the partition of unity used to define the @xmath1321 operator , and @xmath1322 for all @xmath1323 .
let @xmath1318 , @xmath782 , and @xmath1324 .
then @xmath1325 = \sum_{|i - j| \leqslant 1 } \mathbb{e } [ \delta_{\ell } ( \delta_i ( p_t \xi ) \delta_j \xi ) ( x)],\ ] ] where exchanging summation and expectation is justified because it can be easily verified that the partial sums of @xmath1326 are uniformly @xmath185bounded for any @xmath1327 .
now @xmath1328 , and therefore we get from @xmath1329 & = ( 2 \pi)^{- 1 } \sum_{k , k ' \in \mathbb{z}^2 } e_{k + k'}^{\ast } ( x ) \rho_{\ell } ( k + k ' ) \rho_i ( k ) e^{- t |k|^2 } \rho_j ( k ' ) \mathbb{e } [ \hat{\xi } ( k ) \hat{\xi } ( k ' ) ] \\ & = ( 2 \pi)^{- 2 } \sum_{k \in \mathbb{z}^2 } \rho_{\ell } ( 0 ) \rho_i ( k ) e^{- t |k|^2 } \rho_j ( k ) = \frac{\delta_{\ell = - 1}}{2 \pi } \sum_{k \in \mathbb{z}^2 } \rho_i ( k ) \rho_j ( k ) e^{- t |k|^2 } . & \end{aligned}\ ] ] for @xmath1330 we have @xmath1331 and therefore @xmath1319 = ( 2 \pi)^{- 2 } \sum_{k \in \mathbb{z}^2 } \sum_{i , j } \rho_i ( k ) \rho_j ( k ) e^{- t |k|^2 } = ( 2 \pi)^{- 2 } \sum_{k \in \mathbb{z}^2 } e^{- t |k|^2},\ ] ] while @xmath1332 = 0 $ ] .
let @xmath45 be a schwartz function on @xmath1218 and set @xmath1333 for @xmath1318 .
write @xmath1334 .
show that @xmath1335 = \mathbb{e } [ \delta_{- 1 } ( p_t \xi_n { \,\mathord{\circ}\,}\xi_n ) ( x ) ] = ( 2 \pi)^{- 2 } \sum_{k \in \mathbb{z}^2 } e^{- t |k|^2 } | { \mathscr{f}}_{\mathbb{r}^2 } \varphi ( k / n ) |^2 .\ ] ] * hint : * use poisson summation .
the diverging time integral motivates us to study the renormalized product @xmath1336 , where @xmath1337 is an `` infinite function '' : [ lem : pam area]set @xmath1338 then @xmath1339 <
\infty$ ] for all @xmath1340 , @xmath1327 , @xmath409 . moreover , if @xmath45 is a schwartz function on @xmath1218 with @xmath1341 , if @xmath1342 with @xmath1343 for @xmath93 , and @xmath1344 , then @xmath1345 = 0\ ] ] for all @xmath1327 , where for all @xmath1318 @xmath1346 = \mathbb{e } [ ( x_n ( t ) { \,\mathord{\circ}\,}\xi_n ) ( x ) ] \\ & = ( 2 \pi)^{- 2 } \sum_{k \in \mathbb{z}^2 \setminus \ { 0 \ } } \frac{| { \mathscr{f}}_{\mathbb{r}^2 } \varphi ( k / n ) |^2}{|k|^2 } ( 1 - e^{- t | k |^2 } ) + ( 2 \pi)^{- 2 } t. \end{aligned}\ ] ] to lighten the notation , we will only show that
@xmath1347 < \infty$ ] . the convergence of @xmath1348 to @xmath1311 is shown by applying dominated convergence , and we leave it as an exercise .
let @xmath782 and define @xmath1349 .
using gaussian hypercontractivity we will be able to reduce everything to estimating @xmath1350 $ ] for @xmath1351 and @xmath1318 .
lemma [ l : anderson area expectation ] yields @xmath1352 $ ] for @xmath1353 and @xmath1318 , and @xmath1354 $ ] , so that @xmath1350 = { \ensuremath{\operatorname{var } } } ( \delta_{\ell } ( p_t \xi { \,\mathord{\circ}\,}\xi ) ( x))$ ] .
we have @xmath1355 hence @xmath1356 where exchanging summation and expectation can be justified a posteriori by the uniform @xmath185boundedness of the partial sums
. now wick s theorem ( @xcite , theorem 1.28 ) gives @xmath1357 -\mathbb{e } [ \hat{\xi } ( k_1 ) \hat{\xi } ( k_2 ) ] \mathbb{e } [ \hat{\xi } ( k'_1 ) \hat{\xi } ( k'_2 ) ] \\ & = \mathbb{e } [ \hat{\xi } ( k_1 ) \hat{\xi } ( k_2 ) ] \mathbb{e } [ \hat{\xi}(k'_1 ) \hat{\xi } ( k'_2 ) ] + \mathbb{e } [ \hat{\xi } ( k_1 ) \hat{\xi } ( k_1 ' ) ] \mathbb{e } [ \hat{\xi } ( k_2 ) \hat{\xi } ( k_2 ' ) ] \\ & \hspace{10pt } + \mathbb{e } [ \hat{\xi } ( k_1 ) \hat{\xi } ( k_2 ' ) ] \mathbb{e } [ \hat{\xi}(k_2 ) \hat{\xi } ( k'_1 ) ] -\mathbb{e } [ \hat{\xi } ( k_1 ) \hat{\xi } ( k_2 ) ] \mathbb{e } [ \hat{\xi } ( k'_1 ) \hat{\xi } ( k'_2 ) ] \\ & = ( \delta_{k_1 + k_1 ' = 0 } \delta_{k_2 + k'_2 = 0 } + \delta_{k_1 + k'_2 = 0 } \delta_{k_2 + k_1 ' = 0 } ) , \end{aligned}\ ] ] and therefore @xmath1358 .
\end{aligned}\ ] ] now there exists @xmath1359 such that @xmath1360 for all @xmath1361 and for all @xmath1362 . in the remainder of the proof the value of this strictly positive @xmath1363 may change from line to line . if @xmath898 , then we also have @xmath1364 for all @xmath1365 .
thus @xmath1366 where we used that @xmath1367 for all @xmath1368 .
consider now @xmath1369 .
we have for all @xmath1370 @xmath1371 = \sum_{\ell } 2^{2 p \ell ( 2 \gamma - 2 ) } \int_{\mathbb{t}^2 } \mathbb{e } [ | \delta_{\ell } ( x { \,\mathord{\diamond}\,}\xi ( t ) - x { \,\mathord{\diamond}\,}\xi ( s ) ) ( x ) |^{2 p } ] { \mathrm{d}}x.\ ] ] since the random variable @xmath1372 lives in the second non - homogeneous chaos generated by the gaussian white noise @xmath38 , we may use gaussian hypercontractivity ( @xcite , theorem 3.50 ) to bound @xmath1373 & \lesssim \mathbb{e } [ | \delta_{\ell } ( x { \,\mathord{\diamond}\,}\xi ( t ) - x { \,\mathord{\diamond}\,}\xi ( s ) ) ( x ) |]^{2 p } \\ & \leqslant \big ( \int_s^t \mathbb{e } [ | \delta_{\ell } \xi_r ( x ) | ] { \mathrm{d}}r \big)^{2 p}. \end{aligned}\ ] ] but we just showed that @xmath1374 & \leqslant \mathbb{e } [ | \delta_{\ell } \xi_r ( x ) |^2]^{1 / 2 } = ( { \ensuremath{\operatorname{var } } } ( \delta_{\ell } ( p_r \xi { \,\mathord{\circ}\,}\xi ) ( x)))^{1 / 2 } \\ & \lesssim r^{- 1 / 2 } 2^{\ell } e^{- \frac{1}{2 } r c 2^{2 \ell } } \leqslant r^{- 1 / 2 } 2^{\ell } e^{- r c 2^{2 \ell } } \end{aligned}\ ] ] ( changing again the value of @xmath1363 ) , and therefore @xmath1375\big)^{1 / 2 p } & \lesssim \big ( \sum_{\ell } \big ( 2^{\ell ( 2 \gamma - 2 ) } \int_s^t r^{- 1 / 2 } 2^{\ell } e^{- r c 2^{2 \ell } } { \mathrm{d}}r \big)^{2 p } \big)^{1 / 2 p } \\ & \leqslant \sum_{\ell } 2^{\ell ( 2 \gamma - 1 ) } \int_s^t r^{- 1 / 2 } e^{- r c 2^{2 \ell } } { \mathrm{d}}r \\ & \lesssim \big ( \int_s^t r^{- 1 / 2 } \int_{- 1}^{\infty } ( 2^x)^{2 \gamma - 1 } e^{- r c 2^{2 x } } { \mathrm{d}}x { \mathrm{d}}r \big)^{2 p } . \end{aligned}\ ] ] the change of variable @xmath1376 leads to @xmath1377 for @xmath1378 , the integral in @xmath1379 is finite and we end up with @xmath1380 provided that @xmath1381 .
so for large enough @xmath184 we can use kolmogorov s continuity criterion to deduce that @xmath1382 < \infty$ ] for all @xmath409 .
the claim now follows from besov embedding , lemma [ lem : besov embedding ] . combining theorem [ thm : pam ] and lemma [ lem : pam area ] , we are finally able to solve ( [ eq : pam - eta ] ) driven by a space white noise .
let @xmath31 and let @xmath1278 and assume that @xmath470 is a random variable that almost surely takes its values in @xmath1296 for some @xmath1276 with @xmath1383 .
let @xmath38 be a spatial white noise on @xmath1309 .
then there exists a unique solution @xmath29 to @xmath1384 up to the ( possibly finite ) explosion time @xmath1385 which is almost surely strictly positive .
if @xmath1386 and @xmath1387 are as described in lemma [ lem : pam area ] , and if @xmath1388 converges in probability in @xmath1298 to @xmath470 , then @xmath29 is the limit in probability of the solutions @xmath1389 to @xmath1390 we even have a stronger result : we can fix a null set outside of which @xmath1391 is regular enough , and once we dispose of that null set we can solve all equations for any regular enough @xmath470 and @xmath206 simultaneously , without ever having to worry about null sets again .
this is for example interesting when studying stochastic flows or when studying equations with random @xmath470 and @xmath206 .
the pathwise continuous dependence on the signal is also powerful in several other applications , for example support theorems and large deviations . for examples in the theory of rough paths
see @xcite .
let us now get to our main example , the `` kpz family '' of equations .
we concentrate here on the stochastics burgers equation sbe , but essentially the same analysis works for the kpz equation . we can also treat the heat equation with the same arguments , although in that case we need to set up the equation in the right way by applying a suitable transformation .
recall that the stochastic burgers equation sbe is @xmath1392 where @xmath1393 , @xmath38 is a space - time white noise , and @xmath1394 denotes the spatial derivative . as we argued before , the solution @xmath29 can not be expected to behave better than the ornstein uhlenbeck process @xmath69 , the solution of the linear equation @xmath1395 , and as we saw in section [ sec : energy ] @xmath1396 is for all @xmath782 a smooth function of the space variable plus a space white noise . by exercise [ exo : wn besov ]
, the white noise in dimension @xmath852 has regularity @xmath885 .
thus @xmath1397 , and in particular @xmath1398 is the square of a distribution and a priori not well defined .
what raises some hope is that in lemma [ lem : ou square ] we were able to show that @xmath1399 exists as a space time distribution .
so as in the previous examples there are stochastic cancellations going into @xmath1400 .
the energy solution approach was designed to take those cancellations into account in the full solution @xmath29 , but while it allowed us to work under rather weak assumptions which easily gave us existence of solutions , it did not give us sufficient control to have uniqueness of solutions . on the other side , a suitable paracontrolled ansatz for
the solution @xmath29 will allow us to transfer the cancellation properties of @xmath69 to @xmath29 and it will allow us to construct @xmath1401 as a _ continuous _ bilinear map , from where existence and uniqueness of solutions easily follows .
in this discussion we consider the case of zero initial condition and smooth noise @xmath38 , and we analyze the structure of the solution .
let us expand @xmath29 around the ornstein uhlenbeck process @xmath69 with @xmath1402 , @xmath1403 .
setting @xmath1404 , we have @xmath1405 let us define the bilinear map @xmath1406 then we can proceed by performing a further change of variables in order to remove the term @xmath1407 from the equation by setting @xmath1408 now @xmath1409 satisfies @xmath1410 we can imagine to make a similar change of variables to get rid of the term @xmath1411 as we proceed in this inductive expansion , we generate a number of explicit terms , obtained by various combinations of @xmath69 and @xmath390 . since we will have to deal explicitly with at least some of these terms , it is convenient to represent them with a compact notation involving binary trees .
a binary tree @xmath1412 is either the root @xmath1413 or the combination of two smaller binary trees @xmath1414 , where the two edges of the root of @xmath1291 are attached to @xmath1415 and @xmath1416 respectively .
for example @xmath1417 then we define recursively @xmath1418 giving @xmath1419 and so on . in this notation
the expansion ( [ eq : naive - exp])([eq : naive - exp-2 ] ) reads @xmath1420 @xmath1421 we observe that formally the solution @xmath29 of sbe can be expanded as an infinite sum of terms labelled by binary trees : @xmath1422 where @xmath1423 is a combinatorial factor counting the number of planar trees which are isomorphic ( as graphs ) to @xmath1291 . for example @xmath1424 ,
@xmath1425 , @xmath1426 , @xmath1427 , @xmath1428 and in general @xmath1429 .
alternatively , we may truncate the summation at trees of degree at most @xmath271 and set @xmath1430 where we denote by @xmath1431 the degree of the tree @xmath1291 , given by @xmath1432 and then inductively @xmath1433 .
for example @xmath1434 , @xmath1435 , @xmath1436 , @xmath1437 .
we then obtain for the remainder @xmath1438 our aim is to control the truncated expansion under the natural regularity assumptions in the white noise case , @xmath1397 . since ( [ eq : truncated expansion ] ) contains the term @xmath1439 which in turn contains the paraproduct @xmath1440 , the remainder @xmath1441 will be at best in @xmath1442 .
but then the sum of the regularities of @xmath69 and @xmath1441 is negative , and the term @xmath1443 is not well defined .
we therefore continue the expansion up to the point ( turning out to be @xmath1444 ) where we can set up a paracontrolled ansatz for the remainder , which will allow us to make sense of @xmath1445 and thus of @xmath1443 .
inspired by the partial tree series expansion of @xmath29 , we set up a paracontrolled ansatz of the form @xmath1446 where the functions @xmath1447 and @xmath992 are for the moment arbitrary , but we assume @xmath1448 and @xmath1449 , where from now on we fix @xmath1450 . for such @xmath29 ,
the nonlinear term takes the form @xmath1451 which gives us an equation for @xmath1452 : @xmath1453 if we formally apply the paraproduct estimate theorem [ thm : paraproduct ] ( which is of course not possible since the regularity requirements for the resonant term are not satisfied ) , we derive the following natural regularities for the driving terms : @xmath1397 , @xmath1454 , @xmath1455 , @xmath1456 , and @xmath1457 . in terms of @xmath1458 , we can encode this as @xmath1459 under these regularity assumptions the term @xmath1460 is well defined and the only problematic term in ( [ eq : uq - equation ] ) is @xmath1461 .
using the paracontrolled structure of @xmath1452 , we can make sense of @xmath1461 as a bounded operator provided that @xmath1462 is given .
in other words , the right hand side of ( [ eq : uq - equation ] ) is well defined for paracontrolled distributions .
next , we should specify how to choose @xmath1463 and which form @xmath1297 will take for the solution @xmath1452 .
we have formally @xmath1464 where we assumed that not only @xmath1465 , but that @xmath1466 ( which implies @xmath1465 , but also the stronger statement @xmath1467 ) . by theorem [ thm : schauder paracontrolled ]
, @xmath1452 is paracontrolled by @xmath1468 , and in other words we should set @xmath1469 .
the derivative @xmath1297 of the solution @xmath1452 will then be given by @xmath1470 .
unlike for pam , here we do not need to introduce a renormalization .
this is due to the fact that we differentiate after taking the square : to construct @xmath1398 , we would have to subtract an infinite constant and formally consider @xmath1471 , or at the level of the approximation @xmath1472 .
but then @xmath1473 so we obtain the following description of the driving data for the stochastic burgers equation .
[ def : burgers rough distribution](sbe enhancement ) let @xmath1450 and let @xmath1474 be the closure of the image of the map @xmath1475 given by @xmath1476 where @xmath1477 we will call @xmath1478 the sbe enhancement of the driving distribution @xmath1258 .
for @xmath409 we define @xmath1479}$ ] and we write @xmath1480 for the norm of @xmath659 in the banach space @xmath1481 .
moreover , we define the distance @xmath1482 .
let @xmath1483 .
then the space of paracontrolled distributions @xmath1484 is defined as the set of all @xmath1485 with @xmath1486 where @xmath1449 .
for @xmath409 we define @xmath1487 if @xmath1488 and @xmath1489 , then we also write @xmath1490 [ thm : rbe]let @xmath1450 .
let @xmath1492 , write @xmath1493 , and let @xmath1494 .
then there exists a unique solution @xmath1495 to the equation @xmath1496 up to the ( possibly finite ) explosion time @xmath1282 .
moreover , @xmath29 depends on @xmath1497 in a locally lipschitz continuous way : if @xmath1284 are such that for all @xmath1285 with @xmath1498 , the solution @xmath29 to the equation driven by @xmath1285 satisfies @xmath1287 , and if @xmath1288 is another set of data bounded in the above sense by @xmath56 , then there exists @xmath1499 for which @xmath1500 by definition of the term @xmath1401 , the distribution @xmath1501 solves ( [ eq : rbe theorem ] ) if and only if @xmath1502 solves @xmath1503 with initial condition @xmath1504 .
this equation is structurally very similar to pam ( [ eq : pam - eta ] ) and can be solved using the same arguments , which we do not reproduce here . for this result to be of any use
we still have to show that if @xmath38 is the space - time white noise , then there is almost surely an element of @xmath1505 associated to @xmath1506 .
while for pam we needed to construct only one term , here we have to construct five terms : @xmath1507 . for details
we refer to @xcite .
alternatively we can simply differentiate the extended data which hairer constructed for the kpz equation in chapter 5 of @xcite .
the same approach allows us to solve the kpz equation @xmath1508 , and if we are careful how to interpret the product @xmath1509 , then also the linear heat equation @xmath1510 . in both cases
the solution will depend continuously on some suitably extended data that is constructed from @xmath38 in a similar way as described in definition [ def : burgers rough distribution ] .
moreover , the formal links between the three equations that we discussed in section [ sec : rbe , kpz , she ] can be made rigorous .
these results will be included in @xcite .
martin hairer , etienne pardoux and andrey piatnitski .
random homogenisation of a highly oscillatory singular potential .
_ stochastic partial differential equations : analysis and computations _ , 1(4):571605 , 2013 . | these are the notes for a course at the 18th brazilian school of probability held from august 3rd to 9th , 2014 in mambucaba .
the aim of the course is to introduce the basic problems of non linear pdes with stochastic and irregular terms .
we explain how it is possible to handle them using two main techniques : the notion of energy solutions @xcite and that of paracontrolled distributions , recently introduced in @xcite . in order to maintain a link with physical intuitions ,
we motivate such singular spdes via a homogenization result for a diffusion in a random potential . |
in recent years many attempts have been made to understand nucleon structure functions as measured in lepton deep - inelastic scattering ( dis ) .
although perturbative qcd is successful in describing the dependence of structure functions on the squared momentum transfer , their magnitude is governed by the non - perturbative physics of composite particles , and is up to now not calculable directly from qcd .
a variety of models have been invoked to describe nucleon structure functions .
the so called `` spectator model '' is a typical covariant approach amongst them@xcite . in this approach the leading twist , non - singlet quark distributions
are calculated from the process in which the target nucleon splits into a valence quark , which is scattered by the virtual photon , and a spectator system carrying baryon number @xmath4 .
furthermore the spectrum of spectator states is assumed to be saturated through single scalar and vector diquarks .
thus , the main ingredient of these models are covariant quark - diquark vertex functions .
until now vertex functions have been merely parameterized such that the measured quark distributions are reproduced , and no attempts have been made to connect them to some dynamical models of the nucleon . in this work
we construct the vertex functions from a model lagrangian by solving the bethe salpeter equation ( bse ) .
however , we do not aim at a detailed , quantitative description of nucleon structure functions in the present work .
rather we outline how to extract quark - diquark vertex functions from euclidean solutions of the bse . in this context
several simplifications are made .
we consider only scalar diquarks as spectators and restrict ourselves to the @xmath5 flavor group .
the inclusion of vector diquarks and the generalization to @xmath6 flavor are relatively straightforward extensions and will be left for future work .
the cross section for dis of leptons from a nucleon is characterized by the hadronic tensor @xmath7 where @xmath8 and @xmath9 are the four - momenta of the target and exchanged virtual photon respectively . for unpolarized dis ,
the hadronic tensor @xmath10 is conventionally parameterized by two scalar functions @xmath11 and @xmath12 . in the bjorken limit ( @xmath13 ; but finite @xmath14 ) in which we work throughout , both structure functions depend ( up to logarithmic corrections ) on @xmath15 only , and are related via the callan - gross relation : @xmath16 . in the scalar diquark spectator model within @xmath5 flavor ,
the valence quark distributions are extracted from the hadronic tensor ( fig .
[ fig : diquarkspectatorfort ] ) : @xmath17 where the isospin matrix element @xmath18 has to be evaluated in the nucleon isospin space .
we define @xmath19 as the target nucleon spinor and we use @xmath20 and @xmath21 to denote the propagators of the quark and scalar diquark , respectively .
the integration runs over the quark momentum @xmath22 , subject to on mass shell conditions for the diquark and the struck quark .
note that the vertex function @xmath23 and its pt conjugate @xmath24 consist of two lorentz scalar functions which depend on @xmath25 only because the diquark is on shell . in the next section
we shall determine the vertex functions using a ladder bethe - salpeter equation .
we consider the following model lagrangian : @xmath26 where we have explicitly indicated color @xmath6 indices only . the symmetric generator @xmath27 of the flavor @xmath5 group acts on the iso doublet field @xmath28 for the constituent quark carrying an invariant mass @xmath29 .
the charged scalar field @xmath30 denotes the flavor
singlet scalar diquark with invariant mass @xmath31 . in this model
the nucleon with the momentum @xmath8 and spin @xmath32 is described by the bs vertex function @xmath23 ( fig .
[ fig : bsvertex ] ) : @xmath33 where we set the weight factors to the classical values : @xmath34 and @xmath35 .
then the vertex function @xmath23 obeys in the ladder approximation the following bse ( fig.[fig : ladderbse ] ) : @xmath36 to solve the bse for positive energy nucleon states we are free to choose the following dirac matrix structure : @xmath37\lambda^{(+)}(p ) , \label{define_f}\ ] ] with @xmath38 , the projector onto positive energy nucleon states .
we assume that the diquark and the nucleon are stable , namely @xmath39 and @xmath40 .
we then perform the wick rotation of the relative energy variable and choose the nucleon rest frame : @xmath41 .
the `` euclidean '' functions @xmath42 in terms of the momentum @xmath43 for a real @xmath44 are then functions of @xmath45 and @xmath46 $ ] .
we expand each of these functions ( @xmath47 ) as follows : @xmath48 where @xmath49 are gegenbauer polynomials and the phase @xmath50 is introduced for convenience since then the radial functions @xmath51 are real .
the bse in eq.([ladderbse ] ) then reduces to the following system of one dimensional equations : @xmath52 where we have introduced the `` eigenvalue '' @xmath53 for the quark
diquark coupling constant .
the kernel function @xmath54 is a matrix whose elements are real and regular functions of @xmath55 and @xmath56 . by terminating the infinite series in eq.([expanf ] ) at sufficiently high order
, we can easily solve eq.([fn_eq ] ) numerically as an `` eigenvalue '' problem for a fixed bound state mass , @xmath57 . to compare the magnitude of the radial functions ,
let us introduce the normalized @xmath58 radial functions @xmath59 and @xmath60 together with the @xmath58 spherical spinor harmonics @xmath61@xcite . in the conventional gamma matrix representation we can write the nucleon solution at rest as : @xmath62 where @xmath63 denotes angles for @xmath64 vector in the four dimentional polar coordinate system .
the factor @xmath65 is introduced such that @xmath66 , while @xmath60 is expressed as a linear combination of @xmath67 and @xmath68 . thus @xmath69 and @xmath70 correspond to the `` upper '' and `` lower '' components of the nucleon dirac field , respectively .
now , let us consider the analytic continuation of @xmath71 .
since we are interested in applying the bs vertex function to the dis process , we need to rotate the relative energy variable from on the imaginary axis back to on the real one .
recall that the sum over @xmath72 in eq.([expanf ] ) converges even for a complex @xmath73 , if @xmath74 .
we can then analytically continue the gegenbauer polynomials rewriting the argument @xmath75 for the momenta satisfying @xmath76 . for the radial functions we introduce new functions @xmath77 and @xmath78 .
and @xmath79 are functions of @xmath80 was confirmed numerically .
] we analytically continue these functions by changing the argument @xmath81 .
we obtain the physical scalar functions @xmath82 : @xmath83 note that the gegenbauer polynomials together with the square root factors are polynomials of @xmath84 , @xmath57 , and @xmath85 , so that each term in the series ( [ f1minkowski ] ) and ( [ f2minkowski ] ) is regular and real as far as @xmath86 and @xmath87 are regular
. we may then impose the on mass shell condition for the diquark and evaluate the sum over @xmath72 .
the resulting @xmath88 and @xmath89 are then functions of the squared quark momentum @xmath25 and can be applied to the dis process . however , the expressions ( [ f1minkowski ] ) and ( [ f2minkowski ] ) are valid only for momenta satisfying @xmath90 .
also we found that a naive numerical sum over @xmath72 , based on eqs.([f1minkowski ] ) and ( [ f2minkowski ] ) does not converge .
nevertheless , it can be shown that the vertex function @xmath23 for any kinematically allowed @xmath25 is regular when the diquark is on mass shell .
this suggests that one may be able to continue some appropriate linear combinations of @xmath88 and @xmath89 outside of this kinematical range .
indeed , we found such a combination which we denote by @xmath91 . in terms of this on shell scalar functions and the quark momentum , @xmath22 , the vertex function @xmath23 together with the diquark on mass shell condition is given by @xmath92 with this on shell vertex function the valence contribution to the structure function @xmath93 can now be calculated from eq.([compton ] ) .
in this section we present our numerical results . for simplicity we considered an equal mass system ; @xmath94 , and we shall use @xmath95 as a unit for dimensionful quantities . to solve the bse we used a @xmath96channel form factor .
we replaced the quark
diquark coupling constant such that @xmath97 with @xmath98 and @xmath96 is the usual mandelstam variable .
this form factor weakens the short range interaction between the constituents and ensures the existence of a discrete bound state spectrum for a large range of @xmath57 .
we solved eq.([fn_eq ] ) as follows .
first we terminated the infinite series in eq.([expanf ] ) at some fixed value , @xmath99 .
next we discretized the euclidean momentum @xmath55 and @xmath56 and performed the integration over @xmath56 numerically together with some initially assumed radial functions @xmath100 .
this integral generated new radial functions and an `` eigenvalue '' @xmath101 associated with them .
we then used these functions as an input and repeated the above procedure until the radial functions and @xmath101 converged . in fig .
[ fig : fandg ] we plot the normalized @xmath58 radial functions , @xmath59 and @xmath60 , for the bound state mass @xmath102 as functions of @xmath55 .
it is clear that the magnitude of the radial functions with higher @xmath58 angular momentum are strongly suppressed compared with the lowest ones .
this justifies the truncation of the series in eq.([expanf ] ) .
it also confirms that the contribution to the `` eigenvalue '' @xmath101 from the @xmath58 radial functions with @xmath103 is less than 1 % .
the dominance of the lowest @xmath58 radial function has been also observed in the scalar
scalar ladder model@xcite . in fig .
[ fig : fon ] we plot the physical , on shell scalar functions @xmath104 for the bound state mass @xmath102 as functions of the squared quark momentum , @xmath25 .
these functions are calculated with the maximum @xmath58 angular momentum @xmath105 .
we found that the magnitude of @xmath106 and @xmath107 are almost the same even for weakly bound states .
this result suggests that so called `` non relativistic '' approximations , in which one neglects the non leading components of the vertex function ( @xmath89 in our model ) are valid only for extremely weakly bound states : @xmath108 .
we have also confirmed that for weakly bound states ( @xmath109 ) the dependence of @xmath104 on @xmath110 is negligible for a small spacelike @xmath25 , e.g. , @xmath111 .
however , for large spacelike @xmath25 , the convergence of the sum over @xmath72 becomes slow for any value of @xmath57 and numerical results for fixed @xmath110 become less accurate . in fig.[fig :
struct ] we plot the valence quark distribution @xmath93 for a weakly ( @xmath102 , solid line ) and for a strongly bound state ( @xmath112 , dashed line ) .
we have used @xmath105 and the distributions are normalized such that the area below the curve is unity . for the weakly bound system ,
the valence quark distribution has a peak , if quark and diquark masses such as @xmath113 are used . ] around @xmath114 . on the other hand ,
the distribution becomes flat for the strongly bound system .
this behavior however turns out to be mainly of kinematic origin , since the distribution function is given by an integral over the squared momentum @xmath25 carried by the struck quark with integration bounds from @xmath115 to @xmath116 , where @xmath117 this kinematical bound determines the global shape of @xmath93 to a large extent .
we have solved a bse for a nucleon , described as a bound state of a quark and diquark within a covariant quark scalar - diquark model .
we have extracted the physical quark - diquark vertex function when the diquark is on mass shell from the euclidean solution .
the vertex function obtained was applied to a diquark spectator model for dis , and the valence quark contribution to the structure function @xmath93 was calculated .
we found that the shape of the unpolarized valence quark distribution is mainly determined by relativistic kinematics and is independent of the detailed structure of the vertex function .
h. meyer and p.j .
mulders , _ nucl
. phys . _ * a528 * , 589 ( 1991 ) ; w. melnitchouk , a.w . schreiber and a.w .
thomas , _ phys .
rev . _ * d49 * , 1183 ( 1994 ) ; w. melnitchouk , a.w .
schreiber and a.w .
thomas , _ phys .
* b335 * , 11 ( 1994 ) ; w. melnitchouk , g. piller and a.w .
thomas , _ phys .
_ * b346 * , 165 ( 1995 ) .
[ fig : fon ] : the on shell scalar functions @xmath106 ( solid ) and @xmath107 ( dashed ) as functions of the quark momentum @xmath25 .
the mass of the bound sate is @xmath120 and the @xmath58 angular momentum is truncated at @xmath105 . | nucleon structure functions , as measured in deep - inelastic lepton scattering , are studied within a covariant scalar diquark spectator model . regarding the nucleon as an approximate two - body bound state of a quark and diquark ,
the bethe - salpeter equation ( bse ) for the bound state vertex function is solved in the ladder approximation .
the valence quark distribution is discussed in terms of the solutions of the bse .
= 15.9 cm = -1 cm adp-96 - 5/t210 , hep - ph/9603260 + _ talk given at the joint japan - australia workshop on quarks , hadrons + and nuclei , adelaide , south australia , november 15 - 24 , 1995 + ( to appear in the conference proceedings ) _ * structure functions of the nucleon + in a covariant scalar spectator model * + k. kusaka@xmath0 , g. piller@xmath1 , a.w .
thomas@xmath2 and a.g .
williams@xmath2 + _ @xmath0department of physics and mathematical physics , university of adelaide , + s.a . 5005 , australia _ + + _
@xmath3institute for theoretical physics , university of adelaide , s.a . 5005 ,
australia _ + e - mail : _ kkusaka , awilliam , [email protected]_ + |
long - duration gamma - ray bursts ( grbs ) are believed to arise when a very massive star collapses to a black hole , powering a hyper - relativistic outflow ( woosley 1993 ; hjorth et al .
2003 ; stanek et al . 2003 ; malesani et al . 2004
; woosley & bloom 2006 ) .
they are the brightest explosions in the present day universe , releasing , in a matter of few tens of seconds , a large amount of energy ( @xmath0 erg ) in the form of a hot , possibly magnetized , outflow .
grbs are able to radiate a large fraction of their energy in the form of photons , with an efficiency sometimes approaching 100% ( zhang et al .
2007 ) .
the commonly accepted model for the dissipation of the jet energy and its conversion into radiation is the internal shock synchrotron model ( rees & meszaros 1994 ) . in this model ,
differential velocities within the outflow are dissipated through collisionless shock waves that generate strong magnetic fields , accelerate a population of relativistic electrons , and eventually release synchrotron radiation ( e.g. , piran 1999 ; lloyd & petrosian 2000 ) .
internal shocks suffer however from poor efficiency ( kobayashi et al .
1997 ; lazzati et al . 1999 ; spada et al . 2000 ) , being capable of radiating only a few per cent of the total energy in the outflow . in their early evolution ,
the ouflows of long - duration grbs interact strongly with the cold and dense material of their progenitor star ( macfadyen & woosley 1999 ; aloy et al .
2000 ; zhang et al . 2003 , 2004 ) .
the interaction strongly modifies the dynamics of the jet while it is still inside the star .
evaluating the effects of the jet - star interaction at larger scales , when the jet is expanding in the circum - stellar material , is of fundamental importance , since a change in the jet dynamics can profoundly affect our understanding of the dissipation and radiation processes at play in the prompt phase of gamma - ray bursts ( rees & meszaros 2005 ; ghisellini et al .
2007 ) . here
we present the results of a hydrodynamic simulation that follows a gamma - ray burst jet out to 65 stellar radii , where the jet material becomes transparent to radiation .
our calculation shows that the interaction of the jet material with the star profoundly affects the subsequent evolution of the jet , generating continuous dissipation through transverse collimation shocks .
this letter is organized as follows : in 2 we present the numerical model adopted for the jet and the progenitor star , in 3 we present our result and compute light curves , and in 4 we discuss our findings .
we considered a 16 solar - mass wolf - rayet progenitor star , evolved to pre - explosion ( model 16ti , woosley & heger 2006 ) .
the jet was introduced as a boundary condition at a distance of @xmath1 cm from the center of the star , with a total luminosity @xmath2 erg / s , an initial opening angle @xmath3 , an initial lorentz factor @xmath4 , and a ratio of internal over rest mass energy @xmath5 , allowing for a maximum lorentz factor of @xmath6 , reachable in case of complete , non - dissipative acceleration .
the jet evolution was computed with the special - relativistic hydrodynamic code flash ( fryxell et al .
2000 ) .
figure [ fig : frames ] shows density images from three frames of the simulation at @xmath7 s , @xmath8 s , and @xmath9 s. the upper panel shows the collimating power of the stellar material .
the jet that was injected with an opening angle of 10 degrees is squeezed into a 2-degree opening angle , with the typical sausage shape due to the presence of strong tangential collimation shocks ( morsony et al .
this simulation extends into a box ten times bigger than any previously performed in a single run and was run for 150 seconds .
the maximum spatial resolution along the jet axis was @xmath10 cm , while the maximum spatial resolution in the outer jet was @xmath11 cm . in the central and bottom panels it is possible to see that the jet remains affected by the shocks out to many stellar radii ( 20 in the central panel and about 60 in the bottom panel ) .
a zoom around the jet region in the last panel of fig .
[ fig : frames ] shows the effect of the interaction of the outflow with the progenitor material more quantitatively ( fig .
[ fig : shocks ] ) .
the head of the jet shows shocks whose normal is parallel to the jet axis and propagate along the jet ( similar to internal shocks or to the scenario of reborn fireballs , ghisellini et al .
2007 ) , while the part of the jet behind the head is characterized by tangential shocks that propagate from the side of the jet to its axis and vice versa .
the middle panel shows the ratio of the internal energy to the rest mass energy .
the dashed blue line shows the theoretical prediction for a freely expanding outflow , neglecting the effect of the interaction with the progenitor material . in that case ,
@xmath12 , where @xmath13 is the ratio of internal to rest energy when the jet is released at a radius @xmath14 .
the red line shows the result of our simulation along the jet axis .
note that the horizontal scales are equal among panels and the one to one correspondence between the shocks and the increases in internal energy .
the internal energy dominates over the rest mass energy much farther out than the simple theory predicts , being @xmath15 times larger in the region between @xmath16 and @xmath17 cm .
the bottom panel of fig .
[ fig : shocks ] shows the lorentz factor along the jet spine .
the opacity for thompson scattering of a relativistically moving medium is ( abramowicz et al .
1991 ) : @xmath18 where the comoving density @xmath19 , the lorentz factor @xmath20 and the angle between the line of sight and the velocity of the fluid ( @xmath21 ) are evaluated at a laboratory time @xmath22 , where @xmath23 is the laboratory time at which the photon is released at a radius @xmath24 .
the photosphere is defined as the radius @xmath25 for which @xmath26 .
knowing the photospheric radius as a function of time and angle we computed the light curve by integrating a thermal spectrum over the angular direction at the comoving temperature boosted by relativistic effects : @xmath27 where @xmath28 erg @xmath29 k@xmath30 is the radiation constant and @xmath31 is the comoving radiation temperature , which is derived from the pressure as @xmath32 .
results are shown in fig .
[ fig : lcur ] .
the photospheric light curve ( panel a ) shows an intensity and a temporal evolution fairly similar to observed grbs .
it is worth noting here that the jet material is fully ionized and opacity is mainly provided by thompson scattering , with a minor contribution from free - free processes . in such conditions ,
the spectrum is slightly broader than a black body ( goodman 1986 ) , but has the same peak frequency and fairly similar asymptotic behavior .
in addition , a population of non - thermal particles could be present due to the shocks that keep the photosphere hot .
such non - thermal electrons would create a non - thermal tail at high enrgies due to inverse compton scattering .
however , if the high energy tail extend to comoving energies @xmath33 kev , photon - photon interaction may become relevant as a source of opacity for high energy photons as well as a source of electron - positron pairs that could alter the location of the photosphere ( e.g. , meszaros et al .
the details of the opacity calculation depend on the assumed non - thermal tail of the spectrum . in this letter
, we concentrate on the thermal emission from the photosphere and therefore the pairs are not an issue .
the explanation of grb spectra extending to energies larger than @xmath34 mev require a more detailed calculation and assumptions on the spectrum of non - thermal particles that go beyond the scope of this letter .
the kinetic energy remaining in the jet beyond the photosphere was used to evaluate the minimum radiative efficiency .
it is about 50 per cent , with an average of 56 per cent in the first 40 seconds of the light curve ( see fig . [
fig : lcur ] , panel b ) . figure [ fig : lcur ] , panel ( c ) , shows the photospheric radius along the jet axis as a function of time , while panel ( d ) shows the observed peak of the photospheric emission in kev .
the peak of a few hundred kev seen in the simulation is typical for long - duration grbs ( kaneko et al .
note that at times longer than @xmath35 s the jet enters the unshocked jet phase " , as discussed in morsny et al .
( 2007 ) . in this phase ,
the photospheric efficiency drops since tangential shock are absent in the unshocked " phase .
our simulation shows that the interaction of the relativistic outflow with the progenitor material in long duration grbs has effects that are important well after the jet has left the star .
the jet does not cool and has a very bright photosphere yielding a high radiative efficiency in the prompt phase .
in addition , the photospheric component is quasi - thermal ( rees & meszaros 2005 ; thompson et al . 2007 ; peer 2008 ) offering an explanation for the hard x - ray spectra in the early phase of gamma - ray bursts , which is inconsistent with synchrotron ( preece et al . 1998 , ghisellini et al .
thermal components have been detected recently in batse spectra ( ghirlanda et al .
2003 , 2007 ; ryde et al . 2005 ; ryde & peer 2008 ) and the fermi gamma - ray telescope will soon give us a wideband view of the bursts prompt emission .
non - thermal components can be added to the spectrum as a result of the reprocessing of photospheric emission and of additional dissipation due to the large inhomogeneities in the jet lorentz factor ( see bottom panel of fig .
2 ) . additional
work is however required to assess the capability of this model to explain the very high - energy photons observed , e.g. , in grb 080916c ( abdo et al .
unless an extremely large value of the lorentz factor of the fireball is assumed ( @xmath36 to explain photons of 10 gev ) , photons of gev energy need to be released at a radius of at least @xmath37 cm ( zhang & peer 2009 ) .
photospheres can not be located at such large radii , even if the opacity is increased by electron - positron pair production .
if the kev to mev photons of grb 080916c are from the photosphere , as we argue , a second emission mechanism is required for the gev component .
kumar & barniol duran ( 2009 ) argue that the @xmath38 mev photons observed in grb 080916c must be produced in the external shock due to their temporal evolution that is different from the one of the kev to mev photons .
in such a scenario , the kev to mev radiation of grb 080916c could be photospheric .
it is also possible that photons of the photospheric emission are inverse compton scattered by hot electrons at large radii ( heated , e.g. , in an internal shock ) to produce a high energy tail in the gev band .
our computation has two main limitations . on the one hand
, it represents one single case of a jet , with given characteristics , interacting with a given star .
it is not straightforward to predict how different would be the photospheric component for a different jet with a different progenitor star .
based on the result of simulations of smaller domain ( fig .
[ fig : other ] ) we can speculate that bigger stars will produce a stronger photospheric component and that the diversity in progenitor population would explain the range of thermal to non - thermal ratios and efficiencies observed in gamma - ray bursts .
higher luminosity jets are also expected to produce brighter photospheres , since they produce thicker cocoons ( aloy et al .
mizuta & aloy ( 2008 ) performed a systematic study of jets into a diverse set of progenitor stars , finding that the early jet evolution is largely affected by the structure of the stellar progenitor . on the other hand
, the simulation does not have a magnetic component , but is still debated whether the launching and collimation of the jet are due to magnetic effects ( e.g. , mckinney 2006 ) or not ( see , e.g. , aloy & obergaulinger 2007 ) . despite these limitations , our result opens a new window for the understanding of radiation from grb jets .
we thank the anonymous referee for a careful review that improved the clarity and content of this letter .
we would like to thank chris fryer , ehud nakar , asaf peer , daniel proga , felix ryde , jay salmonson , and bing zhang for insightful discussions .
the software used in this work was in part developed by the doe - supported asc / alliance center for astrophysical thermonuclear flashes at the university of chicago .
this work was supported in part by nasa atp grant nng06gi06 g and swift gi program nnx06ab69 g and nnx08ba92 g .
we thank ncsa and the nasa nas for the generous allocations of computing time . | we numerically analyze the evolution of a long - duration gamma - ray burst jet as it leaves the progenitor star and propagates to the photospheric radius , where radiation can be released .
we find that the interaction of the relativistic material with the progenitor star has influences well beyond the stellar surface .
tangential collimation shocks are observed throughout the jet evolution , out to about 100 stellar radii , which is the whole range of our simulation .
we find that the jet is internally hot at the photospheric radius and we compute the photospheric emission .
the photosphere is a very efficient radiator , capable of converting more than half of the total energy of the jet into radiation .
we show that bright photospheres are a common feature of jets born inside massive progenitor stars and that this effect can explain the high radiative efficiency observed in long - duration bursts . |
in this paper , we study binary linear @xmath0 $ ] codes ( codes of length @xmath1 and dimension @xmath2 ) used for error detection on the binary symmetric channel . a comprehensive introduction to the field
is given in @xcite .
the basic definitions are given in section [ sub : errordetection ] .
a main quantity is the _ probability of undetected error _ of a code .
if the probability of undetected error is an increasing function on the interval @xmath3 $ ] , the code is known as _ proper _ for error detection .
it is believed that proper codes exist for all lengths @xmath1 and dimensions @xmath2 .
however , this has been shown only for some cases .
in particular , proper @xmath0 $ ] codes are known to exist for any given @xmath2 when @xmath1 is sufficiently large . the best known result in this direction
was given by klve and yari @xcite who showed that proper codes exist for @xmath4 in this paper , we study a particular class of @xmath0 $ ] codes where @xmath5 .
one of our results is that these codes are proper for many values of @xmath1 and @xmath2 where the existence of proper codes was previously unknown . in particular , we improve the bound ( [ kybound ] ) .
we first consider @xmath1 in the range @xmath6 .
the hamming bound proves that the dual of an @xmath0 $ ] code in this case has minimum distance at most 3 .
moreover , an @xmath7 $ ] code with minimum distance 3 can be obtained by shortening the @xmath8 $ ] hamming code .
two @xmath0 $ ] codes @xmath9 and @xmath10 are equivalent if there exists a permutation @xmath11 of @xmath12 such that @xmath13 if two codes are equivalent , then it may happen that the corresponding ( repeatedly ) punctured codes are not equivalent .
let @xmath14 be some @xmath15 matrix having as columns all possible nonzero vectors of length @xmath2 .
the code generated by @xmath14 is the simplex code @xmath16 , and the code having @xmath14 as parity check matrix is the well - known hamming code .
note that the order of the columns is not specified ; all the equivalent codes are named hamming codes .
however , when we want to puncture the code , the order is very important .
we remind the reader that puncturing a code is equivalent to shortening the dual code .
davydov et al .
@xcite determined an ordering of the columns in @xmath14 such that any of the corresponding ( repeatedly ) shortened codes contains a minimal number of codewords of weight three ; the shortened @xmath7 $ ] codes are obtained by removing @xmath17 columns from @xmath14 to get a @xmath18 matrix @xmath19 and use this matrix as the parity check matrix for the @xmath7 $ ] code .
they showed that a possible choice of @xmath19 is to have as columns the vectors that are the binary representation of the numbers from @xmath20 down to @xmath21 .
for example @xmath22 we let @xmath23 denote the code generated by @xmath19 . for our investigation
, we will consider codes that are equivalent ( but not equal ) to these codes ; we will denote them by @xmath24 .
a main reason for considering @xmath24 rather than @xmath23 is that the determination of the weight distribution is easier for @xmath24 . in this paper , we investigate the performance of the codes @xmath24 when they are used for error detection .
we compute their weight distribution that , in turn , permits us to calculate the undetected error probability @xmath25 .
however , when the code length @xmath1 is large ( @xmath26 ) , the polynomial expressing @xmath25 may be difficult to evaluate , even when the weight distribution is known .
for this reason , in the general case , we also find bounds on the length and dimension such that a necessary condition for codes to be satisfactory does not hold , using a method similar to the one proposed in @xcite .
the paper is organized as follows : in section [ sub : errordetection ] we give some preliminaries on error detection ; in section [ sub : dkstconstruction ] we describe the construction of @xmath24 and show that @xmath24 and @xmath23 are equivalent ; in section [ sec : weights - distribution ] we determine the weight distribution of @xmath24 ; in section [ sec : prob ] we study the undetected error probability of @xmath24 and its dual code ; in section [ sec : approximate - analysis ] we give an asymptotic analysis ; in section [ sec : generalization ] we give a generalization of the construction to lenghts @xmath27 , finally , in section [ sec : results ] we summarize our results .
we start by defining @xmath28 , the undetected error probability for an @xmath0 $ ] code @xmath9 when used on the binary symmetric channel with error probability @xmath29 : @xmath30 where @xmath31 is the number of codewords having hamming weight @xmath32 , see e.g. ( * ? ? ?
* section 2.1.2 ) .
one can also express this polynomial in terms of the weight distribution of the dual code , see e.g. ( * ? ? ?
* theorem 2.4 ) .
if @xmath33 is the number of codewords having hamming weight @xmath32 in the dual code @xmath34 , we have : @xmath35 as mentioned in the introduction , if @xmath28 is an increasing function on @xmath36 $ ] , the code @xmath9 is called _ proper _ for error detection . if @xmath37 for every @xmath38 $ ] , @xmath9 is called _ good _ for error detection .
if @xmath39 for every @xmath38 $ ] , @xmath9 is called _ satisfactory _ for error detection , see @xcite .
a code that is not satisfactory is called _
ugly_. when a code is proper then it is satisfactory ; so , if it is ugly it is clearly not proper ( nor good ) .
we first describe a particular parity check matrix @xmath14 for the hamming code . for @xmath40 ,
let @xmath41 be the @xmath42 matrix constructed as follows : * the first @xmath43 rows are all - zero vectors .
* row @xmath44 is the all - one vector . * in the @xmath45 matrix consisting of the last @xmath46 rows , the columns are ordered lexicographically .
then @xmath47.\ ] ] we illustrate this with an example . for @xmath48 , we get @xmath49 @xmath50 and so @xmath51 \\ = & \begin{bmatrix } 111111110000000\\ 000011111111000\\ 001100110011110\\ 010101010101011 \end{bmatrix}.\end{aligned}\ ] ] we let @xmath52 denote the @xmath18 matrix containing the first @xmath1 columns of @xmath14 . for example @xmath53 we let @xmath24 denote the code generated by @xmath52 .
we see that @xmath54 is the first order reed - muller code and @xmath55 , the simplex code .
both of these codes are known to be proper ( and this is easy to show ) .
the hamming code is @xmath56 .
the code having @xmath52 as parity check matrix is a shortened hamming code which we denote by @xmath57 .
we note that @xmath58 . in the rest of the paper ( except section [ sec : generalization ] ) we will assume that @xmath59 .
[ dkst - bs ] the codes @xmath24 and @xmath23 are equivalent .
we first illustrate by the example @xmath48 and @xmath60 , that is , the matrices ( [ h4 ] ) and ( [ d4 ] ) . adding the second row in ( [ h4 ] ) to the third and forth rows , we get @xmath61 this is an alternative generator matrix for @xmath62 .
the last three columns are the same in ( [ hh4 ] ) and ( [ d4 ] ) , and the first eight columns of ( [ hh4 ] ) are a permutation of the first eight columns in ( [ d4 ] ) .
hence , @xmath62 and @xmath63 are equivalent . in the general case , if @xmath64 $ ] for some @xmath46 , @xmath65 , we add row @xmath66 in @xmath52 to all the rows below .
this gives an alternative generator matrix @xmath67 for @xmath24 .
the first @xmath68 columns of @xmath67 are a permutation of the binary representations of @xmath69 $ ] , the next @xmath70 columns of @xmath67 are a permutation of the binary representations of @xmath71 $ ] , etc .
the final @xmath72 columns in @xmath67 and @xmath19 are the same .
hence , @xmath24 and @xmath23 are equivalent .
if @xmath73 for some @xmath46 , @xmath74 , the same argument shows that the columns of @xmath52 are a permutation of the columns of @xmath19 , and so again @xmath24 and @xmath23 are equivalent .
the main question we consider is : for which @xmath1 and @xmath2 is @xmath24 proper for error detection ? we will also in some cases consider the simpler question : for which @xmath1 and @xmath2 is @xmath24 satisfactory for error detection ?
we note that this is equivalent to the question : for which @xmath1 and @xmath2 is @xmath57 satisfactory for error detection ?
the reason is the following known lemma .
* theorem 2.8 ) . [ sdual ]
a code is satisfactory if and only if the dual code is satisfactory . to determine the probability of undetected error for @xmath24
, we have to determine its weight distribution .
this is done in this section .
we break the argument down into a number of lemmas .
we first give some further notations .
we observe that the matrix @xmath75\ ] ] has length @xmath76 for a given @xmath1 , let @xmath46 be determined by @xmath77 since @xmath78 , we have @xmath65 .
let @xmath79 denote the last column of @xmath52 .
[ alpha - lemma ] let @xmath80 .
then @xmath81 and @xmath82 are determined by @xmath83 the last column in @xmath52 is a column in @xmath84 .
hence ( [ alm ] ) follows . moreover , its number in @xmath84 is @xmath85 when we count the first column as number zero .
the columns in @xmath84 are ordered lexicographically and so ( [ alr ] ) follows .
let @xmath86 denote the weight of the @xmath87-th row in @xmath52 . as usual
, @xmath88 denotes the largest integer less than or equal to @xmath89 .
[ w - lemma1 ] let @xmath80 . then @xmath90 and @xmath91 if @xmath92 and @xmath93 , then @xmath94 if @xmath92 and @xmath95 ,
then @xmath96 all the rows of @xmath14 have weight @xmath68 .
the first @xmath46 rows of @xmath52 are obtained from rows in @xmath14 removing some zeros .
hence @xmath97 for @xmath98 .
before we go on with the proof , let us take a closer look at @xmath41 .
row @xmath99 consists of consecutive blocks of zeros and ones , each block of length @xmath100 .
we use the term _
double block _ for a zero - block combined with the following one - block ; it has length @xmath101 .
now , let @xmath102 then row @xmath87 in @xmath52 consists of @xmath103 double blocks of length @xmath101 , each of weight @xmath100 , followed by an incomplete double block of length @xmath104 that has to be considered further ( when @xmath105 , the incomplete double block is , of course , a full double block ) . if @xmath93 , the incomplete double block is all zero , and so ( [ wm2 ] ) follows . if @xmath95 , the incomplete double block consists of a full block ( of length @xmath100 ) of zeros followed by an incomplete block of ones of length @xmath106 .
hence @xmath107 this proves ( [ wm3 ] ) .
the proof of ( [ wm1 ] ) is similar ( and even simpler ) .
[ lem1 ] consider sums of rows from @xmath52 .
\a ) any of the @xmath108 non - zero sums of some of the first @xmath46 rows have weight @xmath68 .
\b ) any of the @xmath109 sums containing row @xmath66 and zero or more previous rows have weight @xmath110 .
\c ) for @xmath111 , @xmath112 sums containing row @xmath87 and some previous rows have weight @xmath113 and the other @xmath112 sums have weight @xmath114
. for each sum of rows from the first @xmath46 , the corresponding sum of rows in @xmath14 are codewords in the simplex code @xmath16 .
these always have weight @xmath68 . since only zeros have been removed to get the corresponding rows in @xmath52 , their sum also has weight @xmath68 .
this proves a ) .
let @xmath115 .
we note that in the set of positions of a double block in row @xmath115 in @xmath116 for @xmath117 , the elements of any previous row are all zero or all one .
therefore , the weight of these positions in any sum of row @xmath87 and a combination of previous rows is @xmath100 .
it remains to consider the contribution to the weight from the last @xmath104 positions ( where @xmath104 is defined by ( [ nu ] ) ) .
case i ) @xmath93 . in this case , all the last @xmath104 elements of row @xmath87 are zeros .
any previous row has all zeros or all ones in these positions , and so the weight of the elements in these positions in any sum is either 0 or @xmath104 .
hence , the weight of the sum is either @xmath86 or @xmath118 , where @xmath86 is given by ( [ wm2 ] ) .
moreover , row @xmath66 has all ones in the last @xmath104 positions .
hence , half of the @xmath119 sums has weight @xmath86 and the other half has weight @xmath118 . case ii ) @xmath95 . in this case , all the last @xmath104 elements of row @xmath87 are @xmath100 zeros followed by @xmath106 ones .
the weight of the last @xmath104 elements in a sum is therefore @xmath106 or @xmath100 .
hence , the weight of a sum is @xmath86 or @xmath118 , where now @xmath86 is given by ( [ wm3 ] ) . as done above ,
considering sums containing row @xmath66 , we can see that the multiplicities of these two weights are the same .
this proves c ) .
finally , consider row @xmath66 .
any previous row has all zeros in the last @xmath104 positions .
hence , the weight of any sum involving row @xmath66 and previous rows is @xmath110 .
this proves b ) .
we next give an alternative expression for @xmath86 .
[ w - alt ] let @xmath80 . + a ) if @xmath92 and @xmath93 , then @xmath120 b ) if @xmath92 and @xmath95 , then @xmath121 c ) further , @xmath122 from ( [ alr ] ) we get @xmath123 hence @xmath124 where @xmath125 hence @xmath126 and so , by ( [ wm2 ] ) , @xmath127 this proves a ) . similarly , if @xmath128 , ( [ wm3 ] ) gives @xmath129 since @xmath95 .
this proves b ) .
finally , c ) follows directly by substituting the expression for @xmath1 in the expression for @xmath110 in ( [ wm1 ] ) .
[ ex0 ] consider @xmath130 , where @xmath131 .
we have @xmath132 using lemma [ w - alt ] , we see that @xmath24 has @xmath108 codewords of weight @xmath133 and @xmath134 codewords of weight @xmath135 in particular , the minimum distance is @xmath136 . hence , the code @xmath24 is proper ( see ( * ? ? ? * theorem 2.2 ) ) .
[ w - diff ] let @xmath137 and @xmath92 .
\a ) if @xmath138 , then @xmath139 .
\b ) if @xmath93 and @xmath140 , then @xmath141 .
c1 ) if @xmath95 , @xmath142 , and @xmath143 for all @xmath144 such that @xmath145 , then @xmath139 .
c2 ) if @xmath95 , @xmath142 , and @xmath146 for at least one @xmath147 , then @xmath141 .
\d ) in all cases , @xmath148 in particular , the minimum distance @xmath149 of @xmath24 is @xmath150 .
\e ) @xmath151 .
\a ) if @xmath152 , then lemma [ w - alt ] gives @xmath153 if @xmath154 , then lemma [ w - alt ] gives @xmath155 \b ) if @xmath93 and @xmath140 , then lemma [ w - alt ] gives @xmath156 \c ) if @xmath95 and @xmath142 , then lemma [ w - alt ] gives @xmath157 we have @xmath158 with equality if and only if @xmath143 for @xmath145 .
\d ) we have @xmath159 and so @xmath160 further , both for @xmath161 and @xmath162 , lemma [ w - alt ] gives @xmath163 for @xmath92 , a ) , b ) , c1 ) , and c2 ) show that @xmath164 .
\e ) equation ( [ wm1 ] ) implies that @xmath165 let @xmath166 the midpoint of the interval @xmath167 $ ] .
[ dmin ] let @xmath149 be the minimum distance of @xmath24 .
\a ) if @xmath168 , then @xmath169 \b ) if @xmath170 , then @xmath171 \a ) we have @xmath172 hence @xmath173 . by lemma [ w - alt]a ) , @xmath174 \b ) we have @xmath175 . from ( [ alr ] ) and lemma [ w - alt]b ) , @xmath176 [ admin ] a )
if @xmath177 , then @xmath178 \b ) if @xmath179 and @xmath115 is given by @xmath180 then @xmath181 in particular , @xmath182 in all cases .
the conditions imply that @xmath183 has value @xmath149 exactly for @xmath184 ( where @xmath185 for case a ) ) . hence @xmath186
from ( [ eq : puecode ] ) , ( [ eq : puedual ] ) , and lemma [ lem1 ] , we get the following theorems .
[ th1 ] let @xmath80 .
then @xmath187 [ th1dual ] let @xmath80
. then @xmath188 [ ex1 ] consider @xmath189 where @xmath190
. then @xmath191 using lemmas [ w - lemma1 ] and [ w - alt ] we get @xmath192 hence , the weight distribution of @xmath24 is given by table [ tabnr ] .
@xmath193 for @xmath194 and @xmath195 in example [ ex1 ] we get @xmath196 and @xmath197 in fig .
[ s320 ] we give the graphs of @xmath198 and the terms @xmath199 and @xmath200 .
the contributions from the last two terms , @xmath201 and @xmath202 are so small that they are not visible on the graph .
the graph illustrates that @xmath203 is ugly . for small @xmath29 ( @xmath29 up to approximately 0.42 ) ,
@xmath199 is the dominating term ; in this region the difference @xmath204 is so small that it is not visible on the graph . for @xmath29 close to 0.5 , the term @xmath200 dominates .
( solid line ) , @xmath199 ( dashed line ) , @xmath200 ( dotted line ) , and @xmath205 ( long dashed ) . ] theorem [ th1 ] can be used to determine if the code @xmath24 is proper and theorem [ th1dual ] if the code @xmath57 is proper .
we just compute @xmath206 and check the presence or absence of real roots in @xmath207 . for moderate values of @xmath1 and @xmath2 ( e.g. @xmath208 ) , this is feasible in a reasonable time . before we give a main general result , we quote two lemmas from @xcite .
[ incr ] a ) @xcite theorem 2.2 : if @xmath209 , then @xmath210 is increasing on @xmath3 $ ] .
\b ) @xcite lemma 3.5 : if @xmath211 , then @xmath212 is increasing on @xmath3 $ ] .
let @xmath213 @xmath214 [ s - proper ] for @xmath215 , if @xmath216 or @xmath217 then @xmath24 is proper . for @xmath73 and @xmath130 ,
@xmath24 is proper by example [ ex0 ] . for @xmath218 , ( [ wm ] ) , lemma [ w - diff]d ) , and lemma [ w - diff]e ) imply that @xmath219 and so @xmath220 is increasing on @xmath3 $ ] for @xmath221 by lemma [ incr]a ) .
now , consider @xmath222 , where @xmath223 . by lemma [ dmin]a )
@xmath224 by lemmas [ w - diff]d ) and [ incr]b ) , @xmath225 is increasing for @xmath92 if @xmath226 , that is , if @xmath227 this is equivalent to @xmath228 and @xmath229 solving this for @xmath230 , we get @xmath231 we see that if ( [ kcon1 ] ) is satisfied , then all the terms in @xmath232 are increasing on @xmath3 $ ] . consequently , @xmath24 is proper .
next , let @xmath233 where @xmath234 .
then , by lemma [ dmin]b ) , @xmath235 we want @xmath236 solving for @xmath230 , we get @xmath237 and so @xmath238 as above , if ( [ kcon2 ] ) is satisfied , then @xmath24 is proper .
when @xmath24 is proper , then it is satisfactory , and so , by lemma [ sdual ] , @xmath57 is satisfactory .
hence we get the following corollary .
[ c - kange ] if @xmath1 is in the range defined by ( [ kcon1 ] ) or ( [ kcon2 ] ) for some @xmath239 , then @xmath57 is satisfactory .
[ p - kange ] a ) if @xmath240 then @xmath24 is proper for all @xmath241 $ ] .
\b ) @xmath24 is proper for all @xmath242 $ ] .
we have @xmath243 if and only if @xmath244 we observe that @xmath245 decreases with increasing @xmath46 and @xmath246 increases with increasing @xmath46 .
. then ( [ full ] ) is equivalent to the following sequence of inequalities @xmath248 @xmath249
@xmath250 for @xmath251 we get @xmath252 and so @xmath253 for all @xmath239
. however , for @xmath254 we get @xmath255 and so @xmath256 hence , ( [ full ] ) is satisfied if and only if @xmath257 , that is when ( [ mb ] ) is satisfied .
therefore , if ( [ mb ] ) is satisfied , then @xmath24 is proper for all @xmath258 $ ] .
next , since @xmath259 , we see that if ( [ full ] ) is satisfied , then @xmath260 hence , @xmath261 , and so @xmath24 is proper also for all @xmath262.\ ] ] this , combined with the result above , proves a ) .
since @xmath24 is proper for @xmath263\ ] ] for all @xmath264 , b ) follows .
based on the previous theorems , we have found a set of values of @xmath1 for which @xmath24 is proper and , hence , satisfactory .
for other values of @xmath1 , the existence of real roots of @xmath265 in @xmath266 must be checked . however , for large values of @xmath2 and @xmath1 in general it may be difficult to numerically compute the polynomial s real roots , or even just to determine the existence of real roots ( e.g. using sturm s chain ) . however , in many cases we can decide that the code @xmath24 ( and hence @xmath57 ) is not satisfactory ( i.e. ugly ) by showing that @xmath267 for some value of @xmath29 . how should the value of @xmath29 be chosen ?
there is no theory that can give an exact answer to this question .
however , it is known that if the minimum distance of the code is @xmath149 , then @xmath268 is often the dominating term of @xmath28 , except for large @xmath29 .
this is well illustrated by the example of @xmath203 given in fig .
[ s320 ] . since @xmath269 has its maximum for @xmath270 , a good choice for @xmath29 may be @xmath270 .
this gives the following sufficient condition for @xmath24 to be ugly : @xmath271 where @xmath272 is the binary entropy function .
we can reformulate this to the following well - known sufficient condition for a code to be ugly ( see e.g.
* theorem 2.11 ) or @xcite ) : @xmath273 we showed in example [ ex0 ] that @xmath274 is proper for all @xmath275 . in general , @xmath24 and @xmath57 may be ugly for some values of @xmath1 when @xmath276 .
we have checked that @xmath24 is proper for all @xmath277 when @xmath278 .
when @xmath279 , @xmath24 is ugly for some values of @xmath1 .
an example is @xmath203 in fig .
[ s320 ] .
[ var ] for a given @xmath2 , let @xmath280 then @xmath281 is increasing with @xmath1 on @xmath282 $ ] and decreasing with increasing @xmath1 on @xmath283 $ ] , where @xmath284 was given in ( [ nrm ] ) . from the definitions of @xmath285 and @xmath281 , we get @xmath286 by lemma [ dmin]a ) , @xmath149 is constant for @xmath287 $ ] .
considering @xmath1 as a real variable for the moment , direct calculations gives @xmath288 since @xmath289 , we get @xmath290 . similarly , for @xmath291 $ ] , @xmath292 is constant by lemma [ dmin]b ) , and so @xmath293 note : the weight distribution of @xmath294 was given in table [ tabnr ] .
in particular , @xmath295 for @xmath296 . moreover , @xmath182 for all @xmath297 $ ] . hence , ( [ per ] ) is satisfied for some such @xmath1 if and only if it is satisfied for @xmath296 . for @xmath298 we have , from table [ tabnr ] , that @xmath299 and so @xmath300 for a fixed @xmath46 , let @xmath301 where we consider @xmath2 a real variable , and where @xmath302 then ( [ per ] ) , for @xmath296 , can be rewritten as @xmath303 we get @xmath304 to analyze @xmath305 further , we first give some relations for @xmath285 .
[ h - lemma ] for @xmath306 , we have @xmath307 and @xmath308 using taylor s theorem , we get @xmath309 for @xmath310 .
the upper bound ( [ ent ] ) follows immediately since @xmath311 next , from ( [ ht ] ) , we get @xmath312 since @xmath313 for @xmath314 and @xmath315 , we get @xmath316 [ u - lemma ] for @xmath239 , we have @xmath317 where @xmath318
we have @xmath319 hence , from ( [ u - def ] ) and ( [ ent ] ) we get @xmath320 this proves the lower bound on @xmath321 . similarly ,
( [ u - def ] ) and ( [ entl ] ) imply the upper bound on @xmath321 .
in particular , ( [ kb ] ) and lemma [ u - lemma ] imply that @xmath322 for all @xmath239 . since @xmath323 for all @xmath2 and @xmath324 when @xmath325 , we see that @xmath326 has a unique root in @xmath327 , we denote it by @xmath328 .
further , @xmath329 for @xmath330 .
also , @xmath331 and @xmath305 is decreasing for @xmath332 .
[ not - int ] for @xmath239 , @xmath328 is not an integer . for @xmath298
, we have @xmath299 @xmath333 and @xmath295 . by definition , @xmath334 if @xmath335 or equivalently , @xmath336 hence , if @xmath334 were an integer , then the exact powers of 2 dividing the two sides of ( [ eq2 ] ) would be the same .
we will show that this is not the case .
the exact power of 2 dividing @xmath337 is @xmath338 the exact power of 2 dividing @xmath339 is @xmath340 that is , we have a contradiction .
hence , @xmath328 is not an integer .
let @xmath341 be the smallest integer @xmath2 such that @xmath331 . then @xmath342 and so we have the following : @xmath343 in table [ k0tab ] we give the values of @xmath341 for @xmath344 .
@xmath345 & [ 5,14 ] & [ 15,36 ] \\
\hline k(m ) & 9 & 2m+8 & 2m+9 & 2m+10 \\ \hline \hline m \in & & [ 37,81 ] & [ 82,172 ] & [ 173,356 ] \\ \hline k(m ) & & 2m+11 & 2m+12 & 2m+13 \\
\hline \end{array}\ ] ] @xmath346&&&\\ \hline 10 & [ 599,676]&&&\\ \hline 11 & [ 1140,1396]&&&\\ \hline 12 & [ 2219,2878 ] & [ 3286,3367]&&\\ \hline 13 & [ 4331,5853 ] & [ 6458,6844]&&\\ \hline 14 & [ 8540,11878 ] & [ 12717,13888 ] & [ 14812,14883]&\\ \hline 15 & [ 16870,23966 ] & [ 25208,28006 ] & [ 29371,30013]&\\ \hline 16 & [ 33486,48290 ] & [ 50034,56408 ] & [ 58305,58368]\,[58370,60396]\ , [ 60416,60461 ] & [ 62460,62468]\\ \hline 17 & [ 66546,66560]\,[66593,97028 ] & [ 99602,113304 ] & [ 116102,121434 ] & [ 124378,125472 ] \\
\hline 18 & [ 132560,194804 ] & [ 198432,227423 ] & [ 231354,231424]\,[231451,243631]\,[243712,243730 ] & [ 247954,251741 ] \\ \hline \end{array}\ ] ] we will next determine good bounds on @xmath328 .
these can in turn be used to determine @xmath341 .
we use the notations @xmath347 [ k1eks ] let @xmath348 .
\a ) if @xmath349 then @xmath350 \b ) if @xmath351 then @xmath352 let @xmath353 .
since @xmath354 , we have @xmath355 by lemma [ u - lemma ] , @xmath356 if ( [ bb ] ) is satisfied , then @xmath331 and so @xmath357 .
this proves a ) and the proof of b ) is similar .
let @xmath358 [ k1eksc ] if @xmath359 , then @xmath360 first we consider the upper bound .
let @xmath361 .
then @xmath362 for @xmath363 , that is , @xmath364 , i.e. @xmath365 . for @xmath366 we get @xmath367 also . in particular , ( [ bb ] )
is satisfied for all @xmath359 .
the upper bound therefore follows from lemma [ k1eks]a ) .
the proof of the lower bound is similar . for @xmath368 we can show it by direct computation .
some calculus shows that @xmath369 for all @xmath359 and @xmath370 for @xmath371 .
we skip the details .
the lower bound therefore follows from lemma [ k1eks]b ) . from corollary [ k1eksc ]
we immediately get the following result .
[ k0main1 ] we have @xmath372 when @xmath373 .
[ k0main2 ] for all @xmath359 , we have @xmath374 or @xmath375 in particular , if there is no integer between @xmath376 and @xmath377 , then @xmath378 since @xmath379 , ( [ k01 ] ) and corollary [ k1eksc ] imply that @xmath380 further , if there is no integer between @xmath376 and @xmath377 , then @xmath381 .
the difference @xmath382 is small , except for small @xmath46 .
hence , to have an integer between @xmath376 and @xmath377 , @xmath377 must be close to and above an integer .
if we denote this integer by @xmath383 , then @xmath46 must be close to @xmath384 .
we will make this statement more precise in the following lemma .
[ theta - n ] let @xmath385 be a positive integer .
\a ) if @xmath386 then @xmath387 .
\b ) if @xmath388 then @xmath389 . in this proof ,
we let @xmath46 be a positive real variable .
we note that @xmath377 and @xmath376 are still well defined .
moreover , simple calculus shows that @xmath390 and @xmath391 for all @xmath392 .
therefore , a ) is equivalent to @xmath393 and similarly for b ) .
proof of a ) .
let @xmath394 .
then @xmath395 , where @xmath396 and so @xmath397 hence @xmath398 if @xmath399 the inequality ( [ om - con ] ) is equivalent to @xmath400 which in turn is equivalent to @xmath401 this is satisfied for all @xmath402 . for @xmath403
, we can show a ) directly by numerical computation .
this completes the proof of a ) .
the proof of b ) is similar .
we give a sketch , leaving out some details .
let @xmath404 .
then @xmath395 , where @xmath405 and so @xmath406 we have @xmath407 we observe that the function @xmath408 is increasing with @xmath89 .
hence , @xmath409 if @xmath410 that is , @xmath411 this is satisfied for @xmath412 .
direct computation shows that b ) is true also for @xmath413 . combining lemmas [ theta - n]a ) and b )
, we see that if there is an integer between @xmath376 and @xmath377 , then this integer is @xmath383 for some integer @xmath385 , and @xmath414 where @xmath415 we have checked ( [ int ] ) for @xmath416 . for @xmath417 , there is no integer satisfying ( [ int ] ) . for @xmath418
, there actually is an integer @xmath419 satisfying ( [ int ] ) .
however , in these cases , direct computations show that @xmath420 .
we can therefore conclude that @xmath421 for @xmath422 .
this is , of course , far beyond what is needed for any practical application .
whether there are any @xmath423 such that there is an integer satisfying ( [ int ] ) remains an open question .
however , the length of the interval in ( [ int ] ) is @xmath424 and @xmath425 therefore , it is highly unlikely that there is an integer satisfying ( [ int ] ) for some @xmath423 .
based on this , we conjecture that @xmath421 for all @xmath359 .
if @xmath426 , we define @xmath427 to be the smallest integer and @xmath428 the largest integer such that : * @xmath429 , * ( [ per ] ) is satisfied for @xmath430 . in the next section ,
we give estimates for @xmath427 and @xmath428 . for @xmath431 ,
we have computed the values of @xmath1 in the range @xmath432 $ ] for which ( [ per ] ) is satisfied . for @xmath278 , this never happens ; and we have checked that @xmath57 is always proper for @xmath278 . for @xmath433 ,
the values of @xmath1 for which ( [ per ] ) is satisfied are given in table [ pek - n ] . since @xmath434
, we only have to consider @xmath435 when @xmath431 .
we see that , in general , for any given @xmath2 and @xmath46 the set of @xmath1 where ( [ per ] ) is satisfied consists of zero or more intervals .
typically , we have several intervals , except for small values of @xmath2 .
we describe @xmath436 , @xmath195 as an example to illustrate why this is the case .
we first give a small list of values in table [ tab17 ] .
@xmath437 we see that ( [ per ] ) is satisfied for @xmath438 , but not for @xmath439
. since ( [ per ] ) turns out to be satisfied for @xmath440 , we get @xmath441 . for @xmath1 in the range
@xmath442 $ ] we have @xmath443 .
since @xmath444 is decreasing with increasing @xmath1 , ( [ per ] ) is not satisfied for @xmath1 in the range @xmath445 $ ] .
however , we see that for @xmath446 , we have a jump in the value of @xmath447 compared to @xmath448 , and ( [ per ] ) is again satisfied for all @xmath1 in the range @xmath449 . for @xmath450 , ( [ per ] )
is again not satisfied .
[ ex16 ] an interesting example occurs when @xmath451 and @xmath452 ; @xmath453 is an isolated value of @xmath1 for which ( [ per ] ) is not satisfied .
we have @xmath454 we have @xmath455 and @xmath456 and so ( [ per ] ) is not satisfied
. however , @xmath457 and @xmath458 so the contribution from this term alone is sufficient to conclude that @xmath459 is not satisfactory after all .
this shows that if ( [ per ] ) is not satisfied , but the second lowest weight of @xmath24 is close the minimum weight @xmath149 , it may be a good idea to consider the contribution from this weight also . for @xmath460
we have checked if the codes @xmath24 and @xmath57 are proper when ( [ per ] ) is not satisfied .
it turns out that this is always the case for @xmath57 , but not for @xmath24 . as an illustration , in table [ tab : tab3 ]
we give the range of values @xmath1 for @xmath195 and @xmath461 , such that @xmath24 is proper .
@xmath462 & [ 331,384 ] \\
\hline 10 & [ 513,587 ] & [ 688,768 ] \\
\hline 11 & [ 1025,1124 ] & [ 1424,1536 ] \\
\hline 12 & [ 2049,2195 ] & [ 2904,3072 ] \\
\hline 13 & [ 4097,4298 ] & [ 5908,6144 ] \\
\hline 14 & [ 8193,8489 ] & [ 11937,12288 ] \\ \hline 15 & [ 16385,16798 ] & [ 24081,24576 ] \\
\hline 16 & [ 32769,33376 ] & [ 48422,49152 ] \\
\hline 17 & [ 65537,66388 ] & [ 97245,98304 ] \\
\hline 18 & [ 131073,132321 ] & [ 195096,196608 ] \\ \hline \end{array}\ ] ]
in table [ pek - n ] we see that for a given @xmath46 , an increasing fraction of the codes are ugly when @xmath2 increases .
define @xmath463 and @xmath464 by @xmath465 and @xmath466 clearly , @xmath467 .
let @xmath468 [ ny - con ] we have @xmath469 let @xmath470 , where @xmath471 .
by lemma [ dmin ] , @xmath472 , and by lemma [ admin ] , @xmath182 . by ( [ per ] ) we see that if @xmath473 then @xmath474 , and so @xmath24 is ugly .
we have @xmath475 by ( [ ent ] ) , @xmath476 hence @xmath477 by ( [ perapx ] ) , if @xmath478 then @xmath24 is ugly .
since ( [ par2 ] ) is equivalent to @xmath479 , and @xmath480 is an integer , that is @xmath481 , we can conclude that @xmath482 .
this proves the theorem . _
remark_. we see that @xmath483 is an upper bound on the number of @xmath1 in @xmath484 $ ] such that @xmath24 is satisfactory .
therefore , a main corollary of theorem [ ny - con ] is that , for any fixed @xmath46 , @xmath485 converges to 0 exponentially fast when @xmath2 increases . for @xmath486
, we have a similar result .
let @xmath487 [ ny - con2 ] we have @xmath488 the proof is similar to the proof of theorem [ ny - con ] .
let @xmath489 where @xmath490 .
from lemma [ dmin ] we get @xmath491 and so @xmath492 therefore , analogously to ( [ par2 ] ) we get @xmath493 is a sufficient condition for @xmath24 to be ugly .
since this is equivalent to @xmath494 , theorem [ ny - con2 ] follows .
[ cor ] let @xmath495 be the number of @xmath496 $ ] such that @xmath24 is satisfactory
. then @xmath497 the number @xmath498 of satisfactory codes for @xmath1 in the interval @xmath499 $ ] is at most those for @xmath500 $ ] plus those for @xmath501 $ ] .
the number of @xmath1 in the first interval is @xmath502 the number of @xmath1 in the second interval is @xmath503 we see that , for @xmath65 , we have @xmath504 hence @xmath505 and so @xmath506 @xmath507 when @xmath325 .
theorem [ cor ] shows that @xmath57 is ugly for most values of @xmath1 . on the other hand ,
@xmath57 is satisfactory for many values of @xmath1 as shown by corollary [ c - kange ] .
the matrix @xmath14 was defined by concatenating @xmath508 for @xmath509 .
we can generalize this by concatenating @xmath510 copies of @xmath511 , followed by @xmath512 copies of @xmath513 , @xmath514 copies of @xmath515 , etc .
for any sequence @xmath516 of positive integers .
most of the previous results carries over , with obvious modifications . for
now , we only consider the construction with @xmath517 for @xmath314 , and we write @xmath518 . as before , we use the notation @xmath24 for the codes generated by the first @xmath1 columns of the matrix . for large @xmath519 ,
these codes have low rate .
the dual codes will have very high rate and minimum distance 2 .
consider @xmath520 generated by @xmath521 , and let @xmath24 , where @xmath522,\ ] ] be the code generated by the matrix @xmath523 we see that we get a code @xmath24 for each @xmath277 . also , given @xmath1 , the values of @xmath519 and @xmath524 are uniquely determined by ( [ tn ] ) . from its definition
, we immediately get the following lemma .
[ lenging ] a ) the weight of the first row of @xmath52 is @xmath525 larger than the weight of the first row of @xmath521 .
\b ) for any other non - zero codeword in @xmath24 , the weight is @xmath526 larger than the weight of the corresponding codeword in @xmath521 .
in particular , we see that * the minimum distance of @xmath24 is @xmath526 larger than the minimum distance of @xmath521 . * for a non - zero codeword of @xmath24 of weight @xmath32 , either @xmath209 or there is a unique other codeword in the code of weight @xmath527 .
this last property was used to prove theorem [ s - proper ] .
therefore , this theorem can be directly generalized by a similar proof .
let @xmath528 @xmath529 note that @xmath530 and @xmath531 .
[ s - propek - t ] for @xmath532 and @xmath215 , if @xmath533 or @xmath534 then @xmath24 is proper .
the proof is similar to the proof of theorem [ s - proper ] and is omitted .
[ tp - kange ] a ) if @xmath239 and @xmath535 then @xmath24 is proper for all @xmath536.\ ] ] \b ) @xmath24 is proper for all @xmath537.\ ] ] similarly to theorem [ p - kange ] , we see that if @xmath538 where @xmath247 , then @xmath24 is proper for all @xmath539.\ ] ] we see that if @xmath540 , then @xmath541 and so ( [ tx ] ) is not satisfied
. however , if @xmath542 then @xmath543 and so @xmath544 for @xmath545 , that is , all @xmath239 . since ( [ txx ] ) is equivalent to @xmath546 , we get the theorem .
[ skr ] for @xmath547 there exists an integer @xmath548 such that @xmath24 is proper for all @xmath549 . for @xmath547 and @xmath550
, we get @xmath551 by theorem [ tp - kange]b ) , @xmath24 is proper for all @xmath552.\ ] ] this implies that @xmath24 is proper for all @xmath553 [ skr1 ] if @xmath547 , then @xmath24 is proper for all @xmath554 by theorem [ skr ] , @xmath24 is proper for all @xmath555 .
next , theorem [ tp - kange]b ) for @xmath556 shows that @xmath24 is proper for @xmath557.\ ] ] finally , theorem [ s - propek - t ] for @xmath556 , @xmath547 , and @xmath195 implies that @xmath24 is proper for @xmath558.\ ] ] it remains to show that @xmath559 we have @xmath560 and so , for @xmath547 , @xmath561 this proves ( [ tau - verdi ] ) . until recently , the best general result of this kind was ( * ? ? ? * theorem 2.64 ) : if @xmath562 and @xmath563 then there exists a proper @xmath0 $ ] code .
this bound was recently improved in @xcite to the following : if @xmath562 and @xmath564 then there exists a proper ( and self complementary ) @xmath0 $ ] code .
clearly , theorem [ skr ] above gives a further improvement and is now the best known such bound .
we can also find lower bounds on @xmath565 .
we consider @xmath24 for @xmath1 in the middle of the interval with @xmath195 , that is @xmath566 where @xmath567 was defined in ( [ nrm ] ) .
when we consider only the term in @xmath24 of lowest degree , we know that the case @xmath568 is the worst case ( cfr .
lemma [ var ] ) . moreover , this term of lowest degree is the dominating one in @xmath24 .
therefore , it is reasonable to consider these values of @xmath1 when we look for non - proper @xmath24 .
[ t - low ] for @xmath547 we have @xmath569 proof a ) .
we see that if @xmath570 is not proper , then by the definition of @xmath565 , @xmath571
. therefore , ( [ bound1 ] ) follows .
proof b ) .
again , consider @xmath24 for @xmath572 . then , by table [ tabnr ] and lemma [ lenging ] , @xmath573 since @xmath574 , ( [ per ] ) implies that the code is ugly if @xmath575 we have @xmath576 and so , by ( [ ent ] ) , @xmath577 hence , if @xmath578 that is @xmath579 then @xmath24 is ugly
. therefore , @xmath571 , and the theorem follows .
we now give a lemma that is useful for studying when @xmath24 codes in general are proper for a given @xmath2 .
[ small - n2 ] let @xmath580 and let @xmath32 be the weight of the first row of @xmath52 . if @xmath581 is increasing on @xmath3 $ ] , then @xmath582 is proper for all integers @xmath583 .
the weight of the first row of @xmath584 is @xmath585 .
the weight of any other non - zero codeword in @xmath582 is @xmath586 larger than the corresponding codeword in @xmath24 .
hence @xmath587 by assumption , @xmath581 is increasing on @xmath3 $ ] . since @xmath588 and @xmath589 are increasing on @xmath3 $ ]
, we can conclude that @xmath590 is increasing on @xmath3 $ ] , that is , @xmath582 is proper . for the use of this lemma , it is useful to observe that the conclusion of theorem [ s - propek - t ] can be improved : if ( [ tcon1 ] ) or ( [ tcon2 ] ) hold , then @xmath581 is increasing on @xmath3 $ ] .
the proof carries over immediately .
using lemma [ small - n2 ] and computations , we have determined @xmath565 for @xmath591 .
these values are given in table [ t0tab ] together with the lower and upper bounds on @xmath592 in theorems [ t - low ] and [ skr ] .
we have also included the bounds for @xmath593 .
we see that the upper bound is very loose , but @xmath565 equals the implicit lower bound @xmath594 for all @xmath595 .
we conjecture that this may be the case for all @xmath2 .
table [ t0tab ] shows that the explicit lower bound @xmath594 is also loose ( but substantially better than the upper bound ) and the ratio @xmath596 is increasing slowly with @xmath2 .
for @xmath597 the ratio is 1.375 , for @xmath451 it is 1.485 , and for @xmath598 it is 1.555 .
theorem [ tilde ] below shows that the ratio is always less than 2 .
the lower bound @xmath594 has the advantage that it is explicit and that it shows that @xmath565 grows exponentially with @xmath2 . in appendix 1
we prove the following theorem .
[ tilde ] let @xmath547 and @xmath599 .
let @xmath600 be the positive real number defined by @xmath601 where @xmath602 then @xmath603 combining theorem [ tilde]a ) and theorem [ tilde]b ) , we get the following corollary .
we have @xmath604 moreover , the first alternative is the most likely . we have included @xmath605 in table [ t0tab ] .
for the range of values we have computed , i.e. @xmath606 , we have @xmath607 .
if the conjecture that @xmath608 is true , then @xmath605 is a sharp upper bound on @xmath565 .
further , if the conjecture is true , then @xmath24 is proper for all @xmath609 , a substantially stronger result than what we have been able to show in corollary [ skr1 ]
. @xmath610 @xmath611 for @xmath547 , let @xmath612 be the number of @xmath613 $ ] such that @xmath24 is proper .
we have shown by direct computation and the use of lemma [ small - n2 ] that for @xmath614 , @xmath24 is proper for all @xmath615 $ ] .
hence , @xmath616 for @xmath614 . for @xmath617 we know that there are values of @xmath1 where @xmath24 is not proper . for given @xmath519 , @xmath2 , @xmath46 , let @xmath618 be the set of @xmath619 $ ] for which @xmath24 is not proper .
the set @xmath618 may be empty .
in particular , theorem [ tp - kange]a ) shows that @xmath620 if ( 38 ) is satisfied . on the other hand , table [ tab : tab3 ]
shows that @xmath621 for @xmath433 . by direct computation and the use of lemma [ small - n2 ]
, we have shown that the values given in table [ noprop ] are the only values @xmath580 for which @xmath24 is not proper .
the computations have been extended up to @xmath436 . in general , the values in @xmath618 are not necessarily consecutive . for example @xmath622\cup [ 93181,93184].\ ] ] this is similar to what we have in table [ pek - n ] , and the underlying reason is the same .
@xmath623 \\ \hline 10 & 1 & [ 588,687],\,[1127,1175],\,[1661,1667 ] \\ \hline 11 & 1 & [ 1125,1423 ] , \,[2200,2402],\,[3255,3397],\\ & & [ 4306,4396],\,[5353,5398],\,[6399,6401 ] \\ & 2 & [ 1661,1667 ] \\
\hline 12 & 1 & [ 2196,2903],\,[4301,4902],\,[6393,6893 ] , \\ & & [ 8500,8901],\ , [ 10582,10917],\ , [ 12661 , 12935 ] , \\ & & [ 14738 , 14955],\,[16813 , 16977],\,[18887 , 19000 ] , \\ & &
[ 20959 , 21024],\,[23030 , 23050 ] \\ & 2 & [ 3255,3397],\,[5353,5398 ] \\ \hline \end{array}\ ] ] the conjecture that @xmath624 can be reformulated as follows : if @xmath625 then @xmath626 .
in general , we conjecture that if @xmath627 , then @xmath628 . from table [ noprop ]
we get the explicit values of @xmath612 given in table [ theta - tab ] .
we have included in the table the lower bound given in theorem [ del ] below .
@xmath629 [ del ] when @xmath547 , @xmath24 is proper for @xmath630 of the values of @xmath613 $ ] .
the proof is given in appendix 2 .
_ comment .
_ we clearly have @xmath631 .
the discussion on @xmath565 above indicates that we may have @xmath632 .
if this is the case , then we have @xmath633 .
in this paper we have analyzed the codes @xmath24 and their duals @xmath57 ( which are shortened hamming codes ) to investigate if they are proper or satisfactory codes for error detection .
we have determined the weight distribution of the codes @xmath24 and computed the undetected error probability . for @xmath634 ,
the codes @xmath24 are proper for all @xmath65 . however , for @xmath2 greater than 8 , there are values of @xmath1 such that @xmath24 and @xmath57 are ugly ( not satisfactory ) for error detection
. for increasing @xmath2 , the percentage of such codes is increasing . on the other hand
, we have shown that the number values @xmath1 such that @xmath24 is proper grows exponentially with @xmath2 .
we have given a generalization of the construction which defines codes @xmath24 for all lengths greater than @xmath68 , and we have shown that @xmath24 is proper for all @xmath635 .
moreover , @xmath24 is proper for at least @xmath636 of the shorter lengths .
a plausible conjecture ( @xmath608 ) implies that @xmath24 probably is proper for all @xmath637 .
an open question for future work is the following : is it the case that if there is an @xmath638 $ ] for which @xmath24 is not proper , then @xmath639 is such an @xmath1 ? if the answer is yes ( which we believe it is ) , then this would in particular imply the conjecture referred to above .
further work may concern searching for modifications of the construction that will extend the range of lengths where the codes are proper or satisfactory .
in particular , one line of investigation could be to consider lengths less than @xmath68 by looking at the the duals of the best known codes of minimum distance 4 .
let @xmath640 the expression for @xmath641 that was used to determine the bound @xmath642 is easily obtained by combining table [ tabnr ] ( with @xmath195 ) and lemma [ lenging ] . since @xmath599 , we have : @xmath643^{4rt+r } \nonumber \\ & + 4\,p^{4rt+2r}(1-p)^{4rt } + p^{8rt}(1-p)^{2r}. \label{exp}\end{aligned}\ ] ] we note that @xmath641 is well defined for any positive real number @xmath519 . for large values of @xmath519 , @xmath641 is increasing on @xmath3 $ ] . for small values of @xmath519
, @xmath641 is first increasing , then decreasing , then again increasing .
there is a limiting @xmath519 such that , for this @xmath519 , there is a @xmath644 such that @xmath645 and @xmath646 for all other @xmath29 in @xmath647 . in particular
, this implies that @xmath648 . in principle , the two equations @xmath645 and @xmath648 can be used to determine @xmath644 and @xmath649
. however , the equations are complicated , and we will consider an approximation which is easier to handle .
we remark at this point that for @xmath519 close to @xmath650 , @xmath648 is not possible . in this appendix
we often drop @xmath2 from @xmath565 and write just @xmath651 when the value of @xmath2 should be clear from the context .
similarly we write @xmath652 for @xmath653 , @xmath654 for @xmath655 , etc . in @xmath641 ,
the last three terms are increasing on @xmath3 $ ] whereas the first term is increasing on @xmath656 $ ] and decreasing on @xmath657 $ ] .
for all @xmath658 $ ] , the first two terms are dominating .
therefore , we first consider the sum of these two terms : @xmath659^{4rt+r}\ ] ] and determine the @xmath660 and @xmath661 such that @xmath662 .
we expect @xmath660 to be a good approximation to @xmath654 ( and @xmath661 to be a good approximation to @xmath644 ) .
the remaining two terms in @xmath663 are much smaller .
moreover , they are increasing on @xmath3 $ ] .
therefore , @xmath664 .
in particular , @xmath665 .
this proves theorem [ tilde]a ) .
we have @xmath666^{4rt+r-1}(1 - 2p ) \nonumber \\ & = 4r\,p^{4rt-1}(1-p)^{4rt+r-1 } \nonumber \\ & \quad \cdot \bigl\{[2t-(4t+1)p](1-p)^r \nonumber \\ & \qquad \quad + ( 4r-2)(4t+1)(1 - 2p ) p^r \bigr\}.\label{fdir}\end{aligned}\ ] ] _ remark .
_ we see that @xmath667 for all @xmath519 .
hence , the equation @xmath668 will , in addition to @xmath661 , have at least one more solution in @xmath647 .
this second solution will be closer to @xmath669 .
however , it does not reflect a property of @xmath663 , but only the approximation @xmath670 . since @xmath671 and @xmath672 , we have @xmath673 where @xmath674 similarly , @xmath675 where @xmath676,\end{aligned}\ ] ] and so @xmath677 combining ( [ equation1 ] ) and ( [ equation2 ] ) , we get @xmath678 since @xmath679 , we have @xmath680 , and so @xmath681 solving this for @xmath29 , we get two solutions @xmath682 @xmath683 where @xmath684 we see that if @xmath685 , then the roots are not real .
this reflects the fact that @xmath686 is not possible in this case .
the smaller of the two roots is the @xmath661 we are looking for , the larger @xmath687 occurs because we have neglected the two smallest terms in ( [ exp ] ) as explained above .
therefore , it is not relevant for our analysis of @xmath663 . since @xmath688 , ( [ fdir ] ) implies @xmath689(1-p_1)^r+(4r-2)(4{\vartheta}+1)(1 - 2p_1)p_1^r=0.\ ] ] substituting the value of @xmath661 into ( [ llikk ] ) and simplifying , we get ( [ tildef ] ) .
we can not find a closed expression for @xmath660 , but , for a given @xmath2 , we can determine the value numerically .
we note , however , that @xmath690 ( which is the quantity we want ) actually is the least integer @xmath519 such that @xmath691 where @xmath692 and this observation simplifies the numeric determination of @xmath690 since we do not have to solve the equation ( [ tildef ] ) , but only search for @xmath690 . to prove theorem [ tilde]b ) , we first give a couple of lemmas . [ kat1 ] if @xmath693 , then @xmath694 the function @xmath695 is decreasing when @xmath89 is increasing .
we have @xmath696 and so @xmath697 .
hence , @xmath698 } > 1-\frac{1}{2t}.\end{aligned}\ ] ] [ kat2 ] if @xmath693 , then @xmath699 we have @xmath700 from lemmas [ kat1 ] and [ kat2 ] , we see that if @xmath701 then ( [ lik2 ] ) is satisfied and so @xmath702 .
taking logarithms of ( [ ll2 ] ) , we get the equivalent expression @xmath703 since @xmath704 for @xmath705 , we see that if @xmath706 then ( [ ll2 ] ) is satisfied , and so @xmath702 . since @xmath599
, we have @xmath707 . solving ( [ ll4 ] ) for @xmath519 , we get the following relation : @xmath708 in the proof of lemma [ kat2 ] , we used that @xmath709 . however , using ( [ tilweak2 ] ) we get a better bound on @xmath710 and hence a stronger version of lemma [ kat2 ] and a better bound on @xmath652 .
[ kat3 ] if @xmath711 then @xmath712 by ( [ nyb ] ) @xmath713 for @xmath714 .
hence , @xmath715 therefore , if @xmath716 then @xmath702 .
taking logarithms and solving as above , we get theorem [ tilde]b ) exactly as we obtained ( [ tilweak2 ] ) from ( [ ll2 ] ) . for @xmath717
, we can show that theorem [ tilde]b ) is true by direct computation . to prove theorem [ tilde]c ) , we first give another lemma .
[ kat1a ] @xmath718 using ( [ deltaskr ] ) , we get @xmath719 } { 8r\vartheta + 3r+ r[1 - 4/(k-1 ) ] } \\ & = 1-\frac{k-3}{2(k-1)\vartheta + ( k-2)}.\end{aligned}\ ] ] this is lemma [ kat1a]a ) .
moreover , this implies that @xmath720 by theorem [ tilde]b ) @xmath721 for @xmath722 , and so @xmath723 .
direct computations show that @xmath724 also for @xmath725 .
hence lemma [ kat1a]b ) is proved . to prove theorem [ tilde]c ) , let @xmath726 and @xmath727 . we get @xmath728p^{4r(t-\varepsilon ) } ( 1-p)^{4r(t-\varepsilon ) + 2r-1 } \\ & \quad + ( 16r-8)[4r(t-\varepsilon ) + r ] \\ & \qquad \cdot [ p(1-p)]^{4r(t-\varepsilon ) + r-1}(1 - 2p ) \\ & = f'_{t}(p)-8r\varepsilon p^{4r(t-\varepsilon ) -1}(1-p)^{4r(t-\varepsilon ) + 2r } \\ &
\quad + 8r\varepsilon p^{4r(t-\varepsilon ) } ( 1-p)^{4r(t-\varepsilon ) + 2r-1 } \\ & \quad -4r\varepsilon ( 16r-8)[p(1-p)]^{4r(t-\varepsilon ) + r-1}(1 - 2p).\end{aligned}\ ] ] since @xmath729 and @xmath730 , we get @xmath731.\end{aligned}\ ] ] let @xmath732 then @xmath733 . we get @xmath734\pi^{2r } -16r({\vartheta}-\varepsilon)\pi^{2r+1 } \\ & \quad \quad + 8r({\vartheta}-\varepsilon)\pi^{4r({\vartheta}-\varepsilon ) } -2r\pi^{4r({\vartheta}-\varepsilon)+1}\bigr\ } \\ & < 8r(1-p_1)^{8r({\vartheta}-\varepsilon ) + 2r-1 } \pi^{4r({\vartheta}-\varepsilon)-1 } \\ & \quad\cdot \bigl\ { \bigl [ 2{\vartheta}(1-\pi)+1\bigr]\pi^{2r } + \vartheta \pi^{4r({\vartheta}-\varepsilon)}\bigr\}.\end{aligned}\ ] ] clearly
, @xmath735 if @xmath736\pi^{2r } \le \varepsilon(1-\pi),\ ] ] and @xmath737 equation ( [ pico1 ] ) is equivalent to @xmath738 we choose the @xmath739 which gives equality in ( [ pico11 ] ) , that is @xmath740 equation ( [ pico2 ] ) is equivalent to @xmath741 by ( [ vardef ] ) , @xmath742 further , by theorem [ tilde]b ) , @xmath743 and , finally , @xmath744 and so @xmath745 hence , ( [ pico2 ] ) is also satisfied
. therefore @xmath746 where @xmath739 is given by ( [ vardef ] ) . by lemma [ kat1a ] , @xmath747 combining ( [ eqvar ] ) and ( [ epp ] ) we get theorem [ tilde]c ) for @xmath714 .
direct computations show that it is true also for @xmath717 .
this completes the proof of theorem [ tilde ] .
let @xmath748 be the number of @xmath1 in @xmath749\ ] ] such that @xmath24 is proper
. clearly , @xmath750 [ tau - bounds ] let @xmath547 and @xmath65 .
\a ) if @xmath751 and @xmath752 , then @xmath753 \b ) if @xmath751 and @xmath754 , then @xmath755 \c ) if @xmath756 and @xmath532 , then @xmath755 by theorem [ s - propek - t ] , @xmath757 in the proof of theorem [ tp - kange ] we showed that if @xmath758 then @xmath759 this proves b ) and c ) .
now , consider @xmath751 and @xmath752 .
then @xmath760 and so @xmath761 similarly , @xmath762 combining these two inequalities with ( [ tet ] ) , a ) follows .
[ sum3 ] for @xmath547 we have @xmath763 first we see that @xmath764 hence @xmath765 similarly , @xmath766 the lemma follows from these results and lemma [ tau - bounds]a ) .
[ sum1 ] for @xmath547 we have @xmath767 and @xmath768 the result follows directly from lemmas [ tau - bounds]c ) and [ tau - bounds]b ) respectively
. we can now combine these results into a proof of theorem [ del ] . from ( [ tetsum ] ) and lemmas [ tau - bounds][sum1 ]
, we get @xmath769 theorem [ del ] follows from this expression .
the authors are grateful to mario blaum for pointing out reference @xcite .
the construction of @xmath52 is also essentially due to him .
a. a. davydov , l. n. kaplan , yu .
b. smerkis , and g. l. tauglikh , `` optimization of shortened hamming codes '' , _ problems of inform .
4 , pp . 261267 , oct .- dec .
1981 ( transl . from russian ) . | binary linear @xmath0 $ ] codes that are proper for error detection are known for many combinations of @xmath1 and @xmath2 . for the remaining combinations , existence of proper codes
is conjectured . in this paper , a particular class of @xmath0 $ ] codes
is studied in detail . in particular
, it is shown that these codes are proper for many combinations of @xmath1 and @xmath2 which were previously unsettled .
error detection , proper codes , satisfactory codes , simplex codes , punctured codes , ugly codes . |
two - dimensional ( @xmath7 ) lattice spin models , besides being interesting on their own right as models of ferromagnets can be viewed as mathematically well defined scheme of the nonperturbative regularization of quantum continuum field theories like @xmath0 principal chiral model or @xmath8 nonlinear sigma model . being a discrete theory lattice models can be simulated on computers by monte - carlo method , and this is nowadays the most important tool of obtaining physical results near the continuum limit . among major analytical methods one could mention the strong coupling expansion and the perturbation theory ( pt ) .
the pt is essentially the only analytical tool which provides systematic expansion of different physical quantities in the region of weak coupling , i.e. in the region relevant for the construction of the continuum limit .
however , at the best pt is only applicable for studying short - distance quantities and the most important and interesting phenomena , like the mass gap generation can not be described in its frameworks . when attempting to go beyond pt in the region of weak bare coupling , one runs into various mathematical problems , the most important being an absence of any analytical control over exponentially small contributions and , hence the absence of any reliable method to study long - distance physics where nonperturbative effects dominate even if they are exponentially suppressed .
these facts impelled people to look for different , though equivalent representations which would allow to study the long - distance physics .
one of such popular and promising representations is known as a dual formulation and is based on a certain non - classical change of variables .
this formulation deals with dual lattice and appears to be very fruitfull for abelian models @xcite .
namely , within such formulation 1 ) have been obtained various important analytical results on the long - distance dynamics of abelian spin fields and 2 ) the topological structure of the vacuum is transparent and can be more easily studied . in the context relevant to this paper
we would like to mention the dual of the abelian @xmath4 model @xcite which has been used to prove the existence of a soft phase at low temperatures with power - like decay of the correlation function in two - dimensional @xmath4 model @xcite . in this case
the dual of the @xmath4 model is a local theory for certain discrete variables .
the conventional dual transformations @xcite for non - abelian models are perfectly well defined in mathematical terms and also lead to a local dual theory for integers which label irreducible representations of the symmetry group
. nevertheless , these transformations are not complete if one compares to the abelian case .
first of all , the resulting dual variables are not independent but are subject to some constraints known as triangular conditions , and as such they can not really be associated with elements of a dual lattice ( of course , introduction of the dual lattice is not the crucial point but rather matter of convenience ; e.g. , in the abelian case one can work with dual variables on the original lattice ) .
it is to say , however that it is not really clear what are true independent degrees of freedom of the duals of non - abelian models .
secondly , such formulation is so mathematically involved that today it is even far from obvious if it can be useful for any kind of the study of the model , except for the strong coupling expansion . in particular , the positivity of the boltzmann weight remains unclear ( see discussion in @xcite ) .
even the precise definition of what can be termed the dual boltzmann weight is not really obvious , so the possibility of numerical simulations is at present rather uncertain .
most importantly , however is that it is not clear how one could proceed in an analytical study of the model at low - temperatures . ] . on the other hand
, there exists representation of @xmath0 and @xmath1 spin models in terms of link variables @xcite , and this representation can be formulated directly on the dual lattice .
it is a first goal of the present paper to use the link representation for the derivation of an exact dual formulation of non - abelian spin models .
here we give two such formulations which appear to be quite different from the conventional formulation mentioned above . in our opinion
the most essential advantages of our formulations are that 1 ) one of them can be definitely used for the monte - carlo simulations in @xmath7 and 2 ) they are much more suitable for an analytical investigation of the model in the low - temperature region .
the second fact follows from the properties of the link formulation and we refer for the details of why it is so to our papers @xcite . in the last of those papers we have already presented a dual of @xmath7 @xmath5 spin model and proposed approximate representation for the dual partition function at low temperatures . another important feature of one of our formulations is that it carries a close analogy with the abelian @xmath9 model .
for instance , for @xmath7 @xmath5 model there are 3 independent degrees of freedom per ( dual ) site . at low temperatures it possesses an analog of the vortex spin - wave representation , etc .
the deficiency of our approach is that it can be straightforwardly applied only to models in which spins are elements of the lie group .
we do not know any obvious generalization of link formulation to , say , @xmath8 nonlinear sigma models .
low - temperature properties of @xmath7 non - abelian models are crucial for construction of their continuum limit .
it is commonly recognized that models possess no phase transition , correlation function has exponential decay at any coupling and models are asymptotically free . despite being more
than twenty five years old this expectation has not been proven rigorously . on the contrary
, certain percolation theory arguments suggest that all non - abelian models have soft low - temperature phase with power - like decay of the correlation function @xcite .
it is thus another important motivation of the present investigation , namely to get deeper insight into the nature of the mass gap in non - abelian spin models .
for example , in many papers devoted to @xmath7 non - abelian models it is written that `` there is a nonperturbative mass gap generation at arbitrarily small couplings '' .
it is not really clear , however what is precise meaning of this `` nonperturbative generation '' .
it can not be a simple consequence of the link decorrelation which happens in @xmath10 models .
then , one could ask if this `` nonperturbative generation '' follows from the existence of some non - trivial background of defects like vortices of the @xmath4 model or is due to the strong but smooth disorder of non - abelian spins , e.g. like center vortices @xcite .
as is well known , the dual formulation of abelian models have been extremely useful in clarifying all these important physical problems @xcite .
we think that the dual formulation given here is a good starting point for obtaining reliable analytical results in the low - temperature limit of non - abelian models .
this is our next goal to develop a technique within dual formulation of non - abelian models which would allow to investigate them in the limit of the weak coupling . in this context
we study here the following two approaches on the simplest example of @xmath5 spin model * we derive an asymptotic expansion of the dual boltzmann weight when @xmath11 uniformly valid in fluctuations of the dual variables .
the same procedure is also done for the two - point correlation function . as will be seen , at low temperatures
the boltzmann weight converges to a certain gaussian ensemble the fact which ensures a possibility to define an analog of the vortex spin - wave representation for the partition and correlation functions in the semiclassical limit .
this step provides a major simplification of the model as @xmath11 but even in this case it can not be yet solved exactly .
the formulation obtained permits very simple calculation of the leading perturbative contribution to the correlation function .
this contribution results in the power - like decay of the correlation function . *
the second approach consists in replacing @xmath5 matrix elements by @xmath6 ones in the vicinity of the identity element of @xmath5
. we shall give a proof that such replacement is valid at low temperatures , compute the asymptotics of the dual boltzmann weight and present simple numerical evidence that practically for all configurations this way of calculations produces very reasonable approximation for the original boltzmann weight .
this paper is organized as follows . in the section 2
we review briefly the conventional dual transformations .
after introduction of the link formulation we develop original approach to the dual transformations and present two forms of the duals of the principal chiral model . in the section 3
we investigate the low - temperature properties of the dual boltzmann weights of @xmath5 model in two dimensions along the lines described above .
some technical details of this investigation are given in the appendix . in the section 4
we summarize our results and outline some perspectives for future investigations .
also , we calculate here the leading perturbative contribution to the correlation function .
we work on a @xmath2-dimensional hypercubic lattice @xmath12 with lattice spacing @xmath13 and a linear extension @xmath14 .
we impose either free or periodic boundary conditions ( bc ) .
let @xmath15 ; @xmath16 and @xmath17 denote the haar measure on @xmath18 .
@xmath19 and @xmath20 will denote the character and dimension of irreducible representation @xmath21 of @xmath18 , correspondingly .
we treat models with local interactions , i.e. interactions between nearest neighbours .
let @xmath22 $ ] be a real invariant function on @xmath18 such that @xmath23 \mid \
\leq h(i ) \label{huleq}\ ] ] for all @xmath24 and coefficients of the character expansion of @xmath25 @xmath26 \ = \ \int du
\ \exp \left ( \beta h[u ]
\right ) \chi_r(u ) \label{charexpdef}\ ] ] exist .
the partition function ( pf ) of the principal chiral model with the symmetry group @xmath27 is defined as @xmath28 \right ] \ .
\label{pfdef}\ ] ] conventional dual transformations can be defined as a sequence of transformations consisting of the following steps : 1 .
fourier expansion of the boltzmann weight @xmath29 .
this is an essentially character expansion on the group .
2 . exact integration over original degrees of freedom . as a result one obtains a set of constraints on the summation variables ( which label representations of the group @xmath18 and matrix elements of @xmath30 ) in the character expansion .
solution of the constraints in terms of new dual variables .
performing the first step one finds for the pf @xmath31 \right ] \prod_{x\in\lambda } q(x ) \ , \label{pfst1}\ ] ] where @xmath32 denotes links of the lattice and @xmath33 is a group integral
. let @xmath34 $ ] be a matrix element of @xmath24 in representation @xmath35 .
then @xmath33 can be written as @xmath36 d_{r(x - e_n , e_n)}^{m(x - e_n , e_n)k(x - e_n , e_n)}[u^{\dagger } ] \right ] \ .
\label{qxdef}\ ] ] the second step in the most general settings ( for arbitrary graph and for all @xmath18 ) has been accomplished in @xcite . through the rest of this subsection
we consider only the simplest cases of @xmath37 and @xmath38 in two dimensions .
these examples already show a marked difference between abelian and non - abelian models , on the one hand . on the other hand , we want to have relatively simple formulae which can be compared with our formulations . for the @xmath9 model
we have @xmath39 \right ) \ , \label{qxu1}\ ] ] where @xmath40 is the kronecker delta . in @xmath0 case
the result for @xmath33 in @xcite is expressed in terms of the haar intertwiners .
we re - state this result in more familiar form expanding @xmath33 in the clebsch - gordan ( cg ) series @xmath41 where @xmath42 , @xmath43 , @xmath44 , @xmath45 and similar notations are used for @xmath46 and @xmath47 .
the third step can be readily accomplished for the @xmath9 model .
we skip all the details of the derivation which are well known and can be found , for example in @xcite .
the result for the pf reads @xmath48 where the product runs over all links of the dual lattice @xmath49 and the dual boltzmann weight is given by @xmath50 \ , \label{dualbwxy}\ ] ] which in the case of the standard choice @xmath51 leads to @xmath52 with @xmath53 - the modified bessel function .
unfortunately , situation is much more difficult for non - abelian case . @xmath33 for @xmath5
contains not just one but 4 conditions : 2 kronecker delta s which constrain magnetic numbers @xmath46 and @xmath47 @xmath54 and 2 triangular conditions on representations @xmath55 we are not aware of any attempts in the literature to resolve these constraints simultaneously for all lattice sites . of course , this does not cause much problems for the strong coupling expansion where only lowest values of @xmath56 contribute .
but situation becomes hopeless in the low temperature region where probably all @xmath56 become relevant .
eventually , this reduces to the problem of finding the appropriate summation technique when @xmath11 which we could not solve .
moreover , @xmath33 as defined in ( [ qxsu2 ] ) is not strictly positive quantity : on a number of configurations it is negative .
we do not know how to perform resummation over the magnetic numbers to get positive boltzmann weight .
most probably , only complete summation over all magnetic numbers will produce positive quantity .
one of the possible ways of how to proceed further is to go over to the triangular lattices identifying representations with the length of the sides of triangles and magnetic numbers with the orientation of the triangles , and we have undertaken some attempts to go along this line .
detailed description of this procedure is however beyond the scope of the present paper .
we turn now to a different approach to duality transformations based on so - called link representation for partition and correlation functions .
that is , the three steps formulated in the beginning of the previous subsection are replaced by the following ones 1 .
change of variables in the partition function ( [ pfdef ] ) @xmath57 integration over original degrees of freedom .
this integration generates a set of constraints on the link matrices @xmath58 known as the bianchi identities . 2 . implementation of constraints into the partition function by making use of the invariant delta - function on the group .
this step introduces new degrees of freedom which can be associated with plaquettes of the original lattice and which label the irreducible representations of the group @xmath18 . at this step one
can go over to the dual lattice so that the new variables belong to the elements of the dual lattice .
integration over link variables .
these two steps give already some dual version of the principal chiral model .
3 . before integration over link variables one can decouple summations over matrix indices employing certain orthogonality relations for the cg coefficients ( or for some other objects ) of the group and multiplying the group matrices entering
bianchi identity in some specially ordered way . though details depend on the dimension of the lattice
the result is that only independent dual variables remain in the theory and their interaction is confined to a subset of all hypercubes of the dual lattice .
the first step of this program has been accomplished twenty years ago in @xcite .
we therefore restrict ourselves only to the description of the result .
the pf ( [ pfdef ] ) can be exactly reformulated in terms of link variables as ( with periodic bc ) @xmath59
\right ] j[v ] \ , \label{lpf}\ ] ] with jacobian @xmath60 $ ] given by @xmath61 = \prod_{p\in\lambda } j[v(p ) ] \
\prod_{n=1}^d j[h(n ) ] \ , \label{jacobdef}\ ] ] where @xmath62 $ ] and @xmath63 $ ] are given by @xmath64 \ = \
\sum_{\ { r \ } } d(r ) \chi_r \left ( \prod_{l\in p}v(l ) \right ) \ , \label{jacob}\ ] ] @xmath65 \ = \ \sum_{\ { r \ } } d(r ) \chi_r \left ( \prod_{l_n=1}^lv(l_n )
\right ) \ , \ l=(x , e_n ) \ . \label{holonconstr}\ ] ] @xmath66 is a product over all plaquettes of lattice @xmath67 , the sum over @xmath21 is sum over all representations of @xmath68 , @xmath69 is the dimension of @xmath35-th representation .
the group character @xmath70 depends on a product of the link matrices @xmath71 along a closed path ( plaquette in our case ) : @xmath72 the expression @xmath73 is the invariant group delta - function which reflects the fact that the product of @xmath74 around plaquette equals @xmath75 @xmath76 this constraint is called local bianchi identity while the constraint on holonomy operators @xmath77 is usually called global bianchi identity .
the only solution of two constraints ( [ bloc ] ) and ( [ bglob ] ) is a pure gauge ( [ linkfunc ] ) which restores the equivalence of the standard and the link formulations . on the lattice with free bc the global constraint
should be omitted .
furthermore , in the thermodynamic limit ( tl ) the contribution from the constraints on the holonomy operators vanishes very fast . in the perturbation theory in @xmath7 it vanishes like inverse powers of @xmath14 @xcite , and similar behaviour
is expected for @xmath78 .
since we are eventually interested in the tl we neglect these global constraints ( in fact , the presence of global constraints does not cause any principal problems as will be seen from the procedure described below but their absence makes the technical details somewhat simpler ) .
thus , for both free and periodic bc we work with the pf @xmath59
\right ] \prod_{p\in\lambda } j[v(p ) ] \ . \label{linkpffin}\ ] ] the two - point correlation function in representation @xmath79 takes the form @xmath80 where @xmath81 is some path connecting points @xmath82 and @xmath83 and @xmath84 if along the path the link @xmath85 points positive direction and @xmath86 , otherwise . for more details on the link formulation
we refer the reader to our paper @xcite , where we have developed a weak coupling expansion for @xmath0 spin models using the link representation . here
we are going to accomplish the step 2 described in the previous subsection .
the idea that the true dual variables in the link formulation are the fourier conjugate to the local bianchi identity has been proposed also in @xcite but the formulae have been explicitely written only for abelian @xmath9 model in @xmath7 . we thus want to reformulate the model ( [ linkpffin ] ) on the dual lattice only in terms of discrete variables that are in our case representations @xmath87 and magnetic quantum numbers @xmath88 .
we shall use the following conventions for the dual lattice .
@xmath89 will denote an object dual to site @xmath82 , i.e. @xmath2-cell of the dual lattice , @xmath90 - @xmath91-cell of the dual lattice , @xmath92 - @xmath3-cell of the dual lattice and so on . by definition , object dual to k - th cell is @xmath93-cell .
as follows from ( [ linkpffin ] ) and from the definition of the character which enters the group delta - function in ( [ jacob ] ) @xmath94 on the dual lattice the pf may be written as @xmath95 \sum_{m_i(\omega_p)}^{d[r(\omega_p ) ] } \right ] \prod_{\omega_l } \xi_0(\omega_l ) \ . \label{pfsundual}\ ] ] due to the trace there are 4 variables @xmath46 at each @xmath92 , thus @xmath96 .
the dual weight @xmath97 is given by the following one - link integral @xmath98 } \
\prod_{\nu = 1}^{d-1 } \left [ v_{r(\omega_p)}^{m_im_{i+1 } } \ v_{r(\omega_p^{\prime})}^{\dagger \
n_in_{i+1 } } \right ] _ { \nu } \ , \label{dualweight}\ ] ] where @xmath99 is a matrix element of @xmath35-th representation .
@xmath3-cells @xmath92 and @xmath100 share @xmath91-cell @xmath90 .
@xmath101 runs over @xmath91 such pairs .
similar form for the two - point correlation function in the representation @xmath79 reads @xmath102 \sum_{m_i(\omega_p)}^{d[r(\omega_p ) ] } \right ] \nonumber \\ \times \prod_{\omega_l\in \overset{\star}c_{xy } } \xi^{s_is_{i+1}}_j(\omega_l ) \ \prod_{\omega_l\notin \overset{\star}c_{xy } } \xi_0(\omega_l ) \ , \label{2func}\end{aligned}\ ] ] where @xmath103 and the link integral on @xmath104 is @xmath105 } \
\prod_{\nu = 1}^{d-1 } \left [ v_{r(\omega_p)}^{m_im_{i+1 } } \ v_{r(\omega_p^{\prime})}^{\dagger \
n_in_{i+1 } } \right ] _ { \nu } \
v_j^{s_is_{i+1 } } \ .
\label{2funcdualw}\ ] ] here , @xmath106 is a path dual to the path @xmath81 between points @xmath82 , @xmath83 on the original lattice . to vizualize last formulae
let us give expression for the pf in two dimensions with @xmath38 .
in @xmath7 we have @xmath107 , @xmath108 and @xmath109 . placing as usually dual sites in the centers of original plaquettes we obtain @xmath110 \prod_{l\in\overset{\star}\lambda } \xi_0(l ) \ , \label{pflsu2}\ ] ] where the dual weight @xmath111 becomes @xmath112 } \
v_{r(x)}^{m_i(x ) \ m_{i+1}(x ) } \ v_{r(x+e_n)}^{\dagger \
t_i(x+e_n ) \
t_{i+1}(x+e_n ) } \ . \label{linkint}\ ] ] it is very easy to make an integration in ( [ linkint ] ) expanding the result into the cg series .
one then finds the following representation for the dual weight of @xmath5 model @xmath113 \
c_{r_1 m_1 \
j k}^{r_2 t_2 } \ c_{r_1 m_2 \
j k}^{r_2 t_1 } \ , \label{xio2dexact}\ ] ] where we have denoted @xmath114 , @xmath115 , @xmath116 and so on . for the standard choice @xmath117=\chi_{1/2}(v)$ ] the coefficients @xmath118 $ ] can be computed explicitely @xmath119 \
\frac{2j+1}{\beta } i_{2j+1}(2\beta ) \ . \label{cj2d}\ ] ] for the correlation function in eq.([2func ] ) using the cg expansion one gets @xmath120 last expressions can be compared with the dual of the @xmath9 model given in ( [ dualxy ] ) and with the standard dual formulation of @xmath5 model described by formulae ( [ pfst1 ] ) and ( [ qxsu2 ] ) .
certainly , the dual weight for @xmath5 model @xmath121 as given in ( [ linkint])-([cj2d ] ) is very close to what we have for the dual weight in @xmath9 model : 1 ) dual variables , i.e. representations @xmath122 and magnetic numbers @xmath123 , live on sites of the dual lattice ; 2 ) the full gibbs measure is factorized into product of dual weights @xmath121 over links of the dual lattice . unlike the standard dual formulation
there is no complicated triangular conditions on representations : triangular constraint on @xmath124 in ( [ xio2dexact ] ) is only one - link problem which does not lead to any significant complications .
nevertheless , there is also a difference from the abelian case . as follows from the properties of the cg coefficients we have one constraint on the magnetic numbers on every link , namely summation over @xmath125 in ( [ xio2dexact ] )
produces the following condition @xmath126 this shows that not all dual variables are in fact independent , and this is the main difference from the @xmath9 model .
there is no simple way to make a change of variables such that one gets only independent variables and keeps the locality of interaction . in the next subsection we present a different way of the construction of the dual theory from the link formulation in such a way that the dual weight depends only on independent dual variables .
we finish this subsection with the brief description of some important features of @xmath121 : * as follows from the properties of the coefficients of the expansion @xmath118 $ ] the series in @xmath124 in ( [ xio2dexact ] ) gives directly the strong coupling expansion of the model written in closed and compact form .
much less trivial task is to get weak coupling expansion for @xmath121 since all @xmath124 in the series ( [ xio2dexact ] ) become relevant .
* on all configurations @xmath127 @xmath128 is strictly positive @xmath129 .
though we could not prove it rigorously this claim is supported by the following facts : 1 ) the first term in the strong coupling expansion is strictly positive , thus at sufficiently small @xmath130 where the series converges very fast @xmath128 is positive ; 2 ) the leading term of the asymptotic expansion of @xmath128 at large @xmath130 is strictly positive on all configurations ; 3 ) numerical computations of @xmath128 on a number of configurations and in a wide region of @xmath130 also support this conclusion ( we have checked this statement using mathematica on more than @xmath131 configurations ) . if @xmath129 on all configurations this gives a chance for a numerical monte - carlo simulations of the dual model . * the dominant contribution to @xmath128 at large @xmath130 comes from diagonal components of the rotation matrices , non - diagonal contribution is suppressed roughly as @xmath132^{-1}$ ] .
this is , of course a consequence of the fact that when @xmath11 the link matrix performs only small fluctuations around unity . in turn
, this property gives a possibility to compute low - temperature asymptotic expansion of @xmath128 .
this will be subject of the next section .
the main idea of the following approach is to introduce the independent dual variables from the beginning , namely to find a representation for the group invariant delta - function in terms of quantities which would be free of any conditions .
we explain the realization of this idea for @xmath7 @xmath5 model and then show how it can be extended to other groups and to higher dimensions .
we start with the representation for @xmath5 matrix elements in which the dependence on the magnetic numbers enters only through the clebsch - gordan coefficients @xmath133 @xmath134 where @xmath135 is expressed through the spherical harmonics @xmath136 and generalized characters @xmath137 of @xmath5 @xmath138 generalized character of rank @xmath139 in representation @xmath35 can be defined through the relation @xmath140 we remind that the invariant measure in this parameterization is @xmath141 and the fundamental trace becomes @xmath142 the basic formula which we need is the following representation for @xmath5 characters @xmath143 @xmath144 where @xmath145 is the legendre polynomial and @xmath146 this representation can be easily proved if one uses orthogonality relations for the spherical harmonics and then the addition theorem for the generalized characters . as the next step , we divide all sites of the dual lattice into sets of even and odd sites .
site @xmath147 is even if both @xmath148 and @xmath149 are even or odd , and is odd if one of @xmath150 is odd .
thus , all even sites are surrounded by odd sites and vice - versa .
using cyclic properties of the trace we can couple the four matrices entering local bianchi identities in two different ways , namely @xmath151 \ , \label{chareven}\ ] ] if the site @xmath82 is even and @xmath152 \ , \label{charodd}\ ] ] if the site @xmath82 is odd .
we introduced here obvious notations @xmath153 with help of ( [ su2char ] ) and ( [ chareven ] ) , ( [ charodd ] ) the jacobian @xmath60 $ ] defined in ( [ jacobdef ] ) and ( [ jacob ] ) can be written on the dual lattice as @xmath61 = \prod_{x , \mbox{even } } j[v(x ) ] \
\prod_{x , \mbox{odd } } j[v(x ) ] \ , \label{jacobdual}\ ] ] where in even sites @xmath154 while for odd sites @xmath155 a set of variables @xmath156 is chosen to be dual variables at site @xmath82 .
finally , we decompose all plaquettes of the dual lattice into sets of even and odd plaquettes in such a way that all even plaquettes are surrounded by odd ones and vice - versa . due to the special coupling of matrices as described above , the integration over link matrices is now coupled only within every , for example even plaquette .
this is shown in fig.[matrcoupl ] .
substituting ( [ su2deltaeven ] ) and ( [ su2deltaodd ] ) in ( [ lpf ] ) , re - denoting notations for links @xmath157 uniformly for all even plaquettes as shown in fig.[matrcoupl ] we obtain dual expression for the partition function @xmath158 \
\nonumber \\ & \times & \prod_{p,\mbox{even } } b_0\left [ \xi ( x),\xi ( x+e_1 ) , \xi ( x+e_2 ) , \xi ( x+e_1+e_2 ) ; \ \beta \right ] \ , \label{dual2su2}\end{aligned}\ ] ] where the dual boltzmann weight @xmath159\ ] ] is given by @xmath160 \right \ } \ \phi ( \xi ( x);v_1 v_2 ) \\ & \times & \phi ( \xi ( x+e_2);v_2^{\dagger}v_3 ) \ \phi ( \xi ( x+e_1+e_2);v_3^{\dagger } v_4^{\dagger } ) \ \phi ( \xi ( x+e_1);v_4 v_1^{\dagger } ) \ .
\nonumber \end{aligned}\ ] ] dual expression for the correlation function can be easily re - constructed from eq.([corfuncdef ] ) and last formulae .
consider , for simplicity the shortest path between points @xmath161 and @xmath162 .
the corresponding dual path is a set of links which point direction @xmath163 .
let @xmath164 be even .
since every link belongs to one and only one even plaquette we obtain @xmath165 \ \nonumber \\ & \times & \prod_{p^{\prime } } b_0(p^{\prime } ) \
\sum_{s_1=-j}^{j } \cdots \sum_{s_r =- j}^{j } \
\prod_{p(c ) } b_j(p(c);s_{i-1}s_is_{i+1 } ) \ , \label{corrfd1}\end{aligned}\ ] ] where @xmath166 denotes a set of even plaquettes which do not contain links @xmath167 and @xmath168 - a set of even plaquettes which contain such links . in notations of fig.[matrcoupl ] we have @xmath169 \right \ } \
v_j^{s_{i-1}s_i}(l_1 ) \
v_j^{s_{i}s_{i+1}}(l_3 ) \\ & & \times \phi ( \xi ( x);v_1 v_2 ) \ \phi ( \xi ( x+e_2);v_2^{\dagger}v_3 ) \ \phi ( \xi ( x+e_1+e_2);v_3^{\dagger } v_4^{\dagger } ) \ \phi ( \xi ( x+e_1);v_4 v_1^{\dagger } ) \ . \nonumber \end{aligned}\ ] ] finally , the disorder operator in the dual representation can be readily obtained from its expression in the link formulation .
let @xmath170=-h[v(l)]$ ] .
consider the following disorder operator @xmath171 \right ] \
, \label{disparamdef}\ ] ] where @xmath172 $ ] .
expectation value of this operator can be written as a ratio of two partition functions .
with help of the change of variables @xmath173 this ratio can be presented as the following expectation value in the dual formulation @xmath174 generalization of these representations to higher dimensions is straightforward . in this case
we couple link matrices in such a way as to decouple integration over them into a set of integrations within each even @xmath2-cell of the dual lattice . then the partition function can be written as @xmath175 \nonumber \\ & \times & \prod_{h_d , \mbox{even } } b(h_d ) \ , \label{dualsu2d}\end{aligned}\ ] ] where the dual boltzmann weight reads @xmath176 \right \ }
\right ] \prod_{\omega_p \in h_d } \phi ( \xi ( \omega_p ) ; \gamma ( \omega_p ) ) \ . \label{dualbwdef}\ ] ] here , @xmath177 are dual variables which belong to @xmath3-cells of the dual lattice and @xmath178 are combined angles arising from multiplication of @xmath91-cell matrices ( original link matrices ) which belong to the same plaquette trace . to our opinion ,
the dual representation presented above bears the closest analogy to the abelian case .
for example , in two - dimensional model there are three independent dual variables per site as could be expected on general grounds .
further , as follows from the eq.([dualbw ] ) the interaction between dual variables is local though it is not exactly the same as in abelian models : in addition to nearest - neighbour interaction , there is an interaction over diagonals of even plaquettes .
the deficiency of this dual representation is that the dual weight is in general complex valued .
this makes the numerical mc computations impossible . on the other hand ,
we believe it is well suited for an analytical investigation of the model at low temperatures . in the described above scheme the realization of ideas leading to the local dual formulation relies on the special representation for @xmath5 matrices ( [ vrmn ] ) , ( [ su2matr ] ) .
this representation appears to be very efficient at large @xmath130 as we shall show in the next section .
we do not know if similar parameterization exists for any group @xmath18 .
therefore , for arbitrary group @xmath18 we proceed as follows .
instead of ( [ su2char ] ) we use unitarity relations for the cg coefficients of @xmath1 or @xmath0 group @xcite @xmath179 which allows to present group characters as @xmath180 where @xmath181
we then proceed exactly as for @xmath5 . for
the sake of simplicity let us consider the two - dimensional theory .
identifying @xmath182 as new dual variables the pf can be written as @xmath183 \prod_{p,\mbox{even } } \
b(p ) \ , \label{dualsun}\ ] ] where the dual boltzmann weight takes the form @xmath184 \right \ } \ g ( \xi ( x);v_1 v_2 ) \\ & \times & g ( \xi ( x+e_2);v_2^{\dagger}v_3 ) \ g ( \xi ( x+e_1+e_2);v_3^{\dagger } v_4^{\dagger } ) \
g ( \xi ( x+e_1);v_4 v_1^{\dagger } ) \ . \nonumber\end{aligned}\ ] ] substituting ( [ sungdef ] ) into the last equation one can express @xmath185 in terms of one - link integrals defined in ( [ dualweight ] ) .
since the cg coefficients of @xmath0 group can be chosen real @xcite , the dual weight in this representation is also real .
we have not studied if it is positive .
one of the most important application of the dual formulation is the investigation of the low - temperature region of two - dimensional non - abelian models .
the first step in this investigation is to establish an asymptotic expansion for dual weights at large @xmath130 . clearly , it is necessary to get asymptotics uniformly valid in all fluctuations of dual variables .
it turns out that such asymptotics can indeed be constructed , and this is , in our opinion one of the most important advantages of our dual formulation .
calculation of the asymptotic expansion essentially relies on the fact that when @xmath11 the link matrix performs only small fluctuations around unity both in the finite volume and , most importantly in the thermodynamic limit .
note , that this is not the case for the original ( @xmath30 ) degrees of freedom : in the large volume limit their fluctuations are not bounded .
in this subsection we construct an effective low - temperature theory for the dual formulation ( [ dual2su2 ] ) .
the corresponding dual weight is given by eq.([dualbw ] ) .
we modify it by introducing sources @xmath186 which can be related to the correlation function @xmath187 + j_k(l)\omega_k(l ) \right ] \right \ } \
\phi ( \xi ( x);v_1 v_2 ) \\ & \times & \phi ( \xi ( x+e_2);v_2^{\dagger}v_3 ) \ \phi ( \xi ( x+e_1+e_2);v_3^{\dagger } v_4^{\dagger } ) \ \phi ( \xi ( x+e_1);v_4 v_1^{\dagger } ) \ , \nonumber \end{aligned}\ ] ] where @xmath188 it is well known ( see , e.g. @xcite ) that when @xmath11 the link matrix performs only small fluctuations around unit matrix .
our next calculations relay on this fact .
let @xmath117 $ ] has a unique maximum at @xmath189 .
we also require that @xmath117 $ ] possesses character expansion .
then , in the vicinity @xmath189 and in the parameterization ( [ vrmn ] ) , ( [ su2matr ] ) one can always construct the following expansion @xmath190 \ = \ h[i ] -\frac{\gamma}{2}\sin^2\frac{\omega ( l)}{2 } + { \cal o}(\omega^3 ) \ .
\label{hexp}\ ] ] for the standard choice @xmath117={\rm tr } v_{1/2}=2\cos\frac{\omega}{2}$ ] one finds @xmath191=2 $ ] and @xmath192 .
next , we need uniform asymptotic expansion of the function @xmath193 .
we derive such an expansion in the appendix a. the leading term of this expansion takes the form @xmath194 where the classical angular momentum @xmath195 is introduced in ( [ clangm ] ) . as follows from the semiclassical expansion ( [ phisemiclexp ] ) , the remainder is actually bounded as @xmath196 for all @xmath195 .
therefore , corrections @xmath197 and higher are in fact negligible in the limit @xmath11 so we treat them perturbatively .
lastly , we substitute all these expressions into ( [ dualbwsrc ] ) , express combined angles @xmath198 , @xmath199 , @xmath200 in terms of link angles and perform integration .
we omit all these simple but rather cumbersome calculations and give only the final result . introducing notations @xmath201 - 3l^2\ln ( \gamma\beta ) \right ] \ , \label{constant}\ ] ] @xmath202 and making the change of variables @xmath203 so that @xmath204 takes integer values , we present the result for the dual partition function at large @xmath130 in the form @xmath205
\exp [ s_{eff } ] \ . \label{dualeff}\ ] ] an expression for the effective action @xmath206 is too large to be given here in full , therefore we present only its expansion . rescaling sources as @xmath207 the effective action can be expanded as @xmath208 we have computed first three terms in the above expansion . here
we give @xmath209 and @xmath210 @xmath211 @xmath212 @xmath213 @xmath214 we have introduced here the following notations @xmath215 @xmath216 due to the term @xmath217 in the measure , summation over @xmath204 can be extended to include point @xmath218 .
we use poisson summation formula , perfrom change of variables @xmath219 and make the following expansion @xmath220 this expansion is valid when @xmath221 .
we call this expansion semiclassical approximation .
it allows to define an obvious analog of the vortex spin - wave representation for the pf @xmath222 \\ & \times & \exp \left [ s_{eff}(r_k(x ) ) + 2i\pi \sqrt{\gamma\beta}\sum_x m(x ) \left ( \sum_kr^2_k(x ) \right ) ^{1/2 } \right ] \ .
\nonumber \end{aligned}\ ] ] here we investigate the dual model given by eq.([pflsu2 ] ) with the corresponding dual weight ( [ linkint ] ) .
our approach is to replace the matrix elements in the integrand of ( [ linkint ] ) by the @xmath6-like matrix elements .
justification of such replacement will be given in the course of calculations .
for @xmath5 matrix elements we use the wigner @xmath2-function parametrized by euler angles @xmath223 for the sake of simplicity we consider here only the standard action @xmath117=\sum_nd_{1/2}^{nn}$ ] , i.e. the fundamental character which in this parameterization reads @xmath224 the invariant measure on the group takes the form @xmath225 substituting last expressions into ( [ linkint ] ) one can exactly integrate over @xmath226 and @xmath227 angles .
this gives @xmath228 to get the asymptotics when @xmath11 we use the fact ( described in the previous subsection ) that the link matrix in this region performs small fluctuations around unit matrix , namely @xmath229 it seems to be unfeasible task to solve saddle - point equation for the integrand of ( [ xio1int ] ) .
we therefore take a different approach to the problem , and replace this integrand by its asymptotics at @xmath230 such that @xmath231 here @xmath232 is classical angular momentum defined in eq.([clangm ] ) .
the last condition states that the main contribution to the integral comes from diagonal components of @xmath5 matrix elements , non - diagonal contributions can be treated perturbatively . to be precise
, we adjust the following three approximations : 1 .
we use the following asymptotics for the modified bessel function @xmath233 when @xmath234 and such that @xmath235 .
it leads to @xmath236 due to the property ( [ fluct ] ) the remainder is bounded like @xmath196 .
2 . the integration region over @xmath198
is extended to infinity after proper change of variables ( this introduces only exponentially small corrections which can be properly bounded ) .
the main approximation concerns the behaviour of matrix elements @xmath237 in the vicinity @xmath230 .
the standard results on the asymptotic expansion of @xmath237 assert that when @xmath238 , @xmath239 the @xmath5 matrix elements approach the @xmath6 matrix elements @xcite @xmath240 where @xmath241 is the bessel function .
we have found , however that this approximation is oversimplified .
in particular , it does not lead to the correct gaussian distribution of the dual boltzmann weight that is clearly seen from the effective theory obtained above .
moreover , eq.([iso2asympt ] ) states that , e.g. the asymptotics of all diagonal elements is the same . to get more precise asymptotics one can use representation of @xmath242-function in terms of the jacobi function .
the latter can be treated with help of the tricomi expansion into degenerate hypergeometric functions as described in @xcite , section 10.14 .
finally , we use uniform tailor expansion for the degenerate hypergeometric functions @xcite , section 6.13 .
this leads to the following asymptotics @xmath243 where all notations are given in ( [ finnote ] ) .
it can be easily checked with help of mathematica that the last equation gives better approximation to @xmath242-function compared to eq.([iso2asympt ] ) , especially at large values of @xmath244 and is valid , in fact for all values of @xmath35 .
moreover , it gives very reasonable approximation in rather wide region of @xmath198 , except region @xmath245 .
in the appendix b we give independent and simpler proof of this asymptotic expansion calculating also the first correction @xmath246 to eq.([iso2likeasympt ] ) .
the final result is given in eq.([dfuncasympt ] ) .
we term this asymptotics as @xmath6-like approximation of @xmath5 matrix elements .
the only ( but essential ) difference from the exact @xmath6 matrix element is the presence of @xmath247 in the argument of the bessel function . combining eqs.([bessas1 ] ) , ( [ iso2likeasympt ] ) we arrive finally at the following asymptotic representation for the one - link integral @xmath248 where we have used notations @xmath249 @xmath250 making change of variables @xmath251 in the last integral and extending the integration region over @xmath83 to infinity the last integral can be written as @xmath252 where @xmath195 and @xmath253 are defined in eq.([finnote ] ) . calculating the last integral we find @xmath254 \nonumber \\ \times \ i_k\left(\frac{r_1\sin\theta_1 r_2\sin \theta_2}{2\beta } \right ) \ , \label{xio2das}\end{aligned}\ ] ] where @xmath255 .
again , we have compared the present asymptotics with the exact expression given by eq.([xio2dexact ] ) using matemathica .
suprisingly , if @xmath130 is sufficiently large the asymptotic formula gives very reasonable approximation for all configurations we have studied .
it is interesting to stress that the same asymptotics can be obtained directly from ( [ xio2dexact ] ) , where the cg coefficients are replaced by their semiclassical limit by ponzano - regge @xcite .
the method presented above has however two advantages .
first of all , we have computed corrections to the leading behaviour ( [ iso2likeasympt ] ) , therefore @xmath256-corrections to the asymptotic expansion of one - link integral can be computed in a straightforward manner .
corrections to the ponzano - regge formula are in general unknown .
secondly , we find that our method is much simpler .
in this article we proposed dual formulation of non - abelian spin models .
our approach to the dual transformations is summarized in the beginning of the section 2.2 .
main formulae of the present paper , eqs.([pfsundual])-([2funcdualw ] ) , ( [ dual2su2])-([dualbw1 ] ) give two versions of the dual formulation . to demonstrate the usefulness of our formulations we have computed the low - temperature asymptotics of the dual boltzmann weights for two - dimensional model .
we reckon that the results for the dual weights provide significant simplification of the partition function at large values of @xmath130 .
nevertheless , even in this case the model is still too complicated to be solved exactly . in our next paper @xcite
we consider some physical applications of the dual representation .
in particular , we attempt to give a saddle - point solution of the dual model and derive effective @xmath10 model for the two - point correlation function . also , we calculate continuum limit of the dual representation and solve classical continuum equations of motion . some simple consequences which worth noting can be obtained from ( [ vortspwrepr ] ) .
let us perform a shift of variables in the integrand of ( [ vortspwrepr ] ) @xmath257 this shift allows to extract the main contribution from the sources to the free energy which emerges from the gaussian term in the effective action , i.e. @xmath209 given by eq.([sefflead ] ) @xmath258 \tilde{z}(j,\beta ) \ , \label{zjl}\ ] ] where @xmath259 is a remaining part of the pf . here
@xmath260 and @xmath261 are link green functions introduced in @xcite . in @xcite
we show that the following sources @xmath262 and @xmath263 , otherwise , can be related to the correlation function in the representation @xmath79 . substituting these sources into ( [ zjl ] )
one finds after simple analysis @xmath264 \ , \label{gjlead}\ ] ] where @xmath265 , for @xmath266 . by no means
we claim that this result establishes power - like decay of the correlation function : in fact , we do not know the @xmath195-dependence of the remainder . therefore , at the moment we can only claim that the shift ( [ shift ] ) allows to extract the main perturbative contribution .
we would like to stress that the shift ( [ shift ] ) is precisely the same as the one used in @xcite .
unfortunately , we do not know if the methods of @xcite can be applied to the partition function ( [ zjl ] ) to bound the remainder : the essential difference from the @xmath4 model is that even in the absence of vortices both full and effective dual actions are complex .
nevertheless , we think that such possibility deserves further investigations .
in this appendix we calculate the asymptotic expansion of the function @xmath267 , where @xmath268 is given by eq.([phidef ] ) @xmath269 here , @xmath145 is the legendre polynomial and @xmath270 is given in eq.([combangle ] ) .
@xmath5 matrix is taken in the parameterization ( [ vrmn ] ) , ( [ su2matr ] ) .
also , we define and compute the semiclassical limit of @xmath271 . to this end
we introduce classical angular momentum as @xmath272 we are interested in the expansion of @xmath193 when @xmath230 uniformly valid in all variables @xmath273 . because @xmath274 and @xmath226 are compact variables the problem of uniformity concerns mainly the value of the variable @xmath35 .
we face two situations : 1 ) @xmath35 is fixed and 2 ) @xmath239 . in the second case
we look for the asymptotic expansion such that @xmath275 . as the first step we express the generalized characters entering @xmath193 through the associated legendre functions of the first kind taken on the cut @xmath276 $ ] @xmath277^{1/2 } { \rm p}^{-\lambda-1/2}_{2 r+1/2 } ( \cos\frac{\omega}{2 } ) \ . \label{chileg}\ ] ] when @xmath35 is fixed , we construct tailor expansion of @xmath193 . it is more convenient to expand in powers of @xmath278 rather than in powers of @xmath198 . from the representation ( [ chileg ] )
it follows that @xmath279^{1/2 } \left ( \frac{y}{2 } \right ) ^{\lambda } \left [ 1 - \frac{r^2-(\lambda + 1)^2}{2\lambda + 3}\frac{y^2}{2 } + { \cal{o}}(y^4 ) \right ] \ . \label{chiexp}\ ] ] substituting last expression into eq.([phidef1 ] ) , it is straightforward to find the following expansion @xmath280 where the coefficients of the expansion @xmath281 are @xmath282 @xmath283 \ , \label{f2}\ ] ] @xmath284 \nonumber \\ & + & \frac{3}{5}(r^2 - 4)-r^2 + 1 - \frac{3}{5}\sqrt{r^2 - 4}\sqrt{r^2 - 9 } + \sqrt{r^2 - 1}\sqrt{r^2 - 4 } \}\ . \label{f3}\end{aligned}\ ] ] when @xmath285 we expand coefficients @xmath286 in powers of @xmath195 .
this defines the semiclassical expansion of @xmath193 when @xmath238 , @xmath266 such that @xmath275 @xmath287 it would seem that the bound @xmath288 in ( [ phiexp ] ) and , therefore in ( [ phisemiclexp ] ) does not hold uniformly in @xmath195 .
it turns out however that the bound is correct both for fixed @xmath35 and for @xmath239 . to prove this we need an expansion for the associated legendre functions at large @xmath35 uniformly valid in the neighbourhood of the point @xmath289 .
such an expansion is given by macdonald s formula @xcite , section 3.5 @xmath290^{-\mu } \\
& \times & \left \ { j_{\mu}(\alpha ) + \sin^2\frac{\omega}{2}\left [ \frac{\alpha}{6}j_{\mu + 3}(\alpha ) - j_{\mu + 2}(\alpha ) + \frac{1}{2\alpha}j_{\mu + 1}(\alpha ) \right ] + { \cal o}(\omega^4 ) \right \ } \ , \nonumber \end{aligned}\ ] ] where @xmath291 .
the ratio of factorials in ( [ chileg ] ) is expanded as @xmath292^{1/2 } = r^{\lambda + 1/2}\left [ 1-\frac{\lambda ( \lambda^2+\frac{3}{2}\lambda + \frac{1}{2})}{6r^2 } + { \cal o}(r^{-4 } ) \right ] \ . \label{ratiofact}\ ] ] with these results , and using recursion relations for the bessel functions to reduce order to @xmath293 we obtain the representation for the generalized character of the form @xmath294^{1/2 } \
b(r,\lambda ; \omega,\partial_h)\mid_{h=1 } \ j_{\lambda + \frac{1}{2}}(\alpha h ) \ . \label{chiasymp}\ ] ] the function @xmath295 has the following asymptotic expansion @xmath296 + { \cal o } \left ( \sin^4\frac{\omega}{4 } \right ) \ , \label{basymp}\ ] ] where @xmath297 . substituting last expressions into eq.([phidef1 ] )
we can easily calculate all the sums and derivatives .
this leads to @xmath298 re - expanding last formula in powers of @xmath299 one sees that eq.([phirwexp ] ) coincides with eq.([phisemiclexp ] ) .
due to the bound @xmath300 in ( [ legasymp ] ) , the same bound holds in ( [ phirwexp ] ) and , consequently in ( [ phisemiclexp ] ) and in ( [ phiexp ] ) .
we thus conclude that the tailor expansion ( [ phiexp ] ) does provide expansion at small @xmath198 which is valid both for fixed values of @xmath35 and for @xmath239 .
here we compute the asymptotic expansion of the @xmath5 matrix elements @xmath237 in the classical region @xmath301 uniformly valid in the vicinity of the point @xmath289 for all allowed values of @xmath302 and @xmath303 , @xmath304 is held fixed . to get such an asymptotics we first present @xmath242-function in terms of hypergeometric function @xmath305 @xmath306^{\frac{1}{2 } } \left(\sin \frac{\omega}{2 }
\right)^k \left(\cos \frac{\omega}{2 } \right)^{-p } \nonumber \\
\times f(s+k+1,-s - p;k+1;\sin^2 \frac{\omega}{2 } ) \ , \label{dasf}\end{aligned}\ ] ] where @xmath307 if @xmath308 , @xmath309 otherwise , and @xmath310 as is seen from the arguments of the hypergeometric function the infinite series in @xmath311 terminates so that right - hand side of ( [ dasf ] ) is polynomial in @xmath312 @xmath313^{\frac{1}{2 } } \left(\sin \frac{\omega}{2 } \right)^k \left(\cos \frac{\omega}{2 } \right)^{-p } \nonumber \\
\times \sum_{l=0}^{r+\frac{1}{2}(p - k)}(-1)^l \frac{(\sin^2\frac{\omega}{2})^l}{\gamma ( k+1+l)l!}{\cal f}_l(x , y ) \ , \label{dseries}\end{aligned}\ ] ] where we have introduced the following notations @xmath314 @xmath315 @xmath316 the second step consists in expanding the ratio of gamma functions .
this can be done with help of the following formula @xmath317 where @xmath318 are the generalized bernoulli polynomials .
important point concerns the large expansion parameter we use .
we take not merely classical angular momentum @xmath319 but rather quantities @xmath82 and @xmath83 defined above .
such a choice gives more accurate asymptotics valid in a wider region of parameters .
then , in the case of quantity @xmath320 the series in ( [ gratio ] ) terminates because @xmath321 and representation ( [ gratio ] ) becomes exact .
it leads to @xmath322 ^ 2}{(l+k - s_1)!(l+k - s_2 ) ! } \nonumber \\ \times b_{s_1}^{(k+l+1)}(\frac{1}{2}(k+1 ) ) b_{s_2}^{(k+l+1)}(\frac{1}{2}(k+1)+l ) \ \label{flxyexp}\end{aligned}\ ] ] what is essentially the desired expansion at large @xmath82 and @xmath83 .
it follows from the last representation that @xmath323 \ . \label{flxyasym}\end{aligned}\ ] ] for @xmath324 following the same procedure one finds @xmath325 \ . \label{akxasym}\ ] ] substituting last expressions into ( [ dseries ] ) we get after some algebra @xmath326 here we have extended summation over @xmath85 to infinity since this introduces corrections of the order @xmath327 or less .
recalling now the series representation for the bessel function @xmath328 we can easily sum up all series in the last formula .
finally , we arrive at the following asymptotic expansion for @xmath242-function @xmath329 + o ( \sin^4\frac{\omega}{2 } ) \ } \ . \label{dfuncasympt}\end{aligned}\ ] ] we have introduced here the following notations : @xmath330 r. savit , phys.rev.lett .
39 ( 1977 ) 55 ; rev.mod.phys . 52 ( 1980 ) 453 .
j. frhlich , t. spencer , commun.math.phys . 81 ( 1981 ) 527 .
h. pfeiffer , j.math.phys .
44 ( 2003 ) 2891 .
g. batrouni , m.b .
halpern , phys.rev .
d30 ( 1984 ) 1775 .
o. borisenko , v. kushnir , a. velytsky , phys.rev .
d62 ( 2000 ) 025013 .
o. borisenko , v. kushnir , low - temperature behaviour of @xmath7 lattice @xmath5 spin model , proc . of nato workshop
`` integrable structures of exactly solvable two - dimensional models of quantum field theory '' , ed . by s.
pakuliak , and g. von gehlen , kluwer academic publishers , 2001 , 55 .
a. patrascioiu , e. seiler , j.statist.phys .
69 ( 1992 ) 573 ; j.statist.phys .
106 ( 2002 ) 811 - 826 .
o. borisenko , p. skala , phys.rev .
d62 ( 2000 ) 014502 .
t. banks , j. kogut , r. myerson , nucl.phys .
b121 ( 1977 ) 493 . m. gpfert , g. mack , commun.math.phys .
81 ( 1981 ) 97 ; 82 ( 1982 ) 545 .
varshalovich , a.n .
moskalev , v.k .
khersonskii , quantum theory of angular momentum , world scientific publishing co.pte.ltd .
, singapore - new jersey - hong kong , 1988 .
vilenkin , a.u .
klimyk , representation of lie groups and special functions , kluwer academic publishers , dordrecht - boston - london , vol.316 , 1995 .
j. bricmont and j .- r .
fontaine , j.stat.phys .
26 ( 1981 ) 745 .
vilenkin , special functions and theory of group representations , nauka , moscow , 1991 .
h. bateman , a. erdelyi , higher transcendental functions , new york - toronto - london , 1953 ; russian edition , nauka , moscow , vol.1,2 , 1973 .
g. ponzano , t. regge , semiclassical limit of racah coefficients , in spectroscopic and group theoretical methods in physics , north - holland publ .
co. , amsterdam , 1968 , 1 ; k. schulten , r.g .
gordon , journal of math.phys . , v16 ( 1975 ) , 1971 .
o. borisenko , v. kushnir , saddle - point solution of the dual of @xmath7 @xmath5 chiral model , in preparation . | non - abelian lattice spin models with symmetry group @xmath0 or @xmath1 can be formulated in terms of link variables which are subject to the bianchi constraints .
using this representation we derive exact and local dual formulation for the partition function of such models on a cubic lattice in arbitrary dimension @xmath2 .
locality means that the dual action is given by a sum over some subset of hypercubes of the dual lattice and the interaction between dual variables ranges over one given hypercube .
dual variables are in general discrete - valued and live on @xmath3-cell of the dual lattice , in close analogy with the @xmath4 model .
we use our construction to study in details the dual of @xmath5 principal chiral model in two dimensions .
we give dual expressions also for two - point correlation function in arbitrary representation and for the free energy of defects .
leading terms of the asymptotic expansion of the dual boltzmann factor are computed and it is proven that at low temperatures it converges to a certain gaussian distribution uniformly in all fluctuations of dual variables . this result enables us to define the semiclassical limit of the dual formulation and to determine an analog of the vortex spin - wave representation for the partition function .
such representation is used to extract leading perturbative contribution to the correlation function which shows power - like decay at weak coupling .
we also present some analytical evidences that the low - temperature limit of the dual formulation is completely described by @xmath6-like approximation of @xmath5 matrix elements .
* dual formulations of non - abelian spin models : local representation and low - temperature asymptotics * * o. borisenko , v. kushnir * + |
the following is a special case of theorem 8.5a of ivanov s survey article @xcite .
it is a consequence of ideas laid out in the papers of ivanov on the mapping class groups of higher genus orientable surfaces and the work of korkmaz @xcite on the complex of curves associated to an @xmath69-punctured sphere . for simplicity we shall write @xmath72 for the 2-sphere with @xmath69 points removed .
if @xmath73 is an orientable surface ( without boundary ) , we denote by @xmath74 the group of all diffeomorphisms of @xmath73 ( not necessarily respecting orientation ) modulo diffeotopy .
[ ivanovkthm ] let @xmath75 .
if @xmath76 is an isomorphism between finite index subgroups @xmath77 , then @xmath78 is the restriction to @xmath79 of an inner automorphism of @xmath80 ( conjugation by some element @xmath81 ) .
[ defncommg ] let @xmath82 be a group .
we define the _ abstract commensurator group of @xmath82 _ to be @xmath83 where @xmath84 if they agree on a finite index subgroup of @xmath82 .
we note that the group structure on @xmath85 is given by composition of isomorphisms after appropriate restriction of their domains to finite index subgroups .
[ normalizers ] let @xmath86 , @xmath75 , let @xmath87 denote a finite index subgroup and @xmath88 the normalizer of @xmath79 in @xmath89 .
then * @xmath90 , and * @xmath91 .
\(i ) there is a natural homomorphism @xmath92 which takes @xmath93 to conjugation by @xmath94 restricted to the finite index subgroup @xmath95 .
it follows from theorem [ ivanovkthm ] that this map is surjective . to show that it is injective , we observe that @xmath89 has trivial `` virtual centre '' , that is , any element which centralizes a finite index subgroup is necessarily trivial .
this is because any mapping class @xmath96 is determined by its action on a finite number of isotopy classes of simple closed curves on the surface @xmath73 , and therefore by its action on the @xmath6-th powers @xmath97 of the corresponding dehn twists for any choice of @xmath6 .
given a finite index subgroup @xmath98 centralized by @xmath99 we can always find a sufficiently large @xmath6 such that @xmath100 contains all @xmath97 .
then @xmath99 is determined by its action on @xmath100 and so must be the identity in @xmath89 .
\(ii ) there is now an obvious homomorphism @xmath101 which factors through the inverse of the isomorphism just described , and realizes each automorphism of @xmath79 as conjugation by an element of @xmath89 .
it follows easily that this map is injective with image precisely the normalizer @xmath88 of @xmath79 in @xmath89 .
let @xmath102 denote the group of permutations of the set @xmath103 , for @xmath104 . if @xmath105 then we consider @xmath106 as a subgroup of @xmath102 consisting of those permutations which fix the subset @xmath107 . notice that each mapping class of the @xmath69-punctured sphere @xmath72 induces a permutation of the punctures . moreover , the mapping class group acts on the orientation class of the surface .
thus we have a surjective group homomorphism @xmath108 with kernel the group of pure orientation preserving mapping classes .
note that , strictly speaking , the definition of @xmath109 depends on a labelling of the punctures by the numbers @xmath110 through @xmath69
. * notation .
* fix @xmath0
. we shall write @xmath111 and @xmath112 , @xmath113 , @xmath114 respectively , for the finite index subgroup of @xmath89 generated by those orientation preserving diffeomorphisms which fix the last 1 , 2 , 3 punctures of @xmath47 respectively . these three subgroups are , in other words , the preimages under @xmath109 of the subgroups @xmath115 , with @xmath116 respectively .
in particular we have @xmath117=2(n+2)$ ] , @xmath118=2(n+1)(n+2)$ ] and @xmath119=2n(n+1)(n+2)$ ] .
it is well known that the @xmath120-string braid group , @xmath36 , is isomorphic to the mapping class group of the @xmath120-punctured disk _ relative _ to the boundary of the disk .
( that is , diffeotopies are required in this case to fix the boundary of the disk _ pointwise _ ) .
the centre @xmath41 of this mapping class group is generated by the dehn twist about a curve parallel to the boundary , and is precisely the kernel of the natural homomorphism to the mapping class group @xmath111 ( induced by inclusion of the punctured disk in @xmath47 ) .
thus we may realise the group @xmath44 as the finite index subgroup @xmath112 of @xmath89 .
it is less well - known ( see @xcite ) that the artin groups @xmath66 and @xmath43 , respectively , are isomorphic to the subgroups of the braid group @xmath36 leaving fixed one , respectively two , of the punctures in the disk .
we note that the latter of these two groups intersects the centre of the braid group trivially , from which we deduce that * @xmath121 , * @xmath122 , and * @xmath123 .
these isomorphisms may be described explicitly as follows .
distribute the punctures @xmath124 ( @xmath125 ) evenly along half the equator in the order @xmath126 so that @xmath127 and @xmath128 are antipodal ( see figure [ fig2 ] )
. then the @xmath24th standard generator of @xmath36 maps to the positive braid twist @xmath129 exchanging @xmath124 with @xmath130 .
the @xmath24th standard generator of @xmath66 maps to @xmath131 if @xmath132 and @xmath129 otherwise .
the @xmath24th standard generator of @xmath43 maps to @xmath133 if @xmath134 and @xmath129 if @xmath135 .
( this maps @xmath43 into the subgroup of @xmath89 fixing @xmath136 , and @xmath137 .
renumbering the punctures gives @xmath114 . )
isomorphic to @xmath44 , @xmath45 and @xmath43 , @xmath0 , fix the subsets @xmath138 and @xmath139 respectively.,width=302 ] [ abc ] ( i ) : : @xmath140 ( ii ) : : there exist short exact sequences @xmath141 let @xmath79 denote one of @xmath112 , @xmath113 , or @xmath114 , and @xmath142 the set of 1,2 or 3 punctures fixed by all elements of @xmath79 .
it is easily checked that @xmath88 is exactly the group of all mapping classes ( orientation preserving or not ) which leave @xmath142 setwise invariant .
it follows that @xmath143 is a direct product of @xmath144 with the group @xmath145 of permutations of the set @xmath142
. statements ( i ) and ( ii ) now follow immediately from the corresponding statements in corollary [ normalizers ] .
it is easy to see , with the aid of figure [ fig2 ] , that the short exact sequences in the proposition above are split in the cases @xmath112 and @xmath113 .
the group @xmath146 may be generated by a reflection in the equatorial plane and a rotation about an axis in the equatorial plane which exchanges the points @xmath127 and @xmath128 .
only the reflection leaves @xmath112 invariant . for @xmath114 , the exact sequence splits providing @xmath147 mod @xmath28 .
to see this , think of @xmath47 as a sphere with @xmath148 punctures arranged symmetrically along the equator and , if @xmath149 , the remaining two punctures at the north and south poles .
then there is an orientation - preserving action of @xmath150 on @xmath47 generated by an order three rotation @xmath151 about the vertical axis through the poles and an order two rotation @xmath152 about a horizontal axis through one ( @xmath153 odd ) or two ( @xmath153 even ) punctures at the equator .
if @xmath154 is a puncture fixed by @xmath152 , then the orbit of @xmath154 under @xmath151 is preserved by @xmath150 , hence for an appropriate numbering of the punctures ( namely , so that @xmath155 ) the action of @xmath150 lies in @xmath156 and is faithful on @xmath142 .
this splits the @xmath150 factor .
the @xmath63 factor is realized by reflection through the equatorial plane . [ nosplit ] if @xmath157 with @xmath158 mod @xmath28 , then @xmath156 does not contain a subgroup isomorphic to @xmath150 . in this case ,
the exact sequence for @xmath114 in the previous proposition does not split .
suppose @xmath159 contains a subgroup @xmath100 isomorphic to @xmath150 and let @xmath79 be the subgroup of @xmath100 generated by a 3-cycle .
then @xmath79 is a subgroup of @xmath46 , hence @xmath79 acts by permutations on the set of punctures @xmath160 .
every @xmath79 orbit consists of either @xmath110 or @xmath28 punctures . since @xmath161 , the number of fixed points of @xmath79 must also be congruent to @xmath110 mod @xmath28 .
if @xmath79 fixes @xmath162 or more punctures , then it lies in a subgroup of @xmath89 isomorphic to @xmath114 which is impossible since @xmath114 is torsion - free ( since , by @xcite , it is a subgroup of the braid group on @xmath37 strings which is known to be torsion free ) .
thus @xmath79 must have a unique fixed puncture .
since @xmath79 is normal in @xmath100 , all of @xmath100 must fix this puncture and hence @xmath100 lies in a subgroup of @xmath89 isomorphic to @xmath112 . by a theorem of bestvina , @xcite thm 4.5 , every finite subgroup of @xmath112 is cyclic , so we arrive at a contradiction .
it has been observed by several authors @xcite@xcite@xcite that there exists a semidirect product decomposition @xmath163 ( see @xcite for further discussion of this decomposition . ) the coxeter graph of @xmath42 is an @xmath6-cycle and the generator of the cyclic factor in the semi - direct product acts on @xmath42 via an order @xmath6 rotation of this graph .
the centre of @xmath66 is the subgroup @xmath164 of the cyclic factor .
thus we also have @xmath165 here we interpret @xmath42 as the subgroup of @xmath113 consisting of mapping classes of `` zero angular momentum '' about the axis through the two fixed punctures .
more precisely , suppose that the punctures of @xmath47 are arranged so that @xmath127 and @xmath128 are placed at the north and south pole respectively , and the remaining points @xmath166 are equally spaced ( in that cyclic order ) around the equator , see figure [ fig3 ] .
define @xmath167 to be the subgroup of @xmath89 generated by the braid twists @xmath129 which exchange the points @xmath124 and @xmath130 , for @xmath60 , indices taken mod @xmath6 .
the isomorphism @xmath168 is given by sending the @xmath24th standard generator of @xmath42 to @xmath129 . note that @xmath167 is a finite index subgroup of @xmath89 of index @xmath169=n[\ga:\ga_b]=2n(n+1)(n+2)$ ] .
in particular , @xmath170 . of @xmath111 isomorphic to @xmath42.,width=302 ] [ atilde ]
let @xmath171 be as above .
then @xmath172 . consequently @xmath173 and
there is an exact sequence @xmath174 where @xmath58 denotes the dihedral group of order @xmath175 .
moreover , a splitting of this sequence is given by realizing the group of outer automorphisms as the direct product of the group of graph automorphisms of @xmath42 with the group of order @xmath176 generated by the inversion @xmath177 which inverts each standard generator .
we suppose that the punctures of @xmath47 are arranged as shown in figure [ fig3 ] .
the outer automorphism group of @xmath113 may then be generated by the reflection in the equatorial plane and a rotation exchanging the two poles ( this gives a splitting of the exact sequence of proposition [ abc ] slightly different to the one previously mentioned ) .
it follows easily that @xmath167 is invariant by these mapping classes , and hence a characteristic subgroup of @xmath113 .
in particular , we have @xmath178 .
let @xmath179 . to prove that @xmath172 it suffices now to check that every mapping class which normalizes @xmath167 leaves @xmath142 invariant .
but if @xmath180 then @xmath181 normalizes @xmath182 , and this is just to say that @xmath99 induces a symmetry of the puncture set which leaves @xmath142 invariant , as required .
the remaining statements are now easily deduced from corollary [ normalizers ] and proposition [ abc ] . in summary ,
propositions [ abc ] and [ atilde ] give us explicitly the automorphism groups of @xmath44 , @xmath45 , @xmath42 , and @xmath43 , for all @xmath0 , as well as the fact that the abstract commensurator group is in each case isomorphic to @xmath46 .
this proves theorem [ main ] with the exception of the first isomorphism in each of part ( ii)(a ) and ( b ) . in the next section
we shall recover the group @xmath183 from the above result on @xmath184 . in a similar fashion , the result of @xcite on automorphisms of the braid group is easily recovered from the computation of @xmath185 .
these steps will complete the proof of theorem [ main ] .
let @xmath79 be a group with nontrivial centre @xmath41 . by a _
transvection _ of @xmath79 we mean a homomorphism @xmath186 given by @xmath187 , where @xmath188 denotes a function @xmath189 .
the fact that @xmath190 is a homomorphism requires that @xmath188 is also a homomorphism , and it is easily checked that the composite of two transvections is again a transvection .
we note , however , that transvections are typically not automorphisms .
it is easily shown that a transvection is an automorphism ( resp .
injective , resp .
surjective ) if and only if its restriction to the centre @xmath41 is an automorphism ( resp .
injective , resp .
surjective ) .
let @xmath191 denote the group of automorphisms of @xmath79 which are transvections .
then , almost by definition , we have an exact sequence @xmath192 note that @xmath193 always maps isomorphically onto @xmath194 . in general there is an issue , however , as to whether _ every _ outer automorphism of @xmath195 lifts to an automorphism of @xmath79 , and as to whether the extension splits . in the case of an irreducible finite type artin group @xmath32 ,
the extension splits at least over a finite index subgroup @xmath196 of @xmath197 defined as follows .
let @xmath198 be the length homomorphism which maps each generator of @xmath32 to @xmath110 .
recall that the centre @xmath41 of @xmath32 is infinite cyclic .
let @xmath199 be a generator of @xmath41 and let @xmath200 .
the length homomorphism on @xmath32 descends to a `` length homomorphism '' @xmath201 .
define @xmath202 , respectively @xmath196 , to be the group of length preserving or reversing automorphisms , that is , automorphisms @xmath203 such that @xmath204 ( respectively @xmath205 ) .
[ autstar ] suppose @xmath32 is an irreducible artin group of finite type .
then the natural map @xmath206 is an isomorphism .
if , in addition , the abelianization of @xmath32 is infinite cyclic ( i.e. , @xmath32 is not of type @xmath207 , or @xmath208 ) , then @xmath209 is trivial and @xmath210 .
note that a non - trivial transvection @xmath211 of @xmath32 can never be length preserving , and can only be length reversing if @xmath212 ( and @xmath213 ) . since @xmath214 for all irreducible finite type artin groups
we conclude that @xmath215 is trivial . for any length preserving @xmath216
, we define a lift @xmath217 as follows .
( the lift for length reversing automorphisms is defined analogously ) . by definition , for any @xmath218 ,
@xmath219 for some @xmath220 with @xmath221 mod @xmath222 .
since @xmath41 is generated by an element of length @xmath222 , the coset @xmath223 contains a unique representative @xmath224 satisfying @xmath225 .
define @xmath226 . to verify that @xmath227 is a homomorphism ,
note that if @xmath228 and @xmath229 , then @xmath230 so it follows that @xmath231 .
moreover , it is straightforward to verify that if @xmath232 and @xmath233 are two elements of @xmath196 , then @xmath234 lifts to @xmath235 . in particular
, the lift of an automorphism is also an automorphism .
thus , @xmath236 defines a section for the projection @xmath109 , and @xmath109 is an isomorphism .
if the abelianization of @xmath32 is infinite cyclic , then the length homomorphism @xmath237 can be identified with the abelianization map @xmath238 .
hence every automorphism of @xmath32 is either length preserving or length reversing .
the second statement of the proposition follows .
we remark , however , that while @xmath242 when @xmath32 has infinte cyclic abelianization , it is not always the case that @xmath243 ( or equivalently that @xmath244 ) . in particular , for @xmath32 of type @xmath68 with @xmath69 odd , @xmath245 , and @xmath246 is isomorphic to the group of units of @xmath247 , while @xmath248 .
[ trans ] if @xmath32 is an irreducible artin group of type @xmath249 , @xmath0 , or @xmath208 , then @xmath250 . if @xmath32 is of type @xmath251 then @xmath252 .
moreover , in all these cases @xmath253 contains @xmath254 as a finite index subgroup .
we refer the reader to @xcite and @xcite for descriptions of the coxeter graphs , and central elements respectively , for each of the irreducible finite type artin groups . in each case
listed above , the abelianization of @xmath32 is @xmath255 , and the generator @xmath199 of @xmath41 maps to an element @xmath256 , where @xmath257 for type @xmath258 , @xmath259 for type @xmath208 , and @xmath260 for type @xmath31 .
any homomorphism @xmath261 can be obtained by composing the abelianisation homomorphism with a map @xmath262 given by @xmath263 for a pair of integers @xmath264 .
the associated transvection then satisfies @xmath265 where @xmath266 , or @xmath267 , depending on the case ( @xmath268 for type @xmath208 ) .
on the other hand @xmath190 is an automorphism if and only if @xmath269 . when @xmath0 this is only possible if @xmath270 and @xmath271 , or @xmath272 , respectively .
this allows exactly an infinite cyclic group of transvection automorphisms ( all of which act by the identity on @xmath41 ) . in the remaining case , type @xmath251
, we have the constraint @xmath273 , and so @xmath271 or @xmath274 .
we note that @xmath275 is given by the presentation @xmath276 , where @xmath277 .
a transvection of the second type ( @xmath278 ) is realized by the automorphism @xmath279 and @xmath280 .
since in general we have @xmath281 it is easily checked that @xmath209 is an infinite dihedral group with @xmath282 acting as a direction reversing involution .
the subgroup @xmath196 is finite index in @xmath197 ( since @xmath283 takes values in the finite group @xmath284 and @xmath285 , being finitely generated , admits only a finite number of homomorphisms to @xmath284 ) .
hence the inverse image @xmath286 is a finite index subgroup of @xmath33 . by the previous proposition , this subgroup splits as a semi - direct product @xmath287 . for @xmath32 of type @xmath68 , we have the following presentation : @xmath288 in this case the outer automorphism groups were computed in @xcite : @xmath289 where @xmath94 denotes the graph involution @xmath290 , and the @xmath63 factor in each case is generated by the inversion automorphism @xmath291 . in the case
@xmath69 even , the infinite cyclic factor @xmath65 is generated by the automorphism @xmath292 and @xmath293 , and contains the group of transvections as a subgroup of index @xmath294 , when @xmath295 .
( note that @xmath152 fixes both elements @xmath296 and @xmath297 , and that the centre is generated by @xmath298 ) .
it follows that , in the case @xmath299 with @xmath295 even , the subgroup @xmath287 has index @xmath294 in the entire automorphism group .
this behaviour seems to be repeated in the case @xmath300 , for @xmath0 , where , as we will see below , @xmath301 is a subgroup of index @xmath176 in @xmath33 .
the type @xmath251 artin group is however exceptional in this regard . in this case
the group of transvections is generated by @xmath302 and the element @xmath303 , from which it follows that @xmath287 is the whole automorphism group .
we do not know what the index is ( or whether the subgroup is proper ) in the case of @xmath304 .
we first show that the map @xmath306 is surjective and splits .
recall from section [ sect2 ] that @xmath307 .
it is clear from the proof of proposition [ atilde ] that elements of @xmath308 act on this semidirect product by leaving the characteristic subgroup @xmath167 invariant and by mapping the generator of the cyclic factor @xmath309 either to itself or its inverse .
any such automorphism lifts uniquely to an automorphism of @xmath310 .
thus we have a section to the map @xmath311 .
we claim that the @xmath63 factor generated by the inversion @xmath177 also commutes with transvections .
this follows from the fact that @xmath177 restricts to a length reversing automorphism of the center @xmath41 and hence takes the generator @xmath199 of @xmath41 to @xmath313 .
for any transvection @xmath314 and any generator @xmath16 , @xmath315 for some @xmath153 , so @xmath316 by proposition [ trans ] , @xmath250 and the second claim of the proposition now follows .
* the @xmath65 factor in the decomposition @xmath317 is generated by the element @xmath318 of @xmath66 and it acts on @xmath167 via rotation of the @xmath39 graph .
these rotations thus become inner automorphisms in @xmath319 .
the remaining graph automorphisms ( the reflections ) give rise to one of the @xmath63 factors in @xmath319 .
( the other @xmath63 factor is generated by the inversion @xmath177 . ) a generator of this @xmath63 is represented by the automorphism @xmath320 defined by @xmath321 and hence @xmath322 . in particular
, @xmath320 does not lie in @xmath323 ( as long as @xmath0 ) , and this latter group is index 2 in the full automorphism group .
( by contrast , when @xmath32 is of type @xmath324 , the involution @xmath320 just described is exactly the exceptional transvection @xmath325 already mentioned in the preceding discussion ) .
if @xmath79 is a group whose centre @xmath41 is a finite rank free abelian group , the injective transvections which are not automorphisms of @xmath79 have as image a proper finite index subgroup of @xmath79 and therefore represent infinite order elements of @xmath326 ( _ not _ equivalent to automorphisms ) . in the case of a finite type artin group
, this implies that the abstract commensurator group must be large .
it suffices to prove the proposition for irreducible @xmath32 .
let @xmath198 be the length homomorphism .
let @xmath199 be the generator of @xmath328 of positive length @xmath200 .
then for every @xmath329 , there is a homomorphism @xmath330 defined by @xmath331 .
the associated transvection takes @xmath332 to @xmath333 .
in particular , it takes @xmath199 to @xmath334 .
we denote this transvection by @xmath335 .
it is straightforward to verify that @xmath336 .
thus , @xmath335 and @xmath337 are commuting elements of @xmath327 and to find an infinitely generated free abelian subgroup of @xmath327 , it suffices to find an infinite sequence of integers @xmath338 which are relatively prime and satisfy @xmath339 mod @xmath222 .
this can be done inductively by setting @xmath340 and @xmath341 .
the transvections @xmath342 are then linearly independent . | we observe that , for fixed @xmath0 , each of the artin groups of finite type @xmath1 , @xmath2 , and affine type @xmath3 and @xmath4 is a central extension of a finite index subgroup of the mapping class group of the @xmath5-punctured sphere .
( the centre is trivial in the affine case and infinite cyclic in the finite type cases ) . using results of ivanov and korkmaz on abstract commensurators of surface mapping class groups we are able to determine the automorphism groups of each member of these four infinite families of artin groups . #
1#1 hs a rank @xmath6 _ coxeter matrix _ is a symmetric @xmath7 matrix @xmath8 with integer entries @xmath9 where @xmath10 for @xmath11 , and @xmath12 for all @xmath13 . given any rank @xmath6 coxeter matrix @xmath8 , the _ artin group _ of type @xmath8 is defined by the presentation @xmath14 adding the relations @xmath15 to this presentation
yields a presentation of the coxeter group of type @xmath8 generated by standard reflections @xmath16 and such that the rotation @xmath17 has order @xmath18 , for all @xmath19 . a coxeter matrix @xmath8 and its artin group @xmath20
are said to be of _ finite type _ if the associated coxeter group @xmath21 is finite , and of _ affine ( or euclidean ) type _ if @xmath21 acts as a proper , cocompact group of isometries on some euclidean space with the generators @xmath22 acting as affine reflections .
the information contained in the coxeter matrix @xmath8 is often displayed in the form of a graph , the _ coxeter graph _ , whose vertices are numbered @xmath23 and which has an edge labelled @xmath18 between the vertices @xmath24 and @xmath25 whenever @xmath26 or @xmath27 . with this particular convention ,
one usually suppresses the labels which are equal to @xmath28 ( but not the corresponding edges ! ) .
note that the absence of an edge between two vertices indicates that the corresponding generators of @xmath20 commute .
we say that an artin group is _ irreducible _ if its coxeter graph is connected and observe that every artin group is isomorphic to a direct product of irreducible artin groups corresponding to the connected components of its coxeter graph . in this paper
we concern ourselves with the following four infinite families of artin groups @xmath29 ( see figure [ fig1 ] ) : the finite types @xmath30 , @xmath31 and affine types @xmath3 and @xmath4 , of rank @xmath0 in each case .
( we refer to @xcite for the classification of irreducible finite and affine type coxeter systems . ) for each artin group @xmath32 on this list , we determine its automorphism group @xmath33 , its outer automorphism group @xmath34 , and the abstract commensurator group @xmath35 of the group modulo its centre ( see definition [ defncommg ] ) . , @xmath31 and affine type @xmath3 , @xmath4 , with rank @xmath0 .
note that , by convention , the subscript in each case indicates the geometric dimension of the artin group.,width=453 ] the artin group @xmath36 is well - known as the braid group on @xmath37 strings , and the automorphism groups of the braid groups were determined by dyer and grossman in @xcite . in this paper
we exploit the fact that the artin groups of type @xmath38 , @xmath39 and @xmath40 also have descriptions as braid groups ( see @xcite ) in order to study their automorphism groups .
note that the centre of any irreducible artin group of finite type is an infinite cyclic group , which we generally denote by @xmath41 .
( on the other hand , the groups @xmath42 and @xmath43 have trivial centres . ) as explained in section [ sect1 ] , each of the groups @xmath44 , @xmath45 , @xmath42 and @xmath43 is a finite index subgroup of the mapping class group @xmath46 of the @xmath5-punctured 2-sphere @xmath47 . using this fact , and appealing to the work of ivanov and korkmaz on the structure of surface mapping class groups @xcite
, we show : [ main ] for each @xmath0 , we have * @xmath48 for each @xmath49 . + * * * _ ( dyer , grossman @xcite ) _ @xmath50 * * @xmath51 and @xmath52 * * @xmath53 * * @xmath54 .
in addition , we show that the exact sequence @xmath55 splits in each of the above cases with the exception of the case @xmath56 with @xmath57 ( see proposition [ nosplit ] ) . in the case of @xmath42 , for example , the dihedral factor @xmath58 is realised by the group of `` graph automorphisms '' induced from symmetries of the coxeter graph , which in this case is an @xmath6-cycle .
every artin group admits an involution @xmath59 for all @xmath60 which is never inner ( because it reverses the sign of the length function @xmath61 , defined by @xmath62 ) .
this accounts for the central @xmath63 factor in each of the outer automorphism groups above .
note that all of the outer automorphism groups in the theorem are finite with the exception of @xmath64 .
the infinite cyclic factor in @xmath64 is generated by a `` transvection '' which multiplies each generator @xmath16 by some element of the center @xmath41 . in section [ sect3 ] , we study the group of transvections of a finite type artin group .
transvection homomorphisms exist in any artin group with non - trivial center , but in general , they are not automorphisms . in particular , transvection automorphisms do not occur when the artin group has abelianisation @xmath65 , as in the case of the braid groups . however , @xmath66 has abelianisation @xmath67 and likewise for artin groups of dihedral type @xmath68 when @xmath69 is even .
( the groups @xmath70 , for @xmath71 , were computed in @xcite and are discussed in section [ sect3 ] ) . by contrast , transvection homomorphisms appear in the abstract commensurator group of _ every _ finite type artin group . in proposition
[ bigcomm ] , we show that the abstract commensurator group of any finite type artin group contains an infinitely generated free abelian group generated by transvections . the first author would like to thank the institut de mathematiques de bourgogne for their hospitality during the development of this paper . |
globular clusters ( gcs ) were once thought to be composed of stars of a single composition and a single age , with few exceptions ( such as @xmath3 centauri ) .
it had long been known that most clusters contained red - giant stars with anomalies in their light - element abundances ( kraft 1979 ) , but it was not clear whether this came as a result of internal mixing or a variation in the primordial abundances .
high - resolution spectroscopy ( e. g. ramirez & cohen 2002 , carretta et al .
2009 ) has shown that these abundance anomalies in the giants almost usually manifest themselves as an anticorrelation between na and o abundances , which is indicative of contamination from high - temperture hydrogen - burning products ( denisenkov & denisenkova 1989 ) .
a similar anticorrelation has also been observed in some un - evolved main - sequence ( ms ) stars ( gratton et al .
2001 , ramirez & cohen 2002 ) , where the internal temperatures do not allow hot cno - cycle burning .
this fact suggested the presence of two stellar generations in the some cluster .
indeed , high - precision photometry from _ hubble space telescope _
( _ hst _ ) images has shown that @xmath3 centauri and ngc 2808 host multiple distinct mss ( anderson 1997 , bedin et al .
2004 , piotto et al .
2007 ) , which have been associated with stellar populations with different helium ( norris 2004 , dantona et al .
2005 , piotto et al .
2005 ) , with the bluer mss having a higher he - content than the redder ones ( norris 2004 , piotto et al .
2005 , dantona et al .
more recently , a split ms has been observed also in 47 tucanae and ngc 6752 ( anderson et al . 2009 , milone et al . 2010 , 2011a ) .
stellar evolutionary models predict that h - burning at high temperatures through the cno cycle should result in enhanced production of he .
in addition to he , such models also predict enhanced production of n , and na and depletion of c , and o. this pattern of enhancement / depletion has recently been confirmed among the blue and red ms stars in ngc 2808 by bragaglia et al .
( 2010 ) .
multiple populations have also been identified in the color - magnitude diagram ( cmd ) of some clusters in the form of multiple sub - giant branches ( sgbs , milone et al .
2008 , marino et al .
2009 ) or multiple or anomalously wide red - giant branches ( rgbs , marino et al .
2008 , yong et al .
2008 , lee et al .
multiple populations with discrete helium abundance may also offer an explanation for the complex hb morphology exhibited by some clusters ( busso et al .
2007 , dantona & caloi 2008 , catelan , valcarce & sweigart 2009 ) .
a direct confirmation of a connection of the hb shape with the chemical content of hb stars comes from marino et al .
( 2011 ) , who have found that stars on the blue side of the instability strip of the cluster m4 are na - rich and o - poor , whereas stars on the red hb are all na - poor . in this paper
, we present a study of multiple stellar populations in the nearby gc ngc 6397 .
evidence for a large spread in light - elements abundance , and a clear na - o anticorrelation in this gc have been extensively documented in literature ( bell et al .
1979 , briley et al . 1990 ,
pasquini et al .
2004 , gratton et al .
2001 , carretta et al . 2005 , 2009 , lind et al . 2009 , 2011 , l11
) . recently , richer et al .
( 2006 ) used the advanced camera for surveys ( acs ) of the _ hst _ for 126 orbits to image an outer region of ngc 6397 in the f606w and f814w bands .
the spectacular cmd they obtained reaches the deepest intrinsic luminosities for a gc achieved thus far , and reveals an extermely tight ms in the @xmath4 color system ( anderson et al .
these observations are consistent with negligible he variation among ngc 6397 stars : @xmath1y @xmath5 0.02 ( di criscienzo et al .
here , we take advantage of the multi - band _ hst _ photometry of the central region to study the ms of ngc 6397 for signs of multiple populations . the paper is organized as follows .
the data and the data reduction are described in sect .
[ sec : data ] , results are shown in sect . [
sec : ms ] , and discussed in sect .
[ sec : discussion ] .
for this project we used _ hst _
images taken at three different epochs with the wide field channel of the acs ( acs / wfc ) and the ultraviolet / visible channel of the wide - field camera 3 ( wfc3/uvis ) . in particular , we used the following data sets ( program - id , camera , date , filters , number of exposures , exposure time ) : ( 1 ) go-11633 , wfc3/uvis , march 2010 , f336w , 6@xmath6620 s and f225w , 22@xmath6680 s ; ( 2 ) go-10257 , acs / wfc , august 2004-june 2005 , f435w , 5@xmath613 s @xmath7 5@xmath6340 s ; ( 3 ) go-10775 , acs / wfc , may 2006 , f606w , 1 s @xmath74@xmath615 s and f814w , 1 s @xmath74@xmath615 s. the field covered by these images is centered on the cluster core .
their footprint are shown in fig .
[ foot ] , together with an image of the 10@xmath610 arcsec central field taken through the f814w filter . clearly , crowding is not severe , even in the center of this nearby cluster .
star positions and fluxes in the acs / wfc images were measured with the software described by anderson et al .
( 2008b ) .
the program analyzes a set of exposures simultaneously in order to produce a catalogue of stars over the field of view .
stars are measured in each image independently by means of a spatially variable point - spread - function model from anderson & king ( 2006 ) , plus a spatially constant perturbation of the psf that accounts for the effects of focus variations .
the photometry has been calibrated as in bedin et al .
( 2005 ) using the encircled energy and zero points of sirianni et al .
( 2005 ) .
the wfc3/uvis images were reduced with a software that is adapted from img2xym_wfi ( anderson et al .
astrometry and photometry were corrected for pixel area and geometric distortion as in bellini , anderson & bedin ( 2011 ) .
the proper motions were derived as in milone et al .
( 2006 ) .
this paper is mainly based on high - precision photometry , and is limited to a sub - sample of stars that are relatively isolated , are well - fit by the psf , and also have small photometric and astrometric errors .
the photometric software provides a number of quality indices that can be used as diagnostics of the reliability of photometric measurements .
specifically , these are : ( 1 ) the rms of the positions measured in different exposures trasformed in a common reference frame ( @xmath8 and @xmath9 ) , ( 2 ) the residuals to the psf fit for each star ( @xmath10 ) , and ( 3 ) the ratio between the estimated flux of the star in a 0.5 arcsec aperture and the flux from neighbouring stars within the same aperture ( @xmath11 , see anderson et al .
2008b for details . ) to select this high - precision sub - sample of stars , we followed the procedure described by milone et al .
( 2009 , sect .
2.1 ) and illustrated in the upper panels of fig . [ m1 ] for ngc 6397 . as an example
, in the lower panels we show the @xmath2 vs. @xmath12 cmd of all the measured stars , of rejected stars , and of stars that pass the adopted criteria of selection .
since ngc 6397 ( @xmath13 , @xmath14 ) is located in a galactic field richly populated by bulge and disk stars , we must make every effort to remove from our final list stars that do not belong to the cluster .
we measured proper motions using the f435w images of go-10257 data set and the go-10775 images , collected in two epochs separated by @xmath02 years .
figure [ m2 ] shows the vector - point diagram of proper motions .
the separation between field stars and cluster stars is very well defined and cluster members were selected on the basis of their common proper motion . in all cmds discussed in the following , field stars have been removed using the proper motion selection criterion . the average reddening for ngc 6397 is @xmath15 ( harris 1996 , 2010 ) .
usually , such a large reddening is not uniform across an acs field .
a visual inspection at the cmds of ngc 6397 indeed reveals that the sequences are broadened by differential reddening .
we corrected the effect of differential reddening by means of a procedure that is described in detail in milone et al .
( 2011b ) , and used in many other papers by these authors . briefly , we define the fiducial ms ridge - line in a given color system for the cluster and estimate , for each star , how the observed stars in its vicinity may systematically lie to the red or to the blue of the fiducial sequence ; this systematic color and magnitude offset , measured along the reddening line , is indicative of the local differential reddening .
we find that the reddening variations are typically smaller than @xmath1e(b - v)@xmath00.01 mag , and never exceed 0.026 mag .
the reddening map we obtained is plotted in fig .
[ m3b ] , where we divide the field of view into 8 horizontal slices and 8 vertical slices , and plot @xmath16 as a function of the y ( upper panels ) and x coordinate ( right panels ) .
we have also divided the whole field of view into 32@xmath632 boxes and calculated the average @xmath16 within each of them .
the resulting reddening map is shown in the lower - left panel , where each box is represented as a gray square .
the levels of gray are indicative of the amount of differential reddening as shown in the upper - right plot . as an example
, a comparison of the original @xmath2 vs. @xmath12 cmd is provided in fig .
figure [ cmds ] shows cmds with two different color baselines : @xmath17 vs. @xmath18 and @xmath17 vs. @xmath19 .
it is clear that the ms of ngc 6397 is split into two distinct sequences in both diagrams .
it is interesting to note that in the @xmath18 color , the majority of the stars are on the blue ms branch ( hereafter msb ) , while in the @xmath20 color system , the majority of stars are on the redder sequence
. we will explore and quantify this effect below .
the presence of a split ms is even more evident from the two - color diagram shown in the right panels .
we have separated the two sequences into a msa ( green colors in the right upper panel of fig .
[ cmds ] ) , and a msb ( magenta colors ) .
the insets of left and middle panel of fig .
[ cmds ] show that the two mss appear to merge close to the turn - off , and that the turn - off / subgiant - branch part of the cmd is narrow and well defined , implying that any age difference between the two populations must be less than a few 100 myrs .
putting a better upper limit on the possible age spread will require a measurement of the overall c+n+o content for the stars in the two populations .
we note that the narrow magnitude distribution of sgb stars rules out the possibility that the double ms could be due to residual spatial variations of reddening that are beyond the sensitivity of the method that we used to correct them , as the reddening line is nearly perpendicular to the sgb sequence .
there is additional evidence that the ms splitting is real .
first of all , the two mss are visible everywhere in the field .
furthermore , reddening would move stars in the same direction in any cmds ( e. g. higher reddening would move stars towards redder colors and fainter magnitudes , at odds with the observed behaviour of the msa and msb , which invert their relative position in different cmds of fig .
[ cmds ] . )
the upper - left panel of figure [ m5 ] shows the distribution of msa and msb stars in the proper motion vector - point diagram , with the colors that were assigned in fig .
the two samples show no difference in their proper - motion distributions . to more quantitatively test
this , in the upper - right panel we show the cumulative distribution of proper motions .
a kolmogorov - smirnov test shows a vey high probability ( p=66% ) for the null hypothesis that the two populations have a similar proper - motion distributions the spatial distribution is plotted in the lower - left panel . in this case ,
as well , the kolmogorov - smirnov test indicates that there is no evidence for difference in the radial distributions ( p=67% ) . in order to estimate the fraction of stars in each ms we followed the procedure illustrated in fig .
[ msratio ] , which has already been used in several previous papers ( e. g. piotto et al .
the left panel shows the same @xmath21 vs. @xmath22 cmd of fig .
[ cmds ] , zoomed around the upper ms region , where the split is most evident .
the red line is the fiducial ridge line of msb .
to determine it , we started by selecting a sample of msb stars by means of a hand - drawn first - guess ridge line , and choosing the stars within a limited color range about this line .
we then calculated the median color and magnitude of msb stars in intervals of 0.2 magnitude in f336w , interpolated these median points with a spline , and did an iterated sigma - clipping of the straightened sequence . to obtain the straightened ms of the middle panel
, we subtracted from the color of each star the color of the fiducial sequence at the f336w magnitude of the star .
the color distribution of the points in the middle panel were analysed in four magnitude bins over the interval [email protected] .
the distributions have two clear peaks , which we fit with two gaussians ( magenta for the msb and green for the msa ) . from the areas under the gaussians ,
71@xmath243% of stars turns out to belong to the msb , and @xmath25% to the msa . within the statistical uncertainties
these fractions are the same in all magnitude intervals .
since we have photometry through five different @xmath26 filters , we can study the relative location of the two mss of ngc 6397 in a variety of cmds constructed using all possible combinations of the photometric bands , as shown in fig .
[ multiwavea ] .
the upper panels show the cmds @xmath17 vs. @xmath27 or @xmath28 ( where x represents f225w , f336w , f435w , f606w , f814w ) .
we used the same color code to plot the msa and msb stars identified in the color - color diagram in fig .
the msa has a @xmath29 color that is redder than the msb , but the relative color is opposite of this in all the other @xmath28 colors .
the lower panels of fig .
[ multiwavea ] show the @xmath30 vs. @xmath31 cmds . here , the msa is bluer than the msb , with the exception of the already mentioned @xmath30 vs. @xmath32 cmd .
the separation of the two sequences increases for larger color baselines in the remaining cmds . to better visualize the relative positions of the two mss , in fig .
[ multiwaveb ] we plotted the fiducials of the msa and the msb derived for each cmd of fig .
[ multiwavea ] following the same procedure above described .
in this paper we have shown that the ms of ngc 6397 is clearly split into two distinct sequences . broadened or multiple
mss have been observed in other clusters , including 47 tuc , @xmath3 centauri , ngc 2808 , and ngc 6752 ( anderson 1997 , bedin et al .
2004 , piotto et al .
2007 ) . in @xmath3
centauri and ngc 2808 the multiple mss have been associated with stellar populations that have large difference in their helium abundance , with the bluer mss having a higher he - content than the redder ones ( norris 2004 , piotto et al .
2005 , dantona et al .
stellar evolutionary models predict that h - burning at high temperatures through the cno cycle results in an enhanced production of he .
in addition to he , such models also predict enhanced production of n , and na and depletion of c , and o. this pattern of enhancement / depletion has recently been confirmed among the blue and red ms stars in ngc 2808 by bragaglia et al .
( 2010 ) . since the seventies
, spectroscopic studies of ngc 6397 giants have shown that the rgb stars exhibit a large spread in their c and n abundances ( bell et al .
significant star - to - star variations in the na content have also been detected among unevolved to stars ( gratton et al .
2001 , lind et al .
2009 ) , thus demonstrating that such abundance variations must have an intrinsic ( primordial ) rather than an evolutionary origin .
ngc 6397 stars also show na - o anticorrelation ( carretta et al .
2005 , 2009 , l11 ) .
even more relevant for the present discussion , l11 analysed the chemical composition of a large number of elements in 21 rgb stars of ngc 6397 , and found that na and o abundances have a bimodal distribution , with about the 75% of stars being enriched in na and depleted in o , while the remaining @xmath025% of the stars have a abundance patterns similar to field stars .
similarly , carretta et al .
( 2009 ) found that 30% of the stars in their sample are na - poor / o - rich .
l11 also demonstrated that the rgb of ngc 6397 is bimodal in the strmgren color index @xmath33 , with the na - poor and na - rich stars populating the blue and red rgb sequences , respectively .
it is interesting to note that , in sect .
3 we found that @xmath34% of the ms stars populate the msb .
this fraction is consistent with the 75% of na - rich , o - poor stars found by l11 .
therefore , we suggest a connection between the msb and the na - rich , o - poor population of ngc 6397 .
this would imply that the msa represents the first generation , and its intermediate - mass stars polluted the cluster with gas rich in he , n , and na to make the second generation , msb . in an effort to quantify the observed ms split , we have characterized the fiducial sequences by measuring the color difference between the msa and msb at the reference magnitude of @xmath35 ( fig . 3 ) .
for an assumed apparent distance modulus of @xmath36 = 12.2 , this corresponds to an absolute magnitude of @xmath37 = 5.7 . to compare these observations against expectations from synthetic photometry
, we adopted the basti isochrones ( pietrinferni et al .
2004 ) for the populations listed in table 1 , and determined @xmath38 and @xmath39 for ms stars at @xmath40 .
we adopted the average c , n , and o abundance of n - poor stars as measured by carretta et al .
( 2005 ) for dwarf and sgb stars in ngc 6397 . these temperatures and gravities ( listed in table 1 )
were then used to calculate model atmospheres with the atlas12 code ( kurucz 2005 , castelli 2005 , sbordone 2005 ) , which allowed us to use specific chemical compositions .
we then used the synthe code ( sbordone et al .
2007 ) to synthesize the spectrum from 1000 to 10000 and the resulting synthetic spectra were integrated over the transmission curve of each of our filters to produce the synthetic magnitudes and colors .
we did this separately for an msa star , using the composition listed in table 1 , and for an and msb star , using the three different composition options given in table 1 .
the left - panel of fig .
[ models ] shows a comparison of the observed @xmath41 color differences between msa and msb against the synthetic ones .
the blue squares indicate the color differences corresponding to option i , where we assumed for the two mss the same c , n , o mixture , but different he content .
there is a good agreement with the observed color differences in most bands , but a signficant disagreement with the f336w band .
we conclude that helium alone can not account for the observed ms split .
it is worth noting that nitrogen mainly affects the f336w photometry via the nh band around @xmath42 ( e. g. marino et al .
2008 ) . in option
ii , we assume that the msb stars in ngc 6397 are n - enhanced , but have the same he content .
we assumed for the msb the average c , n , and o abundance found by carretta et al .
( 2005 ) for n - rich stars .
the colors that result from option ii are plotted as gray triangles in fig .
[ models ] .
there remains a significant discrepancy between the simulated and the observed color differences in the f336w and f225w bands , now in the opposite sense to that seen in option i. turning instead to option iii , with differences in both helium and the cno elements , we see that the red asterisks in fig .
[ models ] are in better agreement with all of the observed color differences , yet it is still not perfect for the f336w filter . a careful fine - tuning of the composition differences between the two mss might remove the residual discrepancies between observed and theoretical points , but it is unclear whether that would tell us what is really going on without high - resolution spectra .
the difference between the synthetic spectra of a msa and a msb stars as calculated for option iii is shown in the upper - right panel of fig .
[ models ] , while the band - passes of our filters are plotted in the bottom - right panel . in conclusion ,
the observed color differences between the two mss are consistent with the presence of two populations with different helium and light - element content .
specifically , the msa would correspond to the first stellar generation with primordial he , and o - rich / n - poor stars , while the msb would be made of stars enriched in he and n but depleted in o. while the bluer mss of @xmath3 centauri and ngc 2808 imply extremely high he abundances ( up to [email protected] ) , the ms of ngc 6397 is consistent with a significantly smaller helium enhancement of second - generation stars at a level of @xmath43=0.01 , as already suggested by di criscienzo et al .
note that we reached similar conclusions , based on the same technique described above , for 47 tuc ( milone et al .
2011a ) where we found a @[email protected] .
finally , we recall that light - element variations have been detected in all gcs studied to date ( see gratton et al .
2004 for a review ) .
if the interpretation of the split ms of ngc 6397 given above is correct , then either multiple or spread mss should be observed in most ( all ? ) gcs . in this paper ( and in a similar paper on 47 tuc ) , we have demonstrated that the combination of uv and visual photometry , such as @xmath44 , @xmath45 , and @xmath46 , is a powerful tool to isolate stellar populations in the color - magnitude and color - color diagrams , allowing us to infer their he and cno content .
g.p . and s.c .
acknowledge partial support by prin inaf formation and early evolution of massive star clusters and by asi under the program asi - inaf i/016/07/0 .
acknowledge partial support by the universita di padova with the progetto di ateneo `` multiple stellar populations in star clusters '' .
j.a . acknowledges the support of stsci grant go-11233 . | high - precision multi - band _ hst _ photometry reveals that the main sequence ( ms ) of the globular cluster ngc 6397 splits into two components , containing @xmath030% and @xmath070% of the stars .
this double sequence is consistent with the idea that the cluster hosts two stellar populations : ( _ i _ ) a primordial population that has a composition similar to field stars , and containing @xmath030% of the stars , and ( _ ii _ ) a second generation with enhanced sodium and nitrogen , depleted carbon and oxygen , and a slightly enhanced helium abundance ( @[email protected] ) .
we examine the color difference between the two sequences across a variety of color baselines and find that the second sequence is anomalously faint in @xmath2 .
theoretical isochrones indicate that this could be due to nh depletion . |
covering theorems are known to be among some of the fundamental tools of measure theory .
they reflect the geometry of the space and are commonly used to establish connections between local and global behavior of measures .
covering theorems and their applications have been studied for example in @xcite and @xcite .
there are several types of covering results , all with the same purpose : from an arbitrary cover of a set in a metric space , one extracts a subcover as disjointed as possible .
we will consider more specifically here the so - called besicovitch covering property ( bcp ) which originates from the work of besicovitch ( @xcite , @xcite , see also @xcite , @xcite , @xcite ) in connection with the theory of differentiation of measures .
see subsection [ subsection : bcp ] for a more detailed presentation of the besicovitch covering property and its applications .
the geometric setting in which we are interested is the setting of carnot groups equipped with so - called homogeneous distances , and more specifically here the heseinberg groups @xmath0 . our main result in this paper , theorem [ thm : main ] , is the fact that bcp holds for the homogeneous distances on @xmath0 whose unit ball centered at the origin coincides with an euclidean ball centered at the origin .
these distances are bi - lipschitz equivalent to any other homogeneous distance on @xmath0 and in particular to the commonly used cygan - kornyi ( also usually called kornyi or gauge ) and carnot - carathodory distances .
recall that two distances @xmath1 and @xmath2 are said to be bi - lipschitz equivalent if there exists @xmath3 such that @xmath4 .
cygan - kornyi and carnot - carathodory distances are known not to satisfy bcp ( @xcite , @xcite , @xcite ) .
we indeed stress that the validity of bcp depends strongly on the distance the space is endowed with , and more specifically on the geometry of its balls . to put some more evidence on this fact and to put our result in perspective
, we also prove in the present paper two criteria that imply the non - validity of bcp .
they give two large families of homogeneous distances on @xmath0 that do not satisfy bcp and show that in some sense our example for which bcp holds is sharp .
see section [ section : distancesw / obcp ] , theorem [ thm : nobcpingoingcorners ] and theorem [ thm : nobcpoutgoingcorners ] . as a matter of fact
, our first criterion applies to the cygan - kornyi and to the carnot - carathodory distance , thus giving also new geometric proofs of the failure of bcp for these distances .
it also applies to the so - called box - distance ( the terminology might not be standard although this distance is a standard homogeneous distance on @xmath0 , see ) thus proving the non - validity of bcp for this latter homogeneous distance as well .
going back to the distances considered in the present paper and for which we prove that bcp holds , hebisch and sikora showed in @xcite that in any carnot group , there are homogeneous distances whose unit ball centered at the origin coincides with an euclidean ball centered at the origin with a small enough radius . in the specific case of the heisenberg groups , these distances are related to the cygan - kornyi distance .
they can indeed be expressed in terms of the quadratic mean of the cygan - kornyi distance ( at least for some specific value of the radius of the euclidean ball which coincides with the unit ball centered at the origin ) together with the pseudo - distance on @xmath0 given by the euclidean distance between horizontal components .
these distances have been previously considered in the literature .
lee and naor proved in @xcite that these metrics are of negative type on @xmath0 .
recall that a metric space @xmath5 is said to be of negative type if @xmath6 is isometric to a subset of a hilbert space .
combined with the work of cheeger and kleiner @xcite about weak notion of differentiability for maps from @xmath0 into @xmath7 , which leads in particular to the fact that @xmath0 equipped with a homogeneous distance does not admit a bi - lipschitz embedding into @xmath7 , this provides a counterexample to the goemans - linial conjecture in theoretical computer science , which was the motivation for these papers .
let us remark that the cygan - kornyi distance is not of negative type on @xmath0 .
we refer to subsection [ subsection : heisenberg ] for the precise definition of our distances and their connection with the cygan - kornyi distance and the distances of negative type considered in @xcite .
let @xmath5 be a metric space .
when speaking of a ball @xmath8 in @xmath9 , it will be understood in this paper that @xmath8 is a closed ball and that it comes with a fixed center and radius ( although these in general are not uniquely determined by @xmath8 as a set ) .
thus @xmath10 for some @xmath11 and some @xmath12 where @xmath13 .
one says that the besicovitch covering property ( bcp ) holds for the distance @xmath1 on @xmath9 if there exists an integer @xmath14 with the following property .
let @xmath15 be a bounded subset of @xmath5 and let @xmath16 be a family of balls in @xmath5 such that each point of @xmath15 is the center of some ball of @xmath16 .
then there is a subfamily @xmath17 whose balls cover @xmath15 and such that every point in @xmath9 belongs to at most @xmath18 balls of @xmath19 , that is , @xmath20 where @xmath21 denotes the characteristic function of a set @xmath15 .
when equipped with a homogeneous distance , the heisenberg groups turn out to be doubling metric spaces . recall that this means that there exists an integer @xmath22 such that each ball with radius @xmath12 can be covered with less than @xmath23 balls with radius @xmath24 .
when @xmath5 is a doubling metric space , bcp turns out to be equivalent to a covering property , strictly weaker in general , that we call the weak besicovitch covering property ( w - bcp ) ( the terminology might not be standard ) and with which we shall work in this paper .
first , let us fix some more terminology with the following definition .
[ besicovitch balls ] we say that a family @xmath16 of balls in @xmath5 is a _ family of besicovitch balls _ if @xmath25 is a finite family of balls such that @xmath26 for all @xmath8 , @xmath27 , @xmath28 , and for which @xmath29
. one says that the weak besicovitch covering property ( w - bcp ) holds for the distance @xmath1 on @xmath9 if there exists an integer @xmath14 such that @xmath30 for every family @xmath16 of besicovitch balls in @xmath5 .
the validity of bcp implies the validity of w - bcp .
we stress that there exists metric spaces for which w - bcp holds although bcp is not satisfied .
however , when the metric is doubling , both covering properties turn out to be equivalent as stated in characterization [ characterization : equivalentbcp ] below .
this characterization can be proved following the arguments of the proof of theorem 2.7 in @xcite .
[ characterization : equivalentbcp ] let @xmath5 be a doubling metric space . then bcp holds for the distance @xmath1 on @xmath9 if and only if w - bcp holds for the distance @xmath1 on @xmath9 . as already said , covering theorems and especially the besicovitch covering property and the weak besicovitch covering property play an important role in many situations in measure theory , regularity and differentiation of measures , as well as in many problems in harmonic analysis .
this is particularly well illustrated by the connection between w - bcp and the so - called differentiation theorem .
the validity of bcp in the euclidean space is due to besicovitch and was a key tool in his proof of the fact that the differentiation theorem holds for each locally finite borel measure on @xmath31 ( @xcite , @xcite , see also @xcite , @xcite ) .
moreover , as emphasized in theorem [ thm : diff ] , the validity of w - bcp turns actually out to be equivalent to the validity of the differentiation theorem for each locally finite borel measure as shown in @xcite .
* preiss ) [ thm : diff ] let @xmath5 be a complete separable metric space .
then the differentiation theorem holds for each locally finite borel measure @xmath32 on @xmath5 , that is , @xmath33 for @xmath32-almost every @xmath11 and for each @xmath34 if and only if @xmath35 where , for each @xmath36 , w - bcp holds for family of balls centered on @xmath37 with radii less than @xmath38 for some @xmath39 .
as already stressed , the fact that bcp holds in a metric space depends strongly on the distance with which the space is endowed . on the one hand , with very mild assumptions on the metric space ( namely , as soon as there exists an accumulation point ) , one can indeed always construct bi - lipschitz equivalent distances as close as we want from the original distance and for which bcp is not satisfied , as shown in the following result .
[ thm : destroybcp ] let @xmath5 be a metric space .
assume that there exists an accumulation point in @xmath5 .
let @xmath40 .
then there exists a distance @xmath2 on @xmath9 such that @xmath41 and for which w - bcp , and hence bcp , do not hold .
a slightly different version of this result is stated in theorem 3 of @xcite . for sake of completeness
, we give in section [ section : destroybcp ] a construction of such a distance as stated in theorem [ thm : destroybcp ] . on the other hand , the question whether a metric space can be remetrized so that bcp holds is in general significantly more delicate .
as already explained , the main result of the present paper , theorem [ thm : main ] , is a positive answer to this question for the heisenberg groups equipped with ad - hoc homogeneous distances , namely those whose unit ball at the origin coincides with an euclidean ball with a small enough radius . as a set we identify the heisenberg group @xmath0 with @xmath42 and we equip it as a topological space with the euclidean topology .
we choose the following convention for the group law @xmath43 where @xmath44 , @xmath45 , @xmath46 and @xmath47 belong to @xmath48 , @xmath49 and @xmath50 belong to @xmath51 and @xmath52 denotes the usual scalar product in @xmath31 .
this corresponds to a choice of exponential and homogeneous coordinates .
the one parameter family of dilations on @xmath0 is given by @xmath53 where @xmath54 these dilations are group automorphisms .
[ def : homogeneousdistance ] a distance @xmath1 on @xmath0 is said to be _ homogeneous _ if the following properties are satisfied .
first , it induces the euclidean topology on @xmath0 .
second , it is left invariant , that is , @xmath55 for all @xmath56 , @xmath57 , @xmath58 . and third , it is one - homogeneous with respect to the dilations , that is , @xmath59 for all @xmath56 , @xmath60 and all @xmath61 .
it turns out that homogeneous distances on @xmath0 do exist in abundance and make it a doubling metric space .
it is also well known that any two homogeneous distances are bi - lipschitz equivalent .
see for example @xcite for more details about the heisenberg groups and more generally carnot groups .
the ( family of ) homogeneous distance(s ) we consider in this paper can be defined in the following way . for @xmath62 , we denote by @xmath63 the euclidean ball in @xmath64 centered at the origin with radius @xmath65 , that is , @xmath66 where @xmath67 denotes the euclidean norm in @xmath68 and we set @xmath69 hebisch and sikora proved in @xcite that if @xmath62 is small enough , then @xmath70 actually defines a distance on @xmath0 .
more generally this holds true in any carnot group starting from the set @xmath63 given by the euclidean ball centered at the origin with radius @xmath62 small enough , where one identifies in the usual way the group with some @xmath71 where @xmath72 is its topological dimension .
it then follows from the very definition that @xmath70 turns out to be the homogeneous distance on @xmath0 for which the unit ball centered at the origin coincides with the euclidean ball with radius @xmath65 centered at the origin .
the geometric description of arbitrary balls that can then be deduced from the unit ball centered at the origin via dilations and left - translations is actually of crucial importance for understanding the reasons why bcp eventually holds for these distances .
on the other hand , it is particularly convenient to note that in the specific case of the heisenberg groups , one also has a fairly simple analytic expression for such distances whose unit ball at the origin is given by an euclidean ball centered at the origin . this will actually be technically extensively used in our proof of theorem [ thm : main ] .
this also gives the explicit connection with the cygan - kornyi distance and the distances of negative type considered by lee and naor in @xcite .
set @xmath73 for @xmath74 .
then one has @xmath75 see section [ section : preliminaries ] .
first , note that @xmath76 is a left - invariant pseudo - distance on @xmath0 that is one - homogeneous with respect to the dilations .
next , when @xmath77 , @xmath78 is nothing but the cygan - kornyi norm which is well known to be a natural gauge in @xmath0 .
it can actually be checked by direct computations that @xmath79 satisfies the triangle inequality for any @xmath80 and hence defines a homogeneous distance on @xmath0 .
this was first proved by cygan in @xcite when @xmath77 .
one then recovers from the analytic expression that @xmath70 actually defines a homogeneous distance on @xmath0 for any @xmath80 , giving also an explicit range of values of @xmath65 in @xmath0 for which this fact holds and was first observed in @xcite for general carnot groups and for small enough values of @xmath65 . for any @xmath81 , @xmath70 defines a homogeneous distance on @xmath0 .
note that there might be other values of @xmath82 such that @xmath70 defines a homogeneous distance on @xmath0 .
these distances turn out to be those considered by lee and naor in @xcite .
the authors actually proved in @xcite that @xmath83 is of negative type in @xmath0 to provide a counterexample to the so - called goemans - linial conjecture .
let us mention that it can easily be checked that the proof in @xcite extend to the distances @xmath70 for all @xmath81 .
let us now state our main result .
[ thm : main ] let @xmath62 be such that @xmath70 defines a homogeneous distance on @xmath0
. then bcp holds for the homogeneous distance @xmath70 on @xmath0 . for technical and notational simplicity
, we will focus our attention on the first heisenberg group @xmath84 .
we shall point out briefly in section [ section : hn ] the non - essential modifications needed to make our arguments work in any heisenberg group @xmath0 .
the rest of the paper is organized as follows . in section [ section : preliminaries ]
we fix some conventions about @xmath85 and the distance @xmath70 and state three technical lemmas on which the proof of theorem [ thm : main ] is based .
the proof of these lemmas is given in sections [ section : prooflemma : axis ] and [ section : prooflemma : comparisonincone ] .
section [ section : proofmainthm ] is devoted to the proof of theorem [ thm : main ] itself . in section [ section : distancesw / obcp ] we prove two criteria , theorem [ thm : nobcpingoingcorners ] and theorem [ thm : nobcpoutgoingcorners ] , for homogeneous distances on @xmath85 that imply that bcp does not hold . theorem [ thm : destroybcp ] is proved in section [ section : destroybcp ] .
as already stressed we will focus our attention in sections [ section : preliminaries ] to [ section : distancesw / obcp ] on the first heisenberg group @xmath84 for technical and notational simplicity .
the modifications needed to handle the case of @xmath0 for any @xmath86 will be indicated in section [ section : hn ] .
we first fix some conventions and notations .
next , we will conclude this section with the statement of the main lemmas on which the proof of theorem [ thm : main ] will be based .
recall that we identify the heisenberg group @xmath85 with @xmath87 equipped with the group law given in and we equip it with the euclidean topology .
we define the projection @xmath88 by @xmath89 when considering a specific point @xmath90 , we shall usually denote by @xmath91 its coordinates and we set @xmath92 from now on in this section , as well as in sections [ section : proofmainthm ] , [ section : prooflemma : axis ] and [ section : prooflemma : comparisonincone ] , we fix some @xmath93 such that @xmath94 as given in defines a homogeneous distance on @xmath85 . thus all metric notions and properties will be understood in these sections relatively to this fixed distance @xmath94 . in particular we shall denote the closed balls with center @xmath90 and radius @xmath12 by @xmath95 without further explicit reference to the distance @xmath94 with respect to which they are defined . remembering , we have the following properties . for @xmath96
, we have @xmath97 and @xmath98 from which we get @xmath99 for a point @xmath90 , we shall set @xmath100 using left - translations , we have the following properties for any two points @xmath101 , @xmath102 and @xmath103 where @xmath104 and @xmath105 by definition of the group law . note that if @xmath106 then @xmath107 .
let us point out that balls in @xmath108 are convex in the euclidean sense when identifying @xmath85 with @xmath87 with our choosen coordinates .
indeed , the unit ball centered at the origin is by definition the euclidean ball with radius @xmath65 in @xmath109 and thus is euclidean convex .
next , dilations are linear maps and left - translations ( see ) are affine maps , hence @xmath110 is also an eucliden convex set in @xmath109 .
this will be of crucial use for some of our arguments in the sequel and we state it below as a proposition for further reference . [ prop : ball_convex ] balls in @xmath108 are convex in the euclidean sense when identifying @xmath85 with @xmath87 with our choosen coordinates .
we shall also use the following isometries of @xmath108 .
first , rotations around the @xmath49-axis are defined by @xmath111 for some angle @xmath112 .
next , the reflection @xmath113 is defined by @xmath114 using , one can easily check that these maps are isometries of @xmath108 .
we state now the main lemmas on which the proof of theorem [ thm : main ] will be based . for @xmath115 , @xmath116 and @xmath117 ,
we set ( see figure [ fig1 ] ) @xmath118 [ fig : p ] .,title="fig:",height=226 ] [ fig : c_theta ] .,title="fig:",height=226 ] [ lemma : x : axis0 ] there exists @xmath119 , which depends only on @xmath120 , such that for all @xmath121 , there exists @xmath122 such that for all @xmath123 and for all @xmath124 , the following holds .
let @xmath125 and @xmath126 be such that @xmath127 and @xmath128 . then at most one of these two points belongs to @xmath129 . for @xmath116 and @xmath117 , we set ( see figure [ fig2](a ) ) @xmath130 [ fig : t ] and @xmath131.,title="fig:",height=264 ] [ fig : c ] and @xmath131.,title="fig:",height=226 ] [ lemma : z : axis0 ] there exists @xmath132 and @xmath133 , depending only on @xmath120 , such that for all @xmath134 and all @xmath135 , the following holds .
let @xmath125 and @xmath126 be such that @xmath127 and @xmath128 . then at most one of these two points belongs to @xmath136 .
these two lemmas will be proved in section [ section : prooflemma : axis ] .
for @xmath137 , we set ( see figure [ fig2](b ) ) @xmath138 [ lemma : comparisonincone ] there exists @xmath139 , which depends only on @xmath120 , such that for all @xmath140 the following holds .
let @xmath90 and @xmath141 be such that @xmath142 then we have @xmath143 and @xmath144 this lemma will be proved in section [ section : prooflemma : comparisonincone ] .
this section is devoted to the proof of theorem [ thm : main ] . recall that we consider here the case @xmath145 equipped with a homogeneous distance @xmath94 as defined in ( see section [ section : hn ] for the general case @xmath0 , @xmath86 ) . recall also that due to characterization [ characterization : equivalentbcp ] , theorem [ thm : main ] will follow if we find an integer @xmath14 such that @xmath30 for every family @xmath16 of besicovitch balls .
see definition [ besicovitch balls ] for the definition of a family of besicovitch balls .
we first reduce the proof to the case of some specific families of besicovitch balls . in what follows , when considering families of points @xmath146 we shall simplify the notations and set @xmath147 , @xmath148 and @xmath149 . recall that @xmath150 is defined in .
[ lemma : reduction ] let @xmath151 and let @xmath152 be a family of besicovitch balls .
then there exists a finite family of points @xmath146 such that @xmath153 is a family of besicovitch balls with the following properties . for every point @xmath154 in the family ,
we have @xmath155 and @xmath156 let @xmath157 be a family of besicovitch balls where @xmath158 .
take @xmath159 . set @xmath160
remembering that left - translations are isometries and that , by convention , we set @xmath149 , we get that @xmath161 and @xmath162 hence @xmath163 is a family of besicovitch balls .
since balls are euclidean convex ( see proposition [ prop : ball_convex ] ) and since @xmath164 for all @xmath165 , there are at most two balls in @xmath166 with their center on the @xmath49-axis . next , up to replacing the family @xmath146 by @xmath167 ( see for the definition of the reflection @xmath113 ) and up to re - indexing the points
, one can find @xmath168 points @xmath169 that satisfy , such that @xmath170 ( see for the definition of the projection @xmath171 ) , and with @xmath172 .
finally , up to a rotation around the @xmath49-axis ( see for the definition of rotations ) and up to re - indexing the points , we get by the pigeonhole principle that there exists an integer @xmath173 such that @xmath174 and such that @xmath154 satisfies for all @xmath175 . then the family @xmath176 gives the conclusion .
we are now ready to conclude the proof of theorem [ thm : main ] using lemma [ lemma : x : axis0 ] , lemma [ lemma : z : axis0 ] and lemma [ lemma : comparisonincone ] .
* _ proof of theorem [ thm : main ] . _ * we fix some values of @xmath177 , @xmath116 , and @xmath117 so that the conclusions of lemma [ lemma : x : axis0 ] , lemma [ lemma : z : axis0 ] and lemma [ lemma : comparisonincone ] hold . next , we fix some @xmath178 large enough so that @xmath179 , \;|z_p|<b,\ ; |y_p|<x_p \tan \theta\ } \subset u(0,r)\ ] ] and @xmath180,\ ; \rho_p < b\ } \subset u(0,r),\ ] ] where @xmath181 denotes the open ball with center 0 and radius @xmath182 in @xmath183 .
such an @xmath182 exists since in the above two inclusions , the sets on the left are bounded . as a consequence , we have @xmath184 and @xmath185 recall for the definition of @xmath186 and for the definition of @xmath136 .
let us now consider a family of besicovitch balls @xmath187 where , as defined by convention , we have @xmath149 and where the centers @xmath154 satisfy and . noting that the family @xmath188 also satisfies the same properties for all @xmath189 , one can assume with no loss of generality that @xmath190 up to a dilation by a factor @xmath191 .
let @xmath192 and @xmath193 be defined as @xmath194 and @xmath195 we will bound @xmath196 in terms of the constants @xmath72 , @xmath9 , @xmath197 and @xmath198 .
we re - index the points so that @xmath199 let @xmath200 be such that @xmath201 . by choice of @xmath168 and by definition of @xmath72 and @xmath9 ,
we have @xmath202 let @xmath203 be a large enough integer such that @xmath204 .
then we have @xmath205 .
indeed , otherwise we would get from in lemma [ lemma : comparisonincone ] that @xmath206 and hence @xmath207 . then , by choice of @xmath182 ( remember ) , @xmath208 and @xmath209 would be distinct points in @xmath136 which contradicts lemma [ lemma : z : axis0 ] .
let @xmath210 be a large enough integer such that @xmath211 .
then we have @xmath212 .
indeed , otherwise we would get from in lemma [ lemma : comparisonincone ] that @xmath213 and hence @xmath214 . then , by choice of @xmath182 ( remember ) , @xmath215 and @xmath216 would be distinct points in @xmath186 which contradicts lemma [ lemma : x : axis0 ] . all together
we get the following bound on @xmath217 , @xmath218 combining this with in lemma [ lemma : reduction ] , we get the following bound on the cardinality of arbitrary families @xmath152 of besicovitch balls , @xmath219 which concludes the proof of theorem [ thm : main ] .
this section is devoted to the proof of lemma [ lemma : x : axis0 ] and lemma [ lemma : z : axis0 ] .
we begin with a remark that will be technically useful . given @xmath90 and @xmath141
, we set @xmath220 recall that , following , we have @xmath221 by convention .
[ a(p , q ) ] we have @xmath222 if and only if @xmath223 recalling , we have @xmath224 combining this with , which gives @xmath225 we get the conclusion . [ lemma : pp ] there exist constants @xmath226 and @xmath227 , depending only on @xmath120 , such that , for all @xmath228 , all @xmath116 and @xmath117 such that @xmath229 , we have @xmath230 for all @xmath231 .
by , we always have @xmath232 . on the other hand ,
we can bound from above @xmath233 using that @xmath234 , since @xmath235 , and that @xmath236 if @xmath231 ( see for the definition of @xmath129 ) .
namely , we have @xmath237 for @xmath238 , @xmath117 and @xmath115 , we set ( see figure [ fig3](a ) ) @xmath239 [ fig : r_in_p ] and @xmath240.,title="fig:",height=245 ] [ fig : r ] and @xmath240.,title="fig:",height=207 ] [ lemma : x : axis ] there exists @xmath119 , which depends only on @xmath120 , such that for all @xmath241 , there exists @xmath242 such that for all @xmath123 and for all @xmath124 , we have @xmath243 for all @xmath244 and all @xmath245 $ ] .
take @xmath228 , @xmath246 , @xmath244 , @xmath247 and consider @xmath248 . by lemma [ a(p , q ) ] , showing that @xmath249 is equivalent to prove that @xmath250 is negative .
since @xmath251 , we have @xmath252 note that all terms in the last inequality are positive except @xmath253 , since both @xmath254 and @xmath255 are positive .
we now use the conditions @xmath256 , @xmath257 , @xmath258 , @xmath259 , @xmath260 , @xmath261 and @xmath262 , since @xmath263 , to get @xmath264 we consider now separately the case @xmath265 and @xmath266 . for @xmath265 , we bound using lemma [ lemma : pp ] @xmath267 hence @xmath268 as @xmath255 goes to infinity .
thus , choosing @xmath198 small enough so that @xmath269 , we get that @xmath270 provided @xmath255 is large enough . for @xmath266
, we use once again lemma [ lemma : pp ] and get @xmath271 hence @xmath272 as @xmath255 goes to infinity .
thus , choosing @xmath198 small enough so that @xmath273 , we get that @xmath270 provided @xmath255 is large enough .
all together we have showed that one can find @xmath274 , depending only on @xmath65 , and for all @xmath275 , some @xmath242 , such that for @xmath123 and @xmath261 and for all @xmath276 , we have @xmath277 since @xmath278 is euclidean convex by proposition [ prop : ball_convex ] , we conclude the proof noting that @xmath279 , for @xmath280 $ ] , is in the euclidean convex hull of @xmath281 and @xmath282 .
_ * proof of lemma [ lemma : x : axis0 ] . *
_ let @xmath119 be given by lemma [ lemma : x : axis ] .
let @xmath283 and let @xmath242 be given by lemma [ lemma : x : axis ] .
let @xmath123 and @xmath124 .
let @xmath90 and @xmath141 be such that @xmath127 and @xmath284 .
let us assume with no loss of generality that @xmath285 . then
, if both @xmath56 and @xmath57 were in @xmath186 , by lemma [ lemma : x : axis ] we would have @xmath286 since @xmath287 $ ] . but this would contradict the assumptions . [ lemma : tt ] let @xmath288 and @xmath117 .
then for all @xmath289 , we have @xmath290 let @xmath289 ( see for the definition of @xmath136 ) .
since @xmath291 and @xmath292 , we have ( recall ) @xmath293 for @xmath294 and @xmath117 , we set ( see figure [ fig3](b ) ) @xmath295 [ lemma : z : axis ] there exists @xmath132 and @xmath133 , depending only on @xmath120 , such that for all @xmath134 and all @xmath135 , we have @xmath296 for all @xmath297 and all @xmath298 $ ] .
take @xmath246 , @xmath297 , @xmath299 and consider @xmath300 . by lemma [ a(p , q ) ] , showing that @xmath301 is equivalent to prove that @xmath250 is negative .
since @xmath302 , we have @xmath303 note that all terms in the last inequality are positive except @xmath304 , assuming both @xmath254 and @xmath305 negative .
we bound using lemma [ lemma : tt ] and using that the absolute value of each of the first two components of @xmath56 and @xmath57 is smaller than @xmath197 , @xmath306 where in the last inequality we assumed that @xmath197 is small enough , @xmath307 for some @xmath308 which depends only on @xmath65 .
we consider now separately the case @xmath309 and @xmath310 . for @xmath309 , we need @xmath311 which is true as soon as @xmath312 . for @xmath310 , we need @xmath313 which is true as soon as @xmath314 is large enough .
all together we showed that one can find @xmath315 and @xmath316 , depending only on @xmath65 , such that , for all @xmath134 and @xmath135 and all @xmath289 , we have @xmath317 recall that the set @xmath318 is euclidean convex by proposition [ prop : ball_convex ]
. therefore we conclude the proof since @xmath319 , for @xmath320 $ ] , is in the euclidean convex hull of @xmath321 and @xmath322 .
* _ proof of lemma [ lemma : z : axis0 ] . _ * let @xmath315 and @xmath316 be given by lemma [ lemma : z : axis ] .
let @xmath134 and @xmath135 .
let @xmath90 and @xmath323 be such that @xmath127 and @xmath284 . assume with no loss of generality that @xmath324 . then ,
if both @xmath56 and @xmath57 were in @xmath136 , by lemma [ lemma : z : axis ] we would have @xmath325 since @xmath326 $ ] . but
this would contradict the assumptions .
this section is devoted to the proof of lemma [ lemma : comparisonincone ] .
we first fix some notations . for @xmath327 , we set @xmath328 . for @xmath329 , @xmath90 and @xmath330 ,
let @xmath331 denote the two dimensional euclidean half cone in @xmath332 contained in the plane @xmath333 with vertex @xmath334 , axis the half line starting at @xmath334 and passing through @xmath335 and aperture @xmath336 .
see figure [ fig4](a ) . for @xmath329 , @xmath90 and @xmath330 ,
let @xmath337 denote the two dimensional euclidean equilateral quadrilateral contained in the plane @xmath333 with vertices @xmath334 , @xmath338 , @xmath339 and @xmath340 .
note that it is the euclidean convex hull in @xmath332 of these four points .
see figure [ fig4](b ) .
[ fig : c_thetha_p_z ] and @xmath341.,title="fig:",height=188 ] ( -37,66)@xmath342 ( -47,63.5)@xmath343 ( -115,117)@xmath56 ( -106.56,114)@xmath343 ( -94,46)@xmath198 [ fig : q ] and @xmath341.,title="fig:",height=207 ] ( -166,90)@xmath342 ( -15,86)@xmath344 ( -19,77)@xmath343 ( -95,126)@xmath345 ( -86.5,111)@xmath343 ( -104,84)@xmath335 ( -86.5,77)@xmath343 ( -95,29)@xmath346 ( -86.5,42.5)@xmath343 recall for the definition of @xmath150 .
note that @xmath347 if and only if @xmath348 .
we have the following properties , @xmath349 and @xmath350 for @xmath351 , we have @xmath352 this follows from elementary geometry noting that the angle between the half lines starting at @xmath345 and passing through @xmath334 and @xmath353 respectively is larger than @xmath354 .
[ lemma : sev1 ] there exists @xmath355 , which depends only on @xmath120 , such that @xmath356 for all @xmath357 , all @xmath358 and all @xmath327 such that @xmath359 .
recalling proposition [ prop : ball_convex ] , we only need to prove that the vertices @xmath334 , @xmath345 , @xmath346 and @xmath353 of @xmath337 belong to @xmath360 .
we have @xmath359 and , recalling and , @xmath361 hence @xmath362 that is , recalling , @xmath363 .
similarly we have @xmath364 hence @xmath365 . next ,
let us prove that @xmath366 .
set @xmath367 we need to prove that @xmath368 .
we have @xmath369 where the last inequality follows from the fact that @xmath370 which implies in particular that @xmath371 hence we get that @xmath372 choosing @xmath373 small enough so that @xmath374 for all @xmath357 , we get the conclusion .
the fact that @xmath375 is proved in a similar way .
* _ proof of lemma [ lemma : comparisonincone]_. * let @xmath376 where @xmath377 is given by lemma [ lemma : sev1 ] .
let @xmath378 and let @xmath90 and @xmath141 satisfying , , and .
let us first prove .
assume by contradiction that @xmath379
. then @xmath380 .
hence @xmath381 according to lemma [ lemma : sev1 ] . on the other hand
, it follows from , , , that @xmath382 and hence @xmath383 which contradicts
. thus we have @xmath384 and thus @xmath385 .
it follows from , and that @xmath386 . finally we get from lemma [ lemma : sev1 ] that @xmath387 and
then follows from .
in this section we prove two criteria which imply the non - validity of bcp .
this shows that in some sense our example of homogeneous distance @xmath70 for which bcp holds is sharp . roughly speaking the first criterion applies to homogeneous distances whose unit sphere centered at the origin has either inward cone - like singularities in the euclidean sense at the poles ( i.e. , at the intersection of the sphere with the @xmath49-axis ) or is flat at the poles with @xmath388 curvature in the euclidean sense .
the second one applies to homogeneous distances whose unit sphere at the origin has outward cone - like singularities in the euclidean sense at the poles .
note that the unit sphere centered at the origin of our distance @xmath70 is smooth with positive curvature in the euclidean sense .
let @xmath1 be a homogeneous distance on @xmath85 and let @xmath8 denote the closed unit ball centered at the origin in @xmath389 . in this subsection
we shall most of the time identify @xmath85 with @xmath87 equipped with its usual differential structure . for @xmath125 , @xmath390 , @xmath391 , and @xmath392 , let @xmath393 denote the euclidean half - cone in @xmath85 , identified with @xmath87 , with vertex @xmath56 , axis @xmath394 and opening @xmath395 .
we say that @xmath396 , @xmath391 , _ points out of @xmath8 at @xmath397 _ if there exists an open neighbourhood @xmath398 of @xmath56 and some @xmath392 such that @xmath399 let @xmath400 denote the left translation defined by @xmath401 . we consider it as an affine map from @xmath85 , identified with @xmath87 , to @xmath87 whose differential , in the usual euclidean sense in @xmath87 is thus a constant linear map and
will be denoted by @xmath402 .
let @xmath403 be defined by @xmath404 .
for @xmath405 , @xmath391 , and @xmath406 , let @xmath407 denote the set of points @xmath408 such that @xmath409 points out of @xmath8 at @xmath410 and let @xmath411 denote the set of points @xmath412 such that @xmath413 for some @xmath414 such that @xmath415 ( here @xmath416 denotes the euclidean norm in @xmath87 ) .
[ thm : nobcpingoingcorners ] assume that there exists @xmath417 , @xmath391 , and @xmath418 such that @xmath419 for all @xmath420 .
then bcp does not hold in @xmath389 .
we first construct a sequence of points @xmath421 in @xmath422 such that @xmath423 for all @xmath424 and @xmath425 points out of @xmath8 at @xmath426 for all @xmath86 and all @xmath427 . note that if @xmath412 then there exists @xmath428 such that @xmath429 points out of @xmath8 at @xmath410 for all @xmath430 such that @xmath431 ( note that the set of vectors that points out of @xmath8 at some point @xmath397 is open ) .
let us start choosing some @xmath432 .
by induction assume that @xmath433 have already been chosen .
let @xmath434 where each @xmath435 is associated to @xmath436 as above .
then we choose @xmath437 .
we have @xmath438 for some @xmath61 and some @xmath439 such that @xmath440 .
hence , by choice of @xmath441 and of the @xmath442 s , we have that @xmath443 points out of @xmath8 at @xmath426 for all @xmath444 as wanted .
next , we claim that if @xmath445 , @xmath446 are such that @xmath447 and that @xmath448 points out of @xmath8 at @xmath410 , then there exists @xmath449 such that @xmath450 for all @xmath451 .
indeed the curve @xmath452 is a smooth curve starting at @xmath410 and whose tangent vector at @xmath453 is given by @xmath448 . since this vector points out of @xmath8 at @xmath410 , it follows that @xmath454 for all @xmath189 small enough and hence @xmath455 as wanted .
then it follows that for all @xmath86 , one can find @xmath456 such that for all @xmath457 and all @xmath458 , one has @xmath459 then we set @xmath460 and by induction it follows that we can construct a decreasing sequence @xmath461 so that @xmath462 for all @xmath86 and all @xmath458 . for @xmath424 , we set @xmath463 . by construction
we have @xmath464 for all @xmath465 and @xmath424 such that @xmath466 .
it follows that @xmath467 is a family of besicovitch balls for any finite set @xmath468 and hence bcp does not hold .
let us give some examples of homogeneous distances for which the criterion given in theorem [ thm : nobcpingoingcorners ] applies . a first class of examples is given by rotationally invariant homogeneous distances @xmath1 that satisfy that there exists @xmath469 such that @xmath470 and such that @xmath471 by rotationally invariant distances , we mean distances for which rotations @xmath472 , @xmath473 , are isometries ( see for the definition of @xmath472 ) .
indeed , consider @xmath474 and , for @xmath475 , set @xmath476 then consider @xmath477 . by rotational and left invariance ( which implies in particular that @xmath478 for all @xmath141 ) , one has @xmath445 . on the other hand , since @xmath479 , any vector with a positive third coordinate points out of @xmath8 at @xmath410 .
in particular @xmath480 points out of @xmath8 at @xmath410 . hence @xmath481 .
this class of examples includes the so - called box - distance @xmath482 defined by @xmath483 with @xmath484 for which the fact that bcp does not hold was not known .
it also includes the carnot - carathodory distance and hence this gives a new proof of the non - validity of bcp for this distance .
see @xcite for a previous and different proof .
other examples of homogeneous distances @xmath1 for which the criterion given in theorem [ thm : nobcpingoingcorners ] applies can be obtained in the following way .
assume that @xmath8 , respectively @xmath422 , can be described as @xmath485 , respectively @xmath486 , for some @xmath487 real valued function @xmath488 in a neighbourhood of a point @xmath397 .
then the outward normal to @xmath422 at some point @xmath445 is given in a neighbourhood of @xmath56 by @xmath489 ( here it is still understood that we identify @xmath85 with @xmath87 and @xmath490 denotes the usual gradient in @xmath87 ) .
then theorem [ thm : nobcpingoingcorners ] applies if one can find a vector @xmath491 , @xmath391 , such that for all @xmath492 small enough , the following holds .
there exists @xmath408 such that @xmath413 for some @xmath414 such that @xmath415 and such that @xmath410 lies in a neighbourhood of @xmath56 and @xmath493 where @xmath52 denotes the usual scalar product in @xmath87 .
a particular example is given when @xmath8 , respectively @xmath422 , can be described near the north pole ( intersection of @xmath422 with the positive @xmath49-axis ) as the subgraph @xmath494 , respectively the graph @xmath495 , of a @xmath496 function @xmath497 whose first and second order partial derivatives vanish at the origin .
indeed , in that case one can choose for example @xmath498 and for a fixed @xmath499 , one looks for some @xmath481 of the form @xmath500 for some @xmath189
. then @xmath501 lies near the north pole for @xmath189 small and we have @xmath502 that is equivalent to @xmath503 when @xmath189 is small enough .
hence @xmath504 .
this argument applies to the cygan - kornyi distance @xmath505 , and more generally to @xmath506 for all values of @xmath62 such that @xmath506 defines a distance , thus in particular for all values of @xmath507 .
recall from that @xmath508 where @xmath509 and that @xmath505 is the cygan - kornyi distance .
hence theorem [ thm : nobcpingoingcorners ] gives in particular a new geometric proof of the fact that bcp does not hold for the cygan - kornyi distance on @xmath85 , see @xcite and @xcite for previous analytic proofs .
let @xmath1 be a homogeneous distance on @xmath85 and let @xmath8 denote the closed unit ball centered at the origin in @xmath389 . set @xmath510 .
[ thm : nobcpoutgoingcorners ] assume that there exists two sequences of points @xmath511 and @xmath512 and some @xmath116 and @xmath513 such that @xmath514 then bcp does not hold in @xmath389 .
the geometric meaning of the above assumptions is the following . in some vertical plane ( here
we take the @xmath515-plane for simplicity ) one can find two sequences of points @xmath516 and @xmath517 , each one of them on a different side of the @xmath49-axis .
such points are on the unit sphere centered at the origin and are converging to the north pole .
the slope between @xmath517 and @xmath516 is assumed to be bounded away from zero .
we further assume that at the north pole the intersection of the sphere and the @xmath515-plane can be written both as graph @xmath518 and @xmath519 .
see figure [ out : cone ] .
-plane and the unit sphere at the origin of the distance @xmath520 when @xmath521 and @xmath77.,title="fig:",height=151 ] ( -86,102)@xmath516 ( -142,109.4)@xmath517 theorem [ thm : nobcpoutgoingcorners ] applies in particular if the intersection of @xmath8 with the @xmath515-plane can be described near the north pole as @xmath522 for some function @xmath488 of class @xmath487 on @xmath523 such that @xmath524 and @xmath525 exist and are finite with @xmath526 .
this is for instance the case of the following distances built from the cygan - kornyi distance , and more generally from the distances @xmath506 , and given by @xmath527 with @xmath528 for some @xmath529 .
see for the definition of @xmath530 and @xmath531 .
figure [ out : cone ] is exactly the intersection of the @xmath515-plane and the unit sphere at the origin when @xmath521 and @xmath77 .
note that it follows in particular that the @xmath532-sum of the pseudo - distance @xmath533 with the distance @xmath506 does not satisfy bcp in contrast with their @xmath534-sum which is a multiple of the distance @xmath70 .
_ * proof of theorem [ thm : nobcpoutgoingcorners ] . * _ by induction , we construct a sequence of points @xmath535 such that @xmath536 for all @xmath537 , where @xmath538 , and such that @xmath539 for all @xmath537 and all @xmath540 .
then , we will have @xmath541 for all @xmath542 and @xmath537 such that @xmath543 , so that @xmath544 is a family of besicovitch balls for any finite set @xmath468 .
hence bcp does not hold .
we start from a point @xmath545 with @xmath546 .
next assume that @xmath547 have been constructed and choose @xmath548 large enough so that @xmath549 we set @xmath550 note that @xmath551 since @xmath552 .
we have @xmath553 by choice of @xmath548 ( see ) .
we also have @xmath554 hence it remains to check that @xmath555 and that @xmath556 for @xmath540 .
using dilation , left translation and the assumption @xmath557 , it follows that @xmath558 hence , taking into account the fact that @xmath559 and that @xmath560 , to prove that @xmath555 and that @xmath556 for @xmath540 , we only need to check that @xmath561 , which follows from , and that @xmath562 . using the fact that @xmath563 , and , we have @xmath564 which gives the conclusion .
the case of @xmath0 for @xmath86 arbitrary can be easily handled similarly to the case of @xmath85 adopting the following convention . for @xmath565 , we set @xmath566 where @xmath567 , @xmath568 and @xmath569 .
note that this is different from the more standard presentation adopted in the introduction ( section [ section : introduction ] ) .
to avoid any confusion , the explicit correspondance between theses two conventions is the following . if @xmath570 , @xmath571 and @xmath330 denote the exponential and homogeneous coordinates of @xmath572 as in , by denoting @xmath566 with @xmath567 , @xmath568 and @xmath569 , we mean @xmath573 , @xmath574 and @xmath575 .
it follows that @xmath576 should be replaced by @xmath577 and @xmath578 by @xmath579 where @xmath580 denotes the euclidean norm in @xmath581 .
in particular , we get @xmath582 and setting @xmath583 and @xmath584 one can easily check that lemma [ lemma : x : axis0 ] and lemma [ lemma : z : axis0 ] hold true in @xmath0 with essentially the same proofs .
lemma [ lemma : comparisonincone ] and its proof extend to the case of @xmath0 setting @xmath585 and considering the analogue of the sets @xmath331 and @xmath337 ( introduced in section [ section : prooflemma : comparisonincone ] ) defined in the following way .
the set @xmath331 is now defined as the @xmath586-dimensional euclidean half cone contained in the hyperplane @xmath587 with vertex @xmath588 , axis the half line starting at @xmath334 and passing through @xmath335 and aperture @xmath336 .
the set @xmath337 is defined as the @xmath586-dimensional euclidean convex hull in the hyperplane @xmath587 of @xmath334 , @xmath589 and the @xmath590-dimensional euclidean ball @xmath591 . here
@xmath171 denotes the obvious analogue of the map defined in , @xmath592 , @xmath593 .
this section is devoted to the proof of theorem [ thm : destroybcp ] .
the construction is inspired by the construction given by the first - named author in theorem 1.6 of @xcite where it is proved that there exist translation - invariant distances on @xmath51 that are bi - lipschitz equivalent to the euclidean distance but that do not satisfy bcp .
let @xmath5 be a metric space .
assume that @xmath594 is an accumulation point in @xmath5 and let @xmath595 be a sequence of distinct points in @xmath9 such that @xmath596 for all @xmath86 and such that @xmath597 .
set @xmath598 next , we will prove that @xmath594 is an isolated point of @xmath606 for all @xmath607 .
more precisely , by definition of @xmath2 , we have , for all @xmath607 , @xmath608 hence @xmath609 for all @xmath607 . on the other hand , we will prove in lemma [ lem : isolatedpoint ] that @xmath610 for all @xmath607 .
then let us extract a subsequence @xmath611 starting at @xmath612 in such a way that @xmath613 for all @xmath614 and all @xmath615 .
it follows from that @xmath616 for all @xmath614 and all @xmath615 ( remember that the sequence @xmath617 is assumed to be decreasing ) . by definition of @xmath198 , one has @xmath620 for all @xmath621 and @xmath622 .
it follows that @xmath623 note that since @xmath1 is a distance , one indeed has @xmath624 which follows from one side from the triangle inequality and for the other side from the fact that one can consider @xmath625 , @xmath626 and @xmath627 , so that @xmath628 .
we get from lemma [ lem : equivalentdist ] that if @xmath633 then @xmath634 and hence @xmath635 .
since @xmath636 , one has @xmath637 . to prove the triangle inequality ,
let us consider @xmath44 , @xmath45 and @xmath49 in @xmath9 and two arbitrary chains of points @xmath638 , @xmath639 .
since @xmath640 is a chain of points from @xmath44 to @xmath45 , one has @xmath641 and hence @xmath642 first , we claim that @xmath648 for all @xmath649 provided @xmath441 is small enough . indeed , otherwise , with no loss of generality , we would have @xmath650 and @xmath651 , and hence @xmath652 which implies that @xmath653 on the other hand , together with the triangle inequality would give @xmath654 which is impossible as soon as @xmath655 . indeed , first , if @xmath657 for some @xmath658 , then we must have @xmath659 . otherwise , since @xmath617 is decreasing , we would have @xmath660 . hence we would get @xmath661 which is impossible as soon as @xmath662 .
* acknowledgement . *
the authors are very grateful to jeremy tyson for useful discussions and in particular for pointing out the link between the distances @xmath70 and the distances of negative type considered by lee and naor .
the second author would like to thank for its hospitality the department of mathematics and statistics of the university of jyvskyl where part of this work was done . | our main result is a positive answer to the question whether one can find homogeneous distances on the heisenberg groups that have the besicovitch covering property ( bcp ) .
this property is well known to be one of the fundamental tools of measure theory , with strong connections with the theory of differentiation of measures .
we prove that bcp is satisfied by the homogeneous distances whose unit ball centered at the origin coincides with an euclidean ball .
such homogeneous distances do exist on any carnot group by a result of hebisch and sikora . in the heisenberg groups ,
they are related to the cygan - kornyi ( also called kornyi ) distance .
they were considered in particular by lee and naor to provide a counterexample to the goemans - linial conjecture in theoretical computer science .
to put our result in perspective , we also prove two geometric criteria that imply the non - validity of bcp , showing that in some sense our example is sharp .
our first criterion applies in particular to commonly used homogeneous distances on the heisenberg groups , such as the cygan - kornyi and carnot - carathodory distances that are already known not to satisfy bcp . to put a different perspective on these results and for sake of completeness
, we also give a proof of the fact , noticed by d. preiss , that in a general metric space , one can always construct a bi - lipschitz equivalent distance that does not satisfy bcp . |
in this note we complete the work begun in the sequence of papers @xcite , @xcite-@xcite about equilateral triangles , regular tetrahedra , cubes , and regular octahedrons all with vertices having integer coordinates . very often we will refer to this property by saying that the various objects are in @xmath1 . strictly speaking these geometric objects
are defined as being more than the set of their vertices that determines them , but for us here , these are just the vertices .
so , for instance , an equilateral triangle is going to be a set of three points in @xmath2 for which the euclidean distances between every two of these points are the same .
the main purpose of the paper is to take a close look at the regular octahedrons in @xmath2 .
the simplest example of a regular octahedron with integer coordinates for its vertices is @xmath3 , [ 1 , 0 , 1 ] , [ 1 , 1 , 0 ] , [ 1 , 1 , 2 ] , [ 1 , 2 , 1 ] , [ 2 , 1 , 1]\},\ ] ] which can be obtained from the usual unit cube in @xmath4 , multiplying the vertices by a factor of two and then taking the coordinates of the centers of the new faces .
it turns out that this procedure gives all such octahedrons as shown in @xcite : [ fig1 ] @xmath5 @xmath6 [ oldtheorem ] every regular octahedron in @xmath2 is the dual of a cube that can be obtained ( up to a translation with a vector with integer coordinates ) by doubling a cube in @xmath1 .
referring to figure [ fig1 ] ( b ) , we showed that if the regular octahedron @xmath7 is in @xmath2 , then so is the cube @xmath8 and vice versa .
this defines a one - to - one correspondence between the classes of cubes ( invariant under integer translations ) and the classes regular octahedra ( invariant under integer translations ) in @xmath2 . in @xcite
we determined a sequence of irreducible cubes , one from each of the classes of cubes invariant under integer translations and cube symmetries .
for each one of these cubes we can construct as before a regular octahedron , obtaining this way a sequence of irreducible regular octahedrons .
in @xcite , we improved and adapted the earlier code for counting all cubes with vertices in @xmath9 and extended the sequence a098928 . in this paper ,
the usual techniques and ideas are going to be the same except some counting procedure that is very efficient in comparison to what we had before .
we are going to treat this in the general case so , let us suppose that these objects can be either equilateral triangles , regular tetrahedrons , cubes or regular octahedrons with vertices in @xmath2 .
for such an object , say @xmath10 , we can translate it , within @xmath2 , to @xmath11 that is in the positive octant and in such way each plane of coordinates contains at least one vertex of @xmath11 .
let us denote by @xmath12 the number of objects in @xmath13 obtained by applying to @xmath11 all 48 possible symmetries of the cube @xmath13 .
these symmetries are generated in the following way : first we have symmetries with respect to the middle planes and compositions , for example @xmath14 in a total of eight including the identity , then each one of these is coupled with one of the six permutations of the variables ( @xmath15 ) .
these transformations form a group isomorphic with the group of all @xmath16 orthogonal matrices having entries @xmath17 and it is also known as the group of symmetries of a cube or of a regular octahedron .
it is isomorphic to @xmath18 .
we are going to denote this group by @xmath19 although it is usually known under the name of ( extended ) octahedral group and denoted simply by @xmath20 .
if we think of @xmath12 as the cardinality of @xmath21 which is , by the first theorem of isomorphism of groups , the same as the cardinality of the group factor @xmath22 , where @xmath23 is the subgroup of @xmath19 of those symmetries that leave @xmath11 invariant . the structure of subgroups of @xmath19 is known and for each divisor of 48 there is a subgroup of that order .
hence , we expect @xmath12 to be in the set @xmath24 and most of the time to be @xmath25 since an arbitrary object @xmath11 in @xmath13 is unlikely to be invariant under any of the symmetries of @xmath26
. then , we denote by @xmath27 , the cardinality of the set of all the objects counted in @xmath12 and their ( all possible ) integer translations that leave the resulting objects in @xmath13 .
also , we denote by @xmath28 the objects counted in @xmath27 which are in @xmath29 . finally , let us denote by @xmath30 the number of objects counted in @xmath28 which are in @xmath31 .
then , we found a formula that gives the number of objects obtained from @xmath10 , under all symmetries and translation that leaves the resulting object in @xmath32 , @xmath33 .
this fact has been essentially proved in theorem 2.2 in @xcite .
the formula that gives this number is @xmath34 let us suppose that the object @xmath10 can be squeezed within a box of dimensions @xmath35 ( @xmath36 ) , i.e. up to symmetries and translations , @xmath10 can be transformed to @xmath37 fitting snugly into @xmath38 we can similarly consider all eight reflections compatible with the box @xmath39 of the form @xmath40 let us denote the group of these transformations by @xmath41 .
we notice that each one of these transformation leaves the object @xmath11 inside the box @xmath39 . from case to case
, depending of what the values @xmath42 , @xmath43 and @xmath44 are , we may have the result of some or all of the permutation transformations applied to @xmath37 still in @xmath39 .
hence , we will denote by @xmath45 the cardinality of the set @xmath46 ( { \cal o ' } ) \in b_{m , n , p}|s_1\in { \cal s}_b , s_2\in { \cal s}_6\}.\ ] ] let us look at an example
. suppose @xmath10 ( equal with @xmath11 ) is the equilateral triangle given by its vertices : @xmath47 , [ 5 , 7 , 0 ] , [ 7 , 0 , 1]\}.\ ] ] we observe that @xmath48
. then one can check that @xmath49 is the collection of eight triangles @xmath50 , [ 2 , 7 , 0 ] , [ 7 , 2 , 2]\ } , \{[0 , 0 , 1 ] , [ 2 , 7 , 2 ] , [ 7 , 2 , 0]\ } , \{[0 , 2 , 0 ] , [ 5 , 7 , 2 ] , [ 7 , 0 , 1]\},\\ \\ \{[0 , 5 , 0 ] , [ 5 , 0 , 2 ] , [ 7 , 7 , 1]\ } , \{[0 , 5 , 2 ] , [ 5 , 0 , 0 ] , [ 7 , 7 , 1]\ } , \{[0 , 7 , 1 ] , [ 2 , 0 , 0 ] , [ 7 , 5 , 2]\ } , \{[0 , 7 , 1 ] , [ 2 , 0 , 2 ] , [ 7 , 5 , 0]\ } , \end{array}}\ ] ] so @xmath51 .
it turns out that @xmath52 , @xmath53 , @xmath54 and @xmath55
. formula ( [ oldformula ] ) becomes @xmath56 it turns out the this factorization is not accidental and the following alternative to ( [ oldformula ] ) is true .
[ coj ] given @xmath10 , one of the objects mentioned before , and @xmath39 the smallest box containing a translation of @xmath10 ( @xmath36 ) , we let @xmath57 , @xmath58 , and @xmath59 then the number of distinct objects in the cube @xmath60 ( @xmath61 ) , obtained from @xmath10 by all possible integer translations and symmetries is equal to @xmath62 .
the case @xmath63 implies @xmath64 and @xmath65 because there is no room to shift the orbit @xmath66 inside of @xmath67 .
the formula follows from ( [ oldformula ] ) .
let us look into the case @xmath68 and @xmath69 .
we begin by observing that each integer translation of the box @xmath39 in all possible ways inside @xmath60 will give @xmath70 more copies of @xmath10 .
there is no overlap between these copies because neither one of them can be inside of two distinct translations of @xmath39 .
this is due to the minimality of @xmath42 , @xmath43 and @xmath44 .
we get @xmath71 such copies by counting all possible translations . since @xmath42 , @xmath43 and @xmath44
are all distinct , the box @xmath39 can be positioned first with the biggest of its dimensions along one of the directions given by the axis of coordinates , that is three different ways , and for each such position the next largest dimension can be positioned along the two remaining directions .
the minimality of @xmath42 , @xmath43 and @xmath44 makes the six different situations generate distinct objects .
this explains the factor of six that appears in ( [ newformula ] ) for this situation . in the last case ,
the box @xmath39 has two of its dimensions the same , so there are only three possibilities to arrange the box before one translates it . to see that we get all possible translates and symmetries of @xmath10 by this counting
, we can start with one copy @xmath11 .
construct the minimum box around it . in terms of its position and dimensions , we know in what of the six or three cases we are .
we transform it into the standard standard position , @xmath39 , and look at the corresponding object , @xmath72 .
the transformations involved form a group of transformations generated by the permutations of the coordinates , the reflections into the axes and integer translations .
every transformation in this group , say @xmath73 with @xmath74 a permutation , @xmath75 a reflection or a composition of reflections and @xmath76 a translation , which satisfies @xmath77 determines a representation @xmath78 with @xmath79 , @xmath80 as in the definition of @xmath70 .
this can be done by taking @xmath81 and @xmath82 .
this is true again because of the minimality of the box @xmath39 , i.e. there is only one integer translation that takes a reflected box @xmath83 into @xmath39 .
this new way of counting is more efficient from a computational point of view because @xmath84 is simply no bigger than 48 , as opposed to the previous situation when @xmath27 , @xmath28 and @xmath30 could turn out to be big numbers and so the number of iterations for computing them would be also large . roughly speaking , this counting factors out fast the problem with the integer translations . as an example , let us consider @xmath85 , [ 0 , 3 , 4 ] , [ 1 , 4 , 0 ] , [ 3 , 0 , 4 ] , [ 4 , 1 , 0 ] , [ 4 , 4 , 3]\}.\ ] ] the minimal box here is @xmath86 and after rotating @xmath87 in all possible ways ( figure [ fig2 ] ( b ) ) we get @xmath88 .
[ fig2 ] @xmath89
@xmath90 the idea of calculations is basically the same as in @xcite , in which we have constructed a list of irreducible cubes that are used to generate all the other cubes in @xmath60 . here
, we are using theorem [ oldtheorem ] , to construct a similar list of irreducible regular hexahedrons .
as expected , an _ irreducible _ regular hexahedron is one whose coordinates can not be obtained from a strictly smaller one with vertices in @xmath2 by integer dilations and translations .
one simple consequence of theorem [ oldtheorem ] is the next corollary .
[ sideofoctahedrons ] the sides of every irreducible regular octahedron in @xmath2 are of the form @xmath91with @xmath92 .
we used the same way of finding the parameterizations of the equilateral triangles as in @xcite : [ fig3 ] @xmath93 @xmath94 @xmath95 @xmath96 where @xmath97 and @xmath98 is a suitable solution of @xmath99 that makes all the numbers in ( [ paramtwo ] ) integers .
the sides - lengths of @xmath100 are equal to @xmath101 and the triangle can be completed to a regular tetrahedron ( in space ) with integer coordinates if and only if @xmath102 for some @xmath103 .
related to this fact we have the following proposition .
[ numberofro ] for a prime @xmath104 , the number of irreducible regular octahedrons in @xmath2 ( up to translations and symmetries ) having side lengths equal to @xmath105 , is at most @xmath106 , where @xmath107 with @xmath108 and @xmath109 defined as in @xcite .
the exact number of such objects is yet a big mystery to us .
we wrote the code using the maple software and so we took advantage of the build in commands available for a various number of functions .
the beginning is pretty standard : * step 1 . *
the next three procedures calculate all possible values of @xmath110 , then the parameters @xmath42 , @xmath43 , and finally the normal vector @xmath111 .
the values of @xmath110 are determined by using a characterization theorem for the quadratic form involved here , i.e. all values of @xmath110 less than @xmath43 such that @xmath110 is of the form a product of primes of the form @xmath112 , @xmath113 or @xmath114 . ....
> kvalues:=proc(n ) local i , j , k , l , a , p , q , r , m , mm ; l:={};mm:=floor((n+1)/2 ) ; for i from 2 to mm do a:=ifactors(2*i-1 ) ; k:=nops(a[2]);r:=0 ; for j from 1 to k do m:=a[2][j][1 ] ; p:=m mod 3 ; if m=3 then r:=1 fi ; if r=0 and p=2 then r:=1 fi ; od ; if r=0 then l:=l union { 2*i-1};fi ; od ; l:=l union { 1 } ; l:=convert(l , list ) ; end : .... next , we are interested in the solutions @xmath116 of the equation @xmath102 , which are primitive in the sense that @xmath117 , @xmath118 , @xmath119 , and @xmath120 .
we apply this procedure to only those @xmath110 s which are the output of @xmath121 . ....
> listofmn:=proc(k ) local a , b , i , nx , x , m , n , l ; x:=[isolve(k^2=m^2-m*n+n^2 ) ] ; nx:=nops(x ) ; l:= { } ; for i from 1 to nx do if lhs(x[i][1])=m then a:=rhs(x[i][1 ] ) ; b:=rhs(x[i][2 ] ) ; else b:=rhs(x[i][1 ] ) ; a:=rhs(x[i][2 ] ) ; fi ; if gcd(a , b)=1 and a>=0 and b>0 and 2*a < b then l:=l union { [ a , b]};fi ; od;l ; end : .... a prime of the form @xmath112 has a unique primitive decomposition as described , for example , @xmath122\}$ ] since @xmath123 and @xmath124 is a prime .
in general , the number of solutions is equal to @xmath125 where @xmath126 is the number of distinct factors of @xmath110 which are of the form @xmath112 .
this choice of @xmath42 and @xmath43 helps identify the irreducible octahedrons : all the other solutions @xmath116 of the equation @xmath102 lead to the same regular octahedron .
we included the case @xmath127 to obtain the output @xmath128\}$ ] for @xmath129 which is necessary later on . in the next procedure ,
we find only the solutions @xmath111 of @xmath130 that satisfy @xmath131 and @xmath132 .
the number of such solutions can be calculated precisely in terms of the prime factorization of @xmath133 ( see @xcite ) . ....
abcsol:=proc(d ) local i , j , k , m , u , x , y , sol , cd;sol:= { } ; for i from 1 to d do u:=[isolve(3*d^2-i^2=x^2+y^2)];k:=nops(u ) ; for j from 1 to k do if rhs(u[j][1])>=i and rhs(u[j][2])>=i then cd:=gcd(gcd(i , rhs(u[j][1])),rhs(u[j][2 ] ) ) ; if cd=1 then sol:=sol union { sort([i , rhs(u[j][1]),rhs(u[j][2])])};fi;fi ; od ; od;convert(sol , list ) ; end : .... for @xmath134 , for instance , the procedure _ abcsol _ gives the four solutions : [ 1 , 5 , 29 ] , [ 7 , 17 , 23 ] , [ 11 , 11 , 25 ] and [ 13 , 13 , 23 ] .
we observe that two of them have the property that @xmath135 , in which case , we know ( see @xcite ) that the formulae ( [ paramtwo ] ) simplify quite a bit .
* step 2 . *
the next seven procedures implement the new way of finding @xmath136 and @xmath137 which appear in the parametrization ( [ paramtwo ] ) .
it is followed by the calculation of the of the parametrization of equilateral triangles ( [ paramtwo ] ) and only one regular octahedron is constructed based on theorem [ oldtheorem ] .
this octahedron is then translated minimally into the positive octant and the minimal cube containing it is computed . ....
> uniquedecomposition:=proc(p ) local out , s , out1,out2 ; if p=2 then out:=[1,1 ] ; fi ; if p>2 then s:=isolve(p = x^2 + 3*y^2 ) ; out1:=abs(rhs(s[1][1]));out2:=abs(rhs(s[1][2 ] ) ) ; if out1 ^ 2 + 3*out2 ^ 2=p then out:=[out1,out2 ] ; else out:=[out2,out1];fi ; fi ; out ; end : .... for @xmath140 this procedure has the needed output @xmath141
$ ] . as an example , @xmath142 $ ] since @xmath143 is a prime and @xmath144 .
the next procedure is calculating the factorization in @xmath145 $ ] of a number of the form @xmath146 . ....
> factorovereisensteinintegers:=proc(u , v ) local i , n , m , k , a , x , f1,f2,g , y , y1,y2,l , nn , mm , uu , vv ; a:=sqrt(3)*i;nn:=gcd(u , v);uu:=u / nn;vv:=v / nn ; n:=uu^2 + 3*vv^2;x:=uu+vv*a;m:=ifactors(n);k:=nops(m[2 ] ) ; for i from 1 to k do f1:=m[2][i][1 ] ; f2:=m[2][i][2 ] ; if f1>2 then g:=uniquedecomposition(f1 ) ; else g:=[1,1 ] ; f2:=1 ; fi ; y:=expand(rationalize(x/(g[1]+a*g[2 ] ) ) ) ; y1:=re(y);y2:=type(y1,integer ) ; if y2=true then l[i]:=[g[1]+a*g[2],f2 ] ; else l[i]:=[g[1]-a*g[2],f2 ] ; fi ; od;[nn , seq(l[i],i=1 .. k),expand(nn*product(l[ii][1]^l[ii][2],ii=1 .. k))];end : ....
@xmath149 perhaps one word of caution is necessary at this point .
the decomposition in general is not unique in the usual sense since @xmath150 . as in the example
shown , we go for the second representation when something like this happens .
we are interested in this decomposition because it provides suitable values for @xmath136 and @xmath137 that we find next .
this turns out to be the greatest common divisor between @xmath151 and @xmath152 with @xmath153 , @xmath154 and @xmath97 ( see @xcite ) . ....
> findgcd:=proc(a , b , q ) local a , i , j , f , f1,f2,common , m , qq , fac , nfac , s , rs , p ; a:=sqrt(3)*i ; f:=factorovereisensteinintegers(a , b);m:=nops(f)-1 ; common:=gcd(f[1],2*q ) ; qq:=2*q / common^2 ; p:=common ; fac:=ifactors(qq);nfac:=nops(fac[2 ] ) ; for i from 1 to nfac do f1:=fac[2][i][1];f2:=fac[2][i][2 ] ; if f1=2 then f2:=1;fi ; s:=uniquedecomposition(f1 ) ; for j from 1 to m do if s[1]+a*s[2]=f[j][1 ] then p:=p*(s[1]+a*s[2])^(min(f[j][2],f2 ) ) ; fi ; if s[1]-a*s[2]=f[j][1 ] then p:=p*(s[1]-a*s[2])^(min(f[j][2],f2 ) ) ; fi ; od ; od;p:=expand(p);rs:=[re(p),im(p)/sqrt(3 ) ] ; [ rs , a*rs[1]+3*b*rs[2 ] mod 2*q , a*rs[2]-b*rs[1 ] mod 2*q ] ; end : .... in the case we have seen before , where @xmath134 , if we take @xmath155 , @xmath156 and @xmath157 , then we have @xmath158 , @xmath159 and @xmath160 .
the procedure above gives @xmath161 and @xmath162 and checks that @xmath163 ( mod @xmath152 ) and @xmath164 ( mod @xmath152 ) .
these two conditions are enough to insure that the expressions in ( [ paramtwo ] ) , all give integer values for the coordinates of @xmath165 and @xmath166 . in the next procedure
we use the @xmath136 and @xmath137 determined earlier and construct an irreducible regular octahedron in @xmath2 , given a vector @xmath111 and the possible values for @xmath116 in the decomposition of @xmath102 .
we are using simple formulae which can be derived easily from figure 1(b ) , thinking that the point @xmath167 is the origin , @xmath168 $ ] , @xmath169 $ ] and @xmath170 $ ] are the points @xmath171 , @xmath172 and @xmath173 . then the other three vertices are simply given by the addition of every two of these there vectors .
although there are six possible equilateral triangles that one may start with in this construction , one can see that in the end , essentially the same regular octahedron ( up to symmetries and translations ) is obtained . ....
> findpar:=proc(a , b , c , mm , nn ) local q , d , r , s , rs , a , b , k , mx , nx , my , ny , mz , nz , mu , nu , mv , nv , mw , nw , u , v , w , x , y , z , t , r1,r2,dd , e , f ; q:=a^2+b^2;k:=sqrt(mm^2-mm*nn+nn^2);d:=sqrt((a^2+b^2+c^2)/3);a:=a*c;b:=b*d ; rs:=findgcd(a ,- b , q);r:=rs[1][2 ] ; s:=rs[1][1 ] ; mx:=-(d*b*(3*r+s)+a*c*(r - s))/(2*q);nx:=-(r*a*c+d*b*s)/q ; my:=(d*a*(3*r+s)-b*c*(r - s))/(2*q);ny:=-(r*b*c - d*a*s)/q;mz:=(r - s)/2;nz:=r ; mu:=nx;mv:=ny;mw:=nz;nu:=nx - mx;nv:=ny - my ; nw:=nz - mz ; u:=mu*mm - nu*nn;v:=mv*mm - nv*nn;w:=mw*mm - nw*nn ; x:=mx*mm - nx*nn;y:=my*mm - ny*nn;z:=mz*mm - nz*nn ; r1:=[(x+u-2*a*k)/3,(v+y-2*b*k)/3,(z+w-2*c*k)/3 ] ; r2:=[(x+u+2*a*k)/3,(v+y+2*b*k)/3,(z+w+2*c*k)/3 ] ; if r1[1]=floor(r1[1 ] ) then t:=[[u , v , w],[x , y , z],r1 ] ; else t:=[[u , v , w],[x , y , z],r2 ] ; fi ; dd:=[t[1][1]+t[2][1],t[1][2]+t[2][2],t[1][3]+t[2][3 ] ] ; e:=[t[1][1]+t[3][1],t[1][2]+t[3][2],t[1][3]+t[3][3 ] ] ; f:=[t[2][1]+t[3][1],t[2][2]+t[3][2],t[2][3]+t[3][3 ] ] ; [ t[1],t[2],t[3],dd , e , f ] ; end : .... since for @xmath174 we get @xmath175\}$ ] , we checked to see what regular octahedron is obtained for @xmath176 : [ 2301 , -1421 , -880 ] , [ 880 , -2301 , 1421 ] , [ 2401 , 100 , 1521 ] , [ 3181 , -3722 , 541 ] , [ 4702 , -1321 , 641 ] , and [ 3281 , -2201 , 2942 ] . since the set @xmath177 has 336 elements in it , there are essentially at most 337 irreducible regular octahedra in @xmath2 with side lengths equal to @xmath178 as we pointed out in proposition [ numberofro ] .
next , we have a short function for subtracting two vectors @xmath179 and @xmath180 . in order to compare various octahedrons
it is easier if they are all translated to the positive octant in such a way each plane of coordinates contains at least a vertex of the octahedron .
this is accomplished with the next routine . ....
> tmttopqoctahedron:=proc(t ) local i , a , b , c , v , c ; a:=min(t[1][1],t[2][1],t[3][1],t[4][1],t[5][1],t[6][1 ] ) ; b:=min(t[1][2],t[2][2],t[3][2],t[4][2],t[5][2],t[6][2 ] ) ; c:=min(t[1][3],t[2][3],t[3][3],t[4][3],t[5][3],t[6][3 ] ) ; v:=[a , b , c];c:={subtrv(t[1],v),subtrv(t[2],v),subtrv(t[3],v ) , subtrv(t[4],v),subtrv(t[5],v),subtrv(t[6],v ) } ; end : ....
so , for instance , the octahedron mentioned earlier of sides lengths @xmath178 becomes : [ 2401 , 1521 , 3822 ] , [ 3822 , 2401 , 1521 ] , [ 2301 , 0 , 1421 ] , [ 1521 , 3822 , 2401 ] , [ 0 , 1421 , 2301 ] , and
[ 1421 , 2301 , 0 ] .
the smallest cube @xmath181 ^ 3 $ ] containing an octahedron positioned in the positive octant as specified earlier is computed in the following procedure . ....
> mscofmoctahedron:=proc(q ) local a , b , c , t ; t:=convert(q , list ) ; a:=max(t[1][1],t[2][1],t[3][1],t[4][1],t[5][1],t[6][1 ] ) ; b:=max(t[1][2],t[2][2],t[3][2],t[4][2],t[5][2],t[6][2 ] ) ; c:=max(t[1][3],t[2][3],t[3][3],t[4][3],t[5][3],t[6][3 ] ) ; max(a , b , c ) ; end : .... * step 3 . * in our construction of these octahedrons
we end up with essentially the same octahedron if we proceed from a different face of it . to eliminate the possibility of counting an octahedron twice or more than one time
, we would like to have a way of distinguishing between octahedra and so an invariant to translation and symmetries , like the side lengths , will be good .
such an invariant is the set of @xmath110 values which are given by the four pairs of opposite ( parallel ) faces of a regular octahedron .
we must have @xmath182 so , knowing @xmath183 and the @xmath184 s will give us @xmath185 s .
the set @xmath186 is clearly and invariant to translations and the symmetries we have talked about at the beginning of the paper .
hence , two octahedra with different sets of k - values will be essentially different .
we determine these k - values from the first three points given in the procedure @xmath187 .
the following calculation finds @xmath188 , given @xmath111 such that @xmath130 . ....
> fourkvalues:=proc(t ) local n1,n2,n3,n4,x , length ; length:=distance(t[1],t[2])/sqrt(2 ) ; n1:=unitvector(addvec(t[1],t[2],t[3 ] ) ) ; n2:=unitvector(addvec(t[1],t[2],multbyfactorv(t[3],-3 ) ) ) ; n3:=unitvector(addvec(t[1],t[3],multbyfactorv(t[2],-3 ) ) ) ; n4:=unitvector(addvec(t[3],t[2],multbyfactorv(t[1],-3 ) ) ) ; { length / n1,length / n2,length / n3,length / n4 } ; end : .... .... >
orbitbox1:=proc(t ) local i , k , t1,a , b , c , x , t2,t3,t4,t5,t6,t7,t8,t9,t10,t11,t12,t13,t14,t15,t16,t17,t18 , t19,t20,t21,t22,t23,t24,s , q ; q:=convert(t , list ) ; a:=max(q[1][1],q[2][1],q[3][1],q[4][1],q[5][1],q[6][1 ] ) ; b:=max(q[1][2],q[2][2],q[3][2],q[4][2],q[5][2],q[6][2 ] ) ; c:=max(q[1][3],q[2][3],q[3][3],q[4][3],q[5][3],q[6][3 ] ) ; t1:=t ; t2:={seq([q[i][1],q[i][2],c - q[i][3]],i=1 .. 6 ) } ; t3:={seq([q[i][1],b - q[i][2],q[i][3]],i=1 .. 6 ) } ; t4:={seq([a - q[i][1],q[i][2],q[i][3]],i=1 .. 6 ) } ; t5:={seq([a - q[i][1],b - q[i][2],q[i][3]],i=1 .. 6 ) } ; t6:={seq([a - q[i][1],q[i][2],c - q[i][3]],i=1 .. 6 ) } ; t7:={seq([q[i][1],b - q[i][2],c - q[i][3]],i=1 .. 6 ) } ; t8:={seq([a - q[i][1],b - q[i][2],c - q[i][3]],i=1 .. 6 ) } ; s:={t1,t2,t3,t4,t5,t6,t7,t8 } ; s ; end : .... .... > orbitbox:=proc(t , p ) local s , q , tt , a , b , c , m , ss , i , d , u , v , w , y , mm , nn , x , k ; q:=convert(t , list ) ; a:=max(q[1][1],q[2][1],q[3][1],q[4][1],q[5][1],q[6][1 ] ) ; b:=max(q[1][2],q[2][2],q[3][2],q[4][2],q[5][2],q[6][2 ] ) ; c:=max(q[1][3],q[2][3],q[3][3],q[4][3],q[5][3],q[6][3 ] ) ; d:=max(a , b , c);u:=d - a;v:=d - b;w:=d - c;y:={u , v , w};k:=p - d ; tt[1]:={seq([q[i][3],q[i][2],q[i][1]],i=1 .. 6 ) } ; tt[2]:={seq([q[i][2],q[i][3],q[i][1]],i=1 .. 6 ) } ; tt[3]:={seq([q[i][1],q[i][3],q[i][2]],i=1 .. 6 ) } ; tt[4]:={seq([q[i][2],q[i][1],q[i][3]],i=1 .. 6 ) } ; tt[5]:={seq([q[i][3],q[i][1],q[i][2]],i=1 .. 6 ) } ; s:=orbitbox1(t ) ; for i from 1 to 5 do s:=s union orbitbox1(tt[i ] ) ; od ; s:=convert(s , list);m:=nops(s);ss:= { } ; for i from 1 to m do if s[i][1][1]<=a and s[i][2][1]<=a and s[i][3][1]<=a and s[i][4][1]<=a and s[i][5][1]<=a and s[i][6][1]<=a and s[i][1][2]<=b and s[i][2][2]<=b and s[i][3][2]<=b and s[i][4][2]<=b and s[i][5][2]<=b and s[i][6][2]<=b and s[i][1][3]<=c and s[i][2][3]<=c and s[i][3][3]<=c and s[i][4][3]<=c and s[i][5][3]<=c and s[i][6][3]<=c and s[i][1][1]>=0 and s[i][2][1]>=0 and s[i][3][1]>=0 and s[i][4][1]>=0 and s[i][5][1]>=0 and s[i][6][1]>=0 and s[i][1][2]>=0 and s[i][2][2]>=0 and s[i][3][2]>=0 and s[i][4][2]>=0 and s[i][5][2]>=0 and s[i][6][2]>=0 and s[i][1][3]>=0 and s[i][2][3]>=0 and s[i][3][3]>=0 and s[i][4][3]>=0 and s[i][5][3]>=0 and s[i][6][3]>=0 then ss:=ss union { s[i ] } ; fi ; od ; mm:=nops(ss ) ; nn:=(k+u+1)*(k+v+1)*(k+w+1 ) ; if nops(y)=3 then x:=6*mm*nn;fi ; if nops(y)=2 then x:=3*mm*nn;fi ; if nops(y)=1 then x:=mm*nn;fi ; [ x , ss , nops(ss),[u , v ] ] ; end : .... the next two routines calculate the orbit of an octahedron within its minimal cube @xmath13 .
this orbit has at most 48 elements and it is needed in the process of elimination of octahedrons that have already appeared in the construction . in comparison with the previous orbit ,
it is bigger and invariant to all the symmetries . ....
> orbit1octahedron:=proc(t ) local i , k , t1,a , b , c , x , t2,t3,t4,t5,t6,t7,t8,t9,t10,t11,t12,t13,t14 , t15,t16,t17,t18,t19,t20,t21,t22,t23,t24,s , q , d , a1,b1,c1 ; q:=convert(t , list ) ; d:=mscofmoctahedron(t ) ; t1:=t ; t2:={seq([q[k][2],q[k][3],q[k][1]],k=1 .. 6 ) } ; t3:={seq([q[k][1],q[k][3],q[k][2]],k=1 .. 6 ) } ; t4:={seq([q[k][1],q[k][2],d - q[k][3]],k=1 .. 6 ) } ; t5:={seq([q[k][2],q[k][3],d - q[k][1]],k=1 .. 6 ) } ; t6:={seq([q[k][1],q[k][3],d - q[k][2]],k=1 .. 6 ) } ; t7:={seq([q[k][1],d - q[k][2],q[k][3]],k=1 .. 6 ) } ; t8:={seq([q[k][2],d - q[k][3],q[k][1]],k=1 .. 6 ) } ; t9:={seq([q[k][1],d - q[k][3],q[k][2]],k=1 .. 6 ) } ; t10:={seq([d - q[k][1],q[k][2],q[k][3]],k=1 .. 6 ) } ; t11:={seq([d - q[k][2],q[k][3],q[k][1]],k=1 .. 6 ) } ; t12:={seq([d - q[k][1],q[k][3],q[k][2]],k=1 .. 6 ) } ; t13:={seq([q[k][1],d - q[k][2],d - q[k][3]],k=1 .. 6 ) } ; t14:={seq([q[k][2],d - q[k][3],d - q[k][1]],k=1 .. 6 ) } ; t15:={seq([q[k][1],d - q[k][3],d - q[k][2]],k=1 .. 6 ) } ; t16:={seq([d - q[k][1],d - q[k][2],q[k][3]],k=1 .. 6 ) } ; t17:={seq([d - q[k][2],d - q[k][3],q[k][1]],k=1 .. 6 ) } ; t18:={seq([d - q[k][1],d - q[k][3],q[k][2]],k=1 .. 6 ) } ; t19:={seq([d - q[k][1],q[k][2],d - q[k][3]],k=1 .. 6 ) } ; t20:={seq([d - q[k][2],q[k][3],d - q[k][1]],k=1 .. 6 ) } ; t21:={seq([d - q[k][1],q[k][3],d - q[k][2]],k=1 .. 6 ) } ; t22:={seq([d - q[k][1],d - q[k][2],d - q[k][3]],k=1 .. 6 ) } ; t23:={seq([d - q[k][2],d - q[k][3],d - q[k][1]],k=1 .. 6 ) } ; t24:={seq([d - q[k][1],d - q[k][3],d - q[k][2]],k=1 .. 6 ) } ; s:={t1,t2,t3,t4,t5,t6,t7,t8,t9,t10,t11,t12,t13,t14,t15,t16 , t17,t18,t19,t20,t21,t22,t23,t24 } ; s ; end : > orbitoctahedron:=proc(t ) local s , q , t1 ; q:=convert(t , list ) ;
t1:={seq([q[k][3],q[k][2],q[k][1]],k=1 .. 6 ) } ; s:=orbit1octahedron(t ) union orbit1octahedron(t1 ) ; s ; end : .... * step 5 .
* at this point we are ready to build a list of minimal , irreducible octahedrons . with each entry
we keep , in order , @xmath133 , the value @xmath42 which is the size of the minimal cube @xmath192 , the six vertices of the octahedron , the @xmath110-values and the corresponding vector @xmath111 . in the end , another list is created in a new procedure , to get the list of all corresponding reducible cubes needed in a calculation for a given dimension @xmath43 .
the construction is a little complicated because there are octahedrons for which all @xmath110-values are different of @xmath193 . ....
> extendlist:=proc(n , n , l::list , mm , nn , orb::array ) local d , i , sol , nsol , nel , ttpcube , orb::array , ll::list , tnel , nl::list , cio , o , po , k , kvalues , m , exception , int , nn ; nn:=floor((n+1)/2 ) ; orb:=array(1 .. 2*nn+1 ) ; nel:=nops(l ) ; ll:=l ; k:=sqrt(mm^2-mm*nn+nn^2 ) ; m:=n*k ; if m<=n then sol:=abcsol(n);nsol:=nops(sol ) ; tnel:=nel;orb:=orb ; for i from 1 to nsol do o:=findpar(sol[i][1],sol[i][2],sol[i][3],mm , nn);po:=tmttopqoctahedron(o ) ; kvalues:=fourkvalues([o[1],o[2],o[3 ] ] ) ; exception:=evalb(1 in kvalues ) ; if k=1 or exception = false then int:=orbitoctahedron(po ) intersect orb[m ] ; cio:=nops(int ) ; if cio=0 then orb[m]:=orb[m ] union orbitbox(po , n)[2 ] ; d:=mscofmoctahedron(po ) ; nl[tnel+1]:=[n , d , po , kvalues , sol[i ] ] ; tnel:=tnel+1 ; fi ; fi ; od ; ll:=[seq(l[i],i=1 .. nel),seq(nl[j],j = nel+1 .. tnel ) ] ; fi;ll , orb ; end : .... .... > extendlistupton:=proc(n ) local i , j , k , l , kv , kvn , nn , l , orb::array , mn , nmn , e , n , ii ; kv:=kvalues(n);kvn:=nops(kv ) ; nn:=floor((n+1)/2 ) ; l:=[];orb:=array(1 .. 2*nn+1 ) ; for ii from 1 to 2*nn+1 do orb[ii]:= { } ; od ; for i from 1 to kvn do k:=kv[i];mn:=listofmn(k);nmn:=nops(mn ) ; for j from 1 to nmn do for l from 1 to nn+1 do n:=2*l-1 ; e:=extendlist(n , n , l , mn[j][1],mn[j][2],orb ) ; l:=e[1 ] ; for ii from 1 to 2*nn+1 do orb[ii]:=e[2][ii ] ; od ; od ; od ; od ; l;end : .... .... > l:=extendlistupton(100 ) : extendlistuptonmultiples:=proc(n , l ) local i , j , x , lc , m , mm , d , dd , c , cc , ll ; m:=nops(l);i:=1;ll:= { } ; while i<=m do d:=l[i][2 ] ; if d<=n then mm:=floor(n / d);c:=l[i][3];j:=2 ; while j<=mm do cc:=multbyfactoroctahedron(c , j ) ; dd:=d*j;lc:=nops(ll ) ; ll:=ll union { [ l[i][1]*j , dd , cc , l[i][4 ] ] } ; j:=j+1 ; od ; fi ; i:=i+1 ; od ; convert(ll , list ) ; end : .... > addupnew:=proc(n , l , ll ) local i , j , k , nc , mm , m , d , c , cc , x , dd , nt ; nc:=0 ; m:=nops(l ) ; i:=1 ; while i<=m do d:=l[i][2 ] ; if d<=n then x:=orbitbox(l[i][3],n)[1 ] ; nc:=nc+x ; fi ; i:=i+1 ; od ; m:=nops(ll ) ; i:=1 ; while i<=m do d:=ll[i][2 ] ; if d<=n then x:=orbitbox(ll[i][3],n)[1 ] ; nc:=nc+x ; fi ; i:=i+1 ; od ; nc;end : .... no : = [ [ 1 , 0 ] , [ 2 , 1 ] , [ 3 , 8 ] , [ 4 , 32 ] , [ 5 , 104 ] , [ 6 , 261 ] , [ 7 , 544 ] , [ 8 , 1000 ] , [ 9 , 1696 ] , [ 10 , 2759 ] , [ 11 , 4296 ] , [ 12 , 6434 ] , [ 13 , 9352 ] , [ 14 , 13243 ] , [ 15 , 18304 ] , [ 16 , 24774 ] , [ 17 , 32960 ] , [ 18 , 43223 ] , [ 19 , 55976 ] , [ 20 , 71752 ] , [ 21 , 90936 ] , [ 22 , 113973 ] , [ 23 , 141312 ] , [ 24 , 173436 ] , [ 25 , 210960 ] , [ 26 , 254587 ] , [ 27 , 305000 ] , [ 28 , 364406 ] , [ 29 , 432824 ] , [ 30 , 511421 ] , [ 31 , 600992 ] , [ 32 , 702556],[33 , 817200 ] , [ 34 , 946131 ] , [ 35 , 1090392 ] , [ 36 , 1251238 ] , [ 37 , 1430072 ] , [ 38 , 1629391 ] , [ 39 , 1850064 ] , [ 40 , 2094276 ] , [ 41 , 2363616 ] , [ 42 , 2659813 ] , [ 43 , 2984600 ] , [ 44 , 3341660 ] , [ 45 , 3731720 ] , [ 46 , 4156689 ] , [ 47 , 4618480 ] , [ 48 , 5119292 ] , [ 49 , 5661600 ] , [ 50 , 6248705 ] , [ 51 , 6882808 ] , [ 52 , 7568126 ] , [ 53 , 8306520 ] , [ 54 , 9104339 ] , [ 55 , 9962320 ] , [ 56 , 10888762 ] , [ 57 , 11882896 ] , [ 58 , 12949661 ] , [ 59 , 14090952 ] , [ 60 , 15311286 ] , [ 61 , 16613736 ] , [ 62 , 18001975 ] , [ 63 , 19479680 ] , [ 64 , 21052826 ] , [ 65 , 22724576 ] , [ 66 , 24500175 ] , [ 67 , 26383240 ] , [ 68 , 28387456 ] , [ 69 , 30510616 ] , [ 70 , 32758963 ] , [ 71 , 35136544 ] , [ 72 , 37656214 ] , [ 73 , 40317328 ] , [ 74 , 43125329 ] , [ 75 , 46085496 ] , [ 76 , 49207224 ] , [ 77 , 52493112 ] , [ 78 , 55954267 ] , [ 79 , 59592272 ] , [ 80 , 63415296 ] , [ 81 , 67428832 ] , [ 82 , 71642127 ] , [ 83 , 76059704 ] , [ 84 , 80701546 ] , [ 85 , 85565064 ] , [ 86 , 90662451 ] , [ 87 , 95997360 ] , [ 88 , 101592122 ] , [ 89 , 107443264 ] , [ 90 , 113561009 ] , [ 91 , 119951832 ] , [ 92 , 126644136 ] , [ 93 , 133629672 ] , [ 94 , 140916757 ] , [ 95 , 148513712 ] , [ 96 , 156444624 ] , [ 97 , 164706400 ] , [ 98 , 173308509 ] , [ 99 , 182260568 ] , [ 100 , 191575248 ] ] r. guy , _ unsolved problems in number theory _ , springer - verlag , 2004 e. grosswald , _ representations of integers as sums of squares _ , springer verlag , new york , 1985 . i. j. schoenberg , _ regular simplices and quadratic forms _ , j. london math . soc .
12 ( 1937 ) 48 - 55 . | in this paper we describe a procedure for calculating the number of regular octahedrons that have vertices with coordinates in the set @xmath0 . as a result
, we introduce a new sequence in _ the online encyclopedia of integer sequences _ ( a178797 ) and list the first one hundred terms of it .
we adapt the method appeared in @xcite which was used to find the number of regular tetrahedra with coordinates of their vertices in @xmath0 .
the idea of this calculation is based on the theoretical results obtained in @xcite .
a new fact proved here helps increasing the speed of all the programs used before .
the procedure is put together in a series of commands written for maple .
[ section ] [ theorem ] proposition [ theorem ] corollary [ theorem ] lemma [ theorem ] definition [ theorem ] problem [ theorem ] conjecture [ theorem ] example [ theorem ] remark # 1#1 # 1#1 # 1*#1 * |
in this paper , we explore the connection between the schubert calculus and the volume polynomial on spaces of convex polytopes .
we give various representations of schubert cycles in a complete flag variety by sums of faces of the gelfand
zetlin polytope .
our work is motivated by the rich interplay between algebraic geometry and convex polytopes , originally explored for toric varieties and recently extended to a more general setting in @xcite .
one of our main tools is a construction of @xcite , which , to every convex polytope @xmath0 , associates a graded commutative ring @xmath1 ( called the _ polytope ring _ ) satisfying the poincare duality ( see @xcite or section [ s.polytopes ] ) . for an _ integrally simple _ polytope @xmath2 ( simple means that there are exactly @xmath3 edges meeting at each vertex , and integrally simple means that primitive integer vectors parallel to the edges generate the lattice @xmath4 ) , the ring @xmath1 is isomorphic to the chow ring of the corresponding smooth toric variety @xmath5 @xcite .
faces of @xmath2 give rise to certain elements of @xmath1 , which generate @xmath1 as an additive group .
if @xmath6 $ ] is the element of @xmath1 corresponding to a face @xmath7 , then @xmath6\cdot [ g]=[f\cap g]$ ] in @xmath1 , provided that @xmath7 and @xmath8 are transverse .
single faces of @xmath2 represent cycles given by the closures of the torus orbits in @xmath5 . in this paper , we are primarily interested in the case , where @xmath2 is not simple .
kiumars kaveh has related the polytope rings of some non - simple polytopes to the chow rings of smooth non - toric spherical varieties @xcite . in particular
, he observed that the ring @xmath1 for the _ gelfand zetlin polytope _ @xmath9 ( which is not simple ) associated with a strictly dominant weight @xmath10 of the group @xmath11 is isomorphic to the chow ring of the variety @xmath12 of complete flags in @xmath13 .
when @xmath2 is not simple , there is no straightforward correspondence between faces of @xmath2 and elements of @xmath1 .
one of the results of the present paper is a general construction that with every element of @xmath1 associates a linear combination of faces of @xmath2 ( though not every face of @xmath2 corresponds to an element of @xmath1 ) .
namely , we embed the ring @xmath1 into a certain @xmath14-module @xmath15 , whose elements can be regarded as linear combinations of arbitrary faces of @xmath2 modulo some relations ( see section [ s.polytopes ] ) .
the module @xmath15 depends on the choice of a _ resolution _ of @xmath2 . on the algebro - geometric level
, @xmath1 can be regarded as the subring of the chow ring of the singular toric variety @xmath5 generated by the picard group and @xmath15 can be constructed using a resolution of singularities for @xmath5 .
however , we describe @xmath15 in elementary terms using convex geometry
. a crucial feature of such representations by sums of faces is that we can still multiply elements of @xmath1 by intersecting faces ( assuming the faces we intersect are transverse ) .
while our construction applies to any convex polytope @xmath2 , it is especially interesting to study the case , where @xmath9 is a gelfand
zetlin polytope , due to the isomorphism @xmath16 for the flag variety @xmath12 .
recall that @xmath17 ( as a group ) is a free abelian group with the basis of schubert cycles .
in particular , our construction allows to represent schubert cycles as linear combinations of faces of the gelfand
zetlin polytope in many different ways ( see theorem [ t.fk ] , proposition [ p : linrels ] , corollary [ c.fk ] ) , which has applications to schubert calculus . the relation between schubert varieties and faces of the gelfand
zetlin polytope was first investigated in @xcite , and then by different methods also in @xcite and @xcite .
the approach of @xcite ( via degenerations of schubert varieties to subvarieties of a singular toric variety ) seems to be the closest to ours .
however , the results of @xcite can not be directly used for purposes of schubert calculus , since only one degeneration is constructed for each schubert variety .
the polytope ring @xmath1 and the @xmath14-module @xmath15 allow us to bypass toric degenerations and do all calculations with faces directly in @xmath1
. given two schubert cycles @xmath18 $ ] and @xmath19 $ ] , we can represent @xmath18 $ ] and @xmath19 $ ] as sums of faces so that every face appearing in the decomposition of @xmath18 $ ] is transverse to every face appearing in the decomposition of @xmath19 $ ] ( see corollary [ c.richardson ] ) .
this allows us to represent the intersection of any two schubert cycles by linear combinations of faces with nonnegative coefficients .
this might lead to a positive formula for the structure constants ( which are triple products @xmath18[x^{w'}][x^{w''}]$ ] ) by counting vertices of the gelfand
zetlin polytope . the connection between schubert calculus and gelfand zetlin polytopes stems from the representation theory of @xmath11 .
recall that by definition of gelfand zetlin polytopes the integer points inside and at the boundary of @xmath20 parameterize a natural basis ( a _ gelfand zetlin basis _ ) in the irreducible highest weight @xmath21module @xmath22 with the highest weight @xmath23 . in particular , with every integer point @xmath24 we can associate its weight @xmath25 in the character lattice of @xmath21 . to derive some presentations of schubert cycles by sums of faces we establish the following relation between
_ demazure submodules _ of @xmath22 and faces of @xmath20 . for every schubert variety @xmath26 and a strictly dominant weight @xmath23
, we realize the corresponding demazure character as the exponential sum @xmath27 where @xmath28 runs over integer points in the union of all _ rc - faces , or reduced kogan faces _ ( see section [ s.gz ] ) of @xmath29 with permutation @xmath30 ( see theorem [ t.demazure ] ) .
this generalizes the identity from ( * ? ? ?
* corollary 15.2 ) for the demazure character of a _ @xmath31avoiding _ , or _ kempf _
, permutation @xmath30 ( such permutations are also called _
dominant _ , but we will use the term `` kempf '' instead ) .
note that a permutation is kempf if and only if there is a unique face with this permutation ( see ( * ? ? ?
* proposition 2.3.2 ) ) , and this is exactly the face considered in @xcite . to prove our formula for the demazure character we use elementary convex geometry together with a simple combinatorial procedure for dealing with divided difference operators ( called _ mitosis _ ) introduced in @xcite ( see also @xcite for an elementary exposition ) .
in particular , our proof yields a geometric realization of mitosis ( see subsection [ ss : paramitosis ] ) . as a byproduct
, we construct a minimal realization of a simplex as a cubic complex different from those previously known ( see proposition [ p.cubic ] ) .
this paper is organized as follows . in section [ s.polytopes ] ,
we recall the definition of the polytope ring @xmath1 , discuss its properties and construct a module @xmath15 for a non - simple @xmath2 . in section [ s.gz ] ,
we study the polytope rings of the gelfand zetlin polytopes . in section [ s.cycles ] ,
we represent schubert cycles by faces . in section [ s.demazure ] ,
we give formulas for demazure characters , hilbert functions and degrees of schubert varieties in terms of faces and deduce from these formulas some of the results of section [ s.cycles ] . in section [ s.keylemma ] , we introduce a simple geometric version of mitosis ( _ paramitosis _ ) and use it to prove formulas of section [ s.demazure ] for demazure characters .
_ acknowledgements .
_ this project was started when the first , second and third authors , respectively , were affiliated with the max planck institute for mathematics ( mpim ) , hausdorff center for mathematics in bonn and jacobs university bremen .
the project was continued when the first and the third author visited the freie universitt berlin and the mpim , bonn .
we would like to thank these institutions for hospitality , financial support and excellent working conditions .
the authors are grateful to michel brion , askold khovanskii and allen knutson for useful discussions .
following @xcite , we associate a graded commutative ring with any polynomial
. we will later specialize to the case of the volume polynomial on a space of polytopes with a given normal fan .
let @xmath32 be a lattice , i.e. a free @xmath14-module , and @xmath33 a homogeneous polynomial on the real vector space @xmath34 containing the lattice @xmath32 .
the symmetric algebra @xmath35 of @xmath32 can be thought of as the ring of differential operators with constant integer coefficients acting on @xmath36 $ ] , the space of all polynomials on @xmath37 . if @xmath38 and @xmath39 $ ] , then we write @xmath40 $ ] for the result of this action .
define @xmath41 as the homogeneous ideal in @xmath35 consisting of all differential operators @xmath42 such that @xmath43 .
set @xmath44 .
we call this ring _ the ring associated with the polynomial @xmath33_. let @xmath45 be another lattice , and @xmath46 a homomorphism of lattices . define the polynomial @xmath47 $ ] as @xmath48 .
we want to describe a relation between the associated rings @xmath49 and @xmath50 .
unfortunately , there is no natural homomorphism between these rings .
however , [ p : mod ] there is a natural abelian group @xmath51 , a natural epimorphism @xmath52 and a natural monomorphism @xmath53 such that @xmath54 whenever @xmath55 and @xmath56 .
this proposition can be used in the following way .
elements of @xmath50 can be embedded naturally into @xmath51 . although elements of @xmath51 can not be multiplied in general , we consider the lifts to @xmath49 of two elements coming from @xmath50 , multiply them in @xmath49 , and project the product back to @xmath51 . in many cases ,
this is easier than multiplying two elements of @xmath50 directly .
consider a @xmath14-submodule @xmath57 of @xmath35 consisting of all operators @xmath42 such that @xmath58 .
set @xmath59 .
clearly , @xmath60 ; thus we obtain a natural projection @xmath52 .
let @xmath61 be the homomorphism induced by the map @xmath62 .
for a differential operator @xmath63 , let @xmath64 $ ] denote the class of @xmath42 in the ring @xmath50 .
we define @xmath65)$ ] as the class in @xmath51 of the operator @xmath66 .
to verify that @xmath65)$ ] is well defined , we need the following formula @xmath67 for every @xmath39 $ ] . indeed , this formula is obviously true if @xmath68 , and both parts of this formula depend multiplicatively on @xmath42 . in particular
, we have @xmath69 which is equal to zero whenever @xmath42 is in @xmath70 .
it follows that the element @xmath65)$ ] is well defined : if @xmath71 , then @xmath72 .
it also follows from the same formula that @xmath73 is injective : if @xmath65)=0 $ ] , i.e. @xmath72 , then @xmath71 .
it remains to prove that @xmath74 whenever @xmath55 and @xmath56 .
but this is an immediate consequence of the formula @xmath75 .
consider the set of all convex polytopes of dimension @xmath76 in @xmath77 .
this set can be endowed with the structure of a commutative semigroup using _ minkowski sum _ @xmath78 it is not hard to check that this semigroup has cancelation property .
we can also multiply polytopes by positive real numbers using dilation : @xmath79 hence , we can embed the semigroup of convex polytopes into its grothendieck group @xmath80 , which is a real ( infinite - dimensional ) vector space .
the elements of @xmath80 are called _ virtual polytopes_. recall that two convex polytopes are called _ analogous _ if they have the same normal fan , i.e. there is a one - to - one correspondence between the faces of @xmath2 and the faces of @xmath81 such that any linear functional , whose restriction to @xmath2 attains its maximal value at a given face @xmath82 has the property that its restriction to @xmath81 attains its maximal value at the corresponding face of @xmath81 ( the set of linear functionals , whose restrictions to @xmath2 attain their maximal values at a face @xmath83 , form a cone @xmath84 ; the _ normal fan _ of @xmath2 is defined as the set of cones @xmath84 corresponding to all faces @xmath85 ) .
a virtual polytope is said to be analogous to @xmath2 if it can be represented as a difference of two convex polytopes analogous to @xmath2 .
all virtual polytopes analogous to @xmath2 form a finite dimensional subspace @xmath86 . on the vector space @xmath80 , there is a homogeneous polynomial @xmath87 of degree @xmath88 , called the _
volume polynomial_. fix a constant ( translation invariant ) volume form on @xmath77 .
if an integer lattice @xmath89 is fixed , we will always choose this volume form to take value 1 on the fundamental parallelepiped of @xmath4 .
the volume form on @xmath77 being fixed , the volume polynomial on the space @xmath80 is uniquely characterized by the property that its value @xmath90 on any convex polytope @xmath2 is equal to the volume of @xmath2 .
we will be interested in the restriction @xmath91 of the volume polynomial @xmath87 to the subspace @xmath92 of all virtual polytopes analogous to @xmath2 .
consider an integer convex polytope @xmath2 ( i.e. a convex polytope with integer vertices ) of dimension @xmath88 , not necessarily simple .
let @xmath93 be a lattice in @xmath92 generated by some integer polytopes analogous to @xmath2 ( we do not assume that @xmath93 contains all integer polytopes analogous to @xmath2 ; thus this lattice may depend on some extra choices rather than only on @xmath2 ) .
suppose that @xmath81 is a convex polytope with integer vertices , whose normal fan is a simplicial subdivision of the normal fan of @xmath2 . in this case
, @xmath81 is called a _ resolution _ of @xmath2 ( note that , since the normal fan of @xmath81 is simplicial , the polytope @xmath81 is simple ) . with the volume polynomial @xmath91 restricted to the lattice @xmath93 , we associate the _ polytope ring _ @xmath94 .
similarly , for the simple polytope @xmath81 , we consider the ring @xmath95 associated with the volume polynomial @xmath96 on the lattice @xmath97 ( we always assume that this lattice is generated by _ all _ integer polytopes analogous to @xmath81 )
. we will use the @xmath14-module @xmath98 introduced in proposition [ p : mod ] together with the homomorphisms @xmath99 and @xmath100 . since @xmath73 is a canonical embedding
, we will identify elements of @xmath1 with their @xmath73-images in @xmath101 . with every face @xmath102 of @xmath81 , we can associate a face @xmath7 of @xmath2 called the _ @xmath2-degeneration _ of @xmath102 ( or just _ degeneration _ if @xmath2 is fixed ) .
a face @xmath7 of @xmath2 is called _ regular ( with respect to @xmath81 ) _ if there is only one face @xmath102 of @xmath81 such that @xmath7 is the degeneration of @xmath102 .
[ p : reg ] suppose that @xmath103 is a simple vertex of @xmath2 , i.e. exactly @xmath3 facets of @xmath2 meet at @xmath103 .
moreover , suppose that no facet of @xmath81 degenerates to a face of smaller dimension .
then any face of @xmath2 containing @xmath103 is regular .
let @xmath104 , @xmath105 , @xmath106 be all facets of @xmath2 containing the vertex @xmath103 ( which are clearly regular ) .
denote by @xmath107 , @xmath105 , @xmath108 the corresponding ( parallel ) facets of @xmath81 .
note that the intersections of different subsets of @xmath109 are different faces of @xmath2 .
clearly , any intersection of facets @xmath110 degenerates into the intersection of the corresponding facets @xmath111 ( which has the same dimension ) , and all other faces of @xmath81 degenerate to faces of @xmath2 not containing the vertex @xmath103 .
this proves the desired statement .
we now give more details on the structure of the ring @xmath112 .
for every facet @xmath113 of @xmath81 , there is a differential operator @xmath114 such that , for every convex polytope @xmath115 analogous to @xmath81 , the number @xmath116 is the @xmath117-dimensional volume of the facet of @xmath115 parallel to @xmath113 .
the ideal @xmath118 is very easy to describe .
it is generated ( as an ideal ) by the following two groups of differential operators @xcite : * the images of integer vectors @xmath119 under the natural inclusion of @xmath4 into @xmath120 such that @xmath121 is the parallel translation of @xmath81 by the vector @xmath122 ; * the operators of the form @xmath123 , where @xmath124 .
the volume polynomial on the spaces @xmath125 was previously used in @xcite to describe the cohomology rings of smooth toric varieties .
we briefly recall this description .
every integer polytope @xmath81 defines a polarized toric variety @xmath126 .
if @xmath81 is integrally simple , then @xmath126 is smooth . in this case , the chow ring of @xmath126 ( or , equivalently , the cohomology ring @xmath127 ) is isomorphic to @xmath112 @xcite .
this description is very useful .
firstly , it is functorial . secondly , it is clear from the definition that the nonzero homogeneous components of the ring @xmath112 have degrees @xmath128 ( since the volume polynomial has degree @xmath76 ) and that @xmath112 has a non - degenerate pairing ( poincar duality ) defined by @xmath129 for any two homogeneous elements @xmath130 , @xmath131 of complementary degrees .
the poincar duality on the ring @xmath112 is a key ingredient in the proof of the isomorphism between @xmath112 and @xmath127 ( see @xcite for more details ) .
note that there is another functorial description @xcite of the chow ring of @xmath126 via piecewise polynomial functions on fans but for this description the upper bound on the degrees and the poincar duality are harder to check directly .
also , a first known ( non - functorial ) description of the chow ring ( by generators and relations ) follows easily from the definition of the ring @xmath112 ( see e.g. @xcite ) .
so it seems that the polytope rings give the most convenient description of the chow rings of smooth toric varieties .
note that if a polytope @xmath2 is not simple , the ring @xmath1 makes sense , has all nonzero homogeneous components in degrees @xmath128 and satisfies the poincar duality .
however , its relation to the chow ring of ( now singular ) toric variety @xmath5 is unclear partly because the latter no longer enjoys the poincar duality . on the other hand ,
the ring @xmath1 for non - simple polytopes is sometimes related to the chow rings of smooth non - toric varieties as was noticed by kaveh @xcite .
we now discuss some important properties of the isomorphism @xmath132 for a simple polytope @xmath81 .
this isomorphism allows us to identify the algebraic cycles on @xmath126 with the linear combinations of the faces of @xmath81 .
the dimension of the space @xmath125 is equal to the number @xmath133 of facets of @xmath81 ( since we can shift all support hyperplanes of @xmath81 independently ) . note that for a non - simple polytope @xmath2 the dimension of @xmath92 is strictly less than @xmath134 ( e.g.
if @xmath2 is an octahedron , then @xmath92 has dimension 4 ) . for simple @xmath81 ,
the space @xmath125 has natural coordinates called the _
support numbers_. there are as many support numbers as facets of @xmath81 .
the support numbers are defined by fixing @xmath135 linear functionals @xmath136 on @xmath77 corresponding to facets @xmath137 of @xmath81 such that every facet @xmath137 of @xmath81 is contained in the hyperplane @xmath138 for some constants @xmath139 , and the polytope @xmath81 satisfies the inequalities @xmath140 . if @xmath104 , @xmath105 , @xmath141 are all facets of @xmath81 , then any collection of real numbers @xmath142 defines a unique ( possibly virtual ) polytope in @xmath125 .
when dealing with integer polytopes , we always choose @xmath136 to be a primitive integer covector orthogonal to @xmath137 . in this case
, @xmath139 is ( up to a sign ) the integer distance between the origin and the hyperplane containing @xmath137 .
if we choose the volume forms and the linear functionals @xmath136 consistent with the integer lattice ( in the sense explained above ) , then the differential operators @xmath143 coincide with the partial derivatives with respect to the support numbers @xmath139 . for a face @xmath144 of codimension @xmath145 , we set @xmath146 , and denote by @xmath6 $ ] the class of @xmath147 in the ring @xmath112 .
the elements @xmath6 $ ] corresponding to the faces of @xmath81 generate @xmath112 as an abelian group .
moreover , it suffices to take some special faces , called _ separatrices _ in @xcite .
there is an explicit algorithm to represent the product @xmath148\cdot [ f_2]\in r_q$ ] as a linear combination of faces , i.e. of elements of the form @xmath6 $ ] corresponding to faces @xmath7 of @xmath81 .
this algorithm resembles the well - known algorithm from intersection theory : we need to replace @xmath148 $ ] by a linear combination of faces that are transverse to @xmath149 .
the linear relations between facets of @xmath81 follow immediately from the description of the ideal @xmath150 given above .
they have the form @xmath151=0 , \eqno(4)\ ] ] where @xmath152 is any vector , and the sum is over all facets of @xmath81 . indeed , the volume polynomial is invariant under parallel translations
. therefore , the @xmath153-derivative of @xmath154 is zero ( where @xmath155 replaces any fixed element of @xmath125 ) . by the chain rule ,
this derivative is equal to @xmath156 .
any linear relation between the elements @xmath157 $ ] has this form ( see @xcite ) . if @xmath81 is a resolution of @xmath2 , we will be interested in representations of elements @xmath158 by linear combinations of faces of @xmath81 i.e. in the following form @xmath159\right),\ ] ] where the summation is over some set of faces of @xmath81 .
then proposition [ p : mod ] allows us to compute the product of two elements @xmath160 as follows .
if we find a representation @xmath161\right),\ ] ] such that all @xmath162 are transverse to all @xmath7 , then @xmath163\right).\ ] ] in the sequel , we will also use the following lemma , which is a direct corollary of the definition of @xmath101 .
[ l : poincare]let @xmath164 and @xmath165 be two homogeneous elements in @xmath112 of the same degree .
we have @xmath166 in @xmath101 iff @xmath167 for all homogeneous @xmath168 of complementary degree such that @xmath169 .
consider the polytope @xmath2 in @xmath170 given by the following linear inequalities : @xmath171 this is a 3-dimensional gelfand
zetlin polytope , see figure 1 .
the defining system of linear inequalities for @xmath2 is usually represented schematically as follows : @xmath172 and its resolution , title="fig:",width=188 ] and its resolution , title="fig:",width=188 ] the polytope @xmath2 can be obtained from the parallelepiped @xmath173\times [ b , c]\times [ a , c]$ ] by removing two prisms : @xmath174 therefore , the volume of @xmath2 is equal to @xmath175 ( it can also be seen geometrically , without any computation , that the part we are removing is half the volume of the entire parallelepiped ) .
the ring @xmath1 is spanned by the classes of partial differentiations @xmath176 , @xmath177 and @xmath178 . moreover , since the volume of @xmath2 will not change if we shift @xmath122 , @xmath179 and @xmath180 simultaneously by the same real number , we have @xmath181 in @xmath1 .
a distinguished set of additive generators of @xmath1 is given by schubert polynomials in @xmath182 and @xmath183 , i.e. @xmath184 @xmath185 now consider a simple polytope @xmath81 given by the following inequalities : @xmath186 where @xmath187 is a fixed small number .
the polytope @xmath81 can also be obtained from the parallelepiped @xmath173\times [ b , c]\times [ a , c+\eps]$ ] by removing two prisms : @xmath188 therefore , the volume of @xmath81 is equal to @xmath189 @xmath190 this is a polynomial in @xmath122 , @xmath179 , @xmath180 and @xmath191 .
the ring @xmath112 is multiplicatively generated by partial differentiations @xmath192 , @xmath193 and @xmath194 ( the tildes are just to distinguish these elements of @xmath112 from elements @xmath176 , @xmath177 , @xmath195 ) .
we have @xmath196,\quad \tilde\d_b=-[y = b]+[x = b],\quad \tilde\d_c=[y = c].\ ] ] formula ( 4 ) gives three linear relations between facets of @xmath81 : @xmath197+[x = b]+[z = x]=0,\ ] ] @xmath198-[z = y+\eps]+[y = c]=0,\ ] ] @xmath199+[z = y+\eps]=0.\ ] ] we can represent the schubert polynomials of @xmath176 and @xmath177 as the @xmath200-images of certain elements of @xmath112 as follows : @xmath201,\quad \sf_{s_2}=\pi([x = a]+[y = b]),\ ] ] @xmath202,\quad \sf_{s_1s_2}=\pi[x = a , y = b].\ ] ] all faces of @xmath81 that appear in the right hand sides of these identities degenerate to regular faces of @xmath2 .
for instance , the expression for @xmath203 is obtained as follows : @xmath204\cdot [ x = a])=\ ] ] @xmath205\cdot([x = b]+[z = x ] ) ) = \pi([x = a]\cdot[x = b]+[x = z = a]).\ ] ] the first term in the right hand side vanishes , because the faces @xmath206 and @xmath207 are disjoint . in this way , it is easy to justify all heuristic calculations with faces in ( * ? ? ? * section 4 ) .
we now consider the ring @xmath1 for the _ gelfand zetlin polytope _ @xmath9 associated with a strictly dominant weight @xmath10 of the group @xmath11 , i.e. with an @xmath88-tuple of integers @xmath208 such that @xmath209 for all @xmath210 .
recall that the gelfand
zetlin polytope @xmath20 is a convex integer polytope in @xmath211 , where @xmath212 , with the property that the integer points inside and at the boundary of @xmath20 parameterize a natural basis in the irreducible representation of @xmath11 with the highest weight @xmath23 .
it can be defined by inequalities @xmath213 where @xmath214 are coordinates in @xmath77 , and the notation @xmath215 means @xmath216 .
see figure 1 for a picture of the gelfand
zetlin polytope for @xmath217 .
note that gelfand
zetlin polytopes @xmath29 and @xmath218 are analogous for any two strictly dominant weights @xmath219 and @xmath220 . for what follows ,
we set @xmath221 for some strictly dominant weight @xmath23 , and define @xmath222 as the lattice spanned by all gelfand zetlin polytopes @xmath218 , where @xmath220 runs through all strictly dominant weights .
the correspondence @xmath223 establishes a natural isomorphism between the lattices @xmath224 and @xmath93 .
in other words , virtual polytopes in @xmath93 are parameterized by arbitrary @xmath88-tuples of integers , not necessarily strictly increasing .
one can show that the ring @xmath1 does not change if @xmath93 is replaced by the lattice generated by _ all _ polytopes analogous to @xmath20 but we will not need this .
recall that , to every complete flag @xmath225 in @xmath13 , one associates one - dimensional vector spaces @xmath226 .
the disjoint union of @xmath227 with its natural projection to @xmath12 given by the formula @xmath228 has a structure of a line bundle over @xmath12 .
this line bundle @xmath229 is called a _
tautological quotient line bundle _ over @xmath12 .
[ t : iso ] the ring @xmath1 is isomorphic to the chow ring ( and to the cohomology ring ) of the complete flag variety @xmath12 for @xmath11 ( note that @xmath230 ) so that the images in @xmath1 of the differential operators @xmath231 get mapped to the first chern classes of the tautological quotient line bundles @xmath232 over @xmath12 .
this theorem can also be deduced directly from the borel presentation for the cohomology ring @xmath233 using that the volume of @xmath20 ( regarded as a function of @xmath23 ) is equal to @xmath234 times a constant . along with the gelfand
zetlin polytope @xmath2 , we consider its resolution @xmath81 such that the number of facets in @xmath81 is the same as the number of facets in @xmath2 , and every support hyperplane of @xmath81 intersecting @xmath81 by a facet is sufficiently close to a parallel support hyperplane of @xmath2 intersecting @xmath2 by a facet .
this establishes a one - to - one correspondence between facets of @xmath81 and facets of @xmath2 such that the corresponding facets are parallel .
nothing in what follows will depend on a particular choice of @xmath81
. it will be convenient to represent faces of @xmath2 by _
face diagrams_. first , replace all @xmath235 and @xmath236 in table @xmath237 by dots .
every face of @xmath2 is given by a system of equations of the form @xmath238 , where @xmath122 and @xmath179 are coordinates represented by adjacent dots in two consecutive rows . to represent such an equation
, we draw a line interval connecting the corresponding dots ( these line intervals go from northeast to southwest or from northwest to southeast ) .
thus a system of equations defining a face of @xmath2 gets represented by a collection of line intervals called the face diagram .
_ rows _ of a face diagram are defined as the collections of dots corresponding to the coordinates @xmath236 with a fixed @xmath239 , and _ columns _ are by definition collections of dots with a fixed @xmath240 ( columns look like diagonals in our pictures ) .
let @xmath7 be a regular face of @xmath2 and @xmath102 the corresponding face of @xmath81 , so that @xmath7 is the degeneration of @xmath102 .
we will often write @xmath6 $ ] for the class @xmath241 $ ] of @xmath102 in the polytope ring @xmath112 . note that , in general , @xmath242 $ ] _ does not belong to _ @xmath1 .
every facet of @xmath2 is regular . for @xmath243 and @xmath244 , let @xmath245 denote the facet of @xmath2 given by the equation @xmath246 , where we set @xmath247 .
similarly , for @xmath243 and @xmath248 , we let @xmath249 denote the facet @xmath250 . clearly , any facet of @xmath2 is either one of @xmath245 or one of @xmath249 .
the following proposition describes all linear relations between facets of @xmath2 : [ p : linrels ] we have the following linear relations in @xmath112 : @xmath251-[\g^-_{i , j}]-[\g_{i-1,j}]+[\g^-_{i-1,j+1}]=0,\ ] ] where the terms must be ignored if their indices are out of range .
moreover , all linear relations are generated by these .
we call the relation displayed above the _ 4-term relation _ at @xmath252 .
let @xmath253 be the standard basis in @xmath77 .
the 4-term relation at @xmath252 is exactly the relation of the form @xmath254=0,\ ] ] where the summation is over all facets of @xmath81 .
in fact , there are at most four facets @xmath137 of @xmath81 such that @xmath255 ; these are @xmath245 , @xmath256 , @xmath249 and @xmath257 .
it is straightforward to check that the coefficients are as stated . in what follows
, we will mostly consider faces of the gelfand
zetlin polytope given by the equations of the type in notation of @xcite , which is the same as type @xmath258 equation in @xcite ( his @xmath259 is our @xmath236 ) . ] @xmath246 ( i.e. the intersections of facets of the form @xmath245 ) .
we will call such faces _
kogan faces_. to each kogan face @xmath7 , we assign the permutation @xmath260 as follows .
first , assign to each equation @xmath246 the simple reflection @xmath261 .
now compose all simple reflections corresponding to the equations defining @xmath7 by going from left to right in each row of the diagram for @xmath7 and by going from the bottom row to the top one .
we say that a kogan face @xmath7 is _ reduced _ if the decomposition for @xmath260 obtained this way is reduced
. does nt agree with @xcite : his @xmath260 is our @xmath262 , but this difference does not affect the definition of reduced faces . ] reduced kogan faces of the gelfand
zetlin polytopes are in bijective correspondence with reduced _ pipe - dreams _ ( see @xcite for more details ) .
note that the permutations associated with a face and with the corresponding pipe - dream are the same .
all reduced kogan face diagrams for @xmath263 with the corresponding permutations are shown in figure 2 .
+ all kogan faces are regular .
there is a unique kogan vertex .
this vertex is simple and contained in any other kogan face .
now the result follows from proposition [ p : reg ] . using the 4-term relations
, we can express @xmath264 $ ] through kogan facets : @xmath265=[\g_{0,j}]-[\g_{1,j}]+[\g_{1,j}^-]= [ \g_{0,j}]-[\g_{1,j}]+[\g_{1,j-1}]-[\g_{2,j-1}]+[\g_{2,j-1}^-]=\dots\ ] ] @xmath266-[\g_{i+1,j - i}].\ ] ] define the _
@xmath145-antidiagonal sum of facets @xmath267 _ as the sum of all elements of the form @xmath268 $ ] with @xmath269 being fixed ( including the case @xmath270 ) . set @xmath271 and @xmath272 .
the computation we have just made shows that @xmath273-[\g_{j+1}^-]=ad_{j+1}-ad_j.\ ] ] [ p : dla ] we have the following identities in @xmath1 : @xmath274),\ \
\frac{\d}{\d\l_2}=\pi([\g^-_2]-[\g_2]),\ \dots\ \frac{\d}{\d\l_n}=\pi([\g^-_n]).\ ] ] let @xmath275 be the image of the vector @xmath276 under the natural inclusion @xmath277
. denote by @xmath278 and @xmath279 the support numbers corresponding to facets @xmath280 and @xmath281 , respectively .
thus @xmath278 and @xmath282 are linear functionals on @xmath97 . by the chain rule
, we have @xmath283+\sum_{k=2}^n h^-_k(\d_j)[\g^-_j],\ ] ] in @xmath112 since @xmath284=\d/\d h_j$ ] and similarly for @xmath285 $ ] .
it suffices to note that @xmath286 and @xmath287 , where @xmath288 is the kronecker delta .
for the rest of the paper , we set @xmath289 .
let @xmath290 and @xmath291 , respectively , be the subgroups of upper - triangular and lower - triangular matrices in @xmath8 .
the weyl group of @xmath8 is identified with the symmetric group @xmath292 : a permutation @xmath293 corresponds to the element of @xmath8 acting on the standard basis vectors @xmath294 by the formula @xmath295 . for each @xmath293
, we define the _
schubert variety _
@xmath26 to be the closure of the @xmath291orbit of @xmath30 in the flag variety @xmath296 .
it is easy to check that the length @xmath297 of @xmath30 is equal to the codimension of @xmath26 in @xmath12 .
the class @xmath18 $ ] of @xmath26 in @xmath298 is called the _
schubert cycle _ corresponding to the permutation @xmath30 .
[ r.notation1 ] note that notation in @xcite is different from ours .
namely , they consider the flag variety @xmath299 . under the isomorphism @xmath300 that sends @xmath301 to @xmath302 our schubert variety @xmath26
gets mapped to the schubert variety @xmath303 in notation of @xcite . here
@xmath304 denotes the longest permutation @xmath305 .
we now recall the notion of a _
schubert polynomial _ @xcite . for every elementary transposition @xmath306 ,
define the corresponding divided difference operator ( acting on polynomials in @xmath307 , @xmath308 , @xmath105 ) by the formula @xmath309 where @xmath310 is the polynomial @xmath33 with variables @xmath311 and @xmath312 interchanged .
for a permutation @xmath30 , consider a reduced ( i.e. a shortest ) decomposition @xmath313 of @xmath314 into a product of elementary transpositions .
the schubert polynomial @xmath315 is defined by the formula @xmath316 [ t : bgg ] the class @xmath18 $ ] of the schubert variety @xmath26 in @xmath317 is equal to @xmath318 , where @xmath319 is the negative first chern class of the tautological quotient line bundle @xmath229 . under our identification @xmath320 , we have @xmath321=\sf_w\left(-\frac{\d}{\d\l_1},\dots,-\frac{\d}{\d\l_n}\right).\ ] ] we now recall the fomin kirillov theorem @xcite .
assign to each face @xmath7 the monomial @xmath322 in @xmath307 ,
, @xmath323 by assigning @xmath324 to each equation @xmath246 defining @xmath7 and then multiplying them all ( here the order , of course , does not matter ) .
the fomin
kirillov theorem states that the schubert polynomial @xmath325 of the schubert cycle @xmath18 $ ] is equal to @xmath326 where the sum is only taken over reduced kogan faces .
the polytope ring provides a natural setting , where schubert cycles can be immediately identified with linear combinations of faces sidestepping the use of schubert polynomials .
the following theorem is a direct analog of the fomin
kirillov theorem : it shows that each schubert cycle can be represented by the sum of faces in exactly the same way as the respective schubert polynomial can be represented by the sum of monomials .
[ t.fk ] the schubert cycle @xmath18 $ ] regarded as an element of the gelfand
zetlin polytope ring can be represented by the following linear combination of faces : @xmath321=\pi\left(\sum_{w(f)=w}[f]\right),\ ] ] where the sum is only taken over reduced kogan faces ( all these faces are regular ) .
the proof of this theorem will be given in subsection [ ss.degrees ] .
it uses combinatorics and geometry of the gelfand
zetlin polytope together with the demazure character formula .
despite the similarity between this theorem and the fomin
kirillov theorem , the former can not be formally deduced from the latter .
note that a schubert cycle might have a simpler representation by sums of faces than the one given by this theorem ( see example [ e.facets ] ) .
[ e.facets ] using proposition [ p : dla ] and theorem [ t : bgg ] , we can express the schubert divisors @xmath327 $ ] through the elements of the polytope ring corresponding to facets of @xmath2 .
first , we have @xmath328=\sf_{s_{i_0}}\left(-\frac{\d}{\d\l_1},-\frac{\d}{\d\l_2},\dots\right)= \pi\left(\sum_{j=1}^{i_0}[\g_{j}]-[\g^-_{j}]\right),\ ] ] where we drop all terms , whose indices are out of range .
as we have seen , the element @xmath329-[\g^-_{j}]$ ] equals to @xmath330 .
it follows that @xmath328=\pi(ad_{i_0})=\pi\left(\sum_{j=1}^{i_0}[\g_{i_0-j , j}]\right).\ ] ] we obtain a representation of @xmath331 $ ] as a sum of @xmath332 facets .
this representation coincides with the one given in theorem [ t.fk ] .
note , however , that @xmath333 $ ] can be represented by a single facet , namely , we have @xmath334=-\sum_{i=1}^{n-1}\frac{\d}{\d\l_i}=\frac{\d}{\d\l_n}=\pi[\g^-_n].\ ] ] we have used the equality @xmath335 in @xmath112 because the volume polynomial is translation invariant , in particular , it does not change if we add the same number to all @xmath208 .
theorem [ t.fk ] together with relations in the polytope ring @xmath1 implies the following dual presentation of schubert cycles by faces .
dual kogan faces _ of the gelfand
zetlin polytope to be the faces given by the equations of the type in notation of @xcite , which is the same as type @xmath290 equation in @xcite ] @xmath250 ( i.e. the intersections of facets of the form @xmath336 ) . in other words ,
dual kogan faces are mirror images of kogan faces in a vertical line .
to each dual kogan face @xmath337 we can again assign a permutation @xmath338 .
namely , assign to each equation @xmath250 the simple reflection @xmath339 and compose these reflections by going from the bottom row to the top one and from right to left in each row .
note that the permutation @xmath338 is the same as the permutation @xmath260 for the kogan face @xmath7 obtained as the mirror image of @xmath337 in a vertical line ( that is , each equation @xmath250 is replaced by @xmath340 ) .
[ c.fk ] the schubert cycle @xmath18 $ ] regarded as an element of the gelfand
zetlin polytope ring can be represented by the following linear combination of faces : @xmath321=\pi\left(\sum_{w(f^*)=w_0ww_0^{-1}}[f^*]\right),\ ] ] where the sum is only taken over reduced dual kogan faces .
consider the linear automorphism of @xmath77 that takes a point with coordinates @xmath236 to the point with coordinates @xmath341 .
the automorphism takes a gelfand
zetlin polytope @xmath20 to the gelfand
zetlin polytope @xmath342 , where @xmath343 .
thus it induces an automorphism of the space @xmath92 preserving the lattice @xmath93 and hence an automorphism @xmath258 of @xmath1 .
choose a resolution @xmath81 so that the automorphism @xmath258 extends to @xmath112 .
it is clear that the extended automorphism takes the element @xmath242 $ ] corresponding to a regular face @xmath7 of @xmath2 to the element @xmath344 $ ] , where the face diagram of @xmath337 is obtained from the face diagram of @xmath7 by the mirror reflection in a vertical line .
it now suffices to prove that the automorphism @xmath258 of @xmath1 coincides with the automorphism of @xmath17 that sends a schubert cycle @xmath18 $ ] to @xmath345 $ ] .
( the latter automorphism is induced by the automorphism of @xmath12 that sends a complete flag to the flag of orthogonal complements . ) indeed , this is easy to verify for schubert divisors as in example [ e.facets ] ( we basically need to repeat the same computation with dual kogan faces instead of kogan faces ) .
the general case now follows since schubert divisors are multiplicative generators of the cohomology ring of @xmath12 .
note that any kogan face intersects any dual kogan face transversally .
hence , we can represent the cycles given by the richardson varieties as sums of faces .
[ c.richardson ] the product of any two schubert cycles @xmath18 $ ] and @xmath346 $ ] can be represented as the sum of the following faces : @xmath321\cdot [ x^u]= \pi\left(\sum _ { \genfrac{}{}{0pt}{2}{w(f)=w}{w(f^*)=w_0uw^{-1}_0 } } [ f\cap f^*]\right),\ ] ] where @xmath7 and @xmath337 run over reduced kogan and dual kogan faces , respectively .
for each @xmath347 , consider the affine hyperplane @xmath348 with coordinates @xmath349 ,
, @xmath350 given by the equation @xmath351 , where @xmath352 .
choose coordinates @xmath353 ,
, @xmath354 in @xmath355 such that @xmath356 for @xmath357 .
consider the following linear map @xmath358 from the space @xmath77 with coordinates @xmath236 to the hyperplane @xmath348 : @xmath359 in other terms , if we arrange the coordinates @xmath236 into a triangular table as in @xmath237 , then @xmath360 is the sum of all elements in the @xmath239-th row . in
what follows , we identify @xmath361 with the real span of the weight lattice @xmath362 of @xmath8 so that the @xmath239-th basis vector in @xmath361 corresponds to the weight given by the @xmath239-th entry of the diagonal torus in @xmath8 .
then the hyperplane @xmath355 is the parallel translate of the hyperplane spanned by the roots of @xmath8 .
it is easy to check that the image of the gelfand
zetlin polytope @xmath363 under the map @xmath364 is the weight polytope of the representation @xmath22 .
let @xmath365 be a subset of the gelfand
zetlin polytope @xmath20 ( in what follows @xmath365 will be a face or a union of faces ) .
define the _ character _ @xmath366 of @xmath365 as the sum of formal exponentials @xmath367 over all integer points @xmath368 , that is , @xmath369 the formal exponentials @xmath370 , @xmath371 , generate the group algebra of @xmath362 .
thus the character takes values in this group algebra .
consider the linear operators @xmath372 such that the point @xmath373 differs from the point @xmath374 at most in the @xmath239-th coordinate , and the @xmath239-th coordinate of the point @xmath373 is equal to @xmath375 , where @xmath376 .
it is known and easy to verify that the operators @xmath377 are induced by the orthogonal reflections of @xmath378 ( given by the simple roots ) and that they generate an action of the symmetric group @xmath292 on @xmath355 such that the reflection @xmath377 corresponds to the elementary transposition @xmath306 ( we use the same notation for a reflection and a transposition , which is a standard practice when dealing with group actions ) .
we also define the action of @xmath377 on the group algebra of the weight lattice by setting @xmath379 . in what follows
, we identify @xmath355 with the real vector space spanned by the roots of @xmath8 , and @xmath377 with the reflections corresponding to simple roots .
the simple roots correspond to the standard basis vectors in @xmath355 , i.e. the only nonzero coordinate of the simple root @xmath380 is @xmath360 , and this coordinate is equal to one .
let @xmath381 be the demazure @xmath291module defined as the dual space to the space of global sections @xmath382 , where @xmath383 is the very ample line bundle on @xmath12 corresponding to a strictly dominant weight @xmath23 .
note that by the borel
bott theorem @xmath384 is isomorphic to the irreducible representation @xmath22 of @xmath8 with the highest weight @xmath23 .
choose a basis of weight vectors in @xmath385 .
recall that the _ demazure character _ @xmath386 of @xmath385 is the sum over all weight vectors in the basis of the exponentials of the corresponding weights , or equivalently , @xmath387 where @xmath388 is the multiplicity of the weight @xmath220 in @xmath385 .
the main result of this section establishes a relation between the demazure character of a schubert variety and the character of the union of the corresponding faces .
[ t.demazure ] for each permutation @xmath293 , the demazure character @xmath386 is equal to the character of the following union of faces : @xmath389 as usual , @xmath390 runs only over reduced kogan faces in the gelfand
zetlin polytope @xmath20 .
note that in contrast with theorem [ t.fk ] , this theorem and its corollaries below ( describing hilbert functions and degrees of schubert varieties in projective embeddings to @xmath391 ) use exactly the polytope @xmath20 and not just any of the analogous to @xmath20 polytopes .
whenever the choice of @xmath23 matters we indicate this by using notation @xmath390 instead of @xmath7 for the faces . for kempf permutations , theorem [ t.demazure ]
reduces to ( * ? ? ?
* corollary 15.2 ) .
note that by ( * ? ? ?
* proposition 2.3.2 ) a permutation @xmath30 is kempf if and only if there is a unique reduced kogan face @xmath7 such that @xmath392 . hence , @xmath393 in this case . in subsection [ ss.mitosis ]
, we will reduce this theorem to a purely combinatorial lemma [ t : comb ] .
the proof of this lemma is given in subsection [ ss.keylemma ] .
let us now obtain several corollaries from theorem [ t.demazure ] .
first , we can similarly describe the demazure character of @xmath290modules .
let @xmath394 be the demazure @xmath290module defined as the dual space to @xmath395 , where @xmath396 is the closure of the @xmath290orbit of @xmath30 in @xmath12 ( in particular , @xmath397=[x^{w_0w}]$ ] in @xmath17 ) .
[ c.demazure ] for each permutation @xmath293 , the demazure character @xmath398 is equal to the character of the following union of faces : @xmath399 where @xmath400 runs over reduced dual kogan faces in the gelfand zetlin polytope .
the corollary follows immediately from the proof of theorem [ t.demazure ] together with the definition of dual kogan faces since @xmath401 .
another corollary from theorem [ t.demazure ] describes the hilbert function of the schubert variety @xmath26 embedded into @xmath402 .
[ c.ehrhart ] for any permutation @xmath293 , the dimension of the space @xmath382 is equal to the number of integer points in the union of all reduced kogan faces with permutation @xmath30 : @xmath403 in particular , the hilbert function @xmath404 is equal to the ehrhart polynomial of @xmath405 , that is , @xmath406 for all positive integers @xmath145 .
this corollary will be essential for the proof of theorem [ t.fk ] . to prove theorem [ t.fk ] we first prove an analogous identity for the _ degree polynomials _ of schubert varieties .
the _ degree polynomial _
@xmath407 on @xmath361 is uniquely characterized by the property that @xmath408 for all dominant @xmath409 . in particular ,
@xmath410 and @xmath411 .
the degree polynomials originated in the work of bernstein gelfand gelfand @xcite and were recently studied by postnikov and stanley @xcite .
below we prove identities relating the degree polynomial and the volumes of faces of the gelfand zetlin polytope .
denote by @xmath412 the affine span of a face @xmath7 . in the formulas displayed below , the volume form on @xmath413 is normalized so that the covolume of the lattice @xmath414 in @xmath413 is equal to 1
. then the following theorem holds : [ t.deg]@xmath415 @xmath416 for kempf permutations , the first equality of theorem [ t.deg ] reduces to the last formula from ( * ? ? ?
* corollary 15.2 ) .
theorem [ t.deg ] follows easily from corollary [ c.ehrhart ] and hilbert s theorem describing the leading monomial of the hilbert polynomial by the same arguments as in @xcite .
indeed , by hilbert s theorem , @xmath417 is a polynomial in @xmath145 ( for large @xmath145 ) , and its leading term is equal to @xmath418 .
next , note that @xmath417 is the number of integer points in @xmath419 by corollary [ c.ehrhart ] .
finally , use that the volume of @xmath7 for each face @xmath7 is the leading term in the ehrhart polynomial of this face ( since @xmath420 is approximately equal to the number of integer points in @xmath421 for large @xmath145 ) .
[ r.km ] dual kogan faces are exactly the faces considered in @xcite .
note that the definition of @xmath338 in @xcite is different from ours as well as from that of @xcite .
namely , in our notation they associate to a dual kogan face @xmath337 the permutation @xmath422 .
however , since their schubert cycle @xmath397 $ ] is defined so that it coincides with our schubert cycle @xmath345 $ ] ( see remark [ r.notation1 ] ) their theorem ( * ? ? ?
* theorem 8) ( describing the toric degeneration of the schubert variety @xmath396 ) uses exactly the same faces as our the second equality of theorem [ t.deg ] , and the latter can be deduced from the former by standard arguments from toric geometry .
we now deduce theorem [ t.fk ] from theorem [ t.deg ] using lemma [ l : poincare ] . recall that the lattice @xmath222 is a sublattice of @xmath423 .
in particular , the polytope @xmath20 can be regarded as an element of @xmath424 .
let @xmath425 denote the image of @xmath20 under the canonical projection @xmath426 .
it is easy to check that , under the isomorphism of theorem [ t : iso ] , the class @xmath427 corresponds to the first chern class of the line bundle @xmath428 .
hence , we have the following identity in @xmath1 :
@xmath321\pi(l_\l)^{d - l } = ( \deg_\l x^w)[pt],\ ] ] where @xmath429 is the dimension of the variety @xmath26 ( the product in the left - hand side is taken in @xmath1 ; according to our usual convention , we identify elements of @xmath1 with their images in @xmath101 ) . on the other hand , it is easy to check that for any face @xmath430 of codimension @xmath431 we have that the product @xmath432l_\l^{d - l}$ ] in @xmath112 is equal to @xmath433 times the class of a vertex . hence , by theorem [ t.deg ] we have @xmath321\pi(l_\l)^{d - l}=\pi(\sum_{w(f)=w}[f]l_\l^{d - l}).\ ] ] since elements of the form @xmath434 span @xmath435 , we can apply lemma [ l : poincare ] and conclude that @xmath18=\pi(\sum_{w(f)=w}[f])$ ] . theorem [ t.fk ] is proved .
to prove theorem [ t.demazure ] , we use the demazure character formula for @xmath386 together with a purely combinatorial argument .
we now recall the demazure character formula ( see @xcite for more details ) . for each @xmath210 , define the operator @xmath436 on the group algebra of the weight lattice of @xmath8 by the formula @xmath437 similarly , define the operator @xmath438 by the formula @xmath439 [ t.dem_formula ] let @xmath440 be a reduced decomposition of @xmath30
. then the demazure characters @xmath398 and @xmath441 can be computed as follows : @xmath442 @xmath443 the first identity is the standard form of the demazure character formula .
we will use the second identity , which follows immediately from the first one since @xmath401 and @xmath444 .
note that this theorem is similar to theorem [ t : bgg ] ( and especially to its k - theory version @xcite , see also @xcite ) that describes schubert cycles using the divided difference operators .
however , in the former theorem we apply the operators @xmath445 in the same order as elementary transpositions @xmath446 appear in a reduced decomposition of @xmath30 , while in the latter theorem the order is opposite ( that is , the same as in @xmath447 ) .
_ mitosis _ is a combinatorial operation introduced in @xcite that produces a set of kogan faces out of a kogan face .
if we apply mitosis ( in the @xmath239-th column ) to all reduced kogan faces corresponding to a permutation @xmath30 , then we obtain all reduced kogan faces corresponding to the permutation @xmath448 , provided that @xmath449 .
we will need _ mirror mitosis _
, which is obtained from the mitosis by the transposition of face diagrams ( interchanging rows and columns ) .
in other words , mirror mitosis for @xmath30 is the usual mitosis for @xmath447 .
we use mirror mitosis to deduce theorem [ t.demazure ] from the demazure character formula .
we now give a direct definition of the mirror mitosis .
let @xmath7 be a reduced kogan face of dimension @xmath431 .
for each @xmath450 , we construct a set @xmath451 of reduced kogan faces of dimension @xmath452 as follows . for each @xmath453 , , @xmath454 , we say that the diagram of @xmath7 has an edge in the @xmath239-th row if the face @xmath7 satisfies an equation @xmath455 for some @xmath240 .
similarly , we say that the diagram of @xmath7 has an edge in the @xmath239-th column if there is an equation @xmath456 for some @xmath240 .
consider the @xmath239-th row in the face diagram of @xmath7 .
if it does not have an edge in the first column , then @xmath451 is empty .
suppose now that the @xmath239-th row of @xmath7 contains edges in all columns from the first to the @xmath145-th ( inclusive ) , and does not have an edge in the @xmath457-st column .
then for each @xmath458 we check whether the @xmath459-st row has an edge at the @xmath240-th column .
if it does , we do nothing .
the elements of @xmath451 correspond to the values of @xmath240 , for which there is no edge at the intersection of the @xmath459-st row and the @xmath240-th column . for such value of @xmath240 , we delete the @xmath240-th edge in the @xmath239-th row and shift all edges on the left of it in the same row one step south - east ( to the @xmath459-st row ) whenever possible . a new reduced kogan face
@xmath460 thus obtained is called the @xmath240-th _ offspring _ of @xmath7 at the @xmath239-th row .
the set @xmath451 consists of offsprings @xmath460 for all @xmath461 .
the cardinality of @xmath451 is equal to @xmath462 , where @xmath463 is the number of edges in the first @xmath145 places of the @xmath459-st row .
this is the same as the number of monomials in @xmath464 .
an illustration of mirror mitosis is given in figure 3 .
[ f.mitosis ] + the following theorem follows from the properties of the usual mitosis @xcite : [ t.mitosis ] if @xmath465 , then @xmath466 now follows by backwards induction on @xmath297 from the demazure character formula ( the second identity of theorem [ t.dem_formula ] ) , theorem [ t.mitosis ] and the following lemma .
[ t : comb ] for each permutation @xmath293 and an elementary transposition @xmath377 such that @xmath465 we have @xmath467 the proof of this lemma is purely combinatorial .
it is given in subsection [ ss.keylemma ] .
in this section , we reduce the mitosis on the faces of the gelfand zetlin polytope to an analogous operation ( called _ paramitosis _ ) on the faces of a parallelepiped .
the latter is easier to study and has a transparent geometric meaning ( see remark [ r.simposis ] ) .
paramitosis for parallelepipeds and its applications to exponential sums and demazure operators are studied in subsections [ ss : para ] and [ ss : paramitosis ] .
the material therein is self - contained , and all results are proved by elementary methods .
these results are then used in subsection [ ss.keylemma ] to prove key lemma [ t : comb ] .
another application is proposition [ p.cubic ] that gives a new minimal realization of a simplex as a cubic complex .
consider integer numbers @xmath468 , @xmath105 , @xmath469 , @xmath470 , @xmath105 , @xmath471 , such that @xmath472 for all @xmath473 .
define the parallelepiped @xmath474 as the convex polytope @xmath475 for any parallelepiped @xmath476 , consider the following sum @xmath477 this is a polynomial in @xmath153 .
it can be found explicitly : the proof is a straightforward computation .
proposition [ p : dual ] can be restated in combinatorial terms as follows : the number of ways to represent an integer @xmath482 as a sum @xmath483 , in which @xmath484 for all @xmath473 , is equal to the number of ways to represent the integer @xmath485 in the same form .
fix an integer @xmath486 . consider the following linear operator on the space of laurent polynomials in @xmath153 : with every laurent polynomial @xmath33
, we associate the laurent polynomial @xmath487 obtained from @xmath33 by replacing every power @xmath488 with @xmath489 . in other terms
, we have @xmath490 . clearly , @xmath491 for every laurent polynomial @xmath33 .
the duality property of @xmath492 can be restated as follows : if @xmath493 , then @xmath494 .
for the same value of @xmath486 , define the operator @xmath495 by the formula @xmath496 it is not hard to see that , for every laurent polynomial @xmath33 , the function @xmath497 is also a laurent polynomial .
the operator @xmath495 depends on the parallelepiped @xmath498 .
we have @xmath502 .
therefore , by proposition [ p : sformula ] , @xmath503 and @xmath504 proposition [ p : dual ] applied to @xmath113 gives that @xmath505 .
substituting this into the equation gives the desired result .
[ p : tgen ] consider a linear operator @xmath507 defined over integers such that @xmath508 ( the function @xmath62 in the right hand side is the sum of all coordinate functions on @xmath509 ) .
let @xmath498 , @xmath113 and @xmath495 be as in proposition [ p : spi ] .
assume that @xmath510 and @xmath511 for some subsets
@xmath512 such that the restrictions of @xmath362 to @xmath513 and @xmath514 are injective .
then @xmath515 for every @xmath516 set @xmath517 . since @xmath518 , we obtain that @xmath519 ( these two sums coincide term by term ) , and similarly for the right hand side .
thus the desired statement follows from proposition [ p : spi ] .
let @xmath476 be a coordinate parallelepiped in @xmath520 of dimension @xmath521 , so that @xmath522 for all @xmath523 .
we will now discuss combinatorics of @xmath498 .
for every point @xmath524 with coordinates @xmath525 , we can define the _ paradiagram _ ( `` para '' from parallelepiped ) of @xmath526 as the @xmath521-tuple @xmath527 , where consider a face @xmath7 of @xmath498 .
note that all points in the relative interior of @xmath7 have the same paradiagram .
we will call this paradiagram the paradiagram of @xmath7 . a face @xmath7
is called reduced if so is its paradiagram .
parabox _ as a sequence of consecutive positions in a paradiagram .
a parabox , filled with a string ( possibly empty ) of ones , followed by a single star , followed by a string ( possibly empty ) of zeros , is called an _
intron parabox_. a parabox that contains the left end of a paradiagram and that is filled with a string ( possibly empty ) of zeros is called an _ initial parabox_. a parabox that contains the right end of a paradiagram and that is filled with a string ( possibly empty ) of ones is called a _
final parabox_. it is not hard to see that any reduced paradiagram consists of an initial parabox followed by several ( possibly zero ) intron paraboxes , followed by a final parabox . below is an example of how to split a paradiagram into initial , intron and final paraboxes : two reduced faces @xmath535 and @xmath149 of @xmath498 of the same dimension are said to be related by an _
@xmath536-move _ , if their intersection is a non - reduced facet of both @xmath535 and @xmath149 .
we can also define an @xmath536-move of a reduced paradiagram .
this is the operation that replaces a single string @xmath537 in a paradiagram with the string @xmath538 .
note that an @xmath536-move does not affect the decomposition of a paradiagram into initial , intron and final paraboxes .
let @xmath539 be the paradiagram of @xmath535 , and @xmath540 the paradiagram of @xmath149 . since @xmath541 has codimension 1 in @xmath535 , the paradiagram @xmath542 of @xmath541 is obtained from @xmath539 by replacing one star with either 0 or 1 .
consider two cases .
_ case 1 : _ the star is replaced with @xmath534 . then , since @xmath541 is non - reduced , there must be a @xmath533 immediately before this @xmath534 .
since @xmath149 is reduced , this @xmath533 must be replaced with a star in the paradiagram @xmath540 .
therefore , the paradiagram @xmath539 is obtained from @xmath540 by an @xmath536-move .
_ case 2 : _ the star is replaced with @xmath533 . then , since @xmath541 is non - reduced , there must be a @xmath534 immediately after this @xmath533 .
since @xmath149 is reduced , this @xmath534 must be replaced with a star in the paradiagram @xmath540 .
therefore , the paradiagram @xmath540 is obtained from @xmath539 by an @xmath536-move .
two faces of the same dimension are said to be _
@xmath536-equivalent _ if one of them can be obtained from the other by a sequence of @xmath536-moves or inverse @xmath536-moves ( on the level of paradiagrams , inverse @xmath536-moves are defined as the inverse operations to @xmath536-moves ) . for the sake of brevity , we will write @xmath536-classes instead of @xmath536-equivalence classes . throughout the rest of the subsection , we identify @xmath536-classes of faces and their unions ( clearly , an equivalence class can be easily recovered from its union ) . *
any @xmath536-class is homeomorphic to a closed disk , * there is a one - to - one correspondence between @xmath536-classes and the faces of a simplex such that corresponding sets are homeomorphic , and intersections correspond to intersections .
first consider all reduced vertices .
there are exactly @xmath544 of them .
the paradiagram of a reduced vertex consists of a string of zeros followed by a string of ones .
note that different reduced vertices are never @xmath536-equivalent .
next , consider any @xmath536-class @xmath258 of dimension @xmath145 .
it has @xmath145 intron paraboxes . to the @xmath536-class @xmath258
, we assign a set @xmath545 of @xmath546 vertices in the following way : fill first @xmath547 intron paraboxes with zeros , and the remaining intron paraboxes with ones .
clearly , the set @xmath545 is precisely the set of all reduced vertices contained in the class @xmath258 .
it follows that @xmath548 for any two classes @xmath258 and @xmath290 .
note that @xmath258 is determined by the positions and sizes of initial , intron and final paraboxes , that is , by the set @xmath545 .
the set @xmath545 spans a face of the simplex @xmath549 with vertices at all reduced vertices of @xmath498 .
thus we have an injective map @xmath103 from @xmath536-classes of faces of @xmath498 to faces of @xmath549 ; this map takes intersections to intersections .
the map @xmath103 is surjective : any set of reduced vertices has the form @xmath545 for some equivalence class @xmath258 .
indeed , @xmath258 can be defined as the class , in which the boundaries of intron paraboxes are the boundaries between zeros and ones in the vertices from the given set .
it remains to prove that any @xmath536-class is homeomorphic to a closed disk .
first note that an @xmath536-class with only one intron parabox is a broken line , whose straight line segments are parallel to coordinate axes ( every straight line segment of this broken line corresponds to a particular position of the star inside the intron parabox ) .
a broken line is homeomorphic to the interval . in general ,
an @xmath536-class is a direct product of broken lines as above , hence it is homeomorphic to a direct product of intervals , i.e. to a closed cube .
we can now define _
paramitosis_. this is an operation that produces several faces out of a single face @xmath7 .
if the paradiagram of @xmath7 has no initial parabox , then the paramitosis of @xmath7 is empty .
suppose now that the paradiagram of @xmath7 has a nonempty initial parabox
. then we replace it with an intron parabox : the set of all faces obtained in this way ( corresponding to all different ways of filling the new intron parabox ) is the paramitosis of @xmath7 .
below is an example of paramitosis : [ r.simposis ] it is easy to describe the paramitosis of an @xmath536-class using the bijection between @xmath536-classes and faces of the standard simplex defined in proposition [ p.cubic ] .
namely , the @xmath536-classes with non - empty initial paraboxes correspond to the faces of the simplex contained in a facet @xmath551 .
let @xmath103 be the vertex of the simplex that is not contained in @xmath551 .
then the paramitosis of the face @xmath552 coincides with the convex hull of @xmath258 and @xmath103 .
it follows that paramitosis of an @xmath536-class is again an @xmath536-class and that paramitosis of the intersection of two @xmath536-classes with non - empty initial paraboxes coincides with the intersection of their paramitoses .
[ p : paramitosis ] let @xmath495 be the operator associated with @xmath498 as in subsection [ ss : para ] , the function @xmath555 the sum of all coordinates , and @xmath258 an @xmath536-class of faces in @xmath498 with a non - empty initial parabox .
let @xmath290 be the paramitosis of @xmath258 .
then @xmath556 .
consider the paradiagram of a face from @xmath258 .
suppose that this paradiagram has @xmath557 paraboxes total , and the @xmath431-th parabox starts with index @xmath558 ( so that @xmath559 ) .
consider the following linear map @xmath560 : @xmath561 we have @xmath562 , where @xmath62 is the function computing the sum of all coordinates
. we can now apply proposition [ p : tgen ] to the map @xmath563 .
[ p : nonempty ] let @xmath564 , @xmath105 , @xmath565 be @xmath536-classes with nonempty initial paraboxes , and suppose that the @xmath536-classes @xmath566 , @xmath105 , @xmath567 are obtained from @xmath564 , @xmath105 , @xmath565 by paramitosis
. then we have @xmath568 we will use the inclusion - exclusion formula : @xmath569 where the summation is over all nonempty subsets @xmath570 , and @xmath571 is the intersection of all @xmath572 , @xmath573 .
the same formula holds for @xmath574 , and @xmath495 is linear , hence it suffices to show that @xmath575 .
but @xmath571 is also an @xmath536-class with nonempty initial parabox .
thus the first equality follows from proposition [ p : paramitosis ] .
[ p : paramitosis1 ] suppose that @xmath564 , @xmath105 , @xmath565 are @xmath536-classes with nonempty initial parabox , and @xmath566 , @xmath105 , @xmath578 are @xmath536-classes with empty initial parabox .
suppose also that for every @xmath579 , we have @xmath580 . then @xmath581 by the inclusion - exclusion formula , we have for the right - hand side ( @xmath582 : @xmath583 @xmath584 put @xmath585 for every @xmath579 . since @xmath586 , we have that @xmath587 by the second equality of proposition [ p : nonempty ] . hence , @xmath588 and @xmath589 .
[ r : paramitosis ] note that the condition @xmath592 in lemma [ p : paramitosis1 ] is satisfied whenever @xmath593 for an @xmath536-class @xmath594 .
indeed , if @xmath593 then by definition of paramitosis @xmath595 , where @xmath551 is the hyperplane @xmath499 .
since @xmath551 contains all @xmath572 , we always have the inclusion @xmath596 . on the other hand ,
the condition @xmath594 implies the opposite inclusion @xmath597 .
we now apply the general results for parallelepipeds to mitosis on faces of the gelfand
zetlin polytopes @xmath29 .
fix some @xmath239 .
we will now consider mirror mitosis in the @xmath239-th row ( in what follows , by mitosis , we always mean mirror mitosis ) . define the linear projection @xmath598 forgetting all entries in the @xmath239-th row , i.e. forgetting the values of all coordinates @xmath599 with first index @xmath239 . define the _ fibers of @xmath20 _ as the fibers of this projection restricted to the gelfand
zetlin polytope @xmath20 .
fix the values of all coordinates @xmath600 with @xmath601 .
this defines a fiber of @xmath20 .
the fiber can be given in coordinates @xmath602 by the following inequalities : @xmath603 set @xmath604 and @xmath605 , where @xmath606 ( or sufficiently large negative number ) and @xmath607 ( or sufficiently large positive number ) .
therefore , the fiber can be identified with the coordinate parallelepiped @xmath608 .
let @xmath7 be any reduced kogan face of @xmath20 .
we define the fiber of the face @xmath7 as the intersection of @xmath7 with the fiber of @xmath20
. it will be convenient to represent the fiber of @xmath7 by the @xmath239-th _ fiber diagram _ of @xmath7 , i.e. by the restriction of the face diagram of @xmath7 to the union of rows @xmath609 , @xmath239 , @xmath610 .
note that the mitosis in the @xmath239-th row can be seen on the level of the fiber diagram it does not change other parts of the face diagram . with the fiber diagram of every kogan face
, we can associate a paradiagram of a face of @xmath611 as follows .
the fiber of every kogan face is a face of the parallelepiped @xmath611 , and we take the paradiagram of this face ( note that the length of such paradiagram , which is equal to the dimension of @xmath611 , may be strictly less than @xmath612 ) .
it is easy to check that the paradiagram of a reduced kogan face is also reduced , and that the mirror mitosis on the level of fiber diagrams coincides with the paramitosis on the associated paradiagrams . for the convenience of the reader
, we will now recall the definition of _ ladder - move _ of @xcite in the language of reduced kogan faces .
consider rows @xmath609 , @xmath239 and @xmath610 in the face diagram of @xmath7 .
diagonal _ as a collection of 3 dots in rows @xmath609 , @xmath239 and @xmath610 that are aligned in the direction from northwest to southeast , together with all connecting segments between these 3 dots that belong to the face diagram .
diagonals can be of four possible types : @xmath613 , @xmath614 , @xmath615 and @xmath616 .
the first entry is one if the diagonal contains an interval connecting rows @xmath609 and @xmath239 , otherwise the first entry is zero .
the second entry is one if the diagonal contains an interval connecting rows @xmath239 and @xmath610 , otherwise the second entry is zero : the correspondence between fiber diagrams and paradiagrams can now be described in combinatorial terms as follows : diagonals of type @xmath613 , @xmath614 , @xmath615 are replaced by @xmath617 , @xmath533 , @xmath534 , respectively , and diagonals of type @xmath616 are ignored ( each such diagonal decreases by one the dimension of the parallelepiped @xmath611 , that is , the length of the paradiagram ) .
for instance , the 1st fiber diagram of the upper face on figure 3 yields the paradiagram .
define a _ box _ as any sequence of consecutive diagonals in a fiber diagram . in our pictures ,
a box will look like a parallelogram with angles @xmath618 and @xmath619 . by definition , a _ ladder - movable box _ is a box , whose first ( left - most ) diagonal is of type ( 0,0 ) , which follows by any number of type @xmath616 diagonals , and , finally , by a single diagonal of type @xmath615 .
symbolically , we represent such a box as a sum @xmath620 , where @xmath145 is the number of type @xmath616 diagonals .
the _ ladder - move _ of @xcite makes this ladder - movable box into the box @xmath621 : note that ladder moves do not change the permutation associated with a face .
moreover , they take reduced faces to reduced faces .
finally , note that under the correspondence between fiber diagrams and paradiagrams , the ladder - moves are exactly the @xmath536-moves of the previous subsection .
we are now ready to prove key lemma [ t : comb ] .
denote by @xmath622 the set @xmath623 and by @xmath624 the union of all faces that can be obtained from faces in @xmath622 by the mirror mitosis in the @xmath239-th row .
these are sets considered in lemma [ t : comb ] , and to prove the lemma we have to show that @xmath625 [ l : fiber1 ] let @xmath495 be the operator associated with the coordinate parallelepiped @xmath611 as in subsection [ ss : para ] . identifying @xmath626 and @xmath627 with the subsets of @xmath611
, we get @xmath628 note that @xmath629 is a single point @xmath630 ( i.e. all coordinates in all rows except for row @xmath239 are fixed ) . for a point @xmath631 , denote by @xmath632 the coordinates of @xmath526 in row @xmath239 .
let @xmath633 ( for @xmath634 ) be the sum of coordinates in row @xmath240 . by definitions of @xmath635 and @xmath495
, the following identity holds for all @xmath636 after substituting @xmath637 : @xmath638 in lemma [ l : fiber1 ] , replace @xmath153 with @xmath639 , and multiply both parts by the product @xmath640 to obtain key lemma [ t : comb ] , it now suffices to take the sum over all fibers @xmath627 of @xmath624 at the @xmath239-th row .
lemma [ l : fiber1 ] will follow from proposition [ p : paramitosis1 ] once we check that @xmath626 satisfies the hypothesis of the proposition . we know that @xmath626 is closed under @xmath536-moves since @xmath622 is closed under ladder - moves
we can split @xmath626 into the union of @xmath536-classes @xmath641 where @xmath572 and @xmath642 , respectively , have nonempty and empty initial parabox . by remark [ r : paramitosis ]
it suffices to show for each @xmath643 that @xmath644 for some @xmath645 .
this follows from the lemma : [ l.global ] let @xmath7 be a reduced kogan face such that @xmath392 and the @xmath239-th paradiagram of @xmath7 has empty initial parabox .
if @xmath465 , then there exists another reduced kogan face @xmath162 such that @xmath646 , the @xmath239-th paradiagram of @xmath162 has nonempty initial parabox , and @xmath647 is contained in @xmath648 .
recall that the face diagram of @xmath7 defines a reduced decomposition @xmath649 , which by definition splits into two reduced words @xmath650 and @xmath651 as follows
. the word @xmath652 is composed from elementary transpositions by going from the bottom row to the @xmath239-th row inclusively , and @xmath653 is composed by going from the @xmath654-st row to the top row .
in particular , the word @xmath650 contains only @xmath655 with @xmath656 . since the initial parabox of @xmath7 is empty , the fiber diagram of @xmath7 starts with a sequence of length @xmath657 of type @xmath616 diagonals followed by a type @xmath613 or @xmath614 diagonal . if @xmath658 , then the word @xmath650 contains only @xmath655 with @xmath659 , in particular , @xmath660 and @xmath661 is an inversion for @xmath650 unless @xmath662 .
hence , the hypothesis @xmath663 ( which is equivalent to @xmath664 ) implies that @xmath665 .
indeed , @xmath666 ( the last inequality holds because @xmath667 is reduced ) .
if @xmath668 , the word @xmath650 can be further split as @xmath669 , where @xmath670 contains only @xmath655 with @xmath659 and @xmath671 contains only @xmath655 with @xmath672 . by similar arguments we deduce that @xmath673 ( we use the identity @xmath674 ) . by applying to @xmath653 the exchange property we can replace it with a reduced word @xmath675
we now replace @xmath7 with a reduced face @xmath162 with the same permutation and non - empty initial parabox as follows . in the face diagram of @xmath7 , delete the edge corresponding to @xmath676 and add the new edge @xmath677 .
the resulting face diagram defines the face @xmath162 . by construction
, @xmath648 contains @xmath647 .
we now return to the proof of lemma [ l : fiber1 ] .
apply lemma [ l.global ] to a reduced kogan face @xmath678 whose paradiagram @xmath290 lies in @xmath642 .
we get a face @xmath679 such that its paradiagram @xmath258 lies in @xmath680 and @xmath681 .
hence , the @xmath536-equivalence class @xmath682 of @xmath258 also lies in @xmath680 and @xmath683 as desired . | we describe a new approach to the schubert calculus on complete flag varieties using the volume polynomial associated with gelfand zetlin polytopes .
this approach allows us to compute the intersection products of schubert cycles by intersecting faces of a polytope . |
the pioneering discovery by pareto @xcite is not only an important cue for research in fractal but also a significant issue in economics .
the pareto law states that a cumulative number distribution @xmath0 obeys power - law for income @xmath1 which is larger than a certain observational threshold @xmath2 : @xmath3 where the exponent @xmath4 is called pareto index .
recently pareto s law is checked with high accuracy by using digitalized data @xcite .
although the number of persons or companies in the range where pareto s law ( [ pareto ] ) holds is a few percent , the amount of the income occupies a large part of the total income .
high income persons or high income companies can , therefore , influence economics .
it is quite important to investigate the mechanism which governs them . the research on the income distribution is well investigated in econophysics @xcite .
many models to explain the distribution are proposed ( for instance recently refs .
furthermore , fujiwara et al .
@xcite find that , without assuming any model , pareto s law can be derived kinematically from the law of detailed balance and gibrat s law @xcite which are observed in high quality digitalized data .
they derive not only pareto s law but also the reflection law which correlates a positive growth rate distribution to negative one by the pareto index @xmath4 .
this means that pareto s law and index are explained by the law of detailed balance and gibrat s law .
we should note that these arguments are limited in the high income region ( @xmath5 ) . the findings in ref .
@xcite are quite fascinating especially in the idea that the pareto index can be understood by the growth rate distribution . in this paper
, we first investigate the application as follows .
the distribution of high income companies follows pareto s law and the pareto index is about @xmath6 .
this is called zipf s law @xcite .
it is also known that the distributions in most job categories follow pareto s law .
those pareto indices , however , scatter around @xmath6 @xcite .
we examine the reflection law that relates the change of the pareto index to the change of the growth rate distribution in job categories .
there is , however , a problem in estimating the pareto index by the growth rate distribution in each job category .
the number of companies in several job classes is not enough to observe the reflection law .
we clear this difficulty by employing profits data which have no threshold in contrast to high income data announced publicly . by observing all positive profits data , we confirm the law of detailed balance .
we also find that gibrat s law holds in the region where the only initial profits are conditioned to be larger than the threshold . in this case ,
pareto s law is also obtained and the reflection law can be observed in companies data classified into job categories . using these arguments , we confirm the correlation between the pareto index and the growth rate distribution .
in addition , we find that companies in a small scale job category have more possibilities of growing profits than those in a large scale job category .
this kinematically explains that the pareto index for the companies in the small job class is larger than that for the companies in the large scale job class , which is reported in ref .
in this section , we briefly review the proof that the detailed balance and gibrat s law lead to pareto s law and the reflection law @xcite . let the income at two successive points in time be denoted by @xmath7 and @xmath8 .
growth rate @xmath9 is defined as the ratio @xmath10 .
the detailed balance and gibrat s law can be summarized as follows .
* detailed balance ( time - reversal symmetry ) + the joint probability distribution function ( pdf ) @xmath11 is symmetric : @xmath12 * gibrat s law + the conditional probability distribution of growth rate @xmath13 is independent of the initial value @xmath7 : @xmath14 where @xmath13 is defined by using the pdf @xmath15 and the joint pdf @xmath16 as @xmath17 these phenomenological properties , for example , are observed in the data of high income japanese companies in 2002 and 2003 . in japan , companies having annual income more than 40 million yen are announced publicly as `` high income companies '' every year .
the database is published by diamond inc . in fig .
[ income2002vsincome2003 ] , all the companies are plotted , the income of which in 2002 ( @xmath7 ) and 2003 ( @xmath8 ) exceeded 40 million yen ( @xmath2 ) : @xmath18 the number of companies which satisfy this condition is @xmath19 . in fig .
[ income2002vsincome2003 ] , the time - reversal symmetry ( [ detailed balance ] ) is apparent . in fig .
[ incomedistribution ] , we confirm pareto s law ( [ pareto ] ) for the companies , the income of which satisfies the condition ( [ income_condition ] ) .
the pareto indices @xmath4 are very close to @xmath6 , and the case is often referred to as zipf s law @xcite .
we also examine gibrat s law in the regime ( [ income_condition ] ) . here
we divide the range of @xmath7 into logarithmically equal bins as @xmath20 $ ] with @xmath21 . in fig .
[ incomegrowthrate ] , the probability density for @xmath22 is expressed for each bin . in all the cases ,
all the probability density distributions for different @xmath23 collapse on a single curve approximately .
this means that the distribution for the growth rate @xmath22 is approximately independent of the initial value @xmath7 . here
the probability density for @xmath22 defined by @xmath24 is related to that for @xmath9 by @xmath25 pareto s law ( [ pareto ] ) can be derived from the detailed balance ( [ detailed balance ] ) and gibrat s law ( [ gibrat ] ) without assuming any model @xcite . due to the relation of @xmath26 under the change of variables from @xmath27 to @xmath28 , these two joint pdf s
are related to each other , @xmath29 by the use of this relation , the detailed balance ( [ detailed balance ] ) is rewritten in terms of @xmath16 as follows : @xmath30 substituting the joint pdf @xmath16 for the conditional probability @xmath13 defined in eq .
( [ conditional ] ) , the detailed balance is expressed as @xmath31 from the detailed balance and gibrat s law ( [ gibrat ] ) , one finds the following : @xmath32 by expanding this equation around @xmath33 , the following differential equation is obtained @xmath34 where @xmath1 denotes @xmath7 and @xmath35 is defined as @xmath36 .
the solution is given by @xmath37 with @xmath38 .
this is pareto s law , which leads to the cumulative representation in eq .
( [ pareto ] ) as follows : @xmath39 . from eqs .
( [ pareto2 ] ) and ( [ dandg ] ) , one finds the relation between positive growth rate ( @xmath40 ) and negative one ( @xmath41 ) , @xmath42 in terms of @xmath24 in eq .
( [ qandq ] ) , this is written as @xmath43 in fig .
[ incomegrowthrate ] , we examine this relation , which is called the reflection law in ref .
@xcite . by the use of the reflection law to the best - fit line @xcite @xmath44
we obtain the line @xmath45 with @xmath46 where @xmath47 .
the reflected line ( [ rline ] ) seems to match the data in fig .
[ incomegrowthrate ] @xcite .
from the argument in the preceding section , we expect the estimation of the pareto index by using eq .
( [ mu ] ) under the detailed balance and gibrat s law .
there is , however , a difficulty in measuring the slope @xmath48 of the reflected line ( [ rline ] ) .
the reason is that the data for large negative growth , @xmath49 , are not available due to the limit ( [ income_condition ] ) ( fig . [ incomegrowthrate ] ) .
the reflected line ( [ rline ] ) does not match the total data which are not classified into bins for instance ( fig .
[ incomegrowthratetotal ] ) .
indeed , the pareto index is estimated to be about @xmath50 from the best fit lines ( [ line ] ) for @xmath51 and ( [ rline ] ) for @xmath52 . in order to avoid this problem ,
we must remove the threshold @xmath2 in the region ( [ income_condition ] ) .
this is , however , impossible as long as we deal with the income data announced publicly in japan .
we employ profits as another economic quantity similar to the income .
although profits of companies are not announced publicly , the data are available on `` cd eyes '' published by tokyo shoko research , ltd .
@xcite for instance .
this database contains financial information of over 250 thousand companies in japan .
we deal with the companies , the profits of which in 2002 ( @xmath7 ) and 2003 ( @xmath8 ) exceeded 0 yen : @xmath53 in fig .
[ profit2002vsprofit2003 ] , all the companies which satisfy this condition are plotted , the number of which is @xmath54 .
the time - reversal symmetry ( [ detailed balance ] ) is apparent even in fig .
[ profit2002vsprofit2003 ] .
we desire to examine gibrat s law without the threshold .
the condition is , therefore , employed : @xmath55 where @xmath2 is 40 million yen .
the number of the companies which satisfy this limit is @xmath56 . in fig .
[ profitdistribution ] , pareto s law ( [ pareto ] ) for the companies is observed in the region ( [ income_condition ] ) , not in ( [ profit_condition2 ] ) .
let us examine gibrat s law in the regime ( [ profit_condition2 ] ) .
we classify the data for the high profits companies , @xmath57 , into five bins in the same manner in the preceding section .
the result is shown in fig .
[ profitgrowthrate ] .
we confirm gibrat s law without the forbid region caused by the threshold . under the extended limit ( [ profit_condition2 ] )
, we can derive pareto s law . in the derivation
, we should pay attention to the region where gibrat s law holds as follows : @xmath58 in the region ( [ income_condition ] ) where these two conditions are satisfying , eq .
( [ dandg ] ) is relevant and pareto s law ( [ pareto2 ] ) holds as a result .
this is observed in fig .
[ profitdistribution ] .
the reflection law ( [ reflection ] ) is relevant in the region ( [ income_condition ] ) strictly .
here we assume that eqs .
( [ reflection ] ) and ( [ reflection2 ] ) hold approximately near @xmath33 ( @xmath59 ) even in the region ( [ profit_condition2 ] ) , the validity of which should be checked against the results . under this assumption
, we evaluate the pareto index by using eq .
( [ mu ] ) . in fig .
[ profitgrowthratetotal ] , the pareto index is estimated to be @xmath60 from the best fit lines ( [ line ] ) for @xmath51 and ( [ rline ] ) for @xmath52 for the total data which are not classified into bins . this is consistent with the assumption .
we should comment that the pareto index is estimated to be about @xmath50 for the companies , the profits of which in 2002 ( @xmath7 ) and 2003 ( @xmath8 ) exceeded @xmath2 .
it is reported that income distributions in most job categories follow pareto s law , but those pareto indices scatter around @xmath6 @xcite . the same phenomenon is observed in profits distributions . in this section , we classify the companies into job categories in order to verify the results in the preceding section . in japan , companies are categorized by japan standard industrial classification ( jsic ) . this industrial classification is composed of four stages , namely divisions , major groups , groups and details ( industries ) .
the composition includes 14 divisions , 99 major groups , 463 groups and 1,322 industries .
the classification in the database `` cd eyes '' follows jsic .
we classify the @xmath56 companies into the 14 divisions , the profits of which satisfy the condition ( [ profit_condition2 ] ) . because the number of companies classified into 99 major groups is not enough to investigate the growth rate distribution .
the name of each division and the number of companies are follows , a : agriculture ( 84 ) , b : forestry ( 1 ) , c : fisheries ( 14 ) , d : mining ( 3,512 ) , e : construction ( 1,182 ) , f : manufacturing ( 6,428 ) , g : electricity , gas , heat supply and water ( 218 ) , h : transport and communications ( 2,300 ) , i : wholesale , retail trade and eating @xmath61 drinking places ( 8,699 ) , j : finance and insurance ( 1,994 ) , k : real estate ( 293 ) , l : services ( 3,919 ) , m : government , n.e.c .
( 0 ) and n : industries unable to classify ( 0 ) . in each job division , we measure @xmath62 and @xmath48 of the best fit lines ( [ line ] ) and ( [ rline ] ) of all the companies in the class . the pareto index can be estimated from the relation ( [ mu ] ) by assuming the detailed balance and gibrat s law . here
we have excluded a , b , c , m and n divisions , because the number of companies in each division is not enough to observe the growth rate distribution . by comparing the pareto indices estimated by the growth rate with those directly measured in the power - law
, we reconfirm the relation ( [ mu ] ) in fig .
[ paretoindexvstau ] . after identifying the pareto index change as the change of the growth rate distribution in job categories
, we consider an economic quantity which influences the growth rate . in balance - sheet ,
the sum of capital stock and retained earnings is equity capital , and the sum of equity capital and debt is assets .
it is natural to consider that a company works based on the assets and gets profits ( @xmath63 income ) .
the positive correlation between assets and income is reported in ref .
it is also pointed out that assets and sales correlate positively in the same reference . on the other hand ,
the positive correlation between sales and capital stock is observed in the database `` cd eyes '' .
consequently there is a positive correlation between assets and capital stock .
we employ , therefore , average `` capital stock '' as a quantity which statistically characterizes the scale of companies in each job category . in ref .
@xcite , it is reported that there is a relation between the average capital stock of high income companies and the pareto index in 99 major groups .
the larger the average capital becomes , the smaller the pareto index becomes .
the same phenomenon is observed in profits distributions ( fig .
[ averagecapitalvsparetoindex ] ) .
this means that the pareto index is related to the scale of companies in job categories .
we can understand the kinematics by the relation between the average capital and the slope of growth rate distribution @xmath64 . in fig .
[ averagecapitalvstau ] , @xmath62 decreases with the average capital and @xmath48 hardly responds to that .
the companies in the small scale job category ( mining for instance ) have more possibilities of growing profits than the companies in the large scale job category ( finance and insurance for instance ) . on the other hand ,
the possibility of growing down is independent of the scale of the job category for high profits companies , @xmath65 . as a result ,
the relation ( [ mu ] ) explains that the pareto index for the companies in the small scale job category is larger than that for the companies in the large scale job category .
in this paper , we have first shown that the pareto index can be estimated by profits growth rate distribution of companies in japan .
the point is that the data have no threshold ( @xmath2 ) in contrast to the high income data .
we find that the law of detailed balance and gibrat s law hold in the extended region @xmath66 and @xmath67 , respectively . by classifying companies into 14 job divisions ,
we have secondly confirmed the reflection law that relates the change of the pareto index to the change of the growth rate distribution .
we thirdly find that the companies in the small scale job category have more possibilities of growing profits than the companies in the large scale job category by observing the relation between the average capital and the slope of growth rate distribution .
the relation between the average capital and the pareto index is also verified .
consequently it is kinematically explained that the pareto index for the companies in the small scale job category is larger than that for the companies in the large scale job category .
we have concentrated our attention to the region ( @xmath57 , @xmath68 ) in order to examine gibrat s law . beyond this regime ,
the breakdown of gibrat s law is studied , the recent works about that are done by stanley s group @xcite . in this paper , we find the law of detailed balance over the regime where pareto s law holds .
we expect that the distribution in the region ( @xmath69 , @xmath68 ) can be derived kinematically from the law of detailed balance and the results found by stanley s group
. this argument will kinematically decide the distribution for the middle region where pareto s law does not hold .
the author is grateful to the yukawa institute for theoretical physics at kyoto university , where this work was initiated during the yitp - w-03 - 03 on `` econophysics - physics - based approach to economic and social phenomena - '' , and specially to professor h. aoyama for useful comments .
thanks are also due to professor h. kasama for careful reading of the manuscript . 99 v. pareto , cours deconomique politique , macmillan , london , 1897 . h. aoyama , w. souma , y. nagahara , h.p .
okazaki , h. takayasu and m. takayasu , cond - mat/0006038 , fractals 8 ( 2000 ) 293 ; + w. souma , cond - mat/0011373 , fractals 9 ( 2001 ) 463 .
mategna and h.e .
stanley , an introduction to econophysics , cambridge university press , uk , 2000 . h. takayasu , m. takayasu , m.p .
okazaki , k. marumo and t. shimizu , cond - mat/0008057 , in : m.m .
novak ( ed . ) , paradigms of complexity , world scientific , 2000 , p. 243 .
t. mizuno , m. katori , h. takayasu and m. takayasu , cond - mat/0308365 , in : h. takayasu ( ed . ) , empirical science of financial fluctuations : the advent of econophysics , vol .
321 , springer , tokyo , 2003 .
m. anazawa , a. ishikawa , t. suzuki and m. tomoyose , cond - mat/0307116 , physica a335 ( 2004 ) 616 ; + a. ishikawa and t. suzuki , cond - mat/0403070 , physica a343 ( 2004 ) 376 . m. nirei and w. souma , sfi/0410029 .
y. fujiwara , w. souma , h. aoyama , t. kaizoji and m. aoki , cond - mat/0208398 , physica a321 ( 2003 ) 598 ; + h. aoyama , w. souma and y. fujiwara , physica a324 ( 2003 ) 352 : + y. fujiwara , c.d .
guilmi , h. aoyama , m. gallegati and w. souma , cond - mat/0310061 , physica a335 ( 2004 ) 197 ; + y. fujiwara , h. aoyama , c.d .
guilmi , w. souma and m. gallegati , physica a344 ( 2004 ) 112 ; + h. aoyama , y. fujiwara and w. souma , physica a344 ( 2004 ) 117 .
r. gibrat , les inegalites economiques , paris , sirey , 1932 .
gipf , human behavior and the principle of least effort , addison - wesley , cambridge , 1949 .
k. okuyama , m. takayasu and h. takayasu , physica a269 ( 1999 ) 125 .
a. ishikawa , cond - mat/0409145 , physica a349 ( 2005 ) 597 .
tokyo shoko research , ltd .
, http://www.tsr - net.co.jp/. m.h.r .
stanley , l.a.n .
amaral , s.v .
buldyrev , s. havlin , h. leschhorn , p. maass , m.a . salinger and h.e .
stanley , nature 379 ( 1996 ) 804 . l.a.n .
amaral , s.v .
buldyrev , s. havlin , h. leschhorn , p. maass , m.a .
salinger , h.e .
stanley and m.h.r .
stanley , j. phys .
( france ) i7 ( 1997 ) 621 . s.v .
buldyrev , l.a.n .
amaral , s. havlin , h. leschhorn , p. maass , m.a .
salinger , h.e .
stanley and m.h.r .
stanley , j. phys .
( france ) i7 ( 1997 ) 635 . l.a.n .
amaral , s.v .
buldyrev , s. havlin , m.a . salinger and h.e .
stanley , phys .
80 ( 1998 ) 1385 .
y. lee , l.a.n .
amaral , d. canning , m. meyer and h.e .
stanley , phys .
81 ( 1998 ) 3275 .
d. canning , l.a.n .
amaral , y. lee , m. meyer and h.e .
stanley , economics lett .
60 ( 1998 ) 335 . | employing profits data of japanese companies in 2002 and 2003 , we confirm that pareto s law and the pareto index are derived from the law of detailed balance and gibrat s law .
the last two laws are observed beyond the region where pareto s law holds . by classifying companies into job categories
, we find that companies in a small scale job category have more possibilities of growing than those in a large scale job category .
this kinematically explains that the pareto index for the companies in the small scale job class is larger than that for the companies in the large scale job class .
pacs code : 04.60.nc + keywords : econophysics ; pareto law ; gibrat law ; detailed balance ; reflection law |
a great deal of attention has recently been directed to graphene , a monolayer of carbon atoms , both experimentally @xcite and theoretically @xcite .
graphene is marked with its novel charge carriers that behave like massless dirac fermions with effective speed of light @xmath2
m / s @xmath3 .
it thus provides a special opportunity to study @xmath1relativistic " quantum dynamics in condensed - matter systems .
experiments have revealed a number of exotic transport properties of graphene , such as the half - integer quantum hall ( qh ) effect and minimal conductivity .
the dynamics of dirac fermions becomes particularly interesting in a magnetic field and leads to peculiar quantum phenomena , such as fermion number fractionalization @xcite and spectral asymmetry , @xcite intimately tied to the chiral anomaly in 1 + 1 dimensions .
actually the half - integer qh effect and the presence of the zero - energy landau levels observed @xcite in graphene are a manifestation of fermion number fractionalization .
it would be important to explore further possible signatures of relativistic quantum field theory in the low - energy physics of graphene .
an interesting proposal @xcite along this direction is to simulate the klein paradox @xcite ( or tunneling ) in graphene .
calculations @xcite of the dielectric function also reveal that the electromagnetic response of graphene is substantially different from that of conventional two - dimensional systems .
the difference becomes even prominent under a strong magnetic field .
@xcite in particular , for graphene the vacuum state is a dielectric medium and carries an appreciable amount of electric and magnetic susceptibilities over all range of wavelengths ; this reflects the presence of the @xmath1dirac sea " .
curiously the zero - energy landau levels , though distinct in hall conductance , hardly contribute to the susceptibilities .
the purpose of this paper is to study further aspects of the response of graphene in a magnetic field at integer filling factors @xmath0 , with focus on the static structure factor @xmath4 , which is directly related to the cross section in inelastic light scattering by graphene . in particular , we point out that graphene has unusual characteristics of electronic correlations at short distances , which derive from the @xmath1relativistic " nature of massless quasiparticles .
it is also shown that the zero - energy landau levels remain indistinguishable in spectral weight @xmath5 from the vacuum state . in sec .
ii we briefly review the low - energy effective theory of graphene in a magnetic field and derive , as a preliminary , the structure factor for conventional qh systems . in sec .
iii we study the structure factor for graphene .
section iv is devoted to a summary and discussion .
graphene has a honeycomb lattice consisting of two triangle sublattices of carbon atoms with one electron per site .
it is a gapless semiconductor and its low - energy electronic transport is described by an effective hamiltonian of the form @xcite @xmath6 , \nonumber\\ { \cal h}_{\pm}&= & v_{\rm f}\ , ( \sigma_{1}\pi_{1 } + \sigma_{2}\pi_{2 } \pm m \sigma_{3 } ) - e a_{0 } , \label{hzero}\end{aligned}\ ] ] where @xmath7 [ @xmath8 or @xmath9 includes coupling to external electromagnetic potentials @xmath10 ; @xmath11 m / s is the fermi velocity . the two - component spinors @xmath12 and @xmath13 stand for the electron fields near the two inequivalent fermi points ( @xmath14 and @xmath15 ) where the spectrum becomes linear ; @xmath16 reside on the same sublattice and @xmath17 on another . for generality
we have introduced a tiny @xmath1mass " gap @xmath18 which spoils the valley su(2 ) symmetry of @xmath19 .
actually the observed @xmath20 hall plateaus @xcite suggest such a tiny mass gap .
@xcite we keep @xmath21 to clarify the particle - hole character of the lowest landau levels , but practically set @xmath22 .
we suppress the electron spin , which is treated as a global su(2 ) symmetry of @xmath19 by doubling the fields @xmath23 and @xmath24 .
the zeeman splitting , though ignored for simplicity , is readily incorporated .
the coulomb interaction is written as @xmath25 where @xmath26 is the fourier transform of the electron number density @xmath27 ; @xmath28 is the coulomb potential with the fine - structure constant @xmath29 and the substrate dielectric constant @xmath30 .
we shall discuss the effect of @xmath31 later .
let us place graphene in a strong magnetic field and study how the electrons in graphene respond to a weak potential @xmath32 .
we set @xmath33 to supply a uniform magnetic field @xmath34 normal to the sample plane .
when @xmath35 , the eigenmodes of @xmath19 are landau levels of @xmath23 and @xmath24 of energy @xmath36 labeled by integers @xmath37 , and @xmath38 ( or @xmath39 with the magnetic length @xmath40 ) ; @xmath41 is the basic cyclotron frequency . here
@xmath42 specifies the sign of the energy @xmath43 . for @xmath44 , @xmath23 and
@xmath24 have the same spectrum symmetric about @xmath45 .
the @xmath46 level of @xmath23 has negative energy @xmath47 while that of @xmath24 has positive energy @xmath48 ; these @xmath49 levels represent holes and electrons via quantization . with the electron spin taken into account ,
each landau level is thus four - fold degenerate , except for the doubly - degenerate @xmath50 levels .
the @xmath50 eigenmodes have components only on each separate sublattice . to make this landau - level structure explicit , it is useful to pass to the @xmath51 basis , with the expansion @xcite @xmath52 .
( from now on , we shall only display the @xmath23 sector since the @xmath24 sector is obtained by reversing the sign of @xmath53 . )
the hamiltonian @xmath19 thereby is rewritten as @xmath54 and the charge density @xmath55 as @xcite @xmath56 with the coefficient matrix @xmath57 \label{gpsikn}\end{aligned}\ ] ] and @xmath58 ; @xmath59 stands for the center coordinate with uncertainty @xmath60 = i\ell^{2}$ ] . here
the coefficient functions @xmath61 are defined in terms of the harmonic oscillator eigenstates @xmath62 with @xmath63 and @xmath64=1 $ ] ; @xmath65 and @xmath66 . more explicitly ,
@xcite @xmath67 for @xmath68 , and @xmath69^{\dag}$ ] .
actually @xmath70 are the coefficient functions that characterize the charge density for the `` nonrelativistic '' hall electrons @xmath71 expressed in terms of their eigenmodes @xmath72 with energy @xmath73 .
let us first consider , as an exercise , the static structure factor or the spectral weight @xmath74 for a state @xmath75 of free `` nonrelativistic '' hall electrons with integer filling factor @xmath0 ( per spin ) .
one may use eq .
( [ rhonr ] ) and note , in taking the matrix element @xmath76 , that @xmath77 for unoccupied levels @xmath78 so that the result is proportional to the number of electrons per occupied level @xmath79 .
( note here that @xmath80 with @xmath81 . )
this yields the spectral weight @xmath82 for @xmath83 , where @xmath84 denotes the total area .
let us recall that the static structure factor @xcite @xmath85 is defined by @xmath86 with its @xmath87 component isolated , where @xmath88 stands for the total electron number .
accordingly , in general , @xmath89 owing to charge conservation . noting the formula @xmath90 allows one to cast eq .
( [ rhorhonr ] ) into the structure factor at filling factor @xmath0 @xmath91 which agrees with a known result .
@xcite in particular , @xmath92 note first that @xmath93 since @xmath94 .
note also that @xmath95 as @xmath96 for all @xmath0 . to see what this means
let us recall the following : for a collection of classical particles , @xmath97 is written as @xcite @xmath98 for @xmath99 , i.e. , as an average over the relative positions @xmath100 of particles surrounding a given particle . as a result ,
@xmath95 for @xmath96 if @xmath101 , e.g. , for particles formulated on a lattice
. such behavior of @xmath97 thus implies the absence of particle correlations at short distances .
in this section we study the case of graphene . in the present treatment the @xmath23 and @xmath24 sectors are independent and the spectral weight @xmath102 with @xmath103 is given by the sum @xmath104 for @xmath99 . for the @xmath23 sector one may simply replace , in eq .
( [ rhorhonr ] ) , @xmath105 by @xmath106 of eq .
( [ gpsikn ] ) and note that the level indices @xmath107 now run over all integers , @xmath108 .
analogously , @xmath109 for the @xmath24 sector is obtained from @xmath106 by setting @xmath110 ( i.e. , @xmath111 ) , or equivalently @xmath112 let us denote @xmath113 and @xmath114 for short .
they actually are functions of @xmath115 , are thus symmetric in @xmath107 , and have the property @xmath116 with @xmath117 .
the spectral weight is now written as @xmath118 where @xmath119 ; @xmath120 stands for the spin degeneracy .
one can write @xmath121 as @xmath122 when the @xmath23 landau levels are occupied up to the @xmath123th level @xmath124 ; analogously for @xmath125 .
note that eq .
( [ gpsigchi ] ) relates @xmath121 and @xmath126 so that @xmath127 the equalities @xmath128 and @xmath129 then imply the following : ( i ) the @xmath23 and @xmath24 sectors contribute equally to the vacuum spectral weight , @xmath130 .
( ii ) the spectral weight @xmath5 is the same for the charge - conjugate states with @xmath131 . in view of this
we shall focus on the case @xmath132 from now on . in the limit @xmath22 of practical interest ,
distinctions between @xmath133 and @xmath134 disappear : @xmath135 for @xmath136 and @xmath137 for @xmath138 , as seen from eq .
( [ gpsikn ] ) .
this yields @xmath139 in particular , @xmath140 shows that the graphene vacuum and the @xmath141 states , all of zero energy , have the same spectral weight @xmath142 .
actually we earlier noted such degeneracy in electric susceptibility for the @xmath143 and @xmath141 states , @xcite and these states now turn out to be indistinguishable in both the real and imaginary parts of density response ; we discuss this point in more detail later . to proceed further let us note some basic properties of the transition rates @xmath144 with @xmath145 .
it is clear from eq .
( [ gammapsichi ] ) that the transition rates among the positive - energy states and those among the negative - energy states are essentially the same whereas they are different from the rates across the dirac sea ( @xmath146 with @xmath147 ) .
this feature is seen from fig . 1 ( a ) , which illustrates typical profiles of @xmath148 for @xmath149 .
in general , @xmath150 is significantly peaked at some value of @xmath151 and the peak position rises only gradually as the level gap @xmath152 increases .
the structure factor @xmath97 for small @xmath153 is thus governed by virtual transitions to neighboring levels while its property at larger @xmath153 is determined by transitions across larger gaps .
it is readily expected from this property that the sum over an infinite number of negative - energy levels would make the weight @xmath5 cutoff ( @xmath154 ) dependent for small @xmath155 , actually to @xmath156 . in particular , the vacuum weight @xmath157 vanishes at @xmath158 and grows rapidly with @xmath159 for fixed @xmath154 , whereas it diverges as @xmath160 for @xmath161 .
see fig . 1 ( b ) .
indeed , evaluating the @xmath162 terms in @xmath163 shows that the divergence is logarithmic in @xmath154 , @xmath164 with @xmath165 and for @xmath22 .
this ultraviolet divergence is physical .
the infinite depth of the dirac sea , of course , is an artifact of the continuum model ( [ hzero ] ) and the cutoff scale @xmath166 is set by the energy scale above which the model loses its validity .
let us recall here that for conventional qh systems the charge operator trivially annihilates the vacuum , @xmath167 , and the vacuum spectral weight vanishes .
accordingly , the nonzero vacuum weight @xmath168 itself is a `` relativistic '' signature of graphene , @xmath169 which is a consequence of particle - hole pair creation or the presence of the dirac sea .
it is perfectly legitimate to consider the vacuum weight with such a physical cutoff @xmath154 ( apart from its precise value ) .
one can equally well extract cutoff - insensitive information out of it .
one possible way is to consider a variation of @xmath168 for @xmath170 and @xmath171 .
experimentally this means measuring the vacuum weight for @xmath170 and @xmath171 separately .
for @xmath171 the spectral weight for the graphene vacuum @xmath172 with @xmath22 is written as @xmath173 which represents a collection of virtual transitions @xcite from a negative - energy electron state with momentum @xmath174 to a positive - energy state with @xmath175 via the charge density @xmath176 ; here @xmath177 denotes the angle between @xmath174 and @xmath175 ; @xmath120 and @xmath178 denote the spin and valley degeneracy .
this is again ultraviolet divergent .
for regularization we cut off the @xmath174 integral at @xmath179 and choose the `` fermi momentum '' @xmath180 so that the dirac sea accommodates the same number of electrons as in the @xmath181 case , @xmath182 , i.e. , @xmath183 .
a direct calculation yields @xmath184 the cutoff @xmath185 dependence thus correctly disappears from the difference @xmath186 , with the result @xmath187 as seen from fig . 1 ( b ) ,
the vacuum weight @xmath188 differs only slightly for @xmath181 and @xmath171 . actually for all @xmath0
the spectral weight @xmath5 contains the vacuum fluctuations and the static structure factor @xmath97 is necessarily cutoff - dependent .
a possible cutoff - independent measure in experiment is to compare the spectral weights for @xmath189 and @xmath143 ( with @xmath181 ) .
correspondingly let us define by @xmath190 the structure factor with the vacuum contribution subtracted .
for the @xmath191 states ( with @xmath192 ) , @xmath193 in terms of the cutoff - independent deviations @xmath194 and @xmath195 .
for @xmath196 they are rewritten as @xmath197 analogously for @xmath198 . in reaching the second line ,
we have used the formulas ( valid for @xmath22 and @xmath160 ) : @xmath199\ , e^{z } -{1\over{4}}\ , { z^{|n| } \over{|n|!}},\ \ \ \ \ \label{sumgkn}\end{aligned}\ ] ] where @xmath145 and @xmath200 ; \end{aligned}\ ] ] @xmath201 and @xmath202 as @xmath203 ; the decrease is slower for larger @xmath204 , as depicted in fig . 2 ( a ) .
the subtracted structure factor is now written as @xmath205\ ] ] for @xmath191 .
in particular , @xmath206 and @xmath207.\ ] ] in fig . 2 ( b ) and ( c ) we plot @xmath208 for @xmath209 and for higher filling @xmath210 .
there are some notable features : ( 1 ) @xmath211 for @xmath212 .
this is a consequence of the uniformity of charge at long wavelengths .
( 2 ) @xmath211 for @xmath213 .
this shows that the large @xmath153 portion of @xmath214 is common to all @xmath0 and is governed by the vacuum weight @xmath168 . as a result ,
@xmath97 itself would rise with @xmath153 , in sharp contrast to the standard behavior @xmath215 as @xmath216 for `` nonrelativistic '' hall electrons in eq .
( [ spnr ] ) .
( 3 ) @xmath208 tends to saturate around 0.25 over a certain broad range of @xmath153 for higher filling factors @xmath0 .
the subtracted structure factor @xmath208 , unlike @xmath97 , refers to quantum fluctuations of a finite number of filled levels above the vacuum and is governed by virtual transitions , newly allowed or suppressed by the presence of such levels , as seen from eq .
( [ dij ] ) .
saturation is a consequence of a competition between the transitions among positive - energy states and the suppressed vacuum polarization effect . in this connection ,
let us try to retain only the positive - energy states and transitions among them . in fig .
3 ( a ) we plot a projected structure factor @xmath217 calculated from the first term on the right - hand side of eq .
( [ dij ] ) .
note that @xmath218 i.e. , even the positive - energy sector alone does not recover the @xmath215 behavior of conventional qh systems .
while @xmath217 itself is not directly observable in experiment , this feature ( [ splusp ] ) would indicate indirectly that @xmath208 is significantly smaller than 1/2 .
some remarks on the influence of a mass gap are in order here .
the relativistic treatment makes sense when the mass gap is tiny @xmath219 . in the `` nonrelativistic '' limit @xmath220 where the mass gap is large compared with the depth of the dirac sea , in contrast , the landau - level sum is limited to a finite interval @xmath221 and ceases to yield a divergence .
the virtual transition rates across the mass gap scale like @xmath222 ( while @xmath223 for @xmath224 ) , and the vacuum spectral weight @xmath225 tends to zero like @xmath226 as @xmath227 .
figure 3 ( b ) shows the manner how @xmath228 develops as the mass gap gets large for @xmath229 .
the @xmath230 state is now distinguishable in @xmath97 from the vacuum . in the @xmath227 limit
, @xmath231 recovers the standard behavior @xmath215 as @xmath216 for all integer @xmath0 . to explore the origin of the unconventional behavior ( [ splusp ] )
let us note the formula @xmath232 which follows from eq .
( [ sumgkn ] ) .
actually this formula is valid for @xmath21 as well , since it is independently derived from the evaluation of the static weight @xmath233 for a hypothetical state @xmath75 with only one filled landau level @xmath234 and all other levels empty .
it is thus a consequence of completeness of physical states . in the `` nonrelativistic '' limit
the dirac sea effectively disappears . then eq .
( [ completedirac ] ) is reduced to eq .
( [ completenr ] ) , which in turn leads to the behavior @xmath215 for @xmath216 .
for @xmath22 the large-@xmath153 behavior of eq .
( [ completedirac ] ) is governed by the sums over @xmath235 so that @xmath236 as seen also from eq . ( [ sumgkn ] ) .
this precisely accounts for the behavior of @xmath217 in eq .
( [ splusp ] ) .
it is now clear that both the masslessness ( or a tiny mass gap ) of the quasiparticles and the presence of the dirac sea are crucial for the unusual density correlations @xmath237 at short distances . in other words ,
correlations fail to vanish at short distances because of particle - hole pair creation ; the inability of localizing massless particles is a purely `` relativistic '' effect , underlying also the klein paradox .
@xcite equation ( [ completedirac ] ) may suggest that the vacuum spectral weight @xmath168 would have a contribution of @xmath238 from the dirac sea , it actually rises like @xmath239 , as we have seen in eq .
( [ vacuumweight ] ) .
had we defined the `` vacuum '' static structure factor @xmath97 with normalization by the total number of electrons in the dirac sea , it would vanish like @xmath240 .
this shows again that , for massless particles and in the presence of the dirac sea , the standard ( classical ) picture of particle correlations does not necessarily make sense .
finally we wish to comment on some peculiarities of the zero - energy @xmath50 levels , especially on how the coulomb interaction @xmath28 affects their response . in the random - phase approximation ( rpa )
the effect of @xmath31 of eq .
( [ hcoul ] ) is readily included @xcite into the polarization function ( @xmath241 ) , @xmath242 once one knows the polarization function @xmath243 for noninteracting hall electrons .
suppose now that we start with the vacuum state and fill up the @xmath244 level to reach the @xmath245 state .
then @xmath243 will change by an amount proportional to @xmath246 which is easily seen to vanish for zero mass - gap @xmath22 .
this shows that the @xmath143 and @xmath141 states remain indistinguishable in density response even at the rpa level , indicating the robustness of the zero modes against perturbations .
in particular , the static structure factor @xmath247 and the dielectric function @xmath248 remain the same for @xmath143 and @xmath141 .
in this paper we have studied the static structure factor for graphene in a strong magnetic field , with emphasis on revealing possible quantum signatures that distinguish graphene from conventional qh systems . in particular , for graphene the vacuum state is a dielectric medium with full of virtual particle - hole pairs . in other words , @xmath249 for graphene while @xmath250 for standard qh systems .
correspondingly the graphene vacuum state has a nonzero density spectral weight @xmath168 , which actually diverges with the ( cutoff ) depth of the dirac sea and generally rises with @xmath153 significantly .
experimentally the nonzero vacuum weight @xmath168 itself as well as its rise with @xmath153 , measured via inelastic light scattering , would be a clear signal of the quantum nature of the graphene vacuum state . for graphene the static structure factor
@xmath97 grows with @xmath153 for all @xmath0 , in sharp contrast to the behavior @xmath215 as @xmath251 of conventional qh systems ; this is because of pair creation at short distances .
actually , even the transitions among positive - energy states fail to recover the standard behavior and lead to @xmath237 for @xmath216 , as noted in eq .
( [ splusp ] ) .
this feature of @xmath217 itself , unfortunately , is not directly observable .
nevertheless , for large filling factor @xmath0 , it competes with the vacuum polarization effect and would make the observable subtracted structure factor @xmath208 ( which is sensitive to quantum fluctuations of cutoff - independent longer wavelengths ) saturate around 1/4 over a certain broad range of @xmath153 ; this could be a possible signature of the underlying dynamics .
such unusual features of particle correlations at short distances are a `` relativistic '' effect coming from massless dirac particles in graphene . of special interest , in addition , are some peculiar features of the lowest ( @xmath50 ) landau levels of zero energy .
remarkably they hardly affect the density response .
the virtual transitions , both _ allowed _ and _ suppressed _ anew by the presence ( or absence ) of the @xmath50 levels , combine to leave the density response unchanged , and this compensation persists even when the coulomb interaction is taken into account in the rpa . the graphene vacuum ( @xmath143 )
state and the @xmath141 states would thus be indistinguishable in spectral weight @xmath252 as well as in electric susceptibility , although they are distinct in hall conductance .
experimentally one would expect a definite nonzero signal for @xmath5 which largely remains the same over the range @xmath253 , except for some values of @xmath0 where nontrivial dynamics such as the fractional quantum hall effect may come into play .
the author wishes to thank t. morinari for useful discussion .
this work was supported in part by a grant - in - aid for scientific research from the ministry of education science sports and culture of japan ( grant no .
17540253 ) . | a close study is made of the static structure factor for graphene in a magnetic field at integer filling factors @xmath0 , with focus on revealing possible signatures of @xmath1relativistic `` quantum field theory in the low - energy physics of graphene .
it is pointed out , in particular , that for graphene even the vacuum state has a nonzero density spectral weight , which , together with the structure factor for all @xmath0 , grows significantly with increasing wave vector ; such unusual features of density correlations are a ' ' relativistic " effect deriving from massless dirac quasiparticles in graphene .
remarkably it turns out that the zero - energy landau levels of electrons or holes , characteristic to graphene , remain indistinguishable in density response from the vacuum state , although they are distinct in hall conductance . |
with over 30 o - type stars amongst its denizens ( massey , parker & garmany 1989 ; evans et al .
2006 ) , the ngc346 cluster is the site of most intense star formation in the small magellanic cloud ( smc ) as well as one of the most active in the local group .
the massive young stars in ngc346 are responsible for the ionisation of the surrounding n66 nebula , the largest hii region in the smc ( henize 1956 ) .
the location of ngc346 and n66 in the smc , their geometry and the limited amount of foreground extinction have made these regions an ideal place to study the effects of massive objects on the surrounding medium , including whether they can effectively trigger the formation of new generations of stars , as some theories of sequential star formation suggest ( e.g. elmegreen & lada 1977 ) . over the past 20 years , many authors have attempted to give an answer to these questions .
massey , parker & garmany ( 1989 ) conducted a photometric and spectroscopic study of ngc346 , revealing not only 33 o - type stars ( @xmath2 of which earlier than o@xmath3 ) , but also several lower - mass stars ( @xmath4m@xmath5 ) forming a distinct subgroup @xmath6 to the sw of the centre and with an estimated age of order 15myr .
the age difference with respect to the o - type stars led these authors to suggest that sequential star formation might have occurred in the region . using near- and mid - infrared as well as co sub - millimetre observations , contursi et al .
( 2000 ) and rubio et al .
( 2000 ) were able to identify several embedded sources in the bar making up the body of ngc346 , corresponding to strong emission peaks whose presence may reveal recent and/or ongoing star formation . however , while the peak corresponding with the central ngc346 cluster contains unreddened stars , all other peaks are affected by higher reddening , suggesting that the interstellar material has not been completely ejected and that they could be in a younger stage of evolution . since the more reddened peaks appear to be located farther away from the central cluster ,
these authors have suggested that star formation might have taken place in a sequential way along the bar .
more recent observations with the hubble and spitzer space telescopes have resolved the mid - infrared emission peaks , revealing that they are compact clusters made up of a multitude of pre - main sequence ( pms ) stars ( nota et al .
2006 ; sabbi et al .
2007 ; hennekemper et al . 2008 ) and young stellar objects ( yso ; bolatto et al . 2007 ; simon et al . 2007 ) .
hennekemper et al .
( 2008 ) performed a detailed analysis of the locations of these pms stars in the colour magnitude diagram ( cmd ) , which they compare with the pms isochrones of siess et al .
( 2000 ) taking into account the effects of differential reddening , binarity and variability on the age determination .
they conclude that , depending on the amount of reddening present in the field , the observed broadening of the positions of these pms objects in the cmd can be compatible with both a single star formation episode some @xmath7myr ago or with two episodes about 5 and 10myr ago . nevertheless , even in this latter case , the lack of a correlation between the estimated ages and the positions of the objects in the field led hennekemper et al .
( 2008 ) to conclude that there is no obvious signature of sequential star formation in this region . in a subsequent study by the same team , gouliermis et al .
( 2008 ) suggest that signs of sequential star formation might actually be present .
they propose a scenario in which the birth of the three young star clusters on the arc - like structure was triggered by the winds of the massive progenitor of snrb0057724 located at the centre of the arc , @xmath8pc away .
gouliermis et al .
( 2008 ) also suggest that , in a similar manner , the powerful winds of the ob stars at the centre of ngc346 are shaping a dusty arc feature to the south and southwest of the association . on the other hand ,
as we will show in section3 , the dusty arc feature appears to be @xmath8myr old and little or not at all affected by the presence of the ob stars in the centre ( smith 2008 ) , suggesting that there is no causal connection between the arc and the massive stars at the centre of ngc346 .
the weak side of all these studies is that they are mostly qualitative .
they are primarily based on the analysis of the spatial distribution of the objects and on the morphology and geometry of the features present in the field .
however , they do not take into account the actual ages of the low - mass stars that are needed to establish whether there are dependencies and correlations amongst the stellar generations that have formed in the recent past in these regions .
an improvement in this sense is offered by the recent work of cignoni et al .
( 2011 ) . using a classical synthetic cmd procedure
, they concluded that ngc346 has experienced different regimes of star formation , including a dominant and focused `` high density mode '' , which according to these authors led to the formation of rich and massive sub - clusters hosting both pms and massive ms stars , and a subsequent diffuse `` low density mode '' , characterised by the presence of sub - clusters hosting pms stars only .
these different modes of star formation can have an impact on the shape of the mass function ( mf ) , as we discuss further in section3 and 4 .
cignoni et al .
( 2011 ) suggest that the richest sub - clusters formed @xmath9myr ago , with an apparent remarkable synchronisation , while star formation in the sub - clusters mainly composed of pms stars appears to have started @xmath10myr ago , following a multi - seeded spatial pattern . even in works of this type ,
however , the available age range for low - mass stars is limited to the youngest pms objects ( @xmath11myr ) .
this is because ages are based on the comparison of the observations with theoretical evolutionary tracks in the hertzsprung
russell ( h r ) diagram , and at older ages the isochrones become too close to the very populous main sequence ( ms ) of field stars to provide reliable results .
actually , the presence of distinctive emission features in the spectra of pms stars with ages up to @xmath12myr , due to the accretion process to which they undergo , allows us to efficiently and securely detect and identify all objects of this type in a stellar field , regardless of their age and of their position in the h r diagram .
building on the work of romaniello ( 1998 ) and of panagia et al .
( 2000 ) , de marchi , panagia & romaniello ( 2010 , hereafter paperi ) showed that through a suitable combination of broad- and narrow - band photometry it is also possible to derive the mass accretion rate of these objects , with an accuracy comparable to that allowed by spectroscopy . in a companion paper ( de marchi et al .
2011 ; hereafter paperii ) , we applied the method developed in paperi to the high - quality hst photometry of ngc346 ( sabbi et al .
2007 ) and were able to identify two distinct generations of bona - fide pms stars ( about 700 objects ) with a clearly bimodal age distribution in the range from @xmath13myr to @xmath12myr . in this work
we use the accurate physical parameters that we have measured in paperii and correlate them with the spatial distribution of these objects .
the availability of accurate ages for such a large number of stars across the field is the key element that was missing in previous studies of ngc346 and allows us for the first time to study how star formation has proceeded in this area over the past @xmath12myr .
the paper is organised as follows : in section2 we briefly summarise the results of paperii and presents the relevant observational material .
section3 compares the spatial distribution of the two generations of pms stars to one another and to that of young massive stars present in the field . in section4
we look at how the shape of the stellar mass function changes across the field . in section5
we discuss the possible consequences of different processes operating for high mass and low mass star formation on the study of galaxies and the determination of their star formation rates .
a summary of the most important conclusions of the paper is offered in section6 .
in paperii , we have studied the properties of the stellar populations in a field @xmath14 around the centre of ngc346 , making use of observations collected with the advanced camera for surveys on board the hubble space telescope ( details on the observations and on the photometric analysis of the data can be found in nota et al .
2006 and sabbi et al . 2007 ) .
we refer the reader to paperii for a detailed description of the analysis of the stellar populations in this field and of the determination of their properties .
however , for convenience , we offer hereafter a brief summary of the main results that are most relevant to this paper .
thanks to a novel , self - consistent method developed in paperi , it is possible to reliably identify pms stars undergoing active mass accretion , regardless of their age .
the method , fully described in paperi and ii , does not require spectroscopy and combines broad - band @xmath15 and @xmath16 photometry with narrow - band @xmath17 imaging to detect all stars with excess h@xmath18 emission while simultaneously providing an accurate measure of their accretion luminosities @xmath19 and mass accretion rates @xmath20 .
the application of this method to the ngc346 observations allowed us to reveal 791 pms candidates , namely objects with h@xmath18 excess above the @xmath21 level with respect to the reference provided by normal cluster stars observed in the same bands .
the average h@xmath18 luminosity of these pms candidates is @xmath22ergs@xmath23 or @xmath24l@xmath25 . in order to avoid possible contamination due to objects with significant chromospheric activity
, we retained as bona - fide pms stars only those with a large equivalent width of the h@xmath18 emission line , @xmath26 for stars with @xmath27k or @xmath28 for hotter stars
( note that , as customary , a negative equivalent width is used for emission lines ) .
a total of 694 objects satisfy these conditions .
a colour magnitude diagram showing the positions of these objects in the observational plane ( @xmath15 vs @xmath29 ) is provided in paperii . in figure[fig1 ] we show the locations of these objects in the h
r diagram ( thick dots , in red in the online version ) , compared with the pms evolutionary models of the pisa group ( deglinnocenti et al .
2008 ; tognelli , prada moroni & deglinnocenti 2011 ) for metallicity @xmath30 . as noted in paperii ,
although this metallicity is at the lower end of the currently accepted values for the smc , ranging from @xmath31 to @xmath32z@xmath25 ( see russell & dopita 1992 ; rolleston et al .
1999 ; lee et al . 2005 ; perez montero & diaz 2005 ) , it appears better suited to describe the properties of young pms stars in this field . as figure[fig1 ] immediately shows , there are two separate groups of objects , occupying two distinct regions in the diagram , one well above ( or redder than ) the ms and the other at the ms itself .
this distinction is equally clear in the cmd shown in paperii .
comparison with the evolutionary tracks ( solid lines , corresponding to the masses as indicated ) and with the isochrones ( dashed lines , for ages increasing from right to left from @xmath33myr to 32myr doubling at each step ) allows us to determine accurate masses and relative ages for 680 of these objects .
note that previous determinations of these parameters for candidate pms stars in ngc346 that made use of evolutionary models for @xmath34 are necessarily less accurate .
for the interpolation , we followed the procedure developed by romaniello ( 1998 ) , which does not make assumptions on the properties of the population , such as the functional form of the initial mass function ( imf ) . on the basis of the measurement errors ,
this procedure provides the probability distribution for each individual star to have a given value of the mass and age ( the method is conceptually identical to the one presented recently by da rio et al . 2010a ) .
the masses that we obtain range from @xmath35m@xmath5to @xmath36m@xmath5 , with an average value of @xmath0m@xmath5 , whereas the ages show a clear bimodal distribution , already implicit in figure[fig1 ] , where very few objects are seen around ages of @xmath37myr . taking 7myr as a threshold ,
the pms stars can be split into two almost equally populous groups .
one group comprises 350 objects younger than 7myr ( hereafter called `` younger pms stars '' ) , with a median age of @xmath0myr ( formally @xmath38myr ) and a distribution ranging from @xmath39myr to @xmath40myr , respectively the 17 and 83 percentiles .
the other group contains 330 objects older than 7myr ( hereafter called `` older pms stars '' ) , with a median age of @xmath8myr ( formally @xmath41myr ) and a distribution ranging from @xmath42myr to @xmath43myr ( 17 and 83 percentiles , respectively ) .
in fact , the latter value should be considered a lower limit to the age of these objects .
regardless of the statistical approach that one follows , when a star in the h
r diagram is closer to the zero - age ms than its photometric errors , only a lower limit to its age can be derived , since the long tail of the distribution function makes all older ages in principle equally likely .
as discussed in paperi , our age uncertainties are typically less than a factor of two ( hence the choice of the spacing between isochrones in figure[fig1 ] ) and this includes systematic differences arising because of the use of models that might not properly describe the stellar population under study ( e.g. because of the wrong metallicity ) and from differences between models of various authors . yet for a given set of models , our photometric uncertainties result in even smaller uncertainties on the relative ages , typically of order @xmath44 .
thus , the age difference _ between _ the two groups of pms stars is very significant , since the age gap between the two groups is much wider than the uncertainty on the relative ages .
in fact , the age spreads _ within
_ each group might actually be somewhat smaller than what we quote , since our interpolation procedure does not take unresolved binaries into account . da rio et al .
( 2010a ) have shown that when binaries are ignored the derived age spread appears wider , although the average age itself is not affected .
therefore , the two groups of objects may be even better separated in age than our numbers imply .
on the other hand , we believe that interpreting the observed broadening as an age effect is the most conservative assumption as regards our conclusions . we address the properties of these two populations in detail in section3 .
besides the age and mass and h@xmath18 luminosity already mentioned , another other important physical parameter that we derived in paperii for these objects is the mass accretion rate @xmath45 , which has a median value of @xmath46m@xmath5yr@xmath23 .
this value is about 50% higher than that measured in paperi for a population of 133 pms stars in the field of sn1987a , owing to the much younger median age of pms objects in ngc346 .
in fact , the large size of our pms sample and its spread in mass have allowed us to study the evolution of the mass accretion rate as a function of stellar parameters and to conclude that @xmath47 , where @xmath48 is the age of the star , @xmath49 its mass and @xmath50 a quantity that is higher at lower metallicity ( see paperii for details ) .
in this section we use the information on the physical parameters of pms stars of various masses and ages to study how star formation has proceeded in ngc346 over the past @xmath12myr , since this is the time span that we can effectively and accurately probe with our method . as mentioned above , the distribution of our bona - fide pms stars in the h
r diagram reveals a shortage of objects with ages around @xmath51myr , suggesting a likely gap or lull in star formation at that time .
the presence of two so clearly distinct groups of stars with h@xmath18 excess can only be interpreted as the result of distinct star formation episodes .
hillenbrand et al . ( 2008 ) and hillenbrand ( 2009 ) have argued that random luminosity spreads apparent in the h
r diagram of star forming regions and young clusters are often erroneously interpreted as true luminosity spreads and taken as indicative of true age spreads .
this is clearly not the case here , since no random spread in the luminosity of a single age population could produce such a distinctive bimodal distribution in the h
. on the other hand , some of the objects that we label older pms stars could actually be very young stars with a circumstellar disc seen at high inclination ( @xmath52 ) .
objects of this type would appear bluer than their photospheric colour due to light scattering on the circumstellar disc .
however , they would also be several magnitudes fainter than their photospheric brightness due to extinction caused by an almost edge - on disc . according to the models of robitaille et al .
( 2006 ) for the spectral energy distribution of young stars seen at various viewing angles , objects of this type can only account for a few percent of the total young population . furthermore , as we will show in section3.2 , the spatial distribution of the stars with h@xmath18 excess close to the ms is remarkably different from that of the younger pms stars , and this should not be the case if these were all objects of the same type simply viewed at different inclinations .
therefore the vast majority of stars with h@xmath18 excess near the ms must be intrinsically older . in a forthcoming paper ( de marchi , guarcello & panagia , in preparation ) we will address in detail the role played by circumstellar discs seen at high inclination , which can undoubtedly account for a small fraction ( @xmath53% ) of our sample . that work will discuss in detail the theoretical implications that such a geometry can have on the extinction and scattering of the light of the central object and
will address the specific case of ngc6611 in the eagle nebula , where like in ngc346 a population of older ( @xmath7myr ) pms stars is also present .
as regards the age distribution of pms stars in ngc346 , a histogram is shown in figure[fig2 ] , where pms ages are binned using a constant logarithmic step ( a factor of 2 ) that better reflects the relative age uncertainties stemming from the comparison of model isochrones with the actual data .
the solid line in figure[fig2 ] gives the number of stars inside each age bin as a function of time , whereas the dot - dashed line provides an apparent value of the star formation rate , in units of stars per myr , derived by dividing the number of objects in each bin by the width of the bin .
note that at the extremes of the distribution it becomes more difficult to assign an age to the stars , thus the first and last bin are drawn with a dotted line to indicate a larger uncertainty . in particular ,
as mentioned above , for stars that in the hr diagram are closer to the ms than their photometric error , the age that we provide is in practice a lower limit to the true age .
this is due to the fact that the distribution function is characterised by such an extended tail towards older ages that all ages older than the value that we provide are virtually equally likely .
also the dot - dashed line necessarily represents a lower limit to the star formation rate . in this case , the reason is not the age uncertainty but the fact that the we consider exclusively the number of detected pms stars in the range @xmath54m@xmath5that at the time of the observations had h@xmath18 excess emission at the @xmath21 level or above .
one limitation is caused by photometric incompleteness at low masses , which makes it more difficult to detect faint pms stars in crowded environments .
another effect is the uncertainty on the fraction of pms stars that at any given time show excess h@xmath18 emission . for younger pms stars ( @xmath55myr ) , whose position in the h
r diagram is well separated from that of field ms stars , this fraction can be estimated from the ratio of stars with and without h@xmath18 excess in the same region of the diagram .
the data show that at the time of the observations this ratio was @xmath56 ( see also paperii ) . excess emission at the @xmath21 level or above at the time of the observations ) in the range @xmath57m@xmath5are considered in this figure ,
so the dot - dashed line provides a lower limit to the true star formation rate .
the thin solid line shows the age distribution of all stars ( i.e. also those without h@xmath18 excess emission ) , but only up to ages of 8myr , since older objects can not be distinguished from field ms stars .
the thin solid histogram is shifted vertically by @xmath58dex and appears in excellent agreement with the thick solid histogram.,title="fig : " ] the thin solid line in figure[fig2 ] ( green in the online version ) shows the age distribution of all stars in the h
r diagram younger than 8myr , shifted vertically by @xmath58dex ( corresponding to a factor of @xmath59 ) , and is in excellent agreement with the histogram of bona - fide pms stars .
however , it is presently not known how this ratio would change at older ages , and it is also expected to depend on the mass of the stars , so at this stage it is only possible to set a lower limit to the true star formation rate . near the peak of the distribution , at @xmath35myr , this limit corresponds to @xmath60m@xmath5myr@xmath23 ( the median mass of those objects is @xmath61m@xmath5 ) , while at @xmath62myr it drops by an order of magnitude to @xmath8m@xmath5myr@xmath23 ( median mass @xmath0m@xmath5 ) . at face value
, the star formation strength for stars in the range @xmath63m@xmath5thus appears to be much higher in the present burst than in the one that was active @xmath64myr ago and might have ended @xmath65myr ago , while the total integrated output ( i.e. the total number of stars , as shown by the thick solid histogram ) in the two episodes is rather similar . however , there are important selection effects that one must consider . as mentioned above , one effect is the fraction of pms stars with h@xmath18 excess , which most likely varies throughout the pms phase , making it more difficult to compare the number of younger and older pms stars to one another .
another problem is that , while the previous burst has certainly ended , the current one might still continue for a long time , thus making it hard to predict how many stars will eventually be formed .
finally , the accuracy on relative ages being at best of order a factor of @xmath44 , it is not possible to say exactly how long the previous burst lasted nor how many short bursts took place in the time frame covered by one age bin .
this carries the implication that , if there was just one short burst , the star formation strength might have been comparable to or even higher than that of the current episode .
the results of our quantitative age analysis on the star formation efficiency in this field are still necessarily tentative , but it will be possible to lift at least some of the uncertainties plaguing this picture through a systematic comparison of different star forming clusters in similar states of evolution that we plan to conduct in the future . on the other hand , having established that there are at least two star formation episodes in ngc346 , separated by @xmath7myr or more , we still can study whether they are independent of one another or appear to be causally connected .
we compare in figure[fig3 ] the spatial density distributions of the younger and older pms populations by means of contour lines with logarithmic scaling , overlaid on a h@xmath18 image of ngc346 . as mentioned above , the two groups include respectively 350 and 330 objects .
the contour plots have been obtained after smoothing the distribution with a gaussian beam with size @xmath66 or @xmath67pc , as indicated by the circle in the upper right corner of the figure .
the lowest contour level corresponds to a local density of pms stars twice as high as the average pms stars density over the entire field .
the step between contour levels is constant and corresponds to a factor of @xmath68 .
we also show with small circles the positions of 55 young massive stars brighter than @xmath69l@xmath25 and with an implied mass @xmath70m@xmath5(e.g .
iben 1967 ) .
a striking feature in this figure is the difference in the spatial distribution of the three types of stars .
many older pms stars are distributed along the rim of the gas shell to the s and w of the cluster s centre ( hereafter named `` southern arc '' ) and , except for the centre itself , they appear to avoid regions where younger pms stars are located . as for massive stars ,
albeit more abundant near the centre of ngc346 , they also appear at various other locations in the field that are not occupied by pms objects .
it is interesting to compare these contour lines with the maps of mid - ir emission obtained with isocam on board the infrared space observatory by contursi et al .
( 2000 ) and rubio et al .
the strong emission peaks discovered by these authors ( see also introduction ) coincide with regions of recent star formation in our analysis as well , and in particular with the intensity peaks due to younger pms stars in figure[fig3 ] ( darker contour plots , shown in red in the online version ) . as
regards older pms stars , including those in the southern arc , they appear projected against a background of lower ir emission . from the analysis of observations with the spitzer space telescope , simon et al .
( 2007 ) discovered in the arc a few young stellar objects of relatively high mass ( @xmath71m@xmath5 ) . as we will show later ( see figure[fig4 ] ) , there are indeed also some younger pms stars along this gas rim , but the arc appears to be mostly dominated by older pms objects .
the careful reader could be worried that several old pms stars seem to lie along the southern arc , since nebular emission might in principle contaminate their photometry and give us an inaccurate measurement of their h@xmath18 excess emission . however , as already mentioned in paperii , we have carefully inspected the images and removed from the list of bona - fide pms stars all objects whose h@xmath18 photometry might be contaminated by gas filaments . while it is possible that some of the h@xmath18 emission that we detect is due to diffuse nebular emission in the hii region not powered by the accretion process ( see paperi ) , if the emission is extended and uniform over an area comparable to that of the point spread function , its contribution cancels out with the rest of the background when we perform the photometry ( see sabbi et al .
2007 and paperii for details on the photometry ) .
obviously , the subtraction would not work if the emission were not uniform , as for example in the case of a filament that projects over the star but that does not cover completely the background annulus .
for this reason , after applying an unsharp - masking algorithm to highlight and sharpen the details of the h@xmath18 frames , we have carefully inspected all sources with excess h@xmath18 emission and have marked as suspicious and excluded from our bona - fide sample all those with filaments contamination within @xmath72 of the star , for a total of 62 objects .
although some of them might have intrinsic h@xmath18 excess emission , we prefer to adopt a conservative approach and remove all dubious cases .
a detailed example of how well this powerful technique works can be found in beccari et al .
therefore , we are confident that the tight distribution of older pms stars along the rim of the gas shell is not an artefact and suggests instead that these objects have very low velocities or at least a very small velocity spread
. observed values of the velocity dispersions of stars in young clusters and associations typically fall in the range @xmath73kms@xmath23 ( e.g. van altena et al .
1988 ; jones & walker 1988 ; mengel et al . 2009 ; bosch , terlevich & terlevich 2009 ; rochau et al .
the thickness of the projected distribution of the older pms in ngc346 is of order @xmath74 or @xmath9pc . with a median estimated age of @xmath1myr ,
this implies a small value of the projected velocity spread , namely @xmath75kms@xmath23 , corresponding to a three dimensional velocity dispersion of @xmath76kms@xmath23 along the gaseous rim .
this picture is consistent with the very low velocity dispersion ( @xmath77 km s@xmath23 ) of the ionised gas measured by smith ( 2008 ) in this field from high - resolution echelle spectroscopy .
if there is a higher velocity component , it must be linked to the systematic motion of the gas shell .
the match between the location of many old pms objects ( about @xmath78 @xmath2 of them ) and the rim of the gas shell also suggests that the shell itself reflects the distribution of the gas out of which these stars formed and that it has not ( yet ) been significantly affected by the stellar winds and by the ionising radiation of the much younger massive stars at the centre of the field .
furthermore , the fact that the distribution of these massive objects and of the younger pms stars does not appear to trace in any way the geometry of the gas shell indicates quite convincingly that we are seeing two rather different and unrelated generations of stars . from the apparent shape of the rim of the gas shell , gouliermis et al .
( 2008 ) recently argued that there is a relationship between the central ngc346 cluster and the southern arc .
they suggested that the latter outlines the ionisation front of the cloud that is caused by the powerful stellar winds of the young massive stars at its centre . in their scenario
, the photoionisation process of the central ob stars would provide the primary source of mechanical energy that triggers star formation in this region . our analysis does not support this interpretation : not only is the rim of the gas shell unaffected by the central ob stars , but it is also much older ( and it may be much farther away from the ob stars than what the projected distance might seem to suggest ) .
this discrepancy outlines the risks of drawing conclusions on triggered star formation based primarily on the morphology of structures projected on the sky . as watson , hanspal & mengistu ( 2010 )
have recently shown , only 20% of the sample of hii regions that they studied appear to have a significant number of ysos associated with their photodisociation fronts , implying that triggered star formation mechanisms acting on the boundary of the expanding hii region are not common .
a more quantitative characterisation of the relative distribution of younger and older pms objects and massive stars is offered by the maps shown in figure[fig4 ] .
the ( 0,0 ) position in that figure corresponds to ra@xmath79 , dec@xmath80 ( j2000 ) , with north up and east to the left . in panel
a ) we show all young stars using different symbols ( blue pentagrams , red dots and yellow dots respectively for massive young stars , younger pms stars and older pms objects ) , whereas panel b ) gives the number of stars of each type falling within cells of @xmath81 or @xmath82pc on a side , which is the typical size of a star cluster in the magellanic clouds ( @xmath83pc ; hodge 1988 ) .
the difference in the distribution of younger and older pms stars already seen in figure[fig3 ] continues to be present in figure[fig4 ] , revealing that older pms objects are less concentrated and more widely distributed than younger stars .
the other notable feature in figure[fig4 ] ( and already visible in figure[fig3 ] ) is the mismatch between the positions of young massive stars and those of young pms objects of similar age ( both types of objects are younger than 7myr and most are not older than 3myr ) .
the majority of young massive stars are clustered near the adopted centre of ngc346 , where a high concentration of young pms stars is also seen . however
, there is also a number of massive objects that are not surrounded by an overdensity of pms stars .
similarly , many pms objects are clustered in populous groups with no massive stars in their vicinity , although the total mass of the groups can reach @xmath84m@xmath5even when only considering pms objects in the range @xmath85m@xmath5 , once photometric completeness is taken into account .
a similar situation is seen in the cygnus x north complex ( beerer et al . 2010 ) , where the positions of many early b - type stars ( the most massive objects in the region ) do not coincide with those of young star clusters of similar age .
these differences can be easily quantified by using the star numbers shown in the cells of figure[fig4]b .
the numbers provide the count of massive stars on top , of young pms stars in the middle and of older pms object at the bottom of each cell . since in this field there are a total of 55 massive stars ( @xmath70m@xmath5 ) and 350 young pms stars , on average one would normally expect in each cell about 6 times more young pms objects than massive stars of similar age , but as the figure shows this is not always the case . on this basis , assuming poisson statistics , we note at least two regions with many pms stars where the observed absence of massive stars has a probability of less than 5% to occur by chance .
they are centered around ( @xmath86 , @xmath87 ) and ( @xmath88 , @xmath89 ) .
this condition is true for the individual cells with darker shade ( darker pink in the online version ) , while for the light - shaded cells in their vicinities ( lighter pink in the online version ) the condition still applies if they are combined with neighbouring cells of the same or darker colour .
this simple statistical test proves that the observed paucity of massive young stars in these two rather wide areas is significant at a @xmath90 level , at least .
in fact , since our photometric completeness is worse in the most central regions ( see sabbi et al .
2007 ) , the paucity of massive young stars is even more pronounced there .
a very similar result has been recently found by cignoni et al .
( 2011 ) , who analysed the ratio of massive ms stars and low - mass pms objects within the individual sub - clusters detected by sabbi et al .
( 2007 ) in this region .
although the identification of pms objects in the work of cignoni et al .
( 2011 ) is purely based on their broad - band colours and magnitudes , and as such can in principle be affected by interlopers and field objects , these authors conclude that the pms stars in the regions corresponding to our pink cells in figure[fig4]b are over - represented with respect to massive stars , for a typical imf .
similarly , one can find in figure[fig4]b several cells ( shaded in darker green in the online version ) where the number of young pms objects falls below the 5% poisson probability expected from the number of massive stars .
some of these cells are located at the centre of ngc346 , where crowding and photometric incompleteness make it more difficult to detect fainter pms stars , so the statistical significance of our non detection of low - mass stars is lower .
however , most of the isolated massive stars are located in the outer regions of the cluster , where photometric completeness and crowding are not a concern .
it is possible that these objects are not cluster members , but the fact that they are not surrounded by pms stars of similar age remains puzzling and might imply that they have been displaced from their formation region due to dynamical interactions , such as in the case of star 30dor 016 ( evans et al .
2010 ) .
a possibility would be that these massive objects were the lower - mass companions of disrupted binary systems in which the primaries have already ended their evolution .
the radial velocity study of ngc346 by evans et al .
( 2006 ) suggests that at least @xmath91 of the massive stars are in binary systems .
the typical projected separation of the isolated massive stars from the centre of ngc346 is of order 2or @xmath92pc . given the young ages prevalent in ngc346 ( less than 3myr ) , none of the youngest generation stars could have produced runaways .
very few isolated massive stars may indeed be runaways from binary systems if they were part of a @xmath4myr old population , such as the one associated with the subcluster sc16 ( sabbi et al .
2007 ) and corresponding to the three massive objects in the upper left corner of figure[fig3 ] at ( @xmath93 , @xmath94 ) . in this case
, runaway stars could cross the entire field of , say , 2 or @xmath92pc moving at a velocity of about 3kms@xmath23 for about 10myr .
this however seems hardly to be the case for the objects marked by circles in that figure .
firstly , none of them has an age in excess of 7myr , according to our photometry and to the spectroscopy of evans et al .
( 2006 ) for the objects in common with their catalogue .
secondly , the fact that there are only three massive stars in the direct vicinity of the @xmath4myr cluster would require that most massive stars were in binary systems and that almost all secondaries have moved away from it : this explanation appears rather contrived and seems to require a very unlikely occurrence .
therefore , we have to accept that the lack of massive stars associated with a number of pms star clusters simply reflects a genuine property of star formation in those parts of the the ngc346 complex . in summary , from the analysis of figure[fig4 ]
we can conclude in a more quantitative way that there is little or no spatial correlation between the position of young massive objects and that of pms stars of similar age , except for the centre of ngc346 , and that the two populations of older and younger pms objects have rather different distributions ( except again possibly for the regions near the cluster s centre ) .
this implies that the stellar mf varies considerably across the field , as cignoni et al .
( 2011 ) have recently suggested .
therefore , if the mf is sampled over a limited region , its shape will not be statistically representative of the entire cluster . as we discuss in the next section , this has profound implications for the concept of imf .
before proceeding to study the mf of pms stars , we must consider the possible selection effects inherent in our measurements and how they could affect our analysis of the mf .
for this purpose we summarise most of the relevant information in figure[fig5 ] , to which we will refer throughout this section . shown in the figure are the spatial distributions and positions in the h
r diagram of all our bona - fide pms stars , as a function of their ages , masses and l(h@xmath18 ) luminosities as measured in paperii .
as before , we split objects into two age groups , younger or older than 7myr , while for the mass we use three bins , namely @xmath95m@xmath5 , @xmath96m@xmath5and @xmath97m@xmath5 .
we also use symbols of different colours ( most easily discernible in the online version of the paper ) to identify stars with different h@xmath18 luminosities , as indicated by the legends . in all panels
the small open circles refer to the positions of the same 55 young massive stars already shown in figures[fig3 ] and [ fig4]a . as for the h
r diagrams , the solid lines correspond to pms isochrones from the pisa group ( deglinnocenti et al .
2008 ; tognelli et al .
2011 ) for metallicity @xmath30 and ages as indicated in panel i ) with a constant logarithmic step ( @xmath39dex corresponding to a factor of two in age ) .
since the photometric uncertainty is typically smaller than the separation between neighbouring isochrones , with the interpolation procedure explained in section2 we are able to assign relative ages with an accuracy of better than a factor of two , typically of order @xmath44 . a first obvious selection effect , which will however not affect our determination of the mf of young pms stars , is revealed by the paucity of pms objects in the range @xmath98m@xmath5and age @xmath99myr .
this is due to the detection limit of our photometry at @xmath100l@xmath25 .
another interesting characteristic apparent from this figure , and already partly seen in figures[fig3 ] and [ fig4]b , is the compact distribution of older pms stars of all masses along the gaseous rim of the southern arc .
the fact that the majority of these objects are rather luminous ( @xmath101erg s@xmath23 ) confirms that they are not artefacts due to the gas rim itself , as already discussed in section3 .
to illustrate the extent of the mf variations present in this field , we start from panel a ) , where the two conspicuous groups of young low - mass ( @xmath98m@xmath5 ) pms objects already seen in figure[fig4 ] are visible around ( @xmath86 , @xmath87 ) and ( @xmath88 , @xmath102 ) , as indicated by the large circles with a radius of @xmath103 or @xmath104pc . interestingly , very few objects are found in the same regions in panels d ) and g ) , where equally young stars with masses of respectively @xmath105m@xmath5and @xmath106m@xmath5are shown .
in particular , in panel a ) there are 36 and 41 pms stars in these regions , whereas 4 and 8 are found in panel d ) and 4 and 9 in panel g ) , respectively .
based on the number of stars inside the circles in panel a ) and assuming a power - law mf of the type @xmath107 with @xmath108 , as typically observed in galactic star forming regions ( de marchi , paresce & portegies zwart 2010 ) , one would expect respectively 11 and 13 objects inside the circles in panel d ) and 12 and 14 in panel g ) .
these values are considerably larger than the numbers observed , particularly for the region at ( @xmath86 , @xmath87 ) . if we were to derive the slope of the mf in the mass range @xmath85m@xmath5from the ratio of the number of stars occupying the same region in the three panels , we would obtain a rather steep , mf with @xmath109 ( respectively @xmath110 in the first region and @xmath111 in the second ) .
note that the true mf slope is most likely even steeper , owing to the fact that our smaller mass bin is subject to some incompleteness , as mentioned above , even though the completeness of our photometry is well above 50% for all but a handful of pms stars ( see paperii ) .
although instructive , the comparison between panels a ) , d ) and g ) carried out in this way is subject to some bias .
we are using the relative number of stars in different mass bins to characterise the shape of the mf , but the objects that we consider are only those undergoing active mass accretion as witnessed by excess h@xmath18 emission .
this necessarily introduces some selection effects because more massive stars reach the ms more quickly than lower - mass objects ( e.g. palla & stahler 1993 ; see also paperii ) and it becomes increasingly more difficult to derive an accurate age for them . if these objects are associated with older ages , they could be systematically underrepresented in our sample , thereby causing a spurious steepening of the mf .
a possible way to avoid these effects would be to limit the mass range for the analysis , i.e. to consider only panels a ) and d ) since none of the objects in the mass range @xmath112m@xmath5is expected to have reached the ms at an age of @xmath113myr .
alternatively , one could limit the age range and only consider very young stars , e.g. those with @xmath114myr , which are all in the pms phase over the entire mass range considered here ( @xmath57m@xmath5 ) .
this second approach has the advantage of providing more specific information on the imf and we will therefore follow it .
interestingly , however , even when restricting our analysis to stars younger than 1myr we still find a very steep mf : there are in total 35 objects with this age in the two reference areas of panel a ) , but only two objects are found in each of the corresponding regions in panel d ) and g ) .
the implied mf slope in the range @xmath57m@xmath5remains very steep , namely @xmath115 .
one could argue that , if the areas that we are studying are too small , the steep mf slopes that we obtain might be an artefact of small numbers statistics .
however , with a projected radius of @xmath81 or @xmath104pc , the circles in panels a ) , d ) and g ) are twice as large as the typical size of star forming clusters in the magellanic clouds ( e.g. hodge 1988 ) . therefore ,
these regions sample a sufficiently large area to be representative of the local conditions of star formation , that in this case appear to be characterised by the the paucity of massive stars and correspondingly a rather steep mf slope . a similarly steep mf ( @xmath116 )
was found by massey ( 2002 ) in his study of the field of the magellanic clouds , i.e. in regions similarly devoid of massive stars .
the short - dashed line is the best power - law fit in the mass range @xmath117m@xmath5and corresponds to @xmath118 .
the long - dashed line is the fit over the entire mass range , with @xmath119m@xmath5 .
, title="fig : " ] a more `` typical '' value of the mf slope for stars younger than 1myr could be derived if we extended our analysis to the entire area covered by these observations , corresponding to @xmath120pc on a side . in their study of the ngc346 region , sabbi et al .
( 2008 ) determined the present day mf in ngc346 from the same observations that we use in this paper .
when considering all stars in the field in the range @xmath121m@xmath5 , they derived a mf slope @xmath122 , very close the the salpeter ( 1955 ) value .
when we consider all pms stars younger than 1myr in this field we find a good match with the results of sabbi et al .
( 2008 ) over the common mass range , namely @xmath123m@xmath5 .
this is shown in figure[fig6 ] , where the long - dashed line corresponds to @xmath124 and provides a good fit to the mass distribution of pms stars younger than 1myr .
however , we still find a somewhat steeper slope ( @xmath118 , short - dashed line ) in the mass range @xmath125m@xmath5 , where the statistics is more robust thanks to the larger number of objects . in reality
, the true mf slope must be slightly shallower than the @xmath18 values that we have obtained since we only consider as bona - fide pms stars those showing h@xmath18 excess emission at the @xmath21 level or above . as discussed in paperi and ii , the accretion process and with it
the h@xmath18 luminosity is subject to large variations ( a factor of 2 3 in a few days ; e.g. fernandez et al .
1995 ; smith et al . 1999 ; alencar et al .
2001 ) , with the implication that , statistically , not all pms stars in a given region will show at the same time h@xmath18 excess emission above our conservative acceptance threshold . if the fraction of pms stars with h@xmath18 excess emission were constant , considering only these objects would not affect the determination of the mf slope .
however , as we will show in a forthcoming paper ( de marchi et al . , in preparation ) , that fraction appears to become smaller when mass and age increase .
while age is not an issue in the present case , since we are only considering stars younger than 1myr , our objects span about a decade in mass ( @xmath85m@xmath5 ) and in this range variations as large as a factor of 3 are seen in the fraction of pms stars with h@xmath18 excess emission .
if this effect were taken into account , the slope of the mf over the entire field in the range @xmath125m@xmath5would drop slightly , to @xmath126 , a value in line with the typical mf slopes and corresponding uncertainties that are observed in nearby galactic star forming regions ( e.g. de marchi , paresce & portegies zwart 2010 ; bastian , covey & meyer 2010 ) .
nevertheless , even applying this correction to the mfs measured in the two regions discussed above ( see the large circles in figure[fig5 ] ) would still give considerably steeper indices ( @xmath127 ) than those commonly measured in star clusters .
therefore , the general conclusion that we can draw from this analysis is that there are considerable variations in the shape of the mf of young pms stars across the field of ngc346 , as witnessed by largely different values of the power - law index @xmath18 .
these variations remain even when we only consider the mf of stars younger than 1myr , which would normally be taken as representative of the imf .
since these variations are seen when comparing groups of objects comprising several low - mass stars distributed over the typical size of a stellar cluster ( @xmath128pc radius ) , they can not be ascribed to simple statistical fluctuations .
coupled with the remarkable anti - correlation between massive and low - mass stars of similar age ( see section3 and figure[fig4 ] ) , this indicates quite convincingly that the formation of high- and low - mass stars requires at least different initial conditions , and might also be governed by different mechanisms .
this is also the conclusion recently reached by cignoni et al .
( 2011 ) . from the observation
that the apparently youngest sub - clusters , i.e. those composed only by stars in the pms phase , show a deficiency of massive stars , these authors speculate that the imf may be a function of time , with the youngest sub - clusters not having had sufficient time yet to form more massive objects . as already pointed out by panagia et al .
( 2000 ) , while the concept of imf might be meaningful over large areas where it represents the average result of different star formation processes , its predictive power over smaller scales , characteristic of a specific stellar cluster or association , will have to be seriously reconsidered .
an interesting corollary of the apparent separation between massive star formation sites and regions where moderate / low mass stars form is that , in general , massive stars may turn out to not be `` perfect '' indicators of active star formation in galaxies . while it is true that regions where massive stars have formed are actively forming stars , it is not necessarily true that those regions identify the places where most of the stars are formed .
in fact , integrating over a salpeter imf ( @xmath129 ) defined over the range @xmath130m@xmath5it is easy to calculate that objects above 4m@xmath5represent only about 20% of the total mass that goes into stars .
the fraction is even lower in the field of the magellanic clouds , where massey ( 2002 ) found @xmath116 .
our analysis of the ngc346 observations , as well other fields in the lmc and smc ( e.g. panagia et al .
2000 ; romaniello , robberto & panagia 2004 ; de marchi , romaniello & panagia 2010 ; and in preparation ) indicates that low mass stars formation appear to run independently of the formation of massive stars .
this may be due to the fact that even in a big cloud that is subdivided into smaller sub - units , massive stars can only form inside the larger cells whereas low - mass stars may form readily in smaller cells . if the bulk of the mass in the interstellar medium is not in the form of massive clouds , it is possible that the formation of low mass stars be quite active in regions which are not marked by the presence of massive stars .
it is also possible that under suitable conditions this separate channel of low mass star formation be the dominant process to form stars . within this scenario ,
it is clear that measuring the star formation rates of external galaxies , especially in low surface density galaxies , on the basis of massive star diagnostics ( be it direct detection of ob stars , optical and radio emission from hii regions , or their secondary far - infrared radiation for highly opaque clouds ) , may indeed lead to gross underestimates and to misleading results on the nature and the evolution of galaxies . in order to clarify these issues
one should intensively and systematically study regions of local galaxies , selected in an unbiassed way , where low mass stars can be individually detected and characterized , so as to determine their physical parameters including mass , luminosity , age and evolutionary status .
this can be done not only in our galaxy and , even better , in the magellanic clouds in which all stars are approximately at the same distance from us , but also in many other galaxies of the local group . for this purpose , in addition to a number of regions with obvious signs of active star formation , such as the orion nebula cluster ( da rio et al .
2010b , and in preparation ) and ngc3603 in our galaxy ( beccari et al . 2010 ) , 30dor ( de marchi et al .
2011 ) and other selected fields in the lmc , and ngc346 and ngc602 in the smc , we are also analysing a sample of fields selected from the archival pure parallel project that were imaged with the hst - wfpc2 in four broad - band filters ( f300w , f450w , f606w , and f814w ) and a narrow h@xmath18 filter ( f656n ) with a total exposure time of at least 3 orbits , i.e. about 160 minutes ( spezzi et al . 2011 ) .
although these fields are still not completely `` random '' and `` unbiased '' , since they were imaged in parallel with long uv spectroscopic observations of ob stars taken with another hst instrument about 5 to 12 away , they still represent a relatively rich sample ( about 20 fields in each lmc and smc ) of regions with generally marginal massive star formation . for the future it would be helpful to obtain complementary h@xmath18 observations of hst archival fields with deep exposures in the @xmath15 and @xmath16 bands ,
at least , as well as to target suitably selected new fields in the milky way and in the magellanic clouds .
finally , considerable improvements in this field will become possible once pms evolutionary models are properly calibrated .
even though the models currently available are state of the art , they are not as refined and extensively tested as those for ms stars or post - ms evolution . in particular , absolute pms ages are difficult to determine due to the lack of independent calibrators .
improvements in these models would make pms stars even more powerful indicators of how star formation proceeds over time and space .
we have studied the properties of the stellar populations in the field of the ngc346 cluster in the small magellanic cloud , using the results of a novel self - consistent method that provides a reliable identification of pms objects actively undergoing mass accretion , regardless of their age .
we have used the age and other physical parameters measured for these pms stars to study how star formation has proceeded across time and space in ngc346 over the past @xmath131myr .
the main results of this work can be summarised as follows . 1 .
the 680 identified bona - fide pms stars show a bimodal age distribution , with two roughly equally numerous populations with median ages of respectively @xmath0myr and @xmath8myr , although the latter is most likely a lower limit to the true age .
the age separation of the two groups is consideraby wider than the uncertainties on the relative ages and the scarcity of objects with ages around @xmath65myr suggests a lull in star formation at that time . 2 .
although , taken at face value , the colours and magnitudes of the older pms stars are compatible with those of young pms objects whose light is absorbed and scattered by a high - inclination circumstellar disc , this hypothesis is only viable for a few percent ( @xmath132% ) of the stars .
this is confirmed by the remarkaby different spatial distributions of the two age groups .
we set a lower limit to the star formation rate of the ongoing burst of @xmath60m@xmath5myr@xmath23 , while at @xmath133myr the rate drops by an order of magnitude to @xmath8m@xmath5myr@xmath23 .
both values are lower limits since they are based on the number of pms stars in the range @xmath57m@xmath5that were undergoing active mass accretion at the time of the observations .
4 . at face value
, the strength of the current star formation episode appears to be much higher than that of the one ended @xmath65myr ago , while the total integrated output of the two episodes is rather similar .
however , since photometric uncertainty does not allow an age resolution of better than a factor of @xmath134 , we can not establish how many bursts took place between 8 and 30myr ago , nor their duration .
if there was just one short burst , the star formation strength might have been comparable to or even higher than that of the current episode .
5 . except for the regions near the centre of ngc346 , the stars belonging to the two generations have markedly different spatial distributions .
a good fraction ( @xmath78 @xmath2 ) of the older generation occupies an arc - like gas structure to the south and west of ngc346 that had been previously interpreted as the ionisation front caused by the ob stars at its centre .
although the morphology of the arc could have suggested a case of triggered star formation , this is clearly not a viable option since the central massive stars are at least 1020myr younger than the objects on the arc and can not have triggered their formation .
the compact distribution of older pms stars along the arc - like structure suggests that they have formed there from the gas that is still visible and have not ( yet ) been affected by the massive ob stars at the centre of ngc346 .
this picture is consistent with the very low velocity dispersion ( @xmath77 km s@xmath23 ) of the ionised gas measured in this field from high - resolution echelle spectroscopy .
7 . except for the most central regions of ngc346
, we find no correspondence between the positions of young pms stars and massive o - type stars of similar age , suggesting that the conditions ( and possibly also the mechanisms ) for their formation must be rather different .
furthermore , the mass distribution of similarly aged stars shows large variations across the region .
we conclude that , while on a global scale it makes sense to talk about an initial mass function , this concept is not meaningful for individual star forming clumps .
it is possible that a large number of low - mass stars are forming in regions where massive stars are not present and , therefore , they remain unnoticed . for certain low surface density galaxies this might be the predominant way of star formation , which would imply that their total mass based on the luminosity can be severely underestimated and that their evolution is not correctly understood .
we are grateful to an anonymous referee for extensive comments that have helped us to improve the presentation of this work .
np acknowledges partial support by hst - nasa grants go-11547.06a and go-11653.12a , and stsci - ddrf grant d0001.82435 . | we have studied the properties of the stellar populations in the field of the ngc346 cluster in the small magellanic cloud , using the results of a novel self - consistent method that provides a reliable identification of pre - main sequence ( pms ) objects actively undergoing mass accretion , regardless of their age .
the 680 identified bona - fide pms stars show a bimodal age distribution , with two roughly equally numerous populations peaked respectively at @xmath0myr , and @xmath1myr .
we use the age and other physical properties of these pms stars to study how star formation has proceeded across time and space in ngc346 .
we find no correlation between the locations of young and old pms stars , nor do we find a correspondence between the positions of young pms stars and those of massive ob stars of similar age .
furthermore , the mass distribution of stars with similar age shows large variations throughout the region .
we conclude that , while on a global scale it makes sense to talk about an initial mass function , this concept is not meaningful for individual star - forming regions .
an interesting implication of the separation between regions where massive stars and low - mass objects appear to form is that high - mass stars might not be `` perfect '' indicators of star formation and hence a large number of low - mass stars formed elsewhere might have so far remained unnoticed .
for certain low surface density galaxies this way of preferential low - mass star formation may be the predominant mechanism , with the consequence that their total mass as derived from the luminosity may be severely underestimated and that their evolution is not correctly understood . |
experiments with tagged photons , radiated from the initial state in electron proton and electron positron collisions , can become particularly attractive .
the reason is that these radiative processes will permit to extract information about the final states at continuously varying values of the collision energy . to investigate deep inelastic scattering the authors of ref .
@xcite suggested to use radiative events instead of running colliders at reduced beam energies .
the method takes advantage of a photon detector ( pd ) placed in the very forward direction , as seen from the incoming electron beam . the effective beam energy , for each radiative event ,
is determined by the energy of the hard photon is observed in pd .
in fact , radiative events were already used to measure the structure function @xmath3 down to @xmath4 gev@xmath5 @xcite .
the specific theoretical work concerns the evaluation of qed radiative corrections @xcite to the radiative born cross section . with an accurate determination of the cross sections and of the possible sources of background
we believe that the use of radiative events may become particularly useful to carry investigations at various present and future machines .
the important role of the initial state radiation in the process of electron positron annihilation was underlined in a series of papers by v.n .
baier and v.a .
khoze @xcite , where the radiative process was studied in detail in the born approximation . in these papers the mechanism of returning to a resonant region was discovered .
this mechanism consists in the preferable emission of photons from the initial particles , which provides a resonant kinematics of a subprocess
. a utilization of radiative events can become a common type of investigations at various machines . in this paper
we derive explicit formulae for the spectrum of tagged photons .
the calculations are performed having an accuracy of the per
mille order as an aim .
formulae can be used at electron
positron colliders to investigate , for instance , hadronic final states at intermediate energies .
a measurement of the total hadronic cross section at low energies is essential for an high precision test of the standard model particularly for a precise determination of the fine structure constant @xmath6 and of the muon anomalous magnetic moment @xmath7 .
the largest contribution to the errors for these quantities comes from the large indetermination still present on the measurement of the total hadronic cross section in electron
positron annihilation at center of mass energies of a few gevs .
we will consider here the radiatively corrected cross section for the electron
positron annihilation process @xmath8 where @xmath9 is a generic hadronic state .
the hard photon hitting the photon detector has a momentum @xmath10 and an energy fraction @xmath11 with respect to the beam energy . in the following
we assume that the photon detector is placed along the electron beam direction , and has an opening angle @xmath12 , such that @xmath13 , with @xmath14 the electron mass , and @xmath15 the beam energy . to evaluate the process with an accuracy of the per mille requires a careful investigation of the radiative corrections .
this paper is organized as follows . in section 2
we consider the cross section of the process ( [ 1 ] ) in the born approximation .
we give formulae suitable to study as differential distributions in hadronic channels , as well as the total ( in terms of quantity @xmath16 ) and inclusive ( in terms of hadron fragmentation functions ) hadronic cross sections . in sec .
3 we calculate separate contributions into radiatively corrected cross section of process ( [ 1 ] ) within the next to leading accuracy . in sec . 3.1
the contribution due to virtual and soft photon emission is investigated . in sec .
3.2 the case , when additional hard photon hits a photon detector is considered . in sec .
3.3 the contribution due to hard photon emission , which does not hit a photon detector is derived . in sec .
4 we sum up all the contributions and give the final result . in conclusions
we summarize the results and give some numerical illustrations .
in order to obtain the born approximation for the cross section of the process ( [ 1 ] ) , when the pd is placed in front of electron ( positron ) beam , we can use the quasireal electron method @xcite .
it gives @xmath17 where @xmath18 is the probability to radiate photon with energy fraction @xmath11 inside a narrow cone with the polar angle not exceeding @xmath19 around the incoming electron , and @xmath20 is the differential cross section for the radiationless process of electron positron annihilation into hadrons at the reduced electron beam energy .
the form of both , @xmath18 and @xmath21 is well known : @xmath22 we need further the general form of the lowest order cross section @xmath23 for the process @xmath24 hadrons boosted along the beam axis ( @xmath25 ) : @xmath26 where @xmath27 is the full 4momentum of final hadrons , @xmath28 is 4momentum of an individual hadron , @xmath29 is the full center of mass energy squared , and @xmath30 is the hadronic tensor .
the vacuum polarization operator @xmath31 of the virtual photon with momentum @xmath27 is a known function @xcite and will not be specified here .
the tensors @xmath30 and @xmath32 obey the current conservation conditions once saturated with the 4vector @xmath27 .
the differential cross section with respect to the tagged photon energy fraction @xmath33 can be obtained by performing the integration on the hadrons phase space .
it takes the form @xmath34 each hadronic state is described by its own hadronic tensor . the cross section in eqs .
( 3 ) and ( 4 ) is suitable for different uses and , as mentioned above , it can be used to check different theoretical predictions .
the sum of the contributions of all hadronic channels by means of the relation @xmath35 can be expressed in terms of the ratio of the total cross section for annihilation into hadrons and muons @xmath36 . for the @xmath37 final state we get @xmath38 and so , @xmath39 substituting this expression into the right hand side of eqs .
( 3,4 ) results in the replacement @xmath40 .
in experiments of semi inclusive type one fixes an hadron with 3-momentum @xmath41 energy @xmath42 and mass @xmath43 in every event and sum over all the rest . in this case instead of eq .
( [ 6 ] ) we will have ( similarly to the deep inelastic scattering ( dis ) case @xcite ) : @xmath44 where we have introduced two dimentionless functions @xmath45 and @xmath46 in a way similar to the dis case . by introducing the dimentionless variable @xmath47
, we can write the corresponding cross section for radiative events in @xmath48 annihilation in the same form as in the case of deep inelastic scattering with a tagged photon @xcite : @xmath49
for the radiative corrections ( rc ) to the cross section ( 5 ) we will restrict ourselves only to terms containing second and first powers of large logarithms @xmath50 , and omit terms which do nt contain them i.e. we will keep leading and next to leading logarithmic contributions .
we will consider in subsection 3.1 the contribution from one loop virtual photon as well as from the emission of soft real ones . in
3.2 we will discuss the double hard photon emission process . the interference of born and one loop contributions to the amplitude of the initial state radiation in annihilation of @xmath48 into hadrons can be obtained from the analogous quantity of hard photon emission in electron
proton scattering @xcite .
we do that by using the crossing transformation . for the contribution coming from the emission of real soft photons a straightforward calculation
gives : @xmath51 , \quad l_s=\ln\frac{s}{m^2 } = l_0 + l_{\theta } , \quad l_{\theta } = \ln\frac{4}{\theta^2}\ , .\end{aligned}\ ] ] where @xmath52 is the _ photon mass _
, @xmath53 is the energy carried by the soft photon .
the sum of the two contributions is free from infrared singularities .
it reads @xmath54\frac{h_{\rho\sigma}(q ) { { \mathrm d}}\gamma(q)}{(q^2)^2}\frac{\alpha}{4\pi^2}\frac{{{\mathrm d}}^3 k}{\omega},\ ] ] where @xmath55 where @xmath10 and @xmath56 are the 3-momentum and the energy of the hard photon respectively .
the tensors @xmath57 and @xmath58 have a rather involved form .
the first can be obtained from the corresponding expressions of ref .
the tensor @xmath58 coincides with the one of the born approximation . in the kinematical region
where the hard photon is emitted close to the initial electron direction of motion one has @xmath59 where tensor @xmath32 is given in eq .
( 4 ) , @xmath60 , and quantity @xmath61 reads @xmath62,\qquad { { \mathrm li}_2}(x)=-\int_0 ^ 1 { { \mathrm d}}t \frac{\ln(1-tx)}{t}.\end{aligned}\ ] ] further integration over the hard photon phase space can be performed within the logarithmic accuracy by using the integrals @xmath63=\biggl[\frac{1}{2}l_0,\frac{1}{2(1-z ) } , \frac{1}{4}l_0 ^ 2 + \frac{1}{2}l_0\ln(1-z)\biggr]{{\mathrm d}}z . \nonumber\ ] ] the final expression for the born cross section corrected for the emission of soft and virtual photons has the form @xmath64 , \nonumber\\ n&=&-\frac{1+z^2}{1-z}\biggl[(l_0+\ln z)\ln z - \frac{\pi^2}{3 } + 2{{\mathrm li}_2}(z)\biggr]l_0 - 2p_1(z , l_0)\ln\frac{\theta_0 ^ 2}{4}\nonumber \\ & + & \frac{1 + 2z - z^2}{2(1-z)}l_0+\frac{4z}{1-z}l_0\ln z.\end{aligned}\ ] ] if an additional hard photon emitted by the initial state electron hits the pd , we can not use the quasireal electron method and have to calculate the corresponding contribution starting from feynman diagrams .
we can use double hard photon spectra as given in ref .
@xcite for annihilation diagrams only and write the cross section under consideration as follows @xmath65 , \\ \nonumber \xi & = & x(1-x)^2(1-z - x ) , \quad \gamma = 1+(1-x)^2 , \quad \tau = z^2 + ( 1-x)^2.\end{aligned}\ ] ] here the variable @xmath66 under the integral sign is the energy fraction of one hard photon .
the quantity @xmath67 is the energy fraction of the second hard photon provided that their total energy fraction equals to @xmath11 .
we write the index @xmath68 in the left hand side of eq .
( [ 25 ] ) to emphasize that this contribution arises from the collinear kinematics , when the additional hard photon is emitted along the initial electron with 4momentum @xmath69 .
the integration in the right hand side of eq .
( [ 25 ] ) leads to the result @xmath70l_0 \nonumber \\ \label{26 } & + & 6(1-z ) + \frac{3+z^2}{1-z}\ln^2z - \frac{4(1+z)^2}{1-z}\ln\frac{1-z}{\delta } \biggr\},\end{aligned}\ ] ] where the quantity @xmath71 represents the so called @xmath72term of the second order electron structure function : @xmath73 if an additional hard photon , emitted from the initial state , does not hit the pd situated in the direction of motion of the initial electron we distinguish the case when it is emitted in the direction close , within a small cone with angle @xmath74 , to the direction of the initial positron .
in this case we obtain : @xmath75 where @xmath76 , @xmath77 , and @xmath78 is the minimal hadron mass squared .
we suppose that @xmath79 .
we have introduced the additional auxiliary parameter @xmath80 which , together with @xmath81 , separates collinear and semi collinear kinematics of the second hard photon .
contrary to @xmath81 , which is supposed to determine the pd acceptance , @xmath82 will disappear in the sum of the collinear and semi collinear contributions of the second photon .
this last kinematical region gives @xmath83,\qquad x_1=\frac{x}{2}(1-c),\quad x_2=\frac{x}{2}(1+c),\nonumber\\ q_2&=&zp_1+p_2-k_1 , \qquad c=\cos \widehat{{\mbox{\boldmath $ k$}}_1{\mbox{\boldmath $ p$}}_1}.\end{aligned}\ ] ] the phase volume of the second photon is parametrized as : @xmath84 explicitly extracting the angular singularities we represent this expression as @xmath85 , \\
\nonumber \sigma_{sc}&=&\frac{8\pi^2\alpha^2}{z^2}\int\limits_{-1}^{1}{{\mathrm d}}c \int\limits_{\delta}^{\hat x } \frac{{{\mathrm d}}x}{x } \biggl [ \frac{t(c , z , x)-t(1,z , x)}{1-c}+\frac{t(c , z , x)-t(-1,z , x)}{1+c}\biggr].\end{aligned}\ ] ]
the final result in the order @xmath86 for radiative corrections to radiative events can be written as follows : @xmath87\biggr\},\end{aligned}\ ] ] where the last term is defined by eq .
( 23 ) and @xmath88 in order to include the higher order leading corrections to the tagged photon differential cross section and show the agreement of our calculation with the well known drell yan representation for the total hadronic cross
section at electron
positron annihilation @xcite @xmath89 where the electron structure functions include both nonsinglet and singlet parts @xmath90 it is convenient to introduce the quantity @xmath91 note that the shifted cross section in eq .
( [ kf ] ) has just the same meaning as in eq .
( [ 3 ] ) : @xmath92 .
the structure functions @xcite , which enter into the right side of eq .
( [ sf ] ) , are @xmath93 @xmath94,\ ] ] where @xmath95 and the effective electromagnetic couplings in the right side of eq .
( [ lla ] ) are @xmath96 at fixed values of @xmath33 @xmath97 the quantity @xmath98 defines the leading logarithmic contributions into differential cross
section for the events with tagged particles .
that corresponds to only @xmath72terms in the expansion of the structure function @xmath99 before the integral sign in eq .
( [ lla ] ) .
if we consider photonic corrections ( as in previous sections ) it needs to restrict ourselves with the nonsinglet part of the electron structure functions and with the first order terms in the expansion of all effective couplings , namely : @xmath100 it is easy to see that in this case the leading contribution into differential cross section ( [ final ] ) can be obtained as an expansion of the quantity @xmath101 by the powers of @xmath102 , keeping the terms of the order @xmath103 in the production of @xmath104functions . if we want to include the contribution due to @xmath48pair ( real and virtual ) production it is required @xcite to use both nonsinglet and singlet structure functions and effective couplings defined by eq .
( [ 32 ] ) .
note that the insertion into consideration of higher order corrections rises additional questions about concrete experimental conditions concerning registration of events with @xmath48pairs .
the total hadronic cross section in @xmath48 annihilation can be obtained by integration of quantity @xmath98 over @xmath33 @xmath105 we can integrate the expression in the right side of eq .
( [ tots ] ) over the variable @xmath33 provided the quantity @xmath106 is fixed @xmath107 using this result and definition of @xmath108 we indicate the equivalence of the drell yan form of the total cross section as given by eq .
( [ kf ] ) and the representation of the cross section by eq .
( [ tots ] ) .
let us show now that @xmath104functions in expression for the quantity @xmath98 have effective couplings as given by eq .
( [ 32 ] ) . by definition
the nonsinglet electron structure function satisfies the equation @xcite @xmath109 where @xmath110 is the electromagnetic running coupling @xmath111 and @xmath112 is the minimal ( maximal ) virtuality of the particle , which radiate photons and @xmath48pairs .
the structure function @xmath99 describes the photon emission and pair production inside narrow cone along the electron beam direction . in this
kinematics @xmath113 , @xmath114 . the corresponding iterative solution of the eq . ( [ lap ] ) has the form ( [ 30 ] ) with @xmath115 .
the structure function @xmath116 describes the events , when emitted ( by the electron ) particles escape this narrow cone . in this case
@xmath117 , @xmath118 .
the corresponding solution of eq .
( [ lap ] ) gives the structure function with @xmath119 .
at last , the structure function @xmath120 is responsible for the radiation off the positron into the whole phase space . in this case
@xmath121 , @xmath122 . therefore we obtain @xmath104function with @xmath123 . the analogous consideration
can be performed for the singlet part of structure functions .
when writing the representation ( [ tots ] ) for the total cross section we , in fact , divide the phase space of the particles emitted by the electron on the regions inside and outside the narrow cone along electron beam direction . therefore we can use this representation to investigate the events with tagged particles in both this regions . as we saw before the differential cross
section for events with tagged particles inside the narrow cone is defined by the quantity @xmath101 . in order to obtain the corresponding differential cross
section for events with tagged particles outside this narrow cone we have to change the places of @xmath124 and @xmath125 in expression for @xmath126 .
this follows from the symmetry of representation ( [ tots ] ) relative such change .
in sum the formulae ( [ 35],[lla ] ) are the main results of our paper . in fig . 1 we show the cross section @xmath127 as a function of @xmath33 .
the beam energy is chosen to be @xmath128 gev .
the region of @xmath33 values is limited by the pion production threshold at the left , and by the threshold of photon detection ( we choose 50 mev ) at the right .
the peak in middle corresponds to the large contribution of the @xmath129-meson .
values of @xmath16 , used for numerical estimations , were taken from @xcite .
the values of corrections @xmath130 ( see eq .
( [ 35 ] ) ) in percent are shown in fig . 2 .
so , we calculated the cross section of @xmath48 annihilation with detection of a hard photon at small angles in respect to the electron beam .
the general structure of a measured cross section , from which one should extract the annihilation crossection @xmath23 , looks as follows : @xmath131 + { \cal o}(\alpha^3),\end{aligned}\ ] ] where @xmath50 denotes some large logarithm .
we calculated the terms @xmath132 , @xmath133 , @xmath134 , @xmath135 and some contributions to @xmath136 .
the generalized formula ( [ lla ] ) allows to involve the leading terms of the order @xmath137 . in this way
our formulae provide high theoretical precision .
similar formulae can be obtained for an experimental set
up by tagging a definite hadron . by using @xmath48machines such as beps , da@xmath138ne @xcite , vepp , cleo , slac b / factory and others with luminosities of order @xmath139 @xmath140 , one may be in principle able to scan , by measuring the initial state radiation spectrum , the whole energy region of hadron production with an effective luminosity of the order of @xmath141 @xmath140 .
we hope further study will follow on these issues both from the experiments and theory .
we are grateful to s. serednyakov and z. silagadze for a critical reading of the manuscript and for useful comments .
, e.k . and n.p.m .
wish to thank the infn and the dipartimento di fisica of the universit for the hospitality .
this work is supported by the intas grant 931867 ext .
one of us ( n.p.m . ) thanks the ukrainian dffd for the grant n24/379 and the ministry of science and technology of ukraine for support .
we wish also to thank b. shaikhatdenov for help .
brown , w.a .
worstell , phys . rev . *
d 54 * ( 1996 ) 3237 ; + s. eidelman _
_ , z. phys .
* c 67 * ( 1995 ) 585 ; + f. jegerlehner .
the hadronic contribution to @xmath142 .
transparencies of the workshop on " hadron production cross sections at da@xmath138ne , v.1 , p.81 .
karlsrue , germany , november 1 and 2 , 1996 .
editors : u.v .
hagel , w. kluge , j.kuhn , g.pancheri . | events with tagged photons in the process of electron positron annihilation into hadrons are considered .
the initial state radiation is suggested to scan the hadronic cross section with the energy .
qed radiative corrections are taken into account .
the results for the total and exclusive cross sections are given in an analytic form .
some numerical estimates are presented .
@xmath0 _ joint institute for nuclear research , dubna , 141980 , russia + @xmath1 _ institute of physics and technology , kharkov , 310108 , ukraine + @xmath2 _ dipartimento di fisica , universit di parma and infn , + gruppo collegato di parma , 43100 parma , italy _ _ _ + pacs 13.65.+i hadron production by electron positron annihilation , + 12.15.lk electroweak radiative corrections |
the connection between two important disciplines of physics , namely quantum information theory and condensed matter physics , has been the subject of great interest recently , generating much activity at the border of these fields , with numerous interesting questions addressed so far @xcite . in particular , the properties of entanglement in many - body systems , and the analysis of its behavior in critical systems deserve special attention . in this work
we deal with the entanglement of indistinguishable fermionic particles in the one dimensional extended hubbard model ( ehm ) .
we focus in the half - filling case .
the model is a generalisation of the hubbard model @xcite , which encompasses more general interactions between the fermionic particles , such as an inter - site interaction , thus describing more general phenomena and a richer phase diagram .
precisely , it is given by , @xmath0 where @xmath1 is the lattice size , @xmath2 and @xmath3 are creation and annihilation operators , respectively , of a fermion with spin @xmath4 at site @xmath5 , @xmath6 , @xmath7 , and we consider periodic boundary conditions ( pbc ) , @xmath8 .
the hopping ( tunnelling ) between neighbor sites is parametrized by @xmath9 , while the on - site and inter - site interactions are given by @xmath10 and @xmath11 , respectively . despite the apparent simplicity of the model , it exhibits a very rich phase diagram , which includes several distinct phases , namely : charge - density wave ( cdw ) , spin - density wave ( sdw ) , phase separation ( ps ) , singlet ( ss ) and triplet ( ts ) superconductors , and a controversial bond - order wave ( bow ) .
a more detailed description of the model and its phases will be given in the next section .
our numerical analysis is performed employing entanglement measures for indistinguishable particles introduced recently @xcite , in conjunction with the density - matrix renormalisation group approach ( drmg)@xcite , which has established itself as a leading method for the simulation of one dimensional strongly correlated quantum lattice systems .
dmrg is a numerical algorithm for the efficient truncation of the hilbert space of strongly correlated quantum systems based on a rather general decimation prescription .
the algorithm has achieved unprecedented precision in the description of static , dynamic and thermodynamic properties of one dimensional quantum systems , quickly becoming the method of choice for numerical studies .
the paper is organised as follows . in sec .
[ ext.hub.model ] we review the model and its phase diagram . in sec .
[ ent.ind.part ] we present the distinct definitions of entanglement in systems of indistinguishable particles , focusing on the notion of `` entanglement of particles '' . in sec .
[ ent.and.qpt ] we present our results .
we conclude in sec .
[ conclusion ] .
in this section we give a detailed description of the extended hubbard model @xcite , and its distinct phases in the half - filling case .
the reader familiar with the subject may skip this section .
many efforts have been devoted to the investigation of the ehm s phase diagram at half - filling , using both analytical and numerical methods @xcite . despite the apparent simplicity of the model , it exhibits a very rich phase diagram which includes several distinct phases : charge - density wave ( cdw ) , spin - density wave ( sdw ) , phase separation ( ps ) , singlet ( ss ) and triplet ( ts ) superconductors , and a controversial bond - order wave ( bow ) .
see fig.[phase.diagram.ehm ] for a schematic drawing of the phase diagram at half - filling . in the strong coupling limit ( @xmath12 ) , one can qualitatively characterize its phases as given by a charge - density wave , spin density wave and a phase separation . for a strong repulsive on - site interaction ( @xmath13 , @xmath14 ) ,
the ground state avoids double occupancy and the spin density is periodic along the lattice , leading to an antiferromagnetic ordering , namely spin - density wave .
its order parameter is given by , @xmath15,\ ] ] where @xmath16 .
in the limit @xmath17 , the ground state is dominated by the following configurations : @xmath18 where the state is described in the real space mode representation , in which each site can be in the following set of configurations : @xmath19 . considering a strong repulsive inter - site interaction ( @xmath20 , @xmath21 ) ,
a periodic fermionic density is generated , leading to a charge - density wave .
its order parameter is given by , @xmath22.\ ] ] in the limit @xmath23 , the ground state is dominated by the following configurations , @xmath24 in the range of strong attractive interactions ( @xmath25 or @xmath13 , @xmath26 with @xmath27 ) , the fermions cluster together , and the ground state becomes inhomogeneous , with different average charge densities in its distinct spatial regions . such a phase is called phase separated state . in the limit @xmath28 ,
the ground state is dominated by the following configurations , @xmath29 where @xmath30 is the set of translation operators . in the weak coupling limit , different phases appear . for small attractive inter - site interactions ( @xmath26 ) ,
superconducting phases are raised , characterized by the pairing correlations , @xmath31 with the respective order parameter @xmath32 .
if the on - site interactions are lower than the inter - site interactions ( @xmath33 ) , the fermions will pair as a singlet superconductor , characterized by nearest - neighbor ( @xmath34 ) or on - site ( @xmath35 ) singlet pairing correlations given by , @xmath36 where @xmath37 . on the other hand ,
if the on - site interactions are higher than the inter - site interactions ( @xmath38 ) , we have a triplet superconductor , characterized by nearest - neighbor triplet pairing correlations ( @xmath39 ) given by , @xmath40 where @xmath37 .
note that the difference between the singlet and triplet pairing correlations is simply a plus or minus sign .
it can be clarified if we consider , for example , the case of two fermions in a singlet or triplet spin state , given by @xmath41 . expanding this state , we have , @xmath42 where the lower / upper sign corresponds to the singlet / triplet pairing correlation . the last phase in the diagram
is the controversial bond - order - wave ( bow ) . by studying the ehm ground state broken symmetries , using level crossings in excitation spectra , obtained by exact diagonalization , nakamura @xcite argued for the existence of a novel bond - order - wave phase for small to intermediate values of positive u and v , in a narrow strip between cdw and sdw phases .
this phase exhibits alternating strengths of the expectation value of the kinetic energy operator on the bonds , and is characterized by the following order parameter , @xmath43 , \end{aligned}\ ] ] where @xmath44 is the kinetic energy operator associated with the @xmath45th bond .
nakamura argued that the cdw - sdw transition is replaced by two separate transitions , namely : ( i ) a continuous cdw - bow transition ; and ( ii ) a berezinskii - kosterlitz - thouless ( bkt ) spin - gap transition from bow to sdw .
such remarkable proposal was later confirmed by several works @xcite , employing different numerical methods , like dmrg , monte carlo or exact diagonalization .
nevertheless , while the bow - cdw phase boundary can be well resolved , since it involves a standard second order ( continuous ) phase transition , the sdw - bow boundary is more difficult to locate , for it involves a bkt transition in which the spin gap opens exponentially slowly as one enters the bow phase .
the precise location of the bow phase is then still a subject of debate . to the best of our knowledge ,
the best estimates for the transitions , taking @xmath46 , correspond to a cdw - bow transition at @xmath47 @xcite , and to a bow - sdw transition in the range @xmath48 @xcite , or @xmath49 @xcite .
despite widely studied in systems of distinguishable particles , entanglement or more general notions of quantum correlations have received less attention in the case of indistinguishable particles . in this case , the space of quantum states is restricted to symmetric ( @xmath50 ) or antisymmetric ( @xmath51 ) subspaces , depending on the bosonic or fermionic nature of the system , and the particles are no longer accessible individually , thus eliminating the usual notions of separability and local measurements , and making the analysis of correlations much subtler .
in fact , there are a multitude of distinct approaches and an ongoing debate around the entanglement in these systems @xcite . nevertheless , despite the variety , the approaches consist essentially in the analysis of correlations under two different aspects : the correlations genuinely arising from the entanglement between the particles ( entanglement of particles ) @xcite , and the correlations arising from the entanglement between the modes of the system ( entanglement of modes ) @xcite . these two notions of entanglement are complementary , and the use of one or the other depends on the particular situation under scrutiny . for example , the correlations in eigenstates of a many - body hamiltonian could be more naturally described by entanglement of particles , whereas certain quantum information protocols could prompt a description in terms of entanglement of modes .
once one has opted for a certain notion of entanglement , there are interesting methods to quantify it @xcite .
entanglement of modes can be understood by mapping the quantum state in its number representation , namely , @xmath52 where @xmath53 , and @xmath54 is an arbitrary set of @xmath55 fermionic operators describing the single particle modes of the system ( not necessarily the real space modes as in the hamiltonian definition ) .
we will denote hereafter as `` configuration representation ( number representation ) '' the left ( right ) side of the previous equation .
such equation corresponds to a mapping to distinguishable qubits , represented by the occupied ( @xmath56 ) or unoccupied ( @xmath57 ) modes , which then allows one to employ all the tools commonly used in distinguishable quantum systems in order to analyze their correlations .
one could , for example , use the von neumann entropy of the reduced state representing a block with @xmath58 modes , in order to quantify the entanglement between this block with the rest of the modes .
the reduced state is obtained by the partial trace in the number representation ( @xmath59 ) .
notice that , in the mode representation , _ local _ observables may actually involve correlations between particles .
for example , in the hubbard model , although the operator `` @xmath60 '' acts locally at the @xmath5th site ( real space modes ) , it describes pairing correlations between particles .
the algebra of local observables at the modes , defined in the number representation , is generated by , @xmath61 where @xmath62 , with @xmath63 and @xmath64 is the identity operator acting on mode @xmath65 . in this way ,
unentangled states are those which can be completely described by such local observables .
it is known that such states are simply the separable states in the usual tensor product form , @xmath66 .
based on the previous reasoning , we now define the notion of _ entanglement of particles_. notice first that one can not analyze the system under the usual paradigm of _ separability and locality _ , where the reduced states obtained by partial trace are mixed ( @xmath67 ) , whenever the global state is pure and entangled . therefore , in the case of indistinguishable particles in the configuration representation , the use of partial trace to characterize entanglement should be carefully reviewed , since it would suggest that all pure fermionic states are entangled , given that their reduced states are always mixed . in order to generalize the notion of entanglement for systems of indistinguishable particles , the approach based on the algebra of observables sheds light on the problem and allows us to go beyond the paradigm of separability and locality .
we now define the proper algebra of `` local observables '' as the one composed by operators which do not create correlations between the indistinguishable particles .
such algebra , defined in the configuration representation , is generated by the following single particle operators , @xmath68 where @xmath69 is the number of particles .
equivalently , using the second quantization formalism , the above set is given by the number conserving quadratic operators , @xmath70 .
the states that can be completely described by such algebra form , in this way , the set of unentangled states , where any particle is not entangled with any other .
intuitively , we would expect that this set corresponded to single slater determinants with fixed particle number .
more precisely , for a system with @xmath69 fermions , it is given by , @xmath71 where @xmath72 is an arbitrary set of fermionic operators .
recall that these operators can not be quasiparticles with particle - hole superpositions , as usual in a bogoliubov transformation , since the above states have a fixed number of fermions .
in fact , distinct approaches confirmed that such set does indeed correspond to the unentangled states @xcite .
the only non - classical correlation present in such states is the _ exchange _ , due to the antissymetrization , which does not constitute entanglement .
for example , in @xcite the analysis follows by using a very elegant mathematical formalism , called gns ( gelfand - naimark - segal ) construction , for the case of two fermions , each one with hilbert space dimension 3 or 4 , and two bosons with dimension 3 ; in @xcite the authors propose a `` generalized entanglement ( ge ) '' measure , obtaining a simple formula for the `` partial trace '' , and the set of fermionic unentangled states for an arbitrary number of particles ; or also in @xcite , where a general notion of quantum correlation beyond entanglement ( the _ quantumness _ of correlations ) is investigated by means of an `` activation protocol '' , which yields the same set of states with no _ quantumness _ as the above unentangled one .
as in the case of distinguishable modes , the von neumann entropy also provides a good quantifier for the entanglement of indistinguishable particles .
we can define the shifted von neumann entropy of entanglement @xcite as follows , @xmath73 where @xmath74 is the single particle reduced state , the partial trace is taken in the configuration space , and @xmath75 is the von neumann entropy .
such a quantifier is obtained simply by noticing that each extremal state in the one particle reduced space is respective to a unique single slater determinant @xcite . more precisely , the single particle reduced state of a single slater determinant ( as eq . )
is given by , @xmath76 and its particle entanglement is null , @xmath77 .
if the state can not be described by a single slater determinant , its entanglement is necessarily non null , @xmath78 , and at least one of its particles is entangled with another one .
maximally entangled states have their single particle reduced states described by the maximally mixed state , as for example , the strong coupling limit phases sdw , cdw , and ps , as described in eqs .
, , , respectively , whose reduced state is given by , @xmath79 which have maximal von neaumann entropy , @xmath80 , and consequently maximal particle entanglement , @xmath81 .
if the hamiltonian has certain symmetries , its ground state entanglement can be analytically calculated as a simple function of its quadratures @xcite . in our particular case , from both @xmath82 and translational symmetries in the extended hubbard model , formally given by , @xmath83 where @xmath84 is the annihilation ( creation ) fermionic operator of a particle in the @xmath5th site , with spin @xmath4 , and @xmath85 is the ground state of the hamiltonian , we have that its single particle reduced state ( @xmath86 ) is disjoint in the subspaces with distinct spin , @xmath87 , and each of these terms is given by a circulant matrix , @xmath88 @xmath89 where @xmath90 is the lattice size . in this way the matrix is easily diagonalized , and its eigenvalues @xmath91 are given by a fourier transform of the quadratures , @xmath92 .
\label{eigen.mom.space}\ ] ] the entanglement is then directly obtained from eq.([shifted.ent ] ) .
the computation of the single particle correlations , and consequently the entanglement of particles , was numerically performed using dmrg . although dmrg is less accurate for problems with periodic boundary conditions ( pbc ) than with open boundary conditions ( obc ) , from the physical viewpoint pbc are strongly preferable over obc , as boundary effects are eliminated and finite size extrapolations can be performed for much smaller system sizes . in this work
we analyze the extended hubbard model considering pbc .
our simulations were performed for systems up to @xmath93 sites , always keeping a large enough dimension ( @xmath45 ) for the renormalized matrices ( ranging from @xmath94 to @xmath95 ) and number of sweeps ( @xmath96 sweeps ) , in order to obtain an accurate precision . in fig.[error.analysis ] , we see that @xmath45 ranging from @xmath97 to @xmath98 is enough for an entanglement accuracy of the order of @xmath99 . the accuracy for the ground state energy , as well as the truncation error , using such parameters , are of the order of @xmath100 , as a function of @xmath45 ( dimension of the renormalized matrices ) , at the point @xmath101 , and using @xmath102 sweeps in the computation , which is enough for the ground state convergence . ] our results for the entanglement of particles in the extended hubbard model at half - filling are shown in fig.[phase.diagram.ent.part ] , to be dissected below .
it is remarkable that such picture highlights the known phase diagram of the model .
we first note that , as expected , we have a maximum of entanglement at the strong coupling limits ( @xmath103 ) , and as we decrease the interactions between the particles , the entanglement tends also to decrease , until the unentangled case for the non interacting hamiltonian ( @xmath104 ) .
the figure thus presents the shape of a valley around this point .
following then the discontinuities and the local minimum points in the entanglement , we can easily identify the quantum phase transitions , except for both the subtle sdw - bow transition , and the transition between the superconductor phases ts - ss . in the former case
, one needs to recall that the observation of the bow phase is by itself a hard task , since its gap opens exponentially slowly , and also that there are evidences that such transition is of infinite order @xcite .
therefore we believe that a possible detection of such transition by the entanglement of particles would require higher precision numerical analysis as well as the study of larger lattice sizes .
concerning the ts - ss transition , on the one hand the order of the two superconducting phases transition is controversial , being identified as a bkt transition @xcite as well as a second order one @xcite in the literature . on the one hand , we would be led to strengthen the result of a bkt transition , since our entanglement does not detect it . on the other hand , it is reasonable the apathy of the entanglement of particles on distinguishing the two phases , since the correlations between the particles in the two superconducting phases have essentially the same characteristics .
thus it is hard to precisely conclude the reason for the failure to detect such transition with our measure of entanglement . '' as a function of the interaction terms @xmath105 and @xmath106 , in a system with @xmath107 sites at half - filling .
the entanglement behavior in the thermodynamic limit , @xmath108 , keeping fixed the filling @xmath109 , is qualitatively the same , with slight differences of the order of @xmath110 in its magnitude ; see appendix [ sec.finite.size.scaling ] for a detailed discussion .
the ( green ) continuous line denotes the discontinuity at the entanglement function , while the ( red ) dashed line denotes the local minima .
the white dots correspond to the points where we performed a detailed finite - size scaling analysis ( see table i in appendix [ sec.finite.size.scaling ] ) . ]
the discontinuities in the entanglement are directly identified with the first order quantum phase transitions , whereas the minimum points are identified with the second order quantum phase transitions .
when crossing a first order transition , the ground state energy presents a discontinuity and consequently also its observables . in this way , the eigenvalues `` @xmath111 '' of the single particle reduced density matrix ( eq . ) , and the entanglement obtained from them , should present a discontinuity .
the occurrence of the minimum points are due to the divergence of the correlation length when approaching the second order transitions .
as described in the previous section , the eigenvalues `` @xmath111 '' are given in momentum space by the fourier transform of the real space quadratures `` @xmath112 '' ( eq.([eigen.mom.space ] ) ) . in this way , if we are close to the transitions , such real space quadratures tend to become delocalised or spread out along the lattice , thus leading to more localised eigenvalue distributions in momentum space , and consequently to smaller von neumann entropies .
it is worth remarking that such behavior is the opposite of the entanglement of modes , where the sites are maximally entangled at the second order transitions . as an example , see in fig.[momentum.eig ] the eigenvalue distribution for a system with @xmath107 sites when crossing the bow - cdw quantum phase transition . ''
along the lattice sites , and * ( bottom ) * eigenvalue distribution `` @xmath111 '' for the single particle reduced state in a fixed spin sector , as given in eq.([eigen.mom.space ] ) .
we consider a fixed @xmath46 .
the vertical axis is in log - scale , in order to make clearer the visualisation .
as we approach the bow - cdw quantum phase transition point , at @xmath113 , we see that the quadratures tend to delocalise along the lattice , whereas the eigenvalue distribution becomes more localised.,title="fig : " ] '' along the lattice sites , and * ( bottom ) * eigenvalue distribution `` @xmath111 '' for the single particle reduced state in a fixed spin sector , as given in eq.([eigen.mom.space ] ) .
we consider a fixed @xmath46 .
the vertical axis is in log - scale , in order to make clearer the visualisation .
as we approach the bow - cdw quantum phase transition point , at @xmath113 , we see that the quadratures tend to delocalise along the lattice , whereas the eigenvalue distribution becomes more localised.,title="fig : " ] we present now the entanglement behavior in some specific slices of the phase diagram with @xmath107 , in order to clarify the above discussion and results .
more specifically , we show the entanglement behavior in the ps - ss - cdw , ps - ss - sdw , and ps - sdw - cdw transitions . notice that our finite - size scaling analysis showed that in thermodynamic limit the entanglement behavior is qualitatively similar ( see appendix ) , with a scaling inversely proportional to the lattice size , @xmath114 , where @xmath115 and @xmath116 are constants . in fig.[slice.ps.ss.cdw ]
we see the entanglement behavior across the ps - ss - cdw phases .
we clearly see , for any fixed attractive on - site interaction ( @xmath117 ) , a discontinuity in the entanglement followed by a local minimum point , as we increase the value of the inter - site interactions @xmath105 .
the discontinuity is related to the first order transition ps - ss , while the local minimum is related to the second order transition ss - cdw .
we see , however , that the ss - cdw transition is not located exactly at @xmath118 , as expected from the phase diagram described in the literature , but at a value close to this one .
we believe that this discrepancy is related to finite - size effects . ) , is characterized by a discontinuity ( ps - ss transition ) , followed by a local minimun ( ss - cdw transition ) . ] in fig.[slice.ps.ts.sdw ] we see the entanglement behavior across the ps - ts - sdw phases .
we see again the two kinds of behavior for any fixed attractive inter - site interaction ( @xmath119 ) : a first discontinuity , related to the first order transition ps - ts , followed by a local minimum point related to the second order transition ts - sdw .
note that , for large values of the attractive inter - site interaction , @xmath120 , the discontinuity and minimum converge to the same point , and there is no ts phase anymore . ) , is characterized by a discontinuity ( ps - ts transition ) , followed by a local minimun ( ts - sdw transition ) . for large @xmath105 ,
the two transitions shrink at the same point , and there is no ts phase anymore . ] in fig.[slice.ps.sdw.bow.cdw ] we see the entanglement behavior across the ps - sdw - bow - cdw phases . we see that , as we increase the value of the inter - site interactions , for any fixed repulsive on - site interactions ( @xmath121 ) , the entanglement identifies two transitions .
firstly we see a discontinuity , related to the first order transition ps - sdw , followed then by : ( i ) a discontinuity , when considering large @xmath106 , or ( ii ) a local minimum point , when considering small @xmath106 .
such discontinuity is related to the first order sdw - cdw transition , while the minimum points are related to the second order bow - cdw transition ( the sdw - bow transition is not seen , as aforementioned ) .
we see that the transitions to the cdw phase occur at @xmath122 .
performing a finite - size scaling analysis ( see appendix ) we obtain that , for @xmath46 , the bow - cdw transition is located at @xmath123 , which is slightly lower than the literature results , namely @xmath47 @xcite . .
the entanglement , for any fixed repulsive on - site interaction ( @xmath106 ) , is characterized by a discontinuity ( ps - sdw transition ) , followed by : ( i ) a discontinuity for large @xmath105 ( sdw - cdw transition ) , or ( ii ) local minimun points for small @xmath105 ( bow - cdw transition ) .
, title="fig : " ] .
the entanglement , for any fixed repulsive on - site interaction ( @xmath106 ) , is characterized by a discontinuity ( ps - sdw transition ) , followed by : ( i ) a discontinuity for large @xmath105 ( sdw - cdw transition ) , or ( ii ) local minimun points for small @xmath105 ( bow - cdw transition ) .
, title="fig : " ]
we studied the entanglement of indistinguishable particles in the extended hubbard model at half - filling , with focus on its behavior when crossing the quantum phase transitions .
our results showed that the entanglement either has discontinuities , or presents local minima , at the critical points .
we identified the discontinuities as first order transitions , and the minima as second order transitions . in this way
, we concluded that the entanglement of particles can `` detect '' all transitions of the known diagram , except for the subtle transitions between the superconductor phases ts - ss , and the transition sdw - bow .
it is also interesting to compare our results with other entanglement measures , such as the entanglement of modes , which was widely studied in several models , as well as in the extended hubbard model @xcite .
@xcite firstly showed that the entanglement of modes , i.e. , the entanglement of a single site with the rest of the lattice , could detect three main symmetry broken phases , more specifically , the cdw , sdw and ps .
other phases were not identified due to the fact that they are associated to off - diagonal long - range order .
further investigation were performed analysing the block - block entanglement @xcite , i.e. , the entanglement of a block with @xmath124 sites with the rest of the lattice ( @xmath125 sites ) , showing that this more general measure could then detect the transition to the superconducting phase , as well as the bond - order phase .
the measure , however , could not detect the ss - ts transition , besides presenting some undesirable finite - size effects in the ps phase .
on the other hand , the entanglement of particles studied in this work showed no undesirable finite - size effects in the ps phase , but could not detect the superconductor ss - ts transition either . regarding the bow phase , from the above discussion
we see that it would be worth to analyze more general measures for the entanglement of particles , which goes beyond single particle information .
some steps in this direction were made in @xcite , where a notion of entanglement of `` subgroups '' of indistinguishable particles was defined .
we acknowledge financial support by the brazilian agencies fapemig , cnpq , and inct - iq ( national institute of science and technology for quantum information ) .
in this appendix we perform a finite - size scaling analysis in the system entanglement , in order to extract information about the ground state of the model .
we obtained that the entanglement behavior is qualitatively the same for lattices larger than @xmath126 , with just small differences of the order of @xmath127 in its magnitude . in a general way
, the entanglement scales with the inverse of the lattice size , @xmath114 , where @xmath115 and @xmath116 are constants .
see in fig.[figure.finite.scaling ] , for example , the entanglement scaling for the sdw - bow - cdw phase transitions . in tab .
i we show the computed values for the scaling constants at different points in the phase diagram , as highlighted in fig.[phase.diagram.ent.part ] . , we have : * ( upper - left ) * minimum entanglement critical point @xmath128 ; * ( upper - right ) * @xmath129 ; * ( bottom - left ) * @xmath130 . for a fixed @xmath131 * ( bottom - right ) * , there is a discontinuity in the entanglement , highlighted by the scaling at @xmath132 ( bottom curve ) , and @xmath133 ( upper curve ) . ]
.scaling constants for the entanglement , @xmath134 , at different points of the phase diagram , as highlighted in fig.[phase.diagram.ent.part ] .
the symbol `` @xmath135 '' denotes the critical points , and `` @xmath136 '' means that the entanglement is constant , apart from numerical inaccuracy , for the analyzed lattices . [
cols="^,^,^,^",options="header " , ] b. bauer , l.d .
carr , h.g evertz , a. feiguin , j. freire , s. fuchs , l. gamper , j. gukelberger , e. gull , s. guertler , a. hehn , r. igarashi , s.v .
isakov , d. koop , p.n .
ma , p. mates , h. matsuo , o. parcollet , g. pawowski , j.d .
picon , l. pollet , e. santos , v.w .
scarola , u. schollwck , c. silva , b. surer , s. todo , s. trebst , m. troyer , m.l . wall , p. werner , and s. wessel , j. stat .
mech , * 2011 * , p05001 ( 2011 ) .
a. p. balachandran , t. r. govindarajan , a. r. de queiroz and a. f. reyes - lega , phys .
lett . * 110 * , 080503 ( 2013 ) ; a. p. balachandran , t. r. govindarajan , a. r. de queiroz and a. f. reyes - lega , phys . rev .
a * 88 * , 022301 ( 2013 ) .
r. g. parr and w. yang , density matrices : the n - representability of reduced density matrices . in : density - functional theory of atoms and molecules , the international series of monographs on chemistry - 16 , oxford science publications . 1989 . | we investigate the quantum phase transitions of the extended hubbard model at half - filling with periodic boundary conditions employing the entanglement of particles , as opposed to the more traditional entanglement of modes .
our results show that the entanglement has either discontinuities or local minima at the critical points .
we associate the discontinuities to first order transitions , and the minima to second order ones .
thus we show that the entanglement of particles can be used to derive the phase diagram , except for the subtle transitions between the phases sdw - bow , and the superconductor phases ts - ss . |
macro was a large area multipurpose underground detector @xcite designed to search for rare events and rare phenomena in the penetrating cosmic radiation .
the experiment obtained important results on atmospheric @xmath5 oscillations and performed @xmath5 astronomy studies , indirect searches for wimps and searches for low energy @xmath6 from stellar gravitational collapses @xcite .
the detector was located in hall b of the underground gran sasso lab at an average rock overburden of 3700 m.w.e . ; it started data taking with part of the apparatus in @xmath7 ; it was completed in @xmath8 and was running in its final configuration until december 2000 .
the detector had global dimensions of @xmath9 and provided a total acceptance to an isotropic flux of particles of @xmath10 .
the detector was composed of three horizontal layers of liquid scintillation counters , 14 layers of limited streamer tubes and one layer of nuclear track detectors , fig .
vertically it was divided into two parts : the lower part contained 10 layers of streamer tubes , 7 layers of rock absorbers and 2 layers of liquid scintillators ; the upper part was empty , contained the electronics and was covered by 1 layer of scintillators and 4 layers of streamer tubes .
the sides of the detector were covered with 1 layer of scintillators and 6 layers of limited streamer tubes , so as to obtain a closed box structure .
each of the subdetectors could be used in stand - alone and in combined mode . in the following
we shall briefly recall neutrino oscillation formulae and atmospheric neutrinos ; then we shall discuss our main results including the ratio of vertical to horizontal muons and the study of the @xmath2 distribution .
after a detailed analysis of the present situation of the different monte carlos ( mcs ) , we summarize our atmospheric neutrino oscillation results and we conclude with a short discussion of some neutrino astrophysics studies .
* neutrino oscillations . * if neutrinos have non - zero masses , one has to consider the 3 weak _ flavour eigenstates _
@xmath11 , @xmath12 , @xmath13 and the 3 _ mass eigenstates _
@xmath14 , @xmath15 , @xmath16 .
the flavour eigenstates @xmath17 are linear combinations of the mass eigenstates @xmath18 .
neutrino oscillations depend on six independent parameters : two mass - squared differences , @xmath19 and @xmath20 , three mixing angles @xmath21 , @xmath22 , @xmath23 and the cp - violating phase @xmath24 . in the simple case of two flavour eigenstates ( @xmath12 , @xmath13 ) which oscillate with two mass eigenstates ( @xmath15 , @xmath16 ) , and @xmath25 one has : @xmath26 @xmath27 the survival probability of a @xmath12 beam is @xmath28 where @xmath29 and l is the distance travelled by the neutrino of energy @xmath30 from production to detection .
the two - neutrino approximation is adequate for discussing our data .
* atmospheric neutrinos . * high energy ( he ) primary cosmic rays ( crs ) , protons and nuclei ,
interact in the upper atmosphere producing a large number of pions and kaons , which decay yielding muons and muon neutrinos ; then muons decay yielding @xmath12 s and @xmath11 s
. the ratios of their numbers are @xmath31 and @xmath32 .
the atmospheric neutrinos are produced in a spherical shell at about 10 - 20 km above ground and proceed towards the earth .
atmospheric neutrinos have energies from a fraction of gev up to more than 100 gev and they travel distances l from few tens of km up to @xmath33 km .
an underground detector is illuminated by a flux of neutrinos from all directions and it can make oscillation studies for @xmath34 .
macro detected upgoing muon neutrinos via charged current interactions , @xmath35 ; the upgoing muons were identified with the streamer tube system ( for tracking ) and the scintillator system ( for time - of - flight measurement ) .
the events measured and expected for the three measured topologies , indicated in fig . [ fig1 ] and the @xmath2 distribution deviate from monte carlo expectations without oscillations , figs.[fig2 ] , [ fig4 ] and tables 1 , 2 ; the deviations point to the same @xmath36 oscillation scenario @xcite-@xcite .
* monte carlos . * the measured data of fig .
[ fig2 ] were compared with different mc simulations . in the past
we used the neutrino flux computed by the bartol96 group @xcite and the grv94 @xcite parton distribution .
for the low energy channels the cross sections in @xcite were used ; the propagation of muons to the detector was done using the energy loss calculation by lohmann et al . @xcite .
the total systematic uncertainty in the predicted flux of upthroughgoing muons , adding in quadrature the errors , was estimated to be @xmath37 ; this is mainly a scale error that does not change the shape of the angular distribution .
the error on the shape of the distribution is @xmath38 .
the detector was simulated using geant @xcite .
a similar mc ( honda96 ) was used by the superkamiokande collaboration @xcite .
recently new improved mc predictions were made available by the honda @xcite and fluka @xcite groups .
they include three dimensional calculations of hadron production and decays and of neutrino interactions , improved hadronic model and new fits of the primary cosmic ray flux .
the two mcs yield predictions for the non oscillated and oscillated @xmath5 fluxes in complete agreement with each other , equal to within few % , see figs .
[ fig3]a , b. the shapes of the angular distributions for oscillated and non oscillated bartol96 , new fluka and new honda fluxes are the same to within few % , fig .
[ fig3]a , b. the absolute values of our upthroughgoing muon data are about @xmath39 above those predicted by the new fluka and honda mcs ; this situation is also true for the new superkamiokande data @xcite ; the measurements of the flux value at @xmath40 performed by various experiments do not seem to be conclusive @xcite .
note in fig .
[ fig3]a that the new fluka muon flux with an older cosmic ray fit is considerably above the new fluka flux with the new cr fit .
our high energy @xmath5 data suggest that the new honda and fluka predictions should be raised by about @xmath41 , probably because of the used new cr fit .
the inclusion of the new attic collaboration measurements of primary crs seems to lead to the old energy dependence of @xmath42 @xcite .
thus the bartol96 mc may probably still be used for the prediction of the absolute flux , besides predicting well the shape of the angular distribution . in the following we shall use mainly the predictions of this mc .
it should be noted that the evidence for neutrino oscillations rests primarely with the shape of the angular distribution and this is the same in all mc calculations .
the _ upthroughgoing muons _ come from @xmath43 interactions in the rock below the detector ; the @xmath43 s have a median energy @xmath44 ; muons with @xmath45 cross the whole detector .
the data taking livetime was slightly over 6 years ( full detector equivalent ) . the data , table 1 and fig .
[ fig2]a , deviate in shape and in absolute value from the bartol96 mc predictions ( see section 2 ) .
this was first pointed by macro in 1995 @xcite .
\(a ) ( b ) a large number of possible systematic effects and backgrounds that could affect the measurements were studied .
one of the most significant checks was performed using only the scintillator system with two different and independent electronic systems @xcite .
two different final analyses of the upthrough data have been performed : they agree to within @xmath46 . assuming no oscillations , the number of expected upthroughgoing muon events integrated over all zenith angles from bartol96 without oscillations is 1169 ; the measured number is 857 , table 1 .
thus the ratio of the observed number of events to the bartol96 expectation is 0.73 .
* @xmath5 @xmath47 @xmath48 versus @xmath5 @xmath49*. matter effects due to the difference between the weak interaction effective potential for muon neutrinos with respect to sterile neutrinos ( @xmath50 ) would produce a different total number and a different zenith angle distribution of upthroughgoing muons . the ratio @xmath51 vertical / horizontal @xmath52 was used to test the @xmath5 @xmath53 oscillation hypothesis versus @xmath5 @xmath47 @xmath48 @xcite , @xcite , @xcite .
the measured value is @xmath54 ; it should be compared to @xmath55 and @xmath56 , which are the minimum values of the ratios for @xmath5 @xmath47 @xmath48 and @xmath5 @xmath57 oscillations for @xmath58 and maximal mixing .
\(a ) ( b ) .ratios between different event categories : @xmath59 is the measured value , @xmath60 is the expected value from the bartol96 mc without oscillations ; using the new fluka mc the ratios differ by at most @xmath61 .
the last column gives the one - sided probability compatible with a statistical fluctuation computed according to the _ a priori _ assumptions explained in the text . [ cols="^,^,^,^ " , ] for @xmath5 @xmath62 @xmath48 , the one - sided probability to measure a value lower than @xmath63 is @xmath64 . for @xmath5 @xmath49
the probability is @xmath65 ; @xmath66 is about 480 times larger than @xmath67
. therefore @xmath5 @xmath49 oscillations ( with any mixing ) are excluded at about @xmath68 c.l . with respect to @xmath5 @xmath47 @xmath48 oscillations with maximal mixing @xcite .
* oscillation probability as a function of the ratio @xmath2*. @xmath69 was estimated by measuring the muon energy , @xmath70 , by means of the muon multiple coulomb scattering ( mcs ) in the rock absorbers in the lower macro .
the best method employed the streamer tubes in `` drift mode '' , using the special electronics designed to search for magnetic monopoles @xcite .
the space resolution achieved was @xmath71 . for each muon ,
seven variables were given in input to a neural network ( nn ) previously trained to estimate muon energies with mc events of known input energy crossing the detector at different zenith angles .
the distribution of the ratio @xmath72 obtained by this analysis is plotted in fig . [ fig4 ] as a function of @xmath73 @xcite . notice that the data extend from @xmath74 km / gev to @xmath75 km / gev .
the _ internal upgoing _ ( iu ) muons ( @xmath76 157 events , 285 expected by bartol96 and 235 expected by new fluka , table 1 ) come from @xmath77 gev @xmath78 s interacting in the lower apparatus , fig .
compared to the no - oscillation prediction there is a reduction of about a factor of two in the flux of these events , without any appreciable distortion in the shape of the zenith distribution , fig .
[ fig2]b @xcite , @xcite .
the mc predictions for no oscillation in figs .
[ fig2]b and [ fig2]c are given by the dashed lines with a @xmath79 systematic scale band .
notice that at these energies , the new honda and fluka mc predictions are low by about @xmath80 .
the _ upstopping muons _ ( ugs ) are due to @xmath0 gev @xmath43 s interacting below the detector and yielding upgoing muons stopping in the detector .
the _ semicontained downgoing muons _
( i d ) are due to @xmath81-induced downgoing muon tracks with vertex in the lower macro .
the two types of events ( 262 compared to 354 expected by bartol96 and 315 by fluka , table 1 ) are identified by means of topological criteria ; the lack of time information prevents to distinguish the two sub - samples .
the upgoing @xmath5 s should all have oscillated completely , while the downgoing @xmath5 do not .
the zenith distribution shows , as expected , a uniform deficit of about 25 % of the measured number of events with respect to the no - oscillation prediction , fig .
[ fig2]c @xcite , @xcite .
in the past , in order to determine the oscillation parameters , macro made fits to the shape of the upthroughgoing muon distribution and to their absolute flux compared to the bartol96 mc prediction ( with a @xmath82 systematic scale uncertainty ) .
the other data were only used to verify the consistency and make checks .
the result was @xmath83 ev@xmath84 and maximal mixing .
later , in @xcite the high energy data for the zenith distribution , @xmath85 , were combined with the he neutrino energy measurement , @xmath86 .
they are independent of the neutrino absolute flux ; with these two informations the significance of observation of neutrino oscillations is about 4.7 @xmath87 . in order to reduce the effects of possible uncertainties in the mcs ( to about @xmath88 ) we now use the following three independent ratios @xcite . 1 .
high energy data : zenith distribution ratio : @xmath89 2 .
high energy data , neutrino energy measurement ratio : @xmath90 3 .
low energy data : ratio @xmath91 .
table 2 gives for each ratio the measured value , @xmath63 , the prediction , @xmath92 , for no oscillations using either the bartol96 or the new fluka mc ; the probability value for no oscillations obtained allowing the predicted ratio to fluctuate around the mean value ( considering the total number of events equal to the predicted value without oscillations ) .
the no oscillations hypothesis has a probability p @xmath93 ( table 2 ) , ruling out the no - oscillation hypothesis by @xmath94 .
the formula used for combining independent probabilities is @xmath95 @xcite . by fitting the three ratios to the @xmath5 @xmath47 @xmath48 oscillation formulae we obtain @xmath96 , @xmath97 ev@xmath98 and the allowed region indicated in fig .
[ fig5]a .
we could also add the information on the absolute flux values of the 1 .
high energy data , assuming the validity of the bartol96 flux and a systematic scale error of @xmath99 , @xmath100 , @xmath101 2 .
low energy semicontained muons , with a systematic scale error of @xmath102 , + @xmath103 , @xmath104 ( at these low energies the bartol96 , honda 2001 and fluka 2002 mc neutrino fluxes are essentially equal ) . fig .
[ fig5]a shows the macro allowed regions for the @xmath105 oscillation parameters in the @xmath106 plane using only the three ratios @xmath107 , @xmath108 and @xmath109 ( continuous line ) and using also the two estimated absolute values , assuming the validity of the bartol96 fluxes ( dashed line ) . in fig .
[ fig5]b are compared the macro , superkamiokande ( sk ) @xcite and soudan 2 @xcite allowed regions .
all limiting lines are qualitative smoothed interpolations .
* search for astrophysical sources of high energy muon neutrinos .
* high energy @xmath43 s are expected to come from several galactic and extragalactic sources .
neutrino production requires astrophysical accelerators of protons and astrophysical `` beam dumps '' .
a sensitive search was made for upgoing muons produced by neutrinos coming from celestial sources , interacting below the detector .
upper limits were established on the muon fluxes from specific celestial sources ; the limits are in the range @xmath110 . in the case of gx339 - 4 ( @xmath111 , @xmath112 ) and cir x-1 ( @xmath113 , @xmath114 ) ,
we have 7 events : they have been considered as background , therefore the upper flux limits are higher ; but the events could also be indication of signals @xcite .
the pointing capability of the detector was tested by determining the shadows of the moon and of the sun on the primary cosmic ray flux @xcite a search for time coincidences of the upgoing muons with @xmath115-ray bursts was made .
no statistically significant time correlation was found .
a search was also made for a diffuse astrophysical neutrino flux for which an upper limit at the level of @xmath116 was established @xcite .
* * indirect searches for wimps.**weakly interacting massive particles ( wimps ) could be part of the galactic dark matter ; they could be intercepted by celestial bodies , slowed down and trapped in their centers , where wimps and anti - wimps could annihilate and yield neutrinos of multigev energy , in small angular windows from their centers .
one of the best wimp candidate could be the lowest mass neutralino .
a search was made for @xmath5 @xmath117 coming from the center of the earth using @xmath118 cones around the vertical ; we obtained muon flux limits of @xmath119 .
the limits were compared with the predictions of a supersymmetric model by bottino et al . ; our data eliminate a sizable range of parameters used in the model @xcite .
a similar procedure was used to search for muon neutrinos from the sun : the upper limits are at the level of about @xmath120 @xcite . * neutrinos from stellar gravitational collapses . * a stellar gravitational collapse of the core of a massive star is expected to yield a burst of all types of neutrinos and antineutrinos with energies of @xmath121 mev in @xmath122 .
the @xmath6 s can be detected via @xmath123 in the liquid scintillator .
about 120 @xmath6 events could have been detected in our 580 t liquid scintillator for a stellar collapse at the center of our galaxy .
two separate electronic systems were used .
both had an energy threshold of @xmath124 and recorded pulse shape , charge and timing informations of the positron . following a @xmath125 trigger ,
the first system phrase lowered its threshold to @xmath126 for @xmath127 in order to detect ( with a @xmath128 efficiency ) the @xmath129 @xmath115 released in the capture reaction @xmath130 induced by the neutron produced in the primary process .
a redundant supernova alarm system was in operation , alerting the physicists on shift ; a procedure was defined to alert the various supernova observatories around the world @xcite .
macro was completed at the end of 1994 ; the first parts of the detector started operation in 1989 .
no stellar gravitational collapses in our galaxy were observed from @xmath7 to 2000 .
the macro detector took data from 1989 till the end of year 2000 .
important results were obtained in all the items listed in the proposal .
in particular for atmospheric neutrino oscillations we analyzed different event topologies , different energies , exploited muon coulomb multiple scattering in the detector to measure muon energies and studied the effects of using different mcs .
all data are in agreement with the hypothesis of @xmath131 oscillations , with maximal mixing and @xmath132 ev@xmath98 , and rule out the no oscillation hypothesis by @xmath94 .
studies were made on muon neutrino astronomy @xcite , @xcite , seasonal , solar and sidereal variations @xcite @xcite , and on possible dark matter candidates : wimps ( looking for muon neutrinos from the centers of the earth and of the sun ) , nuclearites and q - balls @xcite , @xcite . in the search for gut magnetic monopoles
macro obtained the best existing direct flux upper limit ( @xmath133 ) over the widest @xmath134 range ( @xmath135 ) @xcite .
we would like to acknowledge the cooperation of all the members of the macro collaboration .
macro was a collaboration of us and italian institutions plus one moroccan group ; see ref .
@xcite for the list of authors and institutions .
m1 s. ahlen et al . , nucl .
instr . meth .
+ m. ambrosio et al . , nucl .
instr . meth .
res a486(2002)663 .
g. giacomelli et al . , lectures at the 6th constantine summer school , algeria ( 2002 ) , hep - ex/0211035 .
m. ambrosio et al . , astrop .
9(1998)105 ; lngs status report 2001 , hep - ex/0206027 .
s. ahlen et al . , phys .
p. bernardini , vulcano workshop , 1998 , hep - ex/9809003 .
+ m. ambrosio et al . , phys .
b434(1998)451 ; phys .
+ g. giacomelli et al . , trieste ictp school , hep - ph/9901355 ; hep - ex/0201032
. m. ambrosio et al . , phys .
b517(2001)59 ; hep - ex/0106049 . m. ambrosio et al . , physics/0203018 , nucl .
res a492(2002)376 .
+ m. ambrosio et al .
, hep - ex/0304037 , phys .
m. ambrosio et al .
j. c36(2004)357 .
+ g. giacomelli and a. margiotta , hep - ex/0406037 . v. agrawal et al .
m. gluck et al .
, z. phys .
p. lipari et al .
w. lohmann et al . , `` energy loss of muons in the range 1 - 10000 gev '' , cern 85 - 03 .
http://wwwasd.web.cern.ch/wwwasd/geant/ m. honda et al . , phys .
d64(2001)053011 ; phys .
g. battistoni et al . , astrop .
19(2003)269 ; `` erratum '' , astrop .
+ g. battistoni et al . , `` high energy extension of the fluka atmospheric neutrino flux '' , he2.4 , 1399 , icrc 2003 , tsukuba , japan .
y. hayato , `` status of the super - k , the k2k-2 experiments and jhf nu project '' , hep eps conference , aachen , 17 - 23 july 2003 .
+ y. fukuda et al . , phys .
81(1998)1562 ; phys .
m. f. crouch , proceedings @xmath136 icrc , moscow , ussr 6(1987)165 .
+ m. aglietta et al . , astrop .
+ d. de muth et al .
, hep - ex/0304016 , ( 2003 ) .
r. battiston , rapporteur talk on `` direct measurements and origin of cosmic rays '' , icrc 2003 , tsukuba , japan .
+ t. montaruli , rapporteur talk on `` high energy phenomena '' , icrc 2003 , tsukuba , japan .
m. ambrosio et al . , hep - ex/0207020 , eur .
j. c25(2002)511 .
+ g. giacomelli and l. patrizii , hep - ex/0302011 .
b. roe , probability and statistics in experimental physics , springer ( 1992 ) , pag .
allison et al . , phys .
b449 ( 1999 ) 137 .
+ t. mann et al . , nucl .
91 ( 2001 ) 134 ; nucl .
suppl . 110
( 2002)349 . m.
ambrosio et al .
, astrophys .
j. 546(2001)1038 ; astro - ph/0002492 .
m. ambrosio et al . , phys .
d59(1999)012003 ; hep - ex/9807006 , astro - ph/0302586 . m. ambrosio et al .
, astro - ph/0203181 , astrop .
+ s. ahlen et al .
, astrophys .
j. 412(1993)301 .
m. ambrosio et al . , phys .
d60(1999)082002 , hep - ex/9812020 .
s. ahlen et al . , astrop .
+ m. ambrosio et al .
8(1998)123 ; paper submitted to eur .
. j. c. e. witten , phys .
+ a. de rujula and s.l .
glashow , nature 312(1984)734 .
+ d. bakari et al , hep - ex/0004019 ; m. ambrosio et al . , eur .
j. c13(2000)453 .
m. ambrosio et al . , astro - ph/0211119 , phys . rev .
d 67 , 042002(2003 ) . | the final results of the macro experiment on atmospheric neutrino oscillations are presented and discussed . the data concern different event topologies with average neutrino energies of @xmath0 and @xmath1 gev .
multiple coulomb scattering of the high energy muons in absorbers was used to estimate the neutrino energy of each event .
the angular distributions , the @xmath2 distribution , the particle ratios and the absolute fluxes all favour @xmath3 oscillations with maximal mixing and @xmath4 .
a discussion is made on the monte carlos used for the atmospheric neutrino flux .
some results on neutrino astrophysics are also briefly discussed .
+ _ keywords : astroparticle physics ; underground detectors ; neutrino oscillations .
+ pacs nos . : 13.15.+g ; 14.60.pq ; 96.40.de ; 96.40.tv_
[ 1]\{. |
the bright x - ray nova a062000 ( v616 mon ) was discovered when it erupted in 1975 @xcite .
observations obtained after its return to quiescence revealed that a062000 is an interacting binary system with a k star donating mass to a compact star via an accretion disk @xcite .
early fits to the radial velocity variations of the k star gave a semi - amplitude of @xmath5 km s@xmath6 , which , when combined with the orbital period ( @xmath7 hr ) , yielded a mass function @xmath8 @xmath3 .
the mass function equals the minimum dynamical mass of the compact star .
since the mass function for a062000 is greater than the theoretical maximum mass for a neutron star , a062000 became a strong candidate for a black hole binary @xcite .
follow - up observations have refined the velocity semi - amplitude and determined the mass ratio of the binary : @xmath9 km s@xmath6 and @xmath10 ( marsh , robinson & wood 1994 ; orosz et al .
1994 found similar binary parameters ) . with these values ,
the masses of the stars are @xmath11 @xmath3 and @xmath12 @xmath3 , lacking only a determination of the orbital inclination to be fully determined .
one method for obtaining the orbital inclination of a062000 is to measure the ellipsoidal variations of the roche - lobe - filling k star .
two studies of the ellipsoidal variations have resulted in non - overlapping estimates of the inclination and , consequently , non - overlapping estimates of the mass of the compact star .
haswell et al . ( 1993 ) fit models including a roche - lobe - filling star and an accretion disk to simultaneous ubvr light curves of a062000 and found an inclination range for @xmath13 = 0.067 of @xmath14 .
this corresponds to a compact star mass of @xmath15 @xmath3 ( the compact star masses given here and elsewhere in this manuscript were calculated by us using the mass function determination from marsh , robinson & wood 1994 ) .
@xcite fit a k - band light curve of a062000 with a model including only the lobe - filling star and derived a best - fit inclination of 37@xmath16 , with a range of inclinations @xmath17
( 90% confidence limits for @xmath13 = 0.067 ; their fig . 2 ) .
this corresponds to @xmath18 = 14.2 @xmath3 , and a range of @xmath19 @xmath3 .
all authors agree that the compact star should be a black hole .
the orbital light curves of a062000 have shown clear evidence that the ellipsoidal variations are distorted : the light curve minima and maxima have varied in relative height and depth , the light curve minima have on occasion shown sharp features rather than smooth troughs , and the asymmetry between the two peaks has reversed ( haswell 1996 and sources therein ) .
the ellipsoidal variations can be distorted by star spots on the k star ; by a non - axisymmetric distribution of light across the accretion disk , such as the bright spot where the mass stream impacts the accretion disk ; by superhumps caused by a precessing , elliptical accretion disk ; and by dilution of the star light by non - variable flux from other components of the binary system , such as an axisymmetric accretion disk .
the presence of variable sources of flux in the system can confuse attempts to isolate the ellipsoidal component of the variations , and if a non - varying flux is present in the binary system but not in the models , the models will underestimate the true amplitude of the ellipsoidal variations and , in turn , underestimate the orbital inclination . the difference between the derived inclinations for a062000 stems in part from uncertainty about the magnitude of these distortions .
the light curve of a062000 has been monitored at r and i wavelengths ( e.g. , leibowitz , hemar & orio 1998 ) , but few observations have been published at longer wavelengths . since the contribution of the k star to the observed flux is maximized in the near - infrared , that wavelength region provides a good window to target the ellipsoidal modulation . in order to investigate the remaining uncertainties concerning the mass of its compact star and to measure the long - term behavior of its orbital light curve ,
we have reobserved the orbital light curve of a062000 in the j , h and k bandpasses .
this paper reports the results of our observations .
we observed a062000 on 1995 december 15 17 , 1996 january 27 29 and 1996 december 7 12 on the 2.7-m telescope at mcdonald observatory using rokcam , a near - infrared imaging camera @xcite . in all three observing runs
, we observed extensively in the h filter ( 1.45 1.85 @xmath20 m ) , where our s / n was maximized , and we supplemented these data with observations in the thermal k ( 2.05 2.4 @xmath20 m ) in 1996 january and in the j - band ( 1.1 1.4 @xmath20 m ) in 1996 december .
the dates and total exposure times of the observations are summarized in table [ tab_logobs ] .
individual exposure times were 20 seconds in j and h and 10 seconds in k , with telescope nods between integrations to sample the variable sky background .
after subtracting sky and dark current , we calibrated the images using dome flats ( constructed from images taken with the dome lights off subtracted from images taken with the lights on ) .
we aligned the individual frames and coadded them in groups of four , from which we extracted the instrumental magnitudes of a062000 and three nearby comparison stars .
the positions of the comparison stars relative to a062000 and their nir colors are given in table [ tab_stars ] .
the two brighter comparison stars were averaged and used to correct the observations for seeing and extinction variations .
the third , fainter star is comparable in brightness to a062000 in the near - infrared .
we used the scatter in its measurements to assign an uncertainty to the relative photometry in each filter on each night .
we flux calibrated the h and k data using observations of four standard stars from 1996 january 27 @xcite .
the j - band observations were flux calibrated using a single standard star observed on 1996 december 8 ; the uncertainty in the j - band flux calibration was estimated by using the same standard star to calculate a h - band calibration and comparing that result to the calibration from 1996 january 27 .
because of the paucity of k - band observations of the a062000 field , the transformation equations for each filter were fit with a fixed extinction coefficient @xcite and no color terms .
the mean colors for each observing run are shown in table [ tab_colors ] .
the mean h magnitude of the a062000 lightcurve was stable over the one - year baseline of our observations , changing by less than 0.05 mag . in the final step , we converted the data from magnitudes to fluxes @xcite .
the observations from each run were combined into mean light curves using the linear ephemeris of @xcite . the combined fluxes and error bars were determined from weighted means of the data , where the weights were based on the uncertainties assigned each night from the scatter about the mean for the third field star .
the bin sizes in the combined light curves are 0.01 in orbital phase for j and h and 0.05 for k , corresponding to time intervals of 4.65 min and 23.3 min , respectively .
the orbital phases conform to the standard convention : phase 0 corresponds to inferior conjunction of the k star there is no substantial drift in the phasing of the light curve relative to the 1986 ephemeris ( nor with respect to the refined ephemeris of orosz et al .
1994 ; the two orbital solutions differ by an amount smaller than the bin size of our points ) : the light curve minima occur at phases 0 and 0.5 , and the maxima at phases 0.25 and 0.75 .
the h - band light curves of a062000 are shown in figure [ fig_a0620h ] , and the observations in j from 1996 december and in k from 1996 january are shown in figure [ fig_a0620jk ] .
the light curves all show an asymmetric , double - humped modulation .
there were small fluctuations over the orbit from one observation to the next , but there was no gross variability in the shape or overall flux of the h light curve over the one - year course of our observations .
the minimum at @xmath21 = 0.5 is deeper than the primary minimum at @xmath21 = 0 and the peak at @xmath21 = 0.75 higher than the peak at @xmath21 = 0.25 in all of the h - band light curves , and the same appears to be true in the ( noisier ) k light curve .
the j - band light curve is too poorly sampled to determine the relative amplitudes of the peaks and troughs .
the peaks and troughs in the data are smooth , showing no evidence of the sharp dips seen in b light curves circa 19861989 @xcite .
the shape of the light curves is very similar to that of a k - band light curve obtained in 1990 by @xcite .
the phasing of the light curve peaks and troughs and the sense of the asymmetry in the peaks is the same in both observations .
our mean k and ( j k ) colors are also close to the colors they found .
the shapes of the near - infrared light curves of a062000 are generally consistent with ellipsoidal modulation in a system of moderate to high inclination , but there are two clear deviations from a pure ellipsoidal modulation . first , there are small fluctuations ( @xmath220.05 0.1 mjy ) among the three observations in the amplitudes of the peaks relative to the troughs , indicating a variable source of flux in addition to the ellipsoidal modulation .
the second deviation is the unequal maxima , which an ellipsoidal variation can not produce .
the bright spot is a plausible source for this asymmetry .
it is seen in doppler tomograms of the accretion disk in a062000 @xcite and would be expected to boost the peak at @xmath21 = 0.75 relative to the @xmath21 = 0.25 peak , as we observe .
superhumps from a precessing , non - circular disk or star spots on the k star could also cause asymmetries but , as we will show , they are less plausible sources for the asymmetries in our light curves .
we modeled the near - infrared light curves of a062000 with a rewritten , updated version of the light curve synthesis code described in @xcite .
the code simulates the light curves of binary systems and includes the equipotential geometry , limb and gravity darkening for the stars , and a limb - darkened , flared accretion disk with a bright spot on its edge and top surface .
a significant improvement to the code is the use of specific intensities and quadratic limb darkening coefficients for cool stars obtained from fits to model stellar atmospheres of late - type stars ( allard & hauschildt 1995 ; the fits to the model atmospheres are presented in froning 1999b ) .
we fit three sets of models to the a062000 light curves . in the first set we assumed that only the k star contributes to the observed flux . for the second set we added an accretion disk with a bright spot on its edge .
the third set was a series of models to estimate the effect of a non - varying source of diluting flux on the derived orbital inclinations .
we modeled the h light curves only ; the j and k data are of lower quality and do not warrant detailed modeling .
we initially assumed that the k star is the only source of the observed flux .
we assumed that the asymmetry in the observed light curve is caused by extra flux added to the peak at @xmath21 = 0.75 and , to avoid this extra flux , we fit the models to the light curves only over phases from @xmath21 = 0 to 0.51 .
we generated model light curves of the k star , varying each of the parameters that affect the shape of the ellipsoidal modulation in turn , and fit the light curves to the observed data by least squares .
the parameters that affect this model are : the orbital inclination , @xmath23 , the mass ratio , @xmath13 , the temperature of the k star at the pole , @xmath24 , the quadratic limb - darkening coefficients and the gravity darkening coefficient , @xmath25 .
we varied the inclination from @xmath23 = 1@xmath16 89@xmath16 in 1@xmath16 steps and the k star temperature from @xmath24 = 4000 4500 k in increments of 100 k. the mass ratio in a062000 is known to be @xmath13 = 0.067 @xcite .
we calculated models for this mass ratio but also for @xmath13 = 0.056 , 0.083 and 0.10 to check the dependence of our results on mass ratio .
the limb darkening was modeled using quadratic limb darkening coefficients obtained from the model stellar atmospheres discussed above .
we calculated models for gravity darkening coefficients of 0.05 and 0.08 @xcite and we assumed that the k star fills its roche lobe .
figure [ fig_mod1 ] shows the best - fit models for each of the h - band light curves .
the parameters for the models are given in table [ tab_fits ] .
for the 1996 december light curve , the best - fit model for @xmath13 = 0.067 has an inclination of @xmath23 = 44@xmath16 ; for the 1996 january light curve , @xmath26 gives the best fit ; and for the 1995 december light curve , models with 43@xmath16
45@xmath16 give good fits .
the models are largely insensitive to variations in the mass ratio and are only weakly dependent on the temperature of the k star and the value of the gravity darkening coefficient .
the best fits were typically obtained for @xmath24 = 4100 k and @xmath25 = 0.08 .
more important , changes in these parameters had virtually no effect on the value of the best - fit inclination . for the 1996 january light curve , for example , the reduced @xmath27 of the best fits for each combination of the other parameters tested range from @xmath28 = 1.06 1.58 , but the inclinations for these models vary only from @xmath23 = 38@xmath16 41@xmath16 . similarly , the models calculated for the 1996 december light curve had @xmath28 between 2.62 and 3.01 and inclinations from @xmath23 = 45@xmath16 50@xmath16 , while for the 1995 december light curve , @xmath28 = 1.13 1.18 and @xmath23 = 43@xmath16 48@xmath16 .
if the k star is the sole contributor to the near - infrared emission from @xmath21 = 0 0.51 , then the orbital inclination in a062000 is @xmath23 = 38@xmath16 45@xmath16 ( @xmath23 = 38@xmath16 50@xmath16 for the extreme range of parameter values ) .
this range is consistent with the results of @xcite , who found a best fit to their k light curve of @xmath23 = 37@xmath16 for @xmath13 = 0.067 and a 90% confidence interval of @xmath29 30@xmath16
45@xmath16 for that mass ratio ( their fig . 2 ) .
the assumption that the modulation of the light curve at these orbital phases is purely ellipsoidal is not correct , however .
the amplitude of the @xmath21 = 0.25 peak relative to the light curve minima is smaller in the 1996 january light curve than in 1995 december or 1996 december light curves .
this leads to a lower value for the orbital inclination based on the 1996 january data ( @xmath26 ) and a range of possible inclinations ( 38@xmath16 41@xmath16 ) that does not overlap those determined from the 1995/1996 december light curves ( @xmath23 = 43@xmath16 50@xmath16 ) .
this change in the amplitude of the modulation even at @xmath30 indicates that there is some source of variable contamination of the ellipsoidal modulation .
it also gives a measure of the uncertainty in any determination of the inclination based on just one observation epoch : the inclinations derived from our three light curves have a range of 7@xmath16 . in summary ,
models including only the effect of ellipsoidal variations of the k star give a lower limit to the inclination of @xmath0 and show that there is a variable distortion of the ellipsoidal variations even at @xmath30 .
we next modeled the light curves over the full binary orbit , adding an accretion disk and a bright spot to the k star .
we have complete orbital coverage only in the h band , so we can not constrain system parameters such as the temperature of the accretion disk and bright spot with our models .
rather , our goal was to find simple models with reasonable parameter values that fit the observed light curves in order to determine the range of possible values for the orbital inclination . since the derived orbital inclinations do not depend strongly on the mass ratio , the temperature of the k star , nor the gravity darkening coefficient , we fixed their values at @xmath13 = 0.067 , @xmath24 = 4100 k and @xmath25 = 0.08 .
the inner radius of the accretion disk was fixed at 0.001 r@xmath31 and the outer radius at 0.5 r@xmath31 , which is the outer radius in visible light found by @xcite . to allow for a beamed bright spot on the disk rim , we set the flare half - angle of the disk to a small but non - zero value of 1@xmath32 .
the bright spot also extends onto the top surface of the disk from 0.45 0.5 r@xmath31 @xcite . since the bright spot is not eclipsed and since the disk has only a small flare , the spot component on the top of the disk merely adds to the constant disk flux and is otherwise irrelevant .
the bright spot and accretion disk were each assumed to emit as single - temperature blackbodies , with the linear limb - darkening coefficients for the accretion disk obtained from @xcite .
previous observations of a062000 in quiescence have found no evidence for significant irradiation of the k star by the disk , so we did not include irradiation in our models .
the parameters we varied were : the orbital inclination , @xmath23 , the temperature of the accretion disk , @xmath33 , the temperature of the bright spot , @xmath34 , the azimuthal position of the spot on the disk , @xmath35 and the azimuthal full width of the spot , @xmath36 .
we varied the inclination from @xmath23 = 1@xmath16 89@xmath16 , t@xmath37 from 2000 k
5000 k , t@xmath38 from 5000 k
25,000 k , @xmath35 from 80@xmath16 115@xmath16 ( measured with respect to the line connecting the centers of the stars and increasing in the direction of the orbital motion ) and @xmath36 from 5@xmath16 15@xmath16 .
figure [ fig_mod2 ] shows the h light curves and the models with the lowest @xmath27 for the parameter values we tested .
the output model parameters are given in table [ tab_fits ] .
the fits demonstrate that simple models incorporating ellipsoidal variations from the k star and a two - component constant flux plus a beamed bright spot accretion disk can fully account for the near - infrared light curves of a062000 .
we reiterate that the models summarized above are not intended to provide real constraints on the properties of the accretion disk ( the temperature of the bright spot , for example , is dependent on its assumed size and on the assumed temperatures of the k star and the accretion disk ) .
moreover , the fits presented above are definitely not unique .
virtually any model with @xmath39 will fit the light curves equally well if @xmath33 is increased with inclination .
for example , a model with @xmath40 and @xmath33 = 2000 k fits the 1996 january light curve as well as the model with @xmath41 and @xmath33 = 5000 k shown in figure [ fig_mod2 ] .
our light curves show no evidence of an eclipse , which does set an upper limit to the inclination . for the assumed disk size ,
@xmath42 = 0.5 r@xmath31 , we found that inclinations above 75@xmath16 introduced eclipse features in the models inconsistent with the data .
marsh , robinson & wood ( 1994 ) found a similar limit , @xmath43 , based on the lack of observed rotational disturbance in the h@xmath44 emission line .
our agreement with their upper limit is reassuring but not unexpected , as we used their values for the mass ratio and accretion disk radius in our models .
the sense of the asymmetry of the h - band light curve and the orbital phasing of the light curve peaks was stable over the one year covered by our observations .
this long - term stability argues against a precessing , non - circular disk or star spots as sources of the asymmetry in the light curve peaks , as both are likely to produce a variable asymmetry .
when the asymmetry is modeled by a bright spot on the edge of the disk , both the phase and the extent of the bright spot are the same in all three h - band light curves and , with less confidence , in the k - band light curve : @xmath45 and @xmath46 .
these results strongly support identifying the bright spot as the source of the extra flux at @xmath21 = 0.75 .
the location of the bright spot in our models is not the same as the location of the spot in the doppler maps of @xcite , who found @xmath47 ( their figure 7 ) .
the position of a bright spot as seen in optical line emission will not necessarily coincide with the location of peak spot emission in the near - infrared ( see , e.g. , littlefair et al.2000 , froning et al .
1999 ) ; the difference between the two suggests that the near - infrared bright spot emission in a062000 originates downstream from the initial mass stream impact point .
since there is a direct tradeoff between extra constant flux from the accretion disk ( or any other source ) and the orbital inclination inferred from the amplitude of the ellipsoidal variations , we ran a final set of models to determine how the inclination changes as disk flux is added to the light curve . to do this
, we fit a k star model ( t@xmath48 = 4100 k , @xmath25 = 0.08 ) to the 1996 january data and added increasing amounts of disk flux , specified in the light curve synthesis program as a fraction of the total light curve flux at @xmath21 = 0.25 .
table [ tab_third ] shows the derived orbital inclination as the disk contribution increases .
for no disk contribution , we recover the 38@xmath16 inclination found in section [ sec_star ] .
to allow orbital inclinations approaching 75@xmath16 the upper limit set by the lack of eclipses the disk would need to be contribute 55% of the h - band flux .
table [ tab_third ] also shows that if the accretion disk contribution in the h - band is low , large fluctuations in the disk flux are needed to explain the change in the amplitude of the @xmath21 = 0.25 peak discussed in section [ sec_star ] : the 67@xmath16 difference in the inclinations of our secondary star model fits to the three h light curves implies fluctuations of 2025% in the contaminating flux at the @xmath21 = 0.25 peak .
based on our fits to the h - band light curves , we can constrain the inclination in a062000 to @xmath49 . the upper limit on the inclination
is quite strict , as light curve models with @xmath50 show both primary and secondary eclipse features not seen in the data
. the lower limit on the inclination is also a fairly strict limit .
the amplitude of the light curve modulation was larger in 1996 december and 1995 december than in 1996 january , indicating that the 38@xmath16 lower limit on the inclination derived from the latter is a probable underestimate of the true binary inclination caused by dilution of the ellipsoidal modulation by an accretion disk flux component . from determinations of the orbital period , mass ratio and radial velocity semi - amplitude of the k star , @xcite derived a mass @xmath11 @xmath3 for the compact star in a0620 - 00 .
combined with our limits on the inclination , this limits the mass of the compact star to lie in the range @xmath51 @xmath3 .
the lower limit is close to , but remains larger than , the maximum mass of a uniformly rotating neutron star with the stiffest equation of state , @xmath523.2 @xmath3 @xcite , and it is well above the maximum mass of a nonrotating neutron star , 1.8 2.5 @xmath3 @xcite .
@xcite identified a sharp dip in their ubvr light curves of a0620 - 00 as an eclipse of the k star by a large accretion disk ( at or near the maximum radius for a circular disk inside the roche lobe around the compact star ) .
if this identification is correct , the inclination of a0620 - 00 is @xmath53 and the mass of the compact star is @xmath54 @xmath3(using the inclination range found by haswell et al .
1993 for @xmath13 = 0.067 and masses from the marsh , robinson & wood 1994 equation for @xmath18 given above ) . @xcite also invoked a large accretion disk to model 1986 spectroscopic observations of a062000 .
@xcite explained the large disk and the grazing eclipse in the context of a non - circular , precessing accretion disk model for a062000 : as the orientation of the non - circular disk changes over the long ( @xmath55 ) precession period , eclipses appear and disappear and superhumps move through the light curve , causing changes in the shape of the light curves as was , indeed , seen from 1981 1989 in a062000 .
@xcite found that a062000 exhibited slow fluctuations in its mean optical brightness between 1991 and 1995 with a peak - to - peak amplitude of 0.3 0.4 mag in the r band .
they did not find any periodicities in the fluctuations .
specifically , they found no evidence of a superhump period nor of a beat period between the superhump and orbital periods .
the shape of the orbital light curve varied somewhat with brightness , but the sense of the asymmetry between the two peaks never reversed .
there is , therefore , no evidence for current or recent superhumps in a062000 .
@xcite concluded that , after 1989 the accretion disk had shrunk below the radius necessary to trigger the tidal interactions that drive superhumps .
the most serious challenge to the superhump model comes from @xcite , who re - analyzed the 1986 observations of @xcite and concluded that both the 1986 and the 1991/92 data sets which straddle the observations of @xcite are consistent with a smaller accretion disk , @xmath42 = 0.5 0.6 @xmath56 , one too small to drive superhumps .
they note , however , that their accretion disk radius is based on the radius of h@xmath44 emission and could be smaller than the radius of the optically thick accretion disk used by @xcite .
@xcite also narrowly constrained the derived inclination for a062000 to @xmath17 , corresponding to @xmath57 @xmath3 .
their best - fit value for @xmath13 = 0.067 was @xmath23 = 37@xmath16 , which corresponds to @xmath18 = 14.2 @xmath3 .
their analysis was predicated on the assumption that the accretion disk does not contaminate the k - band flux from the k star . to test this
, @xcite obtained a k - band spectrum of a062000 to which they fit scaled template spectra of stars of known spectral type . from this , they concluded that the k star provides 75@xmath5817% of the k - band flux .
they noted that a 27% accretion disk contribution ( their maximum likely disk fraction ) would increase the minimum inclination in a062000 by 7@xmath16 and decrease the mass of the compact star by 3.6 @xmath3 .
if the accretion disk does contribute @xmath5225% of the near - infrared flux , our three observations would indicate an inclination of @xmath59 53@xmath16 and a corresponding mass of @xmath60 @xmath3 . to determine the contribution of the disk to the infrared spectrum , @xcite fit the template spectra to just the @xmath61 m @xmath4co(2,0 ) bandhead .
the use of this feature to estimate the disk contribution is problematic for several reasons .
first , the co molecular line strengths are both temperature and gravity dependent ( e.g. , kleinman & hall 1986 ) .
while the spectrum of the k star in a062000 is clearly inconsistent with that of a giant star ( luminosity class iii ) , its effective gravity could still be significantly lower than the gravity of a main - sequence star @xcite .
@xcite used only dwarf stars for their template spectra , so their results could not distinguish gravity dependent changes in the @xmath62 m @xmath4co(2,0 ) line strength .
second , observations of cataclysmic variables have shown that in some systems , @xmath4co absorption is weaker relative to the strengths of the atomic absorption lines than expected for their donor star spectral types @xcite .
in addition , the ratio of the @xmath4co bandhead equivalent width to those of the atomic lines can vary with time @xcite .
the reason for these abnormalities in the co absorption line strengths is not yet understood .
thus , using the strength of the @xmath62 m @xmath4co(2,0 ) bandhead to measure the contribution of the disk to the infrared flux gives unreliable results .
the measurements by @xcite do not settle the question of how much flux the disk contributes to the light curve of a0620 - 00 at infrared wavelengths .
the orbital inclination could , therefore , be significantly higher than the lower limit set by the ellipsoidal variations ( @xmath39 ) and the mass of the compact star much less than the upper limit ( @xmath63 @xmath3 ) .
\1 . the infrared light curves of the black hole binary a062000 show an asymmetric , double - humped modulation with extra emission in the peak at phase @xmath21 = 0.75 .
there were no gross changes in the morphology of the light curves over a one year period from 1995 december to 1996 december .
the mean infrared colors and the shape of the light curve are also the same as observed in 1990 january .
there were no sharp dips in the light curve nor reversals of the asymmetry between the two humps as were seen in observations circa 1986 1989 .
the light curves are consistent with ellipsoidal variations from the k star plus beamed flux from a spot on the accretion disk .
a precessing disk is an unlikely source of light curve modulation during the observation period .
star spots are also ruled out unless the spot is fixed in location and size on the k star surface .
3 . based on fits to the lower peak in the light curve ( between @xmath21 = 0 and 0.51 ) , the minimum inclination in a062000 is @xmath64 . from the absence of eclipse features in the light curves , models including an accretion disk of radius @xmath65 give an upper limit to the inclination of @xmath1 .
the mass of the compact star in a062000 is @xmath2 @xmath3 .
ellipsoidal variations provide only a lower limit to the inclination of non - eclipsing binary systems when the contribution of the accretion disk is unknown . a previous attempt to determine the relative contribution of the accretion disk and k star to the near - infrared flux used the strength of the @xmath62 m @xmath4co(2,0 )
this method is not reliable .
lcc 1995 dec 15 & h & 3.72 + 1995 dec 16 & h & 3.46 + 1995 dec 17 & h & 1.92 + 1996 jan 27 & h , k & 3.60 , 0.25 + 1996 jan 28 & h & 0.70 + 1996 jan 29 & h , k & 4.40 , 0.35 + 1996 dec 7 & h & 2.96 + 1996 dec 8 & j , h & 1.16 , 0.72 + 1996 dec 9 & j , h & 1.34 , 1.48 + 1996 dec 10 & j , h & 1.16 , 1.40 + 1996 dec 12 & j , h & 0.20 , 3.56 + ccccc star 1 & 18@xmath66n , 18@xmath66w & [email protected] & [email protected] & [email protected] + star 2 & 19@xmath66n , 5@xmath66w & [email protected] & [email protected] & [email protected] + star 3 & 37@xmath66n , 14@xmath66e & [email protected] & [email protected] & [email protected] + lcccccc 1996 december & 2.62 & 44@xmath16 & & & & + 1996 january & 1.06 & 38@xmath16 & & & & + 1995 december & 1.13 & 43@xmath16 45@xmath16 & & & & + 1996 december & 1.88 & 74@xmath16 & 5000 k & 25,000 k & 100@xmath16 & 5@xmath16 + 1996 january & 0.83 & 70@xmath16 & 5000 k & 20,000 k & 100@xmath16 & 10@xmath16 + 1995 december & 1.32 & 53@xmath16 & 2000 k & 25,000 k & 115@xmath16 & 10@xmath16 + | we measured the near - infrared orbital light curve of the black hole binary a062000 in 1995 and 1996 .
the light curves show an asymmetric , double - humped modulation with extra emission in the peak at orbital phase 0.75 .
there were no significant changes in the shape of the light curve over the one - year observation period .
there were no sharp dips in the light curves nor reversals of the asymmetry between the two peaks as seen in earlier observations .
the light curves are well fit by models incorporating ellipsoidal variations from the mass - losing k - type star plus a beamed bright spot on the accretion disk around the compact star .
the long - term stability of the light curve shape rules out superhumps and star spots as sources of asymmetry when we observed a062000 .
the ellipsoidal variations yield a lower limit @xmath0 on the orbital inclination .
the light curves show no eclipse features , which places an upper limit @xmath1 .
this range of inclinations constrains the mass of the compact object to @xmath2 @xmath3 .
the light curves do not further constrain the orbital inclination because the contribution of the accretion disk to the observed flux is unknown .
we argue that a previous attempt to measure the near - infrared flux from the accretion disk using the dilution of the @xmath4co(2,0 ) bandhead in the spectrum of the k star is not reliable because the band strength depends strongly on surface gravity . |
differential geometric ideas were used in graph coloring problems early on .
wernicke @xcite promoted already gauss - bonnet type ideas to the four color problem and used what we today call curvature to graph theory . the discharging method introduced by heesch @xcite
was eventually used by appel and haken @xcite to prove the four color theorem .
this method is conceptionally related to geometric evolution methods : deform the geometry until a situation is reached which can be understood by classification .
birkhoff @xcite and later tutte attached polynomials to graphs which relate modern topological invariants like the jones polynomial .
cayley already geometrized the problem by completing and modifying planar graphs to make them what we would call cubic so that a planar graph looks like discretization of a two dimensional sphere which by gauss bonnet has euler characteristic @xmath11 forcing the existence of vertices with positive curvature .
studying the positive curvature parts is essential in the proof of the four color theorem for planar graphs @xcite .
+ probabilistic methods in geometry are part of integral geometry as pioneered by mathematicians like crofton , blaschke and santal @xcite .
this subject is known also under the name geometric probability theory and has been used extensively in differential geometry .
for example by chern in the form of kinematic formulae @xcite or milnor in proving total curvature estimates @xcite using the fact that curvature @xmath8 of a space curve is an average of indices @xmath5 ( at minima of @xmath3 on the curve ) using a probability space of linear functions @xmath3 .
banchoff used integral geometric methods in @xcite and got analogue results for polyhedra and surfaces .
we have obtained similar results in @xcite for graphs .
integral geometry is also used in other areas and is an applied topic : the cauchy - crofton formula for example expresses the length of a curve as the average number of intersection with a random line . the inverse problem to reconstruct the curve from the number of intersections in each direction is a radon transform problem and part of tomography , a concept which has also been studied in the concept of graph theory already @xcite .
+ in this note we look at an integral geometric topic when coloring finite simple graphs @xmath0 . in full generality , the scalar curvature function ( [ curvature ] ) satisfies the discrete gauss - bonnet formula @xmath9 . in @xcite ,
the focus had been on even dimensional geometric graphs for which @xmath8 is a discretized euler curvature , a pfaffian of the curvature tensor ( @xcite chapter 12 ) . for odd - dimensional geometric graphs ,
it was shown in @xcite using integral geometric methods that the curvature is always constant zero @xcite , a fact already pioneered in piecewise linear polytop situations @xcite .
given a function @xmath12 on a vertex set @xmath13 of a finite simple graph , let @xmath14 denote the subgraph generated by @xmath15 , where @xmath16 generates a graph called the unit sphere of the vertex @xmath17 . in @xcite , we averaged the index @xmath18 over a probability space functions @xmath19 $ ] in such a way that for a fixed vertex , the random variables @xmath20 are all independent and uniformly distribution in @xmath21 $ ] .
we showed that its expectation is the curvature @xmath22 where @xmath23 is the number of @xmath24 subgraphs in the sphere @xmath25 at a vertex @xmath17 and @xmath26 .
+ we know that @xmath27 if @xmath3 is injective .
this is poincar - hopf @xcite .
it can be proven by induction : add a vertex @xmath17 to a graph and extend @xmath3 so that the new point is the maximum of @xmath3 .
this increases the euler characteristic by @xmath28 using the general formula @xmath29 used in the case where @xmath30 is the graph without the new point @xmath31 is the unit ball of the new point @xmath17 and @xmath32 is @xmath14 .
the expectation of poincar - hopf recovers then the gauss - bonnet theorem @xcite @xmath33 where @xmath34 is the euler characteristic of the graph defined as the super count of the number @xmath35 of @xmath24 subgraphs of @xmath4 .
gauss bonnet also immediately follows from the generalized handshaking lemma @xmath36 while @xcite established poincar - hopf only for injective functions , the same induction step also works for locally injective functions : assume it is true for all graphs with @xmath37 vertices and functions @xmath3 which are locally injective , take a graph with @xmath38 vertices and chose a vertex which is a local maximum of @xmath3 at @xmath17 then remove this vertex together with the connections to @xmath39 .
again this reduces the euler characteristic by @xmath40 .
+ the integral geometric approach to gauss - bonnet illustrates an insight attributed in @xcite to gelfand and his school : _ the main trick of classical integral geometry is the change of order of summation"_. in our case , one summation happens over the vertex set @xmath13 of the graph , the other integration is performed over the set of functions @xmath3 with respect to some measure .
what can be proven in a few lines for general finite simple graphs can also be done for general compact riemannian manifolds .
the integral geometric approach provides intuition what euler curvature is : it is an expectation of the quantized index values or divisors given as poincar - hopf indices which incidentally play a pivotal role also in baker - norine theory @xcite and especially in higher dimensional versions of this riemann - roch theorem . that also in the continuum , curvature is the expectation of index density is intuitive already in the continuum for manifolds @xmath41 : negative index values are more likely to occur at places with negative curvature ; there is still much to explore however .
we need to investigate in particular what happens if we take the probability space @xmath41 itself and chose for every @xmath42 and some fixed time @xmath43 the heat signature function @xmath44_{xy}$ ] , if @xmath45 is the hodge laplacian of the riemannian manifold .
the question is then how the expectation @xmath46 $ ] of the index density of these functions is related to the classical euler curvature @xcite .
this is interesting because it would allow us to work with a small ( finite dimensional ) probability space .
the smallness of the probability space leads us to the topic covered here . + we look at one of the smallest possible probability spaces for which an index averaging result can hold for graphs .
it turns out that we do not need injectivity of the functions @xmath3 at all .
all we need is local injectivity . in other words
, we can restrict the probability space of functions @xmath3 to * colorings*. this reduces the size of the probability space considerably similarly as in the continuum , the choice of linear functions reduced the probability space from an infinite dimensional class of morse functions to a finite dimensional space of linear functions induced onto the manifold from an ambient euclidean space . while poincar - hopf and the index averaging result hold also for all @xmath47 injective functions
, the probability space of @xmath1 colorings has considerably less elements .
if @xmath2 is the chromatic polynomial of the graph introduced in 1912 by birkhoff in the form of a determinant @xcite , then the probability space has @xmath10 elements .
+ for the octahedron for example , a geometric two - dimensional graph @xmath4 with euler characteristic @xmath48 and chromatic number @xmath49 , the probability space of colorings has @xmath50 elements only .
we can see this in figure ( [ fig5 ] ) .
[ mainresult ] let @xmath51 denote the finite set of all graph colorings of @xmath4 with @xmath1 colors , where @xmath1 is bigger or equal than the chromatic number .
let @xmath52 be the expectation with respect to the uniform counting measure on @xmath53 . then @xmath54 = k(x )
\ ; .\ ] ] the proof mirrors the argument for continuous random variables @xcite .
it is simpler as we do have random variables with a discrete distribution .
let @xmath55 denote the number of @xmath56-dimensional simplices in @xmath25 and @xmath57 the number of @xmath56-dimensional simplices in @xmath14 .
given a vertex @xmath17 , the event @xmath58 has probability @xmath59 .
the reason is that the symmetric group of color permutations acts as measure preserving automorphims on the probability space of functions implying that for any @xmath3 which is in @xmath30 there are @xmath60 functions which are in the complement implying that @xmath30 has probability @xmath59 .
this implies @xmath61 = \frac{v_k(x)}{(k+2 ) } \;\ ] ] which is the same identity we also had in with the continuum probability space .
therefore , @xmath62 & = & 1-\sum_{k=0}^{\infty } ( -1)^k { { \rm e}}[v_k^-(x ) ] = 1+\sum_{k=1}^{\infty } ( -1)^k { { \rm e}}[v_{k-1}^-(x ) ] \\ & = & 1+\sum_{k=1}^{\infty } ( -1)^k \frac{v_{k-1}(x)}{(k+1 ) } = \sum_{k=0}^{\infty } ( -1)^k \frac{v_{k-1}(x)}{(k+1 ) } = k(x ) \ ; .\end{aligned}\ ] ] for trees for example where @xmath63 , there are only @xmath64 colorings . at the leaves of the tree
, we have either the index @xmath65 or @xmath66 , where minima @xmath17 always have the index value @xmath67 .
the average index is therefore @xmath68 at the leaves . on an interior vertex @xmath17 with index @xmath69 we have either index @xmath70 if the vertex is a minimum or @xmath71 if it is a maximum .
the curvature there is then @xmath72 .
more examples can be seen in the figures .
\a ) we can think of a graph coloring with @xmath1 colors as a gauge field using the finite cyclic group @xmath73 .
permuting colors is a symmetry or gauge transformation . given a coloring @xmath3 and a fix of direction for every edge
, we can look at the gradient @xmath74 , which is now just realized as a finite @xmath75 matrix from the module @xmath76 to @xmath77 . with the transpose @xmath78 which is a discrete divergence , the laplacian @xmath79 is independent of the directions chosen on the edges .
when looking at such colorings , we get a probability space of gradient vector fields .
if @xmath1 is prime , we could look at the eigenvalues of @xmath80 in an extension field of the galois field @xmath81 .
looking at the situation where @xmath1 is the chromatic number of @xmath4 is natural .
one can now ask spectral inverse problems with respect to such color spectra " or look at random walk defined by the laplacian @xmath80 .
dynamical systems on finite group - valued gauge fields are special cellular automata @xcite .
especially interesting could be to study the discrete markoff process @xmath82 , which is now a cellular automaton .
+ b ) an interesting problem is to characterize graphs with minimal chromatic number in the sense that the * chromatic richness * @xmath83 is equal to @xmath66 .
graphs which satisfy @xmath84 can be colored in a unique way modulo color permutations .
we call them * chromatically poor graphs*. examples are trees ( with @xmath63 ) , the octahedron ( with @xmath85 , the complete graphs @xmath86 ( with @xmath87 ) , wheel graphs @xmath88 with an even number of spikes ( with @xmath89 ) or cyclic graphs @xmath90 or more generally any bipartite graph with @xmath63 . since for cyclic graphs @xmath91 , or wheel graphs
@xmath92 we have @xmath93 , the richness of a graph is expected to be exponentially large in @xmath94 in general .
we have done some statistics with the richness function on erds - renyi probability spaces of graphs .
it appears that a limiting distribution could appear .
but this is very preliminary as experiments with larger @xmath37 are costly . for simply connected geometric graphs , we only need satisfy a local degree condition attributed to heawood ( see theorem 2.5 in @xcite or theorem 9.5 in @xcite ) .
related is the conjecture of steinberg from 1975 which claims that every planar graph without 4 and 5 cycles must be 3 colorable .
more generally , erds asked for which @xmath56 the exclusion of cycles of length @xmath95 renders the chromatic number @xmath96 .
the record is @xmath97 @xcite .
while graphs without cycles of length larger than @xmath98 have @xmath99 already if two triangles are present , such a graph can not be chromatically poor any more .
+ c ) what does the standard deviation of the random variable @xmath100 tell about the graph ? more generally ,
what is the meaning of the moments @xmath101 $ ] ?
like curvature @xmath8 , these are scalar functions on vertices .
the standard deviation is a scalar field which depends on the number of colorings . for the wheel graph @xmath102 for example
which has chromatic number @xmath98 and chromatic polynomial @xmath103 , we measure the standard deviations + @xmath104 , @xmath105 , + @xmath106 , @xmath107 , + @xmath108 , @xmath109 .
+ d ) a general problem in differential geometry is to determine to which extent curvature determines geometry @xcite .
one can now ask for graphs , to what extend the moments @xmath110 determine the graph up to isomorphism . with the index functions @xmath5 one has even more data and
one can ask to which extent the sequence of numbers @xmath111 determines the graph @xmath4 up to isomorphism .
of course , we have @xmath112 already by poincar - hopf . also interesting
are the moments @xmath113 for which we just have established @xmath114 .
one can ask the inverse problem of finding the geometry from the moments @xmath115 $ ] in any geometric setup . while for plane curves , @xmath116 alone determines the curve up to isometry , for space curves ( where total curvature is nonnegative ) , this is already no more the case . the analogue of total curvature would be @xmath117 $ ] which can be expressed with finitely many of the above moments for any graph . still , it is conceivable that the distribution of the moment functions @xmath118 is sufficient to reconstruct the graph up to isomorphism .
+ e ) index averaging allows to see that curvature @xmath8 is identically zero for odd - dimensional geometric graphs @xcite , finite simple graphs for which the unit spheres are discrete spheres , geometric @xmath119 dimensional graphs which become contractible after removing one vertex .
the key insight from @xcite is that the symmetric index @xmath120 is related to the euler characteristic of a geometric @xmath121-dimensional graph . establishing @xmath122 to be zero for all @xmath17 implies curvature @xmath8 is zero .
the same holds for riemannian manifolds @xmath41 where it leads to the insight that euler characteristic @xmath123 for even dimensional manifolds is an average of two dimensional sectional curvatures as @xmath123 is also an average of @xmath124 where @xmath125 runs over a probability space of two - dimensional submanifolds @xmath125 of @xmath41 . as in the case of graphs , a probability measure @xmath126 on the set of scalar fields provides us with with curvature and defines geometry . in physics
, one would select a measure @xmath126 which is invariant under the wave equation and refer to the krylov - bogolyubov theorem for existence .
the same can be done using the finite probability space considered here : any coloring @xmath3 of a four dimensional geometric graph defines a two dimensional geometric graph .
the finite probability space of graph colorings allows computations with rather small number of elements .
the natural measure on functions is the product measure .
+ f ) integral geometry naturally bridges the discrete with the continuum .
this was demonstrated early on in the proof of the fry - milnor theorem @xcite , where a knot was approximated with polygons for which curvature is located on a finite set .
similarly , if a compact smooth riemannian manifold @xmath41 is embedded in @xmath127 and discretized by a finite graph @xmath4 , then the euler curvature of the manifold can be obtained as the expectation of indices @xmath5 where @xmath3 is in a finite dimensional probability space of linear functions in the ambient space equipped with a probability measure which is rotationally symmetric .
the same probability space of functions @xmath3 produces random index functions @xmath5 on @xmath4 . the expectation @xmath128 $ ] is then believed to converge to the classical curvature known in differential geometry .
index averaging becomes so a
monte carlo " method for curvature and choosing different probability spaces allows to deform the geometry .
the measure can be obtained readily without any tensor analysis : assume for example that the riemannian manifold @xmath41 is diffeomorphic to a sphere @xmath129 and nash embedded in an euclidean space , we get from the linear functions in the ambient space a probability measure @xmath126 on scalar functions of @xmath41 . taking the expectation of the index @xmath5 gives curvature .
now push forward the measure @xmath126 to the function space @xmath130 on @xmath131 to get a measure @xmath132 on the set of all scalar functions on @xmath131 . averaging the index function @xmath5 with respect to that measure
gives the curvature function on the standard sphere in such a way that @xmath133 is an isometry - at least in the sense that curvature agrees .
we do nt know yet that the riemannian metric can be recovered from @xmath132 but we believe it to be possible since the probability measure @xmath126 gives more than just the expectation : it provides moments or correlations and enough constraints to force the riemannian metric .
this is an important question because if we would be able to measure distances from the measure @xmath132 , geometry could be done by studying probability measures on the space of wave functions .
classical riemannian manifolds are then part of a larger space of geometries in which curvature @xmath134 is a distribution , but gauss - bonnet still holds trivially as it is just the expectation of poincar - hopf with respect to a measure .
any sequence of riemannian metrics on a space @xmath41 had now an accumulation point in this bigger arena of geometries , as unlike the space of riemannian metrics , the extended space of measures is compact . | given a finite simple graph @xmath0 with chromatic number @xmath1 and chromatic polynomial @xmath2 .
every vertex graph coloring @xmath3 of @xmath4 defines an index @xmath5 satisfying the poincar - hopf theorem @xcite @xmath6 . as a variant to the index expectation result
@xcite we prove that @xmath7 $ ] is equal to curvature @xmath8 satisfying gauss - bonnet @xmath9 @xcite , where the expectation is the average over the finite probability space containing the @xmath10 possible colorings with @xmath1 colors , for which each coloring has the same probability . |
large area redshift surveys such as sdss @xcite and gama @xcite have established a new paradigm regarding the clustering of galaxies @xcite .
the evolutionary effects imposed by this range of environments are not fully understood , largely due to the complexity of n - body interactions on star formation and gas dynamics .
however , the relatively unexplored regime of neutral hi in the context of galaxy groups has the potential to revolutionise our understanding of the mechanisms influencing how galaxies mature .
targeted observations of the hi content in nearby groups have often revealed the presence of intragroup material , for example , with the very large array ( vla e.g. * ? ? ?
* ; * ? ? ?
* ) , the parkes radio telescope ( e.g. * ? ? ?
* ) , the australia telescope compact array ( e.g. * ? ? ?
* ; * ? ? ?
* ) , and nanay @xcite . however , systematic surveys of galaxy groups are rare due to the observational time necessary to achieve the required sensitivities , even at relatively modest redshifts , over large fields of view .
the blind , wide - area arecibo legacy fast alfa ( alfalfa ) survey @xcite , combined with the sloan digital sky survey ( sdss ) dr7 group catalogue @xcite , enabled @xcite to explore the hi content of group galaxies as a function of optical group membership .
they found evidence for gas pre - processing correlated with group assembly .
meanwhile , the blind ultra deep hi survey ( budhies ) on the westerbork synthesis radio telescope ( wsrt ) , a targeted survey at @xmath4 , combines accurate hi content with morphology in environments ranging from voids to cluster centres to show evidence for both ram pressure and pre - processing as mechanisms for gas removal @xcite .
the role of compact groups , in particular , has long been advocated for understanding galaxy evolution processes : their high galaxy densities combined with relatively shallow potential wells prolongs strong gravitational interactions . @xcite
defined compact groups as having 4 or more members within a 3 magnitude range ( @xmath5 mb ) that form a physically isolated system .
the original hickson catalogue has been modified based on radial velocity information with 92 groups having at least three accordant members @xcite .
although other compact group catalogues exist , it remains the most intensively studied sample . the hi deficiency in compact group galaxies @xcite was the focus of a single - dish study of 72 hcgs by @xcite . in combination with a subset resolved by the vla , the authors proposed an evolutionary sequence where galaxies become increasingly hi deficient due to multiple tidal interactions .
the study of compact group evolution was expanded by green bank telescope ( gbt ) observations from @xcite which suggested the presence of copious diffuse , low column - density neutral hydrogen gas between the galaxies . however , even this emission is not always able to account for the hi deficiency of compact groups .
further studies suggest that the most evolved groups , which are also dominated by elliptical galaxies , begin to exhibit signs of hot intragroup gas due to gravitational heating @xcite .
however , strong hi deficiency does not guarantee the presence of intragroup x - ray emitting gas , complicating the interpretation of what mechanisms are responsible for removing the cold gas @xcite . additional evidence of evolution within compact groups has come from photometry from the _ spitzer _ space telescope : bimodality in irac color - color space , appearing to correlate with hi depletion , suggests accelerated evolution from dusty , star - forming galaxies , to gas - poor systems @xcite . detailed studies by @xcite further support a picture of accelerated evolution . evidence of shock - excited h@xmath6 in several compact group galaxies led @xcite to suggest a connection between intragroup tidal debris and rapid quenching in these groups .
hcg 44 is a relatively nearby compact group with redshift @xmath7 and an angular diameter of approximately @xmath8 .
it contains four massive galaxies : three star - forming spirals ( ngc 3185 , ngc 3187 , ngc 3190 ) , and one elliptical which is undetected in hi ( ngc 3193 ) ; and the hi cloud , @xmath9 , which lacks an apparent optical counterpart @xcite .
the galaxies are highly hi - deficient @xcite , but x - ray emission is confined to individual galaxies with no detectable diffuse x - ray emission associated with the intragroup medium @xcite , leaving the missing hi unaccounted for .
further , the lack of diffuse hot gas brings into question the importance of ram pressure stripping as a mechanism for removing gas from the large galaxies in this system .
@xcite found excited h@xmath6 in ngc 3190 , perhaps shocked from impact with the intragroup medium , but gbt observations are unable to resolve individual galaxies , and high resolution interferometry observations have not yet detected intragroup hi within the core of the compact group .
in fact , the evolutionary stage of the compact group has been debated as the hi content suggests it is more advanced , while the lack of intragroup optical light from stars , and intragroup x - rays suggest it is younger @xcite . as part of the atlas@xmath10 survey , @xcite surveyed the neutral gas content of hcg 44 and found evidence for an extended hi distribution well beyond the compact group .
deeper wsrt data , combined with reprocessed hipass data , detected the presence of a @xmath11kpc long tail to the north of hcg 44 containing @xmath12 of hi @xcite .
the authors concluded that this tail is most likely the result of tidal interactions and proposed possible scenarios .
first , the tail is the result of a high speed interaction between ngc 3162 and the compact group which occurred up to 3 gyr in the past .
ngc 3162 lies @xmath131.4 deg ( 620 kpc at the assumed distance ) northwest of hcg 44 .
second , the tail is a result of tidal stripping due to ngc 3187 interacting with the gravitational potential of the compact group , which would explain its very disturbed optical morphology .
however , they noted that deeper hi data over a larger field would benefit detailed modeling of this system . in this paper
we investigate the extended region around hcg 44 and the northern tail by conducting a deep hi mosaic with the karoo array telescope ( kat-7 ) which , due to its sensitivity and resolution , is ideal for detecting low column density hi gas @xcite .
we combine these data with the deep wsrt pointing from @xcite , and an hi map spanning the same sky coverage from the alfalfa survey to improve our overall picture of the hi distribution in the compact group , and nearby surroundings .
the paper is organized as follows . in section [ data ]
we describe the kat-7 observations , and summarize the additional data . in section [ combo ]
we describe our technique for combining interferometric images from kat-7 and wsrt , and the single dish map from alfalfa . in section [ results ]
we compare the measurements from kat-7 and alfalfa and present our results from the combined data cube . in section [ discussion ]
we discuss the possible origins and the fate of the extended hi tail .
we conclude in section [ conclusions ] .
hcg 44 and the hi tail ( @xmath14 ; @xcite ) were observed with the kat-7 telescope @xcite in two pointings : one centered on the group itself ( @xmath15 ; hereafter field 1 ) , and one centered on the northern tail discovered by @xcite ( @xmath16 ; hereafter field 2 ) .
field 1 was observed with all seven antennas in 11 sessions between 2015 jan 9 2015 may 20 for a total of 30 hours on source .
field 2 was later observed with six available antennas in 8 sessions between 2015 july 3 2015 july 13 for a total of 36.25 hours on source .
field 1 was observed during the night , whereas field 2 had primarily day time observations and was affected by solar interference which resulted in more intensive flagging during data reduction .
the kat-7 data were reduced in casa 4.4.0 @xcite using standard procedures . as a first step , we flagged all @xmath17 data to remove low level radio frequency interference ( rfi ) that is correlated when the fringe rate is equal to zero ( see @xcite for a complete explanation ) .
after flux and gain calibration , we performed continuum subtraction using a first order fit in the task uvcontsub and imaged each observation individually for quality assurance .
field 1 and field 2 achieved an rms noise of 2.7 and 2.9 mjy beam@xmath18 , respectively , over 7.7 kms@xmath18 channels , despite the differing total observing times between the fields .
we performed self - calibration on the continuum data of the target field and applied it to the spectral line data .
this generally produced a factor of 1.2 improvement in the rms noise , and up to a factor of 1.5 in individual data sets .
the final mosaic improved to 1.7 mjy beam@xmath18 , with a @xmath19 arcsec beam , thanks to self calibration and overlap between the fields .
we converted this to column density using equation 1 from @xcite : @xmath20 where @xmath21 is in mjy , @xmath22 and @xmath23 are the major and minor axes of the beam in arcseconds , and @xmath24 is the channel width in kms@xmath18 .
this corresponds to a 1 @xmath25 hi column density of @xmath26 @xmath1 channel@xmath18 .
figure [ wise ] shows the hi contours of the kat-7 mosaic overlaid on a four - color image from the _ wide - field infrared survey explorer _ ( wise ) .
we describe the method for generating these contours in section [ results ] .
@xcite observed hcg 44 with the wsrt for @xmath27 hours at a single pointing centered between the group and the hi tail ( @xmath28 . the observations achieved an rms of 0.23 mjy beam@xmath18 with a beam of @xmath29 arcsec over 8.2 kms@xmath18 channels .
hcg 44 also falls in the portion of sky observed by the alfalfa survey .
a full description of the survey strategy can be found in @xcite and @xcite but , in short , alfalfa is a two - pass , fixed - azimuth , drift scan survey covering the velocity range from @xmath30 to @xmath31 kms@xmath18 .
we obtained a data cube spanning @xmath32 to @xmath33 and @xmath34 kms@xmath18 .
the cube has an overall rms of 2.67 mjy beam@xmath18 with a @xmath35 arcsec beam and 5.15 kms@xmath18 channels .
we note that the data cube was affected by low level rfi in the declination strip that passed through the compact group itself @xcite , but we are able to take this into consideration with the weights map that is simultaneously created in the alfalfa data reduction process .
traditional methods of combining radio data between different arrays , or between arrays ( interferometers ) and single - dish measurements , may employ methods which operate in either the ( @xmath36,@xmath37)-plane or in the image plane ( @xcite and references therein ) .
the contribution of each measurement is generally weighted by the inverse square of the rms , @xmath38 . in the case of hcg 44 ,
the wsrt observations are an order of magnitude more sensitive than the kat-7 data in jy beam@xmath18 , making an attempt at traditional combination relatively fruitless . however , recognizing that the wsrt beam is more than an order of magnitude smaller in area than the kat-7 beam , @xcite demonstrate an innovative and powerful method of combining kat-7 and wsrt mosaics to reveal the presence of a low surface brightness tail in virgo galaxy , ngc 4424 .
their technique realizes the strengths of both telescopes by converting the cubes to column density units which removes the dependence of the measured rms on the beam size , and allows the image cubes to be combined with more equal weights . as a result ,
the final image of ngc 4424 has the advantage of sensitivity to large - scale structure from the short baselines of kat-7 , and high resolution from wsrt .
the full procedure is described in @xcite and we highlight the details relevant to the successful combination of our data below .
we employed the same technique to combine the interferometry data with the alfalfa maps .
the most common method of combining interferometry and single - dish observations is to `` feather '' the data sets .
this refers to performing a fourier transform ( ft ) on the individual data to the @xmath39-plane , scaling and combining them , then performing an inverse ft back to the image plane @xcite .
this is typically done because the single - dish and interferometers measure different spatial scales , and the single - dish data is being combined to get the zero - spacing measurement .
however , the short baselines of kat-7 ( 26185 m ) measure effectively the same spatial scales as arecibo ( illuminated diameter of 208 m ) , and the alfalfa drift scans have been synthesized to data cubes from which to extract hi detections .
thus , we use the alfalfa beam information to convert the alfalfa cube to column density units , and combine with the kat-7 and wsrt data in the image plane . formally , as long as the data is weighted appropriately , the methods produce comparable quality results @xcite . ) .
the black contour corresponds to the @xmath40 @xmath1 column density contours of the original wsrt observation @xcite .
the dotted line is the half power of the kat-7 mosaic .
the beams from each telescope are shown in the lower right hand corner , in order of decreasing size : kat-7 ( elongated ) , alfalfa ( round ) , smoothed wsrt .
the wsrt and alflfa data cubes we acquired do not extend as far south as the kat-7 mosaic , so ngc 3177 does not appear in this image . ] to optimize the final weighting scheme versus desired resolution , the wsrt cube from @xcite was smoothed with a @xmath41 arcsec gaussian kernel resulting in an effective restoring beam of @xmath42 arcsec and rms noise of 0.39 mjy beam@xmath18 . as further preparation , the kat-7 , wsrt , and alfalfa cubes were smoothed in velocity using a boxcar kernel of 2 , 2 , and 3 channels , respectively , and decimated in order to achieve the lowest multiple common velocity resolution in independent channels .
the kat-7 and alfalfa cubes were then regridded to match the velocity and spatial axes of the wsrt cube , so that the resulting cubes have the same pixel size and dimensions ( @xmath43 kms@xmath18 ) .
all cubes were converted to column density per unit velocity by the following equation : @xmath44 where @xmath45 is the hi intensity of pixel ( @xmath46,@xmath47 ) in mjy beam@xmath18 kms@xmath18 , and @xmath48 and @xmath49 are the major and minor axes of the restoring beam in arcseconds .
the resulting kat-7 , wsrt , and alfalfa data cubes have rms noise of @xmath50 , @xmath51 , and @xmath52 @xmath1(kms@xmath18)@xmath18 , respectively . before linear combination ,
the kat-7 and wsrt mosaics were primary beam corrected .
casa automatically generates a primary beam pattern when the visibilities are imaged , which we used to correct the kat-7 data . with only the final wsrt image cube
, we assumed the primary beam correction to be a gaussian profile with 35 arcmin width .
the wsrt data was imaged well beyond the primary beam of the telescope , and if combined linearly with the kat-7 and alfalfa data , the final image would be wsrt noise dominated at large distances from the wsrt pointing center .
thus , we only combine with wsrt where the primary beam sensitivity is at least 30% of the peak .
we experimented with combining the image cubes with increasingly complex set of weights to take into account different considerations .
most simply , the kat-7 and wsrt primary beam corrected cubes , and alfalfa cube were weighted by the square of the rms and combined such that : @xmath53 @xmath54 the resulting cube is dominated by the kat-7 observations throughout . despite the relatively large field of view of kat-7 ( approximately 1 degree at the half power beam width )
, it also falls off in sensitivity with distance from the mosaic pointing centers , while the alfalfa sensitivity is fairly flat across the field .
thus , in our final image combination , we take into account not only the relative rms noise at the most sensitive part of the images , but the change in sensitivity across the field of view due to the primary beam @xcite .
we also consider the alfalfa weights map such that : @xmath55 this cube is dominated by the kat-7 observations at the center and the alfalfa observations at the outskirts .
this also means that the effective synthesized beam changes across the field .
however , because the final beam size of kat-7 and alfalfa are similar , we do not consider it to be a significant complication to the interpretation of the final data cube .
see the beams in figure [ mom0 ] , for example .
the peak rms sensitivity of this final cube improves to @xmath56 @xmath1(kms@xmath18)@xmath18 at a velocity resolution of 16.4 kms@xmath18 .
the mean rms across the image is @xmath57 @xmath1(kms@xmath18)@xmath18 .
perhaps the greatest benefit of this combination is that the kat-7 and alfalfa cubes individually contain residual artifacts whose patterns are characteristic to the telescope , but by combining them one is better able to differentiate between noise and diffuse low - level emission . throughout the rest of the paper ,
the hi data presented are measured or calculated from the combined kat-7 , wsrt , alfalfa ( kwa ) image cube , unless otherwise specified ( equation [ finalcube ] ) . [ cols="<,^,^,^,^,^,^,^,^,^ " , ] + @xmath58ngc 3187 and ngc 3190 are blended in the kat-7 data .
masses from the literature are corrected to our assumed distance . where the total hi mass is not reported in the literature
, we add the masses from individually reported group members .
the telescope columns are listed in order of decreasing beam size .
@xmath6015 arcmin beam ; @xcite . @xmath61@xmath62
arcmin beam ; @xcite .
@xmath63@xmath64 arcsec beam ; we have subtracted hcg 44e from the total hi mass reported for hcg 44 because it is background galaxy at 4032kms@xmath18 ; @xcite .
@xmath65@xmath66 arcsec beam ; @xcite .
[ sources ]
in figure [ mom0 ] , we present the hi total intensity map derived from the combined kwa data cube .
the map was made by smoothing the cube with a @xmath67 arcmin beam , and masking regions where the emission was greater than 1.5@xmath25 in the original cube .
the mask was inspected and modified to keep regions with emission spatially contiguous over at least 2 channels , and remove peaks suspected to be residual rfi .
we took particular care to include even potential emission in the region between hcg 44 and the northern hi tail ( see section [ origins ] for further explanation ) .
the mask was similarly inspected and applied to the kat-7 data and was used to generate the contours overlaid in figure [ wise ] .
the addition of the kat-7 and alfalfa observations shows that the hi tail extends further than previously detected with wsrt .
based on low signal - to - noise hipass images , @xcite found that the tail stretched 300 kpc in projection . at the assumed distance of 25 mpc ( see section [ dists ] ) , we find the extended tail spans nearly 450 kpc ( 62 arcmin ) in projection and is arguably resolved into seven to eight clumps .
the extent is measured from the southern extension ( cloud 1 ) , through the brightest part of the tail , to the northwestern cloud 8 .
the physical length of the tail may be even longer than what is seen in projection .
our images reveal the breadth , mass , column density , and velocity dispersion of the tail that was unresolved by hipass . throughout the rest of this work
, we will refer to the tail resolved by the wsrt maps as @xmath14 , and the full extent revealed by the kwa combined data , which includes @xmath14 , as the `` extended tail . ''
in addition to hcg 44 , we detect the suspected background galaxy , sdss 1017723.29 + 214757.9 ( hereafter sdss 1017 ; @xcite ) , and likely foreground galaxy , lsbc f567 - 01 .
we note that the elliptical galaxy , ngc 3193 , remains undetected in hi .
@xcite propose that despite , its velocity ( 1381 kms@xmath18 ) , the absence of planetary nebulae in narrow band observations suggest it lies behind the compact group .
finally , ugc 5574 and ugc 5575 are also detected and , while not formally part of hcg 44 , are likely part of the larger associated group of galaxies @xcite .
we present the properties of all the hi detections in table [ sources ] with comparison to previous detections . in order to convert from column density to solar masses
we use 1 m@xmath3 pc@xmath68 atoms @xmath1 , integrate over the volume of the emission in the cube , and use the pixel scale at the luminosity distance of the object to convert to parsec@xmath61 . for the literature values ,
we take the published integrated flux and convert to hi mass using our assumed distances , and the standard equation : @xmath69 where @xmath70 is the intergrated flux in jy kms@xmath18 , @xmath71 is the distance in mpc , and @xmath72 is the hi mass computed in units of solar mass . these values are given in table [ sources ] .
the greatest source of error on the true hi mass content is the uncertainty in distance to individual galaxies both in and around the compact group .
ned gives estimates corrected for virgo infall of order @xmath73 mpc , while distances assumed in the literature range from 18.4 to 25 mpc @xcite .
the ngc 3190 group itself is part of a broader filamentary structure , so distance estimates from recessional velocities can be significantly distorted by infall onto the filament . following the convention of wsrt and alfalfa studies ,
we assume a distance of 25 mpc to hcg 44 member galaxies ngc 3185 , ngc 3187 , ngc 3190 , to the cloud , @xmath9 , and the extended tail .
we assume the same for ngc 3177 , ugc 5574 , and ugc 5575 which have similar recessional velocities and are at the same distance , within uncertainties , in velocity flow models of the region ( see table [ sources ] ) .
@xcite ) present a detailed discussion of distances to galaxies in the region .
table [ sources ] provides a comparison of the hi masses between the kat-7 observations , the kwa results , and previous observations from a range of telescope facilities .
a cursory look reveals that the interpretation of fluxes from different facilities is not straight forward .
the amount of hi detected by a given telescope is dependent on the size scales to which it is sensitive .
for example , if there is a significant amount of intragroup gas , we expect kat-7 and alfalfa to report more gas than wsrt , perhaps less gas than the gbt , and produce similar results to one another .
errors in the absolute flux calibration would also produce different results between facilities .
recalibrating data from five different telescopes to ensure consistency in the absolute flux calibration is outside the scope of this paper , however our observations give us the opportunity to comment on the relative flux calibration between kat-7 and alfalfa . among systematic sources of error ,
fitting a baseline to the data is particularly critical to account for both broadband continuum emission , and small scale residual bandpass fluctuations that may remain to separate the spectral line emission of individual galaxies @xcite . for both kat-7 and alfalfa
the baseline fitting takes place in a two step process .
there is a first pass to remove broad continuum emission , and a second pass when the detections are measured which fits for local fluctuations of the baseline in emission free channels over the region in which the spectral line flux is extracted .
@xcite show that hi flux densities measurements are typically not better than 15% .
the mean difference between kat-7 and alfalfa hi masses is 21% .
however , there is not a simple universal offset .
generally , the galaxies at the center of the kat-7 mosaic ( hcg 44 members and sdss j1017 ) are underestimated in the kat-7 data compared to alfalfa , and the galaxies on the outskirts of the mosaic ( ngc 3177 , lsbc f567 - 01 , and ugc 5574 ) , where the sensitivity of the kat-7 data are declining , are overestimated .
one possible explanation is that a better primary beam model is required for kat-7 than the gaussian assumed by the casa software .
a full model for the kat-7 primary beam is not currently available , but holography measurements of the individual dishes were taken as part of the commissioning process . in figure [ beam ]
we present the amplitude beam measurement of antenna 1 stokes i for kat-7 compared to the gaussian primary beam used to correct the data .
the engineering holography measurement has been arbitrarily scaled in the radial direction for comparison .
we see that , indeed , it appears the gaussian will under correct flux close to the center of a pointing , and over correct the flux on the outskirts . for the antenna 1 measurement ,
this difference is at the 2% level , but it may be greater when the entire array is considered together .
a study of 22 hcgs , which included hcg 44 , provided evidence of a diffuse neutral gas intragroup medium present in some compact groups @xcite .
however , based on follow - up mapping with the gbt , it was argued from the strong double - horned hi profiles that the majority of hi in hcg 44 is bound in stable disk - like structures @xcite .
these observations covered the three main galaxies in hcg 44 : ngc 3185 , ngc 3187 , ngc 3190 ; and the nearby @xmath74 cloud , and found a total hi mass of @xmath75 m@xmath3 but did not resolve the compact group .
we measure @xmath76 m@xmath3 in hcg 44 , which is consistent with the total hi mass resolved by wsrt @xcite .
while the gbt observations appear to detect 22% more hi in hcg 44 than either the kwa or wsrt observations , this seems to be within the variation measured for individual galaxies by different telescopes , even when those telescopes are sensitive to the same spatial scales .
the gbt and alfalfa values are also consistent with the vla - d observations , which show the hi is confined to individual galaxies @xcite .
thus , we conclude that even with deeper hi observations , there is no strong evidence for a diffuse , neutral igm in hcg 44 .
hcg 44 is known to be highly hi deficient , even when compared to other compact groups .
the expected hi mass for the group is calculated by summing the predicted values from the individual galaxies based on their morphological type @xcite and comparing them to the sum of the observed hi masses : @xmath77-\log[\sigma m(hi)_{\text{obs}}]$ ] .
the predicted value for hcg 44 is @xmath78=10.2\pm0.2 $ ] @xcite .
our observed value is in line with previous measurements : @xmath79=9.35\pm0.2 $ ] .
@xcite point out that the deficiency in hcg 44 is not due to a few highly anemic objects , but is a characteristic shared by all the members .
this argues that the group members have evolved together to account for the high hi deficiency , and against a single interloper which has had its gas completely removed .
the most probable effect produced by tidal interactions is believed to be disk stripping @xcite .
the role of ram - pressure stripping in compact groups has been shown to be small , only capable of removing small amounts of cold gas @xcite , however , turbulent viscous stripping due to hydrodynamical interactions may also be an important mechanism for removing gas from compact group members @xcite .
the advantage of low surface brightness sensitivity and wide field of view of our observations are apparent in observing the hi tail , @xmath14 , where we find @xmath1390% more hi than the wsrt observations alone .
however , the most striking result of our observations is detecting and resolving the gas beyond @xmath14 .
the extended tail is resolved into 7 - 8 distinct knots , and the total hi mass is of the order observed in a typical galaxy : @xmath2 m@xmath3 . nonetheless , even including the extended tail , the total gas content of the hcg 44 system only adds up to @xmath80=9.52\pm0.1 $ ] and can not account for the observed hi depletion of the compact group .
figure [ tailvel ] shows the intensity weighted velocity map of the extended hi tail .
the extended tail is contiguous in velocity with a shallow gradient that is parallel with the long axis of the tail from @xmath81 kms@xmath18 .
in fact , the westernmost extent , at 1270 kms@xmath18 , is not as blueshifted as the center of the tail at 1250 kms@xmath18 .
meanwhile the most massive clumps in the tail , which increase in mass to the east , are continuous from @xmath82 kms@xmath18 .
figure [ pvmap ] shows the extended tail forms a shallow arc in velocity . despite the 350 kpc extent of tail ,
it is quite narrow in velocity with a dispersion of only @xmath83 kms@xmath18 . in the most massive clumps
, @xmath14 has a dispersion of roughly @xmath84 kms@xmath18 , which peaks in cloud 3 . beyond @xmath14 ,
clouds 6 , 7 , and 8 have a dispersion of @xmath85 kms@xmath18 . and @xmath9 .
the data is summed over a slice 1.2 arcmins wide .
the contours have arbitrary values and are included to guide the eye .
there is no strong evidence for an hi bridge to the compact group beyond the southern extension of @xmath14 . ] and
the data is summed over a slice 1.2 arcmins wide .
the contours have arbitrary values and are included to guide the eye .
there is no strong evidence for an hi bridge to the compact group beyond the southern extension of @xmath14 . ] and
the data is summed over a slice 1.2 arcmins wide .
the contours have arbitrary values and are included to guide the eye .
there is no strong evidence for an hi bridge to the compact group beyond the southern extension of @xmath14 . ]
while hi is invaluable for recognizing a history of past interactions , disentangling the chain of events is more complicated .
the origin of the extended hi tail associated with hcg 44 is no less mysterious .
@xcite ruled out the likelihood that the tail is the result of ram pressure from the lack of diffuse x - ray emission .
the authors discussed two possible scenarios .
first , the gas was pulled out from ngc 3162 in a close passage with the compact group .
the elongated morphology of @xmath14 and the hipass detection suggest a tenuous connection between the tail and ngc 3162 , which lies approximately 620 kpc northwest of hcg 44 and the same distance from cloud 2 of the tail .
second , the gas originated from one of the galaxies currently within hcg 44 that has interacted with the group gravitational potential .
within hcg 44 both the stellar and resolved hi component of ngc 3187 show a strong s - shaped distortion ( figure [ wise ] and the wsrt contours in figure [ mom0 ] ) suggesting a strong tidal interaction .
meanwhile , in sdss optical images ngc 3190 has a strong dust lane with a swept - back morphology looking as though it has passed from the direction of @xmath14 , and the galaxy has strong infrared h@xmath6 emission which has been attributed to shocked gas @xcite . an outstanding question has been whether @xcite catch a glimpse of the high column density gas in an hi bridge which connects @xmath14 to the compact group .
if such a bridge exists , it should be visible in both our kat-7 and combined images in the most sensitive region of the cubes .
a connection would add weight to the second origin scenario , and we could trace from which hcg 44 member galaxy the gas was stripped .
we went through a concerted effort to uncover such a bridge , going through multiple iterations of targeted cleaning of the kat-7 cube during the imaging process , inspecting cubes at all stages of post - processing , including the kat-7 cube , the final data cube described in section [ combo ] , the spatially smoothed cube used for masking , and previous versions of the cubes at higher velocity resolution .
finally we inspected position - velocity slices of the cubes at various angles from ngc 3190 and ngc 3187 towards the high column density clumps of @xmath14 .
ultimately , we confirm faint continuous emission between cloud 1 and cloud 2 of @xmath14 , which makes up the southern extension , but we found no convincing evidence for an hi bridge connecting @xmath14 to any of the hcg 44 members at our detection limits ( e.g. figure [ pvmap2 ] ) .
the gap in hi emission between @xmath14 and hcg 44 may be explained if the h@xmath6 emission in ngc 3190 is not the result of shocked gas due to collisions with tidal debris , but comes from the presence of an active galactic nucleus ( agn ; @xcite ) due to jet - interstellar medium ( ism ) interactions @xcite .
the orientation of ngc 3190 is such that ionizing radiation emanating from the nucleus would have a clear path in the direction of where a bridge between hcg 44 and @xmath14 would exist .
thus , it is possible that we do not detect an hi bridge because it has been blown out or ionized , similar to gaps seen in hi tidal features such as around ngc 520 @xcite .
on the other hand , images from alfalfa which cover a larger area of sky show increasing evidence for a connection between the extended tail and ngc 3162 @xcite .
it is not obvious that the gas must have originated from ngc 3162 because there is still a gap in the observed hi distribution between the tail and ngc 3162 .
however , we note that the morphology of the extended tail is consistent with that of gas pulled out of galaxies in other high velocity encounters ( @xmath86 kms ) , such as virgohi 21 near ngc 4254 @xcite . in these simulations , the hi tidal tails that form
are long , narrow , and have a low stellar content .
there are further considerations , although it is unclear as to how they should be interpreted .
first , the tail is resolved into clumps and does not have a large velocity gradient as is seen in the @xcite simulations , although the latter may be projection effects .
second , the most massive hi clouds are those furthest from ngc 3162 , and closest to hcg 44 .
the most massive clouds to result from a tidal interaction are likely those pulled out near periastron , where tidal forces are the greatest .
third , the tail is the messiest to the southeast , close to hcg 44 .
this could be because the gas originated within the compact group , or if the tidal tail was pulled out from ngc 3162 it is now interacting with the compact group . finally , we note that ngc 3162 does not suffer from hi deficiency @xcite , although the uncertainty in expected hi mass for a galaxy of a given morphology and diameter is enough that we can not rule out ngc 3162 as the donor . in any case , it is probable that multiple interactions have occurred in the system .
a third scenario is that multiple interactions within the compact group loosened the cold gas of the galactic ism and allowed it to form a long tail in a high speed encounter with ngc 3162 , as modeled by @xcite .
the internal interactions shake up the hi gas which is then not bound as tightly in the shallow group potential , and would also account for the disturbed morphology of ngc 3190
. the overall hi deficiency of galaxies in the compact group would be the result of their history of group membership and gas depletion through star formation and gas processing , as proposed in other group environments by @xcite or @xcite . if the gas in the extended tail originated in an hcg 44 galaxy , it has not had to travel as far from its starting point as if it originated in ngc 3162 . in short , we present a number of possible scenarios for the origin of the extended tail , but the kwa data presented here is unable to say conclusively which is correct . despite the uncertainty in its origin
, we investigate the possible fate of the hi tail .
our data shows that it has a clumpy distribution with 7 - 8 knots of emission .
if we consider each knot as a distinct hi cloud , we note that the clouds with the highest peak column density and greatest velocity dispersion are the easternmost clumps of the tail .
tidal interactions are known to sometimes produce knots of star formation in stripped gas , and if these structures are self - gravitating , they are referred to as tidal dwarf galaxies @xcite . assuming the clouds are spherical , and the observed velocity dispersion is the thermal velocity of the gas , we can test to see if they will collapse .
the easternmost knot is @xmath87 m@xmath3 and has a velocity dispersion of @xmath88 kms@xmath18 .
the jeans length is @xmath89 kpc .
the wsrt appears to resolve this cloud to the size of 5 arcminutes which corresponds to 36 kpc ( figure [ mom0 ] , or figure 2 in @xcite ) so it is unlikely that stars are forming out of gas in the tail .
additionally , despite deep images from cfht / megacam with a limiting surface brightness sensitivity of @xmath1328.5 mag arcsec@xmath90 in g - band @xcite , there is no known stellar component in the optical , and nothing evident in wise infrared imaging ( figure [ wise ] ) .
if stars were pulled out of the galaxy concurrently with the gas , they would be broadly dispersed and potentially undetectable @xcite .
searches for planetary nebula line emission as a tracer for intragroup stellar light , conclude an upper limit of only 5% for the diffuse light fraction within hcg 44 itself @xcite .
on the other hand , we may investigate cloud survival in the context of photoionization .
observations have shown that both the outer disks of galaxies and tidal debris are protected from the ionizing ultraviolet ( uv ) background if the gas is above a few @xmath91 @xmath1 @xcite .
radiative transfer modeling suggests that clouds can survive in the intergalactic medium if they have a column density of at least @xmath92 @xmath1 @xcite . in our total intensity map ( figure [ mom0 ] )
the extended envelope of gas in the hi tail is closer to @xmath40 @xmath1 or less , but this may be an underestimate since the clouds themselves are largely unresolved .
the peak column density of clouds 2 and 3 are @xmath93 and @xmath94 @xmath1 , respectively .
@xcite point out that a 1 kpc cloud with an ism density of 1 atom @xmath95 expanding at 20 kms@xmath18 would survive for @xmath13500 myr before dropping to a density susceptible to photoionization by the uv background .
however , if the density of the igm is of order a few to tens @xmath96 @xmath95 the cold hi clouds will reach pressure equilibrium before reaching the critical column density , and may survive longer .
these igm densities have been inferred indirectly through the morphology of bent jets in galaxy groups @xcite .
we find that both the column density and velocity dispersion of the hi clouds decrease as one moves along the tail from east to west ( see figure [ pvmap ] and section [ hitail ] ) .
if we consider that the gas has been pulled out of ngc 3187 , as in @xcite s 3d toy model , and/or ngc 3190 , the clouds may represent snapshots of evolution at different times after they have been pulled out of the galaxy . in this case
the hi clouds become fainter as they disperse and the velocity dispersion decreases as what remains is the cold core of the cloud .
our extrapolation of the proposed trajectory from the 3d model implies that the easternmost hi clouds were stripped of order 500 , 600 , and 700 myr ago .
further , the clouds in the extended tail fall at intervals which would imply they were stripped approximately 850 , 1000 , and 1100 myr in the past .
these lifetimes suggest that intragroup hi clouds may have survived by being contained in pressure equilibrium with a warm intragroup medium that has not been directly detected , and which may extend out to quite large distances beyond the compact group perhaps as part of the larger gh58 or ngc 3190 loose group to which hcg 44 apparently belongs @xcite . alternatively
, we can make a simple estimate of the required survival time of the clouds if they are a result of a close encounter with ngc 3162 .
ngc 3162 lies at a projected distance of 620 kpc from hcg 44 , and at relative fly - by velocities of @xmath97kms@xmath18 , the clouds would have had to survive between 1 - 3 gyr .
the lower velocity limit seems challenging , given the long required lifetime of the clouds in the igm .
meanwhile , the high velocity limit has been shown to produce long tidal tails in cluster systems , but may be unusually high for a galaxy group system with few members and an otherwise narrow velocity dispersion ( 115 - 220 kms@xmath18 depending on what is considered part of the group ; @xcite , @xcite )
. simulations of small hi clouds survival in a milky way - like hot ( @xmath98 k ; @xmath99
@xmath95 ) halo suggest that neutral hi which has been stripped can last of order 80200 myr before it is disrupted or ionized @xcite .
the hi clouds of these simulations have a range of mass from ( @xmath100 m@xmath3 ) , but @xcite summarizes that total mass seems to be one of the largest factors which increase their lifetime .
the hydrodynamic simulations also include the effects of ram pressure stripping , in addition to photoionization .
if the effects of ram pressure are less severe on the hcg 44 intragroup clouds @xcite , perhaps they are able to live longer .
if they have already lived in the igm for 1 gyr , they probably started with a larger initial mass than what we currently observe .
hi is a key tracer of physical mechanisms at work in galaxy groups as the gas is both the most extended component of pristine galaxies , and also most easily removed through gravitational and hydrodynamical interactions .
it can provide rigorous constraints on galaxy - galaxy interactions , group dynamics , and indirect evidence of the warm - hot igm which remains otherwise undetected .
tidally stripped hi in groups appears to survive as both filamentary structures and a diffuse component @xcite .
this gas can provide a reservoir for re - accretion through so - called rejuvenation events " , but collisions with this intragroup medium may also produce shocks in the interstellar medium of galaxies @xcite .
associated shock heating may lead to further stripping , or the formation of molecular hydrogen through gas compression at the leading edge of the shock@xcite . to date
the preponderance and evolutionary effects of these processes remain relatively unexplored and poorly understood . in hcg 44 ,
if previously detected h@xmath6 is the result of shocks due to collisions with the igm , the intragroup medium doing the work remains undetected in our deep hi data . on the other hand ,
if the h@xmath6 emission is due to jet - ism interaction shocks from an agn in ngc 3190 , it may explain the absence of an hi bridge between hcg 44 and the extended hi tail .
qualitative analytical arguments for hi cloud survival have estimated cold cloud lifetimes of 500 myr once gas has been removed from a galaxy disk @xcite . however , if the orbit of ngc 3187 modeled by @xcite is representative of a group member interacting with the hcg 44 group potential , the hi clouds at the end of the extended tail may have already survived for up to 1 gyr . if the gas originated in ngc 3162 , the clouds have lived in the circumgroup medium for 1 - 3 gyr .
this result warrants further simulation to understand how the balance of physical processes between the cloud and the circumgroup medium allows for cloud survival .
unfortunately , the warm - hot intragroup medium has thus far evaded direct detection so the density and temperature of the intragroup medium is constrained more by what we do not see than what we do ( e.g. @xcite ) . however , our result demonstrates the potential of hi to provide even stronger constraints on the modeling of galaxy group properties .
future wide - area hi surveys ( e.g. wallaby , dingo , mightee - hi , and apertif surveys ) will uncover many more nearby hi rich groups for deep follow - up . a number of preliminary results from deep kat-7 observations ( oosterloo et al , in prep ; lucero , private communication ) and from askap @xcite ,
show that extended and low - surface brightness hi features may be even more common in galaxy groups than previously realized .
tidal interactions produce material that lead to viscous stripping of the interstellar medium from galaxies , and this process may be important in loose groups in addition to compact groups .
finally , with the expected increase in the gas fraction of galaxies as one moves to higher redshift , group interactions may play an even more significant role in the evolution of galaxies in the past than what is seen locally . in other words , the effect of environment when galaxies were typically gas - rich ( @xmath101 ) may be a key consideration at a time when the universe was experiencing elevated star formation and more efficiently building stellar mass compared to local systems .
we observed the region around hcg 44 and the giant hi tail discovered by @xcite , which extends to the north and west of the compact group , to investigate the extent of intragroup hi .
we combined a deep two - pointing mosaic from kat-7 with imaging from the alfalfa survey , and a single pointing observation of wsrt . despite the very different nature of the telescopes , we successfully combined the data by converting the image units of jy beam@xmath18 to column density per unit velocity , atoms @xmath1(kms@xmath18)@xmath18 , weighting by the sensitivity of the respective observations , and adding them in the image plane .
we find no evidence for copious amounts of neutral hi in the igm within the core of the compact group , which suggests the h@xmath6 emission observed by @xcite in ngc 3190 may be the result of agn activity and jet - ism interaction rather than shocked gas from ngc 3190 plunging through the igm .
the presence of an agn , or star formation in the nucleus of ngc 3190 could account for the lack of an hi bridge between hcg 44 and the hi tail , if hcg 44 is where the extended hi tail originated from .
the combined hi data show that the hi tail extends out to at least 450 kpc and is resolved into 78 clouds at the resolution of kat-7 .
we find more hi mass in all parts of the tail as compared to previous wsrt observations , such that the total hi mass in the tail is @xmath2 m@xmath3 .
however , the gas in the tail is still not able to account for the overall hi depletion in the compact group . the remaining gas is either too diffuse to be measured with current facilities , or more likely
has been ionized and may exist in a warm - hot group halo that is also undetected in x - rays .
we find that if the gas in the tail is stripped from one of the hcg 44 member galaxies , and the timescale for stripping proposed in @xcite is correct , the gas must have been able to live in the intragroup medium without being ionized for 0.5 - 1 gyr , which is longer than previously predicted . if the gas originated in ngc 3162 , it must have been produced in a high speed encounter , and even then must have survived in the igm for well more than 1 gyr .
regardless of the origins of the hi tail , our results demonstrate indirect evidence for a surrounding medium which has confined the expansion of the hi clouds and prevented their destruction by background uv radiation .
targeted simulations are required to explain how these hi clouds can survive for long periods of time in the igm .
the research of kmh has been supported by the european research council under the european union s seventh framework programme ( fp/2007 - 2013)/erc grant agreement nr .
mec acknowledges support from the south african national research foundation .
the work of sy was supported by the 2015 astron / jive summer student programme .
the work of cc is based upon research supported by the south african research chairs initiative ( sarchi ) of the department of science and technology ( dst ) , the square kilometre array south africa ( ska sa ) and the national research foundation ( nrf ) .
we thank thijs van der hulst , dj pisano , lourdes verdes - montenegro for useful discussions .
the authors would like to acknowledge the work of the entire alfalfa collaboration team in observing , flagging , and generating the data products used in this work .
the alfalfa team at cornell is supported by nsf grants ast-0607007 and ast-1107390 to rg and mph and by grants from the brinson foundation .
this paper made use of observations obtained with the westerbork synthesis radio telescope , which is operated by the astron ( netherlands foundation for research in astronomy ) with support from the netherlands foundation for scientific research ( nwo ) .
this publication makes use of data products from the _ wide - field infrared survey explorer _ , which is a joint project of the university of california , los angeles , and the jet propulsion laboratory / california institute of technology , funded by the national aeronautics and space administration ( nasa ) . | extending deep observations of the neutral atomic hydrogen ( hi ) to the environment around galaxy groups can reveal a complex history of group interactions which is invisible to studies that focus on the stellar component .
hickson compact group 44 ( hcg 44 ) is a nearby example and we have combined hi data from the karoo array telescope , westerbork synthesis radio telescope , and arecibo legacy fast alfa survey , in order to achieve high column density sensitivity ( @xmath0 @xmath1 ) to the neutral gas over a large field - of - view beyond the compact group itself . we find the giant hi tail north of hcg 44 contains @xmath2 m@xmath3 of gas and extends 450 kpc from the compact group : twice as much mass and 33% further than previously detected .
however , the additional gas is still unable to account for the known hi deficiency of hcg 44 .
the tail likely formed through a strong tidal interaction and hi clouds in the tail have survived for 1 gyr or more after being stripped .
this has important implications for understanding the survival of neutral clouds in the intragroup and circumgroup medium , and we discuss their survival in the context of simulations of cold gas in hot halos .
hcg 44 is one of a growing number of galaxy groups found to have more extended hi in the intragroup and circumgroup medium than previously measured .
our results provide constraints for simulations on the properties of galaxy group halos , and reveal a glimpse of what will be seen by future powerful hi telescopes and surveys .
[ firstpage ] galaxies : evolution galaxies : interactions galaxies : ism radio lines : galaxies galaxies : groups : individual : hcg 44 |
hydrogen degrades the performance of many metals by lowering the failure stress , leading to fracture at unpredictable loading conditions @xcite .
several mechanisms have been proposed to explain the h - embrittlement of iron : the hydrogen - enhanced decohesion ( hede ) mechanism @xcite , h - vacancy effects @xcite , hydrogen - enhanced localised plasticity ( help ) @xcite , or a combination thereof .
however , none of these explanations alone seems to provide a complete explanation of the experimentally - observed features of h - embrittlement . in the hede mechanism ,
h weakens the cohesive bonds between the metal atoms , leading to failure at interfaces , as h exhibits increased solubility in the tensile strain field of a crack opening .
vacancies containing h can order themselves along critical slip directions , leading to fracture @xcite . within the help mechanism ,
the onset of plasticity with loading occurs at a lower stress as a result of the h shielding of repulsive interactions between dislocations@xcite .
the increased h concentration at dislocations @xcite effectively reduces the dislocation - dislocation distances . together with the experimentally - observed increase of dislocation mobility @xcite
, this may explain the phenomenon of dislocation coalescence , and ultimately , crack advancement at reduced loads .
the importance of increased h - concentration near dislocations and other low - energy trap sites was also indicated by in a recent experimental study of intergranular failure in steel@xcite . in that study ,
the fracture mode changed from ductile to brittle , as the amount of h located at these low - energy trap sites increased , indicating that a study of h at dislocations is important .
the goal of this study is to achieve a microscopic understanding of the changes in the mechanical properties due to h , and their connection with the proposed theories of h - embrittlement .
focussing on @xmath0-fe , we study the influence of interstitial h on the elastic properties , and the behaviour of h within strain fields representative of dislocations . to achieve this , we use density - functional theory ( dft ) calculations to calculate the changes in the elastic properties of @xmath0- ( bcc ) iron , as a function of h - concentration and employ these results in a continuum elasticity theory description of the h - solubility in strain fields . in sec .
[ dft - details ] we describe the details of our dft calculations while sec .
[ elastic - section ] describes the calculations of the modification of the elastic parameters as a function of h - concentration
[ strain - dep ] discusses the solubility of h in strain fields , and we conclude with a discussion in sec .
[ concl ] .
our first - principles density functional theory calculations were performed using the vasp @xcite code .
we used the projector augmented - wave method @xcite and employed pseudopotentials considering the 3p electrons of fe as valence electrons . the generalized - gradient approximation ( gga ) in the pw91 parametrisation @xcite was used for the exchange - correlation functional .
all of our calculations were spin - polarised .
some of our calculations were repeated using the vosko - wilk - nusair @xcite ( vwn ) spin interpoilation in pw91 , or the perdew - burke - ernzerhof @xcite ( pbe ) exchange - correlation functional and are discussed in sec .
[ hconcsection ] . a plane - wave basis with a cutoff of 500 ev and supercell geometry were used .
a @xmath1-centred k - point grid equivalent to 18@xmath218@xmath218 for the two - atom basis bcc unit cell was used for the brillouin - zone sampling except for the 128-atom fe cell where it was 20@xmath220@xmath220 .
our results for the lattice parameter ( 2.832 ) , bulk modulus ( 194 gpa ) , and magnetic moment ( 2.17 @xmath3 ) are in good agreement with other first - principles calculations _
e.g. _ refs . and with experiment .
the plane - wave basis and k - point sampling density used are more than adequate for converging the equilibrium properties , but necessary for converging elastic moduli to within less than one percent .
the ions were relaxed such that the maximum component of the force on each ion was less than 0.01 ev / while the total energies were converged to within 0.01 mev .
the lattice parameter for systems constrained to be cubic ( see following section for a discussion of this approximation ) was determined from fitting the total energies to the murnaghan equation of state @xcite . at times , the correct lattice parameters for a tetragonal unit cell were desired ; these were obtained from a quadratic fit over a two - dimensional grid of total energies . in this study
, we focused our attention on tetrahedral interstitial h , for two reasons : one is that at zero stress we found the tetrahedral interstitial position to be 0.13 ev more stable than the octahedral position , in agreement with previous dft studies @xcite , and secondly , as there are twice as many tetrahedral as octahedral sites per fe atom ( six versus three ) , we expect the former to dominate the mechanical properties at ambient temperatures .
the total energy of a solid at zero stress and equilibrium volume @xmath4 can be expanded about small strains @xmath5 using the voigt notation ( @xmath6 running from 1 to 6 ) : @xmath7 in this section we restrict ourselves to linear elastic behaviour , _ i.e. _ stress linear in the strain the first two terms of eq . [ e_series ] . for the calculation of the elastic parameters in this work ,
cubic symmetry was enforced .
this resulted in an error of less than 2% , as will be described at the end of this section . the number of independent elastic parameters @xmath8 depends on the crystallographic symmetry . for cubic systems
, there are three unique elastic parameters , c@xmath9 , c@xmath10 , and c@xmath11 that describe the change in energy due to applied strain .
the parameters @xmath8 can be determined by applying suitable strain tensors and taking the second derivative of eq .
[ e_series ] with respect to the amount of strain .
the combination @xmath12 is the bulk modulus , found by applying hydrostatic strains and then doing a fit to the murnaghan equation of state , whereupon the pressure derivative @xmath13 is also found .
the combination @xmath14 , and the coefficient @xmath15 are found by applying volume - conserving strains . in the case of @xmath16 ,
an orthorhombic strain @xmath17 is applied , giving a total energy expression @xmath18 , \label{e_en1}\ ] ] while , for @xmath15 , the strain is monoclinic : @xmath19 with a total energy @xmath20 .
\label{e_en2}\ ] ] we fitted our total energies for systems subject to nine values of strains @xmath21 between @xmath22 to eqs .
[ e_en1 ] and [ e_en2 ] in order to obtain @xmath16 and @xmath15 . to estimate the accuracy of the elastic - parameter calculations ,
extensive convergence tests were done .
the quantities varied were the k - point density , cutoff energy , and percent strain ( value of @xmath21 ) . the lattice parameter was fixed at 2.832 for all tests .
an example of these convergence tests for @xmath15 is shown in fig.[converg]a , for @xmath21=3% ( the convergence tests for @xmath16 are similar ) .
the values of cutoff energy and k - point density of the circled point ( 500 ev , 18@xmath218@xmath218 ) were used in ( b ) and ( c ) to check that the applied strains are within the linear regime . as a function of k - point density and energy cutoff for the two - atom bcc - fe cell for a range of @xmath23 .
the error bars are deduced from the least - squares fit to eq .
[ e_en2 ] .
the circled point corresponds to the calculational parameters used in this study .
( b ) @xmath15 and ( c ) @xmath16 fitted between different ranges of @xmath24.,width=336 ] as a result of our extensive convergence tests , we observed less than a 1@xmath25 variation in the elastic parameters upon going from 0.5 - 3@xmath25 change in @xmath21 .
for the calculations presented in the remainder of this study , we chose a value of 3@xmath25 for @xmath21 in order to avoid numerical instabilities and/or the need for very high presicion total energy evaluations .
the volume expansion of the crystal lattice by h in the tetrahedral interstitial site introduces small tetragonal distortions , breaking the cubic symmetry of the fe host lattice .
an h - orientation such as the one shown in fig .
[ tetr - fig ] causes an expansion of the fe nearest - neighbours radially - outward from h ( along the red lines ) this expansion , projected onto the cube axes , is of an identical amount in the _ a _ and _ c _ direction , and greater than the expansion along _ b_. the tetragonal distortion , with lattice parameters @xmath26 , increases the number of unique elastic parameters from three to six : @xmath27 , @xmath28 , and @xmath29 .
the variation among the no - longer equivalent elastic parameters depends on the degree of tetragonal distortion , generally becoming more pronounced with increasing h - concentration , unless there are two or more h in the supercell , in which case the amount of distortion depends also on their relative positions .
generally , h would be expected to be randomly distributed , resulting an isotropic lattice distortion . as well , considering a cubic unit cell facilitates the scale - bridging with mesoscopic approaches such as _ e.g. _ finite - element schemes .
the error due to imposing cubic symmetry can be estimated by explicitly calculating the ( formally - degenerate in cubic symmetry ) elastic parameters , for tetragonal symmetry for a configuration with lattice parameters @xmath26 by straining the system in all three cartesian orientations , all the while maintaining a cubic unit cell .
the reason that the results are not equivalent in all directions even under the assumption of cubic symmetry is because of the local internal tetragonal symmetry around the h atom .
nevertheless , the variation amongst the three orientations was found to be small : in two of the orientations the elastic parameters were identical , as expected , while those from the third differed by about 4% , making the standard error less than 2% for the highest concentration ( 11% ) studied , where the tetragonal distortion @xmath30 was -0.004 , where the two h were oriented such that the distortion is double that caused by a single h. for all of our data above and including a h - concentration of 4% , the formally - degenerate elastic parameters were explicitly calculated by straining in the three cartesian directions and the results presented are the average values along with the associated standard errors . for lower concentrations , the spread amongst the three directions was @xmath31 1% and corresponding errors presented are from the least - squares fitting of the total energy as a function of strain .
in addition , for the system with @xmath32 ( at 11 at.% h ) , the elastic parameters for an explicitly tetragonal lattice were calculated using the strain and total energy expressions given in ref . .
the differences from the simplified case of the cubic unit cell with distortions applied in all three orientations were small , less than 2% for the largest h - concentration studied .
table [ tetrtable ] summarises the average values of the cubic and tetragonal elastic parameters along with the associated standard errors computed from the spread in the values for the different orientations .
only @xmath33 gave a large spread for the tetragonal cell ( separately , the values were @xmath34=1.38 , @xmath35=1.64 ) .
that the agreement is good is not surprising , since the volume of the cubic cell was made to be the same as that of the tetragonal cell and the variation in the elastic parameters arising from different lattice parameters ( sec .
[ nonlin ] ) in the tetragonal cell mostly averages out in the cubic cell .
.elastic parameters ( gpa ) for the highest - considered h - concentration ( 11 at.% ) for a cubic or tetragonal ( 1-b / a=-0.4% ) unit cell , both having the same volume . for the cubic cell
, the strains were still performed in the three , formally - degenerate directions , and averaged , as the results were different due to the lower symmetry locally around the defect . [
cols="^,^,^",options="header " , ] ) for a reference pure fe system with ( a ) same strains , hydrostatic only ( b ) same strains , shear only , and ( c ) same stresses , shear only .
the strain on the horizontal axis is for the feh system.,title="fig : " ] ) for a reference pure fe system with ( a ) same strains , hydrostatic only ( b ) same strains , shear only , and ( c ) same stresses , shear only .
the strain on the horizontal axis is for the feh system.,title="fig : " ] ) for a reference pure fe system with ( a ) same strains , hydrostatic only ( b ) same strains , shear only , and ( c ) same stresses , shear only .
the strain on the horizontal axis is for the feh system.,title="fig : " ] here we will look at the variation of h - solubility as a function of strain direction . as well , we aim to visualise the h - solubility in a typical dislocation strain field .
we will focus on the particular case of a screw dislocation lying along @xmath36 , which is the predominant orientation in a bcc metal @xcite .
the screw dislocation strain field was modelled using isotropic dislocation theory @xcite . in cylindrical @xmath37
coordinates , its only nonzero component is @xmath38 where @xmath39 is a burger s vector equal to the length of @xmath40 . all of the figures and equations for shear strain ( figs . [ sole]c , [ solestress]b - c , eqs .
[ pshreqn]-[hydroeqn2 ] ) look the same whether the shear strain is @xmath41 , @xmath42 , or @xmath43 . for a screw dislocation line situated along @xmath44 ,
the nonzero components are @xmath45 .
therefore , referring to fig .
[ solestress ] , we see that h is trapped towards a screw dislocation core in the case where the pure fe reference system has the same shear strain as the feh system ( fig .
[ solestress]b ) .
we aimed to see whether the repulsive behaviour of h by a screw disclocation at constant volumes ( fig .
[ sole]c ) can be altered , or even reversed in orientations other than the coordinate system defined by the cube axes .
the solubility equations [ hydroeqn]-[hydroeqn2 ] were examined to see whether they yield different values in rotated coordinates corresponding other orientations of the dislocation line .
depending on the choice of coordinate system , it is , in general , possible to get stronger or weaker variations for the solubility than those shown in fig .
already another combined atomistic / dft study has shown that h - trapping in bcc fe is very dependent to the direction of applied shear @xcite .
for the case of purely hydrostatic strain , we find that the solution energy was independent of the coordinate system chosen .
this is because in our model ( eq . [ sol3bcub ] ) both the term linear in strain as well as the quadratic term can be factored in terms of the trace @xmath46 .
since , as can be shown , the trace of the strain tensor is invariant under any unitary transformation , the solution energy for a hydrostatic distortion is thus rotationally - invariant . in the case of the screw dislocation , there is no such restriction ; it is certainly possible to have different coefficients , some leading to h - trapping near a screw dislocation core .
figure [ solscrew ] shows the variation of the solution energy for a fixed , 2% , shear strain where eq .
[ pshreqn ] has been rotated onto other axes . by rotating the entire coordinate system , @xmath47 ,
other directions @xmath48 for the dislocation line ( legend labels ) are achieved .
the label @xmath49 on the horizontal scale defines @xmath50 and hence the shear direction . as the directions @xmath51 , @xmath52 , and @xmath36 are mutually perpendicular , the first two intersect the latter at @xmath53 and @xmath54 respectively .
we expressed the strains @xmath55 in the primed system in terms of the unprimed system and inserted these values into the general expression for the solution energy eq .
[ sol3bcub ] .
note that only the @xmath56 components are considered .
for the full description of the screw dislocation , the @xmath57 components are also active .
the latter components result in the same curves as those depicted except with a @xmath58 phase shift such that the average value , corresponding to all components of a screw dislocation , is flat , independent of shear direction .
in contrast with the study [ ] , we do not find asymmetry of our solution energy @xmath59 about zero strain as a function of applied shear strain . to do
otherwise would mean acquiring linear terms , which are not possible with our model . ) for a dislocation line / burger s vector oriented along different directions ( legend ) .
the direction of the dislocation line/ burger s vector is , black : @xmath60 $ ] , blue dashed : @xmath61 $ ] , red : @xmath62 $ ] , green dashed @xmath63 $ ] .
the horizontal axis is the variation of the solution energy as a function of the choice of @xmath64 and @xmath50 axes ( and hence the shear ) within the plane normal to the burger s vector .
@xmath65 has been arbitrarily chosen to correspond to the @xmath66 direction .
the dotted line is the unstrained value of the h solution energy . ] for each point on a grid in the plane perpendicular to the dislocation line , the solubility corresponding to the complete strain tensor was calculated from eq .
[ sol3bcub ] in the case of equal volumes or analogous equations ( not given due to their complexity ) for the cases of equal stresses and equal strain .
one caveat concerning the case of equal stresses is that we are not modelling the solubility of h in the stress field of a dislocation but its solubility at individual stresses at each point @xmath67 , such that the material is not self - consistently in mechanical equilibrium .
however , this is a first approximation in a self - consistent treatment of h - solubility .
the results , shown in fig . [ solscrewdisl ] , show that h is attracted to the core for equal shear strains or equal volumes and repelled for equal shear stresses
. screw dislocation core for the cases of equal stresses , equal volume , or equal strains for the feh and pure fe systems .
the dotted line denotes the zero - strain solubility . ]
motivated by the strong strain - solubility effects , we aim to calculate the local modification of the h concentration by strain fields . for the purposes of illustration
, we consider only the case of equal volumes in this discussion .
we consider the change in free energy @xmath68 as a result of adding h to fe under strain .
the free energy is @xmath69 where @xmath70 is the temperature , and @xmath71 is the total entropy of the feh system . in order to isolate the effect of adding h to the already - strained fe lattice , we assume that the entropy of the h is independent of that of the fe lattice : _ i.e. _ @xmath72 .
@xmath73 is the total energy of the ( strained ) feh system .
it can be rewritten in terms of the solution energy @xmath59 ( _ c.f . _ eq .
[ solw ] ) as @xmath74 , where we have set the energy per h as @xmath75 , in contrast with eq .
[ solw ] , where it was equal to @xmath76 .
the free energy is minimized with respect to @xmath77 : @xmath78 we first focus on the case of a hydrostatic strain field . for simplicity
we truncate equation [ hydroeqn ] from the previous section to linear order in strain : @xmath79 the expression for the entropy can take on various forms , in the presence of various forms of h - h interactions , all equivalent at lower concentrations . in
what follows , @xmath77 is the number of h , @xmath80 is the number of matrix ( fe ) atoms , @xmath81 is the total number of tetrahedral interstitial lattice sites , and @xmath82 is the occupation concentration defined as @xmath83 .
the simplest form for the entropy , @xmath84 , arises from an unrestricted concentration of h , corresponding not to a lattice but to an ideal gas . in this case
, the entropy derivative is @xmath85 .
the concentration yielded by eq .
[ omega_d ] for the case of hydrostatic strain up to linear order in the strain takes the maxwell - boltzmann form of @xmath86\ ] ] where @xmath87 $ ] .
this reproduces the result from the cottrell picture of enhanced concentration around a dislocation @xcite .
a more realistic form of the entropy takes into account the presence of a lattice and enforces a maximum occupancy of one h per tetrahedral site . in this case
the entropy is @xmath88 $ ] and the entropy derivative is @xmath89 .
the concentration using this form follows fermi - dirac statistics : @xmath90 + 1\right\}^{-1}.\ ] ] however , the maximum occupancy of one h per tetrahedral site is not realistic as the h - h interaction in fe is repulsive .
we have calculated the h - h interaction as a function of distance and find strong h - h repulsion up to third - nearest neighbours . for a realistic representation of the statistics , up to and including third nearest - neighbour h - h spacings should be strongly unfavourable in our model . in the extreme case ,
these sites are left unoccupied , corresponding to infinite repulsion . the expression for the entropy derivative for tetrahedral interstitials in a bcc lattice with total exclusion of first nearest - neighbours , both first and second - nearest - neighbours , and up to third - nearest neighbours
is given by boureau @xcite as @xmath91 where @xmath92 , @xmath93 , and @xmath94 $ ] respectively . in the limits of low concentration , all of the expressions for the entropy ( @xmath84 , @xmath95 , @xmath96 , @xmath97 , @xmath98 ) are the same , the differences being of @xmath99 .
similarly , at small strains , all of the solutions for @xmath82 are identical . a numerical example of the extent of the modification of the h - concentration in bcc fe by applied hydrostatic strain is shown in figure [ conc_low]a .
in this low - concentration regime , all forms of the entropy are identical . in this example
, the background ( zero - strain ) concentration was taken to be equal to 3@xmath100 , in line with typical densities of h in iron @xcite .
at each temperature studied , @xmath101 was adjusted in order to achieve this background concentration . from the figure ,
it is clear that the h - concentration can be dramatically increased by orders of magnitude , even at very small strains .
this increase in concentration is more pronounced at lower temperatures , in agreement with the low thermal desportion temperatures , _ ca _ 400 k , indicated for h trapped at dislocations @xcite .
edge dislocations have a significant hydrostatic strain component @xcite so we take our results to be representative of these . with shear strain oriented in the plane normal to the @xmath36 direction , we obtain the curves in fig . [ conc_low]b .
the effect is not as dramatic , because the solubility energy is quadratic in the shear strain ( _ c.f . _
. [ pshreqn ] , whose coefficients change upon rotation to this coordinate system , but not the functional form ) .
direction , for a chemical potential giving a background concentration of @xmath102 h per fe atom at each temperature . ]
we have studied the modification of elastic properties of h in bcc fe .
the modification quite profoundly affects the elastic moduli .
using this knowledge , we have characterised the strain - dependence for h - solubility in fe . from our model
we find , in agreement with existing experimental studies @xcite , that typical strain fields found near dislocations can trap lots of h. we further show that the local h - concentration may be enhanced by several orders of magnitude near dislocations with tensile strain components , which justifies our use of concentrations in the atomic % range for our calculations . in the case of tensile strain , we find that these areas of enriched local concentration tend to occur mainly at low temperatures , in the hundreds of k , which agrees with the h - trapping energies found in the thermal desorption results of novak . in the case of screw dislocations , we also find trapping of h if the screw dislocation is oriented along the @xmath103 direction but the concentration enhancement is not as great as for the edge dislocation .
screw dislocations along @xmath103 are the predominant types of dislocations in bcc metals driving plasticity @xcite and the concentration enhancement near screw dislocations does not rule out the help mechanism in explaining the mechanism of h - embrittlement of bcc fe .
we acknowledge financial support through thyssenkrupp ag , bayer materialscience ag , salzgitter mannesmann forschung gmbh , robert bosch gmbh , benteler stahl / rohr gmbh , bayer technology services gmbh and the state of north - rhine westphalia as well as the european comission in the framework of the erdf . | the effect of interstitial hydrogen on the elastic properties of @xmath0-iron is investigated using _
ab initio _ density functional theory .
we find that while the overall strength properties are reduced by h , the effects are mainly due to the resulting local volume expansion .
we use these concentration - dependent elastic moduli to model the effects of homogeneous strain fields on h solubility and make extrapolations to h - solubility in the spatially - varying strain fields of realistic dislocations .
we find that h is strongly trapped by dislocations of both the edge and screw variety , leading to a remarkable increase in the local h concentration near the dislocation cores .
the strain dependence of the solution energy leading to accumulation of h near dislocation cores , as well as the reduction of elastic moduli which our calculations predict , is consistent with numerous experimental studies which indicate that the trapping of h near dislocations plays a significant role in the embrittlement of iron and steels . |
a flat manifold is a closed riemannian manifold with zero sectional curvature . from bieberbach theorems
, it is known that in each dimension there is only a finite number of such manifolds , up to affine equivalence , and the problem of classifying them is important .
recently , this has been achieved up to dimension six ( cf.@xcite ) , but it is not possible to do it in general .
hence some interesting families of flat manifolds have been considered , as for instance those with first betti number zero or those with holonomy group @xmath2 .
the purpose of this article is to study another special class of flat manifolds , namely those of dimension @xmath0 with holonomy group isomorphic to @xmath3 . in dimension two
the klein bottle belongs to this family and in dimension three there are three such manifolds : a classical flat manifold first considered by hantzsche and wendt @xcite ( now called _ didicosm _ , see last section ) and two non - orientable ones .
though here they are considered for first time together , these manifolds have been studied partially in @xcite , @xcite , @xcite , @xcite , @xcite , mainly in the oriented case , where we shall call them _ hantzsche - wendt manifolds _ ( hw manifolds for short ; respectively hw ( bieberbach ) groups , if we consider their fundamental groups , which are bieberbach groups ) .
they have many interesting properties , for instance , there are pairs of isospectral hw manifolds non homeomorphic to each other ( cf . @xcite ) .
moreover , hw manifolds have the @xmath4-homology of spheres ( cf .
@xcite ) and hence they are geometrically formal ( cf . @xcite ) .
we shall call the manifolds in our family _
generalized hantzsche - wendt _ manifolds ( ghw manifolds for short ; respectively ghw ( bieberbach ) groups ) .
one of the main results in this article is that the integral holonomy representation of a ghw manifold is of diagonal type ( theorem [ diag ] ) .
this simplifies the study of this family .
an immediate consequence is that the hw manifolds are exactly those considered in @xcite .
another interesting property of these manifolds , or their fundamental groups , is given by the relations between them in different dimensions .
in fact we show that every ghw group has a subgroup isomorphic to a ghw group of one lower dimension ( corollary 3.4 ) .
moreover , from theorem [ exist ] , any @xmath0-dimensional ghw group is a subgroup of some @xmath5-dimensional one . with these observations
we define a graph , whose vertices correspond to ghw groups , connecting the family , which we draw up to dimension four in the last section .
it is clear that the number of non - oriented ghw flat manifolds is much greater than the number of oriented ones ( see for instance the table in section [ no ] ) . in the non - oriented case ,
there are two types of ghw groups , depending on whether the first betti number equals one or zero .
we show that the case with first betti number one ( as well as the oriented case ) has only one holonomy representation , which is completely determined .
hence it makes sense to carry out the calculation of the @xmath4-homology ( see theorem [ betti ] ) .
the situation in the non - oriented case with first betti number zero is much more complicated since the holonomy representation is not unique .
we find the exact number of integral representations occurring in this subclass and show that it grows linearly with the dimension .
this follows as a consequence of theorem [ exist ] , which asserts that every diagonal integral representation can be realized as the integral holonomy representations of a flat manifold . in section [ examples ]
we consider some examples in this subclass which are similar to the hw manifolds from @xcite . in section [ outaut ]
we study some symmetries of ghw manifolds .
in particular , we determine the outer automorphism groups of certain subfamily of ghw groups with non - trivial center .
moreover a general upper bound of the order of the outer automorphism group of ghw groups is given .
there are some problems and questions which we formulate at the end .
we would like to thank wim malfait for discussions regarding section [ outaut ] .
let @xmath6 be a bieberbach group of dimension or rank @xmath0 . by definition
it is a torsion - free , discrete , cocompact subgroup of the group @xmath7 of rigid motions of the @xmath0-dimensional euclidean space .
the multiplication in @xmath8 is given by @xmath9 where @xmath10 and @xmath11 .
the translation part @xmath12 of an isometry @xmath13 shall sometimes be denoted by @xmath14 . if a flat manifold @xmath15 is oriented then by definition @xmath6 is oriented . from bieberbach theorems ( cf.@xcite ) ,
the translation subgroup @xmath16 of @xmath6 is normal , maximal abelian , of finite index .
then there is a short exact sequence of groups @xmath17 where @xmath18 is finite and isomorphic to the holonomy group of the manifold @xmath15 , thus we call it the holonomy group of @xmath19 it is well known that any extension as ( [ short ] ) defines an element @xmath20 , where the @xmath18-module structure of @xmath21 is defined by conjugation in ( [ short ] ) .
we denote by @xmath22 this integral representation and call it the _ integral holonomy representation _ of @xmath23 ( or of @xmath24 ) .
let @xmath6 be a bieberbach group of dimension @xmath0 .
a flat manifold @xmath15 is called _ generalized hantzsche - wendt _ manifold if its holonomy group is @xmath1 . in this case
@xmath23 is called a _ generalized hantzsche - wendt _ group .
we say that a bieberbach group @xmath6 is _ diagonal _ or of _ diagonal type _ if its corresponding lattice of translations @xmath21 has an orthogonal basis in which the integral holonomy representation @xmath22 diagonalizes .
we will also call _ diagonal _ the corresponding flat manifold @xmath15 . in @xcite ,
the previous definition was done with the ( slightly more restrictive ) assumption that the basis of @xmath21 is orthonormal .
we recall that diagonal manifolds have necessarily the holonomy group isomorphic to @xmath25 for some @xmath26 .
consider @xmath27 , \,\,\
, 1\leq i \leq n,\ ] ] where the @xmath28 is placed in the @xmath29 entry . for brevity we will denote @xmath30 , for @xmath31 .
the product of two of these is @xmath32 the elements in ( [ 2prod ] ) generate the group @xmath33 , of cardinality @xmath34 , and of index two in the group @xmath35 of diagonal @xmath36 matrices with entries @xmath37 in the diagonal . in the oriented case with holonomy group @xmath38 , the rationalization of the integral holonomy representation @xmath39 ( i.e. to consider @xmath21 as a @xmath4-module instead of a @xmath40-module ) , gives a unique representation @xmath41 whose image , in an appropriate basis , is exactly @xmath42 .
hence , when @xmath0 is even there are no hw manifolds , since @xmath42 contains @xmath43 , which always corresponds to an element of torsion .
to start with , we present an important structural property of the class of ghw manifolds .
[ diag ] the fundamental group of a generalized hantzsche - wendt manifold is diagonal . for the proof we start with two lemmas .
[ contains ] let @xmath44 be a diagonal faithful integral representation with @xmath45 .
then there is @xmath46 such that @xmath47 .
moreover , if furthermore @xmath48 then there is @xmath46 such that @xmath49 .
first , we regard @xmath35 , @xmath42 and @xmath50 as @xmath51-vector spaces , of dimensions @xmath0 , @xmath52 and @xmath52 respectively . also , @xmath53 .
since the @xmath54 s , for @xmath55 , are linearly independent , they can not lie all in @xmath50 simultaneously , thus there is at least one of them in @xmath56 , or equivalently , there is a @xmath57 in @xmath50 , as claimed .
the last assertion in the statement follows from a similar argument .
[ split ] let @xmath6 be a @xmath0-rank bieberbach group with translation lattice @xmath21 .
suppose that @xmath58 and @xmath59 has eigenvalues @xmath60 , @xmath28 , with corresponding eigenspaces @xmath61 and @xmath62 of dimensions @xmath60 and @xmath52 respectively .
then @xmath63 , and the orthogonal projection of @xmath64 onto @xmath61 lies in @xmath65 we have @xmath66 the torsion - free condition implies that @xmath67 , for every @xmath68 .
now , @xmath69 is the orthogonal projection onto @xmath70 . if we suppose that for some @xmath68 , @xmath71 , then @xmath72 , a contradiction .
hence , the orthogonal projection of @xmath64 ( and also of @xmath73 for every @xmath68 ) onto @xmath61 lies in @xmath74 .
the fact that @xmath61 is one - dimensional now implies that the orthogonal projection of @xmath21 onto @xmath61 is exactly @xmath75 .
hence @xmath21 splits into the direct sum of two lattices , and the lemma follows .
_ proof of theorem [ diag ] .
_ we proceed by induction on the dimension @xmath0 .
if @xmath0 is small , it is known that the theorem holds .
let @xmath22 be the integral representation of @xmath23 defined on @xmath76 and let @xmath77 be the lattice of translations of @xmath6 .
the nature of the holonomy group @xmath18 implies that each @xmath78 diagonalizes , and since they commute , they diagonalize simultaneously , i.e. , there is a @xmath4-basis @xmath79 of @xmath21 such that @xmath80_{\mathcal b}$ ] is diagonal for every @xmath81 ( and the elements in the diagonal are @xmath82 s ) .
then , lemma [ contains ] applied to these integral matrices implies that @xmath83 such that @xmath84_{\mathcal b}=\textrm{diag}(1,-1,\dots,-1)$ ] ( after reordering @xmath79 ) .
hence , @xmath85 has eigenvalues @xmath86 and @xmath28 with multiplicities @xmath60 and @xmath52 respectively .
now we can a apply lemma [ split ] , and thus @xmath21 splits into a direct sum @xmath87 where dim@xmath88 and dim@xmath89 .
also @xmath90 ( resp .
@xmath91 ) is the eigenspace of @xmath85 with eigenvalue @xmath60 ( resp .
@xmath28 ) and the orthogonal projection of the translational part corresponding to @xmath85 lies in @xmath92 . since @xmath93 for every @xmath94 , we have that @xmath95 and @xmath96 are stable under the action of @xmath97 , thus @xmath87 is a @xmath1-module decomposition .
hence we have that @xmath98 , where @xmath99 is a character and @xmath100 has rank @xmath52 , both defined on the same domain as @xmath22 .
thus , @xmath101 r|c|c| c c c r|c|c| & @xmath102 & & & & & & 1 & + @xmath103 & & & & and & & @xmath104 & & + & & @xmath105 & & & & & & @xmath43 + & & & & & & & & + we may conjugate @xmath6 by @xmath106 with @xmath107 and obtain that @xmath108 for every @xmath58 , where @xmath109 is the orthogonal projection onto span@xmath110 ( cf .
@xcite , lemma 1.4 ) .
we define a character @xmath111 by @xmath112 it is well defined since all the elements @xmath113 in the same class in @xmath114 have the same @xmath115 mod @xmath95 .
now we consider @xmath116 where @xmath117 is the canonical projection .
set @xmath118 .
it is a subgroup of @xmath6 of index 2 and its elements @xmath113 have the property that @xmath119 .
the last fact shows that if we take @xmath120 ( i.e. the restriction to the last @xmath121 coordinates of @xmath59 s and @xmath64 s ) , it is still a bieberbach group , it has rank @xmath52 and holonomy group isomorphic to @xmath122 .
( alternatively , we could have considered the subgroup @xmath123 , where @xmath124 , and observed that the restriction of @xmath125 to @xmath91 gives an isomorphism between @xmath126 and @xmath127 . )
then we can apply induction to @xmath127 , hence it is diagonal . by the relation between @xmath6 and @xmath127 it is clear that therefore @xmath6 is diagonal . in
the proof , the group denoted by @xmath128 is a subgroup of @xmath23 ( in the abstract sense , i.e. , there is a monomorphism from @xmath128 into @xmath23 ) .
this gives a way of going from one dimension to the previous one within the class of ghw manifolds , which will be used in section [ graph ] .
in other words we have [ subgroup ] every ghw group has a subgroup isomorphic to a ghw group of one lower rank . in the oriented case ,
the existence of these manifolds in each odd dimension was proved in @xcite and one realization in each odd dimension was shown in @xcite .
in @xcite the oriented flat manifolds with ( orthonormal ) diagonal holonomy and holonomy group @xmath38 were studied and they were parametrized by certain directed graphs ; also lower bounds for their cardinality were shown , growing rapidly with @xmath0 .
[ iii ] \(i ) a direct consequence of theorem [ diag ] is that in the oriented case , the above definition of generalized hantzsche - wendt manifolds coincides with the hantzsche - wendt manifolds considered in @xcite .
\(ii ) generalized hantzsche - wendt manifolds are of minimal dimension among those with holonomy group @xmath129 fixed and @xmath130 ( cf .
furthermore they are the only existing closed riemannian @xmath0-manifolds with holonomy group @xmath1 , since manifolds with finite holonomy group must be flat , by a theorem of cartan - ambrose - singer ( see @xcite , cor.3.4.7 ) .
\(iii ) for every @xmath131 , @xmath132 , there are flat manifolds with holonomy group @xmath25 not of diagonal type
. examples of this situation are easy to construct ( see end of section [ examples ] ) .
\(iv ) somehow opposite to diagonal manifolds are flat manifolds having indecomposable holonomy representation .
we recall that there are many of them , and for example , for holonomy group @xmath133 , they have been classified and the set of their dimensions is unbounded ( cf .
the class of non - oriented ghw manifolds includes two subclasses , depending on whether the first betti number @xmath134 or 1 , or equivalently ( cf . @xcite ) , the classes of bieberbach groups with trivial or non - trivial center .
let us consider first the case with non - trivial center . from theorem [ diag ] and a similar observation as at the end of section [ pn ] for hw manifolds ,
this subclass have the integral holonomy representation uniquely determined .
a simple example is the following ( cf .
let @xmath135 be the subgroup of @xmath8 generated by the set \{@xmath136 , where @xmath137 and @xmath138 are the @xmath36 orthogonal matrices defined in ( [ ci ] ) .
moreover @xmath139 and @xmath140 for other @xmath141 . here
@xmath142 denote the canonical basis of @xmath143 .
in @xcite it was proved that @xmath135 is a bieberbach group and that there exists the following short exact sequence of groups @xmath144 where the epimorphism @xmath145 is given by the composition @xmath146\rightarrow\z.\ ] ] however in general we have the following proposition
. [ chain ] let @xmath6 be a generalized hantzsche - wendt group of dimension @xmath0 with non - trivial center . then the kernel of the composition @xmath147\rightarrow \z$ ] can be an oriented or a non - oriented generalized hantzsche - wendt group . first , it is not difficult to observe that @xmath148 is a bieberbach group ( cf .
@xcite ) . from the assumptions , the abelian group @xmath149\simeq \z \oplus f$ ] where @xmath150 is a finite group ( cf .
let @xmath151 be a generator for some @xmath152 .
we can assume , by using lemma [ contains ] , that @xmath153 has eigenvalues @xmath154 , with corresponding eigenspaces @xmath155 and @xmath156 of dimensions @xmath60 and @xmath52 respectively .
moreover we have an isomorphism @xmath157 , where @xmath158 is the translation lattice of @xmath6 .
we have to show that @xmath159 is the holonomy group of the bieberbach group @xmath148 . but
this follows from the assumptions and lemma [ split ] .
to finish the proof we observe that for odd @xmath0 , every @xmath148 is non - oriented . on the other hand ,
using semidirect product @xmath160 with the integers , for any hw group @xmath161 of dimension @xmath0 , we can define a @xmath5-rank ghw group with non - trivial center such that the kernel of the map onto @xmath40 is isomorphic to @xmath161 . here
@xmath162 is a homomorphism which send @xmath60 to @xmath163 as mentioned above , the holonomy representation of ghw manifolds with @xmath164 is always equivalent to that of the above example , hence all these manifolds have the same rational homology , namely [ betti ] let @xmath165 be a generalized hantzsche - wendt flat manifold of dimension @xmath0 with first betti number one
. then @xmath166 from the definition and assumptions we have a short exact sequence @xmath167 where the action of @xmath1 on @xmath143 is given by the matrices @xmath168 , @xmath169 this defines a representation @xmath170 using a similar method as in @xcite we have a character @xmath171 of @xmath22 given by the formula @xmath172 where @xmath173 with @xmath174 the @xmath141-th of the @xmath121 factors in the product , and @xmath175 here @xmath176 denote respectively a trivial and a non - trivial character of the involution group @xmath177 . from @xcite @xmath178^{\z_2^{\,n-1}}.\ ] ] from its definition , the representation of @xmath1 on @xmath179 has the following character @xmath180 we have @xmath181 hence by induction and an immediate calculation the result follows .
according to @xcite the number of isomorphism classes of ghw manifolds and their holonomy representations in dimensions up to 6 are as in the following table . [ cols="^,>,>,>,>,>,^ " , ] the previous picture gives an idea of the bieberbach group we are constructing . in the middle column is placed @xmath182 .
we need a vector @xmath183 to form a torsion - free element @xmath184 .
we choose @xmath185 ( any other choice @xmath186 for @xmath187 would work well too ) .
now we define new elements by using the elements in @xmath161 in the following way : if @xmath188 , then we define @xmath113 by @xmath189 $ ] and @xmath190 , where @xmath191 , @xmath55 , are the coordinates of @xmath192 .
then @xmath193 is an @xmath0-rank bieberbach group with holonomy group @xmath194 ( it is torsion - free since @xmath161 is so ) .
we claim that @xmath195 is a bieberbach group with integral holonomy representation @xmath22 .
it is clear that its integral representation is @xmath22 , so the only point to be explained is that it is torsion - free .
this condition is clear for elements in @xmath196 and for @xmath184 .
now the products @xmath197 are of the forms @xmath198 with @xmath199 $ ] and @xmath200 .
since @xmath161 is torsion - free , then the elements @xmath198 are torsion - free too .
now we are interested in the holonomy representations of ghw groups .
[ number ] the total number of inequivalent integral holonomy representations occurring in generalized hantzsche - wendt bieberbach groups of dimension @xmath0 equals @xmath201 for @xmath0 even and @xmath202 for @xmath0 odd . by definition , each holonomy representation is defined by a hyperplane in @xmath203 . from linear algebra , it is the kernel of a linear map with matrix @xmath204 $ ] , where @xmath205 .
hence , we have the following @xmath0 classes of diagonal representations : @xmath206 , \,[1,1,0,\dots,0],\dots,[1,1,\dots,1,0 ] , \,[1,1,\dots,1].\ ] ] since we are considering torsion - free groups , all the hyperplanes corresponding to holonomy representations do not have the vector @xmath207 $ ] , since @xmath43 corresponds to an element of torsion .
hence we have to consider only vectors with an odd number of nonzero elements .
we claim that diagonal holonomy representations defined by such vectors are all inequivalent .
in fact , let us consider a hyperplane ( representation ) as the set of @xmath208 diagonal matrices .
it suffices to count the number of matrices @xmath153 such that the group of fix points @xmath209 of the action of @xmath153 is isomorphic to the integers @xmath40 , i.e. matrices of the form @xmath210 .
it is not difficult to see that for different hyperplanes , the number of such @xmath153 s is different .
summing up we have for @xmath0 even , @xmath201 and for @xmath0 odd , @xmath202 possibilities , which indeed occur by theorem [ exist ] . as a consequence ,
the number of integral representations occurring in non - oriented ghw groups with trivial center is @xmath211 for @xmath0 even and @xmath212 for @xmath0 odd .
in this section we present an explicit example of ghw groups with trivial center in each dimension @xmath0 . in the notation of the proof of theorem [ number ]
, its holonomy representation corresponds to hyperplanes @xmath213 $ ] in the odd case and to @xmath214 $ ] in the even case . for @xmath215 ,
let @xmath216 be the subgroup of @xmath8 generated by the set @xmath217 here @xmath218 s are the @xmath36 orthogonal matrices : @xmath219 and @xmath220 for @xmath221 .
these groups have trivial center for @xmath222 . in @xcite ( see also @xcite ) it was proved that @xmath216 are bieberbach groups for odd @xmath0 . for even @xmath0 the groups @xmath216 are also bieberbach groups ( one might also use proposition 2.1 in @xcite to prove it ) in particular because we have [ mono ] for @xmath223 there is a monomorphism @xmath224 of groups .
moreover @xmath225 is not a normal subgroup of @xmath226 .
let \{@xmath227 be a set of generators of @xmath228 we have an inclusion @xmath229 , where @xmath230 is mapped to @xmath231 next , we consider the @xmath232 matrices @xmath233 , \,1\leq i \leq n.$ ] we define @xmath234 finally , it suffices to conjugate by the translation @xmath235 an element @xmath236 to show that the image of @xmath216 is not a normal subgroup . by analogy with @xcite
we have . [ betti ] for @xmath237 , @xmath238 for odd @xmath0
it was proved in @xcite .
the proof for @xmath239 follows the same structure as the proof of theorem [ betti ] .
we have just to replace @xmath0 by @xmath240 , @xmath165 by @xmath241 , and the action of matrices @xmath54 by the action of @xmath242 as in ( [ bi ] ) , thus the characters change accordingly . now from the equivalent to formula ( [ lambda ] )
, an immediate calculation shows that no summand is trivial there , then dim@xmath243^{\z_2^{\,2k-1 } } = 0 $ ] for each @xmath244 , @xmath245 .
moreover @xmath246 @xmath247 here @xmath248 denotes the product of characters @xmath249 without the character @xmath250 hence dim@xmath251^{\z_2^{\,2k-1 } } = 1.$ ] finally , one can calculate that @xmath252 in the formulas above we use the definition of characters on the elements of @xmath253 considered as elements of @xmath177-vector space . a small modification to the groups @xmath135 defined in section [ no ] produces examples as required in remark [ iii](iii ) : we replace the @xmath254 entries in matrices @xmath54 by @xmath255 matrices of the form @xmath256 & \text{if the entry is a $ -1 $ ; and } \\[.4 cm ] \text{the $ 2\times 2 $ identity } & \left[\begin{array}{cc } 1&0\\0&1\end{array } \right ] & \text{if the entry is a 1}. \end{array}\ ] ] now the dimension increases by one , while the holonomy group remains the same .
for example , for @xmath257 , we had for @xmath258 ( the klein bottle group ) , matrix @xmath259 $ ] and vector @xmath260 ; and now we have a @xmath261 matrix @xmath262 $ ] and vector @xmath263 to obtain a 3-dimensional bieberbach group which is not of diagonal type .
hence these are , as claimed , non - diagonal examples in the case that @xmath264 equals the dimension ( where @xmath265 is the holonomy group ) .
the verification that these are indeed bieberbach groups is easy , and analogous to that for @xmath135 .
the smaller the @xmath131 ( with fixed dimension ) , the easier the construction of these examples .
in this section we give an estimate for the order of the outer automorphism group of ghw bieberbach groups . in the case of the groups @xmath135 considered in section [ no ]
, we prove that the group @xmath266 we denote by @xmath267 the order of a group @xmath153 .
[ out ] let @xmath6 be a generalized hantzsche - wendt group of dimension @xmath0 .
then we have @xmath268 .
moreover in the case with non - trivial center we have @xmath269 let @xmath270 denote the normalizer in @xmath271 of the holonomy group @xmath1 .
then @xmath270 acts in a natural way on @xmath272 .
let @xmath273 denote the cohomology class giving rise to the group @xmath274 and let @xmath275 denote its stabilizer in @xmath270 .
then @xmath1 is a normal subgroup of @xmath275 and we have a short exact sequence ( cf .
@xcite , @xcite ) @xmath276 from @xcite for hw groups , @xmath277 is semidirect product of the symmetric group @xmath278 and @xmath203 . for a non - oriented ghw group with non - trivial center , @xmath279 .
in fact , the normalizer @xmath280 because the image of the holonomy representation contains a unique element having as its diagonal entries only one 1 and the remaining are all @xmath28 s , ( see lemma [ contains ] ) .
moreover @xmath281 in the case of non - oriented @xmath6 with trivial center we can write the diagonal group @xmath203 in the group @xmath271 as a sum of a holonomy group @xmath1 and a set @xmath282 . hence @xmath283 . from ( [ ast ] )
we only have to prove that @xmath284 equals @xmath285 in the case of trivial center and @xmath34 in case of non - trivial center .
but this is proved in @xcite . for @xmath286
, @xmath287 from above the normalizer @xmath270 of the holonomy representation is equal to @xmath288 where the permutation group acts on the first @xmath121 coordinates .
let @xmath289 be a cocycle which corresponds to the group @xmath135 of section [ no ] .
then @xmath290 in fact it is enough to observe that for any non - trivial @xmath291 , @xmath292 we have @xmath293 for any generator @xmath294 and @xmath295 see @xcite .
we are looking for the permutation @xmath296 such that the cocycle @xmath297 is trivial .
we have two cases .
first assume that @xmath298 hence @xmath299 from @xmath300 it follows that @xmath301 summing up we have @xmath302 if and only if , for @xmath303 , @xmath304 and @xmath305 hence for @xmath306 we have @xmath307 and by induction @xmath308 is identity .
to finish the proof we observe that @xmath309 and it is generated by the image of @xmath43 .
now it is enough to apply the short exact sequence ( [ ast ] ) . from @xcite , for ghw manifolds @xmath310 of dimension
@xmath0 with first betti number @xmath311 , the group aff@xmath312 of affine diffeomorphisms of @xmath310 is isomorphic to the group @xmath313
in this section we start the discussion of relations between the generalized hantzsche - wendt manifolds of different dimensions . for manifolds with first betti number one we have relations from e. calabi s construction ( see proposition [ chain ] and @xcite , p.125 ) .
but we have also a lot of links between ghw groups of different dimensions in the case when one of them has trivial center .
we must say that this subject needs further research and we want to come back to it in the future . most of the observations in this section are corollaries from previous ones , and we shall omit some proofs . *
the vertices of * g * are the ghw groups ( or manifolds ) .
* we say that a vertex has dimension @xmath0 if it corresponds to a ghw manifold of dimension @xmath0 .
* edges are defined only between vertices in consecutive dimensions .
* there is an edge between two vertices ( in consecutive dimensions ) if one of the corresponding ghw groups is a subgroup of the other ( in an abstract sense , i.e. there is a monomorphism from one into the other ) . on the other hand , from the method in the proof of theorem [ exist ]
it is possible to deduce that any ghw group in dimension @xmath0 is subgroup of a ghw group in dimension @xmath315 .
moreover , we can prove that it is indeed subgroup of at least two different ghw groups in dimension @xmath315 ( we omit the proof for brevity ) . in the graph
@xmath314 this means that from any vertex there are at least two edges ` up ' . in figure
[ fig ] , a point @xmath316 ( resp . a bullet @xmath317 ) denotes a ghw manifold with first betti number zero ( resp .
one ) ; @xmath318 is the klein bottle ; @xmath319 and @xmath320 are the _ first _ and _ second amphidicosm _ respectively , they are the two non - oriented ghw manifolds of dimension 3 ; @xmath321 is the _ didicosm _ ( or the oriented hantzsche - wendt 3-dimensional flat manifold ) .
the new names are due to a proposal by john h. conway to name the ten flat 3-manifolds according to their properties , which we adopt ( this terminology is introduced in @xcite ) . in dimension 5
there are only two hw manifolds ( cf .
@xcite ) . with the method of theorem [ diag ] , it is not difficult to prove that they have both @xmath322 ( defined in section [ examples ] ) as a subgroup , and therefore the distance between them is just two .
we also have denote by @xmath23 the given ghw group . by lemma
[ contains ] there is @xmath323 such that ( after permuting coordinates ) @xmath324 and @xmath325 , where each @xmath326 is in @xmath327 .
since @xmath23 has trivial center there is another element @xmath328 with @xmath329 and the second coordinate of @xmath12 equals @xmath330 . now the integral representation corresponding to @xmath331 is the direct sum of the three non - trivial characters of @xmath332 ( with certain multiplicities ) .
then it is not difficult to check that the subgroup of @xmath23 generated by @xmath113 and @xmath333 is isomorphic to the fundamental group of the didicosm .
finally , we remark that the edges in * g * sometimes correspond to normal subgroups and sometimes not ( or they can correspond to both simultaneously ) . for instance , the chain obtained by proposition [ mono ] is not normal .
also , @xmath336 can be injected as a normal and as a non - normal subgroup in @xmath337 and in @xmath338 but it is never a normal subgroup of @xmath339 .
* are there relations between fibonacci groups and the fundamental groups of hantzsche - wendt manifolds ? ( cf .
@xcite ) * are there other geometric structures on oriented ghw manifolds ?
is it possible to introduce some invariants on them ? ( in analogy with three dimensional rational homology sphere as casson - walker and seiberg - witten invariants )
* is it possible to improve the upper bound given in theorem [ out ] ? could the factor @xmath340 be replaced by @xmath0 ? * what are the geometrical and topological properties of the ` graph ' of ghw manifolds ? | we study the family of closed riemannian @xmath0-manifolds with holonomy group isomorphic to @xmath1 , which we call generalized hantzsche - wendt manifolds .
we prove results on their structure , compute some invariants , and find relations between them , illustrated in a graph connecting the family . |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.